Bocourt’s Mud Snake, Subsessor bocourti (Jan, 1865)



Hypsirhina bocourti Jan, 1865 Archiv per la Zoologia, l’Anatomia e la Fisiologia. 3:258.  Type locality: Bangkok, Thailand. Holotype: is at the MNMH according to Gyi. Collector: unknown.

Ferania sieboldi (non Schlegel) Günther, 1866 Annals and Magazine of Natural History, ser. 3, 18:28.
Hypsirhina gigantea Werner, 1923a Annalen des Naturhistorischen Museum in Wein, 36:163. Type locality: unknown. Collector: unknown. Holotype: Vienna Museum.
Enhydris bocourti – Smith, 1930 Bulletin of the Raffles Museum 3:61.
Hypsirhina bocourti soctrangensis Bourret, 1936 Serpentes de I’Indochine, 1:291. Type locality: Soc Trang, Vietnam. Holotype: MNHN 38144. Collector: unknown.
Subsessor bocourti Murphy and Voris 2014: 34.
Eytmology: This snake was named to honor the 18th-century French herpetologist Marie-Firmin Bocourt. Common Names: Bocourt’s Water Snake (Smith, 1943); Ri voi snake (Thu, 2001) Vietnam
Distribution: This snake is endemic to the Indochinese peninsula, including Thailand, Malaysia, Cambodia, and Vietnam. Deuve (1970) reported it from the river side towns of Ngou Pa and Ngou Leuam  in Vientiane Province, Laos. It is unclear if this is the northern edge of the distribution, an isolated population, or possibly a feral population. I have not yet examined specimens from this area. Theobald (1868) has reported it from India and Burma (Myanmar). Hundley (1964) also reports it from Burma possibly based upon Theobald’s record. These localities are most likely in error based upon confusion with specimens of Enhydris sieboldi. The confusion may extend to erroneous statements made by Günther (1866; 1868) that the illustration labeled sieboldii in Jan and Sordeli (1860 – 1881) was actually bocourti. Specimens documenting its presence west of central Thailand have not been found, but it does occur on the southern edge of Thailand’s Khorat Basin and is likely to be more widespread in the Mekong drainage than current data suggests. It may eventually be shown to be present in southern Myanmar since it is known from southern Thailand. Its distribution forms an arc around the Gulf of Thailand, a distribution similar to that shown by the snakes of the Enhydris jagorii Group and Erpeton tentaculatus. Recent reports of this species from China (Li, 1998) are most likely based upon its presence in markets or feral animals; S. bocourti is imported into China from Cambodia and Vietnam and if Chinese populations are found they are most likely the result of escaped market animals or intentional release.
Diagnosis: A large, robust homalopsid with 27 scale rows at mid-body; divided upper labials behind the eye; 135 ventral scales, or fewer; and an internasal that is single, and may make narrow contact with the loreal scales.
Size: The largest specimen measured was a female from Thale Noi, in southern Thailand, with a total length of 1,230 mm, and a 150 mm tail; this individual was gravid and weighed 2.8 kg. The largest male measured had a total length of 770 mm, with a 124 mm tail. The smallest individual measured was 154 mm in total length with a 27 mm tail; this is probably near the low end size at birth, and this species as well as Homalopsis buccata probably give birth to the largest homalopsid neonates. Gyi (1970) was confused when he said this was the longest homalopsid, and cited USNM 70302 as the largest specimen at 1,375 mm. USNM 70302 is a Homalopsis buccata, collected by H. Smith and discussed and illustrated by Cochran (1927; 1930). While bocourtiis not currently known to be the longest homalopsid, it may the one with the greatest bulk. Smith (1914) reported a 1,120 mm individual. Saint Girons’ (1972) largest specimen was a 950 mm female; his largest male was 812 mm total length, he also described a sexually mature female that was 590 mm. Werner (1923a) described one that had a total length of 1075 mm and a 125 mm tail; this is the type specimen of Hypsirhina gigantea.
External Morphology: The head is short and slightly distinguishable from the neck. The eyes are dorsolateral. The body is extremely thick in girth and slightly compressed. On the head the rostral scale is pentagonal, is as broad as tall, and two notches are barely discernable on the ventral edge. The nasal is semi-divided or divided with the nasal cleft touching the first labial or the first labial and internasal. The internasal is usually single, but divided in one specimen, and smaller than the nasal. The internasal is behind and slightly penetrating the nasals. The prefrontals are in contact with the loreal scales. The frontal is elongated and pentagonal or bullet-shaped.  The parietals are about as long as the frontal and they have light or dark spots or blotches. The loreal is single and is in contact with 1 – 2 or 1 – 3 upper labials. The single supraocular is narrow anteriorly and widened posteriorly. On two specimens there is a fragment of this scale on the posterior ventral edge. The preocular is usually single, but subdivided in one specimen, and there are two postoculars and no subocular scales. Temporal formula is 1 + 2 + 4. Upper labials number eight (rarely nine), the fourth is in contact with the orbit and some of the upper labials posterior to the eye are horizontally divided. The largest upper labial is six or seven, and there are very small tubercles on the labials. On the chin, lower labials number 11 – 14, with the first labial being the largest. The anterior pair of chin shields is 3 – 4 times longer than the second pair. The anterior chin shields are in contact with the first five (or four) lower labials. The second pair of chin shields is separated by a pair of smaller scales. Nine or 10 gular scales separate the chin shields from the most anterior ventral scale. On the body the dorsal scales are in 27 – 29 rows on the neck, these are smooth and ovate and become only slightly lanceolate on the midline; three specimens from peninsular Malaysia have 29 scale rows on the neck, while Cambodian and Vietnam specimens tend to have 26 – 28. Dorsal scale rows at mid-body number 27; they are smooth and or slightly textured, and they are mostly ovate, becoming only slightly elongate toward the midline. The dorsal scales at posterior body are similar and are in 21 – 25 rows. The ventral scales at the anterior of the body are narrow, and about three times the height of a nearby dorsal. The ventrals at mid-body are about 3.5 times the height of a nearby dorsal, and those at posterior body are about three times the height of a nearby dorsal. Ventral scales number 123 – 135 (129 – 133 in males, 123 – 135 in females). The anal plate is divided. On the tail dorsal scales are ovate, smooth, but slightly textured. The subcaudal scales are divided and number 32 – 47 (41 – 47 in males, 32 – 37 in females). At the base of the tail the width is 81% of the height, based on the average of five specimens. Color and pattern. The dorsum is dark brown with 30 – 40 transverse crossbars that are one scale wide; one specimen (MCZ 5969) lacks the crossbars. The crown of the head is dark brown with some dark or light spots on the head shields. There is a post-parietal transverse bar. The upper and lower labials are yellow with some dark seams. The venter is yellow and crossed with 25 bars of dark pigment that are connected to the dark dorsum. Yellow blotches on the sides go to scale rows six or seven and are 4 – 5 scale rows wide. The tail is dark brown with yellow spots on top and below. Boulenger’s (1903) commented on the juvenile coloration, “The superior brilliance of coloration of young specimens is remarkable.” Sexual dimorphism. Tail/SVL ratios for males were 17 – 23%, and for females 13.4 – 16.2%. Males have a slightly constricted area at the base of the tail. Males have 41 – 46 subcaudal scales, while females have 35 – 40 subcaudal scales. Thus both relative tail length and the number of subcaudal scales are sexually dimorphic in this species.
Habitat: This highly aquatic snake uses swamps, shallow lakes, pools, and other stagnant water habitats. This snake is often found with Acrochordus javanicus in swamps and pools; and it may move into deep water during the hottest part of the year (Laidlaw; 1901; Boulenger, 1903).  Campden-Main (1970) stated that it is almost entirely aquatic. Deuve (1970) reported on bocourti in Laos, it was collected in June and November from fish traps; it fed on fish; and was mostly nocturnal. Smith (1943) found “It is not uncommon in the low lying country in the vicinity of Bangkok and in Cochin China.” And, Taylor (1965) reported two specimens from Pattani, Thailand that were obtained from “a pool in that city.” During this study specimens were obtained from reservoirs, rice paddies, and shallow pools in wetland habitats in Thailand. This species also makes up a portion of the Tonlé Sap snake harvet in Cambodia.  Subsessor bocourti inhabits the Indochina Bioregion of Wikramanayake et al. (2002), and extends into peninsular Malaysia (part of the Sunda Shelf and Philippines Bioregion).
Diet and Feeding Behavior: A specimen (FMNH 259282) from Tonlé Sap, Cambodia (male, SVL 49 cm, TL 60.5cm) contained a 10.5 cm catfish, Mystus mysticetus (Siluriformes: Bagridae). A neonate (SVL 210 mm) obtained near Kabin Buri, Thailand contained a 145 mm freshwater eel (Monopterus albus, Synbranchidae). Several other specimens from the same locality contained unidentified fish remains. Flower (1899) reported that captive newborns “…fed readily on small frogs (Ranaand Microhyla).” Saint Girons (1972) reported five fish in the stomach of one specimen.
Reproduction: Flower (1899) wrote, “…While in captivity brought forth seventeen young, alive. They were expelled at intervals of from two to twenty minutes…” He stated that the neonates were “…about 220 mm in length.”  The female was 845 mm total length. Saint Girons and Pfeffer (1972) consider this species to have a “spring hatching” in Cambodia. Saint Girons (1972) reported 6 – 17 eggs in eleven females from Cambodia. A female obtained at Thale Noi, Thailand, had a total length of 1,230 mm and a 150 mm tail, and weighed 2,880 g. It contained 26 embryos, 15 in the right oviduct and one infertile egg, and 11 in the left oviduct with three infertile eggs. The clutch weighed 667 g.  There was one set of twins in the litter. The RCM for this litter was 0.23. A second gravid female (SVL = 910 mm, total length 1030 mm, 1.25 kg) from the Kabin Buri, Thailand area contained 28 near full term embryos on 22 June. The mean SVL of the embryos was 17.8 cm (r = 15.4 – 19.0) and the mean mass was 9.3 g (r = 5.5 – 10.8) (Karns et al., 2005).
Predators and Associated Behavior: Campden-Main (1970) stated that the Vietnamese sell bocourti for food around Can Tho. Thu (2001) recognized that this species was important to local people and economies. In Cambodia, (Stuart et al., 2000; Brooks et al. 2007) bocourtiwas part of the Tonlé Sap snake harvest although it represented a low percentage of the harvest and it was used for human consumption and its skins were used in the leather industry. This species was selectively removed from the catch and sold for export to Vietnam, and presumably re-export to China. Live specimens sold for the equivalent of US $5.21 – 6.51 per kilogram. While dead specimens sold for US$2.61. The live snakes were more valuable because they could be exported. Lau (in Stuart et al., 2000) reported bocourticommon in the wildlife markets of Guangzhou, China, an area outside the distribution of bocourti. And, photographs in Lao (1997) suggest that the Chinese may be culturing this species.  In southern Viet Nam these snakes are also exploited, and Stuart (2004) reported that they are the second most abundant reptile species in trade. It sells for US $10.99/kg, and people in the area raise them in cages until they attain a weight of about 2 kg. In Cambodia, Brooks et al. (2007) report the species sells for US$2.50 to 5.00 per kg.
In Laos, Deuve (1970) found them to be very irritable, and he noted they bite readily.  His samples showed wounds which he attributed to fights, scars were numerous on the ventral scales; these are more likely due to attacks by predators than conspecifics. Smith (1943) commented that “Its temper is uncertain, and its large size enables it to inflict a very serious bite if handled carelessly.” My limited experience with this species suggests that it is docile when handled gently and shows no inclination toward biting.
Population and Abundance: Smith (1914b) considered bocourti rare in Bangkok, but common in the rural areas surrounding it. Tweedie (1950) stated that bocourti is poorly known from the Malayan Peninsula. At the time of writing Tweedie knew of only three specimens, Flower’s specimens, one collected from Trengganu in 1947, and a photo of one from Wellesley Province taken in 1947. My experience with this snake suggests that it is relatively abundant in some areas (Kabinburi, Thailand and Tonlé Sap, Cambodia) but nowhere did it dominate the aquatic snake assembly like E. enhydris or E. plumbea.
Venom: The LD50 for the venom of this species was found to be 7.3 mg/kg of body weight (Sakai et al., 1984; Minton, 1996).