Osteocephalus vilarsi ( Melin, 1941 )

Jungfer, Karl-Heinz, 2010, The taxonomic status of some spiny-backed treefrogs, genus Osteocephalus (Amphibia: Anura: Hylidae), Zootaxa 2407, pp. 28-50 : 41-46

Osteocephalus vilarsi ( Melin, 1941) nov. comb.

( Fig. 7 )

Hyla (Trachycephalus) vilarsi Melin, 1941

Osteocephalus taurinus – Bokermann 1966 (synonymized vilarsi with taurinus ) Osteocephalus leprieurii – Cochran and Goin 1970 (synonymized vilarsi with leprieurii ) Osteocephalus taurinus – Trueb and Duellman 1971

Osteocephalus taurinus – Frost 2009

Hyla vilarsi (GNM 488) was described by Melin (1941) from a single female that he had obtained from indigenous people at Taracuá (on modern maps also Taracua or Missão Taraquá; for information on the type locality see Caldwell et al. 2002), about 100 m a.s.l., Rio Uaupés, in the upper Rio Negro drainage of Estado Amazonas, Brazil, on 7 April 1924. The holotype is a female of 62.2 mm snout-to-vent length (SVL) that contains eggs and therefore must be an adult. A striking feature are two distinct, almost parallel longitudinal frontoparietal ridges on the head, a character found in a few species of Osteocephalus .

Bokermann (1966) listed the taxon without explanation as a synonym of Osteocephalus taurinus Steindachner, 1862 . Cochran and Goin (1970) considered vilarsi and Osteocephalus planiceps Cope, 1874 , to be synonymous with Osteocephalus leprieurii (Duméril and Bibron, 1841) . Trueb and Duellman (1971) included vilarsi and O. planiceps , among others, in the synonymy of O. taurinus , a widespread species ranging from the Guyanas and southern Venezuela throughout the Amazon Basin southward to Bolivia. Duellman and Mendelson (1995) resurrected O. planiceps as a valid species. The status of vilarsi was not addressed by them. Thus, according to those previous authors, the taxon should either be a synonym of O. leprieurii , O. planiceps , or O. taurinus .

Redescription of the holotype: Melin (1941) described the species in some detail. This is expanded here in order to facilitate the comparison between species, for while he mentioned the similarity between vilarsi and O. taurinus , he did not explicitly state the differences between them.

Adult female of 62.2 mm SVL. Snout bluntly rounded in dorsal and lateral view. Head slightly longer than wide. Lateral edges of frontoparietals raised, forming two almost parallel ridges. Canthus rostralis very distinct, almost straight (very slightly curved inwardly). Loreal region deeply concave. Nostrils elevated, opening laterally. Dentigerous processes of vomers in contact with each other, bearing 11 teeth on the left and 15 on the right, angular, in between choanae, their anterior edges at about midlevel of the choanae, their posterior edges reaching slightly beyond the posterior edges of the choanae. The latter are large, oblique and bean-shaped though somewhat angular. Tongue elliptical, about 1.4 times longer than wide. Supratympanic fold from the anterior edge of the tympanum sloping towards the upper edge of the arm insertion in an almost straight line. Tympanum large, conspicuous, elliptical, slightly wider than high and about 77 % of the eye diameter.

Skin on dorsum smooth with a few scattered, small, flat tubercles (under magnification) in the posterior half of the back. Dorsal surfaces of head rugose. Loreal region granulate. Skin loose or missing (removed) above the sphenethmoid and on the right nostril. Skin of body smooth laterally and on lateral and dorsal surfaces of the extremities. Ventrally smooth in thoracic area and on chin, granulate on belly and posterior two thirds of the thighs. Axillary membrane absent. Cloacal opening at about three fourths of thigh height.

The finger and toe discs are desiccated. Melin described them as being scarcely half the size of the tympanum and slightly oblong, but it is not clear whether they were in a better state then. The disc of Finger III is 52 % of the tympanum width. The thumb bears a large elliptical thenar tubercle and there is a prominent supernumerary tubercle proximal to the proximal subarticular tubercle. A few low tubercles are present on the outer edge of the fourth finger. The distal subarticular tubercle on Finger IV is barely bifid, the others are simple. The relative length of the adpressed fingers is I <II <IV <III. The webbing formula is I basal II 2 — 3 + III 2 ¾ — 2 ¾ IV.

On the foot there is a large elliptical inner metatarsal tubercle and a small rounded median plantar tubercle. The subarticular tubercles are single and conical. The proximal segments of Toes III-V bear numerous supernumerary tubercles. The relative length of the adpressed toes is I <II <III = V <IV. The webbing formula is I 1 ¼ — 2 ¼ II 1 + — 2 2/3 III 1 + — 2 ¾ IV 2 ¾ — 1 ¼ V.

Measurements of the holotype (in mm): SVL 62.2; HL 20.1; HW 19.6; TL 34.5; FL 42.0; ED 6.2; TD 4.8; FD 2.3; EN 6.2; IN 4.6; TE 2.5.

Coloration of the holotype: Dorsum uniform tan. Broad dark tan bars on the limbs, two on lower arms, one on hands, two on thighs, one across the knee, two on the tibia, two on the tarsus. A dark tan horizontal supracloacal stripe. A few indistinct dark tan spots laterally. A thin white supralabial line from the posterior edge of the tympanum around the snout, greatly expanded as a subocular spot from anterior edge of eye to midlevel of tympanum on the left, from posterior third of eye level almost to posterior edge of tympanum (reaching eye on lower eyelid) on the right. Throat creamy yellowish with some small tan mental spots. Lower jaw bordered by a thin tan sublabial line. Belly creamy yellowish. A few small tan spots bordering the chest posterior to the clavicles. Arms and legs ventrally light brownish white. Posterior surfaces of thighs uniform tan. Iris color not clearly visible (there are a few black spots on grey ground, straight radiating lines are not visible). The bones, which have been in preservative for 85 years, are white, but may have been green as in many other Osteocephalus .

Comparisons: Osteocephalus leprieurii is a medium-sized species with spiny-backed males up to 53 mm and females with smooth dorsa reaching 63 mm ( Lescure and Marty 2000, Kok and Kalamandeen 2008). Unlike vilarsi , there are no frontoparietal ridges on the head of this species ( Jungfer and Hödl 2002).

Osteocephalus taurinus is a large species of the genus with distinct frontoparietal ridges. Melin (1941) had obtained “ Hyla taurina “ at Taracuá and at the Rio Uaupés, north of Rio Japú (about 60 km WNW of Taracuá). His female specimens were 82–98 mm in SVL. I have not seen them, but from his description there can be no doubt they are O. taurinus . No other similar frog is known from there. The largest preserved female of O. taurinus I have seen is the holotype, NMW 16492, with 103.9 mm SVL, from Barra do Rio Negro, Manaus, Brazil. I have measured a live female of 105 mm SVL at Reserva Adolpho Ducke, 25 km north of Manaus. Thus, the sympatric females of O. taurinus are 30–58 % larger than the adult female holotype of vilarsi , and other O. taurinus from the Rio Negro watershed are up to 69 % larger.

Apart from the considerably larger size, O. taurinus differs from vilarsi in the following characters: The canthus rostralis is more or less straight, distinct and rounded rather than angular, as in vilarsi . The supratympanic fold begins anterior to the tympanum at about 2/3 height of it and curves around the tympanum down towards the arm. In vilarsi it begins at the upper level of the tympanum and continues as a fairly straight sloping fold towards the arm insertion. The back of females is uniformly slightly granulate (smooth anteriorly, a few small granules posteriorly in vilarsi ). The distal subarticular tubercle on Finger IV is distinctly bifid. Although there is some variation in the extent of webbing throughout the range of O. taurinus , there is always more webbing, especially on the outer edge of Finger II and on the inner side of Toe II and both sides of Toe IV. For example, female ZFMK 55900 (77.5 mm SVL) from Manaus, Brazil, has a hand formula of I basal II 1 ½ — 3– III 2 2/3 — 2 ½ IV and a foot formula of I 1 — 2 II 1 — 2 + III 1 — 2 1/3 IV 2 1/3 — 1 V. The bones are green.

Osteocephalus planiceps is known from the western Amazon Basin in Colombia, Peru and Ecuador and a locality south of the Amazon in Brazil, about 300 km WSW Manaus ( Duellman and Mendelson 1995, Jungfer et al. 2000, Lehr 2001, Lynch 2008, unpublished data). Its northernmost locality in Colombia is in the Río Guaviare drainage, other northern localities are in the Río Caquetá drainage ( Lynch 2008). A single Brazilian locality is from the lower Rio Negro ( Gordo and Neckel-Oliveira 2004). No other records exist from the Rio Negro drainage.

cabrerai , O. festae ) and eastern ( O. inframaculatus ) Amazonia.

O. buckleyi O. cabrerai O. festae O. inframaculatus Eleven adult females of O. planiceps from Ecuador and Peru are 67.0 – 88.2 (x = 74.7) mm in SVL. The species shares many characters with vilarsi , especially the frontoparietal ridges, a straight, very prominent canthus rostralis, a more or less straight supratympanic fold, a large white subocular spot and white supralabial line, a dark tan sublabial line on the lower jaw, a few tan spots laterally and bold tan bars on the limbs, and a barely bifid distal subarticular tubercle on Finger IV. The webbing is very similar. That of the hand of vilarsi is almost within the range of that of O. planiceps , which has slightly more webbing on the outer edge of Finger II (1 3/4 – 2–). On the foot O. planiceps has slightly more webbing on the inner edge of Toe II (2 – 2+) and on Toe V (1 – 1+).

Both species can be distinguished by a few characters: The dorsum of O. planiceps females is smooth. Tubercles are lacking on the posterior two thirds of the back. The discs on the fingers of O. planiceps are round and large. Their diameter is about 80 % of the tympanum width. Even though the discs of the holotype of vilarsi are somewhat desiccated, they seem to have been smaller than those of O. planiceps . The tibiofibular bones are green. They are white in vilarsi , but they might just be discolored after such a long time in preservative.

Six other species of Osteocephalus are known from north of the Amazon in Brazil and on the Guaina Shield. Osteocephalus oophagus has two cranial ridges that are not visible superficially, as in vilarsi . The largest known female is 55.6 mm in SVL and thus considerably smaller ( Jungfer and Schiesari, 1995). Osteocephalus cf. buckleyi and O. cf. cabrerai both have rows of tubercles on the tarsus and areolate skin on the flanks. Both characters are absent in vilarsi . A species from Guyana, O. exophthamus Smith and Noonan, 2001 , is considerably smaller (up to 42.5 mm SVL) and lacks cranial ridges ( Kok and Kalamandeen 2008).

Another Guyanan species, O. phasmatus MacCulloch and Lathrop, 2005 , from high elevations> 1400 m on Mount Ayanganna has about the same size as vilarsi and also shares cranial ridges, but vilarsi lacks an axillary fold (one half of humerus length in phasmatus ) and the width of the disc of Finger III is about half the tympanum width (larger than tympanum).

Since vilarsi can be distinguished from all other species in the area it is here considered a valid member of the genus Osteocephalus characterized by (1) medium size (female 62.2 mm SVL), sexual dimorphism unknown; (2) skin on dorsum of females smooth anteriorly, interspersed with small flat tubercles on the posterior 2/3 of the back, unknown in males; (3) skin on flanks smooth; (4) very prominent, almost straight canthus rostralis; (5) frontoparietal ridges present, well visible through skin; (6) dentigerous processes of vomers angular; (7) straight supratympanic fold, sloping in a fairly straight line posterior to the tympanum; (8) web on inner edge of third finger reaching distal end of penultimate subarticular tubercle; (9) distal subarticular tubercle on Finger IV barely bifid; (10) dorsum uniform tan; (11) venter cream with some small brown spots posterior to the clavicle; (12) narrow white labial stripe to posterior edge of tympanum, extended into large subocular spot; (13) flanks brown with a few small irregular tan spots; (14) position of vocal sacs unknown; (15) juvenile coloration unknown; (16) tadpole habitat and labial tooth row formula unknown; (17) color of bones white (?).

Distribution: Osteocephalus vilarsi is still only known from the type locality in the extreme northwest of Brazil, close to the Colombian border.

Figures

FIGURE 7. Osteocephalus vilarsi (holotype): a. dorsal, b. ventral, c. lateral view of head. Not to scale.