Taxonomical review of Prosymna angolensis Boulenger, 1915 (Elapoidea, Prosymnidae) with the description of two new species

Werner Conradie; Chad Keates; Ninda L. Baptista; Javier Lobón-Rovira

doi:10.3897/zookeys.1121.85693

Abstract

African Shovel-snout snakes (Prosymna Gray, 1849) are small, semi-fossorial snakes with a unique compressed and beak-like snout. Prosymna occur mainly in the savanna of sub-Saharan Africa. Of the 16 currently recognised species, four occur in Angola: Prosymna ambigua Bocage, 1873, P. angolensis Boulenger, 1915, P. frontalis (Peters, 1867), and P. visseri FitzSimons, 1959. The taxonomical status and evolutionary relationships of P. angolensis have never been assessed due to the lack of genetic material. This species is known to occur from western Angola southwards to Namibia, and eastwards to Zambia, Botswana and Zimbabwe. The species shows considerable variation in dorsal colouration across its range, and with the lower ventral scales count, an ‘eastern race’ was suggested. In recent years, Prosymna material from different parts of Angola has been collected, and with phylogenetic analysis and High Resolution X-ray Computed Tomography, the taxonomic status of these populations can be reviewed. Strong phylogenetic evidence was found to include the angolensis subgroup as part of the larger sundevalli group, and the existence of three phylogenetic lineages within the angolensis subgroup were identified, which each exhibit clear morphological and colouration differences. One of these lineages is assigned to the nominotypical P. angolensis and the other two described as new species, one of which corroborates the distinct eastern population previously detected. These results reinforce that a considerable part of Angolan herpetological diversity is still to be described and the need for further studies.

Resumo

As cobras-de-focinho-de-pá africanas (Prosymna) são pequenas cobras semi-fossoriais com um focinho único, achatado e em forma de bico, que ocorrem principalmente na savana da África subsaariana. Das 16 espécies actualmente reconhecidas, quatro existem em Angola: Prosymna ambigua Bocage, 1873, P. angolensis Boulenger, 1915, P. frontalis (Peters, 1867), e P. visseri FitzSimons, 1959. O estatuto taxonómico e as relações evolutivas de P. angolensis nunca foram avaliados devido à falta de material genético. A espécie ocorre desde o oeste de Angola, para sul até a Namíbia, e para este em direcção à Zâmbia, Botswana e Zimbábue. Na sua área de ocorrência, esta espécie tem variação principalmente na coloração dorsal, e com base no menor número de escamas ventrais, foi sugerida a existência de uma raça oriental. Recentemente foi amostrado material de Prosymna de várias partes de Angola, e com recurso a análises filogenéticas e a tomografia computadorizada de raios-X de alta resolução, foi possível rever o estatuto taxonómico destas populações. Encontrámos fortes evidências filogenéticas para incluir o subgrupo angolensis como parte do grupo sundevalli. Revelámos a existência de três linhagens filogenéticas no subgrupo angolensis. Atribuímos uma dessas linhagens ao P. angolensis nominotípico, e descrevemos as outras duas como espécies novas, uma das quais corrobora a população oriental previamente detectada. Estes resultados reforçam a ideia de que uma parte considerável da diversidade herpetológica angolana está ainda por descrever, e a necessidade de mais estudos.

Keywords

Africa, Angola, cryptic species, fossorial, Kalahari, Serpentes

Palavras-chave

África, Angola, espécies crípticas, fossorial, Kalahari, Serpentes

Introduction

African Shovel-snout snakes, belonging to the genus Prosymna, are small terrestrial snakes occurring in sub-Saharan Africa, mostly associated with semi-desert, savanna and miombo woodlands (Broadley 1980). They are characterised by a compressed skull with a depressed, upward pointing snout with a sharp horizontal edge, which allows them to live a semi-fossorial lifestyle (Broadley 1990; Branch 1998; Marais 2004; Spawls et al. 2018; Pietersen et al. 2021). The absent or anteriorly reduced dentary teeth, and the unique modified blade-like rear maxillary teeth allow them to slit open soft-shelled reptile eggs, on which they feed almost exclusively, although some species feed on hard-shelled gecko eggs (Broadley 1979, 1980; Heinicke et al. 2020). Due to this unusual skull compression, snout shape and modified maxillary teeth, the higher taxonomical level of these snakes has been in flux. However, modern phylogenetic analyses have allowed them to be assigned to their own family, Prosymnidae (superfamily Elapoidea), sister to the family Psammophiidae (Vidal et al. 2008; Kelly et al. 2009; Pyron et al. 2013; Figueroa et al. 2016; Zaher et al. 2019). Currently, Prosymna is represented by 16 species (Heinicke et al. 2020; Uetz et al. 2022), and recent phylogenetic work has shown that cryptic diversity may exist in the genus, especially within P. ambigua, P. frontalis, and P. stuhlmanni (Heinicke et al. 2020).

The first Prosymna recorded from Angola were documented by Bocage (1873). He assigned material collected from Mossamedes [= Moçâmedes] and Biballa [= Bibala] to Prosymna frontalis (Peters, 1867) and described Prosymna ambiguus Bocage, 1873 (= P. ambigua) from Duque de Bragança (= Calandula) based on the higher number of midbody scale rows (17 vs. 15), the shape of the rostral and the larger parietal scales when compared to P. frontalis. Later, Bocage (1882) once again noted that the Angolan specimens of P. frontalis did not fully agree morphologically with the original description provided by Peters (1867). Afterwards, he provided a more detailed account including additional material collected from Angola, reporting the morphological variation observed when compared to P. frontalis types (Bocage 1895). The main confusion that Bocage faced at the time was that the two type specimens (one adult and one juvenile) of P. frontalis represented two separate species: P. frontalis and P. bivittata (see Mertens 1955; Broadley 1980). His material agrees in part with the juvenile specimen (= P. bivittata) in the number of postoculars (= 1) and lower subcaudal scale count (< 25), while it differed from the larger specimen (= P. frontalis lectotype) in lower subcaudal scale count (50 vs. 17–25) and the number of postorbital scales (one vs. two). The only difference was that the internasal condition (single bandlike scale) of the Angolan material was in agreement with the P. frontalis lectotype and not with the juvenile of P. bivittata. Due to this confusion in the overlapping morphology with the original P. frontalis types, Bocage decided not to take any taxonomic actions and kept referring to material from Angola under the name P. frontalis. It should be noted that the Reptile Database (Uetz et al. 2022) states that P. angolensis is a nomen novum for ‘P. frontalis Bocage, 1895’. This is incorrect, as Bocage (1895) never described P. frontalis as a new species.

Boulenger (1915), with no proper justification, described the Angolan material previously assigned to P. frontalis by Bocage (1873, 1882, 1895) as a new species, i.e., P. angolensis Boulenger, 1915, stating only how it differed from the only other Angolan congener at the time, P. ambigua. Additional material under the names P. ambigua and P. angolensis were recorded from central and western Angola by Monard (1931, 1937), Mertens (1937, 1938) and Bogert (1940). Until the mid-19^th century, this was the only species of Prosymna known to occur in Angola, until Charles Koch collected a specimen from south-western Angola, that was later described as a new species, Prosymna visseri FitzSimons, 1959. When Broadley (1980) reviewed the Prosymna genus, he documented nominotypical P. frontalis from Benguela, collected by Wulf Haacke in the 1970’s. This brought the number of Prosymna species occurring in Angola to four (Branch 2018; Marques et al. 2018).

Over the decades, the relationship of P. angolensis to other Prosymna species was addressed on a morphological basis by several authors, being considered as more closely related to P. sundevalli (Boulenger 1894), P. lineata (Bocage 1895), P. ambigua (Bogert 1940), and P. frontalis (Mertens 1955). Finally, with the whole genus revision, Broadley (1980) was the first to attempt grouping Prosymna species based on shared morphological characters. He identified three main species groups (the ambigua group = ambigua, ornatissima, semifasciata, stuhlmanni; the meleagris group = greigerti, meleagris, ruspolli, somalica; the sundevalli group = bivittata, lineata, sundevalli). Whilst he briefly mentioned it might be part of the sundevalli group, he could not confidently assign P. angolensis to any of these groupings. With the aid of phylogenetics, Heinicke et al. (2020) partly validated Broadley’s groupings, but due to the lack of P. angolensis genetic material, its relationship to its congeners remained unclear.

Since the description of P. angolensis, only a handful of specimens have been recorded from Angola (Monard 1931, 1937; Mertens 1937; Bogert 1940; Hellmich 1957; Baptista et al. 2019; Ceríaco et al. 2021a; Conradie et al. 2021). Specimens from outside of Angola were documented from northern Namibia (Mertens 1955), the Zambezi Region (= Caprivi Strip) in north-eastern Namibia, western Zambia, northern Botswana (Broadley 1980), and north-eastern Zimbabwe (Broadley 1995) (see Fig. 1). However, the specimens from Zambia and the Zambezi Region differ from the nominotypical Angolan form in the number of ventral scales (121–129 vs. 134–142), number of postoculars (two vs. one), and dorsal colouration (large black confluent black blotches vs. mostly two dorsal rows of small paired black spots) (Broadley 1980). Interestingly, one specimen from Vila da Ponte (= Kuvango), in west/central Angola, exhibits the same combination of features (lower ventral scale counts, two postoculars and a dorsal pattern consisting of confluent black blotches posteriorly) (Monard 1937 fide Broadley 1980). Despite noticing these differences, Broadley (1980) did not make taxonomic changes, stating that a larger series was needed to clarify this issue.

Figure 1.

Geographic records of the Prosymna angolensis group, based on all literature records and newly examined material, including nominotypical P. angolensis (blue triangles), Prosymna ‘Eastern’ = P. lisima sp. nov. (red circles) and Prosymna ‘Coastal’ = P. confusa sp. nov. (purple diamonds). White stars represent respective type localities. Question marks represent material tentatively assigned to that species, but needs confirmation. Blue lines represent major river systems. Top right inset represents the eastern Zambezi Region, to show sympatry between P. angolensis and Prosymna ‘Eastern’ = P. lisima sp. nov.

Recent herpetological surveys undertaken in eastern and southwestern Angola led to the collection of several specimens assigned to P. angolensis (Branch 2018; Baptista et al. 2019; Conradie et al. 2021). The first was collected by William R. Branch from the coastal semi-arid lowlands in Namibe Province and was initially assigned to P. ambigua (Branch 2018). This identification was problematic because it had no morphological justification and was collected from a locality well outside the known range of that species. Upon re-examination of the specimen, it was concluded that it was a uniformly coloured P. angolensis (WC unpubl. data), another problematic identification, as it did not agree with the species description and this was the first record of P. angolensis from the more arid coastal plains of Angola. Nominotypical P. angolensis, which fully agrees with the original description, was collected from Bicuar National Park in southwestern Angola (Baptista et al. 2019). Finally, during several expeditions to document the biodiversity of the headwaters of the Angolan Okavango-Cuando-Zambezi river basins in east-central Angola, a series of specimens was collected, and tentatively assigned to P. angolensis based on external morphology (Conradie et al. 2021), but they exhibited the same distinct characteristics reported from western Zambia and north-eastern Namibia specimens by Broadley (1980). This new material allowed us to revisit the taxonomical status and evolutionary relationships of P. angolensis, and investigate the morphological differences observed among these different populations using modern phylogenetic analysis and High Resolution X-ray Computed Tomography (HRCT).

Materials and methods

Sampling

At several sites during the 2016–2019 National Geographic Okavango Wilderness Project surveys, standard Y-shape intercept drift fence funnel trap arrays were deployed to passively collect specimens. Each Y-shaped trap array consisted of three drift fences (each 10 m long and 50 cm high) radiating from a central pitfall trap, with six one-way funnel traps placed on adjacent sides of each drift fence and three one-way funnels at the terminal ends of each drift fence. Trap arrays were installed in varied habitats to ensure the highest possible richness of captured species (Conradie et al. 2021). Consequently, a total of eight Prosymna individuals were captured. Snakes were euthanized by injecting them subcutaneously with tricaine methane sulfonate (MS222) solution (Conroy et al. 2009), after which they were formalin-fixed for 48 hours and transferred to 70% ethanol for long-term storage. Prior to formalin fixing, tissue samples (liver) were preserved in 99% ethanol for subsequent genetic analysis. In addition, we included a recently collected specimen of nominotypical P. angolensis from Bicuar National Park (Baptista et al. 2019) and a specimen that was tentatively assigned to P. ambigua from southwestern Angola (Branch 2018), to complement the morphological and genetic analyses. Voucher specimens are held in the herpetological collection of Port Elizabeth Museum at Bayworld Complex (PEM), Gqeberha, South Africa and Colecção Herpetológica do Lubango (CHL), currently deposited in Instituto Superior de Ciências de Educação da Huíla (ISCED-Huíla), Lubango, Angola.

Additionally, Prosymna angolensis material examined by Broadley (1980) from the following institutions was included in this study:

AMNH American Museum of Natural History, New York, USA;

CAS California Academy of Sciences, Los Angeles, USA;

MBL Museu Bocage, Lisbon, Portugal;

MCZ Museum Comparative Zoology, Harvard, USA;

NMW Naturhistorisches Museum zu Wien, Vienna, Austria;

NMZB/UM National Museum of Zimbabwe, Buluwayo, Zimbabwe;

SAM South African Museum (now Iziko Museums of South Africa), Cape Town, South Africa;

SMF Forschungsinstitut und Natur-Museum Senckenberg, Frankfurt-am-Main, Germany;

TM Transvaal Museum (now Ditsong National Museum of Natural History Northern Flagship Institute);

USBN National Museum of Natural History, Washington, USA.

The original datasheets were made available to the authors by Sheila Broadley. Photographs of the following material were examined by WC: AMNH R50504, MCZ R-32468, NMW 19275.2 and SAM ZR16574.

DNA extraction, amplification, and sequencing

A standard salt extraction method (Bruford et al. 1992) was used to isolate DNA from the tissue sample using ATL lysis and AE elution buffers. Standard PCR procedures were utilised to amplify one partial mitochondrial ribosomal gene (ribosomal ribonucleic acid [16S]), two partial mitochondrial genes (cytochrome b [cyt-b] and NADH-dehydrogenase subunit 2 [ND2]) and one partial nuclear gene (oocyte maturation factor [c-mos]). The specific primer pairs used can be found in Table 1. Each amplification was conducted with a PCR mixture to the total volume of 25 µl containing 12.5 μl TopTaq Mastermix (Ampliqon; containing 2× master mix, 1.5 mM MgCl₂, 0.4 mM dNTPs, and Ampliqon Taq DNA polymerase), 2 µl forward primer (10 µM), 2 µl reverse primer (10 µM), 6.5 µl de-nucleated water and 2 µl genomic DNA (20–50 ng/µl). The cycling profile for all the genes was as follows: initial denaturing step at 94 °C for 5 min, followed by 35 cycles of 94 °C for 30 s, 50–60 °C for 45 s, and 72 °C for 45 s, with a final extension at 72 °C for 8 min. The prepared PCR products were purified and sequenced at Macrogen Corp. (Amsterdam, Netherlands) with the forward primers only.

Table 1.

Download as

CSV

XLSX

Primers and PCR protocols used to generate sequences for the study.

Gene	Primer	Source	Annealing temperature (°C)
16S	L2510: 5’—CGCCTGTTTATCAAAAACAT-—3’	Palumbi (1996)	50
16S	H3080: 5’-CCGGTCTGAACTCAGATCACGT-3’	Palumbi (1996)	50
cyt-b	WWF: 5’—AAAYCAYCGTTGTWATTCAACTAC—3’	Whiting et al. (2003)	52
cyt-b	Cytb‐R2: 5’—GGGTGRAAKGGRATTTTATC—3’	Whiting et al. (2003)	52
ND2	ND2-F1-METF1: 5’—AAGCTTTCGGGCCCATACC—3’	Heinicke et al. (2014)	56
ND2	ND2-R1-TRPR3: 5’—TTTAGGGCTTTGAAGGC—3’	Heinicke et al. (2014)	56
c-mos	S77: 5’—CAT GGACTGGGATCAGTTATG—3’	Slowinski and Lawson (2002)	52
c-mos	S78: 5’—CCTTGGGTGTGATTTTCT CACCT—3’	Slowinski and Lawson (2002)	52

Phylogenetic analyses

The phylogenetic placement of the newly collected Prosymna samples were estimated by comparing them with the sequenced data from Heinicke et al. (2020). In addition to the ingroup taxa (12 of the 16 currently recognised Prosymna species), the dataset was supplemented with sequences from closely related genera that were obtained from GenBank and used as outgroups (Appendix 1: Table A1).

The sequence trace files were checked using BioEdit Sequence Alignment Editor v. 7.2.5 (Hall 1999) and aligned with accessioned GenBank sequences using MEGA v.6.0 (Tamura et al. 2013) and the ClustalW alignment method. Four individual alignments were created (16S, ND2, cyt-b, c-mos) and added to the existing dataset from Heinicke et. al (2020), along with two additional nuclear markers (RAG1 and ENC-1) from the same paper, which were used to resolve the deeper nodes. The congruency of the individual genes was tested using the homogeneity test implemented in PAUP4 v. 4.0a (Swofford 2003). All six gene alignments were not significantly different from one another allowing the creation of a concatenated dataset.

DAMBE v. 6.4.67 (Xia 2013) was used to test for saturation using the individual as well as the combined first and second codon positions of each gene. Saturation was absent from every marker, so a gene-partitioned dataset was created for the phylogenetic reconstruction. The optimal partition scheme and best-fitting models of molecular evolution were selected using ModelFinder implemented in IQ-TREE v.2.1.2 (Minh et al. 2021). The following settings were used: -p partition file (each partition has own evolution rate), a greedy strategy and the FreeRate heterogeneity model excluded (only invariable site and Gamma rate heterogeneity considered) (Chernomor et al. 2016; Kalyaanamoorthy et al. 2017). The best-fitting model scheme selected included the following three partitions and models of evolution: TIM2+I+G (16S); GTR+I+G (cyt-b, ND2); TN+G (c-mos, RAG1, ENC-1). MrBayes v.3.2.7a (Ronquist et al. 2012) and BEAST2 v.2.6.6 (Bouckaert et al. 2019) were not able to implement TIM2 or TN, so the next best alternative (GTR) was used in their place.

Phylogenetic reconstruction

Maximum likelihood (ML) analysis was conducted using IQ-TREE v.2.1.2 (Nguyen et al. 2015). A random starting tree was used using the gene-partitioned scheme mentioned above, the ultrafast bootstrap approximation (UFBoot) method (Hoang et al. 2018) and 1000 bootstrap replicates. Bayesian inference (BI) analysis was implemented using MrBayes v. 3.2.7a (Ronquist et al. 2012) and BEAST2 v. 2.6.6 (Drummond and Rambaut 2007; Suchard and Rambaut 2009) on the CIPRES Science Gateway XSEDE online resource (http://www.phylo.org; Miller et al. 2010; Tamura et al. 2013) using the gene-partitioned scheme mentioned above. For MrBayes, two parallel runs of 20 million generations were performed, with trees being sampled every 1000 generations using BEAGLE (high performance likelihood calculation library). Twenty percent of the generations were discarded as burn-in. For BEAST2, the analysis was run for 50 million generations and sampled every 10,000 generations. Ten percent of the sampled generations were discarded as burn-in. Using Tracer v. 1.6.0. (Rambaut and Drummond 2007), the effective sample size (ESS) was more than 200 for all parameters and the runs reached convergence, indicating that the burn-in for both BI phylogenies was adequate.

Species delimitation

Species delimitation was used to elucidate whether the putative Angolan taxa identified in the phylogenetic tree constituted separate species. Outgroup taxa were removed, leaving only members of Prosymna for single locus species delimitation. The 16S and cyt-b genes were chosen as they had the best representation. The following species delimitation analyses were used: Automatic Barcode Discovery (ABGD), Assemble Species by Automatic Partitioning (ASAP), Poisson Tree Processes (PTP), and Bayesian Poisson Tree Processes (bPTP).

Firstly, a 16S and cyt-b alignment were prepared and uploaded onto the ABGD web interface (abgd web (mnhn.fr), web version 07 July 2022) and the ASAP Web Interface (ASAP web (mnhn.fr), web version 07 July 2022). For ABGD, the followings settings were used: standard p-distance metrics, minimum barcode gap width (1), intraspecific divergence minima (0.001) and maxima (0.1) (Puillandre et al. 2012). For ASAP, the Simple Distance (p-distances) substitution model was used (Puillandre et al. 2021).

Secondly, a 16S and cyt-b ML tree were created in IQTREE, both using the GTR + I + G substitution model and the same settings implemented in the multi-locus phylogeny. The phylogenies were rendered as newick files and uploaded unrooted onto the bPTP web server (http://species.h-its.org/ptp/; Zhang et al. 2013) for PTP and bPTP analysis. The individual gene trees were then rendered, using Figtree v.1.4.2 (Rambaut 2014), and the results from the different single-locus species delimitation analyses were overlaid.

Pairwise distance analysis

Pairwise distance analysis was implemented in MEGA X (Kumar et al. 2018) using the individual 16S and cyt-b alignments, from the phylogenetic reconstruction. Sequences were trimmed to reduce missing data from the datasets and sequences that still had more than 10% data were removed from the alignments to ensure the most accurate p-distance values were attained. For the 16S alignment, the hyper-variable region was retained. Sequences were grouped according to species and pairwise distance analysis was conducted using the following settings: uniform rates, pairwise deletion and 500 bootstrap replicates.

Morphology

Morphological data was gathered from 39 Prosymna angolensis sensu lato (Table 2). Snout-vent length (SVL, measured from the tip of the snout to the posterior end of the cloacal scale or vent opening) and tail length (TL, measured from the cloacal opening to the tip of the tail) were measured to the nearest 1 mm using a flexible ruler or a tape measure. We also expressed the TL as a percentage of the total length (SVL + TL). The following scale counts were recorded using a Nikon SMZ1270 binocular stereo microscope: number of middorsal scale rows (counted one head length behind head, at midbody, and one head length anterior to the cloacal scale), number of preoculars, number of postoculars, the temporal scale arrangement, number of supralabials and the number of supralabials entering orbit, number of infralabials and number of infralabials in contact with 1^st sublinguals, the presence of loreal, the number of ventral scales (Dowling 1951a), number of subcaudal scales (counted from anterior cloaca, excluding the terminal spine) and cloacal scale condition (divided or entire). Scale row reduction was also recorded (Dowling 1951b).

Table 2.

Download as

CSV

XLSX

Summary of morphological features and measurements for Prosymna angolensis group. For abbreviations see Materials and methods. Notes: * including data on the Ebanga (Monard 1937) and Capelongo (Bogert 1940) specimens.

	P. angolensis	P. angolensis	P. lisima sp. nov.	P. lisima sp. nov.	P. confusa sp. nov.*
	Males	Females	Males	Females	Females
Sample size	6	21	7	3	3
SVL (mm)	160–248 (208.7 ± 29.8)	127–305 (224.7 ± 51.1)	138–198 (180.7 ± 20.5)	168–275 (214.3 ± 54.9)	231–240 (235.5± 6.4)
TL (mm)	22–30 (26.5 ± 2.7)	12–27 (29.9 ± 3.7)	17.0–28.9 (24.6 ± 3.7)	19.3–28.0 (22.6 ± 4.7)	23–29 (26.0 ± 4.2)
TL/total length ratio (%)	10.1–13.3 (11.40 ± 1.2)	6.5–9.4 (7.9 ± 0.9)	11.0–13.0 (12.0 ± 0.8)	8.8–10.9 (9.7 ± 1.1)	9.1–108 (9.9 ± 1.2)
Ventral scales	126–155 (138.1 ± 10.4)	134–163 (147.5 ± 8.3)	116–124 (120.2 ± 3.1)	117–129 (122.3 ± 6.1)	143–155 (149.7 ± 6.1)
Subcaudal scales	22–28 (25.4 ± 1.8)	16–25 (18.9 ± 2.5)	22–26 (23.7 ± 1.9)	18–24 (20.0 ± 3.5)	17–26 (21.0 ± 4.6)
Midbody scale rows	17-15-15 (rarely 19-15-15)		17-15-15		17-15-15
Cloacal scale	Entire		Entire		Entire
Preoculars	1		1 (rarely 2)		1
Postoculars	1 (rarely 0 or 2)		2		1
Temporals	1+2 (rarely 2+2, 2, 3)		1+2 (rarely 1+2+3)		1+5
Supralabials (contacting eye)	6 (3,4) [rarely 5 (2,3) or 7 (3,4)]		6 (3,4) [rarely 5 (2,3)]		5–6 (3, 4)
Infralabials (in contact with 1^st chin shield)	7 (3) [rarely 8 (3)]		7 (3)		7 (3)
Loreal	Present		Present		Present

For morphological comparison we preassigned material into three distinct groups based on the shared morphological and colouration differences observed by Broadley (1980) and our personal observations: 1) the nominotypical group which includes material (6 males; 21 females) from west/central Angola, northern Namibia, northern Botswana and north-eastern Zimbabwe (hereafter referred to as P. angolensis sensu stricto), 2) a group which includes material (7 males and 3 females) from eastern Angola, western Zambia and north-eastern Namibian (hereafter refer to as Prosymna ‘Eastern’), and a single female specimen from the coastal Angolan lowlands (PEM R24013) (hereafter refer to as Prosymna ‘Coastal’). We could not include the latter group in significant testing due to that we only had one confirmed sample.

To test if the ventral and subcaudal scale counts differs significantly between Prosymna ‘Eastern’ and P. angolensis sensu stricto as reported by Broadley (1980), we first corrected for size by dividing the ventral scale counts by the SVL and the subcaudal scale counts by the TL. The data was then separated by sex for further analyses. Due to our small sample size, we conducted a Shapiro-Wilk normality test and found that our data is not normally distributed and thus we proceed in conducting a non-parametric Wilcoxon test. We display the results using standard boxplots. All above-mentioned quantitative statistical comparisons were conducted using R v.4.1.0 (R Core Team 2021).

Skull osteology

In order to identify diagnostic osteological characters and evaluate cranial variability within this group, we compared High Resolution X-ray Computed Tomography (HRCT) of newly collected material from Angola with Prosymna data provided by Heinicke et al. (2020), and two additional species (P. janii and P. cf. frontalis) not included in the previous study. We generated and analysed HRCT of three newly collected specimens from Angola (PEM R23512 [male], PEM R23510 [male] and PEM R24013 [female]), one specimen of P. angolensis (SAM ZR16574 [female]), one P. cf. frontalis (PEM R17997 [male]) and one P. janii (PEM R08679 [male]) at Stellenbosch University CT Scanner Facility using a General Electric Nanotom S system, using the settings specified in Appendix 1: Table A2. All specimens were regarded as adults. Three-dimensional segmentation models were generated for the articulated skull in Avizo Lite 2020.2 (Thermo Fisher Scientific 2020). To facilitate visualization, individual bone units for skulls and jaws were coloured following the same colour palette as Lobón-Rovira and Bauer (2021). All rendered files (*.ply) of premaxilla, palatine and maxilla, were analysed in Mesh Lab (2020.2) for colour processing and stacking of reconstructed in lateral, dorsal, ventral or medial views. Annotations were made in Adobe Illustrator CC 22.0.1 (Adobe Systems Incorporated 2017) following the anatomical terminology of Heinicke et al. (2020) and Broadley (1980). CT-scan raw data (.tiff files) have been deposited in MorphoSource (www.morphosource.org; Project ID 435270, Appendix 1: Table A2).

Mapping

To enable the production of up-to-date occurrence maps for the Prosymna angolensis group, data were sourced from published datasets (e.g., Broadley 1980, 1995; Marques et al. 2018) and museum databases. Online virtual museum platforms (http://www.inaturalist.org, http://vmus.adu.org.za, http://www.the-eis.com/atlas/; all accessed 31 July 2022) were also consulted and only a single record from the Namibian Atlas project was found that agreed with the diagnostic features identified by Broadley (1980) for P. angolensis. We gathered a total of 55 records that were assignable to Prosymna angolenis sensu lato. The online GeoNames gazetteer (http://www.geonames.org/) or GEOLocate Web Application (https://www.geo-locate.org/web/WebGeoref.aspx) were used to georeference all historical data. Distribution data were mapped in QGIS v. 3.2 (http://qgis.org).

Results

Phylogenetic analyses

Maximum likelihood (bootstrap support [BS]) and both Bayesian Inference analyses (MrBayes posterior probability [MBPP]; BEAST posterior probability [BPP]) showed strong support for the monophyly of Prosymna (BS 100%, MBPP 1.0, BPP 1.0). (Fig. 2). The genus was characterised by four clades, identical to those found in Heinicke et al. (2020): meleagris group, south-western taxa, ambigua group, and sundevalli group. Whilst the inter-group structuring lacked support across the different algorithms, most of the relationships between sister species within the four groups were well supported.

Figure 2.

a Maximum likelihood (IQTREE) concatenated phylogeny with BI BEAST and BI MrBayes support overlain. Each of the species groups are demarcated by coloured vertical bars at the right margin b maximum likelihood (IQTREE) phylogenies (16S and cyt-b) with gene-specific single-locus species delimitation analyses results overlain. The bars to the right of the phylogeny represent the putative taxa assignments for each analysis and the values beneath the bars denote the total number of putative taxa for each analysis.

The newly sequenced P. angolensis sensu lato material (hereafter ‘angolensis subgroup’) formed a monophyletic group, that was recovered within the sundevalli group by all three algorithms (BS 100%, MBPP 1.0, BPP 1.0). All three algorithms also retrieved identical topological sub-structuring within the sundevalli group, with strong support for the sister relationship between P. lineata and the rest of the species in the group. The sister relationship between P. sundevalli + P. bivittata and the angolensis subgroup was however not supported by any of the algorithms. Whilst all the phylogenies recovered the angolensis subgroup as monophyletic, they differed in their sub-structuring, with ML (BS 100%) and BEAST (BPP 1.0) recovering Prosymna ‘Coastal’ as sister to Prosymna ‘Eastern’ + P. angolensis sensu stricto and MrBayes (MBPP 1.0) recovering P. angolensis sensu stricto as sister to Prosymna ‘Coastal’ + Prosymna ‘Eastern’. All three algorithms recovered similar sub-structuring within the Prosymna ‘Eastern’ lineage with a specimen from Quembo River bridge camp (PEM R27381, Appendix 1: Table A1) being recovered as sister to the rest of the samples (BS 100%, MBPP 1.0, BPP 1.0).

The single locus ML phylogenies (Fig. 2b) produced similar topological structuring to the concatenated dataset, albeit with markedly lower support. Where the topologies do differ, the nodes lack support. Although inter-group support is low, the intra-group support is high and is similar to that observed in multi-locus phylogeny (Fig. 2a). Similar to the multi-locus phylogeny (Fig. 2a), the novel Angolan specimens were recovered as a monophyletic clade with both 16S (BS 100%) and cyt-b (BS 100%) recovering Prosymna ‘Coastal’ as sister to Prosymna ‘Eastern’+ P. angolensis sensu stricto.

The species delimitation analyses employed across both the 16S and cyt-b phylogenies recovered a substantial amount of putative taxa, with vastly different estimates between the different analyses. Across both phylogenies, ABGD and ASAP were more conservative and PTP and bPTP were more liberal with the number of putative taxa. Whilst the differential sampling afforded to the different gene phylogenies resulted in differing numbers of putative taxa, it must be noted that all four species delimitation methods recognised Prosymna ‘Eastern’, Prosymna ‘Coastal’ and P. angolensis sensu stricto as independent species when using both the 16S and cyt-b marker. While not the focus of this study, notable cryptic speciation was also recovered in P. frontalis, P. ambigua, and P. stuhlmanni when using either the 16S or cyt-b marker.

The average 16S pairwise divergence separating species within the genus was 8.11% (± 0.23% s.e. – standard error) when the angolensis subgroup is considered a single species (Table 3). While the average pairwise divergences separating the three lineages of the angolensis subgroup from the rest of the genus varies from 6.99% (± 0.52% s.e.) to 7.99% (± 0.67% s.e.). Although the angolensis subgroup is well delineated from the rest of the genus, when these lineages are compared to one another, the angolensis subgroup was characterised by relatively low pairwise divergences. Prosymna angolensis sensu stricto was separated from Prosymna ‘Eastern’ material by an average pairwise distance of 2.92% (± 0.75% s.e.) and from the Prosymna ‘Coastal’ sample by 3.75% (± 0.89% s.e.), while the Prosymna ‘Eastern’ and Prosymna ‘Coastal’ material were separated by 4.76% (± 0.99% s.e.).

For the cyt-b gene, the average pairwise divergence separating species of the genus was 20.60% (± 0.42% s.e.) (Table 3). The average pairwise divergences separating the three lineages of the angolensis subgroup from the rest of Prosymna varies from 18.39% (± 1.11% s.e.) to 19.08% (± 0.95% s.e.). Like the 16S gene, the angolensis subgroup was characterised by lower pairwise divergences when these three lineages were compared to each other, albeit relatively higher than what was found when using the 16S gene. Prosymna angolensis sensu stricto was separated from Prosymna ‘Eastern’ material by a pairwise distance of 13.90% (± 1.31% s.e.) and from Prosymna ‘Coastal’ material by a pairwise distance of 13.25% (± 1.40% s.e.), while eastern Prosymna ‘Coastal’ material is separated from the Prosymna ‘Eastern’ specimen by 15.55% (± 1.47% s.e.).

Table 3.

Download as

CSV

XLSX

Sequence divergence (uncorrected pairwise distance values) for 16S and cyt-b separating the species of Prosymna. Numbers in the diagonal (in bold) denote intraspecific divergences, numbers below the diagonal denote interspecific divergences and numbers above the diagonal denote the standard error of the interspecific divergences. NA–Not Available.

16S
16S		1	2	3	4	5	6	7	8	9	10	11	12	13
1	P. ambigua	3.02	1.28	1.19	1.23	1.07	1.28	1.15	1.03	1.36	1.31	1.13	1.22	1.19
2	P. angolensis	8.58	NA	1.19	0.89	1.08	1.37	1.21	1.21	0.75	1.39	1.23	1.16	1.30
3	P. bivittata	7.28	6.26	0	1.00	1.12	1.39	1.35	1.09	1.28	1.41	1.23	0.95	1.31
4	P. confusa sp. nov. ‘Coastal’	8.34	3.75	5.07	NA	1.11	1.28	1.29	1.15	0.99	1.30	1.25	1.01	1.32
5	P. frontalis	8.11	7.71	8.01	8.14	5.28	1.22	1.01	0.99	1.15	1.23	0.93	1.08	1.13
6	P. greigerti	9.55	9.96	10.44	8.82	9.84	1.55	1.28	1.30	1.38	0.64	1.34	1.36	1.28
7	P. janii	7.51	7.32	8.63	7.96	6.75	8.35	0	1.02	1.30	1.24	0.93	1.29	1.23
8	P. lineata	6.50	6.73	5.87	6.39	6.93	9.44	5.78	0.69	1.22	1.27	0.92	1.07	1.20
9	P. lisima sp. nov. ‘Eastern’	10.06	2.92	7.53	4.76	8.56	10.85	8.53	7.36	0.63	1.40	1.30	1.25	1.27
10	P. meleagris	9.62	10.11	10.31	9.16	10.26	2.82	8.39	9.47	11.03	2.52	1.28	1.39	1.32
11	P. stuhlmanni	7.58	7.76	7.51	7.52	6.16	9.24	3.96	5.07	8.37	9.31	NA	1.20	1.25
12	P. sundevalli	8.56	7.63	4.97	5.63	7.85	11.11	8.92	6.32	8.16	11.68	8.04	1.54	1.28
13	P. visseri	7.88	8.38	8.01	8.36	8.01	9.07	7.40	7.90	7.76	9.49	7.77	8.64	0.74
*cyt-b*
*cyt-b*		1	2	3	4	5	6	7	8	9	10	11
1	P. angolensis	NA	1.43	1.40	1.38	1.58	1.56	1.31	1.67	1.70	1.39	1.64
2	P. bivittata	15.76	0.33	1.41	1.32	1.60	1.55	1.57	1.70	1.72	1.40	1.64
3	P. confusa sp. nov. ‘Coastal’	13.25	16.35	NA	1.38	1.62	1.60	1.47	1.70	1.72	1.32	1.74
4	P. frontalis	19.95	19.72	19.28	17.97	1.44	1.45	1.44	1.49	1.35	1.34	1.47
5	P. janii	19.64	21.50	19.91	20.17	1.08	1.59	1.57	1.72	1.62	1.51	1.67
6	P. lineata	18.92	17.84	18.36	20.18	20.16	0.22	1.55	1.82	1.73	1.38	1.66
7	P. lisima sp. nov. ‘Eastern’	13.90	18.73	15.55	21.38	19.75	17.67	1.58	1.71	1.69	1.39	1.58
8	P. ruspolii	23.86	24.05	23.75	23.45	24.40	25.26	23.81	NA	1.80	1.67	1.69
9	P. stuhlmanni	20.75	20.92	20.83	18.79	17.97	20.52	20.68	23.47	NA	1.45	1.65
10	P. sundevalli	15.99	16.08	14.63	19.51	18.97	15.39	17.61	23.70	17.52	6.49	1.59
11	P. visseri	21.87	22.29	22.38	21.27	20.33	22.57	21.73	25.04	19.56	20.49	3.09

Morphology

There was a degree of overlap in most morphological features (measurements and scale counts) for all the material examined (Table 2). Although Prosymna ‘Eastern’ has lower number of ventral scales (116–129 [average 121] compared to Prosymna angolensis sensu stricto (126–163 [average 145]), the results of the non-parametric Wilcoxon test showed that there are no significant differences in the ventral and subcaudal scale counts when corrected for size for both sexes (Fig. 3). The only other consistent scalation differences between the P. angolensis sensu stricto and Prosymna ‘Eastern’ material, was the number of postoculars: predominantly has one postocular (only 4 of 31 had two postoculars) vs. always have two postoculars (n = 10), respectively. Due to small sample size, the morphology of the single specimen of Prosymna ‘Coastal’ could not be statistically compared with the other two lineages. It exhibited similar scalation to P. angolensis sensu stricto (i. e., presence of a higher ventral scales count and presence of a single postocular) (Table 2), but differ in dorsal colouration (see below).

Figure 3.

Summary boxplots (top whisker–maximum value; lower whisker–minimum value; dark horizontal line–median; box–1^st and 3^rd quartiles, open circles–data points) comparing ventral and subcaudal scales corrected for size among the species of Prosymna separated by sex: Prosymna angolensis (blue) and Prosymna lisima sp. nov. (red); p-value of non-parametric Wilcoxon test is indicated at bottom of each boxplot. Prosymna confusa sp. nov. is not included due to small sample size.

Colouration

The colouration of P. angolensis sensu stricto (24 out of 28 examined) varied from pale grey to yellow-brown with a large black bar behind the head and a series of smaller paired black spots along the back, similar to P. sundevalli. In some cases (2 out of 28 examined), these black spots are very faint and disappear anteriorly or form continuous faint paravertebral stripes (2 out of 28 examined), similar to P. lineata (Fig. 4). On the other hand, all material from Prosymna ‘Eastern’ (n = 10) exhibited a golden yellow dorsum colouration with a large black bar behind the head, followed by fused irregular black blotches along the back continuing onto the tail (Fig. 5). The only subadult male collected had smaller paired dorsal black spots, similar to the main population but more defined (Fig. 5D). Although we could not statistically compare the morphology of the Prosymna ‘Coastal’ specimen, it has uniform dark grey dorsum with a very faint black bar behind the head (Fig. 6).

Figure 4.

Variation in dorsal colouration of nominotypical Prosymna angolensis A Tundavala, Huíla Province, Angola (Photo: Justin R. Nicolau) B Grootfontein, Namibia (Photo: Francois Theart) C Bicuar National Park (CHL 0521), Huíla Province, Angola (Photo: Ninda L. Baptista).

Figure 5.

Photos of live P. lisima sp. nov. from eastern Angola A PEM R23457 from Quembo River Source, Moxico Province, Angola B PEM R23483 from Cuando River Source, Moxico Province, Angola C PEM R23512 from Cuito Source Lake, Moxico Province, Angola D PEM R27381 from Quembo River bridge camp, Moxico Province, Angola.

Figure 6.

Live P. confusa sp. nov. (PEM R24013) from 20 km west of Lola on the road northwest to Camacuio and on the edge of Bentiaba River, Namibe Province, Angola (Photo: Bill Branch).

Skull osteology (Figs 7–9)

The osteological analysis has shown that material within the angolensis subgroup presents the same common features shared across Prosymna: a compact and rigid skull, anterior reduction of the maxilla, enlargement of the posterior maxillary teeth and reduced palatine, and a unique tooth loci formula with seven reduced tooth loci and four or five posterior lancet-shaped and enlarged tooth loci (see Heinicke et al. 2020). Some specimens in the angolensis subgroup had more than four frontal foramina present, such as in P. angolensis sensu stricto and Prosymna ’Eastern’, vs. one or two in Prosymna ‘Coastal’.

Figure 7.

A Lateral B ventral C medial D dorsal views of skulls of Prosymna angolensis subgroup. Red arrows depict variation characters between the P. lisima sp. nov., P. angolensis, P. confusa sp. nov., P. cf. frontalis, and P. janii.

The angolensis subgroup shares some cranial features with the rest of the sundevalli group, such as an elongated maxillary process, with the premaxilla being in contact with the maxilla and the lack of a postorbital bone (except Prosymna ‘Eastern’, see below). It differs from the sundevalli group by the absence of partial or total fusion between the braincase and the parietal bone and the absence of lateral tubercles in the premaxilla in angolensis subgroup (vs. present in sundevalli group).

Figure 8.

A Frontal and B posterior views of skulls, and C lateral, dorsal, medial and ventral view of jaw of Prosymna angolensis subgroup. Red arrows depict variation in characters between the P. lisima sp. nov., P. angolensis, P. confusa sp. nov., P. cf. frontalis, and P. janii.

The Prosymna ‘Eastern’ material is unique in possessing a well-developed postorbital bone, which is shared with P. cf. frontalis (although much more reduced) and the ambigua group (well developed, including here corroborated for P. janii). Two of the angolensis subgroup lineages (P. angolensis sensu stricto and Prosymna ‘Eastern’) present an unfused braincase (only known to be present in ambigua group, here corroborated for P. janii); however, the Prosymna ‘Coastal’ specimen presents a fused braincase.

Figure 9.

A Dorsal, frontal and ventral view of premaxillae B medial, posterior and dorsal views of palatine; and C lateral, dorsal, and ventral view of maxilla in P. lisima sp. nov., P. angolensis, P. confusa sp. nov., P. cf. frontalis and P. janii (from left to right).

Finally, the osteological comparison demonstrated the presence of unique diagnostic features between the three lineages from the angolensis subgroup, i.e., presence/absence of postorbital bone, fused/unfused braincases, and the number of palatine teeth and frontal foramina, which are addressed in more detail below.

Systematics

Based on the genetic, morphological, and colouration differences discussed above, we recognise all three lineages within the angolensis subgroup as independently evolving lineages and describe two (Prosymna ‘Eastern’ and Prosymna ‘Coastal’) of them as new species. We follow the general lineage-based species concept (de Queiroz 1998).

Reptilia: Squamata: Prosymnidae

Prosymna angolensis Boulenger, 1915

Figs 4, 7, 8, 9, 10 Common names: Angolan Shovel-snout snake (English); Cobra-de-focinho-de-pá-de-Angola (Portuguese).

Chresonymy.1

Prosymna frontalis: Bocage 1873: 218, 1882: 288, 1895: 98; Boulenger 1894: 248, 1896: 641.

Prosymna ambigua: Monard 1931: 104, 1937: 123; Mertens 1937: 13.

Prosymna ambigua ambigua: Mertens 1938: 439; Loveridge 1958: 151.

Prosymna angolensis: Boulenger 1915: 208; Chabanaud 1916: 439; Monard 1937: 114, 122; Bogert 1940: 59; Mertens 1955: 94, 1971: 86; Hellmich 1957: 66; Loveridge 1958: 149; FitzSimons 1962: 161, 1966: 53, 1970: 104; Isemonger 1968: 129; Broadley 1980: 512, 1990: 227, 1995: 48; Auerbach 1987: 178; Branch 1998: 84, 2018: 64; Broadley et al. 2003: 187 (in part); Marais 2004: 236; Broadley and Blaylock 2013: 219; Herrmann and Branch 2013: 8; Wallach et al. 2014: 568; Baptista et al. 2019: 118; Chippaux and Jackson 2019: 283, Ceríaco et al. 2021a: 16 (in part).

When Boulenger (1915) described P. angolensis, he did not designate a precise type locality nor a type specimen for that matter. Later, Loveridge (1958) proposed designating Huíla as the type locality, but it was Broadley (1980) that finally restricted the type locality to Caconda by designating a lectotype of the material he examined on his visit to Museu Bocage, Lisbon (MBL), Portugal in 1968. The reasons for this change in the proposed type locality, were that the Huíla specimen was unaccounted for, as well as the Caconda material was in the best overall condition to represent the species. He initially designated MBL 1606b as the lectotype, but with the destruction of the MBL collection, he designated one of the remaining Caconda specimens in Naturhistorischen Museums in Wien (NMW 19275b) as the replacement neotype (see Gemel et al. 2019).

Material examined

Neotype (Fig. 10). NHMW 19275:2, collected from Caconda (approx. -13.73537, 15.06720, 1662 m a.s.l.), Huíla Province, Angola. Neotype designated by Donald Broadley (1980). Additional material. MBL 1609, Angola (no precise locality), Angola; MBL 1605a, Bibala, Angola; MBL 1605b, Bibala, Angola; CHL 0521, Bicuar NP, Angola; NMW 19275:1, Caconda, Angola; NMW 19275:2, Caconda, Angola; MBL 1606a, Caconda, Angola; MBL 1606b (original lectotype), Caconda, Angola; MBL 1606c, Caconda, Angola; MBL 1608, Caconda, Angola; USBN 20035, Luanda, Angola; CAS 84181, Luanda, 3 mile S of airport, Angola; MBL 1607, Maconjo = Maconge, Angola; MCZ 32468, Missão do Dondi Bela Vista, Angola; MBL 1604, interior of Mossamedes, Angola; UM 20178, Goeverega, Botswana; UM 21271, 15 km WSW of Katima Mulilo, Namibia; SMF 46614, Karakuwisa, Kavango, Namibia; TM 55043, Katima Mulilo, Namibia; UM 24204, Katima Mulilo, Namibia; SAM ZR16574, Namutoni, Namibia; NMZB 9532, NE of Waterberg, Namibia; NMZB 13953, Inyokene, Nyamandhlovo, Zimbabwe; NMZB 13787, Malinbdi Siding, Hwange, Zimbabwe; NMZB 13788, Malinbdi Siding, Hwange, Zimbabwe.

Figure 10.

Neotype of Prosymna angolensis (NMW 19275:2) from Caconda, Huíla Province, Angola (Photos: Alice Schumacher, Natural History Museum Vienna).

General description

See Table 2 for summarised meristic data. Dorsal scales smooth, arranged in 17-15-15 (rarely 19-15-15) rows at midbody, scale row reduction takes place between ventral scales 16–20 (males) and 14–49 (females); one (sometimes two or three on supracaudal scales) apical pits; 126–163 (126–155 males, 134–163 females) smooth ventral scales; 16–28 (22–28 males, 16–25 females) paired subcaudal scales; rostral is acutely angular horizontally; internasal is single and bandlike; 1 preocular; 1 (rarely 0 or 2) postocular; temporals mostly 1+2 (rarely 2+2, 2+3); mostly 6 supralabials, with 3^rd and 4^th entering the orbit (rarely 5 (2, 3) or 7 (3, 4)); 7 infralabials, with first 3 in contact with the chin shield (rarely 8 (3)), cloacal scale entire.

Skull osteology and teeth (Figs 7–9). Based on the examination of a single female specimen (SAM ZR16574, Namutoni, Namibia), P. angolensis presents a compact and rigid skull, which is common among Prosymna species. It has an unfused braincase and nasal bones. Parietals are fused and the fronto-lateral portion presents a sharp edge that forms the edge of the orbital rim. Postorbital bone is absent. Premaxilla has a reduced ascending nasal process with a small groove between the ascending process and frontal portion of the bone. Premaxilla lies between the ventral laminae of nasals with a high profile of the anterior portion which curves shapely to finish on a convex profile. Maxilla and premaxilla are in contact. Nasal bones are reduced and display a wing-shape with a narrower anterior portion. Septomaxilla is a well-developed bone, in broad contact with the premaxilla, frontal, vomer, prefrontal and frontal bones. Vomer well developed with perforated dorsolateral portion of the bone. Maxilla reduced anteriorly with an elongated pick-shaped palatine process, with seven or eight laterally reduced curved tooth loci, followed by four enlarged and lancet-shaped tooth loci. Palatine with three reduced teeth and an enlarged dorsal and curved vomerine process that reinforces the internal portion of the orbit. Pterygoid is a thin elongated bone. Supratemporal is an enlarged bone in broad contact with the quadrate and participates in the lateral movement of the lower jaw. The lower jaw consists of compound, splenial, coronoid, and dentary bones. Coronoid and splenial bones are reduced, almost vestigial. Dentary with eight or nine small sharp tooth loci, with first third clear of any teeth.

Colouration in life (Fig. 4). The head is yellowish-brown with variable darker black markings that can be absent. Most commonly there is an anterior black band across the frontal, followed by a pair of black blotches around the orbits, supraoculars and parietals. A distinct black nuchal spot or collar is often present. The dorsal colouration varies from having mostly small paired black longitudinal vertebral spots (similar to P. sundevalli) to a speckled pattern (similar to P. lineata) on a pale yellowish-brown to grey ground colour. Ventrum and outermost two or three dorsal scale rows yellowish white.

Hemipenis. Short hemipenis with a length that reaches the 9–10^th ventral scales (Broadley 1980).

Size. Males vary from 160–248 (208.7 ± 29.8) mm SVL and 22–30 (26.5 ± 2.7) mm TL, with the largest male measuring 248+28 = 276 mm (NMZB 9532, NE of Waterberg, Namibia). Females vary from 127–305 (224.7 ± 51.1) mm SVL and 12–27 (19.9 ± 3.7) mm TL, with the largest female measuring 305+22 = 327 mm (SMF 32541, Cubal, Angola). Bocage (1895) mentioned an unsexed individual (probably a female) that measured 331+29 = 360 mm, but this specimen was unaccounted for in MBL.

Natural history

This is a semi-fossorial species that feeds exclusively on reptile eggs, using its blade-like rear maxillary teeth to puncture the eggs, similar to other Prosymna species.

Distribution and habitat

Currently the species is known to occur in three main geographic clusters: west-central Angola, north-central Namibia and isolated records from the Zambezi Region in north-eastern Namibia, northern Botswana and north-western Zimbabwe (Fig. 2). However, it is possible that this distribution might be more continuous, given this is a rarely observed species that mostly emerges to the surface only after good rains (Heinicke et al. 2020). The records from Luanda (USBN 20035 and CAS 84181) and northern Angola (IICT/R 14-1957) require verification. This species is associated with savanna with an annual rainfall of 500–1200 mm (Broadley 1980). In southwestern Angola it has been found in miombo woodland in sandy soils (Baptista et al. 2019). In the eastern Zambezi Region and northern Botswana, the species is associated with drier savanna in deep Kalahari sands (Broadley 1980).

Localities

Angola: Bela-Vista (Missão do Dondi), -12.36667, 16.20000 (Hellmich 1957: 66); interior of Benguela (Bocage 1895: 98); Bibala, -14.76667, 13.36667 (Bocage 1873: 218); Bicuar National Park, Woodland trapline 1, -15.09441, 14.83831 (Baptista et al. 2019: 118); Caconda, -13.73537, 15.06720 (Bocage 1895: 151); Cubal, -13.03333, 14.73333 (Mertens 1938: 439); Ebanga, -12.73333, 14.73333 (Monard 1937: 123); Huíla, -15.08333, 13.55000 (Bocage 1895: 98); Luanda and ‘Luanda, 3 mi S of airport’, -8.83333, 13.26667 (Broadley 1980: 515); Maconjo, -15.016667, 13.2000 (Bocage 1895: 98); interior of Mossamedes (Bocage 1873: 218); Quibula, -12.28333, 14.68333 (Bocage 1895: 98); Posto do Milando (-8.81667, 17.56667) (Ceríaco et al. 2021a: 16). Quindumbo, -12.46667, 14.93333 (Bocage 1895: 98); Quissange, -12.43333, 14.05000 (Bocage 1895: 98); Tundavala, -14.82018, 13.404217 (Justin Nicolau photo). Botswana: Joverega (Geoverega), -19.13333, 24.25 (Broadley 1980: 515). Namibia: Grootfontein, -19.55012, 18.10965 (Francois Theart photo); Karakuwisa, -18.933333, 19.733333 (Mertens 1955: 94); Katima Mulilo, -17.5, 24.266667 (Broadley 1980: 515); Namutoni, -18.807624, 16.940288 (FitzSimons 1962: 161); 15 km WSW of Katima Mulilo, -17.61448, 24.205932 (Broadley 1980: 515); Otjozondjupa Region, -19.08800, 18.83300 (http://www.the-eis.com/atlas/?q=details/snake-record&occurrence_id=654386). Zimbabwe: Malindi Siding, Hwange, -18.74885, 27.01852 (Broadley 1995: 48); Inyokene, Nyamandlovu, -19.93333, 28.06667 (Broadley 1995: 48).

Prosymna lisima sp. nov.

https://zoobank.org/70A338BA-ED44-440B-AE3B-863E84B3AFBE

Figs 5, 7, 8, 9, 11, 12 Common names: Kalahari Shovel-snout snake (English); Cobra-de-focinho-de-pá-do-kalahari (Portuguese).

Chresonymy

Prosymna angolensis: Broadley 1971: 82, 1980: 512 (in part); Broadley et al. 2003: 187 (in part); Pietersen et al. 2021: 97, fig.; Conradie et al. 2021: 265.

Material examined

Holotype (male). PEM R23512, collected from Cuito River source lake (-12.68866, 18.36025, 1426 m a.s.l.), Moxico Province, Angola by Werner Conradie and James Harvey on 26 November 2016. Paratypes (five males). PEM R27381, collected from Quembo River lower bridge (-13.526579, 19.278096, 1248 m a.s.l.), Moxico Province, Angola by Werner Conradie, Chad Keates and Timóteo Júlio on 27 November 2019; PEM R23457–8, collected from Quembo River source (-13.13586, 19.04709, 1368 m a.s.l.), Moxico Province, Angola by Werner Conradie on 3 November 2016; PEM R23483, Cuando River source (-13.00164, 19.1296, 1372 m a.s.l.) Moxico Province, Angola by Werner Conradie and James Harvey on 17 November 2016; PEM R23510, collected from Cuito River source lake (-12.68866, 18.36025, 1426 m a.s.l.), Moxico Province, Angola by Werner Conradie and James Harvey on 26 November 2016. Paratypes (two females). PEM R23456, collected from Quembo River source (-13.13586, 19.04709, 1368 m a.s.l.), Moxico Province, Angola by Werner Conradie on 3 November 2016; PEM R23511, Cuito River source lake (-12.68866, 18.36025, 1426 m a.s.l.), Moxico Province, Angola by Werner Conradie and James Harvey on 26 November 2016.

Additional material assigned to the new species

NMZB-UM 10096, Kalabo (approx. -14.99391, 22.67795), Zambia; NMZB-UM 21272, 15 km WSW of Katima Mulilo (approx. -17.61448, 24.20593), Namibia. A specimen from Kuvangu [= Vila-da-Ponte], -14.46667, 16.3000 (Monard 1937: 123) has two postoculars and the characteristic confluent blotched dorsal pattern and might belong to this new species, but this needs verification and is thus tentatively referred to the new species.

Diagnosis

The new species differs from other Prosymna in the following characters: rostral sharply depressed and angular (vs. rounded in P. visseri); presences of a single band-like internasals (vs. paired internasals in P. somalica, P. bivittata, P. sudevalli, P. lineata); dorsal scales smooth (keeled in P. janii); midbody scale rows 15–17 (vs. 19–21 in P. pitmani); 6 supralabials, with 3^rd and 4^th entering orbit (vs. 5 supralabials, with 2^nd and 3^rd entering orbit in P. meleagris and P. greigerti); single apical pits on dorsal scales (vs. paired apical pits in P. ruspolii); lower number of ventral scales in both sexes (116–129 vs. 153–199 in P. frontalis); dorsum with dark black spots (vs. scarlet head and dark body in P. ornatissima; uniform dark brown to grey in P. ambigua and P. stuhlmanni). It further differs from its closest congener, P. angolensis, in having two post oculars (vs. one), dorsal large black blotches mostly fused (vs. mostly small paired dorsal grey to black spots), postorbital bone present (vs. absent) and by the presence of four to five well-developed palatine teeth (vs. three reduced teeth).

Etymology

The name lisima is derived from the locally spoken Luchaze language in the region of the type locality meaning ‘source’. The full phrase used, ‘Lisima Lwa Mwondo’, is translated as “source of life”. This is a reference to central Angola, a high rainfall area where some of the most important rivers in Angola arise. This water makes it its way to the Okavango Delta, sustaining wildlife and local communities in Angola, Namibia and Botswana.

Description of holotype

(Fig. 11). See Table 4 for further details and meristic data for the holotype. The body is cylindrical and elongated, tapering gradually to a very short tail, 13% total length, tail tip with a prominent spike. Dorsal scales smooth with single apical pits (some suprasubcaudal scales have two apical pits) in 17-15-15 scale rows, scale row reduction from 17 to 16 take place at ventral number 17 with the fusion of 3^rd and 4^th dorsal scale rows on left side and from 16 to 15 at ventral 23 with the fusion of 3^rd and 4^th dorsal scale rows on right side; 122 ventral scales; cloaca entire; 13 paired subcaudal scales. Head in dorsal view (Fig. 11B): head narrow and rounded, barely wider than ‘neck’; rostral clearly visible from above, much broader than long (3.39 × 1.06 mm); a single narrow internasal, which is much longer than wide (2.73 × 0.67 mm) and in broad contact with the rostral anteriorly, posteriorly in broad contact with prefrontal and laterally with nasals; single band-like prefrontal which is longer than wide (3.70 × 1.30 mm), in contact laterally with loreal, and posteriorly with the frontal and supraocular scales; frontal pentangular, almost as long as wide (3.09 × 3.00 mm), nearly equal distance to snout (3.30 mm), shorter than prefrontals (3.0 vs. 3.70 mm), but nearly equal in length to the parietal scales (3.00 vs. 3.04 mm), in contact laterally with narrow supraoculars, and posteriorly with two very large parietals; paired parietals as wide as long (3.04 × 3.04 mm), in contact posteriorly with each other and laterally with temporals. Head in ventral view (Fig. 11C): rostral clearly visible from below, protruding well past jawline; mental small, triangular; infralabials seven, first three in contact with single paired chin shields, 1^st infralabials in contact with each other; additional three or four rows of smaller gular scales present before start of ventral scales. Head in lateral view (Fig. 11D): snout sharply pointed, longer than the horizontal diameter of eye (ED/SL = 0.49); rostral large with acutely horizontal angular edge, excavated below; nostril is oval shaped, piercing a fully-divided nasal, and directed backwards; nasal scale longer than wide, with anterior part in full contact with rostral, posterior lower corner in contact with 1^st supralabial and above with internasal scale and prefrontal; nasal suture present and intersecting 1^st supralabial in uppermost corner; single small loreal as long as wide (0.7 × 0.7 mm), in contact below with 1^st and 2^nd supralabial, above with prefrontal, anteriorly with nasal and posteriorly with single preocular; a single preocular on the right side and two on the left side in contact anteriorly with loreal and prefrontal and above with supraocular, posteriorly protruding of loreal overlap with preoculars to create a small flap; eye large 19.36% headlight, vertical diameter (1.47 mm) two thirds as deep as distance between eye and lip (0.99 mm); pupil round; two postoculars, the lower one largest and in contact with 4^th and 5^th supralabials, first temporal scale and parietal, the upper smaller in contact with both supraocular and parietal; temporals 1+2 on both sides; narrow elongated supraocular in contact anteriorly with preocular, posteriorly with upper postocular and parietal and above with frontal; six supralabials, 3^rd and 4^th contacting eye, 5^th and 6^th supralabial the largest.

Table 4.

Download as

CSV

XLSX

Morphological features and measurements for the type series of Prosymna lisima sp. nov. (SVL = snout-vent length, t = truncated).

Catalogue number	PEM R23512	PEM R23456	PEM R23457	PEM R23458	PEM R23483	PEM R23510	PEM R23511	PEM R27381
Type status	Holotype	Paratype	Paratype	Paratype	Paratype	Paratype	Paratype	Paratype
Sex	Male	Female	Male	Male	Male	Male	Female	Male
SVL+TL = total length mm	193+28.9 = 221.9	200+19.3 = 219.3	194+25.8 = 219.8	198+24.6 = 222.6	181+23.8 = 204.8	186+26.3 = 212.3	168+20.6 = 188.6	138+17t = 155
TL/total length ratio (%)	13.0	8.8	11.7	11.1	11.6	12.4	10.9	11.0
Midbody scale rows	17-17-15	17-17-15	17-15-15	17-17-15	17-15-15	17-17-15	17-15-15	17-15-15
Ventral scales	122	121	117	124	121	116	117	122
Subcaudal scales	26	18	22	23	22	23	24	21t
Cloacal Scale	Entire	Entire	Entire	Entire	Entire	Entire	Entire	Entire
Preoculars	1/2	1	2	1	1	2	1	1
Postoculars	2	2	2	2	2	2	2	2
Temporals	1+2	1+2	1+2	1+2	1+2	1+2	1+2	1+2+3
Supralabials (contacting eye)	6 (3,4)	6 (3,4)	6 (3,4)	6 (3,4)	5 (2,3)/6 (3,4)	6 (3,4)	6 (3,4)	6 (3,4)
Infralabials (in contact with 1^st chin shield)	7 (3)	7 (3)	7 (3)	7 (3)	7 (3)	7 (3)	7 (3)	7 (3)
Loreal	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Dorsal colouration	26 fused blotches	30 fused blotches	21 fused blotches	31 fused blotches	22 fused blotches	26 fused blotches	27 fused blotches	36 paired blotches

Figure 11.

Holotype of Prosymna lisima sp. nov. (PEM R23512) from Cuito River source, Moxico Province, Angola A dorsal and ventral full body B dorsal head C ventral head D lateral head.

Colouration. In life (Figs 5C, 11B–D). Dorsum bright yellow-brown with 27 irregular fused black blotches that extend along the back from the nape onto the tail. Each dorsal scale has a darker edge giving it a faint reticulated pattern. Dorsolaterally, between the black vertebral blotches, there is a cluster of 2–4 scales with black edges. The large black nape blotch originates at the posterior margin of the frontal and runs through the parietal onto the dorsal scales, and is approximately nine scale rows deep and eleven scale rows wide. Each vertebral black blotch varies from 4–8 scale rows deep and 3–7 scale row wide. Some of the blotches are fused to form a continuous zig-zag pattern. Frontal and prefrontal sutures have a dark edge forming a pale grey crossbar. Eyes black. Ventrum cream-white, with the two outermost dorsal scale rows same colour as ventrum. In preservative (Fig. 11A). Same as in life, but yellow-brown colouration faded and the dark edges became more noticeable. Ventrum beige.

Paratype and additional material variation. See Table 2 and 4 for full meristic data. Dorsal scales smooth and in 17-17-15 rows at midbody; 116–124 (116–124 males, 117–129 females) smooth ventral scales; 18–26 (22–26 males, 18–24 females) paired subcaudal scales; one (rarely two) preoculars; two postoculars; temporals mostly 1+2; mostly six supralabials, with 3^rd and 4^th entering the orbit; seven infralabials, with first three in contact with the chin shield, cloacal scale entire; 21–36 fused dark dorsal spots. Largest female: 275+28 mm (NMZ UM 21272: 15 km WSW of Katima Mulilo); largest male: 198+25 mm (PEM R23458: Quembo River source). The colouration of the type material is in general in agreement with the holotype, except that the dorsal fused blotches vary in size, number and arrangement (Fig. 5A–C). The nape black blotch always originates at the anterior part of the frontal extending through the parietals to 7–9 dorsal scale rows deep, 11–15 scales wide and start from the 3^rd–5^th lateral dorsal scale row. The dorsum consists of 21–36 confluent black blotches that are 7–11 scales wide and three to four scales deep. One specimen (PEM R23456) exhibits a distinct dark interorbital band and internasal band. The only juvenile collected (PEM R27381, Fig. 5C) has small paired black blotches (two scales deep and four scales wide) on a lighter yellow ground colour, large head blotch starts at posterior frontal through parietals, seven scales wide.

Skull osteology and teeth (Figs 7–9). This species presents a compact and rigid skull, common among Prosymna species with unfused braincase and nasal bones. Parietals are fused. Postorbital bone is present and contributes to the posterior edge of the orbital rim. Premaxilla has a short but robust ascending nasal process that lies between the ventral laminae of the nasals with low profile of the anterior portion which gradually slope ending in a moderate narrow tip and two elongated maxillary processes. Maxilla and premaxilla are in contact. Nasal bones are medium large bones in contact with frontal and premaxilla. Septomaxilla is a well-developed bone, in broad contact with the premaxilla, frontal, vomer, prefrontal and frontal bones. Vomer is well developed with a perforated dorsolateral portion. Maxillary is reduced anteriorly with an elongated pick-shaped palatine process with five or six laterally reduced curved tooth loci, followed by four to five enlarged lancet-shaped tooth loci, on same disposition. Palatine with four to five well developed teeth and an enlarged dorsal curved vomerine process. Pterygoid is a thin elongated bone. Supratemporal is in broad contact with the quadrate and participates in the lateral movement of the lower jaw. The lower jaw presents a compound bone, splenial, coronoid and dentary. Coronoid and splenial are reduced, almost vestigial. Dentary with eight tooth loci.

Hemipenis. Short simple structure, only reaching the 6–9^th ventral scale. Single non-bifurcated sulcus. Ornamentation is flounced. Proximal third is smooth. Distal portion with four to five flounces that starts at the sulcal fold and encircle the whole organ, the most proximal often branched, forming a pocket of which the edges is smooth, tapering into a distal point. Retractor muscle is straight.

Natural history notes

All specimens were caught in late November during the rainy season. At this time, many adult lacertids, Ichnotropis capensis and I. cf. grandiceps, were also observed mating in the same habitat. Prosymna are well known to prey on soft-shell lizard eggs, and P. lisima sp. nov. may actively seek out these lacertids’ eggs. Only two of the females had stomach contents, while all the males had empty stomachs. The largest female (PEM R23456) had three empty lizard egg shells in the hind gut, three empty egg shells at the rear end of the stomach, and four undigested lizard eggs in the main stomach (Fig. 12). Another female (PEM R23511) had four undigested lizard eggs in the main stomach. All eggs measured ~ 11.0 mm in length and each had a lateral cut. The eggs in the main stomach also all had a lateral cut but still maintained their shape and were surrounded by calcified leaked yolk (due to the preservation process). The eggs at the rear and the hindgut were all undecomposed and compressed. One of the paratype females (PEM R23456) was gravid, and had three eggs in early developmental stages (16.8 × 4.2 mm). Interestingly, on two different occasions, three specimens (two males and one female) were caught on the same night in the same trap array. This may indicate that males were following females to breed.

Figure 12.

Stomach and gut contents of Prosymna lisima sp. nov. (PEM R23456) consisting of lizard egg shells. Scale bar: 1 mm.

Distribution and habitat

Currently only known from east-central Angola, western Zambia and the Zambezi Region of north-eastern Namibia. In the region of Katima Mulilo (eastern Zambezi Region, Namibia) it occurs in sympatry with its sister species, P. angolensis (Fig. 2). The tentative assigned material from Kuvango (Monard 1931) needs verification, but it agrees in colouration and morphology to the new species. This species is expected to be much more widely distributed in the Kalahari basin, as it seems to be associated with the deep Kalahari sands. The Angolan material occurs in Angolan moist miombo woodland, while the Zambian and Namibian material occurs in dry miombo woodland. The elevation ranges between 950 and 1450 m a.s.l. All newly collected specimens were captured in trap arrays set in sandy areas next to river source lakes or main rivers in eastern Angola (Conradie et al. 2021).

Prosymna confusa sp. nov.

https://zoobank.org/A4E2E3E8-A658-4007-B5F4-46D875457EA1

Figs 6, 7, 8, 9, 13 Common names: Plain Shovel-snout Snake (English); Cobra-de-focinho-de-pá-lisa (Portuguese).

Chresonymy

Prosymna angolensis: Bogert 1940: 59; Monard 1937: 123 (in part); Broadley 1980: 152 (in part).

Prosymna ambigua: Branch 2018: 64, fig. 24; Pietersen et al. 2021: 96, fig.

Monard (1937) was the first to document a uniformly grey specimen from Ebanga. This was followed by Bogert (1940) who documented a specimen from Capelongo (AMNH R50504) that also exhibited a uniform pale brown dorsum with small white spots (similar to P. meleagris pattern). This uniform dorsum colouration is in agreement with the new specimen collected from coastal Angola (Branch 2018) and this colouration is very distinct from the other two species, yellowish grey with paired small black dorsum spots in P. angolensis and bright yellow with fused black blotches in P. lisima sp. nov.

Material examined

Holotype (female). PEM R24013, collected from 20 km west of Lola on the road northwest to Camacuio, on the edge of Bentiaba River (-14.27583, 13.45806, 791 m a.s.l.), Namibe Province, Angola by William R. Branch, Pedro Vaz Pinto and João S. de Almeida on 2 November 2015.

Additional material tentatively assigned to the new species

AMNH 50504, Capelongo, approx. -14.46645, 16.29241, Huíla Province, Angola (Bogert 1940: 59); Ebanga, approx. -12.73333, 14.73333 , Benguela Province, Angola (Monard 1937: 123).

Diagnosis

The new species differs from other Prosymna species in the following characters: rostral sharply depressed and angular (vs. rounded in P. visseri); presence of a single band-like internasal (vs. paired internasals in P. somalica, P. bivittata, P. sundevalli, P. lineata); dorsal scales smooth (keeled in P. janii); midbody scale rows 15–17 (vs. 19–21 in P. pitmani); six supralabials, with 3^rd and 4^th entering orbit (vs. five supralabials, with 2^nd and 3^rd entering orbit in P. meleagris and P. greigerti); single apical pits on dorsal scales (vs. paired apical pits in P. ruspolii); lower number of ventral scales in both sexes (116–129 vs. 153–199 in P. frontalis); dorsum uniform dark grey (vs. scarlet head and dark body in P. ornatissima). It further differs from its closest congeners in the angolensis group: one postocular (vs. two in P. lisima sp. nov.), dorsum uniform grey (vs. dorsum with large mostly fused black blotches in P. lisima sp. nov. and mostly smaller paired longitudinal rows of grey to black spots in P. angolensis), postorbital bone absence (vs. present in P. lisima sp. nov.), presence of two well-develop palatine teeth (vs. four to five in P. lisima sp. nov. and three reduced teeth in P. angolensis), fused braincase (vs. unfused in P. angolensis and P. lisima sp. nov.) and two frontal foramina (vs. three to four in P. angolensis and P. lisima sp. nov.).

Etymology

When the late Bill Branch collected the holotype, he was unsure of its identification and referred to it as an unusual specimen that could not be assigned to any known species from Angola. He later referred to it as P. ambigua (Branch 2018), presumably based on its uniform grey colouration. The name confusa is a reflection of the confusion this specimen has caused and of the general confusion in the P. angolensis group.

Description of holotype

(Fig. 13). Adult female measuring 240 mm SVL+29 mm TL = 269 mm total length. The body is cylindrical and elongated, tapering gradually to a very short tail, 10.8% total length, tail tip with prominent spike. Dorsal scales smooth, with single apical pits, in 17-15-15 scale rows, scale row reduction from 17 to 16 take place at ventral number 29 with the fusion of 3^rd and 4^th dorsal scale row on right side and from 16 to 15 at ventral 32 with the fusion of 3^rd and 4^th dorsal scale row on left side; 151 ventral scales; cloaca entire; 26 subcaudal scales. Head in dorsal view (Fig. 13B): head narrow and rounded, barely wider than ‘neck’; rostral clearly visible from above, much broader than long (2.84 × 1.47 mm); a single narrow internasal, which is much longer than wide (2.40 × 0.57 mm) and in broad contact with the rostral anteriorly, posteriorly in broad contact with prefrontal and laterally with nasals; single band-like prefrontal which is longer than wide (3.71 × 1.25 mm), in contact laterally with loreal and posteriorly with the frontal and supraocular scales; frontal pentangular, almost as long as wide (2.97 × 3.14 mm), nearly equal in length to the distance to snout (2.91 mm), more than double than prefrontal width (3.14 vs. 1.25 mm), and three quarters the length of the parietals (3.14 vs. 2.43 mm), in contact laterally with narrow supraoculars, and posteriorly with two very large parietals; paired parietals longer than wide (2.28 × 2.43 mm), in contact posteriorly with each other and laterally with temporals. Head in ventral view (Fig. 13C): rostral clearly visible from below, protruding well past jawline; mental small, triangular; infralabials eight on right side and nine on left side, first three in contact with single paired chin shields, 1^st infralabials in contact with each other; additional three rows of smaller gular scales present before the start of ventral scales. Head in lateral view (Fig. 13D): snout sharply pointed, longer than the horizontal diameter of eye (ED/SL = 0.45); rostral large with acutely horizontal angular edge, excavated below; nostril is oval shaped, piercing divided nasal, and directed backwards; nasal scale longer than wide, with anterior part in full contact with rostral, posterior lower corner in contact with 1^st supralabial, upper section in contact with internasal scale and prefrontal and posteriorly with loreal and prefrontal; nasal suture present and intersecting loreal; single small loreal as long as wide (0.84 × 0.84 mm), in contact below with 1^st and 2^nd supralabial, above with prefrontal, anteriorly with nasal and posteriorly with single preocular; single preocular on both sides in contact anteriorly with loreal and above with supraocular and prefrontal, posteriorly with loreal; eye large 16.50% HL, vertical diameter (1.53 mm), two thirds as deep as distance between eye and lip (0.42); pupil round; one postoculars, in contact with 4^th upperlabial, 1^st temporal scale, supraocular, and parietal; temporals 1+2; narrow elongated supraocular in contact anteriorly with preocular and prefrontal, posteriorly with the postocular and above with frontal; five supralabials on both sides with 3^rd and 4^th in contact with eye on right and 2^nd, 3^rd and 4^th on left, 5^th and 6^th supralabial the largest.

Figure 13.

Holotype of Prosymna confusa sp. nov. (PEM R24013) from 20 km west of Lola on the road northwest to Camacuio and on the edge of Bentiaba River, Namibe Province, Angola A dorsal and ventral full body B dorsal head C ventral head D lateral head.

Colouration. In life (Fig. 6). Dorsum uniform grey with the anterior edges of scales with a white spot, outermost two to three scale rows white, with only the first outermost scale row of tail white. Nape with a faint collar that is three scale rows wide. The prefrontal and internasal black compared to the rest of the head being grey. Eye black. Ventrum white. In preservative (Fig. 13). Same as in life, but the grey faded and became brown. Ventrum beige.

Additional material variation. See Table 2 for summarised meristic data. Only data of three females were available, but the assignment of historical material to P. confusa sp. nov. still requires confirmation. Dorsal scales smooth and in 17-17-15 rows at midbody; 143–155 smooth ventral scales; 17–26 paired subcaudal scales; one preoculars; one postoculars; temporals 1+2; five or six supralabials, with 3^rd and 4^th entering the orbit; seven infralabials, with first three in contact with the 1^st chin shield, cloacal scale entire. Largest female: 240+29 mm (holotype PEM R24013). The colouration is similar to the holotype, except that in the Capelongo specimen (AMNH R50504) the white spots are much more conspicuous (Fig. 14). The specimen from Ebanga is unaccounted for (Broadley 1980), but Monard (1937) described the colouration as uniform grey above.

Figure 14.

A specimen (AMNH 50504) form Capelongo, approx. -14.46645, 16.29241; Huíla Province, Angola, assignable to the new species, P. confusa sp. nov. (Photos: Lauren Vonnahme, American Museum of Natural History).

Skull osteology and teeth (Figs 7–9). The holotype presents a compact and rigid skull common among Prosymna species with fused braincase, fused parietal, and unfused frontal and nasal bones. Postorbital bone is absent. Parietal with a fronto-lateral sharp edge that participates virtually as posterior edge of the orbital rim. Premaxilla has a well-developed and robust ascending nasal process that lies between the ventral laminae of the nasals with low profile of the anterior portion which gradually slope to a narrow tip and two elongated maxillary process in contact with the maxilla. Nasal bones are medium large bones in contact with frontal and premaxilla. Septomaxilla is a well-developed bone, in broad contact with premaxilla, frontal, vomer, prefrontal and frontal bone. The vomer is well developed with performed dorsolateral portion of the bone. Maxillary is reduced anteriorly with an elongated pick-shaped palatine process with six to seven laterally reduced curved tooth loci, followed by four to five enlarged lancet-shaped tooth loci. Palatine with two well developed teeth and an enlarged dorsal and curved vomerine process. Pterygoid is a thin elongated bone. Supratemporal is in broad contact with the quadrate and participates in the lateral movement of the lower jaw. The lower jaw is comprised of compound, splenial, coronoid, and dentary bones. Coronoid and splenial bones are reduced, almost vestigial. Dentary bone with six tooth loci.

Hemipenis. Unknown. Bogert (1940) suggested it to be similar to P. ambigua. However, Prosymna ambigua is unique in having a very long ‘telescopic’ hemipenis that is longer than the tail, which is not present in the sundevalli group (Broadley 1980). The latter group, to which P. confusa sp. nov. belongs, is characterised by its short hemipenes (8–10^th ventral scales long vs. longer than tail in P. ambigua), low number of flounces (5–6 vs. more than 50 in P. ambigua) and straight retractor muscle (telescopic in P. ambigua). Prosymna frontalis shares a similar hemipenile structure with P. ambigua.

Natural history notes

The holotype was found actively moving around near a large rock outcrop during the day.

Distribution and habitat

This species is endemic to southwestern Angola (Fig. 2), and appears to be associated with mopane woodlands, dry savannas, and semi-desert shrublands (Barbosa 1970). The new specimen was found in sandy plains with scattered low granite outcrops, with varying degrees of short grass cover and scattered bushes. Vegetation included Colophospermum mopane, Ficus sp., Senegalia (= Acacia) mellifera, Commiphora sp., Boscia foetida, and Salvadora persica. The two additional historical specimens from Ebanga and Capelongo that are tentatively assigned to this species occurred in similar dry habitat.

Discussion

The taxonomic status of Prosymna angolensis has been the subject of debate by several researchers over time. Thanks to the access to new material, we are able to describe two new species and provide the first phylogenetic placement of P. angolensis. Our work recovered a similar topology to that of Heinicke et al. (2020), with four clades characterising Prosymnidae. The addition of the new Angolan samples did not help resolve the alpha taxonomy or provide better support for deeper relationships between and within groups, as in Heinicke et al. (2020). However, our sampling does allow for a broader understanding of the phylogenetic relationships in the sundevalli group.

While the phylogenetic placement between lineages within the angolensis subgroup remains unresolved (particularly P. angolensis sensu stricto) and species pairwise distances are below those of their congeners; the topological support offered by ML and BEAST coupled with species delimitation analyses lends support for the distinctiveness of the two novel taxa at a genetic level. A single, near topotypical, sample of P. angolensis was available for molecular analysis and, based on the disjunct distribution between the populations of Angola and Namibia, Botswana, and Zimbabwe of P. angolensis, cryptic diversity is expected and worth further investigation.

Whilst our phylogeny lacks the resolution necessary to resolve the inter-specific relationships within the sundevalli group, which is likely a product of incomplete taxon sampling, it is clear that the angolensis subgroup is within the larger sundevalli group, as suggested by Broadley (1980). Our work further identified two new lineages within the angolensis subgroup, one of which agrees with differences in morphology and colouration previously reported from western Zambia and the Zambezi Region (Broadley 1980). Although the ‘eastern race’ was partly defined based on the lower ventral scale counts (Broadley 1980), we show here that the differences are not significant and that these might be related to the sizes of the animals collected. Studies have shown positive correlation between ventral and subcaudal scales and body size, and that caution should be taken when used as a diagnostic feature (Lindell 1994; Lee et al. 2016). Based on the clear morphological (head scalation and osteology), dorsal colouration, and genetic differences reported in this study, we here describe these two lineages as P. lisima sp. nov. and P. confusa sp. nov. This raises the total number of Prosymna species to 18, with six species occurring in Angola. This number is, however, expected to grow as further cryptic diversity has been identified in P. frontalis, P. ambigua, and P. stuhlmanni (Heinicke et al. 2020) that may warrant taxonomic re-evaluation.

Because we only compared a small series of skulls for the osteological characterisation (one or two scans per species), further intra-specific variation may exist, thus the osteological differences observed should be taken with caution. That said, there were some striking differences among the groups, as mentioned by Heinicke et al. (2020). The postorbital was only recorded in P. ambigua and P. stuhlmanni and suggested as vestigial in P. greigerti by Heinicke et al. (2020). Here we confirm the presence of a well-developed postorbital bone in P. janii, the only remaining member of the ambigua group. Additionally, we recorded the presence of a postorbital bone in two additional species, P. lisima sp. nov. and P. cf. frontalis, being almost absent in the latter. This shows that the postorbital bone development is not restricted to the ambigua group but rather, present in the south-western taxa and the sundevalli group. This bone was present at a well-developed state in the ambigua group (reaching the maxilla) and P. lisima sp. nov. (slightly reduced, only reaching midway to the maxilla), but being small, almost absent in P. cf. frontalis. The purpose of the postorbital bone is currently unknown, but it may be associated with additional muscle attachment points to aid in feeding or crushing lizard eggs. This may suggest that the postorbital bone has been lost several times from its ancestral trait as a consequence of diet specialisation. Therefore, we recommend further research along this line, to shed light on the evolutionary history of this group.

In Prosymna cf. frontalis, the anterior portion of the maxilla has no teeth, similar to that in P. visseri. Heinicke et al. (2020) speculated, based on the fact that P. frontalis is also more rupiculous than its fossorial relatives, they may also have lost these anterior maxillae to allow them to feed on more hard-shelled gecko eggs which are found sympatrically. Although P. confusa sp. nov. may occur in similar habitat to P. cf. frontalis, it does not share the same maxillary tooth arrangement, which may indicate a different feeding strategy, such as feeding on sympatrically occurring soft-shelled lacertid eggs (e.g., Heliobolus sp. and Pedioplanis sp.). Prosymna confusa sp. nov. shares similar dentary development to P. cf. frontalis, in that the most anterior section of the dentary bone is free of teeth, compared to other species where they extend almost to the tip of the dentary.

The descriptions of two new species, P. confusa sp. nov., endemic from dry habitats in southwestern Angola, and P. lisima sp. nov., associated to the Kalahari sands from Angola to neighbouring countries to the east, are an indication of how much diversity is likely still to be described from these regions. In the last decade a renewed interest in the Angolan herpetofauna has led to numerous expeditions to remote areas. Consequently, the number of species recorded from the country, has increased considerably (Conradie et al. 2012a, 2012b, 2013, 2020; Stanley et al. 2016; Ceríaco et al. 2018, 2020a, 2020b, 2020c, 2021b; Branch et al. 2019, 2021; Marques et al. 2019a, 2019b, 2020, 2022a, 2022b; 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). Among these are four new species of snakes (Conradie et al. 2020; Hallermann et al. 2020). The addition of two new species of Prosymna brings the number of recorded snake species from Angola to ca. 135. This is approximately 69% of the known squamate diversity recorded from the country. As more research is conducted on Angolan herpetofauna, many more new species descriptions and additions are expected. Finally, we raise the importance of further surveys in this poorly studied region of Africa, with the aim of collecting recent material that will allow us to clarify the taxonomic placement of several species’ complexes and poorly understood species.

Acknowledgements

We thank the Wild Bird Trust, which administrates the National Geographic Okavango Wilderness Project (2015–2019 National Geographic Society grant). Material was collected and exported under the following export permits issued by the Angolan Ministry of Environment Institute of Biodiversity (MINAMB): 31/GGPCC/2016, 151/INBAC/MINAMB/2019). Fieldwork in Bicuar National Park was carried out in the framework of the Southern African Science Service Centre for Climate Change and Adaptive Land Management (SASSCAL) project, sponsored by the German Federal Ministry of Education and Research (BMBF) under promotion number 01LG1201M. The genetic component of this work was funded through the Wild Bird Trust grant to WC (grant WBT01 and OWP017). We would also like to acknowledge the use of infrastructure and equipment provided by the NRF-SAIAB Aquatic Genomics Research Platform and the funding channeled through the NRF-SAIAB Institutional Support System. We thank Luke Verburgt, James Harvey, Timóteo Júlio, Pedro Vaz Pinto, and the late Bill Branch for field assistance or collecting valuable specimens. Francois Theart and Justin Nicolau kindly provided comparative photographs from Namibia and Angola, respectively. We thank Enviro-Insight for constructing and donating the drift fences required for the trapping animals. WC thanks the Eastern Cape Province Department of Sport, Recreation, Arts and Culture (DSRAC) and Port Elizabeth Museum (Bayworld) for granting special leave to take part in these surveys. JLR and NLB are currently supported by Fundação para a Ciência e Tecnologia (FCT) contracts PD/BD/140808/2018 and PD/BD/140810/2018, respectively. Ethical clearance for the study was obtained from the Port Elizabeth Museum (Bayworld) ethics committee (Ethical Clearance no. 2013 & 2017-2). We thank Sheila Broadley who kindly made available Don Broadley’s data on the Prosymna angolensis group. Photos of key museum specimens was kindly provided by Georg Gassner and Alice Schumacher (Natural History Museum Vienna), Jose Rosado, Breda Zimkus, and Joseph Martinez (Museum of Comparative Zoology, Harvard University), Lauren Vonnahme (American Museum of Natural History), and Jofred Opperman (Iziko Museums of South Africa). We also further thank Jofred Opperman (Iziko Museums of South Africa) for facilitating the loan of specimens to be CT-scanned. We thank Muofhe Tshibalanganda (Central Analytic Facilities, Stellenbosch University) for performing the CT-scanning. We further want to express our gratitude to the two reviewers, Krystal Tolley and Luís Ceríaco, whose edits helped to improve the paper.

References

Auerbach RD (1987) The Amphibians and Reptiles of Botswana. Mokwepa Consultants, Gaborone, Botswana, 295 pp.

Baptista NL, Vaz Pinto P, Keates C, Edwards S, Roedel M-O, Conradie W (2021) A new species of red toad, Schismaderma Smith, 1849 (Anura: Bufonidae), from central Angola. Zootaxa 5081(3): 301–332. https://doi.org/10.11646/zootaxa.5081.3.1

Baptista NL, António T, Branch WR (2019) The herpetofauna of Bicuar National Park and surroundings, southwestern Angola: a preliminary checklist. Amphibian & Reptile Conservation 13(2)[Special Section]: 96–130[e203].

Barbosa LAG (1970) Carta Fitogeográfica de Angola. Instituto de Investigação Científica de Angola, Luanda, Angola, 323 pp.

Bocage JVB (1873) Mélanges erpétologiques. II. Sur quelques Reptiles et Batraciens nouveaux, rares ou peu connus d’frique occidentale. Jornal de Sciencias Mathematicas, Physicas e Naturaes 4: 209–227.

Bocage JVB (1882) Noticia ácerca de alguns reptis d’Angôche que existem no Museu Nacional de Lisboa. Jornal de Sciencias Mathematicas, Physicas e Naturaes. Academia Real das Sciencias de Lisboa VIII(32): 286–290.

Bocage JVB (1895) Herpétologie d’Angola et du Congo. Ministério da Marinha e das Colónias, Lisbonne, Portugal, 203 pp. [20 pls]

Bogert CM (1940) Herpetological Results of the Vernay Angola expedition. Bulletin of the American Museum of Natural History 77: 1–107.

Bouckaert R, Vaughan TG, Barido-Sottani J, Duchêne S, Fourment M, Gavryushkina A, Heled J, Jones G, Kühnert D, De Maio N, Matschiner M, Mendes FK, Müller NF, Ogilvie HA, du Plessis L, Popinga A, Rambaut A, Rasmussen D, Siveroni I, Suchard MA, Wu C-H, Xie D, Zhang C, Stadler T, Drummond AJ (2019) BEAST 2.5: An advanced software platform for Bayesian evolutionary analysis. PLoS Computational Biology 15(4): e1006650. https://doi.org/10.1371/journal.pcbi.1006650

Boulenger GA (1894) Catalogue of the snakes in the British Museum (Natural History). Volume II. Containing the conclusion of the Colubriae Aglyphæ. British Museum (Natural History), London, [xi,] 382 pp.

Boulenger GA (1896) Catalogue of the snakes in the British Museum (Natural History). Volume III. Containing the Families Colubridae (Opisthoglyphae and Proteroglyphae), Amblycephalidae, and Viperidae. British Museum (Natural History), London, [xiv,] 727 pp.

Boulenger GA (1915) A list of the snakes of the Belgian and Portuguese Congo, northern Rhodesia, and Angola. Proceedings of the Zoological Society of London 1915(2): 193–223. https://doi.org/10.1111/j.1469-7998.1915.tb07411.x

Branch WR (1998) Field Guide to the Snakes and other Reptiles of Southern Africa. Revised Edition. Struik Publishers, Cape Town, 399 pp.

Branch WR (2018) Snakes of Angola: An annotated checklist. Amphibian & Reptile Conservation 12(2)[General Section]: 41–82[e159].

Branch WR, Conradie W, Vaz Pinto P, Tolley KA (2019b) Another Angolan Namib endemic species: a new Nucras Gray, 1838 (Squamata: Lacertidae) from southwestern Angola. Amphibian & Reptile Conservation 13(2): 82–95.

Branch WR, Schmitz A, Lobón-Rovira J, Baptista NL, António T, Conradie W (2021) Rock island melody: A revision of the Afroedura bogerti Loveridge, 1944 group, with descriptions of four new endemic species from Angola. Zoosystematics and Evolution 87(1): 55–82. https://doi.org/10.3897/zse.97.57202

Broadley DG (1971) The reptiles and amphibians of Zambia. The Puku 6: 1–143.

Broadley DG (1979) Predation on Reptile Eggs by African Snakes of the Genus Prosymna. Herpetologica 35: 338–341.

Broadley DG (1980) A revision of the African snake genus Prosymna Gray (Colubridae). Occasional papers of the National Museums of Rhodesia, Series B. Nature and Science 6: 481–556.

Broadley DG (1990) FitzSimons’ Snakes of southern Africa. Revised Edition. Jonathan Ball and Ad. Donker Publishers, Parklands, 376 pp.

Broadley DG (1995) Geographical Distribution - Prosymna angolensis. African Herp News 23: 48.

Broadley D, Blaylock R (2013) The Snakes of Zimbabwe and Botswana. Chimaira, Frankfurt, 387 pp.

Broadley DG, Doria CT, Wigge J (2003) Snakes of Zambia. An Atlas and Field Guide. Edition Chimaira, Frankfurt, 280 pp.

Bruford MW, Hanotte M, Brookfield JFY, Burke T (1992) “Single locus and multilocus DNA fingerprint” in Molecular Genetic Analysis of Populations: A Practical Approach. IRL Press, Oxford, 225–270.

Ceríaco LMP, Marques MP, Bandeira S, Agarwal I, Stanley EL, Bauer AM, Heinicke MP, Blackburn DC (2018) A new earless species of Poyntonophrynus (Anura, Bufonidae) from the Serra da Neve Inselberg, Namibe Province, Angola. ZooKeys 780: 109–136. https://doi.org/10.3897/zookeys.780.25859

Ceríaco LMP, Agarwal I, Marques MP, Bauer AM (2020a) A review of the genus Hemidactylus Goldfuss, 1820 (Squamata: Gekkonidae) from Angola, with the description of two new species. Zootaxa 4746(1): 1–71. https://doi.org/10.11646/zootaxa.4746.1.1

Ceríaco LMP, Heinicke MP, Parker KL, Marques MP, Bauer AM (2020b) A review of the African snake-eyed skinks (Scincidae: Panaspis) from Angola, with the description of a new species. Zootaxa 4747(1): 77–112. https://doi.org/10.11646/zootaxa.4747.1.3

Ceríaco LMP, Agarwal I, Marques MP, Bauer AM (2020c) A correction to a recent review of the genus Hemidactylus Goldfuss, 1820 (Squamata: Gekkonidae) from Angola, with the description of two additional species. Zootaxa 4861(1): 92–106. https://doi.org/10.11646/zootaxa.4861.1.6

Ceríaco LMP, Parrinha D, Marques MP (2021a) Saving collections: Taxonomic revision of the herpetological collection of the Instituto de Investigação Científica Tropical, Lisbon (Portugal) with a protocol to rescue abandoned collections. ZooKeys 1052: 85–156. https://doi.org/10.3897/zookeys.1052.64607

Ceríaco LMP, Santos BS, Marques MP, Bauer AM, Tiutenko A (2021b) Citizen science meets specimens in old formalin filled jars: A new species of Banded Rubber Frog, genus Phrynomantis (Anura: Microhylidae) from Angola. Alytes 38(1–4): 18–48.

Chabanaud P (1916) Revision du genre Prosymna Gray. Bulletin du Muséum National d’Histoire Naturelle 1916: 433–440.

Chernomor O, Von Haeseler A, Minh BQ (2016) Terrace Aware Data Structure for Phylogenomic Inference from Supermatrices. Systematic Biology 65(6): 997–1008. https://doi.org/10.1093/sysbio/syw037

Chippaux J-P, Jackson K (2019) Snakes of Central and Western Africa. Johns Hopkins University Press, Baltimore, 448 pp.

Conradie W, Branch W, Measey GJ, Tolley KA (2012a) A new species of Hyperolius Rapp, 1842 (Anura: Hyperoliidae) from the Serra da Chela mountains, southwestern Angola. Zootaxa 3269(1): 1–17. https://doi.org/10.11646/zootaxa.3269.1.1

Conradie W, Measey GJ, Branch WR, Tolley KA (2012b) Revised phylogeny of African sand lizards (Pedioplanis), with the description of two new species from southwestern Angola. African Journal of Herpetology 61(2): 91–112. https://doi.org/10.1080/21564574.2012.676079

Conradie W, Branch WR, Tolley KA (2013) Fifty shades of grey: giving colour to the poorly known Angolan Ashy reed frog (Hyperoliidae: Hyperolius cinereus), with the description of a new species. Zootaxa 3635(3): 201–223. https://doi.org/10.11646/zootaxa.3635.3.1

Conradie W, Deepak V, Keates C, Gower DJ (2020) Kissing cousins: a review of the African genus Limnophis Günther, 1865 (Colubridae: Natricinae), with the description of a new species from north-eastern Angola. African Journal of Herpetology 69(1): 79–107. https://doi.org/10.1080/21564574.2020.1782483

Conradie W, Baptista NL, Verburgt L, Keates C, Harvey J, Júlio T, Neef G (2021) Contributions to the herpetofauna of the Angolan Okavango-Cuando-Zambezi river drainages. Part 1: Serpentes (snakes). Amphibian & Reptile Conservation 15(2)[General Section]: 244–278[e292].

Conroy CJ, Papenfuss T, Parker J, Hahn NE (2009) Use of Tricaine Methanesulfonate (MS222) for Euthanasia of Reptiles. Journal of the American Association for Laboratory Animal Science [JAALAS] 48: 28–32.

de Queiroz K (1998) The general lineage concept of species, species criteria, and the process of speciation: A conceptual unification and terminological recommendations. In: Howard DJ, Berlocher SH (Eds) Endless forms: Species and speciation. Oxford University Press, Oxford, UK, 470 pp.

Dowling HG (1951a) A proposed standard system of counting ventral scales in snakes. British Journal of Herpetology 5: 97–99. https://doi.org/10.2307/1437542

Dowling HG (1951b) A proposes method of expressing scale reductions in snakes. Copeia 2(2): 131–134. https://doi.org/10.2307/1437542

Drummond AJ, Rambaut A (2007) BEAST: Bayesian evolutionary analysis by sampling trees. BMC Evolutionary Biology 7(1): e214. https://doi.org/10.1186/1471-2148-7-214

Figueroa A, McKelvy AD, Grismer LL, Bell CD, Lailvaux SP (2016) A species-level phylogeny of extant snakes with description of a new colubrid subfamily and genus. PLoS ONE 11(9): e0161070. https://doi.org/10.1371/journal.pone.0161070

FitzSimons VFM (1962) Snakes of southern Africa. Purnell and Sons, Cape Town, South Africa, 424 pp.

FitzSimons VFM (1966) A check-list, with synoptic keys, to the snakes of southern Africa. Annals of the Transvaal Museum 25: 25–79.

FitzSimons VFM (1970) A field guide to the snakes of southern Africa. Collins, London, 231 pp.

Gemel R, Gassner G, Schweiger S (2019) Katalog der Typen der Herpetologischen Sammlung des Naturhistorischen Museums Wien–2018. Annalen des Naturhistorischen Museums in Wien, Serie B 121: 33–248.

Hall TA (1999) BioEdit, a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucleic Acids Symposium Series 41: 95–98.

Hallermann J, Ceríaco LMP, Schmitz A, Ernst R, Conradie W, Verburgt L, Marques MP, Bauer AM (2020) A review of the Angolan House snakes, genus Boaedon Duméril, Bibron and Duméril (1854) (Serpentes: Lamprophiidae), with description of three new species in the Boaedon fuliginosus (Boie, 1827) species complex. African Journal of Herpetology 69(1): 29–78. https://doi.org/10.1080/21564574.2020.1777470

Heinicke MP, Daza J, Greenbaum E, Jackman TF, Bauer AM (2014) Phylogeny, taxonomy and biogeography of a circum-Indian Ocean clade of leaf-toed geckos (Reptilia: Gekkota), with a description of two new genera. Systematics and Biodiversity 12(1): 23–42. https://doi.org/10.1080/14772000.2013.877999

Heinicke MP, Titus-McQuillan JE, Daza JD, Kull EM, Stanley EL, Bauer AM (2020) Phylogeny and evolution of unique skull morphologies in dietary specialist African shovel-snouted snakes (Lamprophiidae: Prosymna). Biological Journal of the Linnean Society. Linnean Society of London 131(1): 136–153. https://doi.org/10.1093/biolinnean/blaa076

Hellmich W (1957) Die reptilienausbeute der Hamburgischen Angola Expedition. Mitteilungen aus dem Hamburgischen Zoologischen Museum und Institut 55: 39–80.

Herrmann H-W, Branch WR (2013) Fifty years of herpetological research in the Namib Desert and Namibia with an updated and annotated species checklist. Journal of Arid Environments 93: 94–115. https://doi.org/10.1016/j.jaridenv.2012.05.003

Hoang DT, Chernomor O, Von Haeseler A, Minh BQ, Vinh LS (2018) UFBoot2: Improving the ultrafast bootstrap approximation. Molecular Biology and Evolution 35(2): 518–522. https://doi.org/10.1093/molbev/msx281

Isemonger RM (1968) Snakes of Africa. 1^st Edn. Books of Africa, Cape Town, 263 pp.

Kalyaanamoorthy S, Minh BQ, Wong TKF, Von Haeseler A, Jermiin LS (2017) ModelFinder: Fast model selection for accurate phylogenetic estimates. Nature Methods 14(6): 587–589. https://doi.org/10.1038/nmeth.4285

Kelly CMR, Barker NP, Villet MH, Broadley DG (2009) Phylogeny, biogeography, and classification of the snake superfamily Elapoidea: A rapid radiation in the late Eocene. Cladistics 25(1): 38–63. https://doi.org/10.1111/j.1096-0031.2008.00237.x

Kumar S, Stecher G, Li M, Knyaz C, Tamura K (2018) MEGA X: Molecular Evolutionary Genetics Analysis across Computing Platforms. Molecular Biology and Evolution 35(6): 1547–1549. https://doi.org/10.1093/molbev/msy096

Lee JL, Thompson A, Mulcahy DG (2016) Relationships between numbers of vertebrae, scale counts, and body zize, with implications for taxonomy in Nightsnakes (Genus: Hypsiglena). Journal of Herpetology 50(4): 616–620. https://doi.org/10.1670/15-066

Lindell LE (1994) The evolution of vertebral number and body size in snakes. Functional Ecology 8(6): 708–719. https://doi.org/10.2307/2390230

Lobón-Rovira J, Bauer AM (2021) Bone-by-bone: A detailed skull description of White-headed dwarf gecko Lygodactylus picturatus (Peters, 1870). African Journal of Herpetology 70(2): 1–20. https://doi.org/10.1080/21564574.2021.1980120

Lobón-Rovira J, Conradie W, Iglesias DB, Ernst R, Veríssimo L, Baptista N, Vaz Pinto P (2021) Between sand, rocks and branches: An integrative taxonomic revision of Angolan Hemidactylus Goldfuss, 1820, with description of four new species. Vertebrate Zoology 71: 465–501. https://doi.org/10.3897/vz.71.e64781

Loveridge A (1958) Revision of five African snake genera. Bulletin of Comparative Zoology 1191: 1–198.

Marais J (2004) A Complete Guide to the Snakes of Southern Africa, 2^nd edn. Struik Publishers, Cape Town, 312 pp.

Marques P, Ceríaco LMP, Blackburn DC, Bauer AM (2018) Diversity and Distribution of the Amphibians and Terrestrial Reptiles of Angola - Atlas of Historical and Bibliographic Records (1840–2017). Proceedings of the California Academy of Science (Series 4) 65: 1–501[Supplement II].

Marques MP, Ceríaco LMP, Bandeira S, Pauwels OSG, Bauer AM (2019a) Description of a new long-tailed skink (Scincidae: Trachylepis) from Angola and the Democratic Republic of the Congo. Zootaxa 4568(1): 51–68. https://doi.org/10.11646/zootaxa.4568.1.3

Marques MP, Ceríaco LMP, Stanley EL, Bandeira SA, Agarwal I, Bauer AM (2019b) A new species of Girdled Lizard (Squamata: Cordylidae) from the Serra da Neve Inselberg, Namibe Province, southwestern Angola. Zootaxa 4668(4): 503–524. https://doi.org/10.11646/zootaxa.4668.4.4

Marques MP, Ceríaco LMP, Buehler MD, Bandeira SA, Janota JM, Bauer AM (2020) A revision of the Dwarf Geckos, genus Lygodactylus (Squamata: Gekkonidae), from Angola, with the description of three new species. Zootaxa 4853(3): 301–352. https://doi.org/10.11646/zootaxa.4853.3.1

Marques MP, Parrinha D, Santo BS, Bandeira S, Butler BO, Sousa CAN, Bauer AM, Wagner P (2022a) All in all it’s just another branch in the tree: A new species of Acanthocercus Fitzinger, 1843 (Squamata: Agamidae), from Angola. Zootaxa 5099(2): 221–243. https://doi.org/10.11646/zootaxa.5099.2.4

Marques MP, Ceríaco LMP, Heinicke MP, Chehouri RM, Conradie W, Tolley KA, Bauer A (2022b) The Angolan Bushveld Lizards, genus Heliobolus Fitzinger, 1843 (Squamata: Lacertidae): integrative taxonomy and the description of two new species. Vertebrate Zoology [in press].

Mertens R (1937) Reptilien und Amphibien aus dem südlichen Inner-Afrika. Abhandlungen der Senckenbergischen Naturforschenden Gesellschaft 20: 425–443.

Mertens R (1938) Amphibien und Reptilien aus Angola gesmmelt von W. Shack. Senckenbergiana 20: 425–443.

Mertens R (1955) Die Amphibien und Reptilien Südwestafrikas. Aus den Ergebnissen einer im Jahre 1952 ausgeführten Reise. Abhandlungen der Senckenbergischen Naturforschenden Gesellschaft 490: 1–172.

Mertens R (1971) Die Herpetofauna Südwest-Afrikas. Abhandlungen der Senckenbergischen Naturforschenden Gesellschaft 529: 1–110.

Miller MA, Pfeiffer W, Schwartz T (2010) Creating the CIPRES Science Gateway for Inference of Large Phylogenetic Trees. In: 2010 Gateway Computing Environments Workshop (GCE), New Orleans, LA, 1–8. https://doi.org/10.1109/GCE.2010.5676129

Minh BQ, Lanfear R, Trifinopoulos J, Schrempf D, Schmidt HA (2021) IQ-TREE version 2.1.2: Tutorials and Manual Phylogenomic software by maximum likelihood. http://www.iqtree.org/doc/iqtree-doc.pdf [accessed 21 March 2022]

Monard A (1931) Mission scientifique Suisse dans l’Angola. Résultats scientifiques. Reptiles. Bulletin de la Société Neuchâteloise des sciences Naturelles 33: 89–111.

Monard A (1937) Contribution à l’Herpétologie d’Angola. Arquivos do Museu Bocage 8: 19–154.

National Geographic Okavango Wilderness Project (2017) Initial Findings from Exploration of the Upper Catchments of the Cuito, Cuanavale, and Cuando Rivers, May 2015 to December 2016. Unpublished report, 368 pp.

Nguyen LT, Schmidt HA, Von Haeseler A, Minh BQ (2015) IQ-TREE: A fast and effective stochastic algorithm for estimating maximum-likelihood phylogenies. Molecular Biology and Evolution 32(1): 268–274. https://doi.org/10.1093/molbev/msu300

Nielsen SV, Conradie W, Ceríaco LMP, Bauer AM, Heinicke MP, Stanley EL, Blackburn DC (2020) A new species of Rain Frog (Brevicipitidae, Breviceps) endemic to Angola. ZooKeys 979: 133–160. https://doi.org/10.3897/zookeys.979.56863

Palumbi SR (1996) The polymerase chain reaction. In: Hillis D, Moritz C, Mable B (Eds) Molecular Systematics. 2^nd Edn. Sinauer Associates, Sunderland, 205–247.

Parrinha D, Marques MP, Heinicke MP, Khalid F, Parker KL, Tolley KA, Childers JL, Conradie W, Bauer AM, Ceríaco LMP (2021) A revision of Angolan species in the genus Pedioplanis Fitzinger (Squamata: Lacertidae), with the description of a new species. Zootaxa 5032(1): 001–046. https://doi.org/10.11646/zootaxa.5032.1.1

Peters WCH (1867) Űber eine Sammlung von Flederthieren und Amphibien aus Otjimbingue in Sűdwest-afrika, welche Hr. Missionär Hahn dem zoolgischen museum zugesandt hat. Monatsberichte der Königlichen Preussische Akademie der Wissenschaften zu Berlin 4: 133–138.

Pietersen DW, Verburgt L, Davies J (2021) Snakes and other reptiles of Zambia and Malawi. Struik Nature, Cape Town, 376 pp.

Puillandre N, Lambert A, Brouillet S, Achaz GJME (2012) ABGD, Automatic Barcode Gap Discovery for primary species delimitation. Molecular Ecology 21(8): 1864–1877. https://doi.org/10.1111/j.1365-294X.2011.05239.x

Puillandre N, Brouillet S, Achaz G (2021) ASAP: Assemble species by automatic partitioning. Molecular Ecology Resources 21(2): 609–620. https://doi.org/10.1111/1755-0998.13281

Pyron RA, Burbrink FT, Wiens JJ (2013) A phylogeny and revised classification of Squamata, including 4161 species of lizards and snakes. BMC Evolutionary Biology 13(1): e93. https://doi.org/10.1186/1471-2148-13-93

R Core Team (2021) R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna. https://www.R-project.org/

Rambaut A (2014) FigTree. Tree Figure Drawing Tool Version 1.4.4. [Available from:] http://tree.bio.ed.ac.uk/

Rambaut A, Drummond AJ (2009) Tracer. [Available at:] http://beast.bio.ed.ac.uk/Tracer

Ronquist F, Teslenko M, Van Der Mark P, Ayres DL, Darling A, Hohna S, Larget B, Liu L, Suchard MA, Huelsenback JP (2012) MrBayes 3.2: Efficient Bayesian Phylogenetic Inference and Model Choice Across a Large Model Space. Software for Systematic and Evolution 61(3): 539–542. https://doi.org/10.1093/sysbio/sys029

Slowinski JB, Lawson R (2002) Snake phylogeny: Evidence from nuclear and mitochondrial genes. Molecular Phylogenetics and Evolution 24(2): 194–202. https://doi.org/10.1016/S1055-7903(02)00239-7

Spawls S, Howell K, Hinkel H, Menegon M (2018) Field guide to East African reptiles. Bloomsbury, London, 624 pp.

Stanley EL, Ceríaco LMP, Bandeira S, Valerio H, Bates MF, Branch WR (2016) A review of Cordylus machadoi (Squamata: Cordylidae) in southwestern Angola, with the description of a new species from the Pro-Namib desert. Zootaxa 4061(3): 201–226. https://doi.org/10.11646/zootaxa.4061.3.1

Suchard MA, Rambaut A (2009) Many-core algorithms for statistical phylogenetics. Bioinformatics (Oxford, England) 25(11): 1370–1376. https://doi.org/10.1093/bioinformatics/btp244

Swofford DL (2003) PAUP* Phylogenetic Analysis Using Parsimony (*and other methods). Version 4.0a 169 Sinauer Associates. https://doi.org/10.1002/0471650129.dob0522

Tamura K, Stecher G, Peterson D, Filipski A, Kumar S (2013) MEGA6: Molecular Evolutionary Genetics Analysis Version 6.0. Molecular Biology and Evolution 30(12): 2725–2729. https://doi.org/10.1093/molbev/mst197

Thermo Fisher Scientific (2020) Thermo Fisher Scientific Reports Fourth Quarter and Full Year 2020 Results.

Uetz P, Freed P, Hošek J [Eds] (2022) The Reptile Database. http://www.reptile-database.org [accessed 11 April 2021]

Vidal N, Branch WR, Pauwels OSG, Hedges SB, Broadley DG, Wink M, Cruaud C, Joger U, Nagy ZT (2008) Dissecting the major African snake radiation: A molecular phylogeny of the Lamprophiidae Fitzinger (Serpentes, Caenophidia). Zootaxa 1945(1): 51–66. https://doi.org/10.11646/zootaxa.1945.1.3

Wagner P, Butler BO, Ceríaco LM, Bauer AM (2021) A new species of the Acanthocercus atricollis (Smith, 1849) complex (Squamata, Agamidae). Salamandra (Frankfurt) 57: 449–463.

Wallach V, Williams KL, Boundy J (2014) Snakes of the World: A catalogue of living and extinct species. [type catalogue] CRC Press, Florida, 1237 pp.

Werner F (1929) Übersicht der Gattungen und Arten der Schlangen der Familie Colubridae. III. Zoologische Jahrbucher. Abteilung fur Systematik, Geographie und Biologie der Tiere 57: 1–481.

Whiting AS, Bauer AM, Sites JW (2003) Phylogenetic relationships and limb loss in sub-Saharan African scincine lizards (Squamata: Scincidae). Molecular Phylogenetics and Evolution 29(3): 582–598. https://doi.org/10.1016/S1055-7903(03)00142-8

Xia X (2013) DAMBE5: A comprehensive software package for data analysis in Molecular Biology and Evolution. Molecular Biology and Evolution 30(7): 1720–1728. https://doi.org/10.1093/molbev/mst064

Zaher H, Murphy RW, Arredondo JC, Graboski R, Machado-Filho PR, Mahlow K, Montingelli GG, Quadros AB, Orlov NL, Wilkinson M, Zhang Y-P, Grazziotin F (2019) Large-scale molecular phylogeny, morphology, divergence-time estimation, and the fossil record of advanced caenophidian snakes (Squamata: Serpentes). PLoS ONE 14(5): e0216148. https://doi.org/10.1371/journal.pone.0216148

Zhang J, Kapli P, Pavlidis P, Stamatakis A (2013) A general species delimitation method with applications to phylogenetic placements. Bioinformatics 29(22): 2869–2876. https://doi.org/10.1093/bioinformatics/btt499

Appendix 1

Table A1.

Download as

CSV

XLSX

List of samples used in this study and their associated metadata. NA–Not Available.

Species	Voucher ID	Locality	Molecular Markers
Species	Voucher ID	Locality	16S	ND2	cyt-b	c-mos	RAG1	ENC-1
Ingroup
Prosymna angolensis	NB 0521/CHL 0521	Bicuar National Park, Huíla, Angola	OP288036	OP289543	OP289533	OP289553	NA	NA
P. confusa sp. nov. ‘Coastal’	AG 014/PEM R24013	20 km W Lola, road northwest to Camacuio, Namibe, Angola	OP288044	OP289552	OP289542	OP289562	NA	NA
P. lisima sp. nov. ‘Eastern’	WC-4694/PEM R23456	Quembo River Source, Moxico, Angola	OP288037	OP289544	OP289534	OP289554	NA	NA
P. lisima sp. nov. ‘Eastern’	WC-4696/PEM R23457	Quembo River Source, Moxico, Angola	OP288038	OP289545	OP289535	OP289555	NA	NA
P. lisima sp. nov. ‘Eastern’	WC-4706/PEM R23458	Quembo River Source, Moxico, Angola	OP288039	OP289546	OP289536	OP289556	NA	NA
P. lisima sp. nov. ‘Eastern’	WC-4810/PEM R23483	Cuando River Source, Moxico, Angola	OP288040	OP289547	OP289537	OP289557	NA	NA
P. lisima sp. nov. ‘Eastern’	WC-4870/PEM R23510	Cuito Source Lake, Moxico, Angola	OP288041	OP289548	OP289538	OP289558	NA	NA
P. lisima sp. nov. ‘Eastern’	WC-4863/PEM R23511	Cuito Source Lake, Moxico, Angola	OP288042	OP289549	OP289539	OP289559	NA	NA
P. lisima sp. nov. ‘Eastern’	WC-4862/PEM R23512	Cuito Source Lake, Moxico, Angola	OP288043	OP289550	OP289540	OP289560	NA	NA
P. lisima sp. nov. ‘Eastern’	WC-6844/PEM R27381	Quembo River bridge camp, Moxico, Angola	NA	OP289551	OP289541	OP289561	NA	NA
P. ambigua	CAS 258670	Cangandala NP, Malanje, Angola	MT453136	MT460633	NA	MT460596	MT460655	NA
P. ambigua	GJ 2762	Malanje, Angola	MT453137	MT460634	NA	MT460597	MT460656	NA
P. ambigua	PEM R16763	Gnimeti River, Klein’s Camp, Loliondo Game Controlled Area, Mara, Tanzania	MT453112	NA	NA	MT460579	MT460635	NA
P. bivittata	AMB (53F6)	NA	MT453113	MT460610	NA	NA	NA	NA
P. bivittata	PEM R17431	Mkhuze Falls Private Game Reserve, KwaZulu-Natal, South Africa	MT453114	MT460611	NA	NA	MT460636	NA
P. bivittata	PEM R17432	Mkhuze Falls Private Game Reserve, KwaZulu-Natal, South Africa	MT453115	MT460612	MT482412	MT460580	MT460637	MT460598
P. bivittata	PEM R17433	Mkhuze Falls Private Game Reserve, KwaZulu-Natal, South Africa	MT453116	MT460613	MT482413	MT460581	MT460638	NA
P. cf. frontalis	MCZ R193166	Farm Omandumba, Erongo, Namibia	MT453117	MT460614	NA	NA	MT460639	NA
P. cf. frontalis	PEM R17996	Espinheira, Namibe, Angola	MT453118	MT460615	MT482414	MT460582	MT460640	MT460599
P. cf. frontalis	MCZ R184857	Hobatere Lodge, Namibia	MT453119	NA	NA	NA	MT460641	NA
P. frontalis	MCZ R185112	Farm Oas, Karas, Namibia	MT453120	MT460616	NA	NA	NA	NA
P. frontalis	MCZ R185003	Geister Schlucht, Klein Aus Vista, Karas, Namibia	MT453121	MT460617	KR814693	KR814680	MT460642
P. greigerti	E 107.19	NA	JF340124	NA	NA	NA	NA	NA
P. greigerti	E 124.1	NA	JF340125	NA	NA	NA	NA	NA
P. greigerti	UTEP 22161	Pian Upe Wildlife Reserve, Hyena Hill, Northern Region, Uganda	NA	MT460632	NA	MT460595	MT460654	NA
P. janii	JM 1045	KwaZulu-Natal, South Africa	MT453122	MT460618	MT482415	NA	MT460643	NA
P. janii	MCZ Z37878	KwaZulu-Natal, South Africa	MT453123	NA	NA	NA	NA	NA
P. janii	AMB (SNH8)	KwaZulu-Natal, South Africa	MT453124	MT460620	MT482417	MT460583	MT460644	MT460601
P. janii	PEM R12072	Madlangula, Kosi Bay Nature Reserve, KwaZulu-Natal, South Africa	FJ404222	NA	FJ404319	FJ404293	NA	NA
P. janii	PEM R17372	uMkhuze, Greater St. Lucia Wetland Park, KwaZulu-Natal, South Africa	NA	MT460619	MT482416	NA	NA	MT460600
P. lineata	MCZ R184472	7.5 km E of Musina on Tishipe Rd., Limpopo, South Africa	MT453126	MT460622	MT482419	MT460585	MT460646	MT460602
P. lineata	JM 1816	Khamai, Limpopo, South Africa	MT453127	MT460623	NA	MT460586	MT460647	NA
P. lineata	MBUR 394	Makgabeng area, Limpopo, South Africa	MT453128	MT460624	MT482420	NA	NA	MT460603
P. lineata	AMB (53F8)	Blouberg, Limpopo, South Africa	MT453125	MT460621	MT482418	MT460584	MT460645	NA
P. meleagris	E 107.22	NA	JF340122	NA	NA	NA	NA	NA
P. meleagris	E 113.2	NA	JF340123	NA	NA	NA	NA	NA
P. meleagris	MVZ245380	Shai Hills, Greater Accra Region, Ghana	MT453135	MT460631	NA	MT460594	MT460653	NA
P. ruspolii	CMRK316	Tanzania	NA	NA	DQ486347	DQ486171	NA	NA
P. stuhlmanni	LHM-000270	3.2 km W of Lesheba Wilderness Reserve, Limpopo, South Africa	NA	MT460625	MT482421	MT460587	NA	NA
P. stuhlmanni	PEM R17402	Mkhuze Game Reserve, KwaZulu-Natal, South Africa	MT453129	MT460626	MT482422	MT460588	MT460648	MT460604
P. sundevalli	MCZ R184401	Farm Newstead, Eastern Cape, South Africa	MT453130	MT460627	MT482423	MT460589	MT460649	MT460605
P. sundevalli	MCZ R184512	23.9 km NE Vaalwater on Rd. to Melkrivier, Limpopo, South Africa	MT453131	MT460628	MT482424	MT460590	MT460650	MT460606
P. visseri	CAS 214753	Opuwo Dist. ca 2 km N of Sesfontein, Kunene Region, Namibia	MT453132	MT460629	MT482425	MT460591	MT460651	MT460607
P. visseri	PEM R17994	Espinheira, Namibe, Angola	MT453133	MT460630	MT482426	MT460592	MT460652	MT460608
P. visseri	MG 302	Farm Kaross, Kunene Region, Namibia	MT453134	NA	MT482427	MT460593	NA	MT460609
Outgroup
Psammophis condanarus	NA	NA	Z46479	AY058991	AF471075	AF471104	NA	NA
Pseudaspis cana	NA	NA	AY611898	AY058992	AY612080	AY611989	NA	NA
Aparallactus werneri	NA	NA	AY188045	NA	AF471035	AF471116	NA	JN881241
Atractaspis corpulenta	NA	NA	AY611837	NA	AY612020	AY611929	NA	JN881242
Boaedon fuliginosus	NA	NA	KX249802	NA	KM519712	AF544686	KM519725	JN881267
Buhoma procterae	NA	NA	AY611818	NA	AY612001	AY611910	NA	NA
Lampropeltis getula	NA	NA	KX694649	MG672874	MG672798	KX694811	MG673016	JN881266
Leioheterodon madagascariensis	NA	NA	AY188061	AY059010	AY188022	AF544685	NA	NA
Naja kaouthia	NA	NA	KX277260	AY059008	AF217835	AY058938	JF412633	JN881273
Oxyrhabdium leporinum			NA	NA	AF471029	DQ112081	NA	NA

Table A2.

Download as

CSV

XLSX

High Resolution X-ray Computed Tomography (HRCT) parameters used to scan Prosymna skulls.

Species	Catalogue Number	Voxel (mm)	Voltage (kV)	Current (μA)	Exposure Time (secs)	MorphoSource
Prosymna angolensis	SAM ZR16574	0.00499991	60	300	0.5	https://doi.org/10.17602/M2/M435305
Prosymna confusa sp. nov.	PEM R24013	0.00799991	90	100	0.5	https://doi.org/10.17602/M2/M435274
Prosymna lisima sp. nov.	PEM R23512	0.00699991	60	300	0.5	https://doi.org/10.17602/M2/M435310
Prosymna lisima sp. nov.	PEM R23510	0.00699991	60	300	0.5	https://doi.org/10.17602/M2/M435343
Prosymna cf. frontalis	PEM R17997	0.00699991	100	80	0.5	https://doi.org/10.17602/M2/M435295
Prosymna janii	PEM R08679	0.00699991	90	80	0.5	https://doi.org/10.17602/M2/M435300

1 Broadley (1980) listed Werner (1929: 142) as chersonomy of Prosymna angolensis, but we have removed it in this work as it refers to Psammophis angolensis (Bocage, 1872).