We here describe a new species of feather-tailed leaf-toed gecko, Kolekanos, from southern Benguela Province, Angola, based on morphological and osteological evidence, supported by phylogenetic analysis of mitochondrial data. The new species adds to the rapidly growing and newly-recognised endemic biodiversity of Angola, doubling the number of Kolekanos species, breaking the pattern observed within other closely-related African members of a clade of circum-Indian Ocean leaf-toed geckos – Ramigekko, Cryptactites and Afrogecko – all of which are presently monotypic. The new species is easily distinguished from K. plumicaudus, based on spine-like (as opposed to feather-like) scales on the margins of the original tail. Phylogenetic analyses also recovered the new taxon as monophyletic, with a well-supported sister relationship to K. plumicaudus, from which it differs by a substantial 24.1% NADH-dehydrogenase subunit 2 mitochondrial gene uncorrected p-distance.

Biodiversity, ct-scan, herpetology, osteology, Reptilia, taxonomy

African leaf-toed geckos are among the most ancient and taxonomically problematic Gekkonidae groups in Africa (Lobón-Rovira et al. 2022a). Not surprisingly, these geckos have been the focus of several studies (Underwood 1954; Kluge 1983; Bauer et al. 1997; Bauer and Menegon 2006; Rocha et al. 2011; Gamble et al. 2012; Heinicke et al. 2014; Lobón-Rovira et al. 2022a), being partially resolved into three main groups: a circum-Indian Ocean group, an Afro-Malagasy group, and Urocotyledon spp. (see Heinicke et al. 2014; Lobón-Rovira et al. 2022a). However, the relationship within each of these groups still remains unresolved in most cases, with poorly-supported deep phylogenetic nodes and several species still excluded from phylogenetic analysis, due to the lack of fresh genetic material or access to new technologies to obtain DNA from formalin-fixed specimens (Lobón-Rovira et al. 2022a, b).

Circum-Indian Ocean leaf-toed geckos, understood as a monophyletic group that attained their current geographic distributions to reflect the landmasses distributed around the Indian Ocean during the Eocene (~40 mya) (Heinicke et al. 2014), have until recently been considered as a group that includes four genera from mainland Africa (Afrogecko [2 spp.], Ramigekko [1 sp.], Cryptactites [1 sp.] and Kolekanos [1 sp.]), Christinus from Australia and Matoatoa from Madagascar (Heinicke et al. 2014). However, a recent phylogenetic analysis that includes, for the first time, material of Afrogecko ansorgii, has demonstrated a paraphyletic status of this species, being consequently described as a new genus (Bauerius) as separate clade to the circum-Indian Ocean leaf-toed geckos and rendering the four mainland Africa circum-Indian Ocean leaf-toed geckos as monotypic genera (Lobón-Rovira et al. 2022a). Nevertheless, phylogenetic analysis (Heinicke et al. 2014) also suggested cryptic diversification within Afrogecko porphyreus that requires further investigation.

This new paradigm for circum-Indian leaf-toed geckos has only been addressed thanks to the rapid growth of new molecular techniques in the last two decades and the intensive surveys in previously poorly or unexplored regions in Africa, like Angola (Vaz Pinto et al. 2019, 2021; Lobón-Rovira et al. 2022a).

Access to newly-collected material from these regions has brought new opportunities to understand the evolutionary patterns of African herpetofauna, especially African gekkonids. This is particularly noteworthy in terms of the remarkable increase in knowledge of Angolan herpetofauna, with the description of 34 new species (Conradie et al. 2012a, b, 2013, 2020a, 2022a; Stanley et al. 2016; Ceríaco et al. 2018a, 2020a, b, c, 2021; Branch et al. 2019a, 2021; Marques et al. 2019a, b, 2020, 2022a, b; Hallermann et al. 2020; Nielsen et al. 2020; Baptista et al. 2021; Lobón-Rovira et al. 2021; Parrinha et al. 2021; Wagner et al. 2021) and several new country records (Branch and Conradie 2013; Conradie and Bourquin 2013; Ernst et al. 2014, 2015; Branch et al. 2019b; Conradie et al. 2020b, 2021; Lobón-Rovira et al. 2022c) in the last decade. This increase has been especially evident within gekkonids, where the number of taxa has risen to over 45 recognised species for the country (Marques et al. 2020; Ceríaco et al. 2020a, b; Branch et al. 2021; Lobón-Rovira et al. 2021, 2022c; Conradie et al. 2022b) including two endemic leaf-toed gecko genera, Kolekanos (Heinicke et al. 2004) and Bauerius (Lobón-Rovira et al. 2022a).

Angolan leaf-toed geckos had previously been considered as members of Afrogecko Bauer, Good & Branch, 1997, represented by two species, A. plumicaudus Haacke, 2008 and A. ansorgii (Boulenger 1902). Until recent studies, both species were poorly known and with very restricted geographical distribution in south-western Angola (Haacke 2008; Agarwal et al. 2017; Marques et al. 2018; Vaz Pinto et al. 2019, 2021). With the availability of new material, both species were subsequently assigned to separate monotypic genera. Heinicke et al. (2014) erected Kolekanos to accommodate A. plumicaudus, while Lobón-Rovira et al. (2022a) created Bauerius to accommodate A. ansorgii. Furthermore, the known distribution of these two species have been extended over 200 km north- and southwards (Vaz Pinto et al. 2021) and 300 km northwards (Lobón-Rovira et al. 2022a), respectively. The range extension was especially remarkable for K. plumicaudus, which is now known to be present from sea level to over 2000 m a.s.l. and covering different ecological zones in south-western Angola (Vaz Pinto et al. 2021).

Scientific studies have been increasing in Angola in recent years, following a long civil war that prevented fieldwork in this region of Africa for several decades until the early 2000s (Huntley and Ferrand 2019). The improved political stability and strengthening of local institutions have motivated further surveys in the coastal regions of Angola, amongst others, with the aim of assessing the distribution of these poorly-known and emblematic taxa. New material of Kolekanos, collected well outside its known distributional range (~180 km north from the northernmost record of K. plumicaudus), prompted the current study aiming to investigate the potential diversification within this poorly-known genus. Due to the relevance of this group to understand the evolutionary history of African leaf-toed geckos, we herein also provide an updated phylogenetic hypothesis of the circum-Indian Ocean leaf-toed geckos with newly-collected material of Kolekanos from Angola to shed light into the taxonomic, distribution and conservation status of this taxon.

While the two phylogenetic analyses (BI and ML) did not retrieve the same topology with regard to the deeper-level topological structuring, both were largely concordant in recovering the monophyletic circum-Indian Ocean group and recognising two clearly distinct sister taxa within Kolekanos (Fig. 1). The two lineages here recognised are molecularly well-differentiated, with 24.1% ND2 uncorrected p-distance from each other and regarded as separate species (Table 2). Both species presented a large disparity in the intraspecific variation. While K. plumicaudus showed lower intraspecific variation (3.8% ± 0.4 s.e.), with the maximum p-distance (5.2%) found between two isolated highlands in Iona National Park (Tchamalindi and Cafema), the new undescribed species presented high intraspecific diversity (7.3% ± 0.6 s.e.) between the two populations found in relatively close proximity in Benguela Province (Table 2).

Figure 1. 

Maximum Likelihood phylogeny, with Bayesian Inference support overlaid. Support values (ML BS = Maximum Likelihood bootstrap values; BI PP = Bayesian Inference posterior probabilities) are shown graphically at the nodes according to the colours shown in the inset key. Kolekanos spinicaudus sp. nov. is highlighted in red.

Morphological analysis revealed morphological differences between the two main clades within Kolekanos. PCA analysis explained a considerable part of the variation within these clades, with PC1 (46.9% of variation) and PC2 (15.6% of variation) showing two well-separated groups (Fig. 2A). These results are supported by the univariate morphometric analysis (ANOVA), which detected significant differences between clades (Fig. 2B), in head width (HW, F_1,14 = 59.451, p = 0.000), forearm length (FL, F_1,14 = 7.764, p = 0.015), snout to eye distance (SE, F_1,14 = 5.905, p = 0.030) and interorbital distance (IO, F_1,14 = 25.834, p = 0.000) (Suppl. material 2). Additionally, visual comparison suggested that both species could be separated also based on the robustness stage, relative to body and hindlimbs. However, we failed to retrieve any statistically significant differences from the morphometric traits analysed to confirm this visual difference (Suppl. material 2).

Figure 2. 

A PCA plots of the first principal component (PC 1) versus the second (PC 2) of morphometric analysis for the two species of Kolekanos. The green polygon denotes the distribution within PCAs of K. spinicaudus sp. nov. and the pink polygon of K. plumicaudus. For loadings of all axis and explained variance, see Suppl. material 3. B boxplots (top whisker – maximum value; lower whisker – minimum value; bold horizontal line – median; box – 1^st and 3^rd quartile) of morphological features where ANOVA t-values where ≤ 0.05; p-value of the one-way ANOVA test is indicated at the bottom of each boxplot. For abbreviations, see Material and methods section.

Although the osteological reconstruction demonstrated the skulls of Kolekanos to be very conserved, we did find differences, mostly in overall shape of the head, supporting the above morphological findings (Fig. 3). While K. plumicaudus presented a more slender and longer-snouted head shape (Fig. 3G), the here recognised new taxon displayed a more rounded and laterally broader head shape (Fig. 3A), with a more compressed head shape in its dorsoventral profile. This modification in the head shape seems to be reflected in osteological features in the northern clade, such as larger jugal bone, more elongated lateral process of the postorbitofrontal, more compressed premaxilla and maxilla bones in its dorsoventral view and wider in the lateral profile of the bones. It is noteworthy that the specimen CAS 248782 had a pair of nasal bones, in contrast with the three specimens analysed in this study, which presented one fused nasal bone.

Figure 3. 

Detailed views in A dorsal B frontal C lateral D posterior and E ventral of skull and F lateral, dorsal, medial and ventral of left jaw (from top to bottom) of K. spinicaudus sp. nov. (MNCN50769). Detailed views in G dorsal H frontal I lateral G posterior and K ventral view of skull and L lateral, dorsal, medial and ventral of left jaw (from top to bottom) of Kolekanos plumicaudus (MNCN50770). Abbreviations: Bc, braincase; Co, coronoid; CB, compound bone; D, dentary; EcP, ectopterygoid; EP, epipterygoid; F, frontal; J, jugal; M, maxilla; N, nasal; Ot, otostapes; P, parietal; PF, prefrontal; Pl, palatine; PM, premaxilla; PO, postorbitofrontal; Pt, pterygoid; Q, quadrate; Sp, splenial; SR, sclerotic ring; V, vomer.

Therefore, the above morphological and phylogenetic differences support the recognition of two different species within Kolekanos and we take the opportunity to describe the second lineage recovered as a new species below. In this manuscript, we have applied the general lineage-based species concept, where we treat all independent evolving lineages represented and supported by multiple lines of evidence, as listed above, as separate species (de Queiroz 1998).

 Kolekanos spinicaudus sp. nov.

https://zoobank.org/80186811-4C1C-4F26-B791-0824BA79E221

Figs 3, 4, 5, 6, 7, Tables 3, 4, Suppl. material 2

Holotype

MNCN 50769, adult male, with regenerated tail and incision in the ventral region, collected in Carivo (-13.19225, 13.42108, 362 m a.s.l.), Benguela Province, Angola, by Pedro and Afonso Vaz Pinto on 19 August 2021.

Paratypes

MNCN 50766, adult male, collected from Ekongo (-13.24940, 13.20650, 636 m a.s.l.), Benguela Province, Angola, by Javier Lobón-Rovira and Pedro Vaz Pinto on 22 November 2021; MNCN 50767 & FKH-0845, adult females, with the same collecting data as the previous. FKH-0645 & FKH-0650, adult females, FKH-0647–8, adult males, MNCN 50768, subadult male, all with the same collecting data as the holotype.

Etymology

The name “spinicaudus” is derived from the combination of the Latin words “spina” and “cauda”, that refers to the spiny appearance of the tail of the new species. The species epithet is used as a singular nominative adjective “-us”.

Diagnosis

Kolekanos can be easily differentiated from other circum-Indian leaf-toed and African leaf-toed geckos, based on its ornamented tail (versus non-ornamented tail in the remaining genera). The new species differs from K. plumicaudus, based on the following characters: different ornamentation of the tail, being composed by modified scales on the margins of the original tail which resemble white lateral spines (versus feathered-like tail in K. plumicaudus); broader head (minimum HW = 7.95 mm versus maximum HW = 7.35 mm in K. plumicaudus); more robust body, with shorter forelimbs (versus thinner and more slender body in K. plumicaudus, Fig. 5); proportionally larger snout to eye distance (SE mean 4.48 mm ± 0.34 s.e. versus 3.99 mm ± 0.22 s.e. in K. plumicaudus) and interorbital distance (IO mean 4.14 mm ± 0.34 s.e. versus 3.33 mm ± 0.28 s.e. in K. plumicaudus); and dorsal pattern is less contrasted, based on zig-zag black patches surrounded by lighter patches (versus dark blocks well contrasted, not surrounded by lighter patches in K. plumicaudus). The new species can also be differentiated from K. plumicaudus by the following osteological characteristics: 1) larger jugal bone (versus reduced jugal); 2) more prominent lateral process of the postorbitofrontal (versus less prominent lateral process of postorbitofrontal); 3) more compressed premaxilla and maxilla bone on its dorsoventral profile and wider in the lateral profile of the bones; 4) ascending process of the premaxilla shorter (versus more elongated); 5) braincase compressed dorsoventrally (versus more rounded in K. plumicaudus); 6) palatine length and width equal (versus unequal); 7) postero-lateral process of parietal rounded and slightly curved (versus flat postero-lateral process of parietal broad and flat that curves downwards posteriorly); 8) anterolateral process of the coronoid markedly enlarged (versus more reduced anterolateral process). Kolekanos spinicaudus sp. nov. also differs from K. plumicaudus by circa 24% (uncorrected p-distance) ND2 mitochondrial DNA.

Holotype description

(Fig. 4). Measurements and meristic characters of the holotype are presented in Table 3. Adult male with a SVL of 44.59 mm and partially (2/3) regenerated tail, tail length (TL) 36.77 mm. Body moderately slender, nape distinct. Head slightly broader than the body and markedly compressed dorsoventrally (HH/HL = 0.27). Canthus rostralis smooth, almost absent. Eye diameter (2.35 mm), with vertical pupil and crenulated margin. Supraciliar scales small and rounded. Ear height (0.47 mm). Ear to eye distance larger than eye diameter (3.72 mm). Snout rounded and slightly pointed. Body relatively slender and elongated (TrunkL/SVL = 0.44). Fore- and hindlimbs moderate and stout, forearm large (FL/SVL 0.23), tibia short (CL/SVL 0.18). Digits elongated and clawed. All digits of manus and pes indistinctly webbed. All digits with granular basal scales and more distal widened divided lamellae. One pair of leaf-like terminal scansors. Number of scansors: 7-10-10-11-10 (right manus) and 7-10-10-11-10 (left manus)/7-9-11-11–10 (right pes) and 7-9-11-11–9 (left pes). Relative length of digits manus I < II < III < IV > V and pes I < II < III > IV > V. Scalation: Frontal scales granular and larger than occipital scales. Occipital scales small and granular. Rostral in direct contact with nostrils, 1^st supralabials, supranasals and one internasal scales. 8/8 supralabial and 9/9 infralabials. First supralabial in contact with the nostril. Nostril circular and surrounded by rostral, 1^st supralabial, supranasal and three reduced postnasals. Lower postnasal half the size of the upper postnasal and supranasal. Two rows of scales between supralabials and the orbit. Mental triangular and rounded posteriorly, with two small rounded postmental scales. 1^st infralabial rectangular and slightly larger than mental. Gular scales small and granular. Ventral scales small and granular. Precloacal pores absent. The dorsal pholidosis present homogenous granular scales from head to tail. The first third of the original tail presents lateral whitish “spine-like” scales, being absent in the last portion of the tail. Post-cloacal scales slightly larger and quadrangular. Osteology: the skull (Fig. 3) displays no co-ossification with the overlying skin. Nasals are fused. Single frontal. Paired parietals. Stapes imperforate. 14 scleral ossicles. 11 premaxillary tooth loci. 36–38 maxillary and 38 dentary tooth loci. Braincase elements fused. Postorbitofrontal arrow-shaped, with lateral process as long as anterior and posterior process. Parietal wider than longer. Jugal small, but visible.

Table 3.

Download as

CSV

XLSX

Morphological (morphometric and meristic) of Kolekanos spinicaudus sp. nov. Measurements are represented in millimetres (mm). For abbreviations, see Material and methods section. R = regenerated tail, M = male, F = female.

Species	K. spinicaudus sp. nov.	K. spinicaudus sp. nov.	K. spinicaudus sp. nov.	K. spinicaudus sp. nov.	K. spinicaudus sp. nov.	K. spinicaudus sp. nov.	K. spinicaudus sp. nov.	K. spinicaudus sp. nov.	K. spinicaudus sp. nov.
Catalogue#	MNCN 50769	FKH0845	MNCN 50767	MNCN 50766	FKH-0645	FKH-0648	FKH-0647	FKH-0650	MNCN 50768
Status	Holotype	Paratype	Paratype	Paratype	Paratype	Paratype	Paratype	Paratype	Paratype
Sex	M	F	F	M	F	M	M	F	M
SVL	44.59	43.19	44.15	40.82	43.71	41.66	42.72	41.65	35.87
TAL	R 36.77	–	R 22.26	R 39.76	R 31.39	36.85	R 28.85	–	39.44
TrunkL (mm)	19.82	17.87	18.65	16.85	19.36	18.44	19.80	18.27	16.71
HL (mm)	10.99	11.67	11.6	11.09	11.10	10.76	11.06	10.88	9.59
HW (mm)	8.63	8.12	8.56	8.48	8.28	7.95	8.21	8.26	7.18
HH (mm)	2.95	4.20	3.80	4.00	3.52	3.77	3.86	3.60	3.17
OD	2.35	2.35	2.50	2.51	2.62	2.35	2.62	2.21	2.08
EL	0.47	0.65	0.85	0.69	0.76	0.64	0.48	0.47	0.54
CL	8.03	8.59	9.01	8.74	8.15	8.01	7.80	8.15	7.94
FL	10.39	10,.68	11.10	9.94	10.93	10.73	10.13	10.40	8.49
NE	3.30	3.40	3.75	3.44	3.61	3.35	3.55	3.43	2.99
SE	4.48	4.50	4.95	4.68	4.56	4.31	4.67	4.46	3.72
EE	3.72	3.21	3.65	3.76	3.66	3.31	3.57	3.10	3.12
IN	1.55	1.55	1.48	1.45	1.45	1.40	1.34	1.39	1.22
IO	4.29	4.33	4.22	4.54	4.33	3.45	4.19	4.21	3.74
N° lamellae 1st toe (Right/Left)	4/3	5/5	4/5	4/5	4/4	4/4	4/3	4/3	5/4
N° lamellae 4th toe (Right/Left)	5/5	7/5	6/6	6/5	6/7	5/5	6/5	6/6	6/6
N° lamellae 1st finger (Right/Left)	3/4	4/	4/4	3/3	4/4	5/4	4/3	3/4	3/3
N° lamellae 4th finger (Right/Left)	6/7	5/6	6/7	6/6	5/6	5/6	6/4	5/7	7/6
N° postmental	2	2	2	2	2	2	2	2	2
N° infralabial	9	8	8	9	9	9	9	9	9
N° supralabial	8	10	10	10	9	9	10	9	9
N° internasal	1	1	1	1	1	1	1	2	1
N° scales ear to eye	17	18	16	17	12	13	14	15	16
N° scales eye to nostril	11	13	11	12	11	10	11	10	12
N° scales eye to eye	17	20	15	16	16	15	17	18	16

Variation

Variation in scalation and body measurements of the paratypes of K. spinicaudus sp. nov. are reported in Table 3. All the material analysed agrees with the holotype description with the exception of the tooth loci, where the specimen MNCN 50766 presented a larger number in the tooth loci of maxilla and dentary (> 40).

Colouration

In life (Fig. 5A): dorsal colouration varies from light pinkish to light brown, with black spots surrounded by lighter brownish regions disposed in zig-zag, from nape to tail. Dorsal reticulated light brownish colouration on tail and fore- and hind-limbs. Anterior part of the tail with marked hourglass-shaped pattern. Ventrum uniformly light cream pink from snout to posterior region of the cloaca. Tail slightly darker than the dorsum dorsally, being even darker in the ventral section, with white lateral spine-like scales on original tail. Last fourth portion of tail black. In preservative (After 4 months in preservation; Fig. 4): dorsal pattern persistent as “in life” with dorsal colouration whitish-greyish. Dark section more marked.

Figure 4. 

Holotype of Kolekanos spinicaudus sp. nov. (MNCN50769) from Carivo, Benguela Province, Angola A dorsal and ventral view of whole specimen B detail of head (from top to bottom) in dorsal, lateral and ventral views C detail of pelvic region and hind-limbs in ventral view D detail of left fingers. Photos by Alberto Sanchez Vialas (MNCN).

Figure 5. 

A dorsal view in life of K. spinicaudus sp. nov. from Carivo and B K. plumicaudus from Omahua. Photos Javier Lobo´n-Rovira.

Distribution

(Fig. 6). This species has only been found at two sites in a very restricted region, in southern Benguela Province. The area lies above the first elevational range recognised for southern Angola’s orographic relief, with specimens retrieved between 400 m and 650 m a.s.l. It can be broadly characterised as a rugged and transitional semi-arid landscape, albeit more vegetated and less arid than the coastal lowlands to the west and less mountainous and forested than eastern regions neighbouring the great escarpment. Despite its unique and unmistakable features, this species had eluded previous surveys conducted in coastal Benguela Province. In the last 5 years our team visited the same area at least five times preceding the discovery, spending at least two days per survey. Even though we found the species to be relatively common at the two referred sites, we failed to confirm its presence in several other locations with presumably suitable habitat, suggesting that it might be highly specialised and sensitive to local environmental conditions. It is possible for the species to be more common and widely distributed in poorly-surveyed regions to the southeast or north of its known range and we recommend further surveys in the region to address the conservation status of this poorly-known species.

Figure 6. 

Geographical records of Kolekanos within Angolan territory. Red circles depict records of K. plumicaudus; green circles represent K. spinicaudus sp. nov. Stars represent type localities. Background grey scale represent elevation (Huntley and Ferrand 2019).

Habitat and natural history notes

(Fig. 7). The local habitat, at both sites where the species was discovered, seems to be a transitional zone in coastal Angola, displaying a rich vegetation mosaic of acacia and mopane savannah, including Senegalia mellifera, Senegalia spp., Colophospermum mopane, Terminalia prunioides, Commiphora spp. and presence of succulents, such as Euphorbia spp. and Aloe littoralis. In contrast, and despite being known from a relatively wider region and across considerable elevational ranges, K. plumicaudus is found in much more arid and sparsely vegetated environments. The new species was mostly found at night foraging in the ecotone between the trees/bushes and moderate to large granite boulders. One individual (not collected) was retrieved while sheltering under a rock flake during the day, behaviour which has been documented for the closely-related K. plumicaudus (Agarwal et al. 2017; Vaz Pinto et al. 2021). When not stretched horizontally, this species curls the tail laterally, but not upwards, while K. plumicaudus often erects the tail upwards and may wave the tip (Agarwal et al. 2017). The first individual observed was seen running fast on the ground between a granite boulder and a tree, but more often, they were found perched on branches and once on a grass stem. Unlike K. plumicaudus, which readily jumps amongst thin branches when disturbed, K. spinicaudus sp. nov. seems to prefer to run along thicker branches or drop to the ground and run for safety. Two individuals were observed mating at night (18 August 2021 19 h 55 m) on a thin branch of Salvadora persica. One female specimen (FKH-0645) collected in November 2021 contained two well-developed eggs. This species has been found in syntopy with another Angolan leaf-toed gecko, Bauerius ansorgii. Finally, due to the complex biogeography of Angola, an updated and stabilized biogeographic classification, especially for south-western Angola, is still lacking. Current schemes depend on the authors interpretation and underlying data used (e.g. phytocoria, centres of endemism, realms, biomes, ecoregions) resulting in different units recognised and sharp boundaries (Burgess et al. 2004; Dinerstein et al. 2017) which often do not match the situation on the ground. Thus, we cannot currently assign any specific biogeographic region to any of these two taxa and are anticipating a better review of Angolan biogeographic units through Huntley (in prep.) in the near future.

Figure 7. 

Habitat of Kolekanos spinicaudus sp. nov. at A Carivo and B Ekongo. Photos Javier Lobón-Rovira.

Table 4.

Download as

CSV

XLSX

Morphological (morphometric and meristic) of Kolekanos plumicaudus. Measurements are represented in millimetres (mm). For abbreviations, see Material and methods section, R = regenerated, M = male, F = female.

Species	K. plumicaudus	K. plumicaudus	K. plumicaudus	K. plumicaudus	K. plumicaudus	K. plumicaudus	K. plumicaudus	K. plumicaudus	K. plumicaudus	K. plumicaudus
Catalogue#	MNCN 50770	FKH-0661	FKH-0663	PEM R18010	PEM R18014	PEM R18011	PEM R18015	PEM R18012	PEM R18013	PEM R18047
Status	–	–	–	–	–	–	–	–	–	–
Sex	F	F	F	F	M	M	F	F	F	M
SVL	41.91	38.45	40.22	42.05	42	36.57	41.84	36.97	41.84	37.38
TAL	R 34.14	R 34.38	41.91	40.34	–	R 26.35	–	33.34	–	34.68
TrunkL (mm)	19.08	17.84	19.07	18.35	18.38	16.20	18.41	16.97	19.38	20.03
HL (mm)	10.04	9.42	9.89	11.11	11.17	10.59	11.59	10.22	11.53	10.98
HW (mm)	7.03	6.91	6.81	6.71	7.07	6.29	7.13	6.92	7.35	6.70
HH (mm)	3.50	3.38	3.27	3.45	3.19	2.76	3.62	2.90	3.21	30.70
OD	2.08	2.28	2.17	2.35	2.31	2.11	2.34	2.14	2.42	2.10
EL	0.59	0.49	0.50	0.71	0.79	0.58	0.63	0.87	0.75	0.62
CL	8.15	7.39	8.11	8.90	8.34	7.33	7.45	8.05	8.41	7.85
FL	10.75	8.77	9.69	10.38	11.76	10.54	12.42	11.85	12.86	12.66
NE	2.91	2.80	2.97	3.15	3.23	3.15	3.49	2.83	3.32	3.06
SE	3.89	3.61	3.65	4.18	4.07	3.98	4.13	3.92	4.24	4.18
EE	3.71	3.00	3.49	3.47	3.45	2.91	3.49	3.19	3.25	2.96
IN	1.30	1.29	1.24	1.30	1.21	1.11	1.25	1.26	1.45	1.43
IO	3.71	3.57	3.78	3.16	3.06	3.03	3.29	3.19	3.46	3.04
N° lamellae 1st toe (Right/Left)	4/4	3/3	4/3	4/4	4/4	4/4	4/4	6/6	5/4	4/4
N° lamellae 4th toe (Right/Left)	6/5	5/6	5/6	6/6	7/7	6/6	7/4	5/6	6/7	7/8
N° lamellae 1st finger (Right/Left)	3/4	3/4	3/4	4/4	4/4	4/4	4/4	4/4	4/4	4/4
N° lamellae 4th finger (Right/Left)	d/6	6/6	6/6	6/6	8/8	7/7	8/8	7/7	7/7	7/7
N° postmental	2	1	2	3	2	2	2	3	2	2
N° infralabial	8	8	9	9	9	9	9	9	9	9
N° supralabial	10	10	10	10	9	10	9	10	9	10
N° internasal	1	1	1	0	1	0	0	1	1	1
N° scales ear to eye	13	14	15	17	17	17	18	16	15	17
N° scales eye to nostril	11	9	11	10	9	10	8	8	8	9
N° scales eye to eye	15	14	15	16	18	15	14	15	15	14

Conservation status

The species seems relatively common, but highly localised. Although the general habitat does not appear to be threatened, more research is needed to confirm if the species’ distribution is larger than currently known. Therefore, following the IUCN Red List guidelines (IUCN 2022), the species should be considered as Data Deficient (DD).

Using molecular and morphological evidence, we herein described a new leaf-toed gecko, from southern Benguela Province, Angola, Kolekanos spinicaudus sp. nov., thereby adding another species to the growing list of gekkonids described in the last decade from this poorly-known African country (Ceríaco et al. 2020a, b; Marques et al. 2020; Branch et al. 2021; Lobón-Rovira et al. 2021; Conradie et al. 2022b). The recognition of K. spinicaudus sp. nov. as a sister species of K. plumicaudus, contradicts the previous knowledge of mainland circum-Indian Ocean leaf-toed geckos as monotypic genera (Lobón-Rovira et al. 2022a) and reinforces the need of further investigation on the potential cryptic diversification within another related species, such as Afrogecko porphyreus (Heinicke et al. 2014). Therefore, we provide a new perspective for future work within this group, which may improve the knowledge regarding Angolan and western African gekkonid diversity.

The molecular analysis, provided in this work, has shown a large divergence of the ND2 mitochondrial gene between K. spinicaudus sp. nov. and K. plumicaudus, being even higher than the molecular divergence found between closely-related genera, such as with Rammigekko and Cryptactites. However, the external morphological similarities between these two taxa support differentiation only at species level. This high divergence can be explained by the ancient character of this group (Heinicke et al. 2014), having persisted through extreme climatological and environmental changes and, consequently, experiencing long isolation periods in southwest Africa (Chase et al. 2009; Garzanti et al. 2018). Both species revealed notable genetic intraspecific variation between close localities, which can, in both cases, be explained by the high ecological specialisation in these geckos, promoting and maintaining isolation and by the relatively fast mtDNA evolutionary rates (Jesus et al. 2006).

Regarding interspecific variation, both species have exhibited a high degree of morphological and ecological differentiation. While K. plumicaudus presented a more slender head and body and seems more strongly associated with the more arid environments of the Angolan Kaokoveld Desert in Namibe Province, the sister species, K. spinicaudus sp. nov. presented a more robust head, body and limbs and is apparently only found in the semi-arid savannahs of Benguela Province (Lobón-Rovira et al 2021). The two species are quite agile and often found foraging in bushes. Kolekanos plumicaudus sometimes wag their tail semi-erected when disturbed and readily jumps amongst thin branches as a primary escape strategy (Agarwal et al. 2017; Vaz Pinto et al. 2021), while K. spinicaudus sp. nov., when threatened, is primarily a swift runner either along branches or on the ground. These behavioural differences may likely reflect subtle adaptations to local conditions, including vegetation cover, predation, foraging and sheltering habits. In addition, both species present a distinctive tail ornamentation that can be used as a clear morphological diagnostic feature between them. These findings seem to be consistent with the idea that habitat diversity leads to species and morphological diversification (Losos and Parent 2009; Tejero-Cicuéndez et al. 2021).

Some of the osteological features provided by Heinicke et al. (2014) seem to be non-homoplastic apomorphic characters for Kolekanos genus, such as ectopterygoid width more or less constant along the length of the bone, prootic contacting the epipterygoid far behind from the posterior process of the postorbitofrontal and groove associated with the surangular foramen and coronoid abutting the dentary. We failed to recover some of the proposed characters for either of the two species of Kolekanos, such as jugal bone being very reduced, almost vestigial (versus moderated size) and anterolateral corner of parietal not clasping the frontal (versus clasping anterolateral section). Furthermore, the K. plumicaudus CT-scanned in this work was not fully concordant with the diagnosis presented in the description of this unique genus (Heinicke et al. 2014), for example, fused nasal bones and well-developed postorbitofrontal bone (versus unfused nasal bones and reduced, almost vestigial, postorbitofrontal bone, Heinicke et al. 2014). However, this difference could be associated with sexual dimorphism, since the only two specimens of K. plumicaudus represent one of each sex. Thus, this work underlines the importance of using larger series of material to fully infer diagnostic characters between species (Lobón-Rovira et al. 2021) and these being even more important to infer osteological variability (Bochaton et al. 2018), including sexual dimorphism. While we consider that the external diagnostic characters are sufficient to identify these species, we suggest caution while using the osteological differences to distinguish these two taxa due to the small sampling size available.

We here provide another example of diversification in south-western Angola, leading to speciation in the more arid desert ecosystems of Namibe Province and in the semi-arid coastal savannahs of Benguela Province, a pattern that has been found in other studies (e.g. Hemidactylus benguellensis-group, Lobón-Rovira et al. 2021). Our findings underline the remarkable herpetological value of coastal Benguela Province and particularly as a potential gekkonid hotspot. We have confirmed at Carivo that at least 14 species, representing all eight Angolan genera, are living in sympatry. This is the highest number of gekkonid species recorded in a single site in the country. In addition, we also report, for the first time, the two endemic Angolan leaf-toed gecko genera (Bauerius and Kolekanos) found in syntopy in both localities where K. spinicaudus sp. nov. has been found.

To conclude, we recommend additional surveys in Benguela Province to study the distribution and abundance of this new species to assess its conservation status and further research is needed in northern Namibe Province to explore potential contact zones between the two Kolekanos species. Due to the high genetic divergence between the two recorded populations of K. spinicaudus sp. nov., we also suggest caution when addressing conservation strategies in western Angola, since it may affect ongoing speciation processes within Kolekanos in this region.

We thank Alvarito Eugénio, the owner of Carivo Farm, where we performed frequent surveys and where topotypical material was recovered, for his very kind hospitality. We also thank Luis Pittagrós and “Bibi” for their logistical support and precious advice. In addition, the INBC – Instituto Nacional da Biodiversidade e Conservação within the Ministry of Culture, Tourism and Environment (MCTE) and especially Director of INBC, Drª Albertina Nzuzi, for issuing research and export permits (no. 01/2021). We thank Fernanda Lages and ISCED – Instituto Superior de Ciências da Educação da Huíla for logistical support. We are grateful to Vladimir Russo from Fundação Kissama for critical administrative and logistical support. We also thank to Alberto Sanchez Vialas from MNCN for providing the photos and detailed information of type material. Afonso Vaz Pinto for unconditional support and collecting in the field. Special thanks to Aaron Bauer for providing the raw files for the CT reconstruction of the CAS specimen. We also thank Centre for Molecular Analysis (CTM) workers (especially Susana Lopes, Sofia Mourão and Patricia Ribeiro) at CIBIO for their tireless work and support in the molecular lab. To Fatima Linares from Centro de Instrumentación Científica of Granada for her incredible work on the CT scanning. JLR and NLB are currently supported by Fundação para a Ciência e Tecnologia (FCT) contracts PD/BD/140808/2018 and PD/BD/140810/2018, and BIOPOLIS 2022-18 and 2022-19. This work was supported by the European Union’s Horizon 2020 Research and Innovation Programme under the Grant Agreement Number 857251 and by National Funds through FCT-Fundação para a Ciência e Tecnologia, under the scope of project UIDP/50027/2020. Finally, we thank to Stuart Nielsen and an anonymous reviewer for the comments, which certainly have improved this manuscript.