Walterinnesia Lataste 1887

The genus Walterinnesia comprises two species of medically-important venomous elapid snakes, which are known for their striking black coloration and neurotoxic venom (Al-Sadoon et al. 2012). Commonly referred to as black desert cobras, Walterinnesia species are primarily fossorial and nocturnal, and are rarely observed in the wild. Such elusive behavior makes them difficult to study, and much of what is known about their ecology comes from indirect observations (Anderson 1963; Haas & Werner 1969; Joger 1984; Gasperetti 1988). Historically, Walterinnesia aegyptia Lataste, 1887 was considered the sole species within the genus. However, a taxonomic revision by Nilson & Rastegar-Pouyani (2007) recognized two distinct species: Walterinnesia aegyptia (type locality al-Qahira, surroundings of El Cairo, Egypt) and Walterinnesia morgani (Mocquard, 1905) (type locality Khuzestan, Iran), which had previously been under the synonym of W. aegyptia. According to these authors, W. morgani could be morphologically differentiated from W. aegyptia by having fewer anterior dorsal scale rows, fewer ventrals, fewer united subcaudals, and by juveniles having dorsal pinkish or reddish crossbars (1–3 scale rows-wide) on their otherwise black bodies. It has to be noted that the range of subcaudals given in the original description of Naja morgani Mocquard, 1905 (40–46), as well as the range of subcaudals (39–47 vs. 43–53), united subcaudals (0–11 vs. 2–13), and ventrals (180–200 vs. 186–198) given by Nilson & Rastegar-Pouyani (2007), clearly show a large to complete overlap with the meristic characters of W. aegyptia. Nevertheless, based on morphological differentiation, Nilson & Rastegar-Pouyani (2007) proposed the following distribution ranges for the two species: W. aegyptia, the western species, would be restricted to Egypt (east of the Nile), the Sinai Peninsula, southern Israel, Palestine, northwestern Saudi Arabia, and the western half of Jordan (Fig. 7 from Nilson & Rastegar-Pouyani 2007). In contrast, W. morgani, the eastern species, would be primarily distributed in Iran, from the western foothills of the Zagros Mountains, extending southeastward along the southern slopes of the range to the southeastern plains (Shafiei Bafti et al., 2023), as well as in southern Anatolia (Urfa province, Turkey), northern and southern Iraq, and central, eastern, and southwestern Saudi Arabia near the border with Yemen (Wadi Qatan). The situation in Syria was not addressed. Recent findings based on molecular data gathered by independent research groups have challenged Nilson & Rastegar-Pouyani’s (2007) hypothesis. Rajabizadeh (2018) compared the published sequence of the complete mitochondrial Cytochrome b (Cytb) gene from one specimen of W. aegyptia from North Africa (Slowinski & Keogh 2000) with Cytb sequences from two specimens from Israel and one specimen of W. morgani from Kermanshah, Iran, and found that all four specimens were genetically identical (Rajabizadeh 2018; work in progress by Rajabizadeh and collaborators). Calvete et al. (2021) compared mitochondrial (16S rDNA— 16S and Cytochrome Oxidase I- COI) and nuclear (melanocortin 1 receptor— MC1R) sequences from a specimen of Walterinnesia from Riyadh in Central Saudi Arabia (W. morgani according to the hypothesis by Nilson & Rastegar-Pouyani 2007) with a specimen of W. aegyptia from the Sinai Peninsula, Egypt. The results showed that the two specimens differed by just one change in 530 base pairs (bp) of 16S, one change in 669 bp of COI, and no changes in 686 bp of MC1 R. More recently, Alshammari et al. (2022) compared partial sequences of the mitochondrial genes 12S rRNA (12S) and 16S from Hail, Saudi Arabia with the Walterinnesia samples from Sinai, Egypt, and Riyadh, Saudi Arabia from Calvete et al. (2021), and one W. morgani from Riyadh from Šmid et al. (2021), and concluded that all Walterinnesia samples were nearly identical in both mitochondrial genes analyzed. All these independent molecular studies challenge Nilson & Rastegar-Pouyani’s (2007) taxonomic hypothesis and especially the delineation of the ranges of W. aegyptia and W. morgani, particularly within Saudi Arabia, and potentially in other regions as well. The apparent absence of crossbanded juveniles in Arabia (Anderson 1963; Haas & Werner 1969; Joger 1984; Gasperetti 1988; Nilson & Rastegar-Pouyani 2007), combined with the genetic similarity of all Walterinnesia specimens analyzed to date (Rajabizadeh 2018; Calvete et al. 2021; Šmid et al. 2021; Alshammari et al. 2022) and the results presented in this study, strongly suggest that Walterinnesia morgani does not represent a valid species, as already indicated by Kamali (2020), Shafiei Bafti et al. (2023), and Moradi et al. (2024). This raises significant doubts about the evolutionary relevance and validity of the pholidotic characters used by Nilson & Rastegar-Pouyani (2007) to resurrect W. morgani and to define the boundaries between the two Walterinnesia species. Although the distribution of Walterinnesia in Arabia was initially thought to be restricted to Saudi Arabia (Sindaco et al. 2013), recent research conducted between April 2015 and May 2024 has led to a remarkable discovery. Ten specimens of Walterinnesia have been found at 10 different localities in the Dhofar region, in the southwestern corner of Oman. This demonstrates the existence of geographically isolated populations of this species, separated by more than 1,000 km from the closest Arabian records in Wadi Qatan (18.10ºN, 44.12ºE), southwestern Saudi Arabia, and Al Huwah (23.05ºN 45.80ºE), 200 km southwest of Riyadh, Saudi Arabia (Sindaco et al. 2013). The geographic distribution of the Omani populations relative to all other Walterinnesia records is presented in Fig. 1A. Data on locality, collection date and time, activity, voucher presence, MorphoBank codes for specimen and habitat photographs, as well as morphological data for three specimens are presented in Appendix I. All specimens and habitat photographs obtained for this study are available from MorphoBank, Project Number 5621: https://morphobank.org/index.php/ Projects / Index. When comparing our meristic measurements with the diagnostic traits for W. morgani provided by Nilson & Rastegar-Pouyani (2007), the female specimens from Oman fall within the range of anterior dorsal scales (24–26) reported for W. aegyptia (24–27) but simultaneously exhibit the exact number of subcaudals (39–44) attributed to W. morgani. This clearly demonstrates that the diagnostic morphological criteria for identifying adult Walterinnesia specimens, as described by Nilson & Rastegar-Pouyani (2007), are inconsistent. DNA was extracted from tissue samples preserved in ethanol from three Walterinnesia specimens collected in Oman, plus one W. aegyptia from Riyadh (CN12091; Šmid et al. 2021; previously assigned to W. morgani based on locality data but see Alshammari et al. 2022) and one W. aegyptia from Sinai, Egypt (CN12020), using the MagAttract HMW DNA Kit (Qiagen, Valencia, CA, USA) following the manufacturer’s protocol. Four mitochondrial genes (COI, 12S, 16S, and Cytb) and one nuclear gene (MC1R) were PCR-amplified and Sanger-sequenced for both strands. The primers and PCR conditions are detailed in Table S1. Chromatograms were processed using Geneious Prime v.2023.0.4 and consensus sequences were combined with sequences from three additional Walterinnesia specimens of known geographic origin available on GenBank (Table 1). Heterozygous positions for the nuclear coding gene fragments were identified based on the presence of two peaks of approximately equal height at a single nucleotide site in both strands and were coded according to IUPAC ambiguity codes. Alignments were performed with MAFFT (Katoh et al. 2019), using the Q-INS-i algorithm for 16S and 12S, and with default settings for the other genes. Pairwise genetic differences were calculated independently for the five gene fragments for all Walterinnesia specimens included in this study using Geneious. For the nuclear gene fragments, SEQPHASE (Flot 2010) and PHASE v.2.1 (Stephens et al. 2001) were used to maintain haplotype information in a haplotype resolved FASTA file. Subsequently, FASTA files from all genes were converted to nexus using the R package ape v.5.7.1 (Paradis & Schliep 2019) and networks for each gene were inferred using the TCS Network Builder (Clement et al. 2000) with the “gap” parameter set to “missing”. The resulting graph file was visualized using TCSbu (Santos et al. 2016). The results of the genetic analyses are presented in Table 2 (genetic distances) and Figure S1 (haplotype networks) and clearly show that all three specimens from Oman are genetically identical to one another and nearly identical to the other specimens from Arabia (Riyadh and Hail) and Egypt (Sinai), confirming that all analyzed specimens belong to Walterinnesia aegyptia. The level of genetic variability across the 1,592 base pairs (bp) of mtDNA and 675 bp of nucDNA analyzed was very low, ranging from 0–2 mutations in 346 bp of the 12S mitochondrial gene (0–0.57% p -distance); 0–3 mutations in 332 bp of the 16S (0–0.9% p -distance); 0–1 mutation in 612 bp of the COI (0–0.16% p -distance); 0–1 mutation in 302 bp of the Cytb (0–0.33% p -distance); and 0–2 mutations in 675 bp of the MC1R nuclear gene (0–0.3% p -distance). Independent nucleotide networks for each gene revealed extensive haplotype sharing and minimal genetic differentiation among individuals, suggesting a high degree of genetic connectivity and limited population structuring. The taxonomic and distributional uncertainty within the genus Walterinnesia underscores the need for further molecular analyses to clarify species boundaries and geographic ranges. The absence of genetic data from Iraq, Turkey, Syria, and southeastern Iran represents a significant gap, as specimens from these regions are essential for a comprehensive understanding of the group’s g