Hypsiboas angelicus, Myers & Donnelly, 2008

Myers, C. W. & Donnelly, M. A., 2008, The Summit Herpetofauna Of Auyantepui, Venezuela: Report From The Robert G. Goelet American Museum-Terramar Expedition, Bulletin of the American Museum of Natural History 2008 (308), pp. 1-147 : 45-51

Hypsiboas angelicus , new species Figures 22–26

HOLOTYPE: EBRG 2733 (field no. CWM 19318), an adult male from summit of Auyantepui at 5 ° 589N, 62 ° 339W ( AMNH – TERRAMAR Camp 4), 1600 m, Bolívar, Venezuela; collected February 19–21, 1994, AMNH –TERRAMAR Expedition. Figure 22.

ETYMOLOGY: The specific name is a Latin adjective pertaining to celestial angels, including, perhaps, legendary pilot Jimmy Angel of Auyantepui fame.

DIAGNOSIS: This species belongs to the former Hyla geographica group (as defined by Duellman, 1973), as indicated by the external characters of calcars, prepollical spines not projecting through the skin, basal webbing between fingers, and general habitus. Presence of a flat mental gland and close resemblance with Hypsiboas roraima places it in the benitezi group of Hypsiboas ( Faivovich et al., 2005: 86–87) .

It is further distinguished by the following suite of characters (in the format of Duellman and Hoogmoed, 1992): (1) body slender, with head distinct from body; (2) skin on dorsal surfaces including upper eyelids weakly granular (in life) or smooth (in preservative); head skin not co-ossified; (3) tympanum distinct; (4) fingers with basal webbing; (5) toes about half webbed; (6) fringes absent on limbs; a short conical calcar present; (7)

13 ‘‘ Boana Gray ’’ was first published by John Edward Gray (1825: 214) but probably not invented by him. One did not purposely create new names to be published as synonyms. Names first published in synonymy usually (if not always) were manuscript or specimen-label names. The name Boana may have been attached to a BMNH specimen of ‘‘ Rana Boans, Lin. ’’ by Edward Whitaker Gray (1748–1806). J. E. Gray (1825: 93) mentioned his ‘‘late [great] uncle, who paid great attention to this department of zoology, and several of whose manuscript species still remain unpublished’’. E. W. Gray, Keeper of Natural Curiosities at the British Museum ( Gunther, 1976), has been all but forgotten. He published at least one herpetological paper, an insightful critique of Linnaeus’ class Amphibia (E. W. Gray, 1789).

axillary membrane absent; (8) dorsum brown with pale lichenlike spots and irregular black spotting; broken black vertebral stripe present (at least in holotype); flanks grayish with small black spots; limbs brown with gray crossbanding; rear of thigh gray; venter pale (gray in life), unmarked; foot webbing brown like dorsum; upper half of palpebrum clear except for brown upper edge and brown flecking, lower half brown; (9) vomerine odontophores angular.

Hypsiboas angelicus is remarkably similar to Hypsiboas roraima (Duellman and Hoogmoed) from Cerro Roraima, differing most conspicuously in lacking pale reticulation on the palpebral membrane.

MORPHOLOGY: An adult male 38.4 mm SVL, with large vocal slits; a shallow subgular vocal sac, irregularly folded against posterior half of throat and chest. A thin, loosely organized mental gland on anterior half of gular region, with a convexly rounded posterior edge (fig. 25, ventral view). Body slender, head approximately as wide as long, prominent, much wider than body (head width/greatest body width 5 1.25); snout slightly sloping, short, truncate in dorsal and lateral views, not projecting beyond margin of upper lip; canthus rostralis rounded; loreal region concave, inclined outward to rounded lip; nostrils protuberant laterally; internarial area slightly depressed; top of head flat; upper eyelid as wide as interorbital distance; eye length greater than its distance to naris (eye/naris-eye 5 1.12), nearly as long as snout; tympanum distinct except for upper edge, circular, deflected dorsolaterally; separated from eye by a distance slightly greater than its length; supratympanic fold weak, but covering upper edge of tympanic annulus.

Axillary membrane absent; forearm slen- der, with row of large, ill-defined, closely spaced low ulnar tubercles situated along ventrolateral edge of arm; fingers slender, with rounded terminal discs; third finger disc subequal with length of tympanum; relative finger length III. IV. II. I; subarticular tubercles rounded, low, barely protuberant; supernumerary tubercles few, indistinct proximally; large palmar tubercle rounded, low and very indistinct; thenar tubercle elongate, not well defined, with median projection distally; basal webbing between fingers II–IV.

Hind limb long and slender; tibia 53.4 % of SVL; heel bearing short conical tubercular calcar; toes with rounded terminal discs subequal in size to those on fingers; relative toe length IV. V. III. II. I; subarticular tubercles low, rounded, only slightly protuberant; supernumerary tubercles few, indistinct proximally; inner metatarsal tubercle oval; outer metatarsal tubercle small and indistinct; no tarsal fold; toes moderately webbed, with formula: I 2– 2 K II 1 K –2 K III 1 K –2 K IV 2–1 K V (fig. 23).

Vomerine odontophores oblique, weakly arched, situated between large choanae and separated by a distinct medial gap, each with about a dozen teeth. Tongue large, covering floor of mouth, only lateral edges free, fully attached behind.

Skin on dorsal surfaces, including upper eyelids, weakly granular in life (fig. 22), becoming virtually smooth in preservative; skin on throat and under arms smooth in preservative, coarsely granular (areolate) on venter; proximal posteroventral surfaces of thighs crowded with smooth tubercles; ventral thigh surfaces covered with large, probably glandular tubercles, low and smooth in preservative. Supracloacal flap short, slightly overlapping vent opening, which is directed posteriorly at upper level of thighs; vent situated immediately above a smooth vertical channel situated between two large, vertically elongate tubercles, which in turn are flanked by large, smoothly rounded tubercles that grade into those on proximal posteroventral surfaces of thighs (fig. 24).

MEASUREMENTS (in mm): Snout to vent 38.4; tibia 20.5; hand from proximal edge palmar tubercle to tip of longest finger 10.5; foot from proximal edge inner metatarsal tubercle to tip of longest toe 14.1; head length on the diagonal from tip of snout to angle of jaw 13.6; greatest head width (near angle of jaws) 13.8; width of upper eyelid 4.0; greatest body width 11.0; width of interorbital area 4.0; internarial distance 3.0; rear edge naris to anterior corner of eye 4.2; eye length from anterior to posterior corner 4.7; eye to tympanum 2.5; horizontal length of tympanum 2.2; width of third finger disc 2.1; width of fourth toe disc 1.9.

COLORATION: In life (fig. 22), grayish brown, with lichenlike spots of pale rose and irregular black spotting on dorsum; a narrow, broken black vertebral stripe from snout to midbody; flanks gray with small black spots; limbs brown with inconspicuous gray crossbanding; rear of thigh gray; foot webbing brown like limbs. Dull grayish yellow under chin, turning uniformly light gray over all remaining ventral surfaces, including palms and soles. Iris pale golden yellow, with very fine, scarcely noticeable, black venation. Upper half of palpebral membrane clear in life (except for brown upper edge and brown flecking, as noted in preservative), lower half brown.

2733). Small arrows point to convex posterior edge of flat mental gland.

In preservative (fig. 25), pattern well retained, with rose coloring faded from lichenous dorsal spots, and with throat and venter turned pale yellowish and undersides of limbs pale brownish.

NATURAL HISTORY AND VOCALIZATION

A single specimen believed to be this species was recorded by Donnelly in dense vegetation next to the river at Camp 1 on February 4; this frog escaped capture. Identification was confirmed by capture of the holotype two weeks later in Camp 4, although further recordings were not obtained.

The advertisement call (fig. 26) was a long train of explosive, short ‘‘beeps’’ given continuously for over a minute, at a rate of 2.6 notes per second, with a mean internote interval of 0.35 sec (range 0.31–0.49 sec, N 5 50). The note is weakly pulsatile or nonpulsatile, 0.03 sec long, and frequency modulated, starting at about 1800 Hz and rising to about 2400 Hz before a slight trail-off.

REMARKS

Faivovich et al. (2005: 86; 2006) postulat- ed a ‘‘flat mental gland’’ as a synapomorphy of the Hypsiboas benitezi group; for additional comments on the composition of the group and illustration of the gland, see Faivovich et al. (2006: fig. 4A–C). This easily overlooked structure is a thin area of glandular tissue and is present in both Hypsiboas roraima and H. angelicus (fig. 25, ventral view). These two species are very similar, with the dorsal color pattern of angelicus seeming to fit comfortably with variation in roraima (compare figs. 25, 27). The presence of pale reticulation on the palpebral membrane is a diagnostic characteristic of various frogs, including H. roraima (Duellman and Hoogmoed, 1992: 6, 10; MacCulloch and Lathrop, 2005: 26–27), but is lacking in H. angelicus . To our knowledge, intraspecific variation has yet to be detected in this character, which is discernible in both living and preserved specimens (the palpebral membrane of the angelicus holotype was examined in both states).

Hypsiboas roraima was indicated by Mac- Culloch and Lathrop (2005: table 2) as lacking all vestiges of finger webbing, but Duellman and Hoogmoed (1992: fig. 4) illustrated basal webbing between fingers III and IV that is similar to that of H. angelicus (fig. 23). Both species have a notably tuberculate cloacal area. The holotype of H. angelicus has a pair of conspicuously large, vertically elongate tubercles that flank a smooth vertical channel below the vent (fig. 24). Although the size and shape of these ridgelike tubercles conceivably might be diagnostic, they might also have resulted from intraspecifically variable fusion of the large round tubercles. The vocalization of Hypsiboas roraima seems to be unknown, but considering other close resemblances and seeming lack of sympatry, it would not be surprising if the two species had somewhat similar calls.

Hypsiboas species ( H. angelicus ?), TADPOLE Fig. 28

MATERIAL: Camp 2, 1750 m: AMNH A-164998 (1 larva), from the 1994 AMNH – TERRAMAR Expedition to Auyantepui.

This curious little tadpole was found in a pool of water at the bottom of a deep (. 10 m) sandstone crevice, into which our colleague Petia Alcócer had descended by rope. Assuming that it does not represent a frog otherwise unknown from the Auyán summit, we tentatively associate it with Hypsiboas angelicus .

HABITUS AND PROPORTIONS: Headbody length 12.6 mm, body width 7.0 mm, body depth 6.0 mm, total length 35.7 mm. Body slightly depressed (body width/body depth 5 1.17); snout acutely rounded in dorsal view, rounded in profile. Nares dorsolateral, directed laterally, 1.5 mm from tip of snout; distance between centers of nares 2.1 mm. Small eyes dorsal, directed laterally, diameter 0.4 mm; interocular plane 2.6 mm from tip of snout. Spiracle a 0.7 mm long tube adhered to body wall, sinistral. Large vent opening dextral to caudal fin; vent 1.5 mm long, with inner wall free of body.

Lateral line system (not illustrated) not pronounced, including: a pair of supraorbital lines diverging in straight lines from center of snout to top of eye; high and middle lateral lines on body; an angular branch with a short anteriorly extending loreal branch; a gular branch extending from the middle lateral line and ventrally converging toward the angular branch.

A single unpigmented bump (gland?), 0.4 mm in diameter, near end of body on each side.

Tail length 64.7 % of total length. Maximum tail height 16.8 % of total length. Fins unequal in height; dorsal fin 1.8 mm and ventral fin 1.6 mm at maximum tail height. Tail distally forked.

PIGMENTATION: In preservative, headbody uniformly covered by rodular melanophores, brown in color. Brown rodular melanophores also overlying the pale yellowish brown tail musculature and extending proximally along both fins, becoming sparse and disappearing before the end of the tail. Well-defined pale yellow chromatophores are sparsely distributed, occurring mainly on the sides of the body and along the base of the upper tail fin.

MOUTHPARTS: Mouth ventral. Oral disc width 2.2 mm, not emarginate. Labial teeth in two anterior and three posterior rows, which are subequal in length except for last posterior row being shortest. Anterior tooth rows medially broken. Edges of upper and lower jaw sheaths with pointed serrations; laterally, the lower jaw sheath is only narrowly keratinized. Oral disc nearly surrounded by pointed marginal papillae, except for an anterior median break in the row. The marginal papillae are disposed in a single row (looking like a double row where folded posteriorly, fig. 28).

REMARKS

A uniform layer comprised solely of epidermal rod-shaped melanophores is a distinctive feature of this tadpole, which is identified only tentatively as Hypsiboas angelicus . Rodular melanophores occur widely in tadpoles of the tribe Cophomantini , but usually other shapes of melanophores are present as well (J. Faivovich, personal commun.). The bumplike structure on the rear of