Abstract
New findings from two Central Libyan oases extend the distribution area of the Saharan gecko Tarentola neglecta considerably to the East. Scale counts falling outside the range of the two known subspecies show that the new specimens belong to a third subspecies which is described here under the name Tarentola neglecta lanzai. A molecular phylogeny with mitochondrial genes presents further evidence that T. neglecta is the sister species of the Egyptian T. mindiae, both of which constitute a monophyletic clade within the T. fascicularis-deserti group.
Introduction
Tarentola neglecta Strauch, 1895, is a morphologically distinct Saharan desert species of Tarentola, characterized by partially reduced digital lamellae, a more slender shape of head and body and lower scale counts than other Tarentola species (Joger, 1984). It is known from sandy deserts with some bushy vegetation and oases in Algeria, Tunisia and extreme western Libya (see Joger, 2003). The nominate subspecies is found around the northern margin of the Grand Erg Oriental. The subspecies T. neglecta geyri Joger, 1984 was described from the area south of the Grand Erg Oriental in southern Algeria. A specimen recorded from Tchad (Wake and Kluge, 1961) was a misidentified T. ephippiata (Joger, 1984). The only known Libyan specimen so far is a T. n. neglecta from Mizda, Tripolitania (BMNH 1954/1.5.50). Here we describe a new subspecies from Central Libya.
Material and methods
Field work and studied specimens
Geckos were collected in Libya between September and November 2008, concentrating on the south-west of the country (Fezzan). The collection area covers at least 21 000 km2, from Bin Ulid in the North up to Ghat and Elperkat in the extreme south-western part of the country. 18 Tarentola specimens were collected during this trip, the specimens were collected from 13 different localities in the Libyan Sahara desert, and represent different species of the genus Tarentola. Three specimens, from Mandria and Gaber Aown Oases, at the Eastern margin of the Murzuq sand sea (fig. 1), were identified as Tarentola neglecta. 299 Tarentola specimens (mainly T. fascicularis) had been collected the year before in northern Libya, and could be used for comparison (fig. 2).
General habitat of Tarentola neglecta lanzai n. ssp. at its type locality, Mandria Oasis, Libya. Photo: I. Bshaena. This figure is published in colour in the online version.
Citation: Amphibia-Reptilia 34, 3 (2013) ; 10.1163/15685381-00002913
Localities of Tarentola collected in Libya. Localities of T. n. lanzai n. ssp. marked by an ellipse. This figure is published in colour in the online version.
Citation: Amphibia-Reptilia 34, 3 (2013) ; 10.1163/15685381-00002913
In addition to the Libyan specimens, Moroccan, Algerian and Tunisian Tarentola specimens from the following collections were studied:
ZFMK – Zoologisches Forschungsinstitut Museum Koenig, Bonn, Germany.
SHNM-BS – Staatliches Naturhistorisches Museum, Braunschweig, Germany.
MNHN – Muséum National d’Histoire Naturelle, Paris, France.
SMF – Senckenberg Forschungsmuseum Frankfurt/M, Frankfurt, Germany.
HLMD – Hessisches Landesmuseum Darmstadt, Darmstadt, Germany.
ZIBU – Zoology Institute Banghazi University, Banghazi, Libya.
BMNH – Natural History Museum, London, Britain.
BEV – University of Montpellier, Montpellier, France.
Preparation of voucher specimens for morphological study
Notes on life coloration were taken from all freshly collected Tarentola specimens. The specimens were then euthanized by injection of an overdose of the anaesthetic agent chlorobutanol (some chlorobutanol crystals dissolved in water and little amount of absolute ethanol), about 5 ml in the lizard’s abdominal region. This is a standard technique for painless euthanizing reptiles and amphibians (McDiarmid, 2012). Immediately after death, tissue samples were taken for molecular study. Vouchers were fixed by injection with absolute ethanol in the body cavity and muscles, then the whole animal was immersed in a bowl containing 90-100% ethanol (absolute ethanol), left in the bowl from 1-2 hours, and subsequently put into plastic jars with 75% ethanol to preserve. The voucher specimens are housed in the herpetological collection of Staatliches Naturhistorisches Museum Braunschweig (SNHM-BS), Germany.
Characters studied
A range of external morphological measurements relating to body proportions, scalation, and color patterns were taken from each specimen, whenever possible. All measurements were taken multiple times until two or three consecutive readings were in agreement.
Morphometric characters
The following linear measurements on the body were taken from each specimen, whenever possible. All linear measurements were taken with a digital caliper (measurement range: 0 mm-150 mm, input-unit: 0.01 mm, accuracy: 0.03 mm, repeatability: 0.01 mm), by the first author to avoid inter-observer variability; used abbreviations and designations from Joger (1984) are:
- – Snout-vent length (SVL), calculated from snout tip to anterior end of cloaca (in mm).
- – Tail length (TL), calculated from anterior end of cloaca to tip of tail (in mm), only from original tails. Tail measurements from individuals with regenerated or broken tails were not included in statistics.
- – Head length (HL), the distance on mid-dorsal line between the tip of the snout and the line connecting the rear tips of the mandibles (in mm).
- – Head width (HW), the distance at the widest part of the head (in mm).
- – Head height (HH), measured as: perpendicular to body longitudinal axis at eye level (in mm).
- – Forelimb length (FLL), from the axillar ring to the end of the fourth finger (in mm).
- – Hindlimb length (HLL), from the groin ring to the end of the fourth toe (in mm).
- – Eye diameter (ED), the longest diameter of the visible part of the eye (in mm).
- – FHLL, distance between forelimb and hindlimb (in mm).
- – (HDL), length of fourth toe, counted from their point of attachment near the third toe up to the end of toe (in mm), not including claw.
- – (HDW), maximum width of the fourth toe (in mm).
- – (ASL), distance from the front edge of eye up to snout tip (in mm).
- – (IOB), distance between the eyes, counted from the closest point (in mm).
- – (AOL), distance from the posterior edge of the eye to the ear opening (in mm).
- – (MOL), distance from the middle point of the imaginary line between ear openings up to the mental (in mm), for details see Joger (1984).
Scalation characters
The following scalation features on the body were taken from each specimen:
- – 1st toe, the number of subdigital lamellae underneath the first toe, counted up to the toe base.
- – 4th toe, the number of subdigital lamellae underneath the fourth toe, counted up to the toe base.
- – 5th toe base, the number of subdigital lamellae underneath the fifth toe, counted up to the toe base.
- – The number of gular scales, counted as the number of scales from mental to the middle-point of an imaginary line between the ears.
- – Interorbital scales, the number of scales between the eyes, counted from the closest point.
- – Body scales, the number of scales around the middle body (including ventral scales).
- – Ventral scales, the number of ventral scales in a transverse row at midbody.
- – Infralabials, the number of scales for one side starting from the angle of the mouth to the mental, not including the latter.
- – Supralabials, the number of scales for one side starting from the angle of the mouth to the rostral scale, not including the latter.
- – The number of dorsal tubercles (tubercles = larger scales), in a transverse row at the body’s center (a tubercle row, fusing with another row, is counted as 1∕2).
Operational taxonomic units (OTUs) were defined using a combination of mitochondrial DNA clades (see below) and geographic proximity. Linear Discriminant Function Analyses (LDFA) were used to find variables that separate the groups. Principal Component Analyses (PCA) were applied to see whether groups are distinguishable without previous definition of OTUs. Significance of character differences were tested with t-tests. Previous to analysis, mensural characters (morphometric characters) were used in their transformed state (relative length); other characters (scalation characters) were used in original state (unchanged).
Molecular techniques
DNA was extracted from the preserved samples using a standard extraction protocol (modified after Bruford et al., 1992). Universal primers were used to amplify mitochondrial 12S rRNA (372 bp) and 16S rRNA (two fragments of 448 and 604 bp), as well as parts of three nuclear marker genes, c-mos (424 bp), RAG2 (387 bp), and Phosducin (361 bp). PCR reaction was performed in a Biometra T-Gradient thermocycler with different temperature profiles depending on the primers and the target fragment. For the rRNAs the following program was used: [94°C 90 s, 94°C 45 s, 52°C 45 s, 72°C 90 s, 72°C 5 min] for 33 cycles. PCR products were electrophoresed in a 1.0% agarose gel containing ethidium bromide and visualized under ultraviolet light. For the subsequent sequencing reactions (10 μl total volume) between 2 and 3 μl of cleaned PCR product were used with 0.5 μl BigDye 3.1 and 0.3 μM primer. Following an initial denaturation for 1 min at 96°C 25 cycles followed by 10 s at 96°C, a 5 s annealing step at 50°C and a 4 min extension at 60°C. Products were cleaned by adding 1 μl of a solution containing 1.5 M sodium acetate and 250 mM EDTA (pH 8) and precipitated with a fourfold volume of 95% ethanol during a 45 min centrifugation step at 1500 g. The dried samples were eluted with 10 μl HiDye before run on an automated sequencer.
Sequences were determined using a 3130XL automated sequencer (Applied Biosystems), and edited and aligned using Clustal-W, as implemented in the program package Codon Code Aligner, ver. 2.0.6, and in MEGA 4.0 (Kumar et al., 2008). Sequences from Genbank were added in some cases. Phylogenetic trees were constructed using a Bayesian Markov Chain Monte Carlo inference applying an evolution model suggested by MODELTEST 3.7 (Ronquist and Huelsenbeck, 2003). Statistical support for branches was indicated by posterior probability values (MrBayes). For a more detailed description of the methods see Busais and Joger (2011). The Bayesian analysis was run with four parallel chains for 1.5 million generations; Markov chains sampled every 100 generations and discarding the first 150 000 generations. Tarentola delalandei from Tenerife, and T. boettgeri from Gran Canaria, Canary Islands, were used as an outgroup. Trees were first constructed separately for mitochondrial genes and nuclear genes combined, respectively.
Results and discussion
Molecular genetic affinities
The Bayesian tree based on 1172 bp nuclear sequences (not shown) had a comb-like shape showing no discernible clades except four paraphyletic ones. One of them was composed of T. n. neglecta (Algeria), T. mindiae (Egypt), and the specimens from Mandria and Gaber Aown Oases in Libya.
In the Bayesian tree based on 1433 bp mitochondrial sequences (12S rRNA, 16S rRNA) a number of distinct clades are obvious (fig. 3). Four of them (clades D, E, G and H) constitute a monophyletic group, and are best characterized as (undescribed) subspecies of T. fascicularis. Depending on the species concept applied, clades D, E, and G + H could also be regarded as separate species. The uncorrected ‘P’ distance between these clades amounts to values between 0.05 and 0.08; their distance from the neglecta clade B lies between 0.07 and 0.1. Clade B includes T. mindiae as well as the Libyan and Algerian T. neglecta. Table 1 shows the intra clade distances of clade B. Clade E has already been described as subspecies T. fascicularis boehmei Joger and Bshaenia, 2010 (Genbank accession number HQ437282). As the description of the other subspecies is in progress, but not yet published, their names and Genbank data are not given here. Genbank accession nos. of the neglecta and mindiae sequences are: KF425708 for Tarentola n. neglecta, El Oued, Algeria, 12S rRNA; KF425709 and KF425710 for T. mindiae, Egypt, 12S rRNA (two specimens), KF425711 and KF425712 for T. neglecta lanzai, southern Libyan oases, 12S rRNA (two specimens); KF425698-702 for the above named five specimens, 16S rRNA 1st fragment; KF425703-707 for the above named five specimens, 16S rRNA 2nd fragment.
50% majority-rule consensus tree obtained from Bayesian MCMC analysis, based on 1433 bp mtDNA sequences, depicting the relationships among haplotypes; with Tarentola delalandii delalandii designated as outgroup. Bayesian posterior probability values are given near branches. Clade of Tarentola neglecta and T. mindiae is Clade B. This figure is published in colour in the online version.
Citation: Amphibia-Reptilia 34, 3 (2013) ; 10.1163/15685381-00002913
The putative T. neglecta does indeed cluster with T. n. neglecta from Algeria, and T. mindiae from Egypt is sister to both of them (clade B in fig. 3). This is reinforced by the fact that the less variable nuclear genes also distinguish this clade. It is also in line with the phylogeny of Rato et al. (2012) who did not have Libyan neglecta at their disposition but likewise found a sistergroup relationship of neglecta and mindiae. Together these two species constitute a distinct clade within the T. fascicularis-deserti group, which is sister to the T. mauritanica group. As we did not have a sample of T. neglecta geyri available for comparison, it is not possible to use the molecular data to judge whether the Libyan neglecta represents a distinct subspecies; this is inferred from morphological data alone.
Uncorrected P distance between main clades, estimated of evolutionary divergence between sequences, based on pairwise analysis of 1127 bp combined mtDNA (12S rRNA and two fragments of 16S rRNA) sequences.
Scale counts and major measurements compared between the subspecies of Tarentola neglecta (mean ± standard deviation). The three specimens of T. n. lanzai are listed individually.
Morphological comparisons
The three specimens from Mandria and Gaber Aown Oases contrast morphologically with other Libyan species except T. neglecta. They are small-sized like neglecta (with 35 mm SVL even smaller than T. n. neglecta from the northern Sahara and T. n. geyri from the Algerian central Sahara which may reach about 50 mm SVL), having a neglecta-like dorsal pattern with dark parallel lines on top of the head and in the dorsal region, and scale counts which are generally lower than in sympatric T. cf. fascicularis: Scale rows around midbody count 128-155 in fascicularis, yet only 72-110 in neglecta, but they are in higher in Libyan neglecta than in western neglecta (see table 2). An exception is seen in ventral scales which are more numerous in neglecta than in fascicularis (see below). Therefore, neglecta generally have a higher difference in size between dorsal and ventral scales than other Tarentola species.
PCA of Libyan Tarentola. Nine OTUs of Libyan Tarentola populations, only males (T. deserti, T. fascicularis, T. sp., [5 distinct populations within the T. fascicularis-deserti-complex], and T. neglecta – Gaber-Aown Oasis Lake) were plotted together. The first and second principal components cover 43.49% of total variance.
Citation: Amphibia-Reptilia 34, 3 (2013) ; 10.1163/15685381-00002913
The Principal Component Analysis (fig. 4) distributes the Libyan samples on a plane defined by the two first axes which covers 43.49% of the total variability. The first component was successful in separating groups T. sp.-Sabha and T. neglecta from the remaining groups, the populations of T. sp.-Sabha and T. neglecta appear on the negative side of the axis. An ANOVA test (not shown) shows the existence of significant differences among different groups in this analysis, T. sp.-Sabha and T. neglecta both having a higher number of ventral scales: 40-43, 42-43 respectively (versus 33-40 in T. f. fascicularis). The second component was successful in separating T. sp.-Sabha from T. neglecta; T. neglecta having a very slim head (relative head breadth in male T. neglecta 0.6-0.7 of head length), and 18 or fewer lamellae underneath the fifth toe, whereas T. sp. from Sabha has a broader head (0.7-0.9 in male T. sp.-Sabha), and 21-23 lamellae under the fifth toe.
Conclusions
The molecular analysis makes clear that the three dwarfish geckos from southern Libyan oases that we deal with are indeed Tarentola neglecta. It does however not provide evidence of their distinctiveness from other neglecta. However, the morphological differences of the southern Libyan neglecta from both known subspecies are evident. Regarding the following characters, the three specimens were found to be outside of the known variation of at least one of the existing subspecies, respectively:
- A) Differences from T. n. neglecta: 15 subdigital lamellae under fourth toe versus 12-14 in T. n. neglecta; 18 subdigital lamellae under fifth toe versus 12-14 in T. n. neglecta; 13 interorbital scales versus 11-12 in T. n. neglecta.
- B) Differences from T. n. geyri: 13-14 dorsal tubercle rows versus more than 16 in T. n. geyri; dorsal scales much smaller than ventral scales (subequal in T. n. geyri). In addition there are color differences.
- C) Differences from both neglecta subspecies: number of scales around midbody 104-110 versus less than 103 in T. n. neglecta and T. n. geyri; number of gular scales 42-52 versus less than 42 in both known subspecies. The head of the Libyan specimens is somewhat broader than in the existing subspecies (table 1).
The known distribution area of T. neglecta is now extending further towards its sister species, T. mindiae in the Egyptian desert, albeit still leaving a considerable gap between the two species, as well as a narrower but significant gap between T. neglecta geyri from Algeria and the Libyan neglecta.
We believe that when a geographically separate population falls outside the range of known subspecies, it deserves the status of a separate subspecies. On the other hand, as T. mindiae is the sister species of both T. n. neglecta and the Libyan form, it also makes sense not to regard the latter two as different species. Therefore we here describe:
Tarentola neglecta lanzai n. ssp.
Holotype
SNHM-BS N 44084, an adult female, collected by Ismail M. Bshaena in October 2008.
(A) Tarentola neglecta lanzai n. ssp., paratype SNHM-BS N 44085. Label shows the initial collection number. Photo: I. Bshaena. (B) Tarentola n. neglecta, southern Tunisia. Photo: U. Joger. This figure is published in colour in the online version.
Citation: Amphibia-Reptilia 34, 3 (2013) ; 10.1163/15685381-00002913
Terra typica
Mandria Oasis (Libyan Sahara desert, N 26°45′40.73″, E 13°25′37.64″, 538 m above sea level), at ruins of an abandoned building.
Paratypes
SNHM-BS N 44085, N 44086, two subadult specimens, from Gaber Aown Oasis (Southwest Libya), collected by Ismail M. Bshaena on trunks of palm trees (fig. 5a).
Description
A small-sized gecko (SVL in adult individuals about 35 mm), very slim, with very long tail (tail length about 15% longer than SVL), head length 30-40% of SVL, head width 56-79% of head length, head height 45-47% of head length; eye diameter about 30% of head length. Dorsum covered by enlarged oval-shaped, simply keeled tubercles arranged in 13-14 transversal rows at midbody; ventral scales smooth, large in size when compared to dorsal scales, forming 40-43 (n = 3) rows at midbody; interorbital scales 12-13; 10 lamellae underneath the first toe (n = 3), 15 lamellae underneath the fourth toe (n = 3), and 18 lamellae underneath the fifth toe (n = 3).
Dorsal color is light grey-brown, with five often interrupted, parallel brown longitudinal stripes along the dorsum. The central stripe and the two lateral of these five stripes are less distinct than the two stripes in between, and tend to dissolve into lines of little streaks (---). The two more distinct stripes continue anteriorly onto the sides of the head. There they proceed above the ear openings to the eye, and from the anterior margin of the eye to tip of the snout, below the nostril. Two slightly s-shaped marks occupy on the upper side of the snout and the interorbital space. From the interorbital space to the neck, a central stripe is present as a quasi-continuation of the central dorsal stripe. Laterally, two parallel dark wavy lines are placed between axilla and groin. On the tail, 12 transverse bands are present.
The described pattern is most obvious in the two subadult paratypes; in the adult holotype the pattern is fading.
Diagnosis
Tarentola neglecta lanzai has a small size (body length 34.76 mm in the only adult of three known specimens versus 51 mm in female T. n. neglecta and 50 mm in female T. n. geyri). This subspecies has a higher number of scale rows around midbody (104-110, versus 72-98 in T. n. neglecta, and 93-102 in T. n. geyri), whereas there are less rows of dorsal tubercles in T. n. lanzai (13-14 versus 13-17 in T. n. neglecta, and 16-18 in T. n. geyri); dorsal tubercles with a single keel, no associate tubercles or transitional scales; tendency towards a higher number of subdigital lamellae under the fourth toe (15 versus 12-14 in T. n. neglecta, and 14-15 in T. n. geyri), and a significantly higher number of subdigital lamellae under the fifth toe (18, versus 13-15 in T. n. neglecta, and 15-18 in T. n. geyri, whilst the Egyptian T. mindiae have an average of 16 subdigital lamellae under the fifth toe). Ventral and lateral scales are much larger than dorsal scales, as in T. n. neglecta, whereas ventral and dorsal scales are of similar size in T. n. geyri. Gular scales are smaller and thus more numerous in T. n. lanzai (42-52, versus 29-41 in T. n. neglecta, and 38-40 in T. n. geyri), interorbital scales (13-14) overlap with the range of T. n. geyri (12-13 versus 11-12 in T. n. neglecta).
Coloration light grey-brown, with a pattern of dark brown near-parallel streaks on the dorsal side, from the top of head to the base of the tail. This pattern is similar to that of other T. neglecta (fig. 5b), while T. mindiae have a light brown color, with a pattern of five-six blackish bands across the back between occiput and sacrum, and two dark near parallel lines running from the rostral to the interorbital region (Baha El Din, 2006). T. neglecta geyri also has a more brownish color in life.
Description of holotype
Snout-vent length 34.76 mm; head length 13.91 mm, head width 10.98 mm, head height 6.62 mm; eye diameter 3.33 mm; ASL 5.93 mm, AOL 4.88 mm, MOL 13.04 mm, IOB 6.51 mm, FLL 18.49 mm, HLL 24.13 mm, HDL 4.22 mm, HDW 1.98 mm, FHLL 25.38 mm; 7 infralabials, 8 supralabials on the right side, 9 on the left side. 13 interorbital scales; dorsal tubercles in the axilla-groin distance 42; 110 scales around midbody; 10 lamellae underneath the first toe, 15 lamellae underneath the fourth toe, 18 lamellae underneath the fifth toe; 44 gular scales from mental scale to gular fold; 3 gular scales in contact with the mental. Nostril in contact with the rostral, the first labial and three other scales. Eye opening in contact with the 7th to 8th (9th) supralabial. Above the eye a first ring of small, projecting (mildly spiny-like) scales, and a second ring of large supraorbital scales: 4 on the right side, 5 on the left side.
Dorsal tubercles are simply, centrally keeled with black spots, not surrounded by a row of secondary tubercles, separated laterally by 1-2 small-sized scales.
Coloration: Basic color of dorsum light grey-brown, with thin dark brown, near-parallel lines of streaks, running from the snout to the sacral region, and linked on the snout to form a V-shape. No central dorsal stripe visible, as in the paratypes. The ventral side is whitish, with little dark spots on each scale.
The tail of the holotype has been freshly autotomized.
Variation of the paratypes
Table 1 shows that in the three specimens known, there is very little variation in scale counts. Only in the gular scale count, we found values between 42 and 52. Scales around midbody were counted 104 and 106. Measurements of the two paratypes are (in mm): SVL 28.1/26.0; TL 37.1/6.6 (partly regenerated) (the generally very long tail of T. neglecta is thus confirmed); HL 8.6/8.3; HW 5.9/6.1; HH 3.9/3.8; FLL 9.8/9.1; HLL 13.9/13.5; ED 2.4/2.2; FHLL 12.5/13.8; HDL 2.4/1.8; HDW 1.2/1.2; ASL 3.6/3.6; IOB 3.8/3.8; MOL 8.3/8.1. The coloration of the paratypes is identical to the description above.
Ecology
The macro-environment of these geckos is sandy desert with dunes – typical for T. neglecta in general. As far as we now, bushy vegetation is chosen as micro-habitat. In oases, they also settle on palm trees and walls of old buildings. Activity is nocturnal.
Derivatio nominis
We name this taxon in honor of Professor Benedetto Lanza, nestor of Italian herpetology, who has invaluable merits in many fields of research on Mediterranean and African fauna.
Acknowledgements
We thank Sherif Baha El Din, Philippe Geniez and Adel Ibrahim for providing specimens, as well as Miguel Vences and Susanne Hauswaldt for providing lab facilities. We also thank two anonymous reviewers for their helpful comments on the manuscript.
References
Baha El Din S. (1997): A new species of Tarentola (Squamata: Gekkonidae) from the western desert of Egypt. Afr. J. Herpetol. 46: 30-35.
Baha El Din S. (2006): A guide to the Reptiles and Amphibians of Egypt. American Univ. Cairo Press.
Bruford M.W., Hanotte O., Brokfield J.F.Y., Burke T. (1992): Single-locus and multilocus DNA fingerprinting. In: Molecular Genetics of Populations: a Practical Approach, p. 225-269. Hoelzel A.R., Ed., Oxford University Press, New York.
Busais S., Joger U. (2011): Molecular phylogeny of the gecko genus Hemidactylus Oken, 1817 on the mainland of Yemen. Zool. Middle East 53: 25-34.
Joger U. (1984): Taxonomische Revision der Gattung Tarentola (Reptilia: Gekkonidae). Bonn. Zool. Beitr. 35: 129-174.
Joger U. (2003): Reptiles and amphibians of southern Tunisia. Kaupia 12: 71-88.
Joger U., Bshaena I. (2010): A new Tarentola subspecies (Reptilia: Gekkonidae) endemic to Tunisia. Bonn. Zool. Bull. 57: 267-274.
Joger U., Busais S. (2011): Molecular phylogeny of the gecko genus Hemidactylus Oken, 1817 on the mainland of Yemen. Zool. Middle East 53: 25-34.
Kumar S., Nei M., Tamura K. (2008): MEGA: A biologist-centric software for evolutionary analysis of DNA and protein sequences. Briefings in Bioinformatics 9: 299-306.
McDiarmid R.W., Foster M.S., Guyer C., Gibbons J.W., Vhernoff N. (2012): Reptile Biodiversity. Standard Methods for Inventory and Monitoring. Univ. California Press, Berkeley.
Rato C., Carranza S., Harris D.J. (2012): Evolutionary history of the genus Tarentola (Gekkota: Phyllodactylidae) from the Mediterranean Basin, estimated using multilocus sequence data. BMC Evol. Biol. 12: 1-12. Article No. 14.
Ronquist F., Huelsenbeck J.P. (2003): MRBayes3: Bayesian phylogenetic inference underixed models. Bioinformatics 19: 1572-1574.
Wake D.B., Kluge A.G. (1961): The Machris expedition to Tchad, Africa. Amphibians and Reptiles. Contributions to Science, L.A. County Mus. 40: 1-12.
Footnotes
Associated Editor: Sebastian Steinfartz
