Zoosyst. Evol. 99 (2) 2023, 345-361 | DOI 10.3897/zse.99.101943 eee BERLIN Diversity of Sand Snakes (Psammophiidae, Psammophis) in the Horn of Africa, with the description of a new species from Somalia wv Jiti Smid’, Sergio Matilla Fernandez!, Hassan Sh Abdirahman Elmi*?, Toma’ Mazuch* Department of Zoology, National Museum, Cirkusova 1740, Prague, Czech Republic Department of Zoology, Faculty of Science, Charles University, Vinicnad 7, Prague, Czech Republic Department of Biology, Amoud University, Boorama, Somaliland Department of Forest Ecology, Mendel University, 61300 Brno, Czech Republic FW NM https://zoobank. org/216E01C8-8854-4763-BC82-9ACA678B697C Corresponding author: Jiti Smid (jirismd@gmail.com) Academic editor: Johannes Penner Received 14 February 2023 # Accepted 19 May 2023 Published 27 June 2023 Abstract The biological diversity of the Horn of Africa is one of the least studied in the world. Yet the Horn supports rich communities of spe- cies that are mostly endemic to the region. Here we study the diversity of Sand Snakes (Psammophis) in East Africa, their phylogeny and systematics. Previous studies have unveiled several cryptic and potentially undescribed species of Psammophis that occur in the Horn and their taxonomic status has remained unclear to this day. We used sequence data from two mitochondrial and one nuclear genes to reconstruct the phylogeny of the genus, in which we included newly obtained samples of six different Psammophis species from Somalia, Ethiopia, Eritrea, Sudan, and Egypt. Our aim was to assess the status of some of the undescribed species, examine the level of intraspecific genetic variation within individual species, improve our understanding of the species distributions, and contrib- ute to the taxonomy of the genus. Our results confirm the existence of two undescribed species, one in eastern Somalia, which we formally describe as new, and one in southern Ethiopia that we refer to as Psammophis cf. sudanensis in accordance with previous studies. Further, we provide first genetic data for the nominotypical subspecies of P punctulatus and confirm the species status for its subspecies P. trivirgatus. In addition, we provide new genetic data for P. tanganicus from Ethiopia and Somalia, and range exten- sion records for P. rukwae from Eritrea and Ethiopia and for P. aegyptius from Somalia. Our findings contribute considerably to our understanding of the diversity and distribution of Psammophis in East Africa. Key Words East Africa, Eritrea, Ethiopia, phylogeny, reptilia, sand racers, Serpentes Introduction The Horn of Africa is the easternmost projection of the Af- rican continent that juts into the Indian Ocean. It supports a broad spectrum of habitats, from the Ethiopian High- lands that exceed 4500 m in elevation to the harsh low- land deserts of Somalia with barely any precipitation. The entire territory of the Horn is considered one of the global biodiversity hotspots, which are classified as regions of exceptionally high species richness and endemism while also suffering considerable habitat loss (Mittermeier et al. 2004). The Horn harbors a high diversity of squamate reptiles, of which snakes with narrow distribution ranges stand out the most (Lewin et al. 2016), implying that the snake fauna is mostly endemic to the region. At the same time, however, large parts of the Horn remain virtually unexplored for their difficult accessibility, and the fauna of the Horn is vastly understudied, especially from a ge- netic perspective (Smid 2022). Sand Snakes of the genus Psammophis are distributed primarily in Africa where there are 27 out of 33 currently recognized species. Six additional species occur strictly in Asia, and one stretches broadly across both continents (Uetz et al. 2023). They are slender and swift diurnal Copyright Smid, J. et al. This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. 346 hunters, with long heads and large eyes with round pu- pils, and relatively large, grooved rear fangs. The genus is typified by the unusually small male copulatory organs (hemipenes), which are short and thin and lack any or- namentation characteristic of other snakes (Steehouder 1984; de Haan 2003). As a result, the sex is almost im- possible to determine externally in Psammophis. Thirteen species of Psammophis occur in and around the Horn of Africa. Five range broadly across the Horn; these are P. biseriatus Peters, 1881; P. pulcher Boulenger, 1895; P. punctulatus Dumeril, Bibron & Dumeéril, 1854 with two subspecies - punctulatus and trivirgatus Peters, 1878; P. sibilans (Linnaeus, 1758); and P. tanganicus Loveridge, 1940, although the distribution of P. pulcher is very limited as the species is known only from a hand- ful of specimens. Psammophis aegyptius Marx, 1958 and P. schokari (Forskal, 1775) are distributed by the Red Sea along which they penetrate south to Eritrea and north- ern Somalia. The more marginal species to the Horn, but otherwise generally widespread in other parts of Africa, are P. angolensis (Bocage, 1872), P. lineatus (Duméril, Bibron & Dumeéril, 1854), P. mossambicus Peters, 1882, P. orientalis Broadley, 1977, P. rukwae Broadley, 1966, and P. sudanensis Werner, 1919 (Figs 1, 2; Lanza 1990; Largen and Spawls 2010; Spawls et al. 2018). This list is, however, apparently far from complete. The existence of additional phylogenetic lineages that may deserve the rank of species has been confirmed in several phylogenet- ic studies of the genus. For instance, P. sibilans has been found to represent two independent lineages unrelated to each other, the nominotypical one distributed in Egypt and western Ethiopia, the other one in central Ethiopia, and termed in this paper P. sp. Ethiopia (Trape et al. 2019). Similarly, P. sudanensis contains two lineages separated phylogenetically by many other species in between, with samples of the real P. sudanensis originating from Chad and those referred here to as P. cf. sudanensis from Kenya and Tanzania (Kelly et al. 2008; Trape et al. 2019). Trape et al. (2019) also recovered an isolated evolutionary enti- ty in western Ethiopia (sample MBUR 8346; termed here P. sp. Ethiopia 2) that is related to the western African P. phillipsii (Hallowell, 1844) and the southern and eastern African P. mossambicus. And lastly, Vidal et al. (2008) included in their analysis a sample of an undetermined species from extreme eastern Somalia. Its taxonomy has remained unsolved to date (Keates 2021). In this study we analyze new material from the Horn of Africa and adjoining countries to provide further in- sight into the phylogenetic relationships, distribution, and taxonomy of the Psammophis snakes of the region. The newly analyzed material originated from targeted herpetological fieldtrips to the Horn countries that were carried out between 2010 and 2022. We assembled a ge- netic dataset of two mitochondrial and one nuclear gene that was based on published data and which included all known Psammophis species, and we supplemented it by 14 newly analyzed specimens, including specimens that could not be determined with certainty on the basis of zse.pensoft.net Smid, J. et al.: Sand Snakes of the Horn of Africa morphology and which likely represented some of the pu- tative new species. We inferred their phylogenetic posi- tion within the genus and investigated genetic variability of all Psammophis species to assess the taxonomic status of the putative new species in the context of a complete Psammophis phylogeny. Methods Data for genetic analyses We assembled all genetic data that were available for the genus Psammophis on GenBank. To ensure suffi- cient overlap of loci we included only genes that were available for most of the Horn species. These were cy- tochrome b (cyt 6) and NADH dehydrogenase subunit 4 (ND4) from the mitochondrial DNA, and oocyte matura- tion factor MOS (c-mos) from the nuclear DNA. Some species are represented in GenBank by many sequenced individuals. We pruned the dataset to contain only up to three samples per species, provided all the pruned species are monophyletic. It should be noted that the ND4 se- quence of P. biseriatus (sample BK10724, GenBank ac- cession DQ486284) and the c-mos sequence of P. tangan- icus (sample CMRK87, GenBank accession DQ486183) contained too many ambiguous nucleotide positions to allow their inclusion in the dataset. Sequences of Malpolon monspessulanus, Psammophylax variabilis, Psammophylax rhombeatus, and Rhamphiophis rostra- tus, all of which belong to the Psammophiidae, were used to root the tree. We de-novo sequenced 14 samples from different lo- calities across the Horn of Africa and neighboring coun- tries and belonging to P. aegyptius (seven samples from Egypt [2], Sudan [1], Somaliland [4]), P punctulatus punctulatus (one sample from Ethiopia), P. rukwae (two samples from Ethiopia and Eritrea), P sudanensis (one sample from Ethiopia), P. tanganicus (two samples from Ethiopia and Somaliland), and one sample from northern Somalia (central Somaliland) tentatively identified as P. cf. tanganicus. See Fig. 3 for the geographic origin of the samples and Suppl. material 1 for precise locality details. DNA extraction, amplification, and sequencing Genomic DNA was extracted from ethanol-preserved tissue samples using the DNA Mini Kit protocol. We PCR-amplified the three genes using primers and PCR conditions detailed in Smid et al. (2019). The PCR prod- ucts were sequenced from both directions at Macrogen Europe (Amsterdam, the Netherlands). Quality of the raw sequences was inspected and contigs were assem- bled using the Geneious v. 11 software (Kearse et al. 2012). The tRNAs that flank the ND4 and that were am- plified along with it were not included in the dataset be- cause they may be problematic to align with certainty. Zoosyst. Evol. 99 (2) 2023, 345-361 347 Figure 1. Sand Snakes of the Horn of Africa, part 1. a. P. aegyptius (Berbera, Somaliland); b. P. angolensis (South Africa); ¢. P. biseriatus (Voi, Kenya); d. P. lineatus (Nigeria); e. P. mossambicus (Baringo, Kenya); f. P. orientalis (Watamu, Kenya); g. P. pulcher (Bisanadi National Reserve, Kenya). Photo credit T. Mazuch (a); WR. Branch (b), S. Spawls (c, e, f, g), G. Dunger (d). zse.pensoft.net 348 Smid, J. et al.: Sand Snakes of the Horn of Africa > . —— en ee 7 re Figure 2. Sand Snakes of the Horn of Africa, part 2. a. P. punctulatus (Gewane, Ethiopia); b. P. trivirgatus (Mwingi, Kenya); c. P. rukwae (Asmara, Eritrea); d. P. schokari (Kasserine, Tunisia); e. P. sibilans (Debre Zeit, Ethiopia); f. P. cf. sudanensis (Yabelo, Ethiopia); g. P. tanganicus (Mado Gashi, Kenya). Photo credit T. Mazuch (a, ¢, d, f, g), S. Spawls (b, e). zse.pensoft.net Zoosyst. Evol. 99 (2) 2023, 345-361 349 Saudi Arabia Somaliland P. aegyptius P. cf. sudanensis Somalia P. cornusafricae sp. nov. P. punctulatus P. rukwae P. tanganicus 500 km Figure 3. Geographic origin of the new material analyzed in this study. The background map shows the terrain surface. It was de- rived from the SRTM elevation data (Farr and Kobrick 2000) using the hillshade function of ArcMap. All three genes were aligned with MAFFT (Katoh et al. 2019) using the auto strategy. The alignments were subsequently treated with Gblocks (Castresana 2000) to remove ragged ends. No stop codons were detected in the cyt 6 and ND4 alignments, indicating no nucle- ar mitochondrial pseudogenes were amplified. The final concatenated dataset contained 106 tips (102 ingroup + 4 outgroup) and was 2,149 base pairs (bp) long. Phylogenetic analyses The phylogenetic relationships within Psammophis were estimated by means of a Maximum Likelihood (ML) analysis and a Bayesian Inference (BI). The ML analysis was conducted in IQ-Tree (Nguyen et al. 2015) using the online interface (Trifinopoulos et al. 2016). The dataset was partitioned by gene and the best-fit model of nucleo- tide substitution was selected automatically for each par- tition during the analysis. Branch support was assessed with 1,000 replicates of the Shimodaira-Hasegawa-like approximate likelihood ratio test (SH-aLRT; Guindon et al. 2010), 1,000 replicates of ultrafast bootstrap (UFBoot; Hoang et al. 2018), and 100 standard bootstrap replicates (Felsenstein 1985). The BI analysis was conducted in MrBayes v. 3.2 (Ronquist et al. 2012). The dataset was partitioned by gene and the best models were estimated by Partition- Finder v. 2.1 (Lanfear et al. 2017). The best models were as follows: GTR+I+G for the cyt b and ND4, HK Y+G for the c-mos. The proportion of invariable sites parameter (+I) was not included as it is accounted for by the +G zse.pensoft.net 350 parameter. We ran three independent runs for 10 million generations, sampling every 10,000 generations. Sta- tionarity was determined by the sequentially calculated standard deviations of the split frequencies being lower than 0.01. We discarded the initial 10% of trees as burnin and generated a 50 percent majority rule consensus tree. Branches with SH-aLRT > 80, UFBoot = 95, standard bootstrap > 70, and a Bayesian posterior probability (pp) > 0.95 were considered strongly supported. Comparisons of genetic distances We assessed whether the genetic distances that separate the samples of the unidentified species from eastern So- malia from its sister species P. tanganicus are similar to or deviate from distances between other sister species across the Psammophis tree following the method of Smid et al. (2018). Pairwise genetic distances between all samples were obtained from the ML tree and were categorized in the following groups: within species, between sister spe- cies, between non-sister species, between tanganicus and the new species. Zero genetic distances between identical haplotypes were removed. Species were considered sister only when the sister relationship was strongly supported in all the phylogenetic analyses. The four outgroup spe- cies were pruned from the tree. We tested for significance between the four groups by means of one-way ANOVA with the Tukey HSD post hoc test for pairwise signifi- cance using R (R Core Team 2013). We also calculated uncorrected patristic distances (p-distances) for the two mitochondrial markers in MEGA X v. 10.2 (Kumar et al. 2018) with the pairwise deletion option. Sex identification We used the method of Laopichienpong et al. (2017) that allows the identification of sex by using molecular markers located on the gametologous genes. The psam- mophiids possess the ZZ/ZW sex determination system where males are the homogametic sex (ZZ), and females are heterogametic (ZW) (e.g., Augstenova et al. 2018). We amplified the CTNNBI1 gametologous gene using the primers Eq-CTNNBI1-11-F1l and Eq-CTNNB1-13-R (Matsubara et al. 2016; Laopichienpong et al. 2017) and PCR conditions detailed in the latter paper. Males were identified based on the presence of a single band on gel electrophoresis (both alleles on the Z sex chromosomes are of the same length), females had two bands present (the allele on the W sex chromosome is shorter than that on the Z chromosome; Laopichienpong et al. 2017). Morphological analyses For morphological comparisons, we examined the two specimens of P. tanganicus that were used in the genet- ic analysis (NMP-P6V 76371-—2) and three specimens zse.pensoft.net Smid, J. et al.: Sand Snakes of the Horn of Africa of the undescribed species from Somalia (NMP-P6V 76373, MVZ:Herp:242772, MVZ:Herp:242773), the former two of which were also included in the genetic analysis. Scale counts and color pattern of the P. tan- ganicus holotype (MCZ R-30380) were obtained from the photographs available at the MCZ collection data- base (https://mczbase.mcz.harvard.edu/guid/MCZ: Her- p:R-30380). General morphological species characteris- tics of P. biseriatus and P. tanganicus were assembled from the literature (Largen and Spawls 2010; Spawls et al. 2018; Spawls et al. 2023). We measured snout-vent length (SVL), measured from the tip of the snout to the anterior margin of the cloaca, and tail length (TL), mea- sured from the posterior margin of the cloaca to the tip of the tail. The following scale counts were recorded: su- pralabials; infralabials; dorsal scales at one head-length behind the head (termed anterior dorsals), at midbody (termed midbody dorsals), and one head-length anteri- or to the vent (termed posterior dorsals); ventrals; sub- caudals. In addition, we collected data on the color and pattern of different body parts of all examined speci- mens. This dataset of color patterns was supplemented by photographs from throughout the distribution of P. tanganicus that were obtained from the literature (Bezy and Drewes 1985), provided by our colleagues (M. Ber- roneau, A. Childs, S. Kirchhoff, M. Menegon, D. Modry, S. Spawls, A. Stein, E. Van Der Westhuizen) or down- loaded from the iNaturalist online database (https:// www. inaturalist.org/taxa/28961-Psammophis-tangan- icus/browse_ photos). Original high-resolution photo- graphs of the examined specimens have been deposited in the MorphoBank (https://morphobank.org/) database where they are available for free download (346 photo- graphs in total; Project No. 4527). MorphoBank acces- sions are provided in Suppl. material 1. Due to a rather small number of specimens available for morphological examinations we did not carry out for- mal statistical analyses to compare the studied taxa. We nonetheless carried out informal comparisons between them to verify whether there are morphological differ- ences consistent with the genetic results. Museum acronyms CAS — California Academy of Sciences, San Francisco, USA; MCZ — Museum of Comparative Zoology, Cambridge, USA; MVZ — Museum of Vertebrate Zoology, Berkeley, USA; MSNG — Museo Civico di Storia Naturale ‘Giacomo Doria’, Genova, Italy; MZUF — University di Firenze, Museo Zoologico ‘La Specola’, Firenze (Florence), Italy; NHMUK — Natural History Museum, London, UK; NMP — National Museum in Prague, Czech Republic; TMHC — Tomas Mazuch herpetological collection, Dfite¢, Czech Republic; UniMoRe — Collezione Franchini nella Sezione Musei Anatomici del Dipartimento del Museo di Paleobiologia e dell’ Orto Botanico dell’ Universita di Modena e Reggio Emilia, Modena, Italy. Zoosyst. Evol. 99 (2) 2023, 345-361 Results Phylogenetic analyses The ML and BI analyses resulted in identical topologies, although some branches were not statistically supported in either of the analysis (Fig. 4). Given that the dataset for the genetic analysis relied mostly on previously published se- quences, the relationships between the Psammophis species remained largely congruent with other recent studies on the genus’ phylogeny (Branch et al. 2019; Chen et al. 2021; Kurniawan et al. 2021; Taft et al. 2022). The subtle topolog- ical differences resulted either from different composition of taxa across different studies or from low branch support. As for the newly analyzed samples, the sample of P. punctulatus punctulatus (JIR513) was recovered as sister to P. punctulatus trivirgatus with a robust branch support (SH-aLRT: 100/UFBoot: 100/Standard boot- strap: 100/MrBayes pp: 1.0; support values are shown in the same order hereafter). The genetic distances between the two taxa are 8.0% in the cytb b and 9.06% in the ND4. The newly analyzed samples of P. rukwae (JIR511, TMHC2012.12.148) clustered within other samples of the species from Chad and Kenya (support 97.4/100/96/1.0), with p-distances within the species ranging between 1.48—5.17% in the cyt 6 and 0.76—5.74% in the ND4. The new sample of P. sudanensis from Ethiopia (TMHC2013.07.250) did not cluster with the other sam- ples of the species included in the analyses, but actually with a lineage termed P. cf. sudanensis from Kenya and Tanzania that was first discovered by Trape et al. (2019) (support 96.4/98/91/1.0). Our new sample was sister to all the other samples of this undescribed species (support 92.3/100/92/1.0), with p-distances between our and the other samples ranging between 5.0—5.49% in the ND4 (cyt b sequence is not available for TMHC2013.07.250). The samples of P. aegyptius from Egypt and Sudan clus- tered with other samples of the species available on Gen- Bank from Egypt and Niger (support 100/100/100/1.0), and the four samples from Somaliland were sister to this clade (support 97.5/99/97/1 .0). Mean p-distances between the Somali samples and those from the north (Egypt, Su- dan, Niger) were 8.03% in the cyt b (range 4.92-10.42%) and 9.24% in the ND4 (range 8.61—10.27%). The two samples of P. tanganicus and that from central Somaliland that we tentatively identified as P. cf. tangani- cus all clustered in a clade with P. biseriatus that was well supported in all analyses (support 100/100/100/1.0). In this clade, the two samples of P. tanganicus from Ethiopia and Somalia JIR508—509 (vouchers NMP-P6V 76371-—2) were recovered to form aclade (support 100/100/100/1.0) that was sister to the remaining samples of P. tanganicus from Tan- zania (support 95.9/99/95/1.0). The sample JIR510 (vouch- er NMP-P6V 76373) from central Somaliland was closely related to a sample of an undescribed Psammophis species from eastern Somalia sequenced by Vidal et al. (2008; sam- ple and voucher codes TP28431 and MVZ:Herp:242772, respectively; support 100/100/100/1.0). Psammophis bise- riatus was then sister to the group of P. tanganicus and the 351 undescribed species (support 85.1/94/74/0.99). Uncorrect- ed p-distances between P. biseriatus, P. tanganicus and the undescribed species are shown in Table 1. Table 1. Mitochondrial genetic distances (uncorrected p-dis- tances in %) between P. biseriatus, P. tanganicus, and P. cor- nusafricae sp. nov. Below the diagonal are values for the cyt b gene, above the diagonal for the ND4. For each comparison, the mean is shown with the min-max range in brackets. P. biseriatus PP. tanganicus P. cornusafricae Sp. nov. P. biseriatus 10.18 12.04 (9.52-10.57) = (11.99-12.08) P. tanganicus 10.78 9.74 (10.33-11.04) (9.26-10.77) P. cornusafricae sp. 9.90 7.79 (7.2-8.58) nov. (9.75-10.06) Comparisons of genetic distances The genus showed clear genetic structuring with significant differences in the genetic distances found between the ex- amined categories: within species, between sister species, between non-sister species, and the distance between the P. tanganicus and the undescribed species from Somalia (ANOVA EF ia PS 0.001). The post hoc tests con- firmed pairwise differences between all categories (p< 0.01), except for the comparison between the categories ‘between sister species’ and ‘tanganicus-Somali species’ (p = 0.999; Fig. 5), indicating that the genetic distance between P. tan- ganicus and this species is statistically comparable to the distance between other sister species pairs in the genus. Taxonomic implications Based on the combined evidence of the genetic and morphological differentiation we recognize the undescribed species from the northeastern Somali regions of Somaliland and Puntland as new and provide its formal description below. Systematics Psammophiidae Bourgeois, 1968 Psammophis Fitzinger, 1826 Psammophis cornusafricae sp. nov. https://zoobank. org/F 1E4DC4C-D816-47A9-B9E 1 -1DAB56323E56 English name: African Horn Sand Snake Somali name: Subxaanyo [pronounced Subhanyo]; a term in the Somali language that refers to all Psammophis species that occur in the region. All Subxaanyo are believed to be harmless and friendly by the locals and are an important part of their folklore Chresonymy. Psammophis biseriatus in Calabresi (1927; in part), Scortecci (1939a), Scortecci (1939b), Lanza (1983; in part), Regnoli et al. (2003; in part); zse.pensoft.net 302 Smid, J. et al.: Sand Snakes of the Horn of Africa 93.1/95/78/1 Malpolon monspessulanus 99.9/100/99/1 Rhamphiophis rostratus 100/100/100/1 Psammophylax rhombeatus Psammophylax variabilis 100/100/100/1 CMRK203 - Zimbabwe r CMRK‘19 - South Africa P. crucifer CMRK70 - South Africa SE SPGISHOe 100/100/100/1 CAS205003 - Myanmar P. condanarus NK2007 - Indonesia P. indochinensis 99.4/100/100/1 100/100/100/1 CAS179682 - Turkmenistan HLMDJ76 - China P. lineolatus 100/100/100/1 IPMB28601 - Kazakhstan 100/100/100/1 ileteanaeltO 2a : GXG1504 - China P. turpanensis ST34 - China 100/100/100/1 CAS214751 - Namibia P 1M83873 - Namibia P. trigrammus 99.8/100/100/1 | 89.9/73/ CMRK256 - Botswana P. jallae 99.9/100/99/1 100/100/100/1 NB560 - Angola a pone NB600 - Angola P. ansorgii 100/100/100/1 PEMR15811 - South Africa 3 F Pe SIGSISOLEGS TM83620 - South Africa P. leightoni 93.8/91/75/0.99 100/100/4.00/1 BEMES G7 a sauih Aitica ; PEMR5669 - South Africa P notostictus PEMR85670 - South Africa 100/100/100/1 CMRK263 - Botswana ’ CMRK283 - Zambia P angolensis 98.9/100/97/1 CMRK392 - Tanzania 100/100/100/1 Vidal aGhana IRD9193S - Senegal P. praeornatus Chir01 - Benin 27.2/65/46/0.8 Fg 100/1 00/1 00/1 TR4504 - Mauritania 222Cl - Ivory Coast P. elegans IRD2761N - Chad 48.6/83/ JIR513.--Ethiopia Sons 100/100/100/1 Baoan ee eee P. punctulatus CMRK391 - Tanzania 93.6/97/87/1 99.9/100/100/1 pbsaeaQman ; JCBO00799 - Morocco P. schokari 63.5/91/ IPMB28602 - Tunisia 99.9/100/100/1 r JIR469.- Somalia 6B/0-72 JIR468 - Somalia JIR497 - Somalia eer JIR498 - Somalia 97.5/99/97/1 BEV11350 - Egypt P A . ae tius 400/100/100/1 erect gyP JCB00340 - Niger EGY07 - Egypt BEV10372 - Egypt 77/85/48/0.99 EGY08 - Egypt 400/100/100/1 CMRK169 - Tanzania See aor P. biseriatus CMRK397 - Tanzania Alea SA 100/100/100/1 JIR510 - Somalia 5 TP28431 - Somalia P. cornusafricae sp. n. 100/100/100/1 JIR508 - Somalia 85.1/94/74/0.99 JIR509 - Ethiopia 95.9/99/95/1 CMRK$90 - Tanzania P tanganicus : CMRK$87 - Tanzania 100/100/100/1 CMRK88 - Tanzania CMRK383 - Burundi 99.5/100/100/1 IRD2120N - Chad | 80.5/83/43/0.93 eae ‘ P. lineatus 400/100/100/1 CAS214727 - Namibia : CAS214763 - Namibia P. leopardinus CMRK‘186. - Zimbabwe 100/100/100/1 , CMRK237 - South Africa P breviro stris 100s O0/100/1 99.9/100/ CMRK238 - South Africa 100/1 CMRK711 - Zambia k . 400/100/100/1 CMRK377 - Burundi P. occidentalis 94.3/89/80/1 0.05 substitutions per site 25.1/77/61/0.89 100/100/ CMRK381 - Burundi MBUR8346 - Ethiopia PEMR15488 - Mozambique CMRK81 - Tanzania P. sp. Ethiopia 2 P. mossambicus 100/100/100/1} 100/1 Met eemcsce 72.7/94/67/0.6 IRD367T. - Togo IRD5002 - Guinea ay, iT 100/100/100/1 7 ag a P. phillipsii 94/99/98/1 CMRK3858 - Ethiopia ——E CMRK382 - Ethiopia P. sp. Ethiopia CMRK364 - Ethiopia 99/100/98/4 ODMGOMOGL MBUR8588 - Ethiopia AUZCR143972 - Egypt P sibilans AUZCR143973 - Egypt 37.4/91/ 100/1 00/1 00/1 IRD2289N - Chad 68/0.51 IRD2231N - Chad P. sudanensis IRD2283N - Chad ZFMK77012 - Benin 98.2/99/95/1 99-3/100t400/4 PEM1 - Senegal P. afroccidentalis 99.9/100/ Vidal - Niger 98/1 BK10620 - Kenya 2121N - Chad 97.4/100/96/1 IRD1888N - Chad P. rukwae 99.9/100/98/1 60.4/85/57/0.78 100/100/100/1 95.3/99/85/1 100/100/100/1 78.2/94/63/0.54 96.4/98/91/1 92.3/100/92/1 JIR511 - Eritrea TMHC2012.12.148 - Ethiopia NMZB4 - Zimbabwe CMRK249 - Botswana CMRK282 - Zimbabwe NMKO3597 - Kenya IPMBJ379 - Tanzania PEMR16132 - Mozambique TMHC2013.07.250.- Ethiopia CMRK91 - Tanzania CMRK386 - Kenya CMRK390 - Tanzania P. subtaeniatus P. orientalis P. cf. sudanensis Figure 4. Maximum Likelihood phylogeny of the genus Psammophis based on a concatenated alignment of the cyt b, ND4 and c-mos genes (2,149 bp in total). The clade containing P. biseriatus, P. tanganicus and P. cornusafricae sp. nov. is highlighted in pink. Codes of samples newly sequenced for this study are in red. Locality details and original references of all samples are given in Suppl. material 1. Branch support is given in the order SH-aLRT/UFBoot/Standard bootstrap/MrBayes posterior probability. zse.pensoft.net Zoosyst. Evol. 99 (2) 2023, 345-361 0.4 = wo = ~) P. tanganicus to P. cornusafricae sp. n. distance Genetic distance (substitutions per site) —) Within species 393 Between non-sister species Between sister species Figure 5. Comparison of genetic distances within the Psammophis species, between sister species pairs, between non-sister species. The distances (y axis) are based on the ML tree. Horizontal bars with asterisks on top of the graph indicate significant differences in the observed genetic distances (p < 0.001; tested by the Tukey HSD post hoc test of ANOVA). The average distance between P. tanganicus and P. cornusafricae sp. nov. (0.127 subst. per site) is shown by the dashed line. The animal depicted is specimen NMP-P6V 76373 (sample JIR510) of P. cornusafricae sp. nov. Psammophis_ biseriatus tanganicus in Loveridge (1940; in part), Lanza (1990; in part); Psammophis tanganicus Largen and Ramussen (1993; in part), Brandstatter (1996; in part), Largen and Spawls (2010; in part), Wallach et al. (2014; in part), Boundy (2020; in part); Psammophis sp. in Vidal et al. (2008), Trape et al. (2019), Keates (2021). Type material. Holotype: MVZ:Herp:242772 (sam- ple code TP28431; Fig. 6), subadult, sex unknown; So- malia, Bari Region, Heela Spring, 11 km NW (by road) of Baargaal (11.371°N, 51.0412°E), elevation 77 m above sea level; collected by T. J. Papenfuss on June 29, 2003. Paratype: MVZ:Herp:242773, adult, sex unknown; same collection data as holotype. Other material. NMP-P6V 76373 (sample JIR510; Fig. 7; MorphoBank accessions: M851260—M851339), subadult female (sex confirmed by genetics); Somaliland, 8.5 km S of Yuffleh (10.3°N, 47.182778°E), elevation 1758 m above sea level; collected by T. Mazuch on Sep- tember 5, 2017. Diagnosis. A Psammophis species sister to P. tangan- icus, with the following combination of morphological characters: 11 infralabials; 9 supralabials, three of which are in contact with the eye; nostril pierced between two equal sized nasals; 15 anterior dorsals; 14-15 midbody dorsals; 11 posterior dorsals; 146-158 ventrals; 95—100 subcaudals. Body brown, grey or beige with only a faint- ly visible pale vertebral stripe but with well pronounced dorsolateral orange stripes on each side of the body. Tail with only a slight trace of or completely lacking any dark vertebral stripe; the posterior half of the tail uniformly pale yellow. Venter white to yellowish with a conspic- uous stipple of irregular dark smears. Head dorsally with a symmetrical pattern of brown markings outlined by black margins which contrast with the grey or beige background coloration. Labials white with dark margins forming a black lip. Dark brown temporal stripe outlined by black blotches and stretching from the eye to the pos- terior end of the head. Description of the holotype (Fig. 6). SVL 42.6 mm, TL 20.3 mm (the very tip of the tail is missing; the orig- inal tail was hence probably a few millimeters longer), total length 62.9 mm. Head flattened dorsoventrally, separated from the body by a distinct neck; nostril pierced between two nasals, major sector-shaped (‘pac-man’-like shape); eyes large, with a round pupil; rostral roughly heptagonal, with a distinct groove at the base for flicking the tongue out; rostral visible from above and followed dorsally by one zse.pensoft.net 354 Smid, J. et al.: Sand Snakes of the Horn of Africa Figure 6. Holotype of Psammophis cornusafricae sp. nov. (MVZ:Herp:242772). pair of internasals, one pair of prefrontals, a long and nar- row frontal with nearly equally as long supraoculars to its side; preoculars in narrow contact with the frontal and wedged between the prefrontals and supraoculars; large pair of parietals contacted laterally by large anterior tem- porals, posterior temporals and enlarged postparietals. The temporal formula of the anterior temporal + posterior temporal scales 1s 1+3. Laterally, rostral followed by two consecutive nasals, an elongated loreal and a large preoc- ular that forms the anterior margin of the eye. Two pos- toculars at the posterior eye margin, the upper of which being contacted by the parietal and the lower by the an- terior temporal. Nine supralabials, of which the 4", 5" and 6" are in contact with the eye; 11 infralabials, five of which are in contact with the anterior pair of chin shields. Two pairs of chin shields. Four (right side) and five (left side) gulars in a longitudinal row between chin shields and preventrals. Anterior dorsals 15, midbody dorsals 14, posterior dorsals 11; 158 ventrals; 85 subcaudals. Coloration in preservation: body brownish beige, dor- sum darker than the flanks, with dark sets of three to five dorsolateral scales separated by two to three paler scale rows. The frontal and supraoculars with dark pigment patches at their anterior sides, posterior margins pale. Pa- rietals with dark oblong blotches by the medial line and at the anterior margin of the scales. Internasals and pre- frontals dark to the sides and paler medially, which forms a narrow white-yellowish stripe at the nose tip. Supral- zse.pensoft.net abials white with contrasting dark brown upper and lower margins. The white of the supralabials continues onto the sides of the neck. Throat with extensive dark spots ar- ranged in longitudinal stripes — one on each side of the mouth on the lower side of the infralabials up to the 6" infralabial, one at the mid-line running through the chin shields and gulars further on the belly. Comparisons. Psammophis cornusafricae sp. nov. can be differentiated from other congeners that occur in the Horn of Africa as follows: from its sister species P. tanganicus (character in brackets) by having 3—5 gulars in a longitudinal row between chin shields and preven- trals (versus 2—3), posterior tip of the frontal moderately pointed (versus rounded), by lacking a clear orange or brownish dorsal stripe along the body and tail (versus or- ange vertebral stripe present), tail lacking a stripe, or hav- ing only a weakly developed dark vertebral stripe (versus tail with a clear dark vertebral stripe), having a clearly demarcated narrow temporal stripe on the head (versus broad stripe without clear demarcation), dark markings on the head dorsum paired and not fused (versus dark marking on the back of the head fused into a ‘W’ shape); from P. biseriatus (character in brackets) by having three supralabials in contact with the eye (versus two), usual- ly one large anterior temporal (versus two), gular region strongly pigmented with dark longitudinal stripes (versus weakly pigmented with small dark spots), head ornate- ly patterned above (versus uniformly colored), tail with Zoosyst. Evol. 99 (2) 2023, 345-361 ee at ie rs site Figure 7. Psammophis cornusafricae sp. nov. in life (specimen NMP-P6V 76373, sample JIR510), with a close-up of the head in Siss) one Seer hy ae the upper left corner, and the habitat at the locality where the specimen was found (8.5 km S of Yuffleh, Somaliland) at the bottom. absent or only a weakly developed dark vertebral stripe (versus tail with a clear dark vertebral stripe). Further morphological comparisons between P. cornusafricae sp. nov., P. tanganicus, and P. biseriatus are summarized in Table 2. From the other Horn species P. cornusafricae sp. nov. differs by having a lower number of midbody dor- sals (14-15 versus 17 or 19 in P. aegyptius, P. lineatus, P. mossambicus, P. orientalis, P. punctulatus, P. rukwae, P. schokari, P. sudanensis a P. cf. sudanensis, P. trivirga- tus), or a higher number of midbody dorsals (11 in P. an- golensis, 13 in P. pulcher). Variation. The studied specimens of P. cornusafricae sp. nov. show some degree of morphological variation (Table 2). Namely, the number of infralabials varies be- tween 10—11 and the number of supralabials between 8-9. The 3-5" supralabials are in contact with the eye in spec- imen NMP-P6V 76373 (the one with 8 supralabials); in the other two specimens it is the 4-6" supralabials that touch the eye. Specimen NMP-P6V 76373 has three postoculars at the posterior eye margin and two anterior temporals (versus two postoculars and one anterior temporal in the other two specimens). Specimens MVZ:Herp:242773 and NMP-P6V 76373 (unilaterally) have two posterior tem- zse.pensoft.net 356 Smid, J. et al.: Sand Snakes of the Horn of Africa Table 2. Morphological comparisons of P. biseriatus, P. tanganicus and P. cornusafricae sp. nov. Superscript letters refer to the original references from which the data was obtained as follows: * - Spawls et al. (2023); > - Largen and Spawls (2010); ° - Bezy and Drewes (1985). NA* indicates characters unavailable due to incomplete tail. P. tanganicus P. cornusafricae sp. nov. biseriatus | General MCZ NMP-P6V NMP-P6V | NMP-P6V MVZ:Herp:242772 MVZ:Herp:242773 R-30380 76371 76372 76373 Holotype Paratype Holotype Total length ~ 50-80 | ~ 50-80 63 90.8 NA * 58.8 62.9 73.8 (cm) (max 100)? | (max 100) SVL (cm) 40.6 61.5 47.4 39,7 42.6 49.7 TL (cm) 22.4 29.3 NA * 19.1 20.3 24.1 SVL/Total 61.7 (56- 66.7 64.4 67.7 6715 67.7 67.3 length (x100) 66)° (64-81)°¢ Infralabials 10-11° 9-11 11 11 10 10/11 11 10 (right/left) Supralabials Ue 9° 9 9 9 8 9 9 Supralabial in Three —‘ Three (4*-6") Three (4-6) ~— Three (4-6") | Three (3'-5") Three (4°-6") Three (4-6") contact with eye (4tr—6th) Midbody dorsals] 15° =| _—_15? 15 15 15 15 14 15 Ventrals 138-168? | 143-169? 151 159 150 tol 158 146 (+1 preventral) (+2 preventrals) (+1 preventral) | (+1 preventral) Subcaudals 81-123? 114 110 NA * 95 85 100 porals (versus three in the holotype). There are 14 mid- body dorsals in the holotype, while the other two spec- imens have 15. The number of ventrals and subcaudals varies between 146-158 and 85-100, respectively. The two specimens examined genetically are closely related (Fig. 4). They are 99.35% identical in the cyt b sequence (different in 7 out of 1087 bp), 98.63% identical in the ND4 (different in 9 out of 662 bp), and 99.75% identical in the c-mos (different in 1 out of 400 bp). Etymology. The species epithet indicates the geo- graphic origin of the species and translates as “The Sand Snake of the Africa’s Horn’. It is a noun in the genitive case that is derived from the Latin words for horn (cor- nu-us) and Africa (Africa-ae). Distribution. Our knowledge on the distribution of P. cornusafricae sp. nov. is very limited because it has so far been confirmed only from two localities. They are both located in Somalia; the type locality is at the very tip of the Horn of Africa (Puntland), the locality of spec- imen NMP-P6V 76373 lies in central Somaliland, some 440 km west-southwest from the type locality (Fig. 8). Puntland and the eastern part of Somaliland are the driest parts of the Horn of Africa. They receive less than 200 mm of precipitation annually (Muchiri 2007), and are covered by sparse vegetation (ESA CCI Team 2017) and geologically formed by the so-called Karkar and Taleh formations that include limestones, dolomites, and gyp- sum (Schreiber 1993; Quiroga et al. 2022). We assume P. cornusafricae sp. nov. to roughly match the extent of the Karkar and Taleh formations and adjoin the range of P. tanganicus from the east. This distribution pattern with central Somaliland being a contact zone of closely relat- ed taxa has already been confirmed for other taxa, e.g. Tomopterna (Pyxicephalidae), Hemidactylus (Gekkoni- dae), Heliobolus, Latastia, Pseuderemias (Lacertidae), Uromastyx, Xenagama (Agamidae), Echis (Viperidae) zse.pensoft.net (Arillo et al. 1965; Lanza 1990; Zimkus and Larson 2011; Wagner et al. 2013; Tamar et al. 2018; Spawls and Branch 2020; Smid et al. 2020: Spawls et al. 2023). Discussion Sand Snakes represent an important element of the snake fauna of Africa and Asia. Their diurnal habits, active forag- ing for prey and a relatively large size make them conspic- uous and frequently encountered by naturalists (e.g., 2008 observations on iNaturalist as of April 27, 2023). This may be the reason why tissue samples are not that difficult to ob- tain for genetic analyses and why our understanding of their phylogenetic relationships is fairly complete, at least from the taxon sampling perspective. Of the 33 currently rec- ognized species (Wallach et al. 2014; Uetz et al. 2023), 30 have been placed in the phylogenetic context. The phyloge- netic studies have not only untangled the evolutionary his- tory of the genus, but also identified several distinct genetic lineages that may represent yet undescribed species (Kelly et al. 2008; Branch et al. 2019; Chen et al. 2021; Keates 2021; Kurniawan et al. 2021; Taft et al. 2022). Coincident- ly, samples of all these undescribed species came from the Horn of Africa and the nearest adjacent regions, suggesting that the fauna of this forsaken region is still overlooked and that it needs further attention of researchers. This study is the first to target the Sand Snakes of the Horn of Africa specifically. We analyze 14 newly ob- tained samples of six species and provide new insights into their distribution, genetic variability, and taxonomy. The phylogenetic reconstruction of the broader relation- ships between the Psammophis species confirmed previ- ous findings (Kelly et al. 2008; Branch et al. 2019; Chen et al. 2021; Keates 2021; Taft et al. 2022), which is not surprising given their largely overlapping genetic data Zoosyst. Evol. 99 (2) 2023, 345-361 el a “=O |CMRK87-88| Lt at, ZUG 357 Nz cornusafricae sp. n. NS ae. \ P.tanganicus — : qi o» P. biseriatus 500km Figure 8. Distribution of P. biseriatus, P. tanganicus and P. cornusafricae sp. nov. in the Horn of Africa. Localities sampled for the genetic analysis are shown along with sample codes. The background layer shows shaded relief and water bodies and drainages and was made with Natural Earth (www.naturalearthdata.com). The species range maps were derived from Roll et al. (2017, 2021), Spawls et al. (2023), and the findings presented in this study. matrices. The general structure of the tree and its poten- tial implications for evolutionary origins of the genus are not new findings and we will not comment on them here. We will rather focus on the individual species analyzed by us, their intraspecific variation, closest phylogenetic relationships, and novel findings on their distribution. The description of P. cornusafricae sp. nov. adds one more species to the list of snakes endemic to Puntland and eastern Somaliland (e.g., Platyceps brevis and P. messa- nai (Schatti and Lanza 1989; Schatti and Charvet 2003), Spalerosophis josephscorteccii (Lanza 1964), Telescopus pulcher (Mazuch et al. 2018), Echis hughesi (Cherlin 1990; Spawls et al. 2023), Rhinotyphlops leucocephalus (Roux-Esteve 1974)). When first analyzed by Vidal et al. (2008), the phylogenetic relationships of this, at that time undescribed, species could not be inferred with cer- tainty because there were only nine Psammophis species included in their study. Only more recent studies with a broader sampling of taxa enabled the identification of its closest relatives — P. tanganicus and P. biseriatus. Psam- mophis tanganicus has consistently been recovered as a sister species to P. cornusafricae sp. nov. (Trape et al. 2019; Keates 2021; this study). Precise delineation of the distribution ranges of P. tanganicus and P. cornusafricae sp. nov. and their potential contact zone is difficult to make at this point. The two closest confirmed localities of both species are in Somaliland and about 240 km apart. We assume the ranges of the two species meet some- zse.pensoft.net 358 where in central Somaliland as observed in other taxa (see the Distribution section above). Ideally, other spec- imens from across Somaliland and Puntland that are de- posited in herpetology collections worldwide and either catalogued as P. tanganicus or P. biseriatus (e.g., CAS, NHMUK, presumably MSNG, MZUF, UniMoRe; Cal- abresi 1927; Scortecci 1939b; Regnoli et al. 2003) and which were unfortunately not available to us should be examined to assess the distribution limits of both species. The new samples of P. tanganicus from Ethiopia and So- malia are slightly genetically different from those from Tanzania, which indicates that there is some level of in- traspecific variation across its range. This genetic differ- entiation is, however, not mirrored by the morphological variation as the specimens from the northern part of the distribution are morphologically identical to other P. tan- ganicus populations from other parts of its range, includ- ing the holotype (MCZ R-30380) from central Tanzania. Besides the description of the new species we also an- alyzed new specimens of other Horn taxa to shed light on their diversity and distributions. The two newly analyzed samples of P rukwae from Ethiopia and Eritrea cluster with other samples of the species from Chad and Kenya. This broadly distributed species ranges from Senegal in the West through the Sahel to East Africa (Trape and Mané 2006; Spawls et al. 2018). Psammophis rukwae has only been confirmed recently from Ethiopia (Trape et al. 2019), and there are no records yet from Eritrea (Largen and Spawls 2010). The genotyped sample from Eritrea present- ed here is thus the first record of P. rukwae for the country. The existence of the cryptic species referred to as P. cf. sudanensis that morphologically resembles P. su- danensis is well established (Trape et al. 2019; Keates 2021; Taft et al. 2022). Quite surprisingly, the two species are not closely related. While the real P. sudanensis forms a clade with P. sibilans and an undescribed species from Ethiopia (termed ‘P. sp. Ethiopia’), P. cf. sudanensis clus- ters with the East African P. orientalis and P. subtaeniatus from southern Africa. The new sample from Ethiopia in- cluded here that we originally identified as P. sudanensis, and which in fact turned out to belong to P. cf. sudanen- sis, proves that the two species are indeed phenotypically similar. Until detailed examinations of specimens of both species are carried out, and a formal description of the cryptic species is presented together with a differential diagnosis, it seems impossible to distinguish them with- out having to use DNA genotyping. We refrained from attempting to sort out the taxonomy of P. cf. sudanensis here due to the lack of comparative material. The species with the most newly analyzed samples was P. aegyptius. The new samples from Egypt and Sudan are genetically very similar to other conspecific samples from the same region. The four samples from Somaliland, on the other hand, show marked genetic differences from the North African samples, as well as within themselves. The range of P. aegyptius covers the eastern part of the Sahara and has until now not been known to occur this far south (Largen and Spawls 2010; Geniez 2018). The new zse.pensoft.net Smid, J. et al.: Sand Snakes of the Horn of Africa localities from Somaliland thus present a considerable range extension for the species. Lastly, the sample of P. punctulatus punctulatus in- cluded in our analysis 1s the first of the taxon to have been analyzed. Previous material of the species belonged to the subspecies P. p. trivirgatus (Kelly et al. 2008) and the two taxa cluster together in the phylogeny. Psammophis p. punctulatus extends from southeastern Egypt to the ex- treme northwest of Somaliland; P. p. trivirgatus is distrib- uted from northern Somaliland to Uganda, South Sudan (Ilemi Triangle) and Tanzania (Largen and Rasmussen 1993; Spawls et al. 2023). The two taxa show consider- able and stable morphological differences, with P. p. tri- virgatus having a lower number of subcaudals than P. p. punctulatus (143-163 in the former and 158-178 in the latter) and a bright orange head and three black dorsal stripes, compared to a grey head and one dorsal stripe in the latter (Fig. 2a, b). On the basis of their apparently partially overlapping ranges Lanza (1990) concluded that they represent different species. Based on the evidence of profound genetic diversification between the two taxa laid out here, in combination with the above mentioned morphological and geographical differences between them we follow Lanza’s (1990) conclusion and recognize them as two distinct species, Psammophis punctulatus Dumeéril, Bibron & Dumeéril, 1854 and Psammophis tri- virgatus Peters, 1878. Sand Snakes rank among the most studied African snakes from the phylogenetic perspective. Yet, as evi- denced by the number of cryptic or undescribed species, we are still far from fully comprehending their diversity. With the currently available sampling it 1s obvious that East Africa, and the Horn of Africa in particular, supports the highest diversity of undescribed species of Psam- mophis. Once their taxonomy is resolved, the diversity in the Horn may surpass that of the currently richest hotspot of the genus in southern Africa. Acknowledgements We are indebted to C. Spencer, J. Ho, and Natalie Ng (MVZ) for morphological examination and photographs of the MVZ specimens. Doubravka Velenska kindly helped with some of the lab work. We thank J. Lokvenc, F. Kovarik, P. Novak, V. Trailin, R. Starha, A. I. Awale, M. Hiis, S. Musse, A. Shabele, and L. Wilson for field work assistance. We are indebted to A. A. Bogore (Vice President, Administration and Finance of Amoud Univer- sity), M. M. Jibril (Vice President, Academic Affairs), S. A. Gulaid (President of Amoud University), M. Y. Muse (President of University of Hargeisa), and Shukri H. Is- mail Bandare (Ministry of Environment and Climate Change, Republic of Somaliland) for issuing the col- lection and export permit (Ref. MOERD/M/I/251/2017, MOERD/M/I/721/2019, .MOERD/M/I/230/2021). We thank W. Branch, G. Dunger and S. Spawls for provid- ing photographs of life Psammophis specimens that we Zoosyst. Evol. 99 (2) 2023, 345-361 used in Figs 1, 2. The editor J. Penner and reviewers W. Conradie, S. Spawls, J.-F. Trape, A. Tiutenko, and P. Wagener provided helpful feedback and fruitful comments on the first version of the manuscript. JS was supported by the Charles University Research Centre program No. 204069, and by the Ministry of Culture of the Czech Re- public (DKRVO 2019-2023/6. VII.e, 00023272). References Arillo A, Balletto E, Spano’ S (1965) II e III spedizione Scortecci in Migiurtinia: il genere Eremias Wiegmann (Reptilia, Lacertidae). Bollettino dei Musei di Zoologia e Anatomia Comparata della R Universita di Genova 33: 85-109. Augstenova B, Johnson Pokorna M, Altmanova M, Frynta D, Rovatsos M, Kratochvil L (2018) ZW, XY, and yet ZW: Sex chromosome evolution in snakes even more complicated. Evolution 72(8): 1701-1707. https://doi.org/10.1111/evo.13543 Bezy RL, Drewes RC (1985) Specific status of the East African colu- brid snake, Psammophis tanganicus. Journal of Herpetology 19(2): 246-253. https://doi.org/10.2307/1564178 Boundy J (2020) Snakes of the World: a supplement. CRC Press, 273 pp. https://doi.org/10.1201/978042946 1354 Branch WR, Baptista N, Keates C, Edwards S (2019) Rediscovery, taxo- nomic status, and phylogenetic relationships of two rare and endem- ic snakes (Serpentes: Psammophiinae) from the southwestern An- golan plateau. Zootaxa 4590(3): 342-366. https://doi.org/10.11646/ zootaxa.4590.3.2 Brandstatter F (1996) Sandrennattern. Westarp Wissenschaften, Mag- deburg, 142 pp. Calabresi E (1927) Anfibi e Rettili raccolti nella Somalia dai Proff. G. Stefanini e N. Puccioni (Gennaio-Luglio 1924). Atti della Societa italiana di scienze naturali e del museo civico di storia naturale di Milano 66: 14-60. Castresana J (2000) Selection of conserved blocks from multiple align- ments for their use in phylogenetic analysis. Molecular Biology and Evolution 17(4): 540-552. https://doi.org/10.1093/oxfordjournals. molbev.a026334 Chen M, Liu J, Cai B, Li J; Wu N, Guo X (2021) A new species of Psammophis (Serpentes: Psammophiidae) from the Turpan Ba- sin in northwest China. Zootaxa 4974(1): 116-134. https://doi. org/10.11646/zootaxa.4974.1.4 Cherlin V (1990) Taxonomic revision of the snake genus Echis (Vipe- ridae) Il. An analysis of taxonomy and description of new forms. Proceedings of the Zoological Institute, Leningrad 207: 193-223. de Haan CC (2003) Extrabuccal infralabial secretion outlets in Dro- mophis, Mimophis and Psammophis species (Serpentes, Colubridae, Psammophiini). A probable substitute for ‘self-rubbing’and cloacal scent gland functions, and a cue for a taxonomic account. Comptes Rendus Biologies 326(3): 275—286. https://doi.org/10.1016/S1631- 0691(03)00073-8 ESA CCI Team 2017 (2017) S2 prototype LC map at 20m of Afri- ca2016. http://2016africalandcover20m.esrin.esa.int/ [accessed Jan 7", 2023] Farr TG, Kobrick M (2000) Shuttle Radar Topography Mission pro- duces a wealth of data. Eos (Washington, D.C.) 81(48): 583-585. https://doi.org/10.1029/EO0811048p00583 359 Felsenstein J (1985) Confidence limits on phylogenies: An approach us- ing the bootstrap. Evolution; International Journal of Organic Evolu- tion 39(4): 783-791. https://doi.org/10.2307/2408678 Geniez P (2018) Snakes of Europe, North Africa and the Middle East: A Photographic Guide. Princeton University Press, Princeton and Oxford, 379 pp. Guindon S, Dufayard J-F, Lefort V, Anisimova M, Hordijk W, Gascuel O (2010) New algorithms and methods to estimate maximum-likeli- hood phylogenies: Assessing the performance of PhyML 3.0. System- atic Biology 59(3): 307-321. https://doi.org/10.1093/sysbio/syq010 Hoang DT, Chernomor O, Von Haeseler A, Minh BQ, Vinh LS (2018) UFBoot2: Improving the ultrafast bootstrap approximation. Molec- ular Biology and Evolution 35(2): 518-522. https://doi.org/10.1093/ molbev/msx281 Katoh K, Rozewicki J, Yamada KD (2019) MAFFT online service: Multiple sequence alignment, interactive sequence choice and visu- alization. Briefings in Bioinformatics 20(4): 1160-1166. https://doi. org/10.1093/bib/bbx108 Kearse M, Moir R, Wilson A, Stones-Havas S, Cheung M, Sturrock S, Buxton S, Cooper A, Markowitz S, Duran C, Thierer T, Ashton B, Meintjes P, Drummond A (2012) Geneious Basic: An integrated and extendable desktop software platform for the organization and anal- ysis of sequence data. Bioinformatics (Oxford, England) 28(12): 1647-1649. https://doi.org/10.1093/bioinformatics/bts 199 Keates C (2021) Integrative systematic structuring of the widespread psam- mophiid snakes (Psammophiidae): a multi-evidence species delineation approach. Rhodes University, Grahamstown, South Africa, 219 pp. Kelly CMR, Barker NP, Villet MH, Broadley DG, Branch WR (2008) The snake family Psammophiidae (Reptilia: Serpentes): Phylogenet- ics and species delimitation in the African sand snakes (Psammophis Boie, 1825) and allied genera. Molecular Phylogenetics and Evolu- tion 47(3): 1045-1060. https://doi.org/10.1016/j.ympev.2008.03.025 Kumar S, Stecher G, Li M, Knyaz C, Tamura K (2018) MEGA X: Mo- lecular evolutionary genetics analysis across computing platforms. Molecular Biology and Evolution 35(6): 1547-1549. https://doi. org/10.1093/molbev/msy096 Kurniawan N, Septiadi L, Fathoni M, Wibawa GS, Thammachoti P (2021) Out of Indochina: Confirmed specimen record and first mo- lecular identification of Psammophis indochinensis Smith, 1943 (Squamata, Psammophiidae) from Bali, Indonesia. Check List 17(6): 1521-1531. https://doi.org/10.15560/17.6.1521 Lanfear R, Frandsen PB, Wright AM, Senfeld T, Calcott B (2017) PartitionFinder 2: New methods for selecting partitioned models of evolution for molecular and morphological phylogenetic anal- yses. Molecular Biology and Evolution 34: 772-773. https://doi. org/10.1093/molbev/msw260 Lanza B (1964) Il genere Sphalerosophis e descrizione di una nuova specie (Reptilia, Serpentes). Monitore Zoologico Italiano 72: 47-64. Lanza B (1983) A list of the Somali amphibians and reptiles. Monitore Zoologico Italiano 8(1): 193-247. https://doi.org/10.1080/0026978 6.1983.11758573 Lanza B (1990) Amphibians and reptiles of the Somali Democratic Republic: Check list and biogeography. Biogeographia 14: 407-465. https://doi.org/10.21426/B614110318 Laopichienpong N, Tawichasri P, Chanhome L, Phatcharakullawarawat R, Singchat W, Kantachumpoo A, Muangmai N, Suntrarachun S, Matsubara K, Peyachoknagul S, Srikulnath K (2017) A novel meth- od of caenophidian snake sex identification using molecular markers zse.pensoft.net 360 based on two gametologous genes. Ecology and Evolution 7(13): 4661-4669. https://doi.org/10.1002/ece3.3057 Largen MJ, Rasmussen JB (1993) Catalogue of the snakes of Ethiopia (Reptilia Serpentes), including identification keys. Tropical Zoology 6(2): 313-434. https://doi.org/10.1080/03946975.1993.10539231 Largen MJ, Spawls S (2010) The Amphibians and Reptiles of Ethiopia and Eritrea. Edition Chimaira, Frankfurt am Main, 693 pp. Lewin A, Feldman A, Bauer AM, Belmaker J, Broadley DG, Chirio L, Itescu Y, LeBreton M, Maza E, Meirte D, Meirte D, Nagy ZT, No- vosolov M, Roll U, Tallowin O, Trape JF, Vidan E, Meiri S (2016) Patterns of species richness, endemism and environmental gradients of African reptiles. Journal of Biogeography 43(12): 2380-2390. https://doi.org/10.1111/jb1.12848 Loveridge A (1940) Revision of the African snakes of the genera Dromophis and Psammophis. Bulletin of the Museum of Compar- ative Zoology 87: 1-70. Matsubara K, Nishida C, Matsuda Y, Kumazawa Y (2016) Sex chro- mosome evolution in snakes inferred from divergence patterns of two gametologous genes and chromosome distribution of sex chro- mosome-linked repetitive sequences. Zoological Letters 2(1): 1-16. https://doi.org/10.1186/s40851-016-0056-1 Mazuch T, Smid J, Price T, Frydlova P, Awale AI, Elmi HSA, Frynta D (2018) New records of one of the least known snakes, 7e/escopus pulcher (Squamata: Colubridae) from the Horn of Africa. Zootaxa 4462(4): 483-496. https://doi.org/10.11646/zootaxa.4462.4.2 Mittermeier RA, Gil PR, Hoffmann M, Pilgrim J, Brooks T, Mittermeier CG, Lamoreux J, Fonseca GAB (2004) Hotspots revisited: Earth’s biologically richest and most endangered terrestrial ecoregions. Conservation International, Washington, 392 pp. Muchiri P (2007) Climate of Somalia. In: (SWALIM) TrW-SWaLIM (Ed), Nairobi, Kenya, 73 pp. Nguyen L-T, Schmidt HA, Von Haeseler A, Minh BQ (2015) IQ-TREE: A fast and effective stochastic algorithm for estimating maxi- mum-likelihood phylogenies. Molecular Biology and Evolution 32(1): 268-274. https://doi.org/10.1093/molbev/msu300 Quiroga E, Bertoni C, Van Goethem M, Blazevic LA, Ruden F (2022) A 3D geological model of the horn of Africa: New insights for hy- drogeological simulations of deep groundwater systems. Journal of Hydrology. Regional Studies 42: 101166. https://doi.org/10.1016/j. ejrh.2022.101166 R Core Team (2013) R: A language and environment for statistical com- puting. Regnoli L, Maramaldo R, Fratello B (2003) I “Colubridi” Opistoglifi della Collezione Franchini. Atti della Societa dei Naturalisti e Mate- matici di Modena 134: 167-215. Roll U, Feldman A, Novosolov M, Allison A, Bauer AM, Bernard R, Bohm M, Castro-Herrera F, Chirio L, Collen B, Colli GR, Dabool L, Das I, Doan TM, Grismer LL, Hoogmoed MS, Itescu Y, Kraus F, LeBreton M, Lewin A, Martins M, Maza E, Meirte D, Nagy ZT, Nogueira CdC, Pauwels OSG, Pincheira Donoso D, Powney GD, Sindaco R, Tallowin OJS, Torres-Carvajal O, Trape J-F, Vidan E, Uetz P, Wagner P, Wang Y, Orme CDL, Grenyer R, Meiri S (2017) The global distribution of tetra- pods reveals a need for targeted reptile conservation. Nature Ecology & Evolution 1: 1677-1682. https://doi.org/10.1038/s41559-017-0332-2 Roll U, Meiri S, Farrell M, Davies J, Gittleman J, Wiens J, Stephens P (2021) GARD 1.5 range shapefiles used in: Global diversity patterns are explained by diversification rates at ancient, not shallow, times- cales. V1 ed. Scholars Portal Dataverse. zse.pensoft.net Smid, J. et al.: Sand Snakes of the Horn of Africa Ronquist F, Teslenko M, Van Der Mark P, Ayres DL, Darling A, Hohna S, Larget B, Liu L, Suchard MA, Huelsenbeck JP (2012) MrBayes 3.2: Efficient bayesian phylogenetic inference and model choice across a large model space. Systematic Biology 61(3): 539-542. https://doi.org/10.1093/sysbio/sys029 Roux-Esteve R (1974) Révision systématique des Typhlopidae d’Af- rique (Reptilia Serpentes). Mémoires du Muséum national d’His- toire naturelle 87: 1-313. Schatti B, Charvet C (2003) Systematics of Platyceps brevis (Boulenger 1895) and related East African racers (Serpentes Colubrinae). Trop- ical Zoology 16(1): 93-111. https://doi.org/10.1080/03946975.200 3.10531186 Schatti B, Lanza B (1989) Coluber messanai, a new species of snake from Northern Somalia. Bollettino del Museo regionale di Scienze naturali, Torino 7: 413-421. Schreiber W (1993) Sedimentological and geochemical aspects of the Taleh Formation (Eocene) in North Somalia. In: Geoscientif- ic Research in Northeast Africa. CRC Press, 449-454. https://doi. org/10.1201/9780203753392-82 Scortecci G (1939a) Gli ofidi velenosi dell’ Africa Italiana. In. Istituto Sieroterapico Milanese, Milano, 292. Scortecci G (1939b) Spedizione zoologica del Marchese Saverio Pa- trizi nel Basso Giuba e nell’Oltre Giuba. Giugno-Agosto 1934. XII. Rettili Ofidi. Annali del Museo Civico di Storia Naturale “G Doria’, Genova 58: 263-291. Smid J (2022) Geographic and taxonomic biases in the vertebrate tree of life. Journal of Biogeography 49(12): 2120-2129. https://doi. org/10.1111/jb1.14491 Smid J, Engelbrecht H, Taft JM, Telford NS, Makhubo BG, Bauer AM, Tolley KA (2018) A contribution to the phylogeny and taxonomy of the Pachydactylus weberi group (Squamata: Gekkonidae): a case of intraspecific colour polymorphism confounding taxonomy. African Journal of Herpetology 67(2): 113-126. https://doi.org/10.1080/215 64574.2017.1398186 Smid J, Go¢men B, Crochet P-A, Trape J-F, Mazuch T, Uvizl M, Nagy ZT (2019) Ancient diversification, biogeography, and the role of climatic niche evolution in the Old World cat snakes (Colubridae, Telescopus). Molecular Phylogenetics and Evolution 134: 35-49. https://doi.org/10.1016/j.ympev.2019.01.015 Smid J, Mazuch T, Novakova L, Modry D, Malonza PK, Elmi HSA, Carranza S, Moravec J (2020) Phylogeny and systematic revision of the gecko genus Hemidactylus from the Horn of Africa (Squamata: Gekkonidae). Herpetological Monograph 33(1): 26-47. https://doi. org/10.1655/HERPMONOGRAPHS-D-19-00010.1 Spawls S, Branch B (2020) The dangerous snakes of Africa. Blooms- bury, London, 336 pp. Spawls S, Howell K, Hinkel H, Menegon M (2018) Field guide to the reptiles of East Africa. Bloomsbury, London, 624 pp. Spawls S, Mazuch T, Abubakr M (2023) Handbook of Amphibians and Reptiles of Northeast Africa. Bloomsbury, London, 640 pp. Steehouder A (1984) Repeated successful breeding of the red striped sand snake Psammophis subtaeniatus sudanensis, and some remarks on the ‘polishing behaviour’ of this species. Litteratura Serpentium 4: 90-103. Taft JM, Maritz B, Tolley KA (2022) Stable climate corridors promote gene flow in the Cape sand snake species complex (Psammophi- idae). Zoologica Scripta 51(1): 58-75. https://doi.org/10.1111/ zsc.12514 Zoosyst. Evol. 99 (2) 2023, 345-361 Tamar K, Metallinou M, Wilms T, Schmitz A, Crochet PA, Geniez P, Carranza S (2018) Evolutionary history of spiny-tailed lizards (Ag- amidae: Uromastyx) from the Saharo-Arabian region. Zoologica Scripta 47(2): 159-173. https://do1.org/10.1111/zsc.12266 Trape J-F, Mané Y (2006) Guide des serpents d’ Afrique occidentale. Sa- vane et désert. IRD Editions, Paris, 226 pp. https://doi.org/10.4000/ books. irdeditions.37382 Trape J-F, Crochet P-A, Broadley DG, Sourouille P, Mané Y, Burger M, Bohme W, Saleh M, Karan A, Lanza B, Mediannikov O (2019) On the Psammophis sibilans group (Serpentes, Lamprophiidae, Psam- mophiinae) north of 12°S, with the description of a new species from West Africa. Bonn Zoological Bulletin 68: 61-91. https://doi. org/10.20363/BZB-2019.68.1.061 Trifinopoulos J, Nguyen L-T, von Haeseler A, Minh BQ (2016) W-IQ- TREE: A fast online phylogenetic tool for maximum likelihood analysis. Nucleic Acids Research 44(W1): W232—W235. https://do1. org/10.1093/nar/gkw256 Uetz P, Freed P, HoSek J (2023) The Reptile Database. http://www.rep- tile-database.org/ [accessed April 25", 2023] Vidal N, Branch WR, Pauwels OSG, Hedges SB, Broadley DG, Wink M, Cruaud C, Joger U, Nagy ZT (2008) Dissecting the major Af- rican snake radiation: A molecular phylogeny of the Lamprophii- dae Fitzinger (Serpentes, Caenophidia). Zootaxa 1945(1): 51-66. https://doi.org/10.11646/zootaxa.1945.1.3 Wagner P, Mazuch T, Bauer AM (2013) An extraordinary tail — integra- tive review of the agamid genus Xenagama. Journal of Zoological Systematics and Evolutionary Research 51(2): 144-164. https://doi. org/10.1111/jzs.12016 Wallach V, Williams KL, Boundy J (2014) Snakes of the world: a cat- alogue of living and extinct species. CRC Press, Boca Raton, 1257 pp. https://doi.org/10.1201/b16901 361 Zimkus BM, Larson JG (2011) Examination of the molecular relation- ships of sand frogs (Anura: Pyxicephalidae: 7Zomopterna) and res- urrection of two species from the Horn of Africa. Zootaxa 2933(1): 27-45. https://doi.org/10.11646/zootaxa.2933.1.2 Supplementary material | List of Psammophis species and samples included in the phylogenetic analysis Authors: Jiti Smid, Sergio Matilla Fernandez, Hassan Sh Abdirahman Elmi, Tomas Mazuch Data type: Genetic data matrix Explanation note: List of Psammophis species and sam- ples included in the phylogenetic analysis, with their sample/voucher number, country and locality of origin (datum WGS84), GenBank accession numbers, Mor- phoBank accessions (MorphoBank Project No. 4527), and original reference. Samples newly sequenced for this study are in bold. The GPS coordinates of P. tan- ganicus from Tanzania that are marked with asterisks are only approximate. Copyright notice: This dataset is made available under the Open Database License (http://opendatacommons. org/licenses/odbl/1.0/). The Open Database License (ODbL) is a license agreement intended to allow us- ers to freely share, modify, and use this Dataset while maintaining this same freedom for others, provided that the original source and author(s) are credited. Link: https://do1.org/10.3897/zse.99.101943.suppl1 zse.pensoft.net