Litoria eungellensis Price, Hoskin, Mahony & Donnellan, 2025, sp. nov.

Litoria eungellensis sp. nov. Eungella Whirring Tree Frog Figs 12, 13, 14 Holotype: QM J35106 (Fig. 12), adult female. Type locality: Thurgood Farm, 18 km from Dalrymple Heights, mid-eastern Queensland, Australia. -21.033º, 148.6º. Collected by G.J. Ingram on 7 December 1978. Diagnosis. Litoria eungellensis sp. nov. can be diagnosed from L. littlejohni and L. watsoni by its smaller size (28–39.5 mm vs 43–61 mm and 42–64 mm, respectively), yellow or orange groin, and posterior thighs with black spots or patches (vs immaculate orange markings on the anterior and posterior surfaces of the femur and tibia, in the groin, and posterior flanks). Can be distinguished from L. ewingii, L. paraewingi and L. verreauxii by expanded finger and toe discs (vs terminal discs similar in width to, or only slightly wider than, the terminal phalanx). Distinguished from L. jervisiensis by the termination of the lateral head stripe near the forearm (vs terminating along flank), dark patches in groin (vs absent), and smaller size (28–39.5 mm vs 37–44 mm). Litoria eungellensis sp. nov. can be diagnosed further from the following species by the greater mean number of notes in the advertisement call (39.3, range 30–35 vs 9, range 4–22 in L. ewingii; 2.3, range 1–4 in L. jervisiensis; 8.8, range 5–16 in L. littlejohni; 4, range 3–8 in L. paraewingi; 9, range 8–11 in L. sibilus; 16.7, 8–23 in L. verreauxii; and 6.5, range 3–14 in L. watsoni). Litoria eungellensis sp. nov. can be diagnosed from L. corbeni and L. revelata by the presence of well-demarked, back spots or patches in all of the axilla, groin, posterior thigh, and underside of the hindlimbs (vs variable presence of spots in any of the first three locations, and any spots present being generally smaller and less conspicuous, and absence of dark marks on the underside of the hindlimb) (Table 5). It can be diagnosed further from L. corbeni and L. revelata by having calls with a lower mean pulse repetition rate (78.1 pps, range 67–86 vs 130.5, range 127.3–136.7 for L. corbeni and 105.3, range 87–127 for L. revelata) (Table 6) and by apomorphic states at 31 and 10 nucleotide sites, respectively, in the ND4 gene alignment (Table 3). Measurements of holotype (mm). SVL 28.8, HL 8.3, HW 8.6, TL 16.3, TD 1.8, ED 3.3 IOD 5.8, IND 2.2, ETD 1.1, THL 14.3, FL 14.2, IMT 1.1, FLL 6.9, Fing3D 1.0. Description (including holotype; Fig. 12). A summary of variation in 14 mensural characters and five ratios is presented in Table 4. Mean SVL: females = 38.5 ± 0.99, 37.3–39.5, males = 29.6 ± 1.19, 28.3–31.8 mm. Snout rounded when viewed from above, blunt or rounded in profile. Head length less than, to about equal to, head width (HL/HW 0.95 ± 0.06, 0.76–1.03) and approximately one-third of SVL (HL/SVL 0.29 ± 0.02, 0.24–0.31). Pupil horizontal when constricted (Fig. 13). Nostrils more lateral than superior; closer to tip of snout than to eye. Canthus rostralis well defined, and straight. Eye relatively large (ED/HL 0.4 ± 0.03, 0.33–0.46). Tympanum distinct; circular; diameter about half eye diameter (TD/ED 0.5 ± 0.06, 0.44–0.58). Vomerine teeth in short straight rows from anterior edge of choanae to midline of palate (assessed in SAMA R72462). Fingers long, narrow; webbing absent. Subarticular tubercles prominent. Terminal discs prominent. Fingers in order of length: 3>4>2>1. Dark brown nuptial pad wraps around inner surface of Finger I from base of finger to base of penultimate phalanx (SAMA R72460–4). Hindlimbs moderately long (TL/SVL 0.5 ± 0.02, 0.52–0.57). Toes in order of length: 4>5=3>2>1. Webbing on all toes reaches base of penultimate phalanx. Subarticular tubercles prominent. Rectangular inner metatarsal tubercle approximately one-quarter length of first toe. Terminal toe discs not prominently expanded (Fig. 12). Dorsum finely granular. Upper surface of limbs smooth or finely granular. Flanks with dense coverage of low tubercles. Chin and ventral surfaces of limbs smooth; abdomen finely granular. Colour in life. Dorsal surfaces tan, cream-brown or bronze, continuing onto flanks. Mid-dorsum with darker shading that starts with well-demarcated margin between eyes; this pattern may not be obvious in males in courting colour. Some individuals with small to minute dark flecks over dorsal surfaces and flanks. A black or dark or light brown stripe, with lower margin often poorly defined, extends from nostril along canthus rostralis through eye and over tympanum to just past forearm. Belly white to cream. Inflated vocal sac near transparent with sparse yellowgreen pigmentation (Fig. 13A, B). Upper iris bright copper-gold colour, lower iris brown copper-gold. Axillae have prominent dark marks in all specimens examined. Groin yellow, with dark patch and large dark spots anterior to groin (Figs 13, 14). Posterior of thighs with black spot at proximal and distal margins, or additionally with black dorsal margin, or with black pattern spread across thigh (Figs 6E, F, 13, 14). Variably-shaped black or dark marks present on the underside of the hindlimbs (Table 5, Fig. 14A, B). Etymology. The species name eungellensis refers to the distribution of this species in the ‘Eungella’ region. According to Kitching (2020) “Eungella is generally supposed to mean ‘Land of Clouds’.” and “was adopted from local indigenous languages [of the Birri and Wiri people] by the first European residents”. Distribution. Restricted to high elevations (> 880 m a.s.l.) from Pease’s lookout, 3km ENE of the Eungella township northeast to Mt David and Mt William in the headwaters of the Cattle Creek North branch, SSW of Mt Dalrymple and the Clarke Range in Eungella National Park and in the west to farm properties west of the Dalrymple Road (Fig. 15). These records fall in the elevational range of about 880–1211 m a.s.l., with most records between 950 and 1000 m a.s.l. It is not known from the fairly large extent of upland rainforest above 1000 m a.s.l. to the north and northwest of the documented historical range, although this may be due to the limited number of sites and frequency of surveys to date in this difficult to access area (Meyer et al. (2020). Extensive surveys at 114 sites through the Eungella region between 2000 and 2015 only found L. eungellensis sp. nov. between 930–980 m a.s.l. in the small area of the headwaters of Cattle Creek North branch and at nearby farm dams (Meyer et al. 2020). Historic records include Pease’s lookout (900 m a.s.l.), as an observation (Covacevich & McDonald 1993), and Snake Road (970 m), as a specimen (QM J45838) (Fig. 15), but there have been no records from these areas in recent decades, despite survey effort (Meyer et al. 2020). Two vouchers (NHM 64.7.8.11–12), listed as paratypes in Ingram et al. (1982) are from Port Denison (= Bowen), north Queensland. This locality is an anomalous record because it is at sea level, is unsuitable dry sclerophyll habitat, and is more than 100 km north of the Eungella region. We take this locality information to be erroneous and the identity of these specimens needs assessing. Ecology and reproductive biology. Most occurrence records come from slow-flowing rainforest creeks, where adults have been found sitting in vegetation along streams and eggs and tadpoles have been observed in small pools in bedrock adjacent to the stream (Retallick et al. 1997, Retallick 1998, Meyer et al. 2020). These small, isolated pools are rare along the streams—for example, breeding habitat was only present along 30 m of a 700 m section of Mt William Creek (Retallick 1998). Litoria eungellensis sp. nov. also occupies farm dams on small gully lines in pasture adjacent to rainforest habitat west of the Dalrymple Road (Meyer et al. 2020). In these situations, adult males call from low vegetation on the margins of the dams. Calling and breeding have been recorded in Spring and early Summer (Retallick et al. 1997, Retallick 1998, CJH observations). Anstis (2017) compared the tadpole morphology of L. revelata and L. eungellensis sp. nov. (then denoted as L. revelata Eungella) but did not find any differences. Hero et al. (1996) described the tadpole in detail, pointing out that a blue sheen was present on the abdomen (vs a dull copper sheen on L. revelata; Anstis 2017), but the consistency of this trait needs verification. Conservation status. Litoria eungellensis sp. nov. has an extremely small spatial and elevational distribution. Rainforest vegetation is present at mid and low elevations at Eungella but all records of the species come from above 900 m, suggesting upland restriction. The highest elevations at Eungella are the summits of Mt David (1200 m), Mt William (1250 m) and Mt Dalrymple (1205 m), and most of the L. eungellensis sp. nov. records are clustered in creek headwaters between these summits. Below we assess conservation status against IUCN Red List (2012) criteria. The extent of occurrence (EOO, measured as a minimum convex polygon around all records in Fig. 15) is approximately 15 km 2, but most records (including those in Meyer et al. 2020) come from an extremely small area of <1 km 2. These values are well under the EOO specified in Criterion B for Critically Endangered (100 km 2); in fact, the EOO for the currently known sites is less than the area of occupancy (AOO) for Critically Endangered under this criterion (10 km 2). Litoria eungellensis sp. nov. occurs at a single ‘location’ because a single threatening event could rapidly affect all individuals (IUCN 2012). Several threats should be considered: fire, clearing, chytridiomycosis, and climate change. Fire may be a threat given the extensive burning of upland rainforest at Eungella in 2018 (Hines et al. 2020). These fires did not burn any habitat known to support L. eungellensis sp. nov. but did burn similar habitat just to the west, so fire could be considered a future threat. Clearing may be a localised threat to sites on private land (e.g., around dams) because there is limited protection for regrowth rainfo