Stegonotus caligocephalus, Kaiser & Lapin & O’Shea & Kaiser, 2020

Kaiser, Christine M., Lapin, Jack, O’Shea, Mark & Kaiser, Hinrich, 2020, Carefully examining Bornean Stegonotus (Serpentes, Colubridae): the montane groundsnake population in Sabah is a new and distinct species, Zootaxa 4894 (1), pp. 53-68 : 58-65

Stegonotus caligocephalus sp. nov.

Dark-headed Sabah Groundsnake

( Figs. 3–6 , Table 2)

Holotype. FMNH 251054 (field number RFI 49331), an adult female from near the Gunung Emas [Gunung Alab] Telekom tower, Crocker Range National Park , Tambunan District, Sabah State, Malaysia, ca. 5.83°N, 116.34°E ( Fig. 1 ), elevation ca. 1950 m, collected on 19 January 1993 by the American herpetologist Robert F. Inger (1920–2019), Curator of Amphibians and Reptiles at the Field Museum, Chicago, Illinois, USA, from 1954–1994 ( Matsui 2019).

Paratype. USNM 130244 , an adult female, from “Bundu Tuhan, Mount Kina Balu, Ranau District, Sabah ” [ Malaysia], ca. 6.05°N, 116.53°E, elevation 1370 m, collected on 1 August 1951 by the American mammalogist David H. Johnson (1912–1996), Curator of Mammals at the National Museum of Natural History, Smithsonian Institution, Washington, D.C., USA, from 1941–1965 ( Perry 2007) .

Referred specimen. ZMB 47866, a juvenile female from “Mt. Kinabalu” (no further locality data), elevation 1500 m, collected on 13 December 1988. Whereas the information in the collection’s ledger indicates that the specimen was collected by the German-South African herpetologist Thomas Ulber, this appears to be an error in both date and time. Ulber (in litt.) visited Mount Kinabalu in November 1989 but collected no snake specimens.

Diagnosis. Stegonotus caligocephalus is a medium-sized (maximum SVL = 1071 mm) snake with a relatively short tail (SCR ♀ = 0.21), the shortest in the genus Stegonotus . It can be distinguished from all other known Stegonotus species by the following combination of characters: (1) rostral extending onto the dorsal part of the head intruding into the internasal area but not reaching the level of the nasals (character state: gull wing +; Fig. 4A , A’); (2) area of prefrontals three times that of the internasals, internasal suture one-fourth to one-third the length of the prefrontal suture (SSR 0.25–0.33, mean = 0.28); (3) frontal clearly pentagonal with well-formed corners and a slightly concave anterior suture; (4) AE lies at the same level as the anterior end of the frontal; (5) length of frontal three-fourths that of the parietal suture (FPR = 0.75); (6) PfS one-fourth to one-third the length of external head-scale length (HSR = 0.25–0.31). (7) PF<± 90° (102°); (8) AP<= 122°, with its lateral ray directed posterolaterally ( Fig. 4A , A’); (9) two temporal scales touching the parietals; (10) three neck scales contacting parietals; (11) loreal as long as wide to one-third longer than wide (LSR = 0.6–1.0); (12) two preoculars, upper preocular 25% larger than lower one ( Fig. 4B , B’); (13) nine supralabials, two (SL4–5) touching the eye; SL5 larger than SL4 ( Figs. 4B , B’); (14) ten infralabials, five (IL1–5) touching the anterior genial ( Fig. 4 C’); (15) three gulars separating the posterior genials and the anterior-most gastrostege; (16) 17-17-15 dorsals; 211–218 (mean = 214 ± 2.9) ventrals, 56–65 (mean = 61 ± 3.7) paired subcaudals; (17) cloacal plate entire; (18) color in preservative (27 years post-collection): dorsum Tawny Olive (17) with some dark spots on lateral edge of a few rows of dorsal scales, head Grayish Olive (273) with some scattered lighter areas in Olive Gray (265). The supralabials and infralabials are Pale Buff (11) with mottled areas of Drab Gray (256) ( Figs. 3A , 4A, B ). The venter is Pale Horn Color (11) with dark areas on the anterior edge of the ventral scales and a dark “dot” on the lateral edge (Burnt Sienna, 38); the anterior one-third of the subcaudals are Burnt Sienna (38), creating a dark tail ( Fig. 3B ).

Comparisons. In the following comparisons, characteristics of S. caligocephalus are listed in parentheses. The number of specimens used to determine ranges of continuous characters follows species names. Based on our previous analysis of Stegonotus ( Kaiser et al. 2019) there does not appear to be any sexual dimorphism in the Bornean or Philippine species we compare in detail below. Given how few specimens of these populations are available in museum collections, combining scale count data for females and males in the absence of sexual dimorphism therefore increases our sample size and improves our analysis. Furthermore, given the geographic position of this Bornean population, most Stegonotus species can be considered extralimital to this comparison. It is highly unlikely that a New Guinean population disperses all the way across Wallacea into Borneo. After presenting an exhaustive comparative section in the description of S. ayamaru ( Kaiser et al. 2019) , we are confident in limiting our detailed, formal comparisons to Stegonotus from Borneo and the Philippines and eliminate other species based on different dorsal scale counts ( S. admiraltiensis , S. diehli , S. dorsalis , S. florensis , S. guentheri , S. poechi , S. sutteri ), significant differences in number of gastrosteges ( S. admiraltiensis , S. aruensis , S. ayamaru , S. derooijae , S. diehli , S. guentheri , S. heterurus , S. keyensis , S. lividus , S. melanolabiatus , S. modestus , S. parvus , S. poechi ), strongly diverging rostral morphology ( S. aruensis , S. australis , S. batjanensis , S. cucullatus , S. dorsalis , S. iridis , S. parvus , S. poechi ), the configuration of supralabials touching the eye ( S. aruensis , S. ayamaru , S. batjanensis , S. derooijae , S. diehli , S. florensis , S. heterurus , S. lividus , S. melanolabiatus , S. modestus , S. parvus , S. poechi , S. sutteri ), and differences in head and body color ( S. batjanensis , S. florensis , S. iridis , S. modestus , S. reticulatus , S. sutteri ). Where appropriate, gender is indicated throughout by subscripted male (³) and female (♀) symbols for ease of reference; when sexes are combined there is no subscript. Given that the three extant specimens of S. caligocephalus are all females, we separately provide data for females of the other species. A listing of those characters most relevant for interspecies comparisons is provided in Table 2.

Stegonotus borneensis (n = 1³, 1♀) has a significantly (unpaired t -test, t = 6.97, P <0.01) lower ventral scale count of 193–196 (211–218) and a significantly (t = 4.98, P <0.05) higher subcaudal count of 78 (56–65). In the single female specimen of S. borneensis , the values for ventral and subcaudal count are 193 and 78, respectively. As a consequence, the SCR in S. borneensis is higher, at 0.28–0.29 (0.21–0.24). Additional differences include (1) SSR = 1/5 (1/4 or 1/3); (2) FPR = 4/5–7/8 (3/4); and (3) a uniform dark body coloration (head darker than body). The closest known locality for S. borneensis is in Sarawak State, Malaysia, a horizontal distance of ca. 560 km by air from the type locality of S. caligocephalus . Furthermore, whereas S. borneensis appears to be a lowland tropical forest species (highest known elevation ca. 56 m at the type locality; Kaiser et al. 2018a), S. caligocephalus occupies montane forest habitat at elevations well above 1000 m. The lowest elevation for which we have found a published sighting ( Malkmus et al. 2002) is 1370 m, at the locality where the paratype was found.

Stegonotus muelleri (n = 13 ♀, 19 ³) is from the south-central Philippine Islands and differs by scale counts of V = 217–241 (t = 5.24, P <0.001), mean = 231 ± 5.5 (211–218, mean = 214 ± 2.9), V ♀ = 217–234, mean = 227 ± 4.5 (211–218, mean = 214 ± 2.9), SC = 78–108 (t = 7.74, P <0.001), mean = 96 ± 7.3 (56–65, mean = 61 ± 3.7), SC ♀ = 81–103, mean = 95 ± 6.5 (56–65, mean = 61 ± 3.7), SCR = 0.29 ( SCR = 0.22), SCR ♀ = 0.30 ( SCR ♀ = 0.22), D = 19-17-15 (17-17-15), SL = 8 in 98.6% or 9 in 1.4% of specimens (9 in 100% of specimens). Additional differences include (1) SSR = 2/5 (1/4 or 1/3); (2) three temporal scales touching the parietals (two); and (3) AE lies behind the anterior end of frontal (same level). The closest known locality for S. muelleri is on the island of Mindanao, ca. 1000 km by air from the type locality of S. caligocephalus .

Description of the holotype (FMNH 251054)—metrics (in mm) and pholidosis. An adult female; length 1071 SVL + 188 TL = 1259 TTL, TL 21% of TTL; head broad, distinct from body; snout rounded in dorsal view, angled in lateral view, mouth subterminal. Rostral character state gull wing + ( Fig. 4A, B ). Internasals shorter than wide, with SSR = 0.25. Prefrontal width 2/3 of prefrontal length, prefrontal area about three times the area of the internasals; HSR = 0.31. Frontal pentagonal with well-formed corners and a slightly concave anterior suture; AE lies at the same level as the anterior edge of the frontal; FPR = 0.81 ( Fig. 4A ); PF<= 100°. Supraoculars relatively large and roughly triangular, barely projecting (8% of supraocular length) beyond the anterior extent of the eye, projecting by 22% of their length posteriorly past the eye. LSR = 0.61. Parietals angular, about two thirds as wide as long; AP<= 122°, with lateral ray directed posterolaterally (lateral ray ca. 65° off the midline). Two equally-sized temporals (an anterior and a posterior temporal) touching the parietals; three neck scales in contact with the parietals.

Nasal two-thirds as tall as wide; naris half the length and the same width as the nasal; loreal about twice as long as wide; two preoculars with the lower half the area of the upper. Nine supralabials; SL4–5 touching the eye, SL5 40% larger than SL4; relative areas of supralabials 7 = 6> 8> 5> 9 = 4> 3> 2 = 1, SL 2 is rectangular whereas SL1 is triangular ( Fig. 4B , B’). Eye length two-thirds of eye-naris distance; pupil oval; internarial distance 60% of interorbital distance. Ten infralabials; IL1–5 contacting anterior genials, with two-thirds of the medial border of IL 5 in contact with the anterior genial; three gulars separating the posterior genial from the anterior-most gastrostege; length of mental groove one-third of ventral head length (VHR = 0.38; Fig. 4 C’). Dorsals in 17-17-15 rows, 214 ventrals, cloacal plate entire, 56 paired subcaudals, SCR ♀ = 0.21.

Description of the holotype (FMNH 251054)—coloration in preservative (27 years after collection). Dorsum Tawny Olive (Color 17) with some dark spots on lateral edge of a few rows of dorsal scales, head Grayish Olive (273) with scattered lighter areas in Olive Gray (265). The supralabials and infralabials are Pale Buff (1) with mottled areas of Drab-Gray (256) ( Fig. 3A , 4A, B ). The venter is Pale Horn (11) with dark areas on the anterior edge of the ventral scales and a dark “dot” on the lateral edge (Burnt Sienna, 38); the anterior one-third of the subcaudals are Burnt Sienna (38), creating a dark tail ( Figs. 3B ).

Key data for the paratype—metrics (in mm) and pholidosis. An adult female ( Fig. 5 ), 965 mm SVL + 191 mm TL = 1156 mm TTL, TL = 17% of TTL. V ♀ = 218, SC ♀ = 62, SCR ♀ = 0.22, D = 17-17-15, SL E = 4+5, SL = 9, IL = 10, IL G = 5.

Coloration in life. Based on a color analysis of the images in Fig. 6 , which have different lighting, we estimated the coloration in life as follows. The dorsal area of the head is Dark Grayish Brown (284) with some scattered Smoky White (261) spots. The color intensity fades laterally, and a lighter brown with elements of Cinnamon-Drab

(50) and Beige (254) dominates the supralabial area. The dorsal coloration of the body is in tones of Mars Brown (25) and much lighter than the head, fading in the area where dorsal and ventral scales meet to a light shade of Light Russet Vinaceous (246). This faded color is also visible on the anterolateral margins of each ventral scale.

Natural history. Whereas during several trips to the Mount Kinabalu area we have never encountered S. caligocephalus , we here provide information obtained from the literature and the field notes of colleagues. The species has only been encountered in forested areas of Crocker Range and Mount Kinabalu National Park above 1300 m, which consists mainly of montane oak-conifer rainforest ( Inger & Tan 1996). Whereas there is a consensus that the species is generally nocturnal ( Malkmus 1989, 1991, 1992, 1996; Manthey & Grossmann 1997; Max Dehling, pers. comm.; Tom Charlton, pers. comm.), there is some disagreement over the movement of this snake. However, Manthey & Grossmann (1997) consider this species to be an exclusively ground-dwelling form, Malkmus (1994) reported on some arboreal activity. In terms of feeding ecology, Malkmus (1994) reported on a predatory attempt on a gecko ( Cyrtodactylus baluensis ) in the only recorded feeding interaction.

Similar species. The species could, in principle, be confused with three similarly colored species ( Stuebing et al. 2014). Lycodon albofuscus 4 has keeled scales, is more slender, and its lowland habitat make it unlikely that this species occurs in sympatry with S. caligocephalus . On Borneo, older individuals of Lycodon subcinctus may have lost the characteristic light bands of younger snakes and may present an overall body coloration similar to S. caligocephalus . As is the case with L. albofuscus , the range of L. subcinctus may not extend into the higher elevations (above 1000 m) where S. caligocephalus is found.

2004; Wallach et al. 2014) but according to Das & Yaakob (2007), Das (2012), and the most recent available phylogenetic

information ( Figueroa et al. 2016; Wostl et al. 2017) the genus Lepturophis should be synonymized with Lycodon and this

species is therefore treated as Lycodon albofuscus here.

Important warning for field encounters. The snakes with which S. caligocephalus are most likely to be confused in the field are younger individuals of the venomous Sunda spitting cobra, Naja sumatrana , which actually appears to be more commonly encountered in the proximity of human habitations than S. caligocephalus . Both species are very similar in color and habits and it is difficult to tell them apart at a glance, especially in the dim cone of a flashlight by night. However, once engaged a spitting cobra will take up its characteristic threat posture with its head and body raised off the ground. In addition, spitting cobras are known to hiss dramatically when threatened. Needless to say, while the venom delivered by the bite of spitting cobras as well as the toxungen 5 delivered to the eyes are dangerous to humans, these snakes are quite harmless when left undisturbed. In general, great care must be taken when encountering any medium-sized dark snake in Borneo.

Etymology. The species name caligocephalus is a compound word derived from the Latin caligo, with meanings including dark(ness), out of sight, fog, and the Greek kephalos (κεϕαλοζ) meaning head. We deliberately select a mixed-language name given that this name is designed to have mixed morphological and metaphorical meanings. The species name first of all references the darker head color in this species. Secondly, the dark, looming shape of Mount Kinabalu, Southeast Asia’s tallest mountain and the montane home of S. caligocephalus , is a dramatic, head-like massif, and we wish to remind people that mountain regions are wonderful places to explore and preserve nature.

Figures

FIGURE 1. Map of Borneo and surrounding archipelagos, including portions of The Philippines, Sulawesi, and the Moluccas. The dotted lines demarcate Wallace’s Line, with larger dots representing the line according to Wallace (1860) and the smaller dots the one modified by Huxley (1868). Marked distributions of Stegonotus include the only known locality of S. borneensis in Sarawak (A; yellow dot); the holotype and paratype localities of S. caligocephalus sp. nov. (B and C, respectively; red dots); the distribution of S. muelleri in blue (Kaiser et al. 2018a), including the Philippine islands of Samar (1), Leyte (2), Dinagat (3), and Mindanao (4); islands with known records for S. batjanensis in gold (Lang 2013), including the northern Moluccan islands of Morotai (5), Halmahera (6), Ternate (7), Tidore (8), Kasiruta (9), Bacan (10), and Obi (11); and islands with confirmed records for S. modestus in green (Kaiser et al. 2018a), including Buru (12), Ambon (13), and Seram (14). The inset globe shows the position of the depicted area between Southeast Asia and Australia. Scale = 500 km.

FIGURE 3. Holotype of Stegonotus caligocephalus sp. nov. (FMNH 251054) in (A) dorsal and (B) ventral views. Note the enlarged vertebral scale row in (A) and the band-like coloration of the ventral scales.

FIGURE 4. Head of Stegonotus caligocephalus sp. nov. (holotype, FMNH 251054) in both photographs (left column) and line drawings (right column). Shown are dorsal (A, A’), right lateral (B, B’), and ventral (C, C’) views. Scale abbreviations for scales in dorsal view include rostral (R), internasals (IN), prefrontals (PF), frontal (F), supraoculars (SO), preoculars (PR), postoculars (PO), anterior temporals (AT), posterior temporals (PT), and parietals (P). In lateral view additional abbreviations include nasals (N), loreal (L), supralabials (SL, numbered), mental (M), and infralabials (IL, numbered). In ventral view anterior genials (AG) and posterior genials (PG) are labeled, as are gulars (Roman numerals).

FIGURE 5. Paratype of Stegonotus caligocephalus sp. nov. (USNM 130244) including the entire specimen in (A) dorsal and (B) ventral views, as well as (C) dorsal, (D) lateral, and (E) ventral views of the head. This specimen is also the paratype of S. borneensis Inger, 1967 and, according to the rules of zoological nomenclature, will retain its original type status with that species. Note that (D) is the correctly placed left lateral head view. This was inadvertently transposed in Fig. 30I in Kaiser et al. (2018a). 4 The species albofuscus was recognized as a member of Lepturophis by a variety of authors (e.g., Grismer et al. 2002,

FIGURE 6. Photographs of Stegonotus caligocephalus sp. nov. in life. Both individuals were encountered at night on the slopes of Mount Kinabalu at elevations> 1300 m and were not vouchered. Note the color offset between the darker head and the lighter body, seen particularly well with the lighting in (B) and (C). The individual in (A) and (B) has generally a more gray-toned body color. Photos by Tom Charlton (A, B) and Maximilian Dehling (C).