Nandus prolixus, Prosanta Chakrabarty, Ronald G. Oldfield & Heok Hee Ng, 2006

Prosanta Chakrabarty, Ronald G. Oldfield & Heok Hee Ng, 2006, Nandus prolixus, a new species of leaf fish from northeastern Borneo (Teleostei: Perciformes: Nandidae)., Zootaxa 1328, pp. 51-61 : 52-59

Nandus prolixus sp. nov.

(Fig. 2)

Nandus nebulosus (non Gray) Inger & Chin, 1962: 164, Fig. 85.

Type material. Holotype. FMNH 44907 , 83.4 mm SL; Borneo: Sabah, Sandakan District, Km 26, North Road, Sandakan ; A.M. Anderson, 3 August, 1950.

Paratypes. FMNH 51964 (6), 47.0-81.3 mm SL; Borneo: Sabah, 26 km NW of Sandakan ; R.F. Inger, 8 August, 1950. FMNH 117232 (2), 71.9-76.8 mm SL; data as for holotype.

Diagnosis. Nandus prolixus is distinguished from its only Sundaic southeastern Asian congener, N. nebulosus , in having a longer, more produced snout (25.7-30.6% HL vs. 18.5-26.1; Fig. 3), more lateral-line scales (33-37 vs. 24-34), more scales below the lateral line (12 vs. 10-11), fewer spines in the dorsal fin (XIV vs. XV -XVI), and fewer pectoral fin rays (15-16 vs. 17-19). It differs from N. nandus (from India) in having fewer lateral-line scales (33-37 vs. 42-55), fewer scales above the lateral line (4-5 vs. 6-7), fewer scales below the lateral line (12 vs. 14-18), a greater number of dorsal spines (XIV vs. XII -XIII), and the absence (vs. presence) of a distinct dark spot at the base of the caudal peduncle. Nandus prolixus differs from N. oxyrhynchus (from mainland southeastern Asia) in having a more slender body (body depth 37.6-40.5% SL vs. 41.3-44.1) and a less steeply sloping predorsal profile (Fig. 3).

Description. Morphometric data as in Table 1; meristic data as in Table 2. Body compressed, moderately elongate; dorsal profile evenly sloping, with noticeable concavity in interorbital region. Snout profile acute. Mouth moderately large, protrusible. Posterior end of maxilla extending just beyond vertical through middle of orbit. Eye large, diameter about one third of head length, circular. Posterior edge of preopercle with fine serrations. Gill rakers short and club-shaped, bearing sharp apical denticles. Teeth short, unicuspid, closely set and in many irregular rows on both upper and lower jaw.

Lateral line divided into two segments, with anterior segment more dorsally located than posterior segment. Upper lateral line beginning at dorsal origin of operculum, rising for distance equivalent to two or three scales rows and reaching greatest height above pectoral fin, thereafter sloping ventrally and ending at vertical through middle of anal-fin base. Lower lateral line beginning at vertical through middle of anal-fin base, vertically centered along length of caudal peduncle, and continuing slightly past end of hypural plate.

Scales ctenoid, imbricate, and nearly uniform in size throughout body. Scales present throughout cheek region, preopercle, opercle, and area around eye, absent along midline of interorbital region. Area around nares and upper lip scaleless. Gular region and ventral region of head immediately adjacent scaleless; sensory pores present in this region. Sheath of scales surrounding proximal regions of dorsal and anal fins, forming ridges along sides of hard rays and attached to soft-fin rays. Caudal fin scaled for about one quarter of length.

Dorsal fin with long base, anterior insertion at vertical through posteriormost extent of opercle and posterior insertion at vertical through base of last anal-fin ray. Length of longest dorsal-fin ray reaching slightly beyond vertical through origin of caudal fin. Pectoral-fin insertion anterior to pelvic-fin insertion; pectoral fin shorter than pelvic fin. Pelvic fins reaching urogenital opening, but not reaching origin of anal fin. Longest ray of anal fin reaching vertical through, and rarely beyond, base of caudal fin.

Coloration. In 70% ethanol: Light brown color on body, with mottled darker areas randomly distributed over body, but never forming distinct vertical bars. Two dark stripes running from eye: one dorsoposteriorly towards dorsal origin of operculum, and second running posteroventrally and passing under posterior edge of maxilla. All fins except pectoral fins with series of small brown spots forming irregular transverse bars across fin membranes.

Distribution. Nandus prolixus is known only from the Sepilok River drainage in northeastern Borneo (Fig. 4). Inger & Chin (1962) also recorded Nandus from the Kinabatangan River drainage to the south, and we surmise that this record may refer to N. prolixus . However, we were unable to examine this material to confirm its identity.

Habitat and biology. Inger & Chin (1962) collected specimens of N. prolixus in “very slow moving water of shallow streams in swampy forested areas…ten were collected hiding in dead leaves that covered the bottom.” The area where the fish were collected consisted of: “Short vegetation primary dipterocarp forest. Roots of trees but no other living submerged or emergent vascular plants. Banks steep; 1-2 meters high. Bottom mud with dead leaves and other plant fragments.” Examination by Inger & Chin (1962) of gut contents of three specimens recovered various insect larvae and a single “unidentifiable fish”. These specimens were captured with Nematabramis everetti (Cyprinidae), Leptobarbus melanotaenia (Cyprinidae), Systomus sealei (Cyprinidae), S. binotatus (Cyprinidae), Hampala macrolepidota (Cyprinidae), Cyclocheilichthys repasson (Cyprinidae), Osteochilus microcephalus (Cyprinidae), Nemacheilus olivaceus (Balitoridae), Pangio mariarum (Cobitidae), Acantopsis choirorhynchus (Cobitidae), Ompok sabanus (Siluridae), Hemibagrus nemurus (Bagridae), Clarias leiacanthus (Clariidae), Dermogenys pusillus (Hemiramphidae), and Channa melasoma (Channidae).

Etymology. The specific epithet comes from the Latin prolixus, meaning stretched out, in reference to the relatively elongate head of this species when compared to its Sundaic congener ( N. nebulosus ). Used as an adjective.

Discussion

The four discrete shape groups recovered in the PCA analysis supports the hypothesis that there are four Nandus species. The Principal Component Analysis reveals that each Nandus species groups separately on a plot of Principal component 1 versus PC2 (Fig. 5). Principal component 1 explains 29% of the variation among specimens; PC2 explains 22%, and PC3 explains 11%. Despite forming a discrete group on PC1 vs. PC2, Nandus prolixus overlaps with N. oxyrhynchus and N.nandus on the PC1 and the PC2 axis. There is no overlap between Nandus prolixus and N. nebulosus on either PC1 or PC2. Nandus prolixus also forms a distinct group separate from the other Nandus species on plots of PC2 vs. PC3, and PC1 vs. PC3 (not shown). Principal component 1 explains much of the variation in mouth size, eye size and body length among specimens. Principle component 2 explains much of the variation in body depth, caudal peduncle depth, preopercular height, and the slope of the head among specimens.

Given the proximity of the type locality of Nandus borneensis Steindachner, 1901 (considered a junior subjective synonym of N. nebulosus by Ng et al., 1996) to that of N. prolixus (both are in northern Borneo), a further discussion of the distinctiveness of the two species is necessary. Ng et al. (1996) considered N. borneensis and N. nandus as conspecific because, apart from a slightly lower number of total lateral line scales (24-30 vs. 29-34) in near-topotypic material (from the Baram River) of N. borneensis , no other distinct differences could be found between populations ascribed to the two nominal species. In any case, the total number of lateral line scales in N. prolixus (33-37) is always higher than that reported in either the original description of N. borneensis (27-29), or in near-topotypic material (24-30). We therefore follow Ng et al. (1996) in treating N. borneensis and N. nebulosus as conspecific.

The freshwater ichthyofauna of northeastern Borneo (Sabah) is different from those of neighboring areas on the Sunda Shelf, with a considerable number of species endemic to this region (Inger & Chin, 1962). It is hypothesized that this high level of endemism is most likely the result of vicariant speciation due to the isolation of the river drainages in northeastern Borneo from the other major southeast Asian river drainages (Ng, 2004). This isolation may have occurred sometime during the late Oligocene (ca. 25 million years ago) as a result of the orogenesis of the central Bornean highlands, which formed a regional drainage divide between drainages to the north and east and those to the south and west (Hall, 1998).