Trimeresurus puniceus ( Kuhl, 1824 )

David, Patrick, Vogel, Gernot, Vijayakumar, S. P. & Vidal, Nicolas, 2006, A revision of the Trimeresurus puniceus-complex (Serpentes: Viperidae: Crotalinae) based on morphological and molecular data, Zootaxa 1293 (1), pp. 1-78 : 22-33

Trimeresurus puniceus ( Kuhl, 1824)

( Figs. 5–11 )

C. [raspedocephalus] puniceus Kuhl, 1824: 80 . Type locality. Java, by inference. Syntypes. RMNH 1557 (female), RMNH 1558 (1)–(2) (females), from “ Java ” (no type specimen was identified in the brief original description. However, the museum tags of these three specimens identify them as “cotypes”). Deposited by H. Boie and H. Macklot.

Atropos acontia Gray, 1842: 49 . Type locality. “ Java ”. Holotype. Specimen depicted in Seba (1735), Pl. 64: Fig. 1 .

Material (40 specimens). INDONESIA. Java Island . Province of Jawa Barat (West Java Province). IRSNB 1417 (male), “Cheribon, Java”, now Cirebon ; IRSNB 1417 β(1)–β(2) (females) , IRSNB 1417 β(3) (male), Bandung ; IRSNB 1417 Ύ (female), “Buitenzorg”, now Bogor ; MNHN 1918.28 (female), “Mt. Gedeh”, now Mt. Gede ; RMNH 6981 (male), “Near Soekaboemi”, now Sukabumi ; RMNH 7259 A–B (females) , RMNH 7259 C (male), “ Kali Baroe, Bandjoewangi ”, now Banjuwangi (or Banyuwangi), near Cilegon ; RMNH 8416 A–B (females) , RMNH 8416 C (male), “Summit of Megamendoeng, half­way between Buitenzorg and Sindanglaja ”, now Mt. Megamendung, between Bogor and Cianjur ; RMNH 8990 (female), “ Near Bandung, Western Java ” ; RMNH 8991 (female), “ Poentjakpass, ca 1400 mn Western Java ” , RMNH 14250 A (female) , RMNH 14250 B (male), “ Poentjakpass, Western Java , ca 1000 m ”, now near Pancak, SE of Sukabumi ; RMNH 11408 (male), “Botanical Garden, Buitenzorg”, now Bogor ; RMNH 14269 , RMNH 14271 (2 females) , RMNH 14270 , RMNH 14272–74 (4 males) , “Tjikadjang, Western Java”, now Cikajang ; USNM 44116 (male), “Mt. Salek, Goenoeng Boender”, now Mt. Salak. Province of Jawa Tengah (Central Java Province). RMNH 8504 (male), “Wonosobo”, ca 1000 m; RMNH 8613 (male), “ Modjotengah, Wonosobo , ca 1100 m ”, now Mojotengah ; RMNH 14261 (female), “ Sapoeran, Wonosobo , ca 600 m ”, now Sapuran, SE of Wonosobo. Province of Jawa Timur. RMNH 8986 (female) , RMNH 8988 (male), “Nongkodjajar, ca 1200 m, Mts. Tengger, East Java”, now Mt. Nongkojajar , near Nongkejajar ; ZMH R­8206–8209, no precise locality.— Java: no precise locality . MNHN 7611 , MNHN 1991.3286 (2 males) , RMNH 1557 , RMNH 1558 (1)–(2) (3 females; syntypes) , ZFMK 32511 (male), “Java” . Sumatra Island . Province of Sumatera Selatan (South Sumatra Province). ZFMK 32512 (male) , Lahat. Province of Bengkulu. S 0011 (male), PSGV 770 (female) (both via the pet trade) ; ZMH R­08204, “ Lebong­Tandai, 3°01'S 101°54'E, Bengkoelen , Sumatra”, now Lebongtandai. Sumatra: no precise locality . MNHN 2002.0405 (female; via the pet trade) .

Taxonomic comments. McDiarmid et al. (1999) discussed the authorship of this species and showed that, although Cophias punicea Boie, 1827 is largely regarded in the literature as the original name of this taxon (see, for example, Boulenger, 1896; De Rooij, 1917; De Haas, 1950; Golay et al., 1993; David & Ineich, 1999), its first valid description should be indeed credited to Kuhl (1824). This original description was very brief, as Kuhl (1824) meant to give a diagnosis of the new genus Craspedocephalus in which was included the sole species “ Trigonocephalus puniceus Reinw ”, a nomen nudum credited to Reinwardt on the basis of a label of the “Musée des Pays­Bas ”. Such a species description by indication, in which a generic description published before 1931 also applies to its originally included species, is valid in regard to the Code (Art. 12.2.6).

The taxonomy of the Trimeresurus puniceus ­group in Sumatra and in Western Indonesia is still provisional. The status of specimens from Natuna and Anamba Islands is discussed below under the account of Trimeresurus cf. puniceus . In Sumatra Island, we identify two different morphotypes among our sample of only seven specimens from this island (see below). Although specimens from the southern provinces of Bengkulu and Lampang cannot be distinguished morphogically from Javan specimens of Trimeresurus puniceus , specimens from Java and South Sumatra differ by a rather high genetic distance (about 9.5%). Furthermore, the DNA sequences of two specimens, unfortunately without more precise locality than Bengkulu Province, South Sumatra, showed a genetic distance between them of about 8.5 %. At the present time, we cannot determine if we are involving a cryptic species from South Sumatra. The morphological analyses do not show significant differences between our available specimens, and we refer them to Trimeresurus puniceus . Two specimens from western Sumatra and North Pagai Islands, morphologically distinct, are referred to as Trimeresurus cf. puniceus (see below).

Diagnosis. A species of the genus Trimeresurus from Java Island and South Sumatra, characterized by the combination of the following characters: (1) an overall grey or brown pattern with 20–30 darker crossbands, distinctly related to the sex: in males, background colour in various shades of grey or yellowish­brown, with darker, irregular, constricted, dorsolateral blotches, often heavily powdered with cream and dark dots, giving an “ashy” or lichen­like appearance; in females, pattern much less contrasted in shades of grey or yellowhish­brown with more or less brown subrectangular dorsolateral blotches; males are more brightly colored, with a more complex and contrasted pattern than females, but not overally darker or lighter than females; (2) a distinctly projected and raised snout, strongly obliquely truncated when seen from the side, rounded or subrectangular seen from above; (3) internasals normal or weakly bilobate, either sligthly raised or upturned, or flat; (4) 21 or 23 (rarely 19) DSR at midbody; (5) 1 st supralabial distinct from nasal; (6) 2 nd supralabials never bordering the loreal pit, topped by a prefoveal scale; (7) 2–5 small, distinctly raised supraoculars; (8) VEN: 153–167, SC: 41–59; (9) occipital and temporal scales distinctly keeled in both sexes; (10) IL of the first pair not in contact each with the other; (11) hemipenes short, reaching 11–13 th SC, nearly entirely spinose.

Description and variation. We did not identify morphological variation correlated to the geographical origin among our sample of 45 specimens.

The maximal known total length known is 920 mm ( Ryabov et al., 2002: 246). The longest specimen examined by us is 790 mm long (SVL 678 mm, TaL 112 mm) for a female (IRSNB 1417 β(1)). The largest examined male is 684 mm long (SVL 577 mm, TaL 107 mm; RMNH 14270).

Body rather slender in males, stout in females. Head distinctly triangular, average, amounting (in adults above SVL 400 mm) for 5.4–7.1 % of SVL (x = 6.1 %) in males, 5.8–7.6 % of SVL (x = 6.8 %) in females, wide at its base, moderately thick and flattened in males, thick and swollen in females when seen from the side. Snout distinctly projected and raised anteriorly, strongly obliquely trunctated when seen from the side, with a distinct canthus rostralis, usually rounded (34/ 45 specimens) or subrectangular with slightly bilobate nasals (11/45) seen from above, rather long, amounting for 22.9–28.3 % (x = 25.5 %) of HL in males and 23.9–28.6 % (x = 25.2 %) of HL in females, or 2.1–2.8 (x = 2.4) times as long as diameter of eye in males, 2.0–3.2 (x = 2.7) times in females. Eye small, amounting for 0.6–0.8 (x = 0.7) times in males and 0.5–0.8 (x = 0.6) times in females of the distance eye–lip. Tail average, tapering progressively and strongly prehensile. Ratio TaL/TL: 0.128 –0.171, with a sexual dimorphism (see below).

DSR: (21)23–27(29)—(19)21–23—15–17, more or less strongly keeled, always smooth on the first DSR in both sexes.

On a sample of 45 specimens, the number of MSR is 19 (3/45), 21 (28/45) and 23 (14/ 45). The number of PSR is 15 (26/ 44 specimens), 16 (2/44) and 17 (16/44).

VEN: 153–169 (plus 1–2 preventrals); SC: 41–59, paired; anal shield entire.

These values are drawn from our own data only, and are limited by the size of our sample. The literature is useless, as no author explicitly restricted the specific nomen to Javanese specimens, at the exception of Hoge & Romano (1974). Unfortunately, these authors did not provide variation of morphological characters of their available specimens.

Rostral barely visible from above, about 1.3–1.6 times broader than high, triangular; nasals subrectangular, undivided; usually 2 (83/90 occurrences), rarely 1 (7/90) internasals on each side, slightly enlarged, normal or weakly bilobate, usually slightly raised or upturned by the sides of the snout; internasals or group of internasals separated by 2 (in 36/ 44 specimens), rarely 1 (4/44) or 3 (4/44) small scales; 3–5 canthal scales, slightly larger than adjacent snout scales, bordering the canthus rostralis between the internasal and corresponding supraocular; 2 elongate subrectangular loreal scales between upper preocular and nasal in all examined specimens; two upper preoculars above the loreal pit, elongated and in contact with the loreal; lower preocular forming the lower margin of loreal pit; 2–3 small postoculars; 2–5 small, narrow supraoculars on each side (2: 7/90 occurrences; 3: 35/90, 4: 38/90, 5: 10/90; total number: 4–10; x = 7.2, s = 1.4) elongate or granular, more or less pyramidal, not or slightly extending out of the margin of the head, distinctly convex (in 8/ 45 specimens) or raised (37/45), in total 3.0–4.0 (x = 3.7) times as long as wide, 0.3–0.4 (x = 0.35) time as wide as the group of internasals, irregularly bordered on their inner margins by the upper head scales; 7–10 (x = 8.1, s = 0.9) slightly enlarged scales on upper snout surface on a line between the scale separating the internasals and a line connecting the anterior margins of eyes, smooth, juxtaposed, irregular in shape (8 in 16/45 and 9 in 13/ 45 specimens); 12–15 (x = 13.0, s = 0.8) cephalic scales on a line between supraoculars (12: 10/45; 13: 28/45; 14: 3/45; 15: 4/45), small, smooth, flat and juxtaposed; occipital scales larger than cephalic scales, in males moderately or strongly keeled, in females usually moderately (17/22) or strongly keeled (2/22), smooth in 3/ 22 specimens; temporal scales small, subequal, in 2 or 3 rows, distinctly or strongly keeled in males and in 19/ 22 females, smooth in 3/ 19 females; usually one thin, elongated, crescent­like subocular, sometimes divided into a long scale and a much shorter one anteriorly; 9–13, usually 10–12 SL (x = 11.1, s = 0.8) (9: 1/90 occurrences; 10: 15/90; 11: 47/90; 12: 25/90; 13: 2/90, with most often the combinations 10–11, 11 or 11–12 SL); 1 st SL, rather short, always separated from nasal; 2 nd SL, short, never bordering the loreal pit, topped by a prefoveal scale that borders entirely the pit and is separated from nasal by 1–3 (usually 2) scales; 3 rd SL longest and highest, 1.0–1.4 times as long as high (x = 1.2), separated from the subocular by 1 (37/90 occurences) or 2 (53/ 90) scales; 4 th SL as long as high, 0.7–1.0 (x = 0.85) time as low as 3 rd SL, separated from subocular by 1 (2/90 occurrences), 2 (74/90) or 3 (14/90) scales on each side; 5 th and posterior SL smaller than 4 th one, 5 th SL separated from subocular by 2 (in 48/90 occurrences) or 3 (42/90) scale rows of similar size, with a sexual dimorphismus (see below); 11–15, usually 12–14 IL (11: 2/90 occurrences; 12: 25/90; 13: 38/90; 14: 21/90; 15: 4/90); scales of the 1 st pair not longitudinally in contact, being divided into two scales; first two pairs in contact with anterior chinshields; 8–11 rows of smooth gular scales; throat shields very irregularly arranged.

In life, the coloration pattern of Trimeresurus puniceus is both quite variable and complex, difficult to describe accurately, and strongly related to the sex. This latter fact has been largely overlooked in the literature before Gumprecht & Tepedelen (1999).

In males, the body background colour is mid­grey, yellowish­grey, greyish­brown, greyish­green or even pinkish­grey, with 20–30 dorsolateral dark brown, reddish­brown, dark purple brown, dark greyish­brown or greenish­brown blotches, reaching downwards 2 nd or 3 rd DSR; the area between the is of a colour distinct(yellowish­brown, reddishbrown, and so on), usually darker than from the main body colour and heavily powdered with both irregular cream dots and minute dark grey or dark brown dots, giving an “ashy” or lichen­like appearance, or marked with small, irregular, dark blotches; the large dorsolateral blotches are much irregular in shape, often strongly constricted in their middle or angulous, distinctly lighter or with a different hue in their centre, darker on their edges, often with white or cream dots (“ashy” or lichen­like pattern), producing the “saddle­like” pattern depicted in Hoge & Romano (1974); the blotches may be separated or sometimes confluent on their upper part; blotches of both sides of body may be either fused or alternate, with all intermediates between perfect opposition and total alternation; often a series of small, dark, elongated ventrolateral blotches on the tips of ventral scales and the 1 st DSR, more or less irregular, making a very irregular ventrolateral stripe. The tail surface is usually of the same colour than the dorsolateral blotches, usually dark brown or dark reddish­brown, with pale greyish­brown or yellowish­brown rings, often edged with cream; the last third or so of the tail is entirely paler reddish­brown or yellowish­brown.

In females, the pattern is much less complex but also much less contrasted than in males, but with the same pattern of dark brown, dark yellowish­brown, dark pinkishbrown, dark greyish­brown or greenish­brown subrectangular blotches, both dark­edged and with a wide lighter centre, arranged on a mid­grey, yellowish­grey, yellow­ochre, fawn, dark greyish­brown or greyish green background; the dorsolateral, of same number than in males but subrectangular and not constricted in their middle, are less distinct than in males and the “ashy” pattern is lacking, making a much more neat pattern; however, many females are heavily dotted with numerous conspicuous, pure white small blotches.

As in males, the tail surface is usually dark brown, chocolate, greyish or pinkishbrown or dark reddish­brown, with pale greyish­brown or yellowish­brown rings, the last third or so being entirely reddish­brown or yellowish­brown. By their intricate “ashy” dorsal pattern, adult males are often very easy to distinguish from females, even more in the case of juvenile snakes (see below).

In both sexes, the dorsal head surface and temporal regions are of the same colours as the dorsolateral blotches, sometimes paler on the anterior supralabials and on the anterior part of the snout; in males, the sides of the head are often iregularly colored, with a mixing of different hues and heavily powdered with dark minute dots, giving an “ashy” or “dirty” pattern; lower limit of supralabials often marked with small irregular cream blotches; scattered pure white dots on the upper head surface in females heavily dotted in the similar way on their body; a more or less distinct, orange­brown, pale pinkish­ or reddish­brown postocular streak extending on upper temporals from the eye to the corner of the mouth in widening and darkening progressively before vanishing in general body colour; behind the eye, this streak may be very narrow or already quite broad. In males, the postocular streak is ofted edged with a more or less distinct dark streak on its both sides, often more dark and conspicuous below than above the postocular streak; sometimes a pale, poorly contrasted oblique streak from below the eye down to the posterior supralalabials.

Venter usually as upper body surface, much irregular in colour and hues, sometimes paler or much darker than body, as dark as dorsolateral blotches, with or without darker blotches; tips of ventrals with irregular elongated blotches of same colour than dorsolateral blotches, making a very irregular, broken ventrolateral stripe. The chin and throat are paler than upper head surface, with numerous black dots and irregular blotches. Eyes are golden or “tiger eye”, more or less dark. According to Gumprecht & Tepedelen (1999), the colour of the tongue is variable and is in relation with the coloration of both the head and eyes. Specimens with a dark upper head surface will have dark tongue and eyes. We have not observed ourselves this character.

In juveniles, the colour and pattern are basically identical, but background colour of females is lighter than in adults, most often in the hue of ochre, pale yellowish­brown or greyish­brown, with darker but rather subdued dorsolateral blotches, whereas males show the strongly contrasted pattern as described in adults, often even more conspicuous with beautiful mixing of grey and brown between dark brown or black dorsal blotches and conspicuous black and white dots. The head of males also show a much more contrasted pattern, with black, complex, irregular markings on upper head surface and head sides.

In alcohol, T. puniceus shows a wide range of brown, greyish­brown, and so on, producing the patterns described above, but much less conspicuous. If the scheme described alive is retained, differences between males and females are much less obvious. White dots and the ashy pattern become barely visible. Males just show a more complex pattern with irregular hues, whereas it is not rare to have females with a barely visible pattern, especially in darker specimens. We have observed only in a few cases the pattern shown in Hoge & Romano (1974: 158: Fig. 12 ). The dark red colour of the distal part of the tail turns to dark brown in preservative.

Hemipenes. The hemipenes are bilobate, rather short, massive, reaching 11–13 th SC, forking opposite 4–6 th SC, entirely covered with spines from the up to the tips, smooth in the vicinity of the sulcus bifurcation point.

Comparison with other species. The differences between Trimeresurus puniceus and the taxon named here Trimeresurus cf. puniceus are detailed under this latter account. The main differences between them are summarized in Tables 7–8.

The salient distinctive characters of Trimeresurus puniceus are (1) the projected but usually normal snout, (2) the small, raised supraoculars, (3) the 2 nd supralabials not bordering the loreal pit, and (4) the range restricted to Java and southern Sumatra.

Sexual dimorphism. It is weakly marked in the relative length of the tail and in the number of subcaudals, but strongly apparent in the pattern:

(1) Difference in the ratio TaL/TL:

males: 0.150 –0.171 (x = 0.158, s = 0.007); females: 0.128 –0.151 (x = 0.140, s = 0.007)

(2) Difference in the number of subcaudals:

45–59 (x = 51.7, s = 3.8) in males vs. 41–49 (x = 45.2, s = 2.3) in females.

(3) Difference in the number of scales separating the 5 th SL from the subocular:

in males: 2 scales in 31/46 occurrences, 3 in 15/46, vs 2 in 17/44 and 3 scales in 27/44 occurrences in females.

(4) Difference in the pattern, as explained above. Males are more brightly colored, with a more complex and contrasted pattern than females.

In our sample of 45 specimens, there is no difference in size and in the numbers of ventral scales nor in other scalation characters between males and females.

Range ( Fig. 4 ). INDONESIA. Java Island. Province of Jawa Barat (West Java) and Province of Jawa Tengah (Central Java): known from about 30 localities (see Hoge & Romano, 1974, and specimens listed above); East Java Province: only a few localities have been recorded in this province (Kediri, Nongkejajar, Puspo; see Hodges, 1993 and Hoge & Romano, 1994). Pulau Tinjil (Tindjil Island). Cited from this small island just south off the western tip of Java by Mertens (1957). Sumatra Island. As conceived here, Trimeresurus puniceus is known from: Jambi Province (no locality [ Orlov et al., 2002b]); Bengkulu Province (Barisan Range: near Bengkulu [ Gumprecht & Tepedelen, 1999; examined specimens], Prefecture of Redjanglebong); Sumatera Selatan Province (Lahat, Palembang and Pendopo [see David & Vogel, 1996 and examined specimens]).

Specimens from Sumatera Barat Province (Alahanpanjang [ Kopstein, 1938] and Singkarak [ Mertens, 1934]), which we did not examine, are tentatively and somewhat arbitrarily referred to Trimeresurus cf. puniceus along with the specimen from Mt. Kerinci (see below).

Biology. According to Hoesel (1959) and Kurniati (2003), this pitviper inhabits rainforests, mountain forests, and tea­ and coffee­estates. Orlov et al. (2002b) met this species in dense rainforests. Mertens (1957) mentioned that this species occurs at higher elevations in Java, but at sea level in Pulau Tinjil. According to Gumprecht & Tepedelen (1999), Orlov et al. (2002b) and Ryabov et al. (2002), in the Barisan Range this rather secretive species inhabits wet montane forests, bamboo forests, bushes and plantations between 700 and 1600 m. Orlov et al. (2002b) recorded specimens in Jambi Province between 1400 and 1600 m asl. Individuals were collected at 0.5 m above the ground in rainforests and at 1–2 m above the ground in coffee trees. Males seem predominantly terrestrial, whereas females are arboreal.

This species is ovoviviparous. According to Kopstein (1938), it may give birth up to 33 offsprings. A wild caught female produced 15 neonates each in a thin, transparent shell, from which the snakes hatched the next day. Orlov et al. (2002b) collected five pregnant females that gave birth to 7–13 babies. A detailed account of the reproduction and growth of juvenile was given in Ryabov et al. (2002).

Comments. Trimeresurus puniceus , as we conceive here, has been depicted alive in Hoesel (1959: 185, Fig. 128), and, more recently and in colour pictures in Gumprecht & Tepedelen (1999: 33, Fig. 16 and 34: Figs. 18–19 ), Orlov et al. (2002a: 195, Fig. 21 ; 2002b: Plate 5: Figs. A–B), Kurniati (2003: 127: Fig.98) and Gumprecht et al. (2004: 277,

Figs. I & III; 278: Fig: I; 279: Figs. I–II; 280: Figs. I–II; 281: Fig. I; 282: Figs. I, III–IV). Ink drawings of the dorsal pattern appeared in Hoge & Romano (1974: 156: Figs. 7–9 , 157: Fig.12 ).

Figures

FIGURE 1. Plot of first two principal canonical variables of the Principal Component Analysis on males of the Trimeresurus puniceus complex. The first two components explained 57.9 % of the variance. Six clusters are identified.

FIGURE 4. Map: distribution of the taxa recognized in the Trimeresurus puniceus­complex. Symbols: dots: Trimeresurus puniceus.—Squares: Trimeresurus cf. puniceus.—Triangles: Trimeresurus borneensis.—Inverted triangles: Trimeresurus wiroti.—Pentagons: Trimeresurus andalasensis.—Stars: Trimeresurus brongersmai.

FIGURE 5. Trimeresurus puniceus. Syntype, RMNH 1557. General view of body. Photograph by Roger Bour.

FIGURE 6. Trimeresurus puniceus. Syntype, RMNH 1557. Lateral view of head. Photograph by Roger Bour.

FIGURE 7. Trimeresurus puniceus. Living male, from Java. Photograph by Rick Hodges.

FIGURE 8. Trimeresurus puniceus. Living male, from Java. Lateral view of the head. Photograph by Rick Hodges.

FIGURE 9. Trimeresurus puniceus. Living female, from Java. Photograph by Gernot Vogel.

FIGURE 10. Trimeresurus puniceus. Living male, from Sumatra. Photograph by Albert Halimi.

FIGURE 11. Trimeresurus puniceus. Living female, from Sumatra. Photograph by Albert Halimi.

FIGURE 12. Trimeresurus cf. puniceus. Specimen from Anamba Island (USNM 26544). General view. Photograph by Gernot Vogel.

FIGURE 16. Trimeresurus borneensis. Syntype, ZMB 7146. Lateral view of head. Photograph by Gernot Vogel.

FIGURE 18. Trimeresurus borneensis. Young female from Poring Hot Spring. Photograph by Björn Lardner.

FIGURE 19. Trimeresurus wiroti. Holotype, SMF 69695. Photograph by Gernot Vogel.

FIGURE 21. Trimeresurus wiroti. Living female, from South Thailand. Photograph by Gernot Vogel.