Bennett’s Mud Snake, Myrrophis bennettii (Gray, 1842)

Hypsirhina bennettii Gray, 1842 Zoological Miscellany, p. 67. Type locality: “China.” Holotype:  BMNH 1946.1.2.49 [This specimen is probably a male, it has a damaged tail.] Collector: unknown.
Hypsirhina maculata  Duméril, Bibron and Duméril, 1854Érpétologie générale…Reptiles, Paris 7(2):950 type locality: “China.” Holotype: MNHN c3453(2). Collector: unknown.
Hypsirhina enhydris var. maculata 1868 Jan and Sordelli, Iconographie Generale des Ophidiens, 30, pl 4, Figure 1.
Enhydris bennetti – Stejneger, 1907 Bulletin of the United States National Museum 58:302.
Myrrophis bennetti – Kumar et al. 2012:484.
Etymology: This snake was named in honor of George Bennett, the 19th-century Australian herpetologist who provided the type specimen to the BMNH. Common Names: Bennett’s Hypsirhina (Gray, 1849); Bennett’s Water Snake, Mangrove Water Snake (Karsten et al., 1986); Bennett’s Water Snake (Romer, 1970).
Distribution: This snake is endemic to the south coast of China from Hong Kong south and west to Hainan. Gyi (1970) reported this species from Java on the basis of a ZMA specimen. This is most certainly an error. Sclater suggested it may be present in Formosa (=Taiwan) based upon a specimen in the Indian Museum; Stejneger (1907) stated that no specimen from Formosa were accessible to him;  Maki (1931) considered it absent from Formosa, but notes an Indian Museum specimen (12692) was received from Hong Kong with Formosa locality data. Romer (1979) wrote about the Hong Kong population, “…it has now been established that this little known snake is by no means uncommon in the tidal marshes (near Pak Hok Chau) along the shore at Deep Bay. Specimens have also been taken at the mouth of the Yuen Long Creek (unofficial name) and by the shore on the Mong Tseng peninsula at Deep Bay.” 
Diagnosis:  This snake has 21 scale rows at mid-body, that are reduced to 17 or 15 in front of the vent, 153 – 164 ventral scales, and an internasal scale that does not contact the loreal. It is most often confused with M. chinensis which has 23 rows of scales at mid-body, 19 rows near the vent, and 153 ventral scales or fewer. It may also be confused with E. enhydris which has 21 scale rows at mid-body, and it has an internasal that contacts the loreal scale. E. enhydris, like bennettii, also has more than 153 ventral scales.
Size: The largest individual measured was a male with a total length of 680 mm, and a 120 mm tail. The largest female measured had a total length of 659 mm with a 96 mm damaged tail. The smallest individual had a 261 mm SVL with a 53 mm tail. Two males from Hong Kong have tails that are 21 and 22% of the SVL; two females from Hainan have tails that are 21 and 24% of the SVL.
External Morphology: The head is distinct from the neck, it is long and depressed. The dorsolateral eyes have a diameter that is slightly more than the eye-nostril distance. On the head the rostral scale is pentagonal and about twice as wide as its height. It is barely visible from above. The nasals are semi-divided with the nasal cleft touching the first labial; at least one specimen has an entire nasal scale without a nasal cleft. The nasals are in contact and the internasal scale intrudes between them on the seam. The internasal is single and less than half the area of a nasal. The prefrontal scales are in contact with the loreal, and each one is larger than the internasal, they also contact the preocular and supraocular. The frontal is almost equal to the interorbital distance. The parietals are about as long as the frontal. The supraocular scale is rectangular with rounded edges, the single preocular has a dorsal edge that is pointed, of the two postocular scales, and the upper scale is more elongated, about twice as high as the lower scale. The temporal formula is 1 + 2 + 2. However, one specimen from Hong Kong has a temporal formula of 1 + 1. Upper labial scales number 7 – 9 (eight is most frequent). The loreal touches the first three upper labial scales. The largest upper labial is six or seven, and the fourth (rarely the fourth and fifth enter the orbit). On the chin the lower labials number 8 – 12, the first four or five lower labials contact the anterior chin shields. The anterior chin shields are the largest. This species has individuals with 1 – 3 pairs of chin shield. At least two specimens have an asymmetrical number of enlarged chin shields (two on one side, three on the other). The type specimen (BMNH 1946.1.2.49) has three pairs of chin shields, the anterior pair are the largest. Gular scales number 4-8. On the body the dorsal scale rows number 21 on the neck and mid-body, scale rows at near the vent are reduced to 17. The scales tend to be ovate and become more so posteriorly, they are smooth and lack ornamentation. Some scales may have tubercles in the anal region.  All scales in the dorsal rows are about equal in size. Ventral scales number 153 – 164 in three females and 158 – 162 in three males. One specimen from Deep Bay, Hong Kong (BMNH 1983.255) has a very large number of divided ventral scales. The ventral scales are wide, at least 3 – 4 times the height of a nearby dorsal scale. The anal plate is divided and 1.5 – 2.0 times the length of a ventral scale. On the tail subcaudal scales are divided and number 47 – 52 in three females (Gyi reported 45 in one female). Three male specimens have 53 – 55 subcaudal scales. The tail is laterally compressed. The limited sample suggests tail length/SVL ratios in this species are not sexually dimorphic, but that subcaudal counts may be.  At the base of the tail the width is 75% of the height.
Color and pattern. The upper surface of the head is gray with some trace of the nape stripe extending onto the crown. The lower three-fourths of the upper labials are yellow. Some labials with yellow centers are outlined in black. The scales of the first dorsal row are divided into a dark anterior and a yellow posterior. Scale rows 1 – 3 or 1 – 4 on the neck and scale rows 2 – 3 at mid-body contain a yellow or white stripe. The rest of the dorsal rows are gray with black spots that sometimes fuse to form a black bar. These number 35 – 40 and may become fused near the tail to form a longitudinal stripe. There is a stripe at the outer edge of each ventral scale as well as a dark anterior and a yellow posterior, and a dark spot on the midline of each ventral scale. On the ventral surface of the tail these form a dark stripe as the yellow is reduced. In terms of morphology and color pattern this species is remarkably similar to M. chinensis.

Habitat: This species inhabits saltwater marshes and estuaries. It apparently climbs into shrubs on occasion, a very unusual behavior for a homalopsid. It is unclear if this behavior is an attempt at basking, or if the snakes were using branches at high tide and were left on a branch as the tide receded. Direct comments from the literature follow. Karsten et al. (1986) wrote, “…common in the tidal marshes and in and around the mangroves of Deep Bay….both salt and brackish water. Is somewhat diurnal but may be nocturnal…Is a good climber and spends much time resting on trees.” Pope (1935) reported, “Dr. Malcolm Smith has told me that his bennettii were taken in the sea along with sea-snakes near the coast of Hainan. This probably means that bennettii inhabits the low coastal plain of Hainan and is frequently washed out to see by freshets… According to Dr. Smith, this is not really a rare snake on Hainan.” Romer (1961) described finding it in a small stream at a point where it discharged into the sea.