Content uploaded by Jiri Smid
Author content
All content in this area was uploaded by Jiri Smid on Feb 08, 2023
Content may be subject to copyright.
Vol.:(0123456789)
1 3
https://doi.org/10.1007/s13127-022-00572-w
ORIGINAL ARTICLE
Diversification ofHemidactylus geckos (Squamata: Gekkonidae)
incoastal plains andislands ofsouthwestern Arabia withdescriptions
andcomplete mitochondrial genomes oftwo endemic species
toSaudi Arabia
JiříŠmíd1,2 · MarekUvizl1,2· MohammedShobrak9· SalemBusais3,10· AlFaqihAliSalim9·
RaedHamoudM.AlGethami9· AbdulazizRaqiAlGethami9· AbdulkarimSalehK.Alanazi4· SaadDasmanAlsubaie4·
MichailRovatsos5· LucieNováková1,2,6· TomášMazuch7· SalvadorCarranza8
Received: 22 December 2021 / Accepted: 13 July 2022
© Gesellschaft für Biologische Systematik 2022
Abstract
The systematic, phylogenetic, and biogeographic aspects of the rich squamate fauna of the Arabian Peninsula are becoming
increasingly well understood. The Arabian members of the gecko genus Hemidactylus, the most diverse genus among Arabian
squamates, have been the subject of several phylogenetic revisions in recent years. However, large parts of the peninsula
lacked thorough sampling, for example, the coastal hyper-arid plains along the Red Sea and some offshore islands. In this
study, we examine the genetic, morphological, and ecological diversification of a Hemidactylus clade that straddles the Red
Sea and contains ten Arabian and three African species. We compiled a genetic dataset of seven markers (two mitochon-
drial and five nuclear) to reconstruct their phylogenetic relationships and collected morphological data to assess the level
of interspecific morphological disparification. Our results support the existence of four yet undescribed species within the
clade – two from Arabia and two from Ethiopia. We provide taxonomic descriptions of the two new Arabian species, one from
the western Asir Mountains foothills and one from the Farasan Islands. The new species from the Asir Mountains foothills
highlights the role of the southern Arabian coastal desert as an important yet often overlooked local biodiversity hotspot. The
new species from the Farasan Islands represents the second vertebrate species endemic to the archipelago. Together with the
descriptions of the diagnostic features of both species, we provide complete annotated mitochondrial genomes of both holo-
types and of holotypes of two other species from the clade to characterize their mitogenomic composition and architecture.
Keywords Afro-Arabia· Genomics· Lizards· Mitogenome· Reptiles· Squamata
* Jiří Šmíd
jirismd@gmail.com
1 Department ofZoology, National Museum, Cirkusová 1740,
Prague, CzechRepublic
2 Department ofZoology, Faculty ofScience, Charles
University, Viničná 7, Prague, CzechRepublic
3 Department ofBiology, Faculty ofEducation, Aden
University, Aden, Yemen
4 Biodiversity Department, National Centre forWildlife,
Riyadh, SaudiArabia
5 Department ofEcology, Faculty ofScience, Charles
University, Viničná 7, Prague, CzechRepublic
6 Department ofMigration, Max Planck Institute ofAnimal
Behavior, Am Obstberg 1, 78315Radolfzell, Germany
7 Department ofForest Ecology, Mendel University,
61300Brno, CzechRepublic
8 Institute ofEvolutionary Biology (CSIC-Universitat
Pompeu Fabra), Passeig Marítim de la Barceloneta, 37-49,
08003Barcelona, Spain
9 Prince Saud Al Faisal Wildlife Research, National Center
forWildlife, P. O Box1086, Taif21944Taif, SaudiArabia
10 Department ofBiology, Faculty ofScience, Jazan University,
Jazan, SaudiArabia
Organisms Diversity & Evolution (2023) 23:185–207
/ Published online: 17 August 2022
1 3
Introduction
The Arabian Peninsula shows a high degree of biodiversity
endemism owing to its geographical and long geological
isolation from other landmasses. The seas that surround
the peninsula from the west, south, and east form natural
barriers to species dispersal. The only contemporary land
connection is in the north with mainland Asia. However,
the north Arabian deserts are a very effective barrier in
their own right (Šmíd etal., 2021a), thus making the pen-
insula a nearly completely isolated evolutionary labora-
tory. As a result, the Arabian biota has, for the most part,
been evolving insitu with occasional colonization events
from other continents (Tejero-Cicuéndez etal., 2021).
Quite naturally, not all parts of Arabia have received
equal attention of scientists. The mountains on the perim-
eter of the peninsula that host the highest diversity of spe-
cies have been fairly thoroughly studied, which contrasts
with the paucity of data available for other regions. This
concerns the interior Arabian deserts but also coastal
plains and some offshore islands.
The Tihama Desert is a coastal plain at a maximum
of 60km-wide and wedged between the Red Sea and the
sharp escarpment of the Asir Mountains of southern Saudi
Arabia and Yemen (Supplementary Fig.S1). It is subdi-
vided into a northern and southern part, which are sepa-
rated by massive lava outflows from the Asir Mountains at
around 18° latitude by the city of Al Qahma. The Tihama
is one of the least hospitable places on earth, with annual
precipitation of only 50–100mm and average annual tem-
peratures exceeding 30 (Edgell, 2006). At the same
time, however, it is one of the most diverse regions of Ara-
bia, mostly in the southern part and particularly in terms of
phylogenetic diversity and endemism (Šmíd etal., 2021a).
Among squamate reptiles, several narrow-ranging endemic
species occur in the Tihama (Arnold, 1986; Gasperetti,
1988), many of which are known only from a handful of
specimens, suggesting the need for further explorations.
The Farasan Islands are a Red Sea archipelago composed
of two large and many smaller islands located about 40km
off the city of Jazan in the southwestern corner of Saudi
Arabia (Supplementary Fig.S1). The islands vary in size,
from the largest Farasan Al-Kebir with over 350 km2 to
tiny islets of a few square meters. Contrary to the Tihama,
the Farasan Islands show a very low level of endemism,
mostly due to the fact that they were connected to main-
land Arabia as recently as in the late Pleistocene during
the last glacial maximum (Sakellariou etal., 2019), which
enabled faunal interchange. Nonetheless, two vertebrate
endemics have been described from the islands: the enig-
matic but sometimes disputed Farasan gazelle (Gazella
gazella farasani Thouless & Al Bassri, 1991; Bärmann
etal., 2013) and the terrestrial colubrid snake Platyceps
insulanus (Mertens, 1965), whose type specimen was
found swimming in the sea near Sarso Island.
The most outstanding example, to our knowledge, of spe-
cies radiation of the Arabian biota is that of Hemidactylus
Goldfuss, 1820 geckos. With its currently recognized 187
species (Uetz etal., 2022), Hemidactylus is the second-
most species-rich gecko genus globally. It is phylogeneti-
cally partitioned into four mostly allopatric clades (Carranza
& Arnold, 2006), with species from Arabia and northern
Africa belonging to the so-called arid clade. The arid clade
is formed by three major radiations: the African, Socotran,
and Arabian (Garcia-Porta etal., 2016; Gómez-Díaz etal.,
2012; Šmíd etal., 2013a). It has been the subject of many
recent systematic and taxonomic revisions and updates that
resulted in recognition of 27 new species during the last
decade (Busais & Joger, 2011; Carranza & Arnold, 2012;
Moravec etal., 2011; Safaei-Mahroo etal., 2017; Šmíd etal.,
2013b, 2015, 2017, 2020; Torki etal., 2011; Vasconcelos
& Carranza, 2014). Nearly half of the diversity of the arid
clade occurs in Arabia. In contrast to the high number of
species in the clade, its diversification is relatively recent.
The crown split of the Arabian radiation has been estimated
to have taken place between 12 and 15 million years ago
(Gómez-Díaz etal., 2012; Šmíd etal., 2013a), when the
peninsula had already been isolated from Africa for about
10 million years and just after the Socotra Archipelago split
off from Arabia and started drifting southwards.
In this study, we analyzed new material of Hemidactylus
geckos obtained during two field trips to southwestern Saudi
Arabia in 2019. We conducted a multilocus genetic analysis
of two mitochondrial and five nuclear markers and examined
morphological traits to compare the new specimens with
previously available data. In particular, we focused on the
lowland coastal and insular populations that had been over-
looked in previous studies and had already been recognized
to bear some unique morphological features, suggesting the
possibility of dealing with new species.
Materials andmethods
Specimen collection
Field work was conducted as part of the project “Systemat-
ics and biodiversity of the reptiles of southwestern Saudi
Arabia,” supported by the Saudi Wildlife Authority (SWA).
Two field trips were carried out in March and June 2019.
A total of 81 Hemidactylus tissue samples and voucher
specimens were collected for this study (Supplementary
TableS1). Genetic and morphological data for additional
specimens and species not sampled in the field were taken
J.Šmíd et al.
186
1 3
from previously published works (Šmíd etal., 2013b, 2015,
2017; Vasconcelos & Carranza, 2014). We also included
four tissue samples of specimens from Ethiopia putatively
closely related to the African member of this clade, H.
awashensis (Šmíd etal., 2015).
The material used in this study is deposited in the fol-
lowing collections: NHMUK (formerly BMNH; Natural
History Museum, London, UK), IBE (Institute of Evolu-
tionary Biology, Barcelona, Spain), NHM-BS (Staatliches
Naturhistorisches Museum Braunschweig, Germany), NMP
(National Museum Prague, Czech Republic), SMF (Senck-
enberg Forschungsinstitut und Naturmuseum, Frankfurt,
Germany), TMHC (Tomas Mazuch private herpetological
collection, Dritec, Czech Republic), TUZC (Taif University
Zoological Collection, Saudi Arabia), and ZFMK (Zoolo-
gisches Forschungsmuseum Alexander Koenig, Bonn, Ger-
many). The newly collected material was deposited in the
NMP collection.
Data forgenetic analysis
We restricted our study to members of the so-called H. saba
and H. robustus groups (Šmíd etal., 2013b, 2015), to which
the putative species from southwestern Arabia were assumed
to belong. Other Arabian Hemidactylus species that did not
belong to these two species groups were not included in the
phylogenetic analyses.
The dataset for the genetic analyses contained a total of
134 samples of the following 13 ingroup species: H. aden-
sisŠmíd etal., 2015 (5 samples), H. alfarraji Šmíd etal.,
2017 (9 samples), H. asirensis Šmíd etal., 2017 (18 samples),
H. awashensis (11 samples), H. granosus Heyden, 1827 (17
samples), H. mandebensis Šmíd etal., 2015 (3 samples), H.
robustus Heyden, 1827 (26 samples), H. saba Busais & Joger,
2011 (3 samples), H. ulii Šmíd etal., 2013a, b (8 samples),
the twoputative species from Saudi Arabia described herein
(15 samples each), and twoputative species from Ethiopia (2
samples each). Samples of H. angulatus Hallowell, 1854, H.
flaviviridis Rüppell, 1835, and H. ruspolii Boulenger, 1896,
that are not part of the Hemidactylus arid clade (Carranza &
Arnold, 2006) were used to root the trees. Figure1 shows the
distribution of the sampling localities.
DNA extraction, sequencing, andalignments
Genomic DNA was extracted from ethanol-preserved tissue
samples using the Geneaid extraction kit. We PCR-amplified
seven genetic markers: 12S rRNA (12S, 391–397 base pairs
[bp]) and cytochrome b (cytb, 307bp) from the mitochon-
drial DNA, and the oocyte maturation factor MOS (cmos,
403bp), melanocortin 1 receptor (mc1r, 668bp), recombina-
tion activating genes 1 and 2 (rag1, 280bp; rag2, 410bp),
and phosducin (pdc; 394bp) from the nuclear DNA. Primers
and PCR conditions for 12S, cytb, cmos, mc1r, rag1, and
rag2 followed those in Šmíd etal. (2013a), pdc was ampli-
fied with the forward primer PHOF2 (5’-GAT GAG CAT
GCA GGA GTA TGAA-3’) and reverse primer PHOR1 (5’-
TCC ACA TCC ACA GCA AAA AAC TCC T-3’; Bauer etal.,
2007), with the annealing temperature of 55 . Both strands
of the PCR products were sequenced at Macrogen (the Neth-
erlands). All samples of the putative species and at least
several samples of the described species were sequenced
for all the markers. Some samples of densely sampled spe-
cies such as H. asirensis, H. granosus, and H. robustus
were sequenced for the 12S gene only, which was used as a
genetic barcode to confirm their species assignment. Alto-
gether, we generated 330 new DNA sequences (Supplemen-
tary TableS1).
Raw sequences were edited and contigs assembled in
Geneious v.11 (Kearse etal., 2012). Sequences of each
gene were aligned with MAFFT (Katoh etal., 2019) using
the auto strategy for all markers except the 12S, for which
the Q-INS-i strategy that considers the secondary structure
of the RNA was applied. The 12S alignment was treated
with Gblocks (Castresana, 2000) to trim poorly aligned
regions with gaps. This shortened the original 430-bp-long
alignment to the length of 381bp. Sequences of cytb were
translated to amino acids, and no stop codons were found,
suggesting that no nuclear mitochondrial pseudogenes were
amplified. Heterozygous positions in the nuclear markers
were identified using the Heterozygote Plugin in Geneious
and confirmed based on the presence of two approximately
equal peaks at a single nucleotide site, and they were coded
using the IUPAC ambiguity codes. The concatenated data
set of all genes contained 2841bp with a 70% complete gene
sampling. After excluding the 12S-barcoded samples, the
gene sampling increased to 91%.
Phylogenetic analyses
We reconstructed the phylogenetic relationships with the
complete data set using maximum likelihood (ML) and
Bayesian analyses. The ML analysis was performed in IQ-
TREE v. 1.6 (Nguyen etal., 2015) through the web interface
(Trifinopoulos etal., 2016). The dataset was partitioned by
gene with the model selected automatically for each par-
tition. Branch support was assessed with the Shimodaira-
Hasegawa-like approximate likelihood ratio test (SH-aLRT;
Guindon etal., 2010) and the ultrafast bootstrap (UFBoot;
Hoang etal., 2018), both with 1,000 replicates, and the stand-
ard bootstrap (Felsenstein, 1985) with 100 replicates.
The Bayesian phylogenetic analysis was set up in BEAST
2.5 (Bouckaert etal., 2014). The dataset was partitioned by
gene, with site and clock models estimated independently
for each partition. Heterozygous positions were included in
the analysis. The best nucleotide substitution model for each
Diversification ofHemidactylus geckos (Squamata: Gekkonidae) incoastal plains andislands… 187
1 3
Fig. 1 Phylogenetic tree resulting from the ML analysis of two mito-
chondrial and five nuclear markers concatenated. Branch support is
given by each node in the following order: SH-aLRT, UFBoot, stand-
ard bootstrap, and posterior probabilities from the Bayesian analysis.
The lengths of the branches leading to the outgroup have been trun-
cated. The maps on the right show sampling localities for each spe-
cies. In most species, the sampling localities cover the entire known
range of the species. Only in H. robustus and H. granosus, the points
show the species’ general distribution and localities sampled for the
genetic analysis of this study are marked with dark centers. Localities
of the two new species are highlighted by dashed circles
J.Šmíd et al.
188
1 3
partition was determined using PartitionFinder 2.1 (Lanfear
etal., 2017), with the branch lengths linked and the best
model selected by the BIC criterion. The best models were
as follows: GTR + I + G for 12S, GTR + G for cytb, K80 + I
for cmos, HKY + I + G for mc1r, K80 for rag1, HKY + I for
rag2, SYM + I for pdc. We used the HKY model instead
of K80 and the GTR model instead of SYM as the clos-
est alternatives available in BEAST. For the partitions that
had both the proportion of invariable sites (+ I) and rate
categories (+ G) parameters suggested by PartitionFinder,
we did not estimate the + I parameter as it was substituted
by the exponentially distributed + G. In the cases where
the + G parameter was not included, the + I parameter was
estimated with a flat prior distribution with an equal prob-
ability between 0 and 1. The transition/transversion param-
eters of the HKY models were estimated using lognormal
distributions (mean = 1.0, sd = 1.25). Priors on the GTR
substitution rates were uniform (min = 0, max = 100). We
applied strict clock models with lognormally distributed
priors (mean = 1.0, sd = 1.25) to all partitions as we did not
assume deviations from clock-likeness at this phylogenetic
scale. We used the coalescent constant population model
tree prior. The analysis ran through the CIPRES Science
Gateway (Miller etal., 2010) in three independent runs, each
for 50 million generations with parameters and trees sampled
every 5,000 generations, which produced 10,000 posterior
samples in each run. We used Tracer v. 1.7 (Rambaut etal.,
2018) to ensure that convergence and stationarity of the runs
had been reached and that all parameters had high effective
sample sizes (> 200). After that, 10% of trees sampled in the
posterior were discarded as burnin, tree files were combined
using LogCombiner, and a maximum clade credibility tree
with mean branch heights was generated with TreeAnnotator
(both programs are part of the BEAST package). Nodes in
the ML and Bayesian phylogenies that had SH-aLRT ≥ 80,
UFBoot ≥ 95, standard bootstrap ≥ 70, and Bayesian poste-
rior probability ≥ 0.95 were considered strongly supported.
We further inferred a phylogenetic network using the
neighbor-net algorithm (Bryant & Moulton, 2004) imple-
mented in SplitsTree v. 4 (Huson & Bryant, 2006). The
nuclear markers were phased for this analysis. We used
SeqPHASE (Flot, 2010) to prepare the input files and
PHASE v. 2.1 (Stephens etal., 2001) to reconstruct the
haplotypes. The phase probability threshold was set to 0.7
(Harrigan etal., 2008). The network support was assessed
with 1,000 bootstrap replicates.
In addition to the concatenation-based analyses described
above, we also inferred a coalescent-based species tree of the
described and putative Hemidactylus species. We used the
StarBeast template of BEAST 2.5. The dataset was pruned
for the species tree analysis – Samples barcoded for the 12S
gene only (37 in total) and outgroup taxa were excluded.
The nuclear loci were phased as described above. Similar
to many other species tree methods, StarBeast requires a
priori assignment of samples to evolutionary units (“spe-
cies”). Samples of the described species were assigned based
on this information; samples of the putative species were
assigned based on the results of the analyses of the con-
catenated dataset (see below). We used the reversible-jump
substitution model (RB; Bouckaert etal., 2013) with four
gamma-distributed rate categories (α = 0.2, β = 5.0) and the
shape parameter estimated. The RB model allows sampling
of a mixture of models during the analysis and thus does not
depend on one model to be specified prior to the analysis.
Ploidy of the mitochondrial genes was set to haploid. A uni-
form prior (min = 0, max = 100) was used for the population
mean parameter. Other settings were similar to the Bayesian
analysis in BEAST described above. Three runs, each of 100
million generations with sampling every 5,000 generations,
produced output files with 20,000 posterior trees and param-
eters. The posterior tree files were processed as described
above for the Bayesian analysis.
We reconstructed allele networks separately for each
nuclear locus to visualize the relationships between the
Hemidactylus species and the degree of its reticulation at
the level of individual nuclear genes. We used the phased
alignments and the TCS algorithm (Clement etal., 2002)
implemented in PopART (Leigh & Bryant, 2015) to infer
the networks.
Patristic distances (p-distances) were calculated with
MEGA X v. 10.2 (Kumar etal., 2018) with the pairwise
deletion option applied.
Data formorphological analysis
We collected new morphological data for 69 specimens of
H. asirensis (19 specimens), H. granosus (8 specimens),
H. robustus (12 specimens), the two putative species from
Saudi Arabia described herein (12 and 14 specimens), and
the two putative species from Ethiopia (2 specimens each).
Juvenile specimens that were included in the morphological
comparisons were used only for analyses of meristic traits.
These data were complemented by measurements that were
published in earlier taxonomic studies on the genus (Šmíd
etal., 2013b, 2015, 2017; Vasconcelos & Carranza, 2014)
and taken by the same person (JŠ; except data taken from
Vasconcelos & Carranza, 2014). For most specimens, we
took high-resolution pictures, either in the field or after
preservation, which are publicly available for anyone to
download at MorphoBank (https:// morph obank. org; Pro-
ject number 4024; a total of 663 photographs). MorphoBank
accession numbers for each specimen are provided in Sup-
plementary TablesS1 and S2.
We scored the following metric and meristic characters:
snout–vent length (SVL; from the tip of snout to vent); tail
length (TL; from vent to tip of original tail); head length
Diversification ofHemidactylus geckos (Squamata: Gekkonidae) incoastal plains andislands… 189
1 3
(HL; from the tip of snout to retroarticular process of the
jaw); head width (HW; at the widest part of the head); head
depth (HD; maximum depth of head); horizontal eye diam-
eter (EYE); axilla-groin distance (AG; distance from the
insertion of the front limb to the insertion of the hind limb);
number of supralabials (SUP) and infralabials (INF); contact
of uppermost nasals (NASCON; none/point/broad); number
of infralabials in contact with anterior postmentals (MEN-
INF); contact of anterior postmentals (MENCON; none/
point/broad); number of longitudinal rows of enlarged dorsal
tubercles (TUBER); number of precloacal pores in males
(PORES); and number of lamellae under the first (TOE1)
and fourth toe (TOE4). We used a digital caliper and dissect-
ing microscope to take the measurements and scale counts,
respectively. Original metric and meristic data are given in
Supplementary TableS2.
Morphological analysis
We tested for the presence of sexual dimorphism in the spe-
cies for which the sample size was sufficient (all except H.
mandebensis and H. saba, which are known only from three
and four specimens, respectively [Busais & Joger, 2011;
Šmíd etal., 2015], and the two putative species from Ethio-
pia, for which we only had two specimens of each). We used
the two-sided t-test with Bonferroni correction for multi-
ple comparisons (adjusted p value = 0.000568) and tested
the non-categorical variables. No sexual dimorphism was
detected in any of the traits for any of the species (Supple-
mentary TableS3). As a result, we pooled both sexes for all
subsequent morphological analyses and also present most
summary statistics for both sexes together.
To examine whether there are metric and/or meristic
differences between the species, we first tested if the vari-
ables were normally distributed, for which we used the
D’Agostino normality test implemented in the “fBasics” R
package (Wuertz etal., 2017). Most variables deviated from
normality (results not shown), and we therefore tested the
significance of between-species differences using the non-
parametric Kruskal–Wallis test. We compared the two puta-
tive Saudi species with the species they are most closely
related to (see the phylogenetic results below). Pairwise
differences were thus tested between the putative species
from the Farasan Islands and H. adensis, H. mandebensis,
and H. robustus, and between the putative species from the
Asir Mountains foothills and H. alfarraji, H. asirensis, H.
granosus, H. saba, and H. ulii.
Mitogenome sequencing, assembly, andannotation
We used DNA extracted from muscle tissue to generate
mitochondrial genome sequences of the holotypes of the
two new species described herein and of holotypes of two
other species from this clade, H. mandebensis (voucher
NMP 74836/2) and H. ulii (voucher NMP 74833/2). The
concentration of the DNA was measured using a Qubit fluo-
rometer (Thermo Fisher Scientific). Genomic libraries with
an insert size of 350bp were prepared, and 150bp paired-
end reads were sequenced on an Illumina NovaSeq 6000
platform at Novogene Europe. We used Trimmomatic 0.39
(Bolger etal., 2014) to remove the adapters and reads of
poor quality and FastQC (Andrews, 2010) to check the qual-
ity of the remaining reads. We mapped the filtered reads on
two other available Hemidactylus mitogenomes, H. frenatus
Duméril & Bibron, 1836 (Jie etal., 2009; GenBank acces-
sion NC_012902), and H. bowringii (Gray, 1845) (Qin etal.,
2014; GenBank accession NC_025938) using the Geneious
software (Kearse etal., 2012). The reads that mapped on the
reference genomes were de novo assembled, and a consensus
sequence was generated. This consensus was then used as
a new reference on which the original reads were mapped.
This procedure was repeated until the expected mitogenome
length had been exceeded. The complete mitogenomes were
annotated using MitoAnnotator (Iwasaki etal., 2013). Anno-
tations of the transfer RNA (tRNA) regions were confirmed
with the tRNA scan-SE online tool (Chan & Lowe, 2019)
and MITOS2 (Donath etal., 2019). Repetitive elements were
inspected using Tandem Repeats Finder (Benson, 1999).
Final boundaries between the mitochondrial genes were
refined manually based on the results of the annotation algo-
rithms and comparisons with other Gekkotan mitogenomes
available in GenBank. The base composition was calculated
in MEGA X (Kumar etal., 2018). The four mitogenomes
have been deposited in GenBank with accession numbers:
OL689327-OL689330.
Results
Phylogenetic analyses
The ML and Bayesian phylogenetic analyses resulted in
largely concordant tree topologies (Fig.1, Supplemen-
tary Fig.S1). Monophyly of the SW Arabian clade was
strongly supported by all support values (SH-aLRT = 99.9/
UFBoot = 100/standard bootstrap = 100/Bayesian poster ior
probability [pp] = 1.0; support values are given in this order
hereafter). The clade is formed by two groups, the so-called
H. robustus group (Šmíd etal., 2015) that consists of H.
adensis, H. awashensis, H. mandebensis, H. robustus, the
putative species from the Farasan Islands, and the two
putative species from Ethiopia (98.5/100/100/1.0), and the
so-called H. saba group (Šmíd etal., 2013b) that contains
H. alfarraji, H. asirensis, H. granosus, H. saba, H. ulii,
and the putative species from the Asir Mountains foothills,
J.Šmíd et al.
190
1 3
although the position of H. ulii in the group was supported
only in the Bayesian analysis (73.3/87/61/1.0).
Within the H. robustus group, there were two subgroups
consistently inferred in both analyses: H. adensis and H.
mandebensis were sister species (98/100/100/1.0), and
the three African species formed a monophyletic group
(99.9/100/100/1.0) with H. awashensis being intermedi-
ately supported as sister to the two putative species from
Ethiopia (61.1/91/62/1.0). Although the topology of the ML
tree suggested a sister relationship between H. robustus and
the putative species from the Farasan Islands, it was only
supported by the SH-aLRT test (94.9), and furthermore, it
differed from the topology of the Bayesian tree in which the
Farasan species was sister to the three African species, but
again without any support (pp = 0.13). The basal relation-
ships within the H. robustus group remained unresolved and
differed between the ML and Bayesian analysis: In the ML
tree, the H. adensis and H. mandebensis pair was sister to the
rest of the group (82.7/64/49), while in the Bayesian analy-
sis, they were sister to H. robustus (pp = 0.77). But none of
these topologies was statistically supported.
The H. saba group had a fully pectinate topology.
Hemidactylus ulii was sister to all the remaining species
(94.1/95/78/1.0); H. saba branched off next (albeit with low
support: 16.8/75/47/0.6), followed by the putative species from
the Asir Mountains foothills (94.1/96/74/1.0) and H. alfar-
raji, which formed a sister group to H. asirensis + H. granosus
(97.9/100/95/1.0). The ML and Bayesian topologies of the H.
saba group were identical, although not always supported.
The phylogenetic network inferred with SplitsTree iden-
tified similar evolutionary units and a topology that was
largely consistent with the ML and Bayesian phylogenies
(Fig.2). Monophyly of all species was strongly supported
(bootstrap 98.4–100). The H. robustus and H. saba groups
were clearly differentiated in the network (bootstrap = 96.9),
and the relationships within the groups matched those in the
trees, including the unresolved basal relationships within the
H. robustus group.
The coalescent-based species tree analysis resulted in a
tree shown in Fig.3. The topology of the H. saba group
matched that of the ML and Bayesian analyses described
above. On the contrary, the topology of the H. robustus
99.9
99.9
100
99.9
96.1
98.4
100
85.2
96.9
99.9
100
100
100
100
100
100
99.5
99.8
88.5
0.005
H. saba
H. ulii
H. almakhwah
sp. n.
H. granosus
H. asirensis
H. alfarraji
H. adensis
H. mandebensis
H. awashensis
H. sp. E
H. sp. E
H. farasani sp. n.
H. robustus
outgroup
taxa
Fig. 2 Phylogenetic network resulting from the SplitsTree analysis. Species are highlighted with colors that match those in Figs.1, 3, and 7. The
branch leading to the outgroup has been truncated, which is indicated by its transparency. Bootstrap values are shown for major clades
Diversification ofHemidactylus geckos (Squamata: Gekkonidae) incoastal plains andislands… 191
1 3
group differed from both the ML and Bayesian trees. The
species tree inferred the three African species (pp = 1.0) as
sister to the rest of the species within the group (pp = 1.0),
but the monophyly of the entire H. robustus group was not
well supported (pp = 0.89; for a tree with all support values,
see Supplementary Fig.S3).
The allele networks reconstructed for each nuclear marker
showed varying levels of allele sharing (Supplementary
Figs.S4, S5, and S6). A close inspection of the positions of
the two putative species from Saudi Arabia shows that all their
alleles are private in the mc1r and rag1 networks, meaning
that there is no overlap with other species in these markers.
The putative species from the Asir Mountains foothills does
not share any allele with any other species in the cmos gene,
but the Farasan species overlaps in this marker with H. aden-
sis and H. robustus. The pdc network shows an intermediate
overlap of alleles between the Asir Mountains foothills spe-
cies and H. alfarraji, and between the Farasan species and the
two putative species from Ethiopia. The rag2 network indi-
cates a very low level of genetic differentiation in this marker
across many Hemidactylus species from both main groups (H.
adensis, H. alfarraji, H. granosus, H. mandebensis, and H.
robustus), including both putative Saudi species.
Morphological analysis
The two putative species from Saudi Arabia showed sig-
nificant differences from the species they are most closely
related to in most of the characters tested. For example,
the putative species from the Asir Mountains foothills was
considerably smaller and had smaller body proportions
than most species of the H. saba group, but in contrast to
the other species of the group, it had a relatively longer
head when compared to its SVL (Fig.3). The putative
Farasan species was the largest of the species of the H.
robustus group with largest body proportions. Details on
the between-species differences are provided in the Com-
parison sections below. Pairwise comparisons of the spe-
cies and their statistical significance are shown in Sup-
plementary Fig.S7.
Mitogenome sequencing
The mitochondrial genomes of H. mandebensis, H. ulii,
and the two new species described herein range in length
from 16,815 to 16,932bp. All four mitogenomes contain 13
protein-coding genes, two ribosomal RNA genes, 22 tRNA
genes, and the control region and conform to the typical
vertebrate gene arrangement (Fig.4). Base composition was
mostly similar across the four species: (H. mandebensis – A:
31.4%, C: 33.1%, G: 14.9%, T: 20.5%; H. ulii – A: 32.0%, C:
31.9%, G: 14.9%, T: 21.1%; the new species from the Far-
asan Islands – A: 32.1%, C: 32.2%, G: 14.6%, T: 21.0%; the
new species from the Asir Mountains foothills – A: 32.5%,
C: 31.9%, G: 14.2%, T: 21.5%).
Body size (mm) Head to body ra th
Lamellae under 4 toeEleva
30 40 50 60 0.24 0.26 0.28 0.30 8 9 10 11 12 0 500 1000 1500 2000
p< 0.05 p< 0.05 p< 0.05
H. mandebensis
H. adensis
H. robustus
H. farasani sp. n.
H. sp. E
H. sp. E
H. awashensis
H. ulii
H. saba
H. almakhwah sp. n.
H. alfarraji
H. granosus
H. asirensis
0.01
> 0.95
0.9-0.95
Branch posterior probability
Fig. 3 Species tree of the southwestern Arabian Hemidactylus clade.
Posterior probability values are indicated only for branches with
pp ≥ 0.9. For a tree with all posterior probability values, see Sup-
plementary Fig.S3. Boxplots to the right of the tree show some key
morphological and ecological characteristics for the species. Body
size is SVL of adult specimens; the head-to-body ratio was calculated
as HL/SVL (adults only); elevation indicates the range of elevations
inhabited by each species (data from Šmíd etal., 2021a). Red vertical
bars in the plots of the morphological traits show significant differ-
ences between species pairs in that trait as assessed by means of the
Kruskal–Wallis test. The two new species were compared with the
other species of their groups, and the African species (H. awashensis,
H. sp. Etio2, H. sp. Etio3) were excluded from the comparisons
J.Šmíd et al.
192
1 3
Taxonomic implications
The combined genetic, morphological, and geographical
evidence indicates that the two putative species from Saudi
Arabia, one from the Farasan Islands and one from the Asir
Mountains foothills, represent phylogenetically distinct line-
ages. Their phylogenetic positions and unique combinations
of morphological characters indicate that they represent two
distinct species that we formally describe below.
The two putative species from Ethiopia that are closely
related to H. awashensis seem genetically well differentiated,
at least at the mitochondrial level. However, they are not the
focus of this study, and their sample sizes are too small to per-
mit drawing any formal taxonomic conclusions at this point.
Hemidactylus farasani sp. n.
NMP 76104/2
16,815 bp
Hemidactylus mandebensis
NMP 74836/2
16,830 bp
Hemidactylus ulii
NMP 74833/2
16,901 bp
Hemidactylus almakhwah sp. n.
NMP 76093/2
16,932 bp
Fig. 4 Maps of the complete mitochondrial genomes of the holotypes
of H. almakhwah sp. n., H. farasani sp. n., H. mandebensis, and H.
ulii. Protein-coding genes are denoted with yellow and green anno-
tations, rRNA genes with red annotations, tRNA genes with pink
annotations, and the control region with orange annotations. Voucher
numbers are given under each species name along with the complete
mitogenome length. The specimens depicted are the holotypes them-
selves; they are not to scale. There is no photograph of H. ulii in life
to date
Diversification ofHemidactylus geckos (Squamata: Gekkonidae) incoastal plains andislands… 193
1 3
a
b
c
dg
f
e
J.Šmíd et al.
194
1 3
Hemidactylus almakhwah sp. n.
Proposed English name: Al Makhwah gecko.
Proposed Arabic name: Burs Al Makhwah, ةومـ .
Chresonymy. Hemidactylus sp. 13 in Carranza etal. (2018);
Vasconcelos and Carranza (2014)
Holotype (Fig.5). NMP 76093/2 (sample code CN15140),
adult male; Saudi Arabia, Makkah Province, dry wadi SW
of Al Ju’aydah (19.657°N, 41.567°E, datum WGS84 here-
after, 548m above sea level [asl]), June 24, 2019; collected
by J. Šmíd, S. Carranza, M. Shobrak, S. Busais, A. F. Salim,
R. H. M AlGethami, A. R. AlGethami, A. S. K. Alanazi,
and S. D. A. Alsubaie. GenBank accession numbers of
the genes used in the phylogenetic analyses and Morpho-
Bank accessions are detailed in Supplementary TableS1.
The complete mitochondrial genome accession number:
OL689327.
Paratypes (Fig.6). Adult males (NMP 76093/1, NMP
76093/6, sample codes CN15129, CN15166), adult female
(NMP 76093/4, sample code CN15168), subadults (NMP
76093/3, NMP 76093/5, sample codes CN15167, CN15169),
same collection data as the holotype. Adult male (NMP
76092/1, sample code CN15719), Saudi Arabia, Makkah
Province, Al Ju’aydah (19.657°N, 41.579°E, 482m asl),
March 30, 2019. Adult males (NMP 76091/2, NMP 76091/3,
sample codes CN15709, CN15710), adult females (NMP
76091/1, NMP 76091/4, sample codes CN15519, CN15711),
Saudi Arabia, Al Bahah Province, Al Makhwah (19.810°N,
41.442°E, 459m asl), March 30, 2019. All paratypes have
the same collectors as the holotype.
Other specimens examined Juvenile NMP 76092/2 (sam-
ple code CN15720) and unvouchered sample CN15693,
Saudi Arabia, Makkah Province, Al Ju’aydah (19.657°N,
41.579°E, 482m asl). Unvouchered samples CN15130 and
CN15131 from the type locality. Juvenile BMNH1992.168,
Saudi Arabia, Makkah Province, Khiyat, 27km E of Qun-
fudhah (ca. 19.240°N, 41.270°E, 63m asl). The unvouchered
samples were used for genetic analyses only. The juvenile
NMP 76092/2 was used for genetic analyses and for analy-
ses of meristic traits, and the BMNH specimen for analyses
of meristic traits only. Data for the BMNH specimen were
obtained from Vasconcelos and Carranza (2014).
Diagnosis A species of the Arabian radiation of the genus
Hemidactylus (Šmíd etal., 2013a, 2020), a member of the
H. saba species group sensu Šmíd etal. (2013b) charac-
terized by the following combination of morphological
traits: (1) enlarged, granular tubercles present dorsolater-
ally, but absent from mid-dorsum (only in large males they
are also present in mid-dorsum, but in these cases, they are
always much smaller and less conspicuous than the dorso-
lateral ones); (2) small size with maximum recorded SVL
43.5mm (mean 38.6 ± 3.2mm standard deviation; range
36.2–43.5mm in males, 34.1–42.9 mm in females); (3)
narrow and flat head (mean HW 7.1 ± 1.1mm, mean HD
3.7 ± 0.6mm) with pointy snout; (4) head relatively long to
the body size (mean HL 28 ± 1.6% of SVL); (5) tail length
being 117–128% of SVL; (6) anterior postmentals in broad
medial contact; (7) anterior postmentals in contact with the
first and second infralabials, less frequently (and in such
cases always unilaterally) with the first infralabial alone; (8)
9–11 supralabials; (9) 7–9 infralabials; (10) 5–6 lamellae
under the first toe, and 9–10 lamellae under the fourth toe;
(11) four precloacal pores in males; (12) tail with whorls
of enlarged tubercles; (13) enlarged subcaudals; (14) in
life pinkish to yellow–brown dorsally with dark markings
either in the form of isolated dark spots or faint transverse
or X-shaped marks. At least some of the granular dorsolat-
eral tubercles are whitish and stand out from the otherwise
darker tone of the body. There is a conspicuous narrow dark-
brown stripe running from the nostril across the eye and
above the ear to the temporal area and on the sides of the
neck, sometimes forming a continuous line from the nostril
to the forelimb insertion. Tail with broad alternating black
and white bands (when original). Body pinkish ventrally.
Comparisons. Hemidactylus almakhwah sp. n. may be most
readily differentiated from other species of the H. saba group
and from other Hemidactylus species that occur in the region
by the absent or minuscule mid-dorsal tubercles (Figs.5 and
6). All other Hemidactylus species in the region have well-
developed dorsal tubercles, which are uniform in size across
the body and often large and keeled. This character also
differentiates it from H. robustus, the most common species
in southwestern Arabia, with which H. almakhwah sp. n.
most likely overlaps geographically (although direct sym-
patry has not been observed by us). The only Hemidactylus
species which may occur in the region and that also lacks
the dorsal tubercles is the introduced H. flaviviridis. It can,
however, be easily distinguished from H. almakhwah sp. n.
by its complete absence of any dorsal tubercles, including
the dorsolateral ones. Compared to H. robustus and the other
H. saba group members, except for H. ulii, H. almakhwah
sp. n. has a relatively longer head compared to its body size
and a lower number of lamellae under the first and fourth
toe (Figs.3, SupplementaryS7). Genetic distances between
H. almakhwah sp. n. and the other Hemidactylus species
included in this study are shown in Supplementary TableS4.
Fig. 5 Holotype of H. almakhwah sp. n. (NMP 76093/2) in lifeA;
lateral B, dorsal C, and ventral Dview of the head; detail of dorsal
scales and tubercles E; precloacal area showing four precloacal pores
F; and detail of left foot F
◂
Diversification ofHemidactylus geckos (Squamata: Gekkonidae) incoastal plains andislands… 195
1 3
Description of holotype Adult male (Fig. 5), SVL
40.5mm, head length 11.1mm, head width 8.1mm, head
depth 3.7mm, eye diameter 3.0mm, axilla-groin distance
17.1mm. Supralabials 10/10 (right/left), infralabials 8/9.
Nostril is surrounded by rostral and three nasals. The
uppermost nasals touch each other at one point. Anterior
postmentals are large, of irregular hexagonal shape, and in
broad contact. Each anterior postmental is in contact with
the first and second infralabial. Digits are slightly dilated
with 6/6 lamellae under the first toe and 9/9 lamellae under
the fourth toe. Four precloacal pores are in a continuous
line. About half of the tail is regenerated. Enlarged body
tubercles are round and slightly pointy on the dorsolateral
sides; in mid-dorsum, they are smaller, entirely flat, and do
not protrude from the body when viewed laterally. There are
whorls of enlarged, posteriorly pointed tubercles on the tail;
the whorls are separated by five rows of small granules. The
tongue has been removed for genetic analysis.
In life, light brown to yellowish dorsally with a promi-
nent dark band from the nostril across the eye to the tem-
poral region. There are three roundish spots in the scapular
area and three more irregularly shaped in front of them
on the nape. Other dark spots on the dorsum are rather
faint and form barely visible transverse bands. Some of
the enlarged dorsolateral tubercles are brightly yellow-
ish, which makes them distinct against the darker body.
a
c
b
d
e
f
Fig. 6 Paratypes of H. almakhwah sp. n. in life and the species’ type locality. A – adult male NMP 76093/6; B – subadult NMP 76093/3; C –
adult female NMP 76093/4; D – subadult NMP 76093/5; E and F – the type locality, a dry wadi SW of Al Ju'aydah (19.657°N, 41.567°E)
J.Šmíd et al.
196
1 3
The color pattern has so far not changed in the preserva-
tive. Original high-resolution photographs of the holotype
may be accessed in MorphoBank, accession numbers:
M822955–M822998.
Morphological and genetic variation There is a certain
degree of variation in the development of the mid-dorsal
tubercles. While the holotype and two other adult male
specimens (paratypes NMP 76091/2–3) have minuscule
and flat tubercles present, none of the other specimens has
any enlarged scales or mid-dorsal tubercles developed. Most
specimens (77%) have anterior postmentals in contact with
the first and second infralabials; only NMP 76093/3 and
NMP 76093/5 have the postmentals on one side in contact
only with the first infralabial. There is no variation in the
number of precloacal pores in males; all have invariably
four. Some specimens (NMP 76093/1, NMP 76091/2) lacked
the dark dorsal markings in life and were uniformly grayish
brown. Paratype NMP 76093/1 has a bifurcated tip of the
tail. Original measurements are provided in Supplementary
TableS2.
Genetically, H. almakhwah sp. n. is quite uniform, most likely
owing to the fact that the sampled localities lie within a 20-km
radius. There is little variation in the mitochondrial genes,
within species p-distances being 0.28 ± 0.23% in the 12S and
1.0 ± 0.6% in the cytb. In the more variable of the nuclear
markers (mc1r, cmos), the species shows the presence of sev-
eral alleles, which are however usually separated by no more
than two substitutions. In the more conserved loci, all the
samples share the same allele (rag2) or there are only three
(pdc) or four (rag1) alleles with most samples sharing one
central allele (Supplementary Figs.S4, S5, and S6).
Etymology The species epithet refers to the city and gover-
norate of Al Makhwah, in the vicinity of which most speci-
mens were collected. It is a noun in apposition.
Distribution, habitat, and ecology. Hemidactylus almakhwah
sp. n. is so far known from four localities of low to mid-
elevation (63–548m asl) in the northern Tihama at the Asir
Mountains foothills in southwestern Saudi Arabia (Figs.1
and 7). It is reasonable to assume the species to be more
widespread in the area; its distribution is most likely tied
to the lower elevations. The isolated record from Khiyat
(specimen BMNH1992.168) indicates that H. almakhwah
sp. n. can inhabit more arid habitats of the Tihama than
those of the foothills. The southern part of the Saudi Tihama
has been fairly well surveyed, and H. almakhwah sp. n. has
not been recorded from there (Fig.7). On the contrary, the
northern edge of the coastal desert has received consider-
ably less attention, and it is thus likely that more detailed
field work in this area would provide additional localities
for the species.
At the two localities in the vicinity of Al Ju’aydah, speci-
mens were found in a dry wadi active after dusk (between
ca. 18.30 and 22.30). The geckos were found on the rocky
slopes of the hills that rimmed the riverbed, either on the
ground hopping among the rubble or hiding in crevices.
The wadi shown in Fig.6 had sparse vegetation composed
mostly of acacia trees (Vachellia tortilis), Arabian boxthorn
(Lycium shawii), and desert rose (Adenium obesum). Air
Fig. 7 Digital elevation model of coastal southwestern Arabia show-
ing the distributions of the Hemidactylus species included in this
study. Note the sharp elevation gradient between the Tihama plain
and the Asir Mountains and the association of species to particular
elevations. Squares indicate major cities. The model was developed in
ArcScene 10 (ESRI, 2011)
Diversification ofHemidactylus geckos (Squamata: Gekkonidae) incoastal plains andislands… 197
1 3
a
b
c
dg
f
e
J.Šmíd et al.
198
1 3
temperature at the time of visit ranged between 34 and 35
and relative humidity between 45 and 50%. Quite interest-
ingly, the site at Al Makhwah was a dumping site at the edge
of the city, and the geckos were found hiding under plywood
slabs and other debris.
Conservation status The extent of occurrence of H.
almakhwah sp. n. is 609 km2, and the area of occupancy is
16 km2 according to the IUCN red-listing criteria. Such low
numbers are a result of the species being known only from
four unique localities that are at most 65km apart (Fig.7).
As mentioned above, the species is likely more widespread
in the Asir Mountains foothills and has been overlooked so
far due to the lack of field work in the area. We found the
species in natural as well as in synanthropic habitats, and
we thus assume that the growing human population does
not pose a direct threat to the species. On the other hand,
the Asir Mountains foothills and the Tihama Desert are rela-
tively small in extent compared to other Arabian ecoregions
(Šmíd etal., 2021a) and are underrepresented in conserva-
tion planning (Cox etal., 2012). Hemidactylus almakhwah
sp. n. could thus serve as a flagship species for future con-
servation strategies in the region.
Hemidactylus farasani sp. n.
Proposed English name: Farasan gecko.
Proposed Arabic name: Al Burs Al Farasani,
ــ
ي
.
Chresonymy. Hemidactylus turcicus in Cunningham (2010);
Masseti (2014); Mertens (1965); Schätti and Gasperetti
(1994)
Holotype (Fig. 8). NMP 76104/2 (sample code
CN15571), adult male; Saudi Arabia, Sajid Island of the
Farasan Archipelago (16.904°N, 41.898°E, 3m asl), March
25, 2019; collected by J. Šmíd, S. Carranza, M. Shobrak, S.
Busais, A. F. Salim, R. H. M AlGethami, A. R. AlGethami,
A. S. K. Alanazi, and S. D. A. Alsubaie. GenBank acces-
sion numbers of the genes used in the phylogenetic analyses
and MorphoBank accessions are detailed in Supplementary
TableS1. The complete mitochondrial genome accession
number: OL689328.
Paratypes (Fig. 9). Adult males (NMP 76104/4,
NMP 76104/7–8, sample codes CN15575, CN15578,
CN15580, respectively), adult females (NMP 76104/1,
NMP 76104/5, NMP 76104/6, sample codes CN15570,
CN15576, CN15577), juvenile (NMP 76104/9, sample
code CN15582), same collection data as the holotype.
Adult males (NMP 76105/2, NMP 76105/3, sample
codes CN15608, CN15609), adult female (NMP 76105/1,
sample code CN15591), Saudi Arabia, Farasan Island
(16.826°N, 41.848°E, 8m asl), March 26, 2019. Juve-
nile (NMP 76103, sample code CN15589), Saudi Arabia,
Farasan Island (16.702°N, 42.055°E, 7m asl), March
25, 2019. All paratypes have the same collectors as the
holotype.
Other specimens examined Juvenile (NMP 76104/3, sam-
ple code CN15572) from the type locality. Juvenile (NMP
76106, sample code CN15594), Saudi Arabia, Farasan
Island (16.747°N, 41.904°E, 12m asl). Unvouchered sam-
ple CN15568, Saudi Arabia, Farasan Island (16.826°N,
41.848°E, 8m asl).Juveniles (SMF SMF60196–7), Saudi
Arabia, Sarso Island of the Farasan Archipelago (ca.
16.84°N, 41.59°E, 8m asl). The unvouchered sample was
used for genetic analyses only; the two juvenile NMP spec-
imens were used for genetic analyses and for analyses of
meristic traits. The SMF specimens were compared based
on photographs but were not included in the morphological
analysis.
Diagnosis A species of the Arabian radiation of the genus
Hemidactylus (Šmíd etal., 2013a, 2020), a member of the
H. robustus species group sensu Šmíd etal. (2015) char-
acterized by the following combination of morphological
traits: (1) large, triangular, posteriorly pointed, and dis-
tinctly keeled dorsal tubercles arranged in 14 regular rows
and interspersed with small granular scales; (2) medium size
with maximum recorded SVL 55.4mm (mean 45.9 ± 5.4mm
st. dev.; range 39.8–55.4mm in males, 37.1–48.5mm in
females); (3) broad head, particularly in males (mean HW
9.2 ± 1.4mm in males, 8.4 ± 0.6 in females); (4) tail length
being 119–133% of SVL; (5) left and right anterior post-
mentals not in contact and separated by an inserted scale;
(6) anterior postmentals in contact with the first and second
infralabials, less frequently with the first infralabial alone;
(7) 8–10 supralabials; (8) 7–8 infralabials (8 only in 11% of
cases, otherwise 7); (9) 6–7 lamellae under the first toe, and
10–11 lamellae under the fourth toe; (10) 4–6 precloacal
pores in males; (11) tail with whorls of enlarged tubercles;
(12) enlarged subcaudals; (13) in life grayish or light brown
dorsally with dark X-shaped markings across the body. The
markings are typically five, the first being in the scapular
region, three across the body, and the last in the pelvic area.
There is a narrow dark-brown stripe running from the nostril
across the eye to the temporal area. Tail with broad alternat-
ing black and white bands (when original). Body pinkish
ventrally.
Fig. 8 Holotype of H. farasani sp. n. (NMP 76104/2) in life A; lateral
B, dorsal C, and ventral Dview of the head; detail of dorsal scales
and tubercles E; precloacal area showing four precloacal pores F; and
detail of left foot F
◂
Diversification ofHemidactylus geckos (Squamata: Gekkonidae) incoastal plains andislands… 199
1 3
Comparisons. Hemidactylus farasani sp. n. is the largest of
its closest relatives, H. adensis, H. mandebensis, and H.
robustus, although the comparisons are statistically signifi-
cant only with H. adensis (Fig.3). Among these species,
it also has the longest head, both absolutely and relatively
compared to the body length, with all comparisons being
significant (Figs.3, SupplementaryS7). Especially impor-
tant are the comparisons with H. robustus and H. flavivir-
idis, the only other Hemidactylus species reported from the
Farasan Islands (Cunningham, 2010). The dorsal tubercles
of H. farasani sp. n. are larger, more prominent, and clearly
triangular compared to H. robustus, in which the tubercles
are smaller, granular, and round. Hemidactylus flaviviridis
does not have any enlarged dorsal tubercles whatsoever; the
dorsum is smooth. Genetic distances between H. farasani sp.
n. and the other Hemidactylus species included in this study
are shown in Supplementary TableS4.
Description of holotype Adult male (Fig.8), SVL 55.4mm,
head length 15.6mm, head width 11.3mm, head depth
6.7 mm, eye diameter 3.7 mm, axilla-groin distance
24.2mm. Supralabials 9/9 (right/left), infralabials 8/7. Nos-
tril is surrounded by rostral and three nasals. The uppermost
nasals are not in contact; there is a scale inserted between
them. Anterior postmentals are large and trapezoid and in
broad contact. Each anterior postmental is in contact only
with the first infralabial. Digits are distinctly dilated with
6/6 lamellae under the first toe and 10/10 lamellae under
the fourth toe. There are four precloacal pores in a slightly
open V-shaped line that is interrupted in the middle by a
pore-less scale. About half of the tail is regenerated. Body
tubercles are very large and protruding and give the speci-
men a spiky appearance. They continue on the tail, where
they form whorls of enlarged, posteriorly pointed tubercles
that are separated by five rows of small granules. The tongue
has been removed for genetic analysis and the intestines to
prevent rotting.
In life, grayish dorsally with light brown, regular,
X-shaped markings across the body. The dark stripe run-
ning across the eye was not that well discernible. The dark
bands on the tail were dark gray. The color pattern is only
faintly visible in the preservative. Original high-resolution
photographs of the holotype may be accessed in Morpho-
Bank, accession numbers: M823285–M823314.
Morphological andgenetic variation The number of infrala-
bials was 7 in most specimens; only three of the 14 morpho-
logically examined specimens had unilaterally 8 infralabials
(7 on the other side). In contrast to the holotype that has the
anterior postmental in contact only with the first infralabial,
Fig. 9 Paratypes of H. farasani
sp. n. in life and the spe-
cies’ type locality. A and B
– the type locality in the Sajid
Island (16.904°N, 41.898°E).
Specimens were frequently
encountered on or close to the
abandoned rocky walls in B.
C – adult male NMP 76104/7;
D – adult female NMP 76104/1,
note the presence of eggs; E
– juvenile NMP 76104/9; F –
juvenile NMP 76103
a
b
c
e
d
f
J.Šmíd et al.
200
1 3
in most specimens (79%), the anterior postmentals touch the
first and second infralabials. There are four adult males in
the set with well-developed precloacal pores, two of which
(holotype and paratype NMP 76104/4) have four pores and
two (paratypes NMP 76105/2–3) have six pores. Original
measurements are provided in Supplementary TableS2.
Hemidactylus farasani sp. n. shows a very low intraspecific
genetic diversity. In the mitochondrial 12S gene, all speci-
mens but one share the same haplotype, the one being dif-
ferent only in one nucleotide position along the 381-bp-long
gene fragment. In the cytb gene, within species, p-distances are
0.39 ± 0.26%. In the nuclear markers, all specimens share the
same allele in the rag1 gene and only two distinct alleles were
found in the cmos and rag2 markers. The pdc and mc1r mark-
ers show a slightly higher degree of variation with five and
six alleles, respectively, but even in these genes most samples
share the same allele with the rest separated from it by one or
two substitutions (Supplementary Figs.S4, S5, and S6).
Etymology The species epithet is a noun in apposition and
refers to the native inhabitants of the Farasan Islands, the
Farasani people.
Distribution, habitat, and ecology During our field work,
we found H. farasani sp. n. on the two largest islands of
the Farasan Archipelago, Farasan Al-Kebir and Sajid
(Fig.10). However, our surveys were limited to these
islands because logistic issues prevented us from visiting
the other islands of the archipelago. It is nonetheless likely
that the species is present on some of the other islands
as well. Two juvenile specimens from Sarso Island that
were originally attributed to H. turcicus (Linnaeus, 1758)
(SMF60196–7; Mertens, 1965) have also turned out to
belong to H. farasani sp. n. (MorphoBank accessions:
M823414–M823417, M823418–M823422). No reports
of H. farasani sp. n. have so far been made from the
mainland, and with the current knowledge, the species is
endemic to the Farasan Islands. Of interest would be to
explore the Dahlak Islands of Eritrea that are across the
Red Sea from the Farasan Islands (ca. 180km by a straight
line) and from where “H. turcicus” has also been reported
(Hoofien & Yaron, 1964).
Hemidactylus farasani sp. n. is nocturnal. Specimens were
found in rocky areas on walls of abandoned buildings as well
as in the closest vicinity of the ruins on rocks and on the
ground. Schätti and Gasperetti (1994) reported it from old
wells. Some specimens were found during the day hiding
inside drainage tunnels under roads. Vegetation in the area
is shown in Fig.9 and consisted of acacia trees (Vachellia
flava), date palms (Phoenix dactylifera), and caper (Capparis
cartilaginea; Al-Qthanin & Al-Yasi, 2021). Air temperature
in the night hours was about 28 and relative humidity 64%.
Fig. 10 Map of the Farasan
Islands with sampled localities
of H. farasani sp. n. and H.
robustus
Diversification ofHemidactylus geckos (Squamata: Gekkonidae) incoastal plains andislands… 201
1 3
We observed H. robustus to live in syntopy with H. far-
asani sp. n. (Fig.10). Other reptiles we found sympatric
with H. farasani sp. n. or confirmed to occur in the Farasan
Islands were Pristurus sp. 1 (Badiane etal., 2014), P. fla-
vipunctatus Rüppell, 1835 (Sphaerodactylidae), Ptyodac-
tylus hasselquistii (Donndorff, 1798) (Phyllodactylidae),
Mesalina austroarabica Sindaco etal., 2018 (Lacertidae),
Trachylepis brevicollis (Wiegmann, 1837) (Scincidae),
Psammophis schokari Forskål, 1775 (Psammophiidae),
and Echis borkini Cherlin, 1990 (Viperidae). All the species
have been confirmed to inhabit the islands before (Masseti,
2014), although the taxonomy of some has been updated
(Pook etal., 2009; Simó-Riudalbas etal., 2019; Sindaco
etal., 2018).
Conservation status Our field data suggest that H. farasani
sp. n. is common on Farasan Al-Kebir and Sajid Islands,
and literature data confirm its presence also on Sarso Island
(Fig.10; Mertens, 1965). Given that the Farasan Archi-
pelago has a very low human population density, which is
mostly concentrated in the main town of Farasan, we do not
assume the species to be under human-induced pressure.
Moreover, the entire Farasan Archipelago was established
as a protected area by the SWA in 1989 (Hall etal., 2010),
and we believe that as long as the protection of the islands
is well managed, the species is not under threat.
Discussion
Hemidactylus is the most species-rich Arabian squamate
genus, and despite major recent advances, its systematics
and taxonomy is still a dynamic field. The pace with which
new species are being discovered does not seem to be slow-
ing down. In the last ten years, an average of 3.3 new Hemi-
dactylus species has been described from Arabia every year.
The descriptions of H. almakhwah sp. n. and H. farasani sp.
n. provided here contribute to this knowledge and further
increase our understanding of the unique Arabian biota.
The Tihama Desert andAsir Mountains foothills
Most of the previous studies dealing with the systematics of
Arabian Hemidactylus have focused on mountainous areas
where the potential for new discoveries was greatest. For
example, Carranza and Arnold (2012) discovered, among
others, two new species confined to the Hajar Mountains
of northern Oman, and Šmíd etal. (2017) described two
narrow-ranging endemics from the Asir Mountains of Saudi
Arabia. The fact that the mountains of Arabia support a high
diversity of squamate reptiles has long been understood
(Cox etal., 2012). The high diversity was recently shown
to be a result of the high topographic heterogeneity of the
mountains that enables the coexistence of many species
(Šmíd etal., 2021a). However, here, we show that the low-
land areas also host yet unforeseen diversity.
The Tihama Desert is a renowned hotspot for reptile
endemics. Several species that represent a wide spectrum
of the squamate tree of life are restricted to the Tihama.
These are skinks (Chalcides levitoni Pasteur, 1978; Scincus
hemprichii Wiegmann, 1837), geckos (Stenodactylus yemen-
ensis Arnold, 1980), agamids (Uromastyx shobraki Wilms &
Schmitz, 2007), varanids (Varanus yemenensis Böhme etal.,
1989), colubrids (Lytorhynchus gasperetti Leviton, 1977),
and vipers (Echis borkini). By describing H. almakhwah sp.
n., we add one more species to this list. That the Tihama
endemics are so phylogenetically diverse suggests that they
originated through various biogeographic processes such
as vicariance (e.g., Stenodactylus yemenensis; Metallinou
etal., 2012), independent dispersals (e.g., Varanus yemen-
ensis; Portik & Papenfuss, 2012), or allopatric speciation
(Scincus hemprichii; Šmíd etal., 2021b), and not by local
radiations within the Tihama. The same seems to be the case
with the Hemidactylus geckos.
Hemidactylus almakhwah sp. n. is a member of the H.
saba species group, and while the other species of the
group inhabit mostly higher elevations in the Asir and
Hejaz Mountains, H. almakhwah sp. n. occurs, to our
knowledge, at 550m asl at maximum. It thus seems that
the present distribution of H. almakhwah sp. n. is a result
of dispersal from a high-elevation ancestor. In contrast to
the Tihama endemics mentioned above that occur in the
southern part of the coastal desert, H. almakhwah sp. n.
inhabits the northern Tihama. Interestingly, despite the
solid sampling of genes included in the phylogenetic anal-
yses, none of the phylogenetic approaches conducted here
provided a conclusive resolution as to the phylogenetic
relationships among species of the H. saba group. This has
been a recurrent phenomenon in all previous phylogenetic
reconstructions of the group regardless of the taxon and
gene sampling (Šmíd etal., 2013a, b, 2017). Although we
compiled the most comprehensive dataset of the H. saba
group so far, the position of H. almakhwah sp. n. within
the group remained unresolved. All analyses generally
recovered H. ulii to be sister to the rest of the H. saba
group, and the existence of the clade of H. alfarraji, H.
asirensis, and H. granosus, but the relationships between
this clade and H. almakhwah sp. n. and H. saba were not
resolved. The age of the crown node of the H. saba group
without H. ulii was estimated to be relatively recent, about
4.8 million years ago (Šmíd etal., 2017), which means
that the separation of H. almakhwah sp. n. from the rest of
the species in the group must have postdated this event. It
may be that the group radiated rapidly, and that the coin-
cidence or quick succession of speciation events did not
leave genetic traces for us to be able to resolve the tree as
J.Šmíd et al.
202
1 3
a fully bifurcating dichotomy, but it is also possible that
denser sampling of loci will help resolve the topology with
confidence.
The character of the dorsal scales of H. almakhwah sp.
n. that lacks or has very minuscule mid-dorsal tubercles is
unique among southwestern Arabian Hemidactylus geckos.
This morphological feature may be found in some other Ara-
bian Hemidactylus species, namely H. paucituberculatus
Carranza & Arnold, 2012, and H. masirahensis Carranza &
Arnold, 2012, which are endemic to southern Oman and to
Masirah Island off the eastern coast of Oman, respectively,
and do not occur in southwestern Saudi Arabia and Yemen
(Carranza & Arnold, 2012; Šmíd etal., 2021a). Given the
notorious difficulty of determining the many Hemidactylus
species in the field or in collections, this feature presents a
neat character that allows the identification of H. almakhwah
sp. n. specimens at first sight, even if the specimens are par-
tially damaged or not mature.
The Farasan Islands
The fauna and flora of the Farasan Islands are becoming
increasingly well documented in recent years. Studies
of various groups of organisms concur that the levels of
endemism are very low, if any, which corresponds to the
general trend of other Red Sea islands and archipelagos
(Masseti etal., 2015). For example, there are no known
endemic plants (Al-Qthanin & Al-Yasi, 2021) or beetles
in the Farasan Islands (Abdel-Dayem etal., 2020), and
the taxonomic status of the only endemic mammal, the
Farasan gazelle, is dubious (Bärmann etal., 2013). These
findings are, however, not surprising considering that the
archipelago is isolated from mainland Arabia by only
about a 40km-wide stretch of sea that rarely exceeds 60m
in depth, indicating the existence of land bridge connec-
tions at times of sea-level lowstands during the Pleistocene
glacial episodes. The species that occur in the Farasan
Islands belong either to widespread taxa (e.g., Psammo-
phis schokari) or to species distributed in the adjoining
mainland Arabia (e.g., Echis borkini, Mesalina austroara-
bica). There are also rare cases of species with an African
distribution that have populations in the Farasan Islands,
but not on the Arabian mainland (e.g., the bat Asellia
patriziide Beaux, 1931; Moeschler etal., 1990 or the
spurge Euphorbia collenetteae; Al-Zahrani & El-Karemy,
2007). Thus, the only known exceptions to this absence of
endemics are the colubrid snake Platyceps insulanus and
Hemidactylus farasani sp. n.
Hemidactylus farasani sp. n. belongs to the H. robustus
species group, together with the widespread H. robustus,
a clade of two south Arabian endemics (H. adensis and H.
mandebensis) and a clade of possibly three African species
(H. awashensis and two putative species from northeastern
Ethiopia whose status needs to be clarified using meth-
ods of integrative taxonomy). Phylogenetic relationships
within the H. robustus group remained largely unresolved.
The species tree analysis placed the African species as
sister to all the Arabian species in the group, but the analy-
ses of the concatenated data did not support this topology
and inferred the two south Arabian endemics H. adensis
and H. mandebensis as sister to the rest, however without
support. It is possible that the diversification in the group
was also rapid as we hypothesize for the H. saba group,
and which is suggested by the star-like topology of the
phylogenetic network.
The difference between the topologies of the species
tree and the concatenated tree was already recovered in
the first phylogenetic analysis of the robustus group (Šmíd
etal., 2015). Despite that the present study had a more
complete sampling of species than the Šmíd etal. (2015)
study and that we sequenced one more variable nuclear
marker (pdc), no method has yet produced a tree with high
nodal support. The cause may be that the nuclear genes
included in the analyses show a large degree of allele shar-
ing between different and often distantly related species
due to the slow substitution rate of these markers. As a
result, individual gene trees derived from these loci pro-
vide little resolution at this phylogenetic depth (results
not shown). We believe that sampling more variable loci
using, for example, RAD or ddRAD sequencing would
help resolving the relationships within the robustus group.
The basal split within the H. robustus group has been
estimated to take place between ca. 4.2 and 7.0 million years
ago (Šmíd etal., 2013a, 2017). This corresponds to the time
when Arabia and Africa were connected by a land bridge in
the Bab-el-Mandeb Strait (Bosworth etal., 2005). This tem-
porary land connection likely facilitated the dispersal of the
ancestor of the Ethiopian H. awashensis and the two putative
species from Arabia (Šmíd etal., 2013a). The fluctuating
sea level at that time might have also been responsible for
the separation of H. farasani sp. n. from the rest of the H.
robustus group. It however remains unclear why the spe-
cies remained confined to the islands despite the frequent
land bridge connections with mainland Arabia throughout
the Quaternary glacial cycles. For future research, it will be
essential to include material from the Dahlak Islands and
coastal northeastern Africa to verify whether H. farasani
sp. n. is indeed endemic to the Farasan Islands or if it is
another example of a satellite population of an otherwise
African species.
Hemidactylusmitogenomes
This is the first study on Hemidactylus geckos for which we
not only generated traditional phylogenetic markers using
Diversification ofHemidactylus geckos (Squamata: Gekkonidae) incoastal plains andislands… 203
1 3
Sanger sequencing but also employed Illumina sequencing to
obtain complete mitochondrial genomes of four species, two
of which were newly described in this study. It is pivotal to
note that all four specimens for which the mitogenomes were
sequenced are holotypes, the name-bearing specimens of the
species, which has crucial implications for future taxonomic
and phylogenetic studies as these sequences will always be
tied with the names of the species. We were inspired by
the recent paper of Köhler etal. (2021), who published a
draft genome and complete mitogenome of a holotype of a
new homalopsid snake. For the time being, the mitogenomic
sequences of the four Hemidactylus species published here
serve to characterize the genomic composition and archi-
tecture of the mitochondrial genomes of the four holotypes.
However, once complete genomes or mitogenomes become
available for more Hemidactylus species (work in progress),
they will provide a powerful tool to reconstruct the evo-
lutionary and diversification history of this highly diverse
gecko genus with confidence and trace the genomic signa-
tures of its highly successful radiation.
Conclusions
Southwestern Arabia supports a high richness of squamate
reptiles, with Hemidactylus being the most diverse genus.
We here identify and describe two new species from previ-
ously little explored regions – coastal deserts and the Red
Sea islands. We show the phylogenetic positions of the new
species within the SW Arabian radiation of the genus and
assess the morphological and ecological disparification
of the clade. As an important addition, we provide com-
plete mitochondrial genomes for the holotypes of the newly
described species as well as for holotypes of two other
species from the clade to characterize their mitogenomic
composition and architecture. Thus, the name-bearing speci-
mens of the species will be permanently associated with
their complete mitogenomic sequences, which will have key
implications for future taxonomic and phylogenetic studies
on the genus.
Supplementary Information The online version contains supplemen-
tary material available at https:// doi. org/ 10. 1007/ s13127- 022- 00572-w .
Acknowledgements We thank the staff of the Saudi Wildlife Authority
(SWA), most notably the SWA vice-president Dr. Hany Tatwany, for
his support, encouragement, and field work and permit arrangements.
Gunther Köhler and Maurice Scheidler kindly provided photographs
of the SMF specimens from Sarso Island. JŠ was supported by the
Charles University Research Centre program No. 204069, and by the
Ministry of Culture of the Czech Republic (DKRVO 2019–2023/6.
VII.d, 00023272). MU was supported by Charles University (SVV
260571/2022). LN was supported by project OP RDE, called Inter-
national mobility of researchers at Charles University (MSCA-IF IV),
with reg. n. CZ.02.2.69/0.0/0.0/20_079/. SC was supported by grant
PGC2018-098290-B-I00 (MCIU/AEI/FEDER, UE), Spain.
Data availability Both Sanger- and Illumina-generated DNA
sequence data are available on GenBank (https:// www. ncbi. nlm. nih.
gov/ genba nk/). Original high-resolution photographs of all specimens
examined have been deposited and are publicly available through
MorphoBank (https:// morph obank. org/). Original morphological data
(metric and meristic traits) are provided as a tab-delimited text file in
Supplementary TableS2.
Declarations
Consent for publication All authors approved the final version of the
manuscript for publication.
Conflict of interest The authors declare no competing interests.
References
Abdel-Dayem, M. S., El-Ghiet, U. M. A., Elsheikh, T. M., Elgharbawy,
A. A., Al-Fifi, Z. I., & Aldhafer, H. M. (2020). The first survey of
the beetles (Coleoptera) of the Farasan Archipelago of the southern
Red Sea, Kingdom of Saudi Arabia. ZooKeys, 959, 17–86. https://
doi. org/ 10. 3897/ zooke ys. 959. 51224
Al-Qthanin, R., & Al-Yasi, H. (2021). Progress towards an updated
checklist of the Farasan Archipelago flora. Journal of Arid Envi-
ronments, 189, 104488. https:// doi. org/ 10. 1016/j. jarid env. 2021.
104488
Al-Zahrani, D., & El-Karemy, Z. (2007). A new succulent Euphor-
bia (Euphorbiaceae) species from the Red Sea coast and islands.
Edinburgh Journal of Botany, 64, 131–136. https:// doi. org/ 10.
1017/ S0960 42860 70008 56
Andrews, S. (2010). FastQC: A quality control tool for high through-
put sequence data. Retrieved from http:// www. bioin forma tics.
babra ham. ac. uk/ proje cts/ fastqc
Arnold, E. N. (1980). Reptiles of Saudi Arabia. A review of the liz-
ard genus Stenodactylus (Reptilia: Gekkonidae). Fauna of Saudi
Arabia, 2, 368–404.
Arnold, E. N. (1986). A key and annotated check-list to the lizards and
amphisbaenians of Arabia. Fauna of Saudi Arabia, 8, 385–435.
Badiane, A., Garcia-Porta, J., Červenka, J., Kratochvíl, L., Sindaco, R.,
Robinson, M. D., Morales, H., Mazuch, T., Price, T., Amat, F., Shobrak,
M. Y., Wilms, T., Simó-Riudalbas, M., Ahmadzadeh, F., Papenfuss,
T. J., Cluchier, A., Viglione, J., & Carranza, S. (2014). Phylogenetic
relationships of semaphore geckos (Squamata: Sphaerodactylidae:
Pristurus) with an assessment of the taxonomy of Pristurus rupestris.
Zootaxa, 3835, 33–58.https:// doi. org/ 10. 11646/ zoota xa. 3835.1.2
Bärmann, E. V., Wronski, T., Lerp, H., Azanza, B., Börner, S., Erpenbeck,
D., Rössner, G. E., & Wörheide, G. (2013). A morphometric
and genetic framework for the genus Gazella de Blainville, 1816
(Ruminantia: Bovidae) with special focus on Arabian and Levan-
tine mountain gazelles. Zoological Journal of the Linnean Society,
169, 673–696. https:// doi. org/ 10. 1111/ zoj. 12066
Bauer, A. M., de Silva, A., Greenbaum, E., & Jackman, T. (2007). A
new species of day gecko from high elevation in Sri Lanka, with a
preliminary phylogeny of Sri Lankan Cnemaspis (Reptilia, Squa-
mata, Gekkonidae). Mitteilungen Aus Dem Zoologischen Museum
in Berlin, Zoologische Reihe, 83, 22–32. https:// doi. org/ 10. 1002/
mmnz. 20060 0022
Benson, G. (1999). Tandem repeats finder: A program to analyze DNA
sequences. Nucleic Acids Research, 27, 573–580. https:// doi. org/
10. 1093/ nar/ 27.2. 573
J.Šmíd et al.
204
1 3
Böhme, W., Joger, U., & Schätti, B. (1989). A new monitor lizard
from Yemen, with notes on ecology, phylogeny and zoogeography.
Fauna of Saudi Arabia, 10, 433–448.
Bolger, A. M., Lohse, M., & Usadel, B. (2014). Trimmomatic: A flexible
trimmer for Illumina sequence data. Bioinformatics, 30, 2114–2120.
https:// doi. org/ 10. 1093/ bioin forma tics/ btu170
Bosworth, W., Huchon, P., & McClay, K. (2005). The Red Sea and Gulf
of Aden Basins. Journal of African Earth Sciences, 43, 334–378.
https:// doi. org/ 10. 1016/j. jafre arsci. 2005. 07. 020
Bouckaert, R., Alvarado-Mora, M. V., & Pinho, J. R. (2013). Evolutionary
rates and HBV: Issues of rate estimation with Bayesian molecular
methods. Antiviral Therapy, 18, 497–503. https:// doi. org/ 10. 3851/
IMP26 56
Bouckaert, R., Heled, J., Kühnert, D., Vaughan, T., Wu, C. -H., Xie, D.,
Suchard, M. A., Rambaut, A., & Drummond, A. J. (2014). BEAST
2: A software platform for Bayesian evolutionary analysis. PLoS
Computational Biology, 10, e1003537. https:// doi. org/ 10. 1371/
journ al. pcbi. 10035 37
Bryant, D., & Moulton, V. (2004). Neighbor-net: An agglomerative
method for the construction of phylogenetic networks. Molecu-
lar Biology and Evolution, 21, 255–265. https:// doi. org/ 10. 1093/
molbev/ msh018
Busais, S., & Joger, U. (2011). Three new species of Hemidactylus
Oken, 1817 from Yemen (Squamata, Gekkonidae). Vertebrate
Zoology, 61, 267–280.
Carranza, S., & Arnold, E. N. (2006). Systematics, biogeography
and evolution of Hemidactylus geckos (Reptilia: Gekkonidae)
elucidated using mitochondrial DNA sequences. Molecular
Phylogenetics and Evolution, 38, 531–545. https:// doi. org/ 10.
1016/j. ympev. 2005. 07. 012
Carranza, S., & Arnold, E. N. (2012). A review of the geckos of the
genus Hemidactylus (Squamata: Gekkonidae) from Oman based
on morphology, mitochondrial and nuclear data, with descrip-
tions of eight new species. Zootaxa, 3378, 1–95.
Carranza, S., Xipell, M., Tarroso, P., Gardner, A., Arnold, E. N., Robinson,
M. D., Simó-Riudalbas, M., Vasconcelos, R., de Pous, P., Amat,
F., Šmíd, J., Sindaco, R., Metallinou, M., Els, J., Pleguezuelos, J.
M., Machado, L., Donaire, D., Martínez, G., Garcia-Porta, J., &
Al Akhzami, S. N. (2018). Diversity, distribution and conserva-
tion of the terrestrial reptiles of Oman (Sauropsida, Squamata).
PLoS ONE, 13, e0190389. https:// doi. org/ 10. 1371/ journ al. pone.
01903 89
Castresana, J. (2000). Selection of conserved blocks from multiple
alignments for their use in phylogenetic analysis. Molecular
Biology and Evolution, 17, 540–552. https:// doi. org/ 10. 1093/
oxfor djour nals. molbev. a0263 34
Chan, P. P., & Lowe, T. M. (2019). tRNAscan-SE: Searching for tRNA
genes in genomic sequences. In M. Kollmar (Ed.), Gene Predic-
tion: Methods and Protocols (pp. 1–14). Springer.
Cherlin, V. (1990). Taxonomic revision of the snake genus Echis
(Viperidae) II. An analysis of taxonomy and description of new
forms. Proceedings of the Zoological Institute, Leningrad, 207,
193–223.
Clement, M., Snell, Q., Walker, P., Posada, D., & Crandall, K. (2002).
TCS: Estimating gene genealogies. Paper presented at the Pro-
ceedings of the 16th International Parallel and Distributed Pro-
cessing Symposium.
Cox, N., Mallon, D., Bowles, P., Els, J., & Tognelli, M. (2012). The con-
servation status and distribution of reptiles of the Arabian Peninsula
(2831715601). Retrieved from https:// www. iucn. org/ conte nt/ conse rvati
on- status- and- distr ibuti on- repti les- arabi an- penin sula
Cunningham, P. (2010). A contribution to the reptile of the Farasan
Islands, Saudi Arabia. African Herps News, 50, 21–24.
de Beaux, O. (1931). Spedizione del barone Raimondo Franchetti
in Dancalia. Mammiferi. Annali del Museo Civico di Storia
Naturale di Genova, 55, 183–217.
Donath, A., Jühling, F., Al-Arab, M., Bernhart, S. H., Reinhardt, F.,
Stadler, P. F., Middendorf, M., & Bernt, M. (2019). Improved
annotation of protein-coding genes boundaries in metazoan
mitochondrial genomes. Nucleic Acids Research, 47, 10543–
10552. https:// doi. org/ 10. 1093/ nar/ gkz833
Edgell, H. S. (2006). Arabian deserts: Nature, origin and evolution.
Springer.
ESRI. (2011). ArcGIS Desktop, Release 10.
Felsenstein, J. (1985). Confidence limits on phylogenies: An
approach using the bootstrap. Evolution, 39, 783–791. https://
doi. org/ 10. 2307/ 24086 78
Flot, J. F. (2010). Seqphase: A web tool for interconverting phase
input/output files and fasta sequence alignments. Molecular
Ecology Resources, 10, 162–166. https:// doi. org/ 10. 1111/j.
1755- 0998. 2009. 02732.x
Garcia-Porta, J., Šmíd, J., Sol, D., Fasola, M., & Carranza, S. (2016).
Testing the island effect on phenotypic diversification: Insights
from the Hemidactylus geckos of the Socotra Archipelago. Sci-
entific Reports, 6, 23729. https:// doi. org/ 10. 1038/ srep2 3729
Gasperetti, J. (1988). Snakes of Arabia. Fauna of Saudi Arabia, 9,
169–450.
Gómez-Díaz, E., Sindaco, R., Pupin, F., Fasola, M., & Carranza,
S. (2012). Origin and insitu diversification in Hemidactylus
geckos of the Socotra Archipelago. Molecular Ecology, 21,
4074–4092. https:// doi. org/ 10. 1111/j. 1365- 294X. 2012. 05672.x
Guindon, S., Dufayard, J. -F., Lefort, V., Anisimova, M., Hordijk, W.,
& Gascuel, O. (2010). New algorithms and methods to estimate
maximum-likelihood phylogenies: Assessing the performance
of PhyML 3.0. Systematic Biology, 59, 307–321. https:// doi. org/
10. 1093/ sysbio/ syq010
Hall, H., Llewellyn, O. A., Miller, A. G., Al-Abbasi, T. M., Al-
Wetaid, A. H., Al-Harbi, R. J., & Al-Shammari, K. F. (2010).
Important plant areas in the Arabian Peninsula 2. Farasan Archi-
pelago. Edinburgh Journal of Botany, 67, 189–208. https:// doi.
org/ 10. 1017/ S0960 42861 00000 89
Harrigan, R. J., Mazza, M. E., & Sorenson, M. D. (2008). Compu-
tation vs. cloning: Evaluation of two methods for haplotype
determination. Molecular Ecology Resources, 8, 1239–1248.
https:// doi. org/ 10. 1111/j. 1755- 0998. 2008. 02241.x
Hoang, D. T., Chernomor, O., Von Haeseler, A., Minh, B. Q., &
Vinh, L. S. (2018). UFBoot2: Improving the ultrafast bootstrap
approximation. Molecular Biology and Evolution, 35, 518–522.
https:// doi. org/ 10. 1093/ molbev/ msx281
Hoofien, J., & Yaron, Z. (1964). A collection of reptiles from the
Dahlak archipelago (Red Sea). Israel South Red Sea Expedi-
tion, 35, 35–40.
Huson, D. H., & Bryant, D. (2006). Application of phylogenetic
networks in evolutionary studies. Molecular Biology and Evolu-
tion, 23, 254–267. https:// doi. org/ 10. 1093/ molbev/ msj030
Iwasaki, W., Fukunaga, T., Isagozawa, R., Yamada, K., Maeda, Y.,
Satoh, T. P., Sado, T., Mabuchi, K., Takeshima, H., & Miya, M.
(2013). MitoFish and MitoAnnotator: A mitochondrial genome
database of fish with an accurate and automatic annotation pipe-
line. Molecular Biology and Evolution, 30, 2531–2540. https://
doi. org/ 10. 1093/ molbev/ mst141
Jie, Y., Jianli, Z., Chao, T., & Kaiya, Z. (2009). Complete nucleotide
sequence and gene organization of the mitochondrial genome of
Common House Gecko, Hemidactylus frenatus. Journal of Nanjing
Normal University Natural Science Edition, 32.
Katoh, K., Rozewicki, J., & Yamada, K. D. (2019). MAFFT online
service: Multiple sequence alignment, interactive sequence
choice and visualization. Briefings in Bioinformatics, 20, 1160–
1166. https:// doi. org/ 10. 1093/ bib/ bbx108
Kearse, M., Moir, R., Wilson, A., Stones-Havas, S., Cheung, M.,
Sturrock, S., Buxton, S., Cooper, A., Markowitz, S., & Duran,
C. (2012). Geneious Basic: An integrated and extendable
Diversification ofHemidactylus geckos (Squamata: Gekkonidae) incoastal plains andislands… 205
1 3
desktop software platform for the organization and analysis of
sequence data. Bioinformatics, 28, 1647–1649. https:// doi. org/
10. 1093/ bioin forma tics/ bts199
Köhler, G., Khaing, K. P. P., Than, N. L., Baranski, D., Schell, T.,
Greve, C., Janke, A., & Pauls, S. U. (2021). A new genus and
species of mud snake from Myanmar (Reptilia, Squamata, Hom-
alopsidae). Zootaxa, 4915, 301–325. https:// doi. org/ 10. 11646/
zoota xa. 4915.3.1
Kumar, S., Stecher, G., Li, M., Knyaz, C., & Tamura, K. (2018).
MEGA X: Molecular evolutionary genetics analysis across com-
puting platforms. Molecular Biology and Evolution, 35, 1547–
1549. https:// doi. org/ 10. 1093/ molbev/ msy096
Lanfear, R., Frandsen, P. B., Wright, A. M., Senfeld, T., & Calcott, B.
(2017). PartitionFinder 2: New methods for selecting partitioned
models of evolution for molecular and morphological phyloge-
netic analyses. Molecular Biology and Evolution, 34, 772–773.
https:// doi. org/ 10. 1093/ molbev/ msw260
Leigh, J. W., & Bryant, D. (2015). POPART: Full-feature software for
haplotype network construction. Methods in Ecology and Evolu-
tion, 6, 1110–1116. https:// doi. org/ 10. 1111/ 2041- 210X. 12410
Leviton, A. E. (1977). A new lytorhynchid snake. Journal of the Saudi
Arabian Natural History Society, 19, 16–25.
Masseti, M. (2014). Herpetological enigmas from the Arabian seas,
with particular reference to the Sarso island racer, Platyceps insu-
laris Mertens, 1965 (Farasan archipelago, Saudi Arabia). In M.
Capula & C. C. (Eds.), Scripta Herpetologica. Studies on Amphib-
ians and Reptiles in honour of Benedetto Lanza (pp. 99–116).
Latina (Italy): Societas Herpetologica Italica.
Masseti, M., De Marchi, G., & Chiozzi, G. (2015). Forbidden islands.
The absence of endemics among the insular non-volant terrestrial
mammalian fauna of the Red Sea. Natural History Sciences, 2,
101–130. https:// doi. org/ 10. 4081/ nhs. 2015. 253
Mertens, R. (1965). Eine neue Natter von einer Insel des Roten Meeres.
Senckenbergiana Biologica, 46, 5–9.
Metallinou, M., Arnold, E. N., Crochet, P. -A., Geniez, P., Brito, J. C.,
Lymberakis, P., Baha El Din, S. M., Sindaco, R., Robinson, M., &
Carranza, S. (2012). Conquering the Sahara and Arabian deserts:
Systematics and biogeography of Stenodactylus geckos (Reptilia:
Gekkonidae). BMC Evolutionary Biology, 12, 1–17. https:// doi.
org/ 10. 1186/ 1471- 2148- 12- 258
Miller, M., Pfeiffer, W., & Schwartz, T. (2010). Creating the CIPRES Sci-
ence Gateway for inference of large phylogenetic trees. Paper pre-
sented at the Proceedings of the Gateway Computing Environments
Workshop (GCE), New Orleans.
Moeschler, P., Nader, I., & Gaucher, P. (1990). First record of Asellia
patrizii De Beaux, 1931 (Chiroptera: Hipposideridae) in Saudi
Arabia. Mammalia, 54, 654–655.
Moravec, J., Kratochvíl, L., Amr, Z. S., Jandzik, D., Šmíd, J., &
Gvoždík, V. (2011). High genetic differentiation within the Hemi-
dactylus turcicus complex (Reptilia: Gekkonidae) in the Levant,
with comments on the phylogeny and systematics of the genus.
Zootaxa, 2894, 21–38.
Nguyen, L. -T., Schmidt, H. A., Von Haeseler, A., & Minh, B. Q.
(2015). IQ-TREE: A fast and effective stochastic algorithm for
estimating maximum-likelihood phylogenies. Molecular Biology
and Evolution, 32, 268–274. https:// doi. org/ 10. 1093/ molbev/
msu300
Pasteur, G. (1978). Notes sur les sauriens du genre Chalcides III.
Description de Chalcides levitoni n. sp., d’Arabie Saoudite. Jour-
nal of Herpetology, 12, 371–372.
Pook, C. E., Joger, U., Stümpel, N., & Wüster, W. (2009). When conti-
nents collide: Phylogeny, historical biogeography and systematics
of the medically important viper genus Echis (Squamata: Serpentes:
Viperidae). Molecular Phylogenetics and Evolution, 53, 792–807.
https:// doi. org/ 10. 1016/j. ympev. 2009. 08. 002
Portik, D. M., & Papenfuss, T. J. (2012). Monitors cross the Red Sea:
The biogeographic history of Varanus yemenensis. Molecular
Phylogenetics and Evolution, 62, 561–565. https:// doi. org/ 10.
1016/j. ympev. 2011. 09. 024
Qin, X., Hou, L., Yang, X., & Zhang, Y. (2014). Complete mitochon-
drial genome of Hemidactylus bowringii. unpublished.
Rambaut, A., Drummond, A. J., Xie, D., Baele, G., & Suchard, M. A.
(2018). Posterior summarization in Bayesian phylogenetics using
Tracer 1.7. Systematic Biology, 67, 901–904. https:// doi. org/ 10.
1093/ sysbio/ syy032
Safaei-Mahroo, B., Ghaffari, H., Ghafoor, A., & Amini, S. (2017). A
new species of Hemidactylus (Squamata: Gekkota: Gekkonidae)
from Qara Dagh Mountains, Kurdistan Region, with a key to the
genus in Iraq. Zootaxa, 4363, 377–392. https:// doi. org/ 10. 11646/
zoota xa. 4363.3.4
Sakellariou, D., Rousakis, G., Panagiotopoulos, I., Morfis, I., & Bai-
ley, G. N. (2019). Geological structure and late quaternary geo-
morphological evolution of the Farasan Islands continental shelf,
south Red sea, SW Saudi Arabia. In N. Rasul & I. Stewart (Eds.),
Geological Setting, Palaeoenvironment and Archaeology of the
Red Sea (pp. 629–652). Springer.
Schätti, B., & Gasperetti, J. (1994). A contribution to the herpetofauna
of southwest Arabia. Fauna of Saudi Arabia, 14, 348–423.
Simó-Riudalbas, M., Tamar, K., Šmíd, J., Mitsi, P., Sindaco, R., Chirio,
L., & Carranza, S. (2019). Biogeography of Mesalina (Reptilia:
Lacertidae), with special emphasis on the Mesalina adramitana
group from Arabia and the Socotra Archipelago. Molecular Phy-
logenetics and Evolution, 137, 300–312. https:// doi. org/ 10. 1016/j.
ympev. 2019. 04. 023
Sindaco, R., Simo-Riudalbas, M., Sacchi, R., & Carranza, S. (2018).
Systematics of the Mesalina guttulata species complex (Squamata:
Lacertidae) from Arabia with the description of two new spe-
cies. Zootaxa, 4429, 513–547. https:// doi. org/ 10. 11646/ zoota xa.
4429.3.4
Šmíd, J., Carranza, S., Kratochvíl, L., Gvoždík, V., Nasher, A. K., &
Moravec, J. (2013a). Out of Arabia: A complex biogeographic
history of multiple vicariance and dispersal events in the gecko
genus Hemidactylus (Reptilia: Gekkonidae). PLoS ONE, 8, e64018.
https:// doi. org/ 10. 1371/ journ al. pone. 00640 18
Šmíd, J., Mazuch, T., Nováková, L., Modrý, D., Malonza, P. K., Elmi,
H. S. A., Carranza, S., & Moravec, J. (2020). Phylogeny and sys-
tematic revision of the gecko genus Hemidactylus from the Horn
of Africa (Squamata: Gekkonidae). Herpetological Monographs,
33, 26–47. https:// doi. org/ 10. 1655/ HERPM ONOGR APHS-
D- 19- 00010.1
Šmíd, J., Moravec, J., Kratochvíl, L., Gvoždík, V., Nasher, A. K., Busais,
S., Wilms, T., Shobrak, M., & Carranza, S. (2013b). Two newly rec-
ognized species of Hemidactylus (Squamata, Gekkonidae) from the
Arabian Peninsula and Sinai. Egypt. Zookeys, 355, 79–107. https://
doi. org/ 10. 3897/ zooke ys. 355. 6190
Šmíd, J., Moravec, J., Kratochvíl, L., Nasher, A. K., Mazuch, T.,
Gvoždík, V., & Carranza, S. (2015). Multilocus phylogeny and
taxonomic revision of the Hemidactylus robustus species group
(Reptilia, Gekkonidae) with descriptions of three new species from
Yemen and Ethiopia. Systematics and Biodiversity, 13, 346–368.
https:// doi. org/ 10. 1080/ 14772 000. 2014. 996264
Šmíd, J., Shobrak, M., Wilms, T., Joger, U., & Carranza, S. (2017).
Endemic diversification in the mountains: Genetic, morphologi-
cal, and geographical differentiation of the Hemidactylus geckos
in southwestern Arabia. Organisms Diversity & Evolution, 17,
267–285. https:// doi. org/ 10. 1007/ s13127- 016- 0293-3
Šmíd, J., Sindaco, R., Shobrak, M., Busais, S., Tamar, K., Aghová, T.,
Simó-Riudalbas, M., Tarroso, P., Geniez, P., Crochet, P. -A., Els,
J., Burriel-Carranza, B., Tejero-Cicuéndez, H., & Carranza, S.
(2021a). Diversity patterns and evolutionary history of Arabian
J.Šmíd et al.
206
1 3
squamates. Journal of Biogeography, 48, 1183–1199. https:// doi.
org/ 10. 1111/ jbi. 14070
Šmíd, J., Uvizl, M., Shobrak, M., AlGethami, R. H. M., Algethami, A.
R., Alanazi, A. S. K., Alsubaie, S. D., Busais, S., & Carranza, S.
(2021b). Swimming through the sands of the Sahara and Arabian
deserts: Phylogeny of sandfish skinks (Scincidae, Scincus) reveals a
recent and rapid diversification. Molecular Phylogenetics and Evo-
lution, 155, 107012. https:// doi. org/ 10. 1016/j. ympev. 2020. 107012
Stephens, M., Smith, N. J., & Donnelly, P. (2001). A new statisti-
cal method for haplotype reconstruction from population data.
American Journal of Human Genetics, 68, 978–989. https:// doi.
org/ 10. 1086/ 319501
Tejero-Cicuéndez, H., Patton, A. H., Caetano, D. S., Šmíd, J., Harmon,
L. J., & Carranza, S. (2021). Reconstructing squamate bioge-
ography in Afro-Arabia reveals the influence of a complex and
dynamic geologic past. Systematic Biology, Early View. https://
doi. org/ 10. 1093/ sysbio/ syab0 25
Thouless, C., & Bassri, K. A. (1991). Taxonomic status of the Farasan
Island gazelle. Journal of Zoology, 223, 151–159.
Torki, F., Manthey, U., & Barts, M. (2011). A new Hemidactylus from
Lorestan Province, western Iran, with notes on Hemidactylus
robustus Heyden, 1827 (Reptilia: Squamata: Gekkonidae). Sau-
ria, 33, 47–56.
Trifinopoulos, J., Nguyen, L. -T., von Haeseler, A., & Minh, B. Q.
(2016). W-IQ-TREE: A fast online phylogenetic tool for maximum
likelihood analysis. Nucleic Acids Research, 44, W232–W235.
https:// doi. org/ 10. 1093/ nar/ gkw256
Uetz, P., Freed, P., & Hošek, J. (2022). The reptile database. Retrieved
from http:// www. repti le- datab ase. org/
Vasconcelos, R., & Carranza, S. (2014). Systematics and biogeogra-
phy of Hemidactylus homoeolepis Blanford, 1881 (Squamata:
Gekkonidae), with the description of a new species from Ara-
bia. Zootaxa, 3835, 501–527.https:// doi. org/ 10. 11646/ zoota xa.
3835.4.4
Wilms, T. M., & Schmitz, A. (2007). A new polytypic species of the
genus Uromastyx Merrem 1820 (Reptilia: Squamata: Agamidae:
Leiolepidinae) from southwestern Arabia. Zootaxa, 1–23.
Wuertz, D., Setz, T., Chalabi, Y., Maechler, M., & Setz, M. T. (2017).
Package ‘fBasics’. Rmetrics-Markets and Basic Statistics. R Foun-
dation for Statistical Computing.
Publisher's Note Springer Nature remains neutral with regard to
jurisdictional claims in published maps and institutional affiliations.
Springer Nature or its licensor holds exclusive rights to this article under
a publishing agreement with the author(s) or other rightsholder(s);
author self-archiving of the accepted manuscript version of this article
is solely governed by the terms of such publishing agreement and
applicable law.
Diversification ofHemidactylus geckos (Squamata: Gekkonidae) incoastal plains andislands… 207
Supplementary Materials - S1
SUPPLEMENTARY MATERIALS
for
Diversification of Hemidactylus geckos (Squamata: Gekkonidae) in coastal plains and islands of
southwestern Arabia with descriptions and complete mitochondrial genomes of two endemic
species to Saudi Arabia
Jiří Šmíd, Marek Uvizl, Mohammed Shobrak, Salem Busais, Al Faqih Ali Salim, Raed Hamoud M
AlGethami, Abdulaziz Raqi AlGethami, Abdulkarim Saleh K Alanazi, Saad Dasman Alsubaie, Michail
Rovatsos, Lucie Nováková, Tomáš Mazuch, Salvador Carranza
Contents:
Supplementary Figures ............................................................................................................................... 2
Figure S1 ................................................................................................................................................. 2
Figure S2 ................................................................................................................................................. 3
Figure S3 ................................................................................................................................................. 4
Figure S4 ................................................................................................................................................. 5
Figure S5 ................................................................................................................................................. 6
Figure S6 ................................................................................................................................................. 7
Figure S7 ................................................................................................................................................. 8
Supplementary Tables ................................................................................................................................ 9
Table S1. ................................................................................................................................................. 9
Table S2 ................................................................................................................................................ 11
Table S3. ............................................................................................................................................... 12
Table S4. ............................................................................................................................................... 13
References for supplementary materials ................................................................................................... 14
Supplementary Materials - S2
Supplementary Figures
Figure S1. Topographic and political features of southwestern Arabia mentioned in the article.
100 km
Farasan Islands
Dahlak Islands
Sudan
Eritrea
Ethiopia
Yemen
Saudi Arabia
Tihama
Asir Mountains
Red Sea
Gulf of Aden
Jazan
southern
northern
Al Qahma
Strait of Bab-el-Mandeb
Supplementary Materials - S3
Figure S2. Phylogenetic tree resulting from the Bayesian analysis of two mitochondrial and five nuclear
markers concatenated. Values by branches indicate posterior probabilities.
0.03
0.6
1
1
1
0.99
1
0.95
1
0.84
1
1
1
0.98
0.98
0.97
1
0.99
1
0.98
1
1
1
1
1
1
1
1
1
1
1
0.99
1
1
1
1
1
1
1
0.69
1
1
1
1
1
alfarraji_HSA40
granosus_CN15017
granosus_HSA54
robustus_CN15588
robustus_CN15579
awashensis_JS242
asirensis_CN15156
awashensis_JS249
granosus_HSA65
robustus_CN15556
farasani_CN15594
sp_Etio2_JS315
almakhwah_CN15167
alfarraji_HSA41
asirensis_HSA2
farasani_CN15571
awashensis_JS247
robustus_CN15649
alfarraji_HSA38
almakhwah_CN15140
awashensis_JS246
sp_Etio3_JS294
granosus_CN15021
asirensis_CN15126
mandebensis_JS27
asirensis_CN15096
almakhwah_CN15168
robustus_JS67
sp_Etio2_JS304
alfarraji_HSA43
alfarraji_HSA36
almakhwah_CN15693
awashensis_JS204
farasani_CN15575
robustus_JS58
robustus_CN15659
saba_BJ29
farasani_CN15608
almakhwah_CN15519
farasani_CN15572
robustus_CN15597
granosus_CN15008
robustus_CN15661
almakhwah_CN15719
farasani_CN15568
flaviviridis_JS119
adensis_BJ15
flaviviridis_JS113
ulii_JS18
almakhwah_CN15130
farasani_CN15570
asirensis_CN15148
adensis_BJ13
asirensis_HSA5
robustus_CN15532
granosus_HSA70
angulatus_JS123
asirensis_CN15088
asirensis_HSA1
almakhwah_CN15129
robustus_CN15647
alfarraji_HSA34
awashensis_JS245
farasani_CN15589
awashensis_JS248
mandebensis_JS39
ulii_JS17
asirensis_HSA52
alfarraji_HSA42
almakhwah_CN15169
farasani_CN15577
robustus_CN15643
robustus_CN15581
robustus_CN15678
awashensis_JS244
granosus_CN15797
farasani_CN15576
ulii_JS48
granosus_JS332
robustus_CN15677
ulii_JS37
almakhwah_CN15709
granosus_CN15796
asirensis_CN15155
granosus_HSA62
robustus_CN15573
alfarraji_HSA37
robustus_CN15583
mandebensis_JS36
ulii_JS32
robustus_CN15606
farasani_CN15580
asirensis_CN15160
granosus_HSA63
granosus_CN15019
robustus_CN15551
awashensis_JS250
asirensis_CN15057
ulii_JS38
granosus_HSA55
granosus_CN15018
saba_BJ27
adensis_BJ14
farasani_CN15609
alfarraji_HSA35
asirensis_CN15161
granosus_CN15020
awashensis_JS243
farasani_CN15591
robustus_JS50
robustus_CN15614
saba_BJ28
robustus_CN15679
robustus_JS106
adensis_BJ11
ulii_JS49
almakhwah_CN15711
almakhwah_CN15710
asirensis_HSA12
farasani_CN15578
granosus_HSA61
asirensis_HSA13
farasani_CN15582
sp_Etio3_JS295
awashensis_JS212
robustus_CN15599
asirensis_CN15223
granosus_CN15185
asirensis_HSA44
almakhwah_CN15131
adensis_BJ10
robustus_CN15567
almakhwah_CN15166
ruspolii_JS177
almakhwah_CN15720
ulii_JS47
asirensis_CN15183
robustus_CN15574
Supplementary Materials - S4
Figure S3. Species tree of the Hemidactylus species included in the study. Numbers above branches
show posterior probability values.
0.01 1
1
0.46
0.89
0.61
0.91
0.99
0.57
0.67
1
1
H. mandebensis
H. adensis
H. robustus
H. farasani sp. n.
H. sp. Etio2
H. sp. Etio3
H. awashensis
H. ulii
H. saba
H. almakhwah sp. n.
H. alfarraji
H. granosus
H. asirensis
Supplementary Materials - S5
Figure S4. Unrooted allele networks for the cmos and mc1r nuclear loci. Species are color-coded and
circle sizes are proportional to the number of individuals with a given allele. Short transverse bars on the
connecting lines indicate the number of mutational steps between alleles. Alleles of H. almakhwah sp. n.
and H. farasani sp. n. are highlighted with blue and red blobs, respectively.
40 samples
cmos
10 samples
1 sample
mc1r
10 samples
1 sample
H. adensis
H. alfarraji
H. asirensis
H. awashensis
H. granosus
H. mandebensis
H. robustus
H. saba H. almakhwah sp. n.
H. farasani sp. n.H. ulii
H. sp Etio2. H. sp Etio3.
H. almakhwah sp. n.
H. almakhwah sp. n.
H. farasani sp. n.
H. farasani sp. n.
Supplementary Materials - S6
Figure S5. Unrooted allele networks for the rag1 and rag2 nuclear loci. Species are color-coded and
circle sizes are proportional to the number of individuals with a given allele. Short transverse bars on the
connecting lines indicate the number of mutational steps between alleles. Alleles of H. almakhwah sp. n.
and H. farasani sp. n. are highlighted with blue and red blobs, respectively.
H. adensis
H. alfarraji
H. asirensis
H. awashensis
H. granosus
H. mandebensis
H. robustus
H. saba H. almakhwah sp. n.
H. farasani sp. n.H. ulii
H. sp Etio2. H. sp Etio3.
H. almakhwah sp. n.
H. farasani sp. n.
H. almakhwah sp. n.
H. farasani sp. n.
10 samples
1 sample
10 samples
1 sample
rag1
rag2
Supplementary Materials - S7
Figure S6. Unrooted allele network for the pdc nuclear locus. Species are color-coded and circle sizes
are proportional to the number of individuals with a given allele. Short transverse bars on the connecting
lines indicate the number of mutational steps between alleles. Alleles of H. almakhwah sp. n. and H.
farasani sp. n. are highlighted with blue and red blobs, respectively.
pdc
H. adensis
H. alfarraji
H. asirensis
H. awashensis
H. granosus
H. mandebensis
H. robustus
H. saba H. almakhwah sp. n.
H. farasani sp. n.H. ulii
H. sp Etio2. H. sp Etio3.
H. almakhwah sp. n.
H. farasani sp. n.
10 samples
1 sample
Supplementary Materials - S8
Figure S7. Morphological features of the 13 Hemidactylus species included in the study. Horizontal bars above boxplots show significant
differences between species pairs in that trait as assessed by means of the Kruskal-Wallis test. The p value of significance of each pairwise
comparison is shown above the bar. Hemidactylus almakhwah sp. n. was compared with the other species of the H. saba group, and H. farasani
sp. n. with the other species of the H. robustus group. The African species (H. awashensis, H. sp. Etio2, H. sp. Etio3) were excluded from the
comparisons. For character abbreviations see Materials and Methods.
0.0091
7.3e−06
0.00041 4.1e−05
0.0075
0.069
0.077
30
40
50
60
70
80
asirensis granosus alfarraji almakhwah saba ulii awashensis Etio3 Etio2 farasani robustus adensis mandebensis
SVL
0.0091
0.00091
4.1e−05
0.00025
0.00032
0.014
8
12
16
20
asirensis granosus alfarraji almakhwah saba ulii awashensis Etio3 Etio2 farasani robustus adensis mandebensis
HL
0.0091
3.7e−06
0.0045 4.1e−05
0.011
0.06
6
9
12
15
18
asirensis granosus alfarraji almakhwah saba ulii awashensis Etio3 Etio2 farasani robustus adensis mandebensis
HW
0.0091
0.00011
0.00078
0.022
0.092
3
5
7
9
asirensis granosus alfarraji almakhwah saba ulii awashensis Etio3 Etio2 farasani robustus adensismandebensis
HD
0.036
0.001
0.00041
0.0005
0.00018
0.007
2
3
4
5
6
asirensis granosus alfarraji almakhwah saba ulii awashensis Etio3 Etio2 farasani robustus adensis mandebensis
Eye
0.0091
0.00028
0.00016
0.00067
0.0075
0.077
20
30
40
asirensis granosus alfarraji almakhwah saba ulii awashensis Etio3 Etio2 farasani robustus adensis mandebensis
AG
0.0072
0.045
0.011
6
8
10
asirensis granosus alfarraji almakhwah saba ulii awashensis Etio3 Etio2 farasani robustus adensis mandebensis
INF
0.044
0.0011
0.017
7.5
10.0
12.5
15.0
asirensis granosus alfarraji almakhwah saba ulii awashensis Etio3 Etio2 farasani robustus adensis mandebensis
SUP
0.0033
2.7e−07
0.0016 1.6e−05
0.00047
0.00095
0.02
6
8
10
asirensis granosus alfarraji almakhwah saba ulii awashensis Etio3 Etio2 farasani robustus adensis mandebensis
TOE1
0.0061 0.024
4.6e−07
0.0001 7.7e−05
0.00045
0.028
0.0057
10.0
12.5
15.0
17.5
asirensis granosus alfarraji almakhwah saba ulii awashensis Etio3 Etio2 farasani robustus adensis mandebensis
TOE4
Hemidactylus saba species group Hemidactylus robustus species group Hemidactylus saba species group Hemidactylus robustus species group
Supplementary Materials - S9
Supplementary Tables
Table S1. Samples used in this study including information on the sample and voucher numbers, country and locality of origin including GPS
coordinates (datum WGS84), altitude, GenBank accession numbers for the mitochondrial genomes and the seven genes, and MorphoBank
project number and picture accessions of the original high-resolution photographs (accessible at www.morphobank.org). Accession numbers of
sequences generated for this study are in bold.
Species
Sample
Voucher
Type
Country
Locality
Lat
Long
Alt
Mitogenome
12S
Cytb
Cmos
Rag1
Rag2
MC1R
PDC
Morphobank
H. adensis
BJ10
NHM-BS N41907
Paratype
Yemen
Sheikh Othman
12.917
44.983
12
KP238276
KP238262
KP238253
KP238239
KP238233
KP238246
OL653906
Project 1172: M329181–M329189
H. adensis
BJ11
NHM-BS N41906
Paratype
Yemen
Sheikh Othman
12.917
44.983
12
KP238271
KP238261
KP238253
KP238237
KP238233
KP238247
OL653907
Project 1172: M329173–M329180
H. adensis
BJ13
NHM-BS N41902
Paratype
Yemen
Lahij
13.055
44.878
79
KP238274
KP238260
KP238253
KP238243
KP238231
KP238249
OL653908
Project 1172: M329131–M329140
H. adensis
BJ14
NHM-BS N41903
Paratype
Yemen
Lahij
13.055
44.878
79
KP238271
KP238267
KP238255
KP238237
KP238233
KP238251
OL653909
Project 1172: M329141–M329153
H. adensis
BJ15
NHM-BS N41904
Holotype
Yemen
Sheikh Othman
12.917
44.983
12
KP238270
KP238268
KP238253
KP238242
KC819063
KP238245
OL653910
Project 1172: M329154–M329164
H. alfarraji
HSA34
IBES 10335
Saudi Arabia
9 km N of Najran
17.599
44.205
1364
KX263481
KX263645
KX263622
KX263546
KX263521
KX263586
OL653911
Project 2227: M390345–M390354
H. alfarraji
HSA35
NMP 75269
Holotype
Saudi Arabia
32 km W of Najran
17.529
43.827
1969
KX263482
KX263646
KX263624
KX263546
KX263522
KX263587
OL653912
Project 2227: M390464–M390480
H. alfarraji
HSA36
NMP 75270
Paratype
Saudi Arabia
32 km W of Najran
17.529
43.827
1969
KX263483
KX263647
KX263622
KX263546
KX263522
KX263588
OL653913
Project 2227: M390450–M390463
H. alfarraji
HSA37
IBES 10266
Paratype
Saudi Arabia
32 km W of Najran
17.529
43.827
1969
KX263482
KX263646
KX263626
KX263546
KX263523
KX263587
OL653914
Project 2227: M390434–M390449
H. alfarraji
HSA38
IBES 10278
Saudi Arabia
32 km W of Najran
17.529
43.827
1969
KX263484
KX263646
KX263624
KX263546
KX263524
KX263587
OL653915
Project 2227: M390419–M390433
H. alfarraji
HSA40
IBES 10288
Saudi Arabia
32 km W of Najran
17.529
43.827
1969
KX263482
KX263646
KX263624
KX263546
KX263522
KX263587
OL653916
Project 2227: M390393–M390405
H. alfarraji
HSA41
IBES 10295
Paratype
Saudi Arabia
32 km W of Najran
17.529
43.827
1969
KX263482
KX263646
KX263624
KX263546
KX263525
KX263586
OL653917
Project 2227: M390379–M390392
H. alfarraji
HSA42
IBES 10302
Saudi Arabia
32 km W of Najran
17.529
43.827
1969
KX263482
KX263646
KX263624
KX263546
KX263524
KX263590
OL653918
Project 2227: M390367–M390378
H. alfarraji
HSA43
IBES 10303
Paratype
Saudi Arabia
32 km W of Najran
17.529
43.827
1969
KX263482
KX263649
KX263624
KX263546
KX263524
KX263587
OL653919
Project 2227: M390355–M390366
H. almakhwah sp. n.
CN15519
NMP 76091/1
Paratype
Saudi Arabia
Al Makhwah
19.810
41.442
459
OL630166
OL653734
OL653767
OL653807
OL653844
OL653876
OL653922
Project 4024: M823060–M823093
H. almakhwah sp. n.
CN15709
NMP 76091/2
Paratype
Saudi Arabia
Al Makhwah
19.810
41.442
459
OL630168
OL653733
OL653769
OL653809
–
OL653878
OL653923
Project 4024: M823094–M823100
H. almakhwah sp. n.
CN15710
NMP 76091/3
Paratype
Saudi Arabia
Al Makhwah
19.810
41.442
459
OL630169
OL653736
OL653770
OL653810
OL653845
OL653879
OL653924
Project 4024: M823101–M823106
H. almakhwah sp. n.
CN15711
NMP 76091/4
Paratype
Saudi Arabia
Al Makhwah
19.810
41.442
459
OL630170
OL653735
OL653771
OL653811
OL653846
OL653880
OL653925
Project 4024: M823107–M823111
H. almakhwah sp. n.
CN15693
–
Saudi Arabia
Al Ju'aydah
19.657
41.579
482
OL630167
OL653724
OL653768
OL653808
–
OL653877
–
–
H. almakhwah sp. n.
CN15719
NMP 76092/1
Paratype
Saudi Arabia
Al Ju'aydah
19.657
41.579
482
OL630171
OL653727
OL653772
OL653812
OL653847
OL653881
OL653926
Project 4024: M823112–M823116
H. almakhwah sp. n.
CN15720
NMP 76092/2
Saudi Arabia
Al Ju'aydah
19.657
41.579
482
OL630172
OL653730
OL653773
OL653813
–
OL653882
OL653927
Project 4024: M823117–M823118
H. almakhwah sp. n.
CN15129
NMP 76093/1
Paratype
Saudi Arabia
wadi SW of Al Ju'aydah
19.657
41.567
548
OL630158
OL653726
OL653759
OL653799
OL653839
OL653868
OL653920
Project 4024: M822926–M822950
H. almakhwah sp. n.
CN15130
–
Saudi Arabia
wadi SW of Al Ju'aydah
19.657
41.567
548
OL630159
OL653723
OL653760
OL653800
OL653840
OL653869
–
Project 4024: M822951–M822954
H. almakhwah sp. n.
CN15131
–
Saudi Arabia
wadi SW of Al Ju'aydah
19.657
41.567
548
OL630160
OL653728
OL653761
OL653801
–
OL653870
–
–
H. almakhwah sp. n.
CN15140
NMP 76093/2
Holotype
Saudi Arabia
wadi SW of Al Ju'aydah
19.657
41.567
548
OL689327
OL630161
OL653731
OL653762
OL653802
OL653841
OL653871
OL653921
Project 4024: M822955–M822998
H. almakhwah sp. n.
CN15167
NMP 76093/3
Paratype
Saudi Arabia
wadi SW of Al Ju'aydah
19.657
41.567
548
OL630163
OL653725
OL653764
OL653804
–
OL653873
–
Project 4024: M823030–M823038
H. almakhwah sp. n.
CN15166
NMP 76093/6
Paratype
Saudi Arabia
wadi SW of Al Ju'aydah
19.657
41.567
548
OL630162
OL653732
OL653763
OL653803
–
OL653872
–
Project 4024: M822999–M823029
H. almakhwah sp. n.
CN15168
NMP 76093/4
Paratype
Saudi Arabia
wadi SW of Al Ju'aydah
19.657
41.567
548
OL630164
OL653729
OL653765
OL653805
OL653842
OL653874
–
Project 4024: M823039–M823050
H. almakhwah sp. n.
CN15169
NMP 76093/5
Paratype
Saudi Arabia
wadi SW of Al Ju'aydah
19.657
41.567
548
OL630165
OL653722
OL653766
OL653806
OL653843
OL653875
–
Project 4024: M823051–M823059
H. asirensis
HSA1
IBES 10021
Saudi Arabia
20 km NE of Al Sir
21.259
40.796
1594
KX263485
KX263650
KX263627
KX263548
KX263526
KX263591
OL653928
Project 2227: M390088–M390101
H. asirensis
HSA12
NMP 75272
Paratype
Saudi Arabia
5 km N of Wadi Shora
19.836
41.776
1750
KX263486
KX263651
KX263630
KX263549
KX263527
KX263592
OL653931
Project 2227: M390048–M390066
H. asirensis
HSA13
–
Saudi Arabia
Khathaam
19.779
41.905
1801
KX263487
KX263652
KX263634
KX263550
KX263526
KX263592
OL653932
–
H. asirensis
HSA2
IBES 10044
Paratype
Saudi Arabia
20 km NE of Al Sir
21.259
40.796
1594
KX263485
KX263650
KX263627
KX263548
KX263526
KX263593
OL653929
Project 2227: M390067–M390077
H. asirensis
HSA44
NMP 75271
Holotype
Saudi Arabia
Al Balhy
18.075
43.083
2376
KX263489
KX263655
KX263629
KX263553
KX263526
KX263592
OL653933
Project 2227: M389997–M390014
H. asirensis
HSA5
IBES 10030
Saudi Arabia
10 km S of Al Sir
21.115
40.599
1696
KX263490
KX263656
KX263631
KX263554
KX263529
KX263596
OL653930
Project 2227: M390078–M390087
H. asirensis
HSA52
IBES 10221
Paratype
Saudi Arabia
7 km S of Ghazaial
20.928
41.123
1453
KX263491
KX263657
KX263633
KX263555
KX263530
KX263597
OL653934
Project 2227: M390031–M390047
H. asirensis
CN15057
NMP 76094/1
Saudi Arabia
Lake Tendaha
18.312
42.886
2046
OL630173
–
–
–
–
–
–
Project 4024: M823119–M823126
H. asirensis
CN15088
NMP 76095/1
Saudi Arabia
Wadi Bishah
19.712
42.482
1328
OL630174
–
–
–
–
–
–
Project 4024: M823158–M823165
H. asirensis
CN15096
NMP 76096/1
Saudi Arabia
Wadi Tharad
20.158
41.722
1506
OL630175
–
–
–
–
–
–
Project 4024: M823175–M823180
H. asirensis
CN15126
NMP 76097/1
Saudi Arabia
Alshorouk
20.079
41.569
1935
OL630176
–
–
–
–
–
–
Project 4024: M823195–M823197
H. asirensis
CN15148
NMP 76098
Saudi Arabia
Taif
21.335
40.430
OL630177
–
–
–
–
–
–
Project 4024: M823198–M823206
H. asirensis
CN15155
NMP 76099
Saudi Arabia
7 km S of Ben Faraj Shalawi
20.557
41.288
1507
OL630178
–
–
–
–
–
–
Project 4024: M823207–M823215
H. asirensis
CN15156
NMP 76097/2
Saudi Arabia
Alshorouk
20.079
41.569
1935
OL630179
–
–
–
–
–
–
Project 4024: M823216–M823222
H. asirensis
CN15160
NMP 76100/1
Saudi Arabia
Al Qara
20.203
41.370
2264
OL630180
–
–
–
–
–
–
Project 4024: M823223–M823227
H. asirensis
CN15161
NMP 76100/2
Saudi Arabia
Al Qara
20.203
41.370
2264
OL630181
–
–
–
–
–
–
Project 4024: M823228–M823236
H. asirensis
CN15183
NMP 76101
Saudi Arabia
Wildlife Research Center, Taif
21.256
40.696
1609
OL630182
–
–
–
–
–
–
Project 4024: M823237–M823250
H. asirensis
CN15223
NMP 76102
Saudi Arabia
Saiysad National Park, Taif
21.294
40.494
1593
OL630183
–
–
–
–
–
–
Project 4024: M823251–M823261
H. asirensis
CN15058
NMP 76094/2
Saudi Arabia
Lake Tendaha
18.312
42.886
2046
–
–
–
–
–
–
–
Project 4024: M823127–M823129
H. asirensis
CN15060
NMP 76094/3
Saudi Arabia
Lake Tendaha
18.312
42.886
2046
–
–
–
–
–
–
–
Project 4024: M823130–M823136
H. asirensis
CN15061
NMP 76094/4
Saudi Arabia
Lake Tendaha
18.312
42.886
2046
–
–
–
–
–
–
–
Project 4024: M823137–M823145
H. asirensis
CN15062
NMP 76094/5
Saudi Arabia
Lake Tendaha
18.312
42.886
2046
–
–
–
–
–
–
–
Project 4024: M823146–M823150
H. asirensis
CN15063
NMP 76094/6
Saudi Arabia
Lake Tendaha
18.312
42.886
2046
–
–
–
–
–
–
–
Project 4024: M823151–M823157
H. asirensis
CN15089
NMP 76095/2
Saudi Arabia
Wadi Bishah
19.712
42.482
1328
–
–
–
–
–
–
–
Project 4024: M823166–M823174
H. asirensis
CN15097
NMP 76096/2
Saudi Arabia
Wadi Tharad
20.158
41.722
1506
–
–
–
–
–
–
–
Project 4024: M823181–M823187
H. asirensis
CN15098
NMP 76096/3
Saudi Arabia
Wadi Tharad
20.158
41.722
1506
–
–
–
–
–
–
–
Project 4024: M823188–M823194
H. awashensis
JS204
–
Ethiopia
Metehara
8.921
39.903
990
KP238272
KP238258
KC818786
KP238240
KC819058
KC818939
–
–
H. awashensis
JS212
NMP 74977
Paratype
Ethiopia
Metehara
8.921
39.903
990
KC818723
KC818873
KC818786
KC818998
KC819058
KC818939
OL653935
Project 1172: M328897–M328937
H. awashensis
JS242
TMHC 2012.07.072
Ethiopia
Metehara
8.908
39.912
964
KC818723
KP238266
KC818786
–
KC819058
KC818939
–
–
Supplementary Materials - S10
Species
Sample
Voucher
Type
Country
Locality
Lat
Long
Alt
Mitogenome
12S
Cytb
Cmos
Rag1
Rag2
MC1R
PDC
Morphobank
H. awashensis
JS243
TMHC 2012.07.073
Ethiopia
Metehara
8.908
39.912
964
KC818723
KP238263
KC818786
–
KP238230
KC818939
–
–
H. awashensis
JS244
NMP 74978/1
Paratype
Ethiopia
Metehara
8.908
39.912
964
KP238269
KP238264
KC818786
–
KC819058
KC818939
–
Project 1172: M328942–M328968
H. awashensis
JS245
NMP 74978/2
Paratype
Ethiopia
Metehara
8.908
39.912
964
KC818723
KP238259
KC818786
–
KC819058
KC818939
–
Project 1172: M328969–M329004
H. awashensis
JS246
TMHC 2012.07.076
Ethiopia
Metehara
8.908
39.912
964
KC818723
KP238257
KC818786
–
KC819058
–
–
–
H. awashensis
JS247
NMP 74978/3
Paratype
Ethiopia
Metehara
8.908
39.912
964
KC818723
KP238263
KC818786
–
KP238230
KC818939
–
Project 1172: M329005–M329042
H. awashensis
JS248
TMHC 2012.07.079
Ethiopia
Metehara
8.923
39.905
990
KP238275
KP238257
KC818786
–
KC819058
KC818939
–
–
H. awashensis
JS249
NMP 74979
Holotype
Ethiopia
Metehara
8.922
39.912
981
KC818723
KP238266
KC818786
–
KC819058
KC818939
OL653936
Project 1172: M329043–M329085
H. awashensis
JS250
NMP 74980
Paratype
Ethiopia
Metehara
8.931
39.905
1020
KP238273
KP238266
KC818786
–
KC819058
KC818939
–
Project 1172: M329086–M329118
H. farasani sp. n.
CN15589
NMP 76103
Paratype
Saudi Arabia
Farasan Island
16.702
42.055
7
OL630194
OL653749
OL653784
OL653824
OL653858
OL653892
–
Project 4024: M823334–M823339
H. farasani sp. n.
CN15570
NMP 76104/1
Paratype
Saudi Arabia
Sajid Island
16.904
41.898
3
OL630185
OL653745
OL653775
OL653815
OL653849
OL653883
–
Project 4024: M823262–M823284
H. farasani sp. n.
CN15571
NMP 76104/2
Holotype
Saudi Arabia
Sajid Island
16.904
41.898
3
OL689328
OL630186
OL653746
OL653776
OL653816
OL653850
OL653884
–
Project 4024: M823285–M823314
H. farasani sp. n.
CN15572
NMP 76104/3
Saudi Arabia
Sajid Island
16.904
41.898
3
OL630187
OL653742
OL653777
OL653817
OL653851
OL653885
OL653938
Project 4024: M823315–M823316
H. farasani sp. n.
CN15575
NMP 76104/4
Paratype
Saudi Arabia
Sajid Island
16.904
41.898
3
OL630188
OL653747
OL653778
OL653818
OL653852
OL653886
OL653939
Project 4024: M823317
H. farasani sp. n.
CN15576
NMP 76104/5
Paratype
Saudi Arabia
Sajid Island
16.904
41.898
3
OL630189
OL653748
OL653779
OL653819
OL653853
OL653887
OL653940
Project 4024: M823318–M823320
H. farasani sp. n.
CN15577
NMP 76104/6
Paratype
Saudi Arabia
Sajid Island
16.904
41.898
3
OL630190
OL653743
OL653780
OL653820
OL653854
OL653888
OL653941
Project 4024: M823321–M823323
H. farasani sp. n.
CN15578
NMP 76104/7
Paratype
Saudi Arabia
Sajid Island
16.904
41.898
3
OL630191
OL653741
OL653781
OL653821
OL653855
OL653889
OL653942
Project 4024: M823324–M823328
H. farasani sp. n.
CN15580
NMP 76104/8
Paratype
Saudi Arabia
Sajid Island
16.904
41.898
3
OL630192
OL653750
OL653782
OL653822
OL653856
OL653890
OL653943
Project 4024: M823329–M823330
H. farasani sp. n.
CN15582
NMP 76104/9
Paratype
Saudi Arabia
Sajid Island
16.904
41.898
3
OL630193
OL653744
OL653783
OL653823
OL653857
OL653891
OL653944
Project 4024: M823331–M823333
H. farasani sp. n.
CN15568
–
Saudi Arabia
Farasan Island
16.826
41.848
8
OL630184
OL653738
OL653774
OL653814
OL653848
–
OL653937
–
H. farasani sp. n.
CN15591
NMP 76105/1
Paratype
Saudi Arabia
Farasan Island
16.826
41.848
8
OL630195
OL653739
OL653785
OL653825
OL653859
OL653893
OL653945
Project 4024: M823340–M823344
H. farasani sp. n.
CN15608
NMP 76105/2
Paratype
Saudi Arabia
Farasan Island
16.826
41.848
8
OL630197
OL653737
OL653787
OL653827
OL653861
OL653895
OL653947
Project 4024: M823350–M823382
H. farasani sp. n.
CN15609
NMP 76105/3
Paratype
Saudi Arabia
Farasan Island
16.826
41.848
8
OL630198
OL653751
OL653788
OL653828
OL653862
OL653896
OL653948
Project 4024: M823383–