New localities and distribution models inform the conservation status of the endangered lizard Anolis guamuhaya (Squamata: Dactyloidae) in central Cuba

Abstract

New localities and distribution models inform the conservation status of the endangered lizard Anolis guamuhaya (Squamata: Dactyloidae) from central Cuba. Anolis guamuhaya is known from seven localities restricted to the Guamuhaya Massif in central Cuba and is always associated with mountane ecosystems above 300 m a.s.l. Previous evaluations of the conservation status of the species based on the estimated number of mature individuals have categorized the anole as Endangered. Eight new records of A. guamuhaya are provided here. These double the number of known localities, and two represent the first records of the species in lowland areas, apart from the Guamuhaya Massif. The new records extend the elevational range of the species from 15 m to above 1000 m. We used ecological niche modeling based on all of the locality records, along with what we considered the most appropriate IUCN criteria according to the available information (Criterion B) to reevaluate the conservation status of the species. These new records of A. guamuhaya increase its area of occupancy up to a total of 60 km 2 , and its extent of occurrence up to 648 km 2. Despite this increase in geographic range, the species meets the IUCN criteria in the category of Endangered. We used ecological niche modeling to predict possible trends for the species under differing scenarios of global climate change, all of which portend a drastic reduction in area climatically suitable for A. guamuhaya.

Full text

13
Phyllomedusa - 19(1), June 2020
Received 23 May 2018
Accepted 20 February 2020
Distributed June 2020
New localities and distribution models inform the
conservation status of the endangered lizard Anolis
guamuhaya (Squamata: Dactyloidae) in central Cuba
Tomás M. Rodríguez-Cabrera,1 Javier Torres,2 Carlos A. Mancina,3 Ruben Marrero,1 Yasel U.
Alfonso,4 and Ernesto Morell Savall5
1 Sociedad Cubana de Zoología. La Habana, Cuba. E-mails: tomasmichel.rodriguez@gmail.com, rubensherp89@gmail.com.
2 Department of Ecology and Evolutionary Biology, University of Kansas. Lawrence, Kansas 66045, USA. E-mail: javiertorres@
ku.edu.
3 División de Zoología de Vertebrados, Instituto de Ecología y Sistemática. La Habana, Cuba. E-mail: mancina@ecologia.cu.
4 Florida Museum of Natural History, Division of Herpetology, University of Florida. Gainesville, Florida 32611, USA.
E-mail: anoles1983cuba@gmail.com.
5 Edicio 4, Apto 7, e/ Oquendo y Silverio, Rpto Virginia, Santa Clara, Villa Clara 50100, Cuba. E-mail: ernestomorell68@
nauta.cu.
Phyllomedusa 19(1):13–33, 2020
© 2020 Universidade de São Paulo - ESALQ
ISSN 1519-1397 (print) / ISSN 2316-9079 (online)
doi: http://dx.doi.org/10.11606/issn.2316-9079.v19i1p13-33
Abstract
New localities and distribution models inform the conservation status of the
endangered lizard Anolis guamuhaya (Squamata: Dactyloidae) from central Cuba.
Anolis guamuhaya is known from seven localities restricted to the Guamuhaya Massif in
central Cuba and is always associated with mountane ecosystems above 300 m a.s.l.
Previous evaluations of the conservation status of the species based on the estimated
number of mature individuals have categorized the anole as Endangered. Eight new
records of A. guamuhaya are provided here. These double the number of known localities,
and two represent the rst records of the species in lowland areas, apart from the
Guamuhaya Massif. The new records extend the elevational range of the species from 15
m to above 1000 m. We used ecological niche modeling based on all of the locality
records, along with what we considered the most appropriate IUCN criteria according to
the available information (Criterion B) to reevaluate the conservation status of the
species. These new records of A. guamuhaya increase its area of occupancy up to a total
of 60 km2, and its extent of occurrence up to 648 km2. Despite this increase in geographic
range, the species meets the IUCN criteria in the category of Endangered. We used
ecological niche modeling to predict possible trends for the species under differing
scenarios of global climate change, all of which portend a drastic reduction in area
climatically suitable for A. guamuhaya.

14
Phyllomedusa - 19(1), June 2020
Keywords: Climate change, ecological niche modeling, natural history, Twig-giant Anole,
West Indies.
Resumen
Nuevas localidades y modelos de distribución informan sobre el estatus de conservación del
lagarto amenazado Anolis guamuhaya (Squamata: Dactyloidae) de Cuba central. Anolis
guamuhaya se conoce de siete localidades restringidas al Macizo de Guamuhaya, en Cuba central,
siempre asociado a ecosistemas de montaña por encima de los 300 m s.n.m. Evaluaciones anteriores
de su estatus de conservación basadas en el número estimado de individuos maduros categorizaron
a la especie En Peligro. En este trabajo damos a conocer ocho nuevos registros de A. guamuhaya.
Con estos se duplica el número de localidades conocidas y dos de ellas constituyen los primeros
registros de esta especie en zonas llanas, fuera del Macizo de Guamuhaya. Estos nuevos registros
expanden el rango altitudinal de la especie desde 15 m hasta por encima de los 1000 m. Teniendo
en cuenta los registros previos y los nuevos, hacemos una reevaluación del estatus de conservación
de la especie empleando la modelación de nicho ecológico y los criterios de la UICN que
consideramos más apropiados de acuerdo a la información disponible (criterio B). Estos nuevos
registros de A. guamuhaya aumentan su área de ocupación hasta un total de 60 km2 y su extensión
de presencia hasta 648 km2. A pesar de este incremento en su área de distribución, la especie se
ajusta a los criterios de la UICN para la categoría de En Peligro. Usamos la modelación de nicho
ecológico para predecir posibles tendencias de la especie bajo diferentes escenarios de cambio
climático, donde todos los modelos auguran una reducción drástica del área climáticamente idónea
para la especie.
Palabras claves: cambio climático, ecomorfo gigante de ramita, historia natural, modelación de
nicho ecológico, Las Antillas.
Resumo
Novas localidades e modelos de distribuição informam sobre o estado de conservação do
lagarto Anolis guamuhaya (Squamata: Dactyloidae) de Cuba central. Anolis guamuhaya é
conhecida de sete localidades restritas ao Maciço de Guamuhaya, no centro de Cuba, sempre
associada a ecossistemas montanhosos acima de 300 m de altitude. Avaliações anteriores de seu
estado de conservação baseadas no número estimado de indivíduos maduros categorizaram a espécie
como Em Perigo. Neste trabalho, fornecemos oito novos registros de A. guamuhaya. Com esses
registros, duplica-se o número de localidades conhecidas, e duas delas constituem os primeiros
registros da espécie em zonas baixas fora do Maciço de Guamuhaya. Os novos registros expandem
a amplitude altitudinal da espécie desde 15 m até acima dos 1000 m. Levando em conta os registros
prévios e os novos, reavaliamos o estado de conservação da espécie empregando modelagem de
nicho ecológico e os critérios da IUCN que consideramos mais apropriados de acordo com a
informação disponível (Critério B). Os novos registros de A. guamuhaya aumentam sua área de
ocupação para um total de 60 km2 e sua extensão de presença para 648 km2. Apesar desse incremento
em sua área de distribuição, a espécie ajusta-se aos critérios da IUCN para a categoria Em Perigo.
Utilizamos modelagem de nicho ecológico para prever possíveis tendências para a espécie em
diferentes cenários de mudanças climáticas globais, todas pressagiando uma redução drástica em sua
área climática adequada.
Palavras-chave: anoles-graveto-gigante, Antilhas, história natural, modelagem de nicho ecológico,
mudanças climáticas.
Rodríguez-Cabrera et al.

15
Phyllomedusa - 19(1), June 2020
Introduction
There are six species of anoles in the
Chamaeleolis clade, all endemic to Cuba
(Rodríguez et al. 2013, Poe et al. 2017). These
large (154–177 mm snout–vent length), slow-
moving, chameleon-like anoles have massive
heads with prominent parietal bones that form a
casque-like structure, and posterior molariform
teeth that may enable the lizards to consume
hard prey, such as gastropod mollusks (Garrido
and Schwartz 1968, Díaz et al. 1998, Herrel and
Holanova 2008). Among the morphological
adaptations of these lizards for a strict arboreal
lifestyle and use of narrow perches are a semi-
prehensile, non-autotomous, blunt tail and short
limbs (Garrido and Schwartz 1968, Garrido
1982, Rodríguez 1999). Given the striking
differences of these anoles from their congeners,
they were placed in their own genus—
Chamaeleolis (e.g., Garrido and Schwartz 1968,
Garrido 1982, Rodríguez 1999). However,
Chamaeleolis currently is considered a junior
synonym of Anolis because it is nested within
the Anolis radiation (Hass et al. 1993, Jackman
et al. 1999, Poe 2013, Pyron et al. 2013, Poe et
al. 2017).
Anolis guamuhaya (Garrido, Pérez-Beato,
and Moreno, 1991) is a montane twig-giant
anole currently thought to be restricted to
ecosystems above 300 m a.s.l. in the Guamuhaya
Massif (= Escambray) (e.g., Garrido et al. 1991,
Rodríguez et al. 2013). The massif is located in
south-central Cuba and is composed of two main
topographical units—the Trinidad Range in the
west (the largest unit) and the Sancti Spíritus
Range in the east (the smallest unit); the
Agabama River Basin separates these mountain
ranges (Mateo-Rodríguez and Acevedo-González
1989; Figure 1). Garrido et al. (1991) stated that
the holotype of A. guamuhaya was collected on
the road between Jibacoa and Topes de Collantes,
Trinidad Range, at about 900 m a.s.l. However,
the highest elevations between these two
localities is 869 m a.s.l. at Sierra Guaniquical (4
km SSW Jibacoa) and 851 m a.s.l. at Loma
Guaniquical (6 km SSE Jibacoa), and the road
does not exceed 800 m a.s.l. at any point.
Elevations of 900 m near to the type locality of
A. guamuhaya occur only at Pico de Potrerillo
(973 m a.s.l., 3.3 km SE Topes de Collantes) and
Pico Tuerto (919 m a.s.l., 4.3 km WSW Jibacoa,
western side of Hanabanilla Dam). Access to
cartographic maps and GPS was limited when
Garrido et al. (1991) described this species, and
in interviews with Orlando H. Garrido, we could
not conrm the exact collecting site of the
holotype. Therefore, according to the
International Code of Zoological Nomenclature
(ITZN 1999; Article 76), the type locality of A.
guamuhaya includes the entire strip of road
between Jibacoa and Topes de Collantes (about
22 km). Additional specimens were collected at
Topes de Collantes and surrounding areas, Sancti
Spíritus Province (Garrido and Schwartz 1968,
Garrido et al. 1991, Rodríguez et al. 2013), and
in the vicinity of Aguacate, Cienfuegos Province
(Garrido 1982, Garrido et al. 1991). The adult
male collected by Wilson (1957) at Mina Carlota,
Cienfuegos Province, was initially assigned to A.
porcus (Cope, 1864) by Garrido and Schwartz
(1968) and later to A. guamuhaya by Garrido et
al. (1991). Most recently, Torres et al. (2015)
reported the species from near Pico San Juan (at
1015 m a.s.l.), Cienfuegos Province.
The detectability of species is sometimes a
critical factor in estimating population density
and other demographic parameters (e.g., Mancina
and Cruz 2017). The coloration and behavior of
twig-giant anoles are highly effective in
camouaging the lizards (Garrido and Schwartz
1968, Leal and Losos 2000). Because it is so
difcult to detect them by day, we assume that
the lizards are more abundant than they seem to
be. Nevertheless, the conservation status of some
species in the Chamaeleolis clade, including A.
guamuhaya, has been evaluated based only on
the IUCN Criterion D, which refers to the
estimated number of mature individuals (IUCN
2001, 2012, IUCN Standards and Petitions
Subcommittee 2014). Rodríguez (1999, 2012)
listed A. guamuhaya as Endangered; however,
New localities and distribution models for Anolis guamuhaya in Cuba

16
Phyllomedusa - 19(1), June 2020
Figure 1. Distribution of Anolis guamuhaya in south-central Cuba with locality records from the literature (red dots)
and this study (green triangles): (1) Santa Martina, (2) Boca Ambuila, (3) Mina Carlota, (4) Carso de Buenos
Aires, (5) Pico San Juan, (6) ca. 6 km SW of Aguacate, (7) Río Negro Touristic Center, (8) Loma Cariblanca,
(9) Trinitario Stream waterfall, (10) La Chispa, (11) Topes de Collantes and vicinity, (12) Vegas Grandes, (13)
Caburní waterfall, (14) midway on the road between Jibacoa and Topes de Collantes, and (15) resurgence
of the Jibacoa River. Note the two main topographical units that form the Guamuhaya Massif (the Trinidad
Range in the west and the Sancti Spíritus Range in the east), divided by the Agabama River Basin.
this author commented that the population size
of the species was unknown and that all records
were restricted to a region of about 14 km2 in the
Sancti Spíritus Province, with a calculated area
of occupancy of 8 km2. Also, this author
overlooked two localities from Cienfuegos
Province—Mina Carlota, and ca. 6 km SW of
Aguacate (Wilson 1957, Garrido and Schwartz
1968, Garrido 1982, Garrido et al. 1991).
However, the reclusive habits and camouage of
twig-giant anoles (e.g., Garrido and Schwartz
1968, Leal and Losos 2000) make it extremely
difcult to estimate population size accurately
by counting in searches by day, in contrast to
gathering data for other more conspicuous anoles
(e.g., trunk-ground anoles). Thus, we consider
that the population size of A. guamuhaya may be
substantially greater than estimates based on
Criterion D suggest.
The Guamuhaya Massif has been intensively
and extensively deforested, particularly during
the last two centuries (e.g., Domínguez-González
and Acosta-Rodríguez 2012, Mancina et al.
2017). Given the highly arboreal habits of twig-
giant anoles, the loss of forest coverage probably
has had negative effects on the population of A.
guamuhaya. Herein we report more than twice
the number of localities previously known for
Rodríguez-Cabrera et al.

17
Phyllomedusa - 19(1), June 2020
the species and reevaluate its conservation status
applying what we consider the most appropriate
IUCN criteria according to the available
information and taking into account the
behavioral particularities of twig-giant anoles.
We used ecological niche modeling to test the
possible response of A. guamuhaya to the effects
of different future scenarios of global climate
change. Last, we discussed implications for the
conservation of the species.
Materials and Methods
Data Collection
The specimens of A. guamuhaya reported here
were collected between 1993 and 2014; additional
surveys were conducted until late-2018. Two
methods were effective in detecting these lizards.
In the rst, small groups composed of two to ve
people separated by about 2 m from one another
walk in line across the forest by day. When a
lizard tries to hide from the rst person by moving
to the opposite side of the trunk or branch, it is
more likely to be seen by the next person. The
second method, with which we obtained the best
results, is searching at night when the anoles are
sleeping; by night, the twig-giant anoles are
whitish in color, and therefore, easily detectable
in darkness in the white light of a headlamp. It is
necessary to check all vegetation strata, because
these anoles may be found sleeping as low as 50
cm above the ground on shrubby vegetation and
vines to more than 5 m high in the canopy (Leal
and Losos 2000, authors pers. obs.).
We included seven literature records
(Rodríguez et al. 2013, Torres et al. 2015) and
eight new localities of A. guamuhaya in our
analyses (Figure 1; Appendix I). To include the
holotype record in this study, we assigned its
locality to a point midway on the road between
Jibacoa and Topes de Collantes (Figure 1). We
collected geographic coordinates directly in the
eld with a GPS Garmin eTrex Vista HC or
obtained them from MapInfo Professional ver.
10.5. Datum for all coordinates is WGS 84. We
also accessed existing GIS data available for
voucher specimens deposited in several
collections. Specimens were collected under the
project “Colecciones zoológicas, su conservación
y manejo II,” hosted by the Instituto de Ecología
y Sistemática and deposited in the zoological
collection of the latter institution and the Museo
de Historia Natural “Felipe Poey” (Appendix I).
Acronyms and abbreviations are as follow: SVL
= snout–vent length, TL = tail length;
CZACC: Instituto de Ecología y Sistemática, La
Habana; MFP: Museo de Historia Natural
“Felipe Poey”, Facultad de Biología, Universidad
de La Habana, La Habana; MNHNCu: Museo
Nacional de Historia Natural de Cuba, La
Habana; MCZ: Museum of Comparative Zoology,
Harvard University, Cambridge, Massachusetts,
USA; USNM: National Museum of Natural
History, Smithsonian Institution, Washington
D.C., USA; NMP: National Museum of Prague,
Czech Republic.
For the purposes of this work and following
the concept of “location” of the IUCN Standards
and Petitions Subcommittee (2014; section 4.11),
we considered all locality records separated by
less than 2 km to be the same location (e.g.,
Topes de Collantes and vicinity; Garrido and
Schwartz 1968, Garrido 1982, Rodríguez et al.
2013; Appendix I) and provided a central
reference coordinate and elevation. After
applying this criterion, the 32 records of A.
guamuhaya represent 15 different locations,
eight of which are new ones reported here
(Figure 1; Appendix I). For the reevaluation of
the conservation status we used the IUCN
Criterion B (IUCN 2001, 2012, IUCN Standards
and Petitions Subcommittee 2014).
Ecological Niche Modeling
Occurrence and study region.—We cons-
tructed Ecological Niche Models (ENMs) of A.
guamuhaya with the maximum entropy algorithm
Maxent (v.3.3.3; Phillips et al. 2006). Maxent
examines the relationship between known
occurrence records of the species and the
New localities and distribution models for Anolis guamuhaya in Cuba

18
Phyllomedusa - 19(1), June 2020
environmental characteristics of that area, and
then extrapolates from the areas where similar
conditions occur in the study region (Phillips and
Dudik 2008). We dened the study region for
model calibration using a convex hull that
enclosed the occurrence records, and then
buffered this at a distance of 50 km (Figure 2).
We considered this extent appropriate as
background selection because it does not include
large regions that the species does not inhabit
Figure 2. Geographical extent (white dots), convex hull (blue polygon) and background sampling (red area) considered
in this study.
Rodríguez-Cabrera et al.

19
Phyllomedusa - 19(1), June 2020
owing to dispersal limitations and/or biotic
interactions, such as competition with similar
species (Barve et al. 2011, Anderson and Raza
2010). We extracted climate information using
ArcGIS v. 10.3 (ESRI, Redlands) and ran the
model by randomly sampling 5000 background
points within this extent.
Environmental variables and climatic
scenarios.—As current and future predictor
variables, we used 19 bioclimatic variables with
a resolution of ca. 1 km2 (0.74 km2 for Cuba)
downloaded from the WorldClim ver. 1.4
database (http://www.worldclim.com) (Hijmans
et al. 2005). We rst ran a model in MaxEnt ver.
3.3.3k (Phillips et al. 2006), including all
variables, to identify those with the highest
signicance to the model and selected the
bioclimatic predictors from the jackknife test of
variable importance. We selected six variables
with low collinearity (Pearson’s r < 0.8) that
could be ecologically important to reptiles, as
follow: mean diurnal temperature range (Bio 2);
temperature seasonality (Bio 4); maximum
temperature of warmest month (Bio 5); preci-
pitation of wettest month (Bio 13); seasonality of
precipitation (Bio 15); and precipitation of driest
quarter (Bio 17). The use of more predictors
could generate overtting and could reduce
temporal transfer (Peterson et al. 2011, Anderson
2013, Radosavljevic and Anderson 2014).
Our prediction is based on bioclimatic
envelope modeling, which changes with coupled
10 general circulation models (GCMs): BCC-
CSM1-1, CCSM4, CNRM-CM5, HadGEM2-ES,
MIROC5, MPI-ESM-LR, MRI-CGCM3, GISS-
E2-R, NorESM1-M, IPSL-CM5A. Different
GCMs and greenhouse gas scenarios will lead to
various changes in distributions of species in the
future. The Intergovernmental Panel on Climate
Change (IPCC 2014) in its Fifth Assessment
Report (AR5) proposes four Representative
Concentration Pathways (RCPs). RCPs may be
better than the emission scenarios developed in
the Special Report on Emissions Scenarios
(SRES); hence, RCPs have replaced SRES
standards (Moss et al. 2010). The two pathways
(RCP 2.6 and RCP 8.5) represent two possible
radiative forcing values (+ 2.6 and + 8.5 W/m2,
respectively) (Moss et al. 2010). We used data
from 1950–2000 as baseline climate data. Ten
GCMs were used for the years 2050s and 2070s.
For each GCM, we used the two RCPs to evaluate
different greenhouse gas scenarios. Hence, the
total number of climate scenarios considered was
40 (10 GCMs, two scenarios and two-time steps).
Ecological niche modeling.—We made a
correlative niche model relating environmental
conditions to 15 presence records using MaxEnt
software (Phillips et al. 2006). We optimized
model complexity and predictive power using
the R package ENMeval (Muscarella et al. 2014)
to select the optimal combination of the
regularization multiplier and the combination of
feature classes. We tested regularization
multiplier values from 0.5–6.0 with intervals of
0.5, and evaluated ve settings of feature classes,
as follow: linear (L); linear and quadratic (Q);
hinge (H); LQH; and LQH plus product (P). The
performance of model was evaluated using
Akaike Information Criterion corrected for small
sample sizes (AICc) (Warren and Seifert 2011).
The nal model was run in Maxent using 50
bootstrapping replicates and the combination of
regularization multiplier and feature classes with
lowest AICc using all occurrence records.
Additionally, we examined the omission rate and
tested AUC to assure that the model selected as
optimal performed well.
We used the “minimum training presence”
threshold value to discriminate suitable from non-
suitable habitat (Liu et al. 2005, Radosavljevic
and Anderson 2014). We used the logistic output
of MaxEnt (continuous model), and applied
thresholds in ArcGIS to convert it into a presence-
absence model. Then we clipped the binary
models by the shape les of natural forests
(Estrada et al. 2011) and protected areas (CNAP
2013) to examine the extent of the ecological
niche covered by both categories. The ecological
niche model generated for A. guamuhaya was
New localities and distribution models for Anolis guamuhaya in Cuba

20
Phyllomedusa - 19(1), June 2020
projected onto two future climate change scenarios
and they were added for each RCPs using the
median value for all projections—i.e., we used the
consensus of only those areas where ve or more
GCM predict the occurrence of environmentally
suitable areas in the future. We subtracted the
future and current niche models from each other,
and areas of contraction, expansion and stability
were calculated.
Reassessment of Conservation Status
Based on the information available for A.
guamuhaya, we consider that the most accurate
evaluation of its conservation status must follow
the IUCN Criterion B—viz., area of occupancy
(AOO) and extent of occurrence (EOO) (IUCN
2001, 2012, IUCN Standards and Petitions
Subcommittee 2014). We calculated the AOO and
the EOO using ArcGIS v. 10.3 (ESRI, Redlands).
For the AOO we used the standardized reference
scale grid of 4 km2 cells proposed by the IUCN
(IUCN 2001, 2012, IUCN Standards and Petitions
Subcommittee 2014). The EOO was estimated
with two approaches: (1) the convex hull (e.g.,
IUCN Standards and Petitions Subcommittee
2014) and (2) the environmentally suitable area
resulting from ecological niche modeling clipped
by the layer of natural forests. Both approaches
are biased toward overestimation. Thus, the
convex hull can include several areas of unsuitable
habitat (IUCN 2001, 2012, IUCN Standards and
Petitions Subcommittee 2014) and ecological
niche modeling is unable to detect environmentally
suitable areas never occupied by the species
(historical biogeography, limited dispersal ability)
and the exact areas occupied by ecologically
similar species (interference competition).
However, ecological niche modeling seems a
more realistic measure of habitat suitability; thus,
it was used here as the nal estimation of EOO.
There is also a bias toward underestimation
because of the highly irregular topography of the
montane region that contains a larger surface area
than that of a at landscape.
Twig-giant anoles are arboreal; therefore, it
is reasonable to expect the species to have a
strong dependence on forested areas, where
canopy connectivity may allow them to disperse.
There are no estimates of the average dispersal
distance in A. guamuhaya or any other member
of the Chamaeleolis clade; nevertheless, using
the factors mentioned above, we could qualify
the species (and probably most species within
this group) as having poor dispersal ability
(reviewed in IUCN Standards and Petitions
Subcommittee 2014). Therefore, we considered
natural forest coverage to be an important
variable in estimating habitat decline and EOO.
Cuba has a long history of forest depletion
resulting from the extensive development of the
sugar industry, agriculture, and stockbreeding,
which resulted in a reduction of forest coverage
of more than 80% between the 16th century and
the mid-20th century (Del Risco 1995, Gutiérrez-
Domech and Rivero-Glean 1997). And there are
few signs of recovery now (ONEI 2015). We
would expect that such an extreme reduction in
forest coverage has had long-lasting negative
effects on populations of native, forest-dwelling
species such as A. guamuhaya. The Guamuhaya
Massif is a mountain range where native
ecosystems have persisted. However, the primary
vegetation has suffered serious impacts from
forest res and intensive felling for coffee
plantations and stockbreeding (e.g., Domínguez-
González and Acosta-Rodríguez 2012, Mancina
et al. 2017). Consequently, we consider that the
IUCN Criterion Bb(iii), which refers to
continuous decline in the effective area extent
and/or quality of habitat, is fully applicable to A.
guamuhaya.
Results
Geographic Distribution and Natural History
With the eight new localities reported here,
the current distribution of A. guamuhaya includes
nearly the entire Trinidad Range and encompasses
Rodríguez-Cabrera et al.

21
Phyllomedusa - 19(1), June 2020
Specimen Age Veg. Perch Observations
Type ØHeight
Cienfuegos
MFP 12515 J SEC SB 1.1 305 Basking, body temperature 28°C, air temperature 27.9°C
No voucher J SEC SB 1.6 240 Pico San Juan, body temperature 28°C, air temperature 27.9°C
CZACC 4.5871 A MRF TB 1.5 400 Sleeping at night in horizontal position.
MFP 12.633 A EGF SB 1.3 50 Swallowing a snail (Zachrysia sp.)
No voucher A EGF ST 2.5 100 Carso de Buenos Aires
CZACC 4.5846 J ECO ST 2.2 110 On vertical stem, by a trail by day
MFP 12.629 A SDF TT 27 200 On smooth palm tree trunk, head down
No voucher J ECO VI 0.3 120 Santa Martina, on oblique twig, head up
MFP 12.603 A SDF TB 1.8 320 Sleeping at night in vertical position, head up
CZACC 4.5850 J ECO SB 0.5 50 Sleeping at night in horizontal position
CZACC 4.5851 J ECO VI 0.4 200 Sleeping at night in horizontal position
Sancti Spíritus
No voucher A EGF VI 2 200 Resurgence of the Jibacoa river,
on a vertical vine near a cave entrance
Table 1. Summary of ecological data of specimens of Anolis guamuhaya examined in the field. Measurements are
presented in centimeters. Juveniles (J) refer to small individuals less than 100 mm SVL and adults (A) refer to
individuals larger than 100 mm SVL. Abbreviations: Veg. = vegetation type; SDF = semideciduous forest;
EGF = evergreen forest; MRF = mountain rainforest; ECO = ecotonal vegetation; SEC = secondary vegetation;
Ø = perch diameter; SB = shrub branch; ST = shrub trunk; TT = tree trunk; TB = tree branch; VI = vine.
the three central provinces—Cienfuegos, Villa
Clara, and Sancti Spíritus (Figure 1; Appendix
I). The three new records from Villa Clara are
the rst for this province. Two of the new
localities are the rst lowland records for A.
guamuhaya, which was found in isolated patches
of subcoastal semideciduous forest, about 20 km
west of the nearest previous mountain record at
Pico San Juan. The altitudinal distribution is 15–
1015 m a.s.l.
The lowland and montane A. guamuhaya
have the same scale pattern described for the
holotype. In life, the dorsum of the montane
lizards is greenish, whereas that of individuals
from the lowlands grayish. The dewlap of
montane lizards has more black areas than that
of the lowland lizards (Figure 3); however,
dewlap coloration can vary. These differences
may indicate local adaptations (dry subcoastal
vegetation vs. mountain rainforest and evergreen
forest) and/or genetic drift. The isolated lowland
records apparently represent outlying occurrences
of the species; these might reect a reduction in
suitable habitat (primary forest) in the transition
zone between subcoastal and mountain areas
(Ecological Niche Modeling below) and
subsequent isolation of the lizards as meta-
populations.
We found Anolis guamuhaya mostly
associated with primary forest (Figure 4).
However, many immature individuals were
associated with ecotonal or secondary shrubby
vegetation, but these areas always are in direct
contact with primary forest (Table 1). The
specimens deposited in the collection of the
NMP have no ecological data.
New localities and distribution models for Anolis guamuhaya in Cuba

22
Phyllomedusa - 19(1), June 2020
A
C
B
D
Figure 3. Selection of specimens of Anolis guamuhaya examined for this study: (A) male CZACC 4.5848 from Santa
Martina; (B) female CZACC 4.5849 from Boca Ambuila; (C) female CZACC 4.5845 from Carso de Buenos
Aires, notice the camouflage; and (D) male from Loma Cariblanca, kept alive at the Parque Zoológico
Nacional, La Habana. Photographs by RM (A, B, D) and TMRC (C).
Ecological Niche Modeling
The ecological niche models for A. gua muhaya
have mean values of AUCtraining = 0.938 ± 0.006,
AUCtest = 0.940 ± 0.005, mean AUCdiff = 0.025 ±
0.0006 SD and mean of minimum training presence
omission rate = 0.227 ± 0.103 SD (Appendix II).
These results indicate good model performance
and low overtting. The lowest values of AICc
(ΔAICc = 0) resulted from regularization multiplier
of 1 with linear and quadratic feature classes. The
three most important bioclimatic variables are (1)
the maximum temperature of the warmest month;
(2) mean diurnal temperature range; and (3)
temperature seasonality (Appendix III). The current
clima tically suitable area encompasses more than
1398 km2 (Figure 5A), but only 46% (648 km2) is
covered by natural forests (Figure 5B).
Under all climate change scenarios, the models
predict a reduction of the suitable area; however,
the amount of reduction is variable (Figure 6). The
models suggest a reduction of suitable areas
between 43–67% in 2050 (Figure 6A, B) and
between 44–94% in 2070 (Figure 6C, D). The
largest estimated percent loss in suitable areas
occurs under RCP 8.5 scenario for 2070 (Figure
Rodríguez-Cabrera et al.

23
Phyllomedusa - 19(1), June 2020
Figure 4. Different habitats of Anolis guamuhaya: (A) subcoastal semideciduous forest on limestone at Boca Ambuila;
(B) trail crossing subcoastal semideciduous forest at Santa Martina; (C) evergreen forest on limestone
at Carso de Buenos Aires; and (D) mountain rainforest at Pico San Juan. Photographs by TMRC (A-C)
and RM (D).
A
C
B
D
6D) in which only 77 km2 (6% of the current
suitable area) will remain with suitable climatic
conditions for the occurrence of A. guamuhaya.
However, all these models are clearly
overestimated, because they do not include the
natural forest coverage, which is expected to be
drastically reduced as a collateral effect of
climate change.
Conservation Status
We assigned a grid size of 4 km2 per locality
and summarized the 15 records from the literature
and this study to estimate the range (AOO) of A.
guamuhaya at 60 km2. The convex hull revealed
an EOO of 622 km2, but the climatically suitable
area clipped by the layer of natural forests
revealed an EOO of 648 km2, considered herein
as the most realistic approach to the EOO (Figure
5B). We consider that the loss of forest coverage
during the last centuries is not only a measure of
“continuing decline in area, extend and/or quality
of habitat,” but also a measure of the extreme
decrease in the EOO. Moreover, only 21% (138
km2) of the climatically suitable area covered by
natural forests (= EOO) and only seven of the
fteen localities (47%) of A. guamuhaya fall
within protected areas (Figure 5B).
New localities and distribution models for Anolis guamuhaya in Cuba

24
Phyllomedusa - 19(1), June 2020
Figure 5. Ecological niche model for Anolis guamuhaya based on presence records (white dots) from the literature and
this study. (A) Binary representation of the current environmentally suitable area (red shading) based on the
“minimum training presence” threshold (consensus of the scenarios where five or more GCM coincide). (B)
Current distribution of environmentally suitable areas clipped by natural forests (green shadings, assumed
herein as the EOO) and then by protected areas (blue contours).
Rodríguez-Cabrera et al.

25
Phyllomedusa - 19(1), June 2020
Figure 6. Ecological niche model percent reduction of environmentally suitable areas for Anolis guamuhaya projected
under two future climate change scenarios (RCP: 2.6 W/m2 and 8.5 W/m2) for the years 2050 and 2070. The
blue shading represents the sum of those areas where five or more GCMs predict environmentally suitable
conditions. The red shading represents the current environmentally suitable areas, illustrating retraction
under future scenarios; no expansion was observed under any scenario.
A
C
B
D
Discussion
Most authors have reported A. guamuhaya
to occur above 300 m a.s.l. (Garrido et al. 1991,
Rodríguez 2012, Rodríguez et al. 2010, 2013);
Henderson and Powell (2009) reported an
elevational distribution for this species from
sea level to 900 m, although no voucher
specimens from lowland localities are
mentioned. The specimens reported here from
the southern coast of Cienfuegos Province are
the rst vouchered records of this species in
lowland ecosystems.
The greater number of voucher specimens
from a cluster of localities in the vicinity of
Topes de Collantes does not necessarily imply
higher population density. It is more likely that it
reects a greater sampling effort because this
area attracts many tourists, as well as researchers.
As observed here and reported in the
literature, A. guamuhaya is both a mountain- and
lowland-dwelling species, ranging from close to
the sea level to above 1000 m. However,
environmentally suitable areas and primary
forest coverage in south-central Cuba are scarce
and highly fragmented apart from the Guamuhaya
New localities and distribution models for Anolis guamuhaya in Cuba

26
Phyllomedusa - 19(1), June 2020
Massif (Figures 5 and 7). Montane ecosystems
are the last natural refuge for this species under
the worst-case scenarios of climate change in the
future (Figure 6).
Climate change and the modication and
fragmentation of natural forests seem to
challenge the survival of A. guamuhaya. Changes
in range size usually are assessed by considering
the climatic characteristics of current distributions
and the projected distribution of these climatic
conditions in the future. However, the
vulnerability of this lizard may be exacerbated
by other factors. These anoles seem to have a
limited dispersal ability; further, they are affected
by the loss and fragmentation of forest patches
within climatically suitable areas, because forest
coverage/canopy connectivity may affect the
conditions (microclimate) and resources (habitat
structure) required by the species.
The model identied environmentally suitable
areas for A. guamuhaya only in the Guamuhaya
Massif and its nearby surroundings, including in
the Sancti Spíritus Range, about 30 km east of the
nearest records of the species (Figure 5). However,
despite extensive surveys through the years, the
only twig-giant anole that has been found so far in
the latter area is A. chamaeleonides Duméril and
Bibron, 1837, which also occurs in the southeastern
portion of the Trinidad Range, close to A.
guamuhaya (Figure 7; Garrido et al. 1991, Estrada
1994, Rodríguez et al. 2013, authors pers. obs.).
All species in the Chamaeleolis clade apparently
have similar ecological requirements, and
sympatry rarely has been reported (e.g., Rodríguez
Figure 7. Records of Anolis guamuhaya (red dots) and A. chamaeleonides (orange squares) in south-central Cuba
(sensu Rodríguez et al. 2013 and this study), superimposed to the layer of current natural forests (green
shading) and protected areas (striped zones). Question mark represents the juvenile male CZACC 4.5846
from Santa Martina, tentatively assigned to A. chamaeleonides.
Rodríguez-Cabrera et al.

27
Phyllomedusa - 19(1), June 2020
et al. 2013), which might be indicative of
competitive exclusion (Morin 2011). Higher
resource requirements in giant anoles may indicate
a stronger interference competition among
ecologically similar species. For example, A.
guamuhaya and A. chamaeleonides apparently
occur sympatrically at Topes de Collantes
(Rodríguez et al. 2013) but the overlap in
distribution seems to be minimal (Figure 7).
Because no ecological studies exist on interspecic
interactions between these species, this scenario
provides a unique opportunity to study resource
partitioning between twig-giant anoles.
We found one juvenile specimen at Santa
Martina (CZACC 4.5846) that we tentatively
assigned to A. chamaeleonides (because it had a
scale row between the supralabials and
infraoculars; Garrido et al. 1991), but its small
size made it difcult to identify unambiguously
(Figure 8; Appendix I). Thus, a second probable
point of contact between A. guamuhaya and A.
chamaeleonides might be Santa Martina. This
seems plausible because the latter species is
known to occur in similar ecological conditions
about 20 km northwestward, near Juraguá, on
the western side of Cienfuegos Bay (Figure 8;
Garrido 1980, Rodríguez et al. 2013). Anolis
chamaeleonides is widespread in Cuba
(Rodríguez et al. 2013), from sea level to above
1000 m (Rodríguez et al. 2010), which suggests
that it is more generalist than A. guamuhaya.
Therefore, its absence throughout most of the
range of A. guamuhaya in the Trinidad Range
supports the hypothesis of strong competition
between both species.
This is an example of the importance of
ground-truthing to assess the incidence of biotic
interactions (e.g., competition, predators, food
availability) and abiotic factors (e.g., geographic
and ecological barriers) in species distributions
when using ecological niche modeling (Soberón
and Nakamura 2009). The climatically suitable
areas generated by the models do not necessarily
represent the full distribution of the species.
Given the apparently low dispersal ability of A.
guamuhaya and strong barrier imposed by the
Figure 8. Juvenile male CZACC 4.5846 (52.2 mm SVL)
from Santa Martina, tentatively assigned to
Anolis chamaeleonides. The inset above
shows details of the scale row between
supralabials and infraoculars (arrow, also the
scale row in was outlined to increase
legibility), which is a diagnostic character for
this species. The inset below shows details of
the head of a juvenile male A. guamuhaya of
comparable size (57 mm SVL) from the
southwestern slope of Pico San Juan, notice
the supralabials in contact with the
infraoculars. Photographs by TMRC.
Agabama Basin, it is possible that this species
never dispersed into the Sancti Spíritus Range,
despite the occurrence of climatically suitable
habitats in these mountains. Another possibility
is that A. chamaeleonides has been progressively
displacing A. guamuhaya from east to west. The
severe deforestation of the Sancti Spíritus Range
in the past (e.g., Domínguez-González and
Acosta-Rodríguez 2012, Mancina et al. 2017)
New localities and distribution models for Anolis guamuhaya in Cuba

28
Phyllomedusa - 19(1), June 2020
translates to strong habitat modications that
might have produced the retreat of A.
guamuhaya and favored the advance of A.
chamaeleonides. This hypothesis could be
tested by using molecular markers to assess if
the populations of A. chamaelonides found in
the Sancti Spiritus range came from nearby
eastern populations.
Another factor that may affect A. guamuhaya
and potentially limit its distribution is competition
with ecologically similar species such as A.
chamaeleonides. The latter species seems more
adaptable; thus, under future scenarios of climate
change and overall loss/modication of suitable
habitats it could displace A. guamuhaya.
Although both forest coverage and interference
competition have the potential to affect the
distribution of A. guamuhaya signicantly, they
cannot be modelled predictably with current
tools.
If we assume that the climatically suitable
areas generated by the ecological niche model
are a measure of the species range, then the
EOO, the number of locations, and its AOO
decrease under all future scenarios of climate
change. Given these changes, the status of A.
guamuhaya would fulll the IUCN criteria in the
category of Endangered (B1ab[i,ii,iii,iv], c[i];
B2ab[i,ii,iii,iv], c[i]). Nonetheless, this estimated
EOO of A. guamuhaya is just a rst approach to
the reality because of the difculty in determining
the exact area occupied by a species on a highly
irregular topography such as that of the
Guamuhaya Massif. Moreover, part of this EOO
is occupied by a closely related and ecologically
similar species (A. chamaeleonides) that
apparently can replace A. guamuhaya and further
reduce its effective EOO (Figure 7; see below).
However, the exact area occupied by A.
chamaeleonides and the degree to which it
reduces the effective EOO of A. guamuhaya is
difcult to estimate.
Anolis guamuhaya remains in the Endangered
category, as previously evaluated by other
authors following the IUCN Criterion D
(Rodríguez 1999, 2012). However, we consider
that IUCN Criterion B is more plausible for
twig-giant anoles because it allows a more
realistic assessment. The ecological niche models
that we obtained for A. guamuhaya predict that
under future scenarios of climate change, the
environmentally suitable area for this species
reaches a critical value under scenario RCP 8.5
W/m2 for 2070, in which only 77 km2 of the
Guamuhaya Mountains would remain compatible
with the species requirements. Moreover, it is
reasonable to associate this reduction in
climatically suitable habitat with a severe decline
in natural forest coverage. For this reason, we
consider that A. guamuhaya may fulll the IUCN
criteria for the Critically Endangered category in
2070. This situation would be exacerbated if, in
fact, the species currently does not occupy all its
potential distribution because of other factors
apart from climatic suitability, such as forest
coverage reduction and interference competition.
The fact that we found several specimens of
A. guamuhaya in secondary vegetation suggests
that the species can survive some degree of
habitat deterioration. However, all the places
where we found the species in secondary
vegetation were ecotonal areas at the edge of
larger patches of primary vegetation
(semideciduous forest or mountain rainforest)
having different degrees of canopy connectivity.
These individuals inhabiting ecotonal areas may
be incipient dispersers, with their presence not
necessarily evidencing ecological success in
deteriorated habitats. Indeed, all individuals we
found in ecotonal secondary vegetation were
juveniles; these lizards may be density-dependent
emigrants that are dispersing in response to
inbreeding and/or kin competition (Begon et al.
2006). In fact, we never found adult A.
guamuhaya in isolated patches of secondary
vegetation; this would seem to suggest that the
species is unable to survive after a relatively low
level of habitat deterioration.
More extensive, long-term studies on the
demographic parameters and ecological
requirements of A. guamuhaya are needed to
assess its conservation status accurately. This is
Rodríguez-Cabrera et al.

29
Phyllomedusa - 19(1), June 2020
particularly important because more than a half
(53%) of all records of the species are outside
the network of protected areas in Cuba. This
consideration should be a matter of utmost
concern when environmental authorities
formulate action plans and conservation
strategies.
Acknowledgments
We thank Julio León, Raimundo López-
Silvero, Alejandro M. Rodríguez, Paidel
Gutiérrez, and Lázaro Fernández for eld
assistance. Alexander Arango kindly allowed us
to take photographs and provided measurements
of the specimen under his care at the Parque
Zoológico Nacional, La Habana, Cuba. We also
thank Luis F. de Armas for providing useful
information regarding the exact locality of the
paratype CZACC 4.1549 collected by him in
1978 and to Orlando H. Garrido for granting us
an interview regarding the holotype of A.
guamuhaya. Manuel Iturriaga and Adonis
González provided access to the specimens
deposited in the Instituto de Ecología y
Sistemática, and Alejandro Barro and Roberto
Alonso to the specimens deposited in the Museo
de Historia Natural “Felipe Poey” of the
Universidad de La Habana. Alejandro Palmarola
provided literature references. The management
of the Cienfuegos Botanical Garden and the
meteorological radar station at Pico San Juan
(Guamuhaya) provided logistical support during
eld expeditions. The Earthwatch Institute, the
Wildlife Conservation Society (WCS), and the
Sociedad Cubana de Zoología provided nancial
and technical support during expeditions made
to the “Lomas de Banao” Ecological Reserve,
Sancti Spiritus range, particularly to Maikel
Cañizares, Lucía Hechevarría, and Natalia Rossi
(WCS) who arranged the expeditions, the many
volunteers that assisted us in the eld and the
staff from the “Lomas de Banao” Ecological
Reserve for providing us with excellent treatment
and accommodations. This work is also part of
the results of the project “Explorando los Efectos
del Cambio Climático sobre la Biota Cubana,”
Programa Nacional de Cambio Climático
(CITMA). Finally, Sheila Rodríguez, Marlon E.
Cobos and Linda Trueb made useful comments
that improved the manuscript.
References
Anderson, R. P. 2013. A framework for using niche models
to estimate impacts of climate change on species
distributions. Annals of the New York Academy of
Sciences 1297: 8–28.
Anderson, R. P. and A. Raza. 2010. The effect of the extent
of the study region on GIS models of species geographic
distributions and estimates of niche evolution: prelimi-
nary tests with montane rodents (genus Nephelomys) in
Venezuela. Journal of Biogeography 37: 1378–1393.
Barve, N., V. Barve, A. Jiménez-Valverde, A. Lira-Noriega,
S. P. Maher, A. T. Peterson, J. Soberón, and F.
Villalobos. 2011. The crucial role of the accessible area
in ecological niche modeling and species distribution
modeling. Ecological Modeling 222: 1810–1819.
Begon, M., C. R. Townsend, and J. L. Harper. 2006. Ecology.
From Individuals to Ecosystems. 4
th Edition. Malden.
Blackwell Publishing. 738 pp.
CNAP (Centro Nacional de Áreas Protegidas). 2013. Plan
del Sistema Nacional de Áreas Protegidas 2014-2020.
La Habana. Ministerio de Ciencias, Tecnología y Medio
Ambiente. 366 pp.
Del Risco, E. 1995. Los Bosques de Cuba: Su Historia y
Características. La Habana. Editorial Cientíco-
Técnica. 94 pp.
Díaz, L. M., N. Navarro, and O. H. Garrido. 1998. Nueva
especie de Chamaeleolis (Sauria: Iguanidae) de la
Meseta de Cabo Cruz, Granma, Cuba. Avicennia
8/9: 27–34.
Domínguez-González, A. Z. and E. Acosta-Rodríguez. 2012.
Características ambientales de la provincia de Sancti
Spíritus. Chapter I. Pp. 1–43 in A. Z. Domínguez
González, M. Torres Martínez, and Y. G. Puerta de
Armas (eds.), Experiencias en la Protección de la
Biodiversidad y el Desarrollo Sostenible en la Provincia
de Sancti Spíritus. La Habana. Ministerio de Ciencia,
Tecnología y Medio Ambiente.
Estrada, A. R. 1994. Herpetofauna de la Cuenca Banao-
Higuanojo, Sancti Spíritus, Cuba. Revista de la
Academia Colombiana de Ciencias 19: 353–360.
New localities and distribution models for Anolis guamuhaya in Cuba

30
Phyllomedusa - 19(1), June 2020
Estrada, R., G. Martín, P. Martínez, S. Vioel, R. Capote, I.
Reyes, S. Galano, C. Cabrera, C. Martínez, L. Mateo, Y.
Guerra, A. Batte, and L. Coya. 2011. Mapa (BD-SIG) de
vegetación natural y seminatural de Cuba v.1 sobre
Landsat etm 7 slc-off gap lled, circa 2011. Memorias
del IV Congreso de Manejo de Ecosistemas y Biodi-
versidad, ISBN 978-959-300-034-5. La Habana, Cuba.
Garrido, O. H. 1980. Los vertebrados terrestres de la
Península de Zapata. Poeyana 203: 1–49.
Garrido, O. H. 1982. Descripción de una nueva especie de
Chamaeleolis (Lacertilia: Iguanidae), con notas sobre su
comportamiento. Poeyana 236: 1–25.
Garrido, O. H. and A. Schwartz. 1968. Cuban lizards of the
genus Chamaeleolis. Quarterly Journal of the Florida
Academy of Sciences 30: 197–220.
Garrido, O. H., O. Pérez Beato, and L. V. Moreno. 1991.
Nueva especie de Chamaeleolis (Lacertilia: Iguanidae)
para Cuba. Caribbean Journal of Science 17: 162–168.
Gutiérrez-Domech, R. and M. Rivero-Glean. 1997.
Minigeografía de Cuba. La Habana. Editorial Cientíco-
Técnica. 142 pp.
Hass, C. A., S. B. Hedges, and L. R. Maxson. 1993.
Molecular insights into the relationships and biogeo-
graphy of West Indian anoline lizards. Biochemical
Systematics and Ecology 21: 97–114.
Henderson, R. W. and R. Powell. 2009. Natural History of
West Indian Amphibians and Reptiles. Gainesville.
University Press of Florida. 513 pp.
Herrel, A. and V. Holanova [Holáňová Zahradníčková].
2008. Cranial morphology and bite force in Chamaeleolis
lizards. Adaptations to molluscivory? Zoology
111: 467–475.
Hijmans, R. J., S. E. Cameron, J. L. Parra, P. G. Jones, and
A. Jarvis. 2005. Very high resolution interpolated
climate surfaces for global land areas. International
Journal of Climatology 25: 1965–1978.
Holáňová Zahradníčková, V., A. Abramjan, K. Palupčíková,
I. Rehák, and D. Frynta. 2017. Discovering an Antillean
Anolis (Squamata: Polychrotidae) with contrasting
sexual dichromatism in otherwise sexually monomorphic
“chamaeleolis” group. Acta Societatis Zoologicae
Bohemoslovenicae 81: 31–47.
IPCC (Intergovernmental Panel on Climate Change). 2014.
Climate Change 2014: Synthesis Report. Contribution
of Working Groups I, II and III to the Fifth Assessment
Report of the Intergovernmental Panel on Climate
Change. R. K Pachauri and L. A. Mayer (eds.). Geneva.
Intergovernmental Panel on Climate Change. 151 pp.
ITZN (International Trust for Zoological Nomenclature).
1999. International Code of Zoological Nomenclature.
4th Edition. London. International Trust for Zoological
Nomenclature. 106 pp.
IUCN. 2001. IUCN Red List Categories and Criteria: Version
3.1. Gland and Cambridge. IUCN Species Survival
Commission. 33 pp.
IUCN. 2012. IUCN Red List Categories and Criteria. Version
3.1. 2nd Edition. Eletronic Database accessible at http: //
www.iucnredlist.org/technical-documents/categories-
and-criteria. Captured on 15 March 2018.
IUCN Standards and Petitions Subcommittee. 2014.
Guidelines for Using the IUCN Red List Categories and
Criteria. Version 11. Eletronic Database accessible at
http: //www.iucnredlist.org/documents/RedListGuideli
nes.pdf. Captured on 15 March 2018.
Jackman, T. R., A. Larson, K. de Queiroz, and J. B. Losos.
1999. Phylogenetic relationships and tempo of early
diversication of Anolis lizards. Systematic Biology
48: 254–285.
Leal, M. and J. B. Losos. 2000. Behavior and ecology of the
Cuban “chipojos bobos” Chamaeleolis barbatus and C.
porcus. Journal of Herpetology 34: 318–322.
Liu, C., P. M. Berry, T. P. Dawson, and R. G. Pearson. 2005.
Selecting threshold of occurrence in the prediction of
species distributions. Ecography 28: 385–393.
Mancina, C. A. and D. D. Cruz Flores (eds.). 2017.
Diversidad Biológica de Cuba: Métodos de Inventario,
Monitoreo y Colecciones Biológicas. La Habana.
Editorial AMA. 502 pp.
Mancina, C. A., R. Fernández de Arcila Fernández, D. D.
Cruz Flores, M. A. Castañeira Colomé, and A. González-
Rosell. 2017. Diversidad biológica terrestre de Cuba. Pp.
8–25 in C. A. Mancina and D. D. Cruz (eds.), Diversidad
Biológica de Cuba: Métodos de Inventario, Monitoreo y
Colecciones Biológicas. La Habana. Editorial AMA.
Mateo-Rodríguez, J. and M. Acevedo-González. 1989.
Regionalización físico-geográca 5. 1: 3 000 000. Pp
XII.2.1 in G. Oliva (ed.), Nuevo Atlas Nacional de Cuba.
Academia de Ciencias de Cuba and Instituto Cubano de
Geodesia y Cartografía. Instituto de Geografía Nacional
de España.
Morin, P. J. 2011. Community Ecology. 2nd Edition. Oxford.
Blackwell Publishing. 407 pp.
Moss, R. H., J. A. Edmonds, K. A. Hibbard, M. R. Manning,
S. K. Rose, D. P. van Vuuren, T. R. Carter, S. Emori, M.
Kainuma, T. Kram, G. A. Meehl, J. F. B. Mitchell, N.
Nakicenovic, K. Riahi, S. J. Smith, R. J. Stouffer, A. M.
Thomson, J. P. Weyant, and T. J. Wilbanks. 2010. The
next generation of scenarios for climate change research
and assessment. Nature 463: 747–756.
Rodríguez-Cabrera et al.

31
Phyllomedusa - 19(1), June 2020
Muscarella, R., P. J. Galante, M. Soley, R. Boris, J. Kass, M.
Iriarte, and R. P. Anderson. 2014. ENMeval: an R
package for conducting spatially independent evaluations
and estimating optimal model complexity for Maxent
ecological niche models. Methods in Ecology and
Evolution 5: 1198–1205.
ONEI (Ocina Nacional de Estadística e Información). 2015.
Anuario Estadístico de Cuba, 2014. Eletronic Database
accessible at http: //www.one.cu. Captured on 14 Mayo
2017.
Peterson, A. T., J. Soberón, R. G. Pearson, R. P. Anderson,
E. Martínez-Meyer, M. Nakamura, and M. B. Araújo.
2011. Ecological Niches and Geographic Distributions.
Monographs in Population Biology 49. Princeton.
Princeton University Press. 314 pp.
Phillips, S. J. and M. Dudik. 2008. Modeling of species
distributions with Maxent: new extensions and a
comprehensive evaluation. Ecography 31: 161–175.
Phillips, S. J., R. P. Anderson, and R. E. Schapire. 2006.
Maximum entropy modeling of species geographic
distributions. Ecological Modeling 190: 231–259.
Poe, S. 2013. 1986 Redux: new genera of anoles
(Squamata: Dactyloidae) are unwarranted. Zootaxa
3626: 295–299.
Poe, S., A. Nieto-Montes de Oca, O. Torres-Carvajal, K. De
Queiroz, J. A. Velasco, B. Truett, L. N. Gray, M. J.
Ryan, G. Köhler, F. Ayala-Varela, and I. Latella. 2017.
A phylogenetic, biogeographic, and taxonomic study of
all extant species of Anolis (Squamata; Iguanidae).
Systematic Biology 66: 663–697.
Pyron, R. A., F. T. Burbrink, and J. J. Wiens. 2013. A
phylogeny and revised classication of Squamata,
including 4161 species of lizards and snakes. BMC
Evolutionary Biology 13: 93–146.
Radosavljevic, A. and R. P. Anderson. 2014. Making better
MAXENT models of species distributions: complexity,
overtting and evaluation. Journal of Biogeography
41: 629–643.
Rodríguez Schettino, L. 1999. Systematic accounts of the
species. Pp. 104–380 in L. Rodríguez Schettino (ed.),
The Iguanid Lizards of Cuba. Gainesville. University
Press of Florida.
Rodríguez Schettino, L. 2012. Chamaeleolis guamuhaya. Pp.
141–142 in H. González Alonso, L. Rodríguez Schettino,
C. A. Mancina, and I. Ramos-García (eds.), Libro Rojo
de los Vertebrados de Cuba. La Habana. Editorial
Academia.
Rodríguez Schettino, L., C. A. Mancina, and V. Rivalta-
González. 2013. Reptiles of Cuba: checklist and
geographic distribution. Smithsonian Herpetological
Information Service 144: 1–96.
Rodríguez Schettino, L., V. Rivalta-González, and E. Pérez-
Rodríguez. 2010. Distribución regional y altitudinal de
los reptiles de Cuba. Poeyana 498: 11–20.
Soberón, J. and M. Nakamura. 2009. Niches and distributional
areas: concepts, methods, and assumptions. Proceedings
of the National Academy of Sciences 17: 19644–19650.
Torres, J., R. Marrero, and T. M. Rodríguez-Cabrera. 2015.
Anolis guamuhaya (Escambray Bearded Anole).
Maximum Elevation. Herpetological Review 46: 86–87.
Warren, D. L. and S. N. Seifert. 2011. Ecological niche
modeling in Maxent: the importance of model
complexity and the performance of model selection
criteria. Ecological Applications 21: 335–342.
Wilson, E. O. 1957. Behavior of the Cuban lizard
Chamaeleolis chamaeleontides (Duméril and Bibron) in
captivity. Copeia 1957: 145.
Editor: Steven Poe
New localities and distribution models for Anolis guamuhaya in Cuba

32
Phyllomedusa - 19(1), June 2020
Appendix I. Material examined and update of localities with coordinates and elevations reported for Anolis guamuhaya and
A. cf. chamaeleonides used in this study, from the literature and this paper
(N = 32 records, 15 locations). An asterisk (*) represents voucher declared in this paper.
Acronyms and abbreviations declared in Materials and Methods.
Anolis guamuhaya
Type locality: the road between Jibacoa [22.01699, -79.98912; 330 m a.s.l.], Villa Clara province, and Topes de Collantes
[21.9186, -80.02060; 700 m a.s.l.], Sancti Spíritus province, including elevations of 800 m [female MNHNCu–500, holotype]
(Garrido et al. 1991).
CIENFUEGOS province: Cumanayagua municipality: — Mina Carlota (22.07171, -80.17762; 300–600 m a.s.l.) (male MCZ
R-53598, dry skeleton) (Wilson 1957, Garrido and Schwartz 1968, Garrido et al. 1991); — southwestern slope of Pico San Juan
(21.98990, -80.14939; 1,015 m a.s.l.) [juvenile male MFP 12515 (Torres et al. 2015); juvenile male, no voucher (57 mm SVL),
8.x.2013; female CZACC 4.5871* (162 mm SVL, 121 mm TL), 25.viii.2014], collected/examined by T. M. Rodríguez-Cabrera,
R. Marrero, J. Torres, P. Gutiérrez-Macías, and A. M. Rodríguez; — ca. 6 km SW of Aguacate (21.88775, -80.12651; 350 m
a.s.l.) [female CZACC 4.1549; formerly IZ-5397 (Garrido 1982) and CZACC:7499 (Garrido et al. 1991)]; — Carso de Buenos
Aires (21.98872, -80.18981; 760 m a.s.l.) [female MFP 12.633* (146 mm SVL, 123 mm TL, 39 g), 19.vii.2014; adult, no
voucher, 21.ii.2015], collected/examined by T. M. Rodríguez-Cabrera and J. León; — near the Trinitario stream waterfall
(22.00705, -80.04250; 550 m a.s.l.) (adult, no voucher), examined by E. Morell; — Santa Martina (22.01833, -80.38369; 40 m
a.s.l.) [male MFP 12.629* (165 mm SVL, 203 mm TL, 87.2 g), 27.viii.2014; juvenile male, no voucher (49 mm SVL, 48 mm
TL)], collected/examined by T. M. Rodríguez-Cabrera, J. Torres, P. Gutiérrez-Macías, A. M. Rodríguez, and L. Fernández;
— Boca Ambuila, eastward of the Gavilanes river mouth (21.96478, -80.33822; 15 m a.s.l.) [female MFP 12.603* (175 mm
SVL, 167 mm TL, 80 g); juvenile female CZACC 4.5850* (64 mm SVL, 57 mm TL, 4.5 g); juvenile female CZACC 4.5851*
(59 mm SVL, 41 mm TL, 4 g)], 24.ix.2014, collected by T. M. Rodríguez-Cabrera, R. López-Silvero, and R. Domínguez.
VILLA CLARA province: manICaragua municipality: — near the Río Negro Touristic Center (22.04567, -80.05894; 400 m
a.s.l.) (adult, no voucher), examined by E. Morell; — southeastern slope of Loma Cariblanca (22.06025, -80.01181; 400 m
a.s.l.) [male (140 mm SVL, 130 mm TL, 47.5 g), currently kept alive at the Parque Zoológico Nacional, La Habana, Cuba],
collected by E. Morell. SANCTI SPÍRITUS province: TrInIdad municipality: — Topes de Collantes (21.91194, -80.02000;
700 m a.s.l.) and vicinity, within a radius of 1 km (650–800 m a.s.l.) (USNM-156786, male NMP6V 34517, female NMP6V
71870/1, male NMP6V 71870/2, male NMP6V 71870/3, female NMP6V 71870/4, male NMP6V 71870/5, male NMP6V
71870/6, male NMP6V 71870/7, female CZACC 4.5506) (Garrido and Schwartz 1968, Garrido 1982, Rodríguez et al. 2013,
Holáňová et al. 2017) — La Chispa (21.92246, -80.03304; 800 m a.s.l.) (no voucher) (Garrido et al. 1991); — Caburní
waterfall (21.92076, -80.00363; 400 m a.s.l.) (no voucher) (Rodríguez et al. 2013); — ca. 2 km southeast of Topes de Collantes,
near Vegas Grandes village (21.90250, -80.00500; 650 m a.s.l.) (male NMP6V 71870/8*, male NMP6V 71871*); — near the
resurgence of the Jibacoa river (21.97158, -79.93372; 200 m a.s.l.) (adult, no voucher), examined by T. M. Rodríguez-Cabrera.
Anolis cf. chamaeleonides.
CIENFUEGOS province: Cumanayagua municipality: Santa Martina (22.01833, -80.38369; 40 m a.s.l.) [juvenile male
CZACC 4.5846* (52.2 mm SVL, 50 mm TL), 22.ii.2014], collected/examined by T. M. Rodríguez-Cabrera, J. Torres, P.
Gutiérrez-Macías, A. M. Rodríguez, and L. Fernández.
Rodríguez-Cabrera et al.

33
Phyllomedusa - 19(1), June 2020
Appendix II. Settings for the best ecological niche models for Anolis guamuhaya based on lowest Akaike’s information
criterion (AICc). ΔAIC represents the difference between the AICc score for each model and
the AICc score of best model; models with ΔAIC < 2 could be considered to have substantial support.
FC RM AUCtrain AUCtest ± SD AUCdiff OR MTP AICc ΔAIC
LQ 1 0.938 0.940 ± 0.005 0.025 0.227 ± 0.103 242.49 0
LQ 1.5 0.938 0.934 ± 0.009 0.030 0.227 ± 0.103 242.92 0.42
L 1 0.937 0.940 ± 0.005 0.024 0.273 ± 0.149 243.50 1.00
LQ 2 0.938 0.924 ± 0.001 0.038 0.227 ± 0.103 243.51 1.01
LQ 2.5 0.938 0.919 ± 0.002 0.042 0.227 ± 0.103 244.25 1.75
L 1.5 0.937 0.935 ± 0.0007 0.028 0.227 ± 0.103 244.30 1.80
LQHP 4.5 0.931 0.939 ± 0.02 0.021 0.227 ± 0.103 244.43 1.94
Appendix III. Estimates of relative contributions and permutation importance of the six bioclimatic
variables selected to build the ecological niche model for Anolis guamuhaya.
Variables Percent contribution Permutation importance
Max. temperature of warmest month (bio 5) 55 0
Mean diurnal range (bio 2) 18.7 39.1
Temperature seasonality (bio 4) 17.4 18.2
Precipitation seasonality (bio 15) 7.3 41.9
Precipitation of driest quarter (bio 17) 1.5 0
Precipitation of the wettest month (bio 13) 0.1 0.9
New localities and distribution models for Anolis guamuhaya in Cuba