Calamaria annamensis Bourret, 1937 stat. nov.

(Table 3; Figs. 2–6; Appendix I)

Calamaria pavimentata annamensis Bourret, 1937: 32–33.

Holotype: MNHN-RA-1958.0456, formerly M. 540 (private Bourret), adult female.

Type locality: “ Dong Tam Ve (province de Quang-tri), Annam ”, now within Bac Huong Hoa Nature Reserve, Quang Tri Province, Vietnam (altitude 780 m asl); collected by Eugène Poilane in 1935 or 1936; deposited by René Bourret.

Specimens examined (n = 4; all within Bac Huong Hoa NR, Quang Tri Province, Vietnam). MNHN-RA-1958.0456, adult female, holotype; see above; VRTC NAP-17841, adult male, Huong Phung Commune, Huong Hoa District (16.8059°N, 106.57759°E; altitude 1,130 m asl.), collected by A.M. Bragin, 18 April 2025; VRTC NAP-17842–43, two adult females, Huong Phung Commune, Huong Hoa District (16.7909°N, 106.60055°E; altitude 1,010 m asl.), collected by A.M. Bragin, 19 April 2025.

Revised diagnosis. Calamaria annamensis stat. nov. (based on Bourret 1937a and this study) can be distinguished from all congeners by the following combination of morphological characters: rostral wider than high; prefrontal shorter than frontal and in contact with the first two supralabials; mental not in contact with the anterior chin shields; dorsal scales in 13-13-13 rows, smooth throughout; a single preocular and a single postocular; four supralabials, with the second and third contacting the eye; five infralabials; six scales surrounding the paraparietal; ventrals 182–204; paired subcaudals 14–22; tail relatively short (3.9–6.6% of total length), uniformly cylindrical anteriorly, tapering gradually posteriorly, and then abruptly tapering near the pointed tip. Dorsal surface uniform dark brown to olive-brown, sometimes with a faint pale or light-coloured nuchal ring on the neck; head dorsum dark brown with irregular yellowish mottling; labials yellowish-brown with dark speckling. Ventral surface light orange to salmon, each ventral with dark brown to blackish outer spots and corners; subcaudal region with a broad, distinct but discontinuous median black stripe. No light-coloured rings or blotches present on the tail.

Description of the holotype. MNHN-RA-1958.0456, adult female see (Fig. 2). Specimen in good condition but somewhat discoloured. Body slender, cylindrical (SVL 526 mm); tail short, thinner than body, uniformly cylindrical in its anterior portion, tapering gradually posteriorly, and then abruptly tapering near the pointed tip (TaL 26 mm, TaL/TL 4.7%). Head small, elliptical in dorsal view (HL 14.05 mm, HW 7.2 mm); eye small, round (ED 1.4 mm), smaller than the eye-nostril distance (EN 1.9 mm).

Head scalation. Rostral broader than high; dorsal portion visible from above approximately half the length of the prefrontal suture. Prefrontal shorter than frontal, not entering the orbit, in contact with the 1 st and 2 nd supralabials. Frontal hexagonal, longer than wide (FT-l 3.8 mm, FT-w 2.6 mm). Temporals absent. Paraparietal surrounded by six scales. One preocular on each side, higher than wide, slightly taller than postocular but shorter than eye diameter. One postocular on each side, also higher than wide. Nasals divided, small, bordered by rostral, prefrontal, and 1 st supralabial. Four supralabials on each side, the 2 nd and 3 rd entering the orbit; the 4 th largest in height (relative size: 4> 2> 3> 1), but the 2 nd one, rather low, as long as the 4 th one. Five infralabials on each side, the first three pairs contacting anterior chin shields; the 1 st pair meeting at the midline; the 4 th largest. Anterior chin shields much longer than wide, pentagonal, meeting at the midline. Posterior chin shields are shorter, in contact anteriorly, and separated posteriorly by first gulars. Mental triangular, not in contact with anterior chin shields.

Body scalation. Dorsal scale rows 13-13-13 throughout the body, all smooth. Ventrals 204 (+ 2 gulars) [VEN 205 in Bourret 1937a]; subcaudals 15 [SC 16 in Bourret 1937a], all paired. Cloacal plate undivided.

Dentition. Maxillary teeth modified (enlarged teeth), 8 at right (not counted at left).

Colouration of the holotype. In preservative, the dorsal surfaces of the head, body and tail are uniform greyish-brown; no longitudinal stripes are present; no collar is visible; the eye is black. Supralabials barely paler than the upper surface. The ventral surface of the body is greyish-brown, barely paler than the dorsum, with many brown spots scattered on the outer parts of the ventral plates, and darker tips of ventrals darker. The underside of the tail is like the venter, with a brown, discontinuous stripe extending from the first subcaudal to the tip.

Description of specimen VRTC NAP-17843. Topotype, adult female (see Figs. 3–4). Specimen VRTC NAP-17843 is in excellent condition. Body slender, cylindrical (SVL 435 mm); tail short, thinner than body, uniformly cylindrical in its anterior portion, tapering gradually posteriorly, and then abruptly tapering near the tip (TaL 21.0 mm, TaL/TL 4.6%) (Fig. 3A–B). Head small, elliptical in dorsal view (HL 11.5 mm, HW 6.5 mm); eye small, round (ED 1.1 mm), smaller than the eye-nostril distance (EN 1.8 mm) (Fig. 3E–F).

Head scalation. (Figs. 3C–F, 4). Rostral broader than high (RT-w 2.0 mm, RT-h 1.5 mm), dorsal portion visible from above approximately half the length of the prefrontal suture (Figs. 3C, 4B). Prefrontal shorter than frontal, not entering the orbit, in contact with the 1 st and 2 nd supralabials (Figs. 3C, 4B). Frontal hexagonal, longer than wide (FT-l 3.1 mm, FT-w 2.3 mm) (Fig. 3C). Paraparietal surrounded by six scales (Figs. 3E–F, 4A–B). One preocular on each side, higher than wide, slightly taller than postocular but shorter than eye diameter (Figs. 3E–F, 4A). One postocular on each side, also higher than wide (Figs. 3E–F, 4A). Nasals divided, small, barely surrounding nostril, bordered by rostral, prefrontal, and 1 st supralabial (Figs. 3E–F, 4A). Four supralabials on each side, the 2 nd and 3 rd entering the orbit; the 4 th largest (relative size: 4> 2> 3> 1) (Figs. 3E–F, 4A). Five infralabials on each side, the first three pairs contacting anterior chin shields; the 1 st pair meeting at midline; the 4 th largest (Figs. 3D, 4C). Anterior chin shields longer than wide, pentagonal, meeting at midline (Figs. 3D, 4C). Posterior chin shields are shorter, in contact anteriorly, and separated posteriorly by first gulars (Figs. 3D, 4C). Mental triangular, not in contact with anterior chin shields (Figs. 3D, 4C).

Body scalation. (Fig. 3A). Dorsal scale rows 13-13-13 throughout, all smooth. Ventrals 196 (+ 3 gulars in midline, the first gular in contact with anterior chin shields); subcaudals 14, all paired, first pair not in contact. Cloacal plate undivided.

Dentition. (Fig. 5B–C). Maxillary teeth modified (basally expanded) flask-shaped, nine on each side (9/9). The palatines bear six unmodified conical teeth each (6/6). The pterygoids bear 11 unmodified conical teeth each, notably smaller in size than those on other bones (11/11), many of teeth on the pterygoids appear to be lost or broken and could only be indentified by respective tooth sockets. The dentaries bear eight unmodified conical teeth each, shrinking in size posteriorly (8/8). The pulp cavity of unmodified tooth is conical and slightly curved, paralleling the shape of the tooth. The pulp cavity of modified tooth is expanded in its basement posteriorly, close to what is described in Inger & Marx (1965).

Skull morphology. The skull of specimen VRTC NAP-17843 is subcylindrical in general shape, with a long postorbital part and rather short orbits (Fig. 5). The snout is composed of the premaxilla, the nasals, the septomaxillars and the vomers. The braincase is composed of the prefrontals, the frontals, the jugals (the homology of this bone is questionable, and some studies interpret it as the postorbital), the parietals, the parabasiphenoid, the basioccipital, the prootics and the otoccipitals. The upper jaw is composed of the maxillae, the palatines, the ectopterygoids and the pterygoids. The supratemporals and the quadrates form suspensoriums, connecting the braincase with the lower jaw. Finally, each half of the lower jaw is composed of the compound bone, the angular, the splenial and the dentary. The maxillaries, the palatines, the pterygoids, and the dentaries are dentigerous. The maxillary teeth are modified, flask-shaped, while all the others are unmodified, conical.

Colouration of the VRTC NAP-17843. In life, the dorsal surfaces of the head, body, and tail are a uniform dark olive-brown, each scale with darker margins, producing a finely reticulated appearance; no longitudinal stripes are present (Fig. 6A). The dorsal surface of the head is dark brown with irregular yellowish to olive mottling; the eye is black. The supralabials are yellowish-orange with dark speckling, especially posteriorly, while the infralabials and anterior chin shields are light orange to salmon with scattered dark mottling. The ventral surface of the body is light orange to salmon, with dark pigmentation confined to the outer margins of the ventral scales, forming a narrow, diffuse lateral line. The underside of the tail is light orange to salmon with a prominent median dark stripe extending from the second subcaudal to the tip.

Variation. Variation in measurements and scalation of the holotype and three newly collected topotype specimens is summarised in Table 3. The longest specimen and the largest female is the holotype MNHN-RA-1958.0456 reaching a TL of 552 mm, whereas the only male, topotype VRTC NAP-17841, reaching a TL of 304 mm. Sexual dimorphism is evident in several characters. Females possess more ventral scales (VEN 196–204) than the single male (VEN 182), but fewer subcaudal scales (SC 14–15 in females vs. 22 in the male). Tail proportions also differ markedly between sexes: the adult male has a relatively longer tail (TaL/TL ratio 6.6%) compared to adult females (TaL/TL ratio 3.9–4.7%). Additionally, Fig. 6A (VRTC NAP-17843) shows a specimen with a faint nuchal ring, while Fig. 6B displays a specimen, sex unknown, not collected from the type locality. This specimen is included as it exhibits the characteristic nuchal ring, contributing to the overall variation within the species.

Etymology. The specific epithet annamensis is a Latinised adjective in the nominative singular, referring to the species distribution, which is currently known only from Bac Huong Hoa NR, Quang Tri Province, Vietnam, in the part of Vietnam formerly called « Annam ». We propose the following common names: “Annamite Reed Snake” (ENglish), “RắN mai gầm HướNg HOá” (VietNamese), “Аннамская карликовая Змея” (ANNamskaya karlikOvaya zmeya; Russian), and “Calamaire de l’Annam” (French).

Comparison. Comparative morphological data for the Calamaria annamensis stat. nov. and currently recognised members of the genus Calamaria from mainland China and the Indo-Burma Region are presented in Table 4.

Calamaria annamensis stat. nov. is readily distinguished from C. lumbricoidea (distributed in Thailand, Malaysia, Singapore, Indonesia, Brunei, and the Philippines) and C. schlegeli (Thailand, Malaysia, Indonesia, and Singapore) by having four supralabials, with the second and third in contact with the eye (vs. five supralabials, with the third and fourth in contact). Furthermore, both species are restricted to areas south of the Isthmus of Kra in Peninsular Malaysia and have not been recorded from mainland Indochina.

Calamaria annamensis stat. nov. also differs from C. lovii ingermarxorum Darevsky & Orlov (Gia Lai Province, Vietnam), C. nebulosa (Phongsaly Province, Laos), C. thanhi (Quang Binh Province, Vietnam), and C. yunnanensis (Yunnan Province, China; possibly Dien Bien Province, Vietnam) by the presence of a preocular scale (vs. absent). It further differs from C. concolor (Hue and Da Nang cities, Vietnam), C. gialaiensis (Gia Lai Province, Vietnam), C. lovii ingermarxorum, C. lumbricoidea, and C. sangi (Quang Ngai Province, Vietnam) in having the mental not contacting the chin shields (vs. contacting).

Calamaria annamensis stat. nov. differs from C. abramovi (Quang Ngai Province, Vietnam) in having slightly higher ventrals in both sexes (VEN 182 in the only known male, 196–204 in females vs. 159 in males, 174 in females), fewer subcaudals in both sexes (SC 22 in male, 14–15 in females vs. 26 in males, 20 in females), a larger maximum total length in males (304 mm vs. 139 mm), a slightly shorter relative tail length in both sexes (TaL/TL ratio 6.6% in male, 3.9–4.7% in females vs. 13.3% in males, 7.1% in females), an obtusely pointed tail tip (vs. sharply pointed), dorsal colouration uniform dark brown to olive-brown (vs. uniformly black), and the presence of light blotches on the neck (vs. absent).

Compared to Calamaria andersoni (Yunnan Province, China), Calamaria annamensis stat. nov. has higher ventral counts in both sexes (VEN 182 in the only known male, 196–204 in females vs. 164–172 in males, 186 in females), a smaller maximum total length in males (max TL 304 mm vs. 351 mm) but a larger maximum total length in females (max TL 552 mm vs. 312 mm), a slightly shorter relative tail length in both sexes (TaL/TL ratio 6.6% in male, 3.9–4.7% in females vs. 8.8–9.2% in males, 5.8% in females), uniform dark brown to olive-brown dorsal colouration (vs. brownish with faint narrow black lateral stripes), and the presence of light blotches on the neck (vs. absent).

It is distinguished from Calamaria arcana (Zhejiang, Guangdong, Fujian, and Hunan provinces, China) by having higher ventral counts in both sexes (VEN 182 in the only known male, 196–204 in females vs. 170–176 in males, 192 in females), slightly higher subcaudal counts in females (SC 14–15 vs. 12), a slightly shorter relative tail length in males (TaL/TL ratio 6.6% vs. 7.2–11.8%), 8–9 maxillary teeth (vs. 10).

Calamaria annamensis stat. nov. further differs from C. berezowskii (Sichuan Province, China) in having higher ventral counts in both sexes (VEN 182 in the only known male, 196–204 in females vs. 149–155 in males, 153–171 in females), a larger maximum total length in females (max TL 552 mm vs. 245 mm), a slightly shorter relative tail length in females (TaL/TL ratio 3.9–4.7% vs. 6.5–6.9%), and light orange to salmonventral colouration with small black spots (vs. light khaki or white).

It is distinct from Calamaria dominici (Lam Dong Province, Vietnam) in having higher ventral counts in females (VEN 196–204 vs. 174), slightly fewer subcaudals in females (SC 14–15 vs. 17–18), a slightly shorter relative tail length in females (TaL/TL ratio 3.9–4.7% vs. 6.2%), uniform dark brown to olive-brown dorsal colouration (vs. dark with irregular yellow blotches), a light orange to salmonventral surface with small black spots (vs. dark with few yellow blotches and bands), and the presence of light blotches on the neck (vs. absent).

It can be distinguished from Calamaria jinggangensis (Jiangxi, Guizhou, and possibly Hunan provinces, China) by having higher ventral counts in both sexes (VEN 182 in the only known male, 196–204 in females vs. 157–158 in males, 179 in females), slightly higher subcaudal counts in females (SC 14–15 vs. 12), larger maximum total length in both sexes (max TL 304 mm in male, 552 mm in females vs. 260 mm in males, 364 mm in females), shorter relative tail length in males (TaL/TL ratio 6.6% vs. 15%), by having 6 palatine teeth (vs. 10), and light orange to salmonventral colouration with small black spots (vs. dark orange).

Calamaria annamensis stat. nov. differs from C. mizoramensis (Mizoram State, India; possibly Bangladesh) by having higher ventral counts in both sexes (VEN 182 in the only known males, 196–204 in females vs. 147–155 in males, 166–175 in females), a smaller maximum total length in both sexes (max TL 304 mm in male, 552 mm in females vs. 251 mm in males, 258 mm in females), shorter relative tail length in males (TaL/TL ratio 6.6% vs. 10.3–13.1%), uniform dark brown to olive-brown dorsal colouration (vs. dark brown to blackish-brown with six narrow dark longitudinal stripes), and light orange to salmon ventral colouration (vs. yellow).

Calamaria annamensis stat. nov. differs from C. pavimentata alleged to originate Java Island, Indonesia, by having higher ventral counts in both sexes (VEN 182 in male, 196–204 in females vs. 151 in male, 136 in female), fewer subcaudal counts in males (SC 22 vs. 27) but higher subcaudal counts in females (SC 14–15 vs. 13). Calamaria annamensis stat. nov. also differs from C. pavimentata from Longzhou County, Chongzuo City, Guangxi Province, China (as defined by Yeung et al. 2022) by having a non-tapering tail (vs. tapering), an obtusely pointed tail tip (vs. sharply pointed), absence of light spots at the base of the tail (vs. present), and absence of distinct narrow dark longitudinal lines on the dorsum (vs. present).

It differs from Calamaria septentrionalis (southern China and northern Vietnam) in having higher ventral counts in both sexes (VEN 182 in the only known male, 196–204 in females vs. 148–166 in males, 168–188 in females), more subcaudals in both sexes (SC 22 in male, 14–15 in females vs. 15–19 in males, 6–11 in females), a smaller maximum total length in males (max TL 304 mm vs. 344 mm) but a larger maximum total length in females (552 mm vs. 384 mm), an obtusely pointed tail tip (vs. broadly rounded), uniform dark brown to olive-brown dorsal colouration (vs. dark brown or black, often with a yellow neck ring), light orange to salmonventral colouration with small black spots (vs. yellow with small black spots), and absence of light spots at the base of the tail (vs. present).

Compared to Calamaria strigiventris (Lam Dong and Khanh Hoa provinces, Vietnam), Calamaria annamensis stat. nov. differs in having higher ventral counts in both sexes (VEN 182 in the only known male, 196–204 in females vs. 130–168 in males, 176–183 in females), fewer subcaudals in both sexes (SC 22 in male, 14–15 in females vs. 29–31 in males, 20–22 in females), a smaller maximum total length in males (max TL 304 mm vs. 362 mm) but a larger maximum total length in females (552 mm vs. 367 mm), shorter relative tail length in both sexes (TaL/TL ratio 6.6% in male, 3.9–4.7% in females vs. 11.2–17.9% in males, 8.4–8.6% in females), light orange to salmonventral colouration with small black spots (vs. bright yellow with black stripes), and the presence of light blotches on the neck (vs. absent).

Notably, Calamaria annamensis stat. nov. differs from C. synergis (Yunnan Province, China) by having six shields surrounding the paraparietal (vs. five), higher ventral counts in males (VEN 182 vs. 161–166), a slightly larger maximum total length in males (max TL 304 mm vs. 284 mm), and light orange to salmonventral colouration with small black spots (vs. light khaki, immaculate).

Lastly, Calamaria annamensis stat. nov. morphologically differs from the three junior synonyms of C. pavimentata occurring in the Indo-Burma region. From Calamaria siamensis Günther, 1864 (type locality: “southern Siam and Laos Mountains, Cochinchina ”, corresponding to the Luangphabang Range in north-western Laos), it differs by having a higher maximum ventral count (VEN 205 vs. 190) and by lacking white spots or light rings on the tail (vs. two or three pairs of white spots in both sexes, depending on tail length). It also differs in dorsal colouration, being uniform dark brown to olive-brown without longitudinal blackish lines (vs. brownish above, minutely dotted with black, sometimes with seven indistinct blackish lines), and in ventral pattern, having light orange to salmon with dark brown to blackish outer corners and a broad, distinct median black stripe on the subcaudal region (vs. ventrals and subcaudals whitish, densely punctulated with brown, with only the hind margins of the shields immaculate) (data for Calamaria siamensis after Günther, 1864). From Calamaria pavimentata var. uniformis Smith, 1921 (type locality: Langbian Plateau, Lam Dong Province, Vietnam), it differs by having higher ventral counts in both sexes (VEN 182 in the only known male, 196–205 in females vs. 143–149 in males, 166–167 in females), lower subcaudal counts (SC 22 in males, 14–16 in females vs. 30–34 in males, 18–19 in females), and by lacking a median longitudinal line along the belly (vs. present). It further differs in ventral colouration, being light orange to salmonwith dark brown to blackish outer corners and a broad (vs. ventrals yellowish white without a dark stripe), present median black stripe on the subcaudal region (vs. absent) (data for Calamaria pavimentata var. uniformis after Smith, 1921). From Calamaria pavimentata banaensis Bourret, 1934 (type locality: Ba Na-Nui Chua NR, Da Nang City, Vietnam), it differs by having higher ventral counts in both sexes (VEN 182 in the only known males, 196–205 in females vs. 157–160 in males, 178–179 in females) and by lacking a dark midventral stripe on the belly (vs. present) (data for Calamaria pavimentata var. uniformis after Bourret, 1934).

Natural history notes. Calamaria annamensis stat. nov. was recorded in Bac Huong Hoa NR, Quang Tri Province, Vietnam, which encompasses a series of ridges in the southern sector of the Northern Annamites (Truong Son Mountains, Fig. 7). The landscape comprised shale slopes at approximately 1,000 m asl., culminating in Sa Mui Peak (1,550 m asl.), which extends from northwest to southeast along the present Quang Tri provincial border (Mahood & Tran 2008). At the foot of the ridge, at elevations of 800 m asl and lower, valleys with limestone karst ridges and outcrops dissected by large rivers were present; however, no observations of the resurrected species were made in these karst areas. Two specimens (VRTC NAP-17842 and VRTC NAP-17843) were found road-killed on the Ho Chi Minh Tay highway [similar to records of Gongylosoma pallidonuchale (Poyarkov, Nguyen & Vogel) reported by Hoang et al. 2025], and one specimen (VRTC NAP-17841) was observed actively moving within accumulated leaf litter in a roadside drainage ditch. The vegetation at the latter site consisted of a trampled, buffalo-grazed roadside clearing dominated by Melastoma L., Leucaena Benth., Schima Reinw. ex Blume, and Dicranopteris Bernh., at 1,050 m asl. Calamaria annamensis stat. nov. likely reached this disturbed site accidentally during movement along the slopes of the ridge.

The species is presumed to inhabit leaf litter on the slopes of the shale ridge at elevations of 1,000 –1,500 m asl., covered by polydominant montane high-canopy forest dominated by Fagaceae Dumort. and Cephalotaxus hainanensis H.L. Li, with an understory of Piper L., Capparis L., and Annonaceae Juss. All specimens were found at dusk (18:00–21:00) following heavy rainfall, with ambient temperatures of 18–23°C. These rains had just commenced after the end of the dry season and likely triggered snake activity. This is a small, crepuscular, and nocturnal species, foraging within the upper soil layer and leaf litter. Sympatric reptile species recorded in the same habitat included Cyrtodactylus cf. pseudoquadrivirgatus Rösler, Vu, Nguyen, Ngo & Ziegler, Scincella cf. rufocaudata (Darevsky & Nguyen), Sphenomorphus cf. maculatus (Blyth), S. cf. indicus Gray, Calotes emma Gray, C. versicolor Daudin, Trimeresurus albolabris Gray, T. vogeli David, Vidal & Pauwels, Fimbrios cf. klossi Smith, and Bungarus candidus (Linnaeus).

Distribution and conservation status. Calamaria annamensis stat. nov. is currently confirmed only from Bac Huong Hoa NR, Quang Tri Province, Vietnam, and is likely to occur in adjacent areas, including Dong Chau-Khe Nuoc Trong NR in the same province. The species inhabits elevations ranging from 780 to 1,500 m asl. While occurring within a protected area, Calamaria annamensis stat. nov. is a secretive and presumably rare forest specialist whose habitats are locally impacted by human activities. These include livestock grazing (goats, cattle, and buffalo) that can alter vegetation structure both along roads and in forest interiors, small-scale collection of young bamboo shoots and associated temporary field camps and fires, as well as occasional felling of large trees to create canopy gaps for bamboo regeneration, and direct mortality from road traffic, as evidenced by specimens found road-killed on the Ho Chi Minh Tay Highway. Although no immediate large-scale threats have been identified, such pressures may affect habitat quality over time. Based on the currently available information, we recommend that the species be assessed as Near Threatened (NT) under the IUCN Red List Categories and Criteria (IUCN, 2025).