Hypsirhina chinensis Gray, 1842 Zoological Miscellany, p. 66. Type locality: “China” holotype: BMNH 1918.104.22.168. Collector: J. R. Reeves.
Hypsirhina sinensis – Stanley, 1914 Journal of the North-China Branch Royal Asiatic Society, Shanghai, ns, 45:30 (in error for chinensis). Chang (1935, Peking Nat. Hist. Bull. 9:143) placed this name in the synonymy of chinensis.
Enhydris chinensis – Smith, 1923 Journal of the Natural History Society of Siam, 6:203.
Myrrophis chinensis – Kumar et al. 2012:484.
Etymology: This species was named chinensis because of its distribution in southern China. Common Names: Tang Water Snake or Chinese Rice Paddy Snake (Kuntz, 1963); Chinese Water Snake (Romer, 1961); “Pum-twa” [local name on Hainan Island] (Gressitt, 1940).
Distribution: Myrrophis chinensisinhabits Southern China inclusive of south central China’s East Hill and Plain, as well as the coastal plain of Fujian-Guangdong, and it occurs in adjacent Vietnam. It also occurs on the continental islands of Taiwan, Hong Kong, and Hainan. Devue (1970) has suggested it may be present in northern Laos. Steindachner (1867) reported chinensis from Singapore and Boulenger (1896), Flower (1899), and Wall (1903) examined specimens presumed to be from Bangkok or Siam (=Thailand). Maki (1931) hypothesized that the Bangkok specimens may have been imported from Formosa (=Taiwan) and obtained in a Bangkok medicine shop. Smith (1943) wrote, “I obtained specimens at sea in the Straits of Hainan …” References to this species from salt water environments are the result of its confusion with its close relative Myrrophis bennettii, which uses salt marsh and estuarine environments on China’s southern coast. Both of these species are similar in pattern, coloration, and scale counts. Myrrophis chinensis is primarily a lowland species; Mell (1929b) described it as common below 200 m. However, Pope (1929, 1935) found it ascends to considerable altitudes on the Chungan plateau and in northeastern Kiangsi, the former area reaches an elevation of 1500 m, this is not to say that chinensis is found at that elevation. Gressitt (1941) reported chinensis from the southwestern part of Fukien Province at Gang-keu (525 m); and southeastern Kiangsi Province at Hong San (825 m). Data accompanying specimens from North Vietnam suggest it reaches 1100 m.
Diagnosis: Scale rows at mid-body number 23, and these are reduced to 19 rows in front of the vent, although a few specimens that did not show this degree of reduction and had 21 rows in front of the vent. The 23 mid-body scale rows, an internasal that is not in contact with the loreal, a ventral count of 153 or less, and a pattern of mid dorsal blotches will distinguish it from Enhydris enhydris. However, this species has been frequently mistaken for its very close relative M. bennettii.
Size: Thirty-two specimens were measured for this study. The largest specimen measured was a female with a total length of 810 mm and a 100 mm tail. The largest male measured was 705 mm in total length with a 125 mm tail. The smallest specimen was 154 mm with a 35 mm tail. Size at birth is most likely in the 150 – 170 mm range for some populations, late term embryos have been reported being this size (Pope, 1929, 1935). However, Karsen et al. (1986) give a size range for neonates as 10 – 15 cm. My observations of late term embryos suggest this species gives birth to relatively small neonates within the range of 100-170 mm. Relative tail length is sexually dimorphic; and Gyi (1970) reported males had significantly longer tails. Combining all populations examined, males had a tail/SVL range of 15 – 19%, while females ranged from 12 -15%. At the base of the tail the width is 79% of the height, based upon an average from five specimens.
External Morphology: The head is slightly depressed, the body is cylindrical to slightly flattened. The eyes are dorsolateral and have a diameter equal to that of the width of the frontal. On the head the rostral scale is pentagonal, broader than high and tuberculate. The nasal is large, sometimes twice the size of the internasal, it may be divided or semi-divided with the cleft touching the internasal and or the first labial. The single internasal is small, about half the nasal, in contact with the loreal, and penetrates the nasals. The prefrontal scales are as large as, or slightly larger than, the internasal. The frontal is lance-shaped and wider than the supraocular, and the apex points posterior. The supraocular is narrow anteriorly, and it widens at the posterior edge; the preocular is single and taller than wide; and the postocular scales number two, with the bottom scale wider and usually shorter than the top scale. No subocular scales present, the fourth upper labial enters the orbit. The temporal scale formula is 1 + 2 + 3. Upper labials number seven, the largest is the sixth or seventh, the fourth enters the orbit. The first three or four labials have numerous tubercles; the labials that follow have fewer tubercles. Yin et al. (1996) reported this species to have two cornified tongue sheaths as opposed to one found in Xenochrophis (Colubridae: Natracinae); they considered this a primitive condition. On the chin, the lower labials number 6 – 9 (usually seven) the first four pair are in contact with the anterior chin shields and five or six are the largest. The anterior pair of chin shields is the longest; the second pair is relatively long, but narrow and separated by a pair of small scales. The gular scales number four to seven.
On the body, the dorsal scale rows on the neck are smooth and in 23 rows, two specimens (10.5%) of 19 had 25 rows on the neck. The first row is ovate and the scales become more lanceolate toward the midline. Dorsal scales at mid-body are similar and are in 23 rows. The dorsal scales at post-body are in 19 rows and they are more ovate than those on the anterior of the body. The type specimen [BMNH 1922.214.171.124] has 22 – 23 – 19 dorsal scales. Dorsal scales on the tail are ovate. The scales in the first dorsal scale row of this species are about 1.5 times the width of a scale on row 4 or 5. The ventral scales on the anterior of the body are 2.5 times the height of a nearby dorsal scale, those at mid-body are four times the height of a nearby dorsal and at posterior body they are about 3.5 times the height of a nearby dorsal scale. The ventral scales number 137 – 153. In a series of 13 specimens (eight males and five females) from Xi Qiao, China the females have 141 – 147 ventral scales, while the males have 150 – 153. In a series of 13 specimens (six males and seven females) from Taiwan the males have 137 – 147 ventral scales and the females have 137 – 147. Thus, at least one continental population appears to show sexual dimorphism in the ventral count while the Taiwan (island) population does not. Two males (CAS 74515, 144017) have four scales lateral to the anal plate that contain swollen glandular structures. The anal plate is divided and is about twice the length of the preceding ventral scale. On the tail the subcaudal scales are divided and number 34 – 50. In the Xi Qiao, China population males have 47 – 50 subcaudal scales (three specimens of five had damaged tails) and females have 37 – 43 subcaudal scales (in four specimens one specimen had a damaged tail). Similarly, in nine specimens from Vietnam (seven males and two females) the subcaudals were 41 – 42 in females and 42 – 52 in the males. However, in the Taiwan population males have 39 -46 subcaudal scales (in five specimens one specimen had a damaged tail) and females have 39 -46 subcaudal scales (in seven specimens). Thus, the subcaudal scale counts appear to be sexual dimorphic in mainland populations but not in the island population on Taiwan.
Color and pattern. The first row of dorsal scales usually has a pigmented spot that is similar to the pigment on the ventral scales. Scale rows 1 – 3, 2 – 3, or 2 – 4 have a cream stripe. The rest of the dorsum is more or less uniform brown or gray with irregular spots that are 1 – 3 scales in size. These spots may form indistinct transverse bars. Anteriorly the stripe on rows 2 – 3 may have a dark upper border. Seven of eight specimens have a mid-dorsal nape stripe, and there is usually is a continuous canthal to postocular stripe. The upper edge of the upper labials is darkly pigmented; the lower portion is lighter in color. The chin is mottled. Each ventral has a dark anterior edge as does each subcaudal. See Figure 27. Specimens examined from Taiwan, mainland China, and the high elevations (900 – 1100 m) of Vietnam show very little variation is scale counts or color pattern. Kuntz (1963) reported its skin oily to the touch. Preserved specimens of this and several other species do in fact seem oily to the touch.
Habitat: This is a lowland, highly aquatic snake that usually occurs between sea level and 200m, but reaches 900 – 1100m in North Vietnam. Pope (1935) said that it avoids true mountains but that it ascends to considerable altitudes on the Chungan plateau and in northeastern Kiangsi, China. It uses fish ponds, sluggish streams, canals, and rice paddies. Many literature accounts report this species associated with rice paddy ecosystems, and this agricultural practice may have contributed to the high abundance and relatively widespread altitudinal distribution of the species. Karsen (1986) noted that it may leave the water on rainy nights, but also reports it active in water during the day. Kuntz (1963) made similar observations and records its presence in suburban Taipei during flooding. Myrrophis chinensis enters the Indochina Bioregion uses aquatic habitats in the South China-Vietnam Subtropical Evergreen Forest ecoregion of Wikramanayake et al. (2002).
Diet and Feeding Behavior. Myrrophischinensis feeds primarily on fish and according to the literature may take amphibians on occasion. Mori (1998) found this species will swallow food underwater, behavior undoubtedly common to all homalopsids. Specific food habits reported in the literature include the following. Chang and Fang (1931) report a specimen with the fish Anabas scandens in its stomach. Pope (1929) wrote, “One stomach contained the remains of a small carp, one of a goldfish Carassius or a carp Cyprinus, one of a small fish Macropodus viridiauratus, and another an entire specimen of Erythroculter aokii.” Pope (1935) and Voris and Murphy (2002) have summarized these records. All of the fish species identified as prey from this snake are associated with rice cultivation and stagnant or slow moving water.
Reproduction. Seven females from Taiwan contained 3 – 32 embryos/eggs (X¯ = 16.3). The smallest SVL of a gravid female in this group was 396 mm, the largest 701 mm. However, another gravid female from Canton, China is 385 mm SVL. The RCM (n = 7) for the Taiwan specimens ranged from 0.12 – 0.55 (X¯ = 0.29). The effort per embryo ranges from 1.3 – 3.8 (X¯ = 2.2) for this population. Three specimens from mainland China had RCMs ranging from 0.15 – 0.28 (X¯ = 0.22). The range of effort per embryo for this population was 1.6 – 2.7 (X¯ = 2.1).
The literature suggests clutch sizes are usually smaller than the three largest described here. The largest litter previously reported in the literature was 18 (Smith, 1914b). Pope’s (1929) comment about females from higher latitudes giving birth to larger litters is also interesting. Parturition seems to occur during the summer months, although Smith (1914b) found a gravid female in December, and reports parturition in April, but does not state where the specimens originated. Literature data on reproduction in chinensis, suggests that this species has a very seasonal parturition, August and early September.
Meixi and Fuying (1988) examined the testes of 144 specimens from the suburbs of Fuzhou and found that spermatogenesis is also seasonal; they found the following. Acid phosphatase increased in September with the development of spermatozoa. The highest levels of acid phosphatsae in the seminiferous tubules were from October to December when spermatogenesis was vigorous. Alkaline phosphatase was most abundant from September to November, which is relative to the growth and differentiation of reproductive cells. The activity of 3-B hydroxysteroid dehydrogenase in the interstitial cells appeared as two peaks, one in October and November and the other in June, the latter being lower. And, they suggest this signals the formation of sexual hormones and sexual activities of the snakes. During the same periods of time (October-November and June) concentrations of lipids and steroids are low, possibly resulting from consumption of these substances during spermatogenesis and sexual hormone production.
Predators and Parasites: Like several other homalopsids, M. chinensisis occasionally used as food by humans (Angel, 1950; Irvine, 1954; Lau et al. 2000). Two authors have described its defensive behavior. Kuntz (1963) reports that when first brought into captivity M. chinensis is prone to strike upon the slightest irritation but it soon becomes more docile. When prodded into activity, it may make false strikes with lateral and/or vertical jerks of the head, accompanying quickly vertical jumps of the body. It is difficult to capture in the wild when disturbed since it has a tendency to disappear or submerge into the soft mud of the rice paddies….When handled, this snake is likely to initiate a profuse flow of semi fluid feces with a pungent odor. Schmidt (1927) also described defense and escape behavior that included biting, jumping, and striking, and flattening the body.
Most of the stomachs examined contained numerous nematode worms. Schmidt (1927) reported nematodes abundant in this species, and Fischthal and Kuntz (1975) described a trematode (Mesostephanoides taiwanensis) from this snake. The only other known species of worm in this genus parasitizes Enhydris enhydris.
Diving and Breathing: Few observations have been made on submergence and diving in homalopsids. Schmidt (1927) kept a large female M. chinensis in a “water jar” and observed the frequency of respiration. Seven intervals between emergence and submergence ranged from 10 – 15 minutes, averaging between 13 – 14 minutes. Breathing at the surface was 1 – 5 minutes, with only the tip of the snout being exposed.
Venom: Envenomation by this species is said to cause headache, nausea, and pain (Karsen et al., 1986). The LD50 for this species has been measured at 2.05 mg/kg of body weight when injected intravenously. This is about seven times more toxic than Homalopsis buccata venom and 3.5 times more toxic than Subsessor bocourtivenom. The LD50 is comparable to the toxicity of venom from the carpet viper, Echis carinatus (Minton, 1996; Sakai et al., 1984).