Callulops gobakula sp. nov.

Gobakula Frog

(Figures 2–4)

Material examined. Holotype. QM J98925, field number conx5942, male, calling when captured, Dauan Island (9.4227° S, 142.5301° E), 12 January 2021, C. J. Hoskin, K. Aland, A. Davies, A. Zwar. Paratypes. QM J98926, field number conx5925, calling male; QM J98927, field number conx5926, female, with eggs; QM J98928, field number conx5927, female, with eggs. Collection details as for holotype.

Diagnosis. A medium-sized Callulops (male SVL 48.5–51.1 mm; females 48.5–52.2 mm), with moderately long hindlimbs (TL–knee/ SVL 0.40–0.41), distinctly enlarged finger and toe discs of approximately equal width (3FD/4 TD 0.90–0.95), all discs with a terminal (circum-marginal) groove (but can be indistinct on 1 st finger), moderately long hands (HandL/ SVL 0.25–0.26) and feet (FootL/ SVL 0.39–0.42), relatively small eye (EYE / SVL 0.094 –0.097), two subarticular tubercles on toe 4, distinct tympanum, relatively smooth skin, adult dorsal colour uniform brown, subadults patterned with gold or orange blotches and flecks, and call a series of 9–12 croaks (notes) uttered at 3.9–4.6 notes/s, for a call duration of 2.2– 2.9 s, with two dominant frequency peaks at about 1.06 and 1.58 kHz.

Description of type series. Measurements and proportions of the type series are presented in Table 1.A medium-sized Callulops (SVL mean 50.5 mm, range 48.5–51.8 mm); the two adult males (calling when collected) and two adult females (with well-developed eggs) similar in size (males 48.5 mm, 51.1 mm; females 50.8 mm, 51.8 mm). Head wide (HW/SVL 0.35–0.36), wider than long (HL/HW 0.88–0.90); broadly triangular in dorsal view, snout moderately long (SN/SVL 0.11–0.12); canthus rostralis rounded; loreal region steeply oblique, almost vertical, slightly concave posterior to naris; nostrils rounded, directed laterally, closer to tip of snout than to eyes; tip of snout broad (IN/SVL 0.078 –0.082); internarial distance slightly greater than, or about equal to, distance from naris to eye (EN/IN 0.82–1.00); snout rounded in lateral view, moderately truncate (but still slightly rounded) in dorsal view; eyes of moderate size (EYE/SVL 0.094 –0.097; EYE/SN 0.78–0.83); tympanum moderate size (TYM/SVL 0.061 – 0.082), obviously distinct (QM J98928), reasonably distinct (QM J98926, QM J98927) or indistinct (QM J98925); with a minimal (QM J98927), moderately distinct (QM J98925), or obvious supratympanic fold (QM J98926, QM J98928) extending from behind eye over tympanum and terminating above forelimb. Relatively long forearm, hand and 3 rd finger (ArmL/SVL 0.43–0.45; FA/SVL 0.17–0.20; HandL/SVL 0.25–0.26; F3L/SVL 0.16–0.17). Relative length of fingers 3>4≅2>1, with fingers 2 and 4 of similar length, and finger 3 conspicuously long; fingers unwebbed; discs present on all fingers, slightly less than twice width of penultimate phalanges; largest on 3 rd finger (3FD/SVL 0.025 –0.028), smallest on 1 st finger (1FD/SVL 0.023 –0.024), rounded on fingers 1 and 2, slightly truncate on fingers 3 and 4; circum-marginal grooves present on all finger discs but typically indistinct (faint) on finger 1 (but distinct on QM J98928), moderately distinct to distinct on finger 2, and distinct on fingers 3 and 4; subarticular tubercles prominent, one on fingers 1 and 2; two on fingers 3 and 4; inner and two outer metacarpal tubercles low, elongate. Relatively long hindlimb, foot and 4 th toe (TL–fold/SVL 0.35–0.38; TL–knee/SVL 0.40–0.41; FootL/SVL 0.39–0.42; T4L/SVL 0.24–0.28). Relative length of toes 4>3>5≅2>1, toe 4 conspicuously long; toes unwebbed; discs present on all toes, twice width of penultimate phalanges, widest on toe 4 (4TD/SVL 0.026 –0.031), narrowest on toe 1 (1TD/SVL 0.020 –0.025), larger than those of fingers (3FD/4TD 0.90–0.95), rounded on toes 1–3, slightly truncate on toes 4 and 5; all toe discs with distinct circum-marginal grooves; one prominent subarticular tubercle on each of toes 1 and 2; two low but prominent subarticular tubercles on each of toes 3, 4 and 5, those on toe 4 are on the two most distal joints; inner metatarsal tubercle small but prominent, oval shaped; no metatarsal tubercle. Skin texture in preservative. Dorsal, lateral and ventral surfaces perfectly smooth (Fig. 2). Colouration in preservative. Dorsal colour uniformly purplish brown (Fig. 2); indistinct paler crescent above shoulder on each side (absent on QM J98927). Ventral surfaces cream or light brown, with purplish tinge around margins of throat and belly and on underside of limbs. Palmar surfaces light brown, with some paler areas and fine white stippling; palmar and subarticular tubercles pale grey. Plantar surfaces brown; metatarsal and subarticular tubercles grey. Pupil horizontal (but usually heavily dilated).

Colour pattern and skin texture in life. Adults. Fairly uniform brown across all dorsal surfaces, with fine mottling of darker and lighter brown (Fig. 3A–D). Faint gold/cream crescent on each shoulder, above forelimb insertion, and faint gold marking between eye and tympanum. Paler brown/grey on tip of snout, along jawline, and on eyelids and knees, and paler patches on tops of hands, feet and digits. Dorsal brown colour becomes lighter on lateral surfaces and grades to light brown or cream on the ventral surfaces, which are unpatterned. Iris black, with heavy copper stippling. Variation among adults is minimal (e.g., Fig. 3A–D) except that some are lighter in all aspects of colour pattern, including a lighter brown or golden–brown dorsum, whiter lateral and ventral surfaces, and more prominent cream or golden crescent above the shoulder and between the eye and tympanum. Dorsal surfaces smooth or finely rugose (Fig. 3A–D); ventral surfaces smooth. Subadults (Fig. 3E–F). Dorsal surfaces more patterned than adults, typically bright golden on top of head and dorsum, becoming more blotched and flecked golden on a dark brown ground colour on the posterior half of the back, and with golden flecks on the dorsal surfaces of the arms and legs. Crescent above arms and small blotch between eye and tympanum prominent golden or orange (Fig. 3E–F). Some individuals have bright orange flecks on the back and tops of hindlimbs. Sides of face dark brown. Ventral markings dark brown or grey, with white flecks. Dorsal surfaces smooth or finely rugose; ventral surfaces smooth.

SVL measurements of additional, unvouchered animals in life. Calling males: 49.1 mm, 50.1 mm; probable females: 48.5 mm, 50.5 mm, 51.9 mm, 52.2 mm; subadults (based on more ornate colour pattern): 38.1 mm, 39.4 mm.

Call. A series of croaks: “rark, rark, rark, rark…”, starting quietly and becoming louder through the call (Fig. 4). Call parameters are presented in Table 2, and have the following average values (with range in parentheses): call duration 2.52 s (2.20–2.91); number of notes 10.2 (9–12), note rate 4.24 notes/s (3.86–4.56); dominant frequency (two clear energy peaks of similar magnitude in the spectrum view), peak one 1057 Hz (1029–1093), peak two 1576 Hz (1447–1704); note length 0.100 s (0.092 –0.107); note interval 0.154 s (0.134 –0.179); call interval 30.56 s (21.17–45.21).

Comparisons. I first compare C. gobakula sp. nov. to all congeners, and then in more detail to the morphologically most similar species, C. omnistriatus.

The medium body size of C. gobakula sp. nov. (male SVL 48.5–51.1 mm; females 48.5–52.2 mm) distinguishes it from five smaller species (which all attain maximum SVL <40 mm, at least for males): C. boettgeri (Méhely, 1901); C. dubius (Boettger, 1895); C. eremnosphax Kraus & Allison, 2009; C. fuscus (Peters, 1867); C. glandulosus (Zweifel, 1972). Body size also distinguishes C. gobakula sp. nov. from many species that attain substantially larger size (all attain SVL> 70 mm): C. argus Kraus, 2019; C. bicolor Kraus, 2019; C. doriae (Boulenger, 1888); C. personatus (Zweifel, 1972); C. robustus; C. stellatus Kraus, 2019; C. stictogaster (Zweifel, 1972); C. taxispilotus Kraus, 2019; C. valvifer (Barbour, 1910). The moderately long hindlimbs of C. gobakula sp. nov. (TL–knee/SVL 0.40–0.41) distinguish it from many species with short hindlimbs (TL–knee/SVL maximum <0.38): C. comptus (Zweifel, 1972); C. fojaensis Oliver, Richards & Tjaturadi, 2012; C. fuscus; C. glandulosus; C. humicola (Zweifel, 1972); C. kampeni (Boulenger, 1914); C. kopsteini (Mertens, 1930); C. personatus; C. sagittatus Richards, Burton, Cunningham & Dennis, 1995; C. stictogaster; C. wilhelmanus (Loveridge, 1948); and from those with particularly long hindlimbs (TL–knee/SVL minimum> 0.43): C. biakensis Günther, Stelbrink & von Rintelen, 2012; C. boettgeri; C. dubius; C. marmoratus Kraus & Allison, 2003; C. mediodiscus Oliver, Richards & Tjaturadi, 2012; C. neuhaussi (Vogt, 1911); C. yapenensis Günther, Stelbrink & von Rintelen, 2012. It further differs from the morphologically similar species, C. mediodiscus, by having shorter hands and feet (HandL/SVL 0.25–0.26 vs 0.28–0.30; FootL/SVL 0.39–0.42 vs 0.48–0.54) and a call that has more notes (9–12 vs 4) that are uttered at a faster note rate (3.86–4.56 notes/s vs 1.42–1.49 notes/s). The relatively smooth dorsal and lateral skin of C. gobakula sp. nov. differs from the pustulose skin of: C. argus; C. bicolor; C. doriae; C. microtis (Werner, 1901); C. neuhaussi; C. robustus; C. stellatus; and C. taxispilotus. Callulops gobakula sp. nov. has circum-marginal grooves on all toe and finger discs (but can be indistinct on 1 st finger), which separates it from the following species with circum-marginal grooves absent from all toe and finger discs: C. argus, C. bicolor, C. glandulosus, C. stellatus, C. stictogaster; and those with grooves absent from finger discs: C. doriae, C. neuhaussi, C. robustus; and those with grooves absent from fingers 1 and 2: C. eremnosphax. The presence of distinctly expanded finger and toe discs distinguishes C. gobakula sp. nov. from four species that lack discs entirely: C. glandulosus, C. sagittatus, C. stictogaster, C. wilhelmanus; and from species that have discs only minimally expanded: C. comptus, C. fojaensis. Callulops gobakula sp. nov. differs from C. wondiwoiensis Günther, Stelbrink & von Rintelen, 2012 and C. yapenensis by having a smaller eye (EYE/SVL 0.091 –0.097 vs 0.107 –0.125 and 0.135, respectively), a greater EN/IN ratio (0.82–1.00 vs 0.64–0.88 and 0.74, respectively), and an obviously different call, including faster note rate (mean 4.24 notes/s vs mean 2.18 notes/s and 2.55 notes/s, respectively) and lower dominant frequency (mean 1.58 kHz vs mean 1.75 kHz and 2.2 kHz, respectively).

Callulops gobakula sp. nov. is most similar to C. omnistriatus, but it differs in many morphological traits, including smaller males (SVL 48.5–51.1 mm vs 55.0– 59.6 mm) and females (SVL 48.5–52.2 mm vs 49.9–66.9 mm); having shorter hindlimbs (TL–knee/SVL 0.40–0.41 vs 0.41–0.47); shorter foot (FootL/SVL 0.39–0.42 vs 0.42–0.48); smaller eye (EYE/SVL 0.09–0.10 vs 0.12–0.14; EYE/SN 0.78–0.83 vs 1.0–1.3); narrower toe discs (4TD/SVL 0.026 –0.031 vs 0.033 –0.039); presence of two subarticular tubercles on fourth toe (vs three); circum-marginal groove on disc of 1 st finger indistinct to moderately distinct (vs well-developed); snout less obviously truncate in dorsal view; iris reddish brown/copper in life (vs light green-bronze); dorsal and ventral surfaces lacking a purplish wash in life (vs violet or lavender wash); call of 9–12 notes (vs 5–6 notes). Data for C. omnistriatus comes from Kraus & Allison (2009a) and Richards (2025).

Etymology. The species name gobakula is derived from the local language (Kalaw Kawaw Ya) words “goba kula”, meaning boulders, and refers to the species being restricted to areas of piled boulders. Local language was provided by Laurie Elisala, Torenzo Elisala, Abi Mooka, Tenny Elisala, and Thomas Mooka.

Distribution. Only known from Dauan Island (Fig. 5). Likely to be endemic to the island, based on the lack of piled boulder habitat on nearby islands or the adjacent New Guinean mainland. All records come from the lower and mid slopes of Mt Cornwallis (Simakal Pad), from an elevational range of 30–140 m a.s.l. Surveys were not conducted at higher elevations.

Natural history. Restricted to areas of deeply piled granite boulders with associated rainforest vegetation (Fig. 6). Males were found calling at night in narrow gaps between surface boulders. Females and sub–adults were found active at night on boulders and leaf-litter among boulders. Frogs were agile on the granite boulders and quickly retreated into deep gaps among the rocks when disturbed. The finger and toe discs were used in an ‘angular’, ‘gripping’ way (e.g., Figs 3A, 3D) when climbing on the rocks. Callulops gobakula sp. nov. was not found during surveys in adjacent habitats without deeply piled boulders. Surveys were conducted during wet weather, and it is assumed the species retreats deep among boulders during dry periods. Callulops gobakula sp. nov. is assumed to be a terrestrial breeder with direct development, like all other asterophryines. This is further supported by the large, pale eggs found in females QM J98927 and QM J98928, which are typical of other asterophryines (e.g., Hoskin 2004; Anstis et al. 2011).

Conservation. If C. gobakula sp. nov. is restricted to Dauan Island, then it has a very small distribution. The island has an area of about 3.4 km 2, of which approximately 1 km 2 is estimated to be potentially suitable piled rock habitat (or 1.05 km 2 measured as a minimum convex polygon). Callulops gobakula sp. nov. fits an IUCN Vulnerable listing based on Criterion D2: ‘restricted area of occupancy (AOO <20 km 2) or number of locations (≤5), with a plausible future threat that could drive the taxon to Critically Endangered or Extinct in a very short time’. The ‘plausible future threat’ is Yellow Crazy Ants Anoplolepis gracilipes (Smith, 1857). This invasive species is not known from Dauan Island but has been accidentally introduced to many tropical islands, globally, where it has had many documented impacts on invertebrates and vertebrates (e.g., Holway et al. 2002; Lach & Hooper-Bùi 2010). Severe impacts have been documented on the abundance of small skinks in rainforest of mainland north Queensland (Lach et al. 2022). Based on likelihood of introduction and documented impacts, particularly on islands, it is reasonable to consider Yellow Crazy Ants as a plausible future threat that could rapidly impact C. gobakula sp. nov. More detailed surveys are required to estimate fine-scale distribution on Dauan Island and to assess other potential threats.