Zoosyst. Evol. 97 (2) 2021, 355-382 | DOI 10.3897/zse.97.59696 > PENSUFT. eee ee BERLIN Description of six new species of Xenorhina Peters, 1863 from southern Papua New Guinea (Amphibia, Anura, Microhylidae) Rainer Giinther!, Stephen Richards 1 Museum fiir Naturkunde, Herpetology Department, Invalidenstr. 43, 10115 Berlin, Germany 2 Herpetology Department, South Australian Museum, North Terrace, Adelaide, South Australia 5000, Australia http://zoobank.org/FB92F5 DF-7FC7-4F01-A 1DD-8E85 B6F5FE67 Corresponding author: Rainer Giinther (rainer.guenther@mfn.berlin) Academic editor: Rafe Brown @ Received 15 October 2020 # Accepted | June 2021 Published 9 July 2021 Abstract We describe six new species of the microhylid frog genus Xenorhina from the southern slopes of Papua New Guinea’s central cor- dillera and adjacent lowlands, based on a combination of morphological (including osteology) and bioacoustic features. All of the new species are fossorial or terrestrial inhabitants of tropical rainforest habitats and belong to a group of Xenorhina having a single, enlarged odontoid spike on each vomeropalatine bone. Advertisement calls and habitat preferences are described for each species, one of which is amongst the smallest hitherto members of the genus. Description of these six species brings the total number of Xe- norhina known to 40 and emphasises the importance of the high-rainfall belt that extends along the southern flanks of New Guinea’s central cordillera as a hotspot of Melanesian amphibian diversity. Key Words acoustics, Asterophryinae, central cordillera, fossorial frogs, morphology, odontoid spike, rainforest, taxonomy Introduction Asterophryine microhylid frogs occur from mainland and insular southeast Asia eastwards through New Guin- ea to New Britain Island and northern Australia (Clulow and Swan 2018, Suwannapoom et al. 2018). The group reaches its greatest diversity in the New Guinea region, where many genera are endemic or near-endemic to the New Guinea mainland (Menzies 2006). Xenorhina Pe- ters, 1863 is a moderately speciose genus of asterophry- ine microhylid frogs that currently contains 34 named species distributed across New Guinea and some nearby islands (Zweifel 1972, Blum and Menzies 1989, Gunther and Richards 2005, Menzies 2006, Kraus 2011, Gtn- ther et al. 2020, Frost 2021). With the exception of three arboreal species, X. arboricola Allison & Kraus, 2000, X. macrodisca Gunther & Richards, 2005 and X. varia Gunther & Richards, 2005, other members of the genus are fossorial or inhabit litter or subterranean burrows and have squat bodies, small, pointed heads, small eyes and short, robust limbs (Menzies and Tyler 1977, Blum and Menzies 1989). The genus Xenorhina includes two main groups: |) species having one or two spikes on each vom- eropalatine and 2) species lacking vomeropalatine spikes. The former group was previously recognised as a separate genus, Xenobatrachus Peters & Doria, 1878 (e.g. Blum and Menzies 1989, Menzies 2006), but Kraus and Allison (2002) noted the lack of synapomorphies distinguishing Xenorhina from Xenobatrachus and suggested that the two genera may need to be combined. Frost et al. (2006) subsequently synonymised Xenobatrachus with Xenorhi- na (the older available name), based on molecular evi- dence, a move supported by Kohler and Gunther (2008). The monophyly of Xenorhina + Xenobatrachus is now well supported (de Sa et al. 2012, Peloso et al. 2015, Ri- vera et al. 2017), but the relationships of Xenorhina to other asterophryine genera remain poorly resolved. For example, molecular studies by Kohler and Gunther (2008) and Pyron and Wiens (2011) concluded that Xenorhina 1s closely related to Asterophrys and some Callulops, while Copyright Gunther R, Richards S. This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. 356 Rivera et al. (2017) found that Xenorhina is the sister tax- on to Callulops, Mantophryne and Hylophorbus. In this paper, we describe six new Xenorhina that be- long to the group of species with enlarged vomeropala- tine spikes. They were collected from localities within and adjacent to the high-rainfall belt extending along the southern flanks of New Guinea’s central cordillera (McAlpine et al. 1983), a region that has previously been reported to support a diverse amphibian fauna (Hyndman and Menzies 1990, Richards and Gunther 2019). Descrip- tion of these frogs brings to 40 the number of recognised Xenorhina, making it the third most speciose microhylid genus in the New Guinea region after Oreophryne (61 species) and Cophixalus (48 species) (Richards and Gunther 2019, Frost 2020). It also reinforces the impor- tance of the extensively forested southern flanks of Papua New Guinea’s central cordillera as a hotspot of Melanesi- an amphibian diversity. Material and methods Fieldwork was conducted in tropical rainforest habitats along the southern flanks of Papua New Guinea’s cen- tral cordillera. Most frogs were located at night by their advertisement calls. Representative specimens were pho- tographed in life and voucher animals were euthanised in an aqueous chlorobutanol solution (Gamble 2014) and subsequently fixed in 10% formalin. All specimens were transferred to 70% ethanol within two days of fixation. Descriptions follow a template developed for taxonom- ic treatments of New Guinea microhylid frogs, including Xenorhina (e.g. Gunther et al. 2014, Gunther et al. 2020). We adopt the biological species concept of E. Mayr (1963 and elsewhere), placing emphasis on reproductive isola- tion and we treat morphological, osteological and etho- logical (acoustic) differences as expressions of genetic differences that are large enough to prevent exchange of genes between the populations concerned. Our approach follows that of other taxonomic studies of this genus (Zweifel 1972, Blum and Menzies 1989, Kraus and Alli- son 2002, Gunther et al. 2014). The following measurements were taken with a dig- ital calliper (> 10 mm) or with a binocular dissecting microscope, fitted with an ocular micrometer (< 10 mm) to the nearest 0.1 mm from preserved specimens using protocols for microhylid frogs adopted previously (e.g. Gunther et al. 2014): SUL — snout-urostyle length from tip of snout to posterior tip of urostyle (SUL is sufficient- ly similar to SVL that, where relevant, we compare our SUL measurements with SVLs presented for members of the genus in some papers); TL — tibia length: external distance between knee and tibio-tarsal articulation (re- ferred to herein also as “shank’”’); TaL — length of tarsus: external distance between tibio-tarsal and tarsal-meta- tarsal joints when held at right angles; T4L — length of 4" toe: from tip of toe to proximal end of inner meta- tarsal tubercle; T4D — transverse diameter of disc of 4" zse.pensoft.net Gunther, R. & Richards, S.: Six new species of Xenorhina toe; T1D — transverse diameter of disc of first toe; F3L — length of 3" finger: from tip of 3 finger to proximal edge of palm; F3D — transverse diameter of disc of 3" finger; F1D — transverse diameter of disc of 1* finger; HL — head length, from tip of snout to posterior margin of tympanum; HW — head width, taken across the tympana; SL — snout length, from an imaginary line connecting the centres of the eyes to tip of the snout; EST — distance from anterior corner of orbital opening to tip of snout; END -— distance from anterior corner of orbital opening to centre of naris; IND — internarial distance between cen- tres of nares; ED — eye diameter, from anterior to poste- rior corner of orbital opening; TyD — horizontal diameter of tympanum. Measurements are presented as arithmetic means + standard deviation and range. Statistical cal- culations were done with the programme Statgraphics Centurion Version 15.2.14 (Statpoint Technologies, Inc., Warrenton, Virginia, USA). All p-values are calculated by the non-parametric Mann-Whitney (Wilcoxon) Test for comparison of medians. Osteological features were determined by superficial dissection. Sex was determined mainly by observations of call- ing in the field and/or the presence of vocal slits or testes (males) or absence of vocal slits and/or presence of eggs (females). Advertisement calls were recorded under natural conditions with a Sony WM-D6C Professional Walkman tape recorder, a Marantz PMD-661 or an Edi- rol RO9 digital recorder and a Sennheiser ME66 shotgun microphone and analysed with the sound-analysis pack- age Avisoft-SAS Lab Pro. Air temperatures adjacent to calling males were recorded using a rapid-reading digital thermometer. Terminology and acoustic analysis proce- dures mostly follow Kohler et al. (2017). All of the spe- cies described here produce calls in groups, which are separated from other groups by periods of silence that are much longer than the inter-call intervals and within which calls are repeated at regular intervals. As such, they meet the definition of a “call series” from Kohler et al. (2017). For all species, each call within a series comprises a single unpulsed note (so call = note); we use the term “call” in preference to “note” throughout to provide consistency. Measurements of call parame- ters are presented predominantly as range and mean + standard deviation. Colour of animals in life was described from digital photographs and of preserved specimens from direct ob- servations. Most colours were determined according to a colour matching system that is created and administrat- ed by the German RAL GmbH (RAL non-profit LLC). It should be stressed, however, that in many cases it was impossible to find an exact match between observed col- ours and RAL colour numbers. In those cases, the most similar RAL number was chosen. Specimens are stored in the South Australian Muse- um, Adelaide, Australia (SAMA) and the Museum fur Naturkunde, Berlin, Germany (ZMB). Paratypes for most species will also be repatriated to the Papua New Guinea National Museum, Port Moresby, Papua New Zoosyst. Evol. 97 (2) 2021, 355-382 Guinea (PNGNM). Abbreviations for other institutions mentioned are: American Museum of Natural History, New York, U.S.A. (AMNH); Bernice P. Bishop Muse- um, Hawaii, U.S.A. (BPBM); Institut Royal des Sciences Naturales de Belgique, Brussels (IRSNB); Museo Civi- co di Storia Naturale di Genova, Genoa, Italy (MSNG); Museum of Comparative Zoology, Harvard, U.S.A. (MCZ); Museum Zoologicum Bogoriense, Cibinong, Indonesia (MZB); National Museum of Natural History, now Naturalis Biodiversity Center, Leiden, The Nether- lands (RMNH); University of Papua New Guinea, Port Moresby (UP); Zoological Museum Amsterdam, now Naturalis Biodiversity Center, Leiden, The Netherlands (ZMA). SJR refers to the original field collection tag of Stephen Richards. Specimens examined for comparative purposes are listed in Appendix 1. Additional morphometric and oth- er data were extracted from original species descriptions and/or recompiled treatises, particularly Zweifel (1972), Blum and Menzies (1989), Kraus and Allison (2002) and Menzies (2006). Systematics Specimens were assigned to the genus Xenorhina on the basis of the following combination of features: jaw symphygnathine; clavicles and procoracoids absent; each vomeropalatine bone with elongated odontoid spike; body squat, head small, triangular, with small eyes; cuta- neous tubercles present dorsolaterally, absent on eyelids; tips of toes 2-5 expanded, with circum-marginal grooves; life style subterrestrial. Xenorhina lacrimosa sp. nov. http://zoobank.org/D78F9340-1032-4D34-976A-B91C7001CC1C Holotype. SAMA R71648 (SJR 14203), adult male, from Rentoul River, Western Province, Papua New Guinea (6.4355°S, 142.5615°E; 380 ma.s.l.), collected on 10-08- 2014 by S.J. Richards. Paratypes. SAMA R71647 (SJR10389), female with ripe eggs, ZMB 91129 (SJR10417) male, Camp 2, up- per Strickland River basin, Western Province, Papua New Guinea (5.9018°S, 142.4360°E; 950 m a.s.1.), col- lected by S.J. Richards on 18-02-2008 and 20-02-2008, respectively, ZMB 91130 (SJR10466) male, Camp 1, upper Strickland River basin, Western Province, Papua New Guinea (5.8078°S, 142.3083°E; 215 m a.s.l.), col- lected by S.J. Richards on 26-03-2008; SAMA R65069 (SJR10902) and R65070 (SJR10949), males, R65071 (SJR10963), (subadult?) female with scarcely developed eggs and R65072 (SJR10985), juvenile, Gugusu Camp, Muller Range, Western Province (5.7290°S, 142.2630°E; 515 mas.l.), all collected by S.J. Richards and C. Dahl between 7—9-09-2009. 357 Referred specimens. SAMA _ R71649, R71650 (SJR2577, 2582), PNGNM (SJR2571), adult males, He- rowana, Eastern Highlands Province, Papua New Guinea (6.6220°S, 145.1962°E; 1,400 m a.s.l.), collected by S.J. Richards between 20 and 24-11-2001. Diagnosis. This species of Xenorhina is characterised by the unique combination of: medium size (SUL of five males 34.5-41.0 mm); vomeropalatines each with one long and acuminate spike; legs moderately long (TL/SUL 0.42-0.46); all fingers without and all toes with expanded discs; eye-naris distance greater than internarial distance (END/IND 1.18—1.48); tympanum diameter smaller than or equal to that of eye (TyD/ED 0.75—1.00); dorsal sur- faces in life different tones of brown or blue or a mix- ture of these colours; ventral surfaces different tones of orange with irregular whitish spots or mouse grey (RAL 7005) with whitish spots and reticulations; advertisement calls uttered in series containing 7-12 single, mournful “hoots” separated by long intervals of about five seconds. Description of the holotype. Measurements are sum- marised in Table 1, a dorsolateral view in life 1s shown in Fig. la and ventral surfaces in life in Fig. 1b. Head broader than long (HL/HW 0.74); snout acuminate from above and below, distinctly protruding in profile; tongue very broad, only its lateral and posterior edges free; pre- pharyngeal ridge with five roundish denticles; left vomer- ine spike very well developed, right spike present, but malformed; loreal region oblique, no canthus rostralis; nostrils near tip of snout, directed more lateral than dor- sal, visible from above, but not from below; eye-naris dis- tance greater than internarial distance (END/IND 1.18); tympanum visible in life and preservative, its diameter slightly less than that of eye (TyD/ED 0.87); supratym- panic fold weakly expressed, extending from behind eye to insertion of fore leg; shank short (TL/SUL 0.42); fingers moderately short, not webbed; tips of all fingers with circum-marginal grooves, all not wider than penul- timate phalanges; relative lengths of fingers 3 > 4 = 1 = 2 (Fig. Ic); all toe tips with circum-marginal grooves and wider than penultimate phalanges; toes not webbed, relative lengths 4 >3 >5 >2> 1 (Fig. 1d); plantar and palmar tubercles (with exception of prominent, oval in- ner metatarsal tubercle; Fig. 1d), as well as subarticular tubercles scarcely visible. Body laterally with numerous distinct tubercles in life, less prominent in preservative; dorsal surfaces of limbs and middle of dorsum with fewer tubercles, all ventral surfaces smooth; tip of snout with several tiny elevations (especially on underside). In life, dorsal surfaces of head and anterior portion of body and fore limbs, uniform bluish-brown; remaining dorsal surfaces and flanks a mixture of saffron-yellow (RAL 1017) and blue-grey; tubercles with brown bases and whitish apices concentrated on flanks; body dorsal- ly with light yellow mid-dorsal line that continues on to hind legs; lumbar region with light yellow semi-circular spot (Fig. la); vent and adjacent region enclosed within zse.pensoft.net 358 Gunther, R. & Richards, S.: Six new species of Xenorhina Figure 1. Holotype (SAMA R71648) of Xenorhina lacrimosa sp. of right hand; (d) Thenar view of right foot. dark brown triangular patch; iris blackish with barely visible golden reticulation; plantar and ventral surfaces of toes predominantly brown, palms and ventral surfaces of fingers predominantly grey and cream; throat, chest, abdomen and ventral surfaces of extremities deep orange (RAL 2011), with some whitish spots (Fig. 1b). In preservative, dorsal surfaces of head, anterior back and fore limbs signal brown (RAL 8002); other dorsal surfaces ivory with diffuse brownish smears, tubercles with terra brown (RAL 8028) bases and whitish apices; ventral surfaces light ivory (RAL 1015); ivory lumbar spot on left side more clearly pronounced than on right. zse.pensoft.net nov. in life: (a) Dorsolateral view; (b) Ventral view; (c) Volar view Morphological variation. Measurements and propor- tions of most paratypes show limited variation (Table 1). An exception is a juvenile (SAMA R65072) measuring 16.6 mm SUL that exhibits some major deviations in pro- portions from the remainder of the type series. As these differences are almost certainly due to allometry, meas- urements of this specimen are disregarded in Table 1. Males and females have the same body size, although some ratios of the adult female (SAMA R71647 differ to a negligible degree (Table 1). The smallest specimen in the series is the just-mentioned adult female with an SUL of “only” 34.3 mm that contains ripe ovarian eggs; Zoosyst. Evol. 97 (2) 2021, 355-382 the largest specimen in the series is a male (ZMB 91130) with an SUL of 41.0 mm. Colour of paratypes in life varies considerably. Dorsal surfaces may be uniform blue-brown (SAMA R65070, Fig. 2a), uniform light red-orange similar to RAL 2008 (ZMB 91129, Fig. 2b), bluish on head and lower flanks, but reddish-brown on back and dorsal extremities (ZMB 91130, Fig. 2c) or dark brown with bluish hue on head, body and thighs, but beige on fore limbs, shanks and tar- si (SAMA R71647, Fig. 2d). Colour of ventral surfac- es 1s also highly variable. Some specimens are uniform deep orange or traffic orange (RAL 2009) interspersed with scattered irregular whitish spots (SAMA R65070, Fig. 2e); others are more extensively spotted (ZMB 91129, Fig. 2f) or exhibit a mixture of whitish, orange and brown spots, but with throat and thighs more or less uniform traffic orange (SAMA R65071, Fig. 2g); others exhibit grey-brown ground colour with many irregular whitish spots, some of them interspersed with small ir- regular red patches (SAMA R71647, Fig. 2h) . In preservative dorsal surfaces of three specimens predominantly violet, of two specimens copper brown, 359 of one specimen beige and of the juvenile specimen beige-brown; ventral surfaces of three specimens al- most completely light ivory, of the four other specimens a light ivory ground colour with a brown-beige pattern of various extent. All paratypes, except SAMA R71647, have a light ivory mid-dorsal line and all specimens in- cluding the juvenile have a greyish snout tip. None of the paratypes has a clearly pronounced lumbar spot in life or in preservative. Distribution and ecological notes. Most records of Xenorhina lacrimosa sp. nov. are from lowland and foot- hill forest in south-central Papua New Guinea (Fig. 8), where this species appears to have a broad distribution at altitudes ranging from near sea level around Kopi to at least 950 m a.s.l. We also refer several specimens from Herowana Village at 1,400 m as.l. (the most easterly location in Fig. 8) to this species pending confirmation of genetic relationships. Males called at night, normally after rain, either from within the leaf litter on the forest floor or down to several centimetres depth in the humus layer beneath the litter. Table 1. Body measurements and body ratios of the type series of Xenorhina lacrimosa sp. nov. SAMA R71648 is the male ho- lotype, others are paratypes. All measurements in mm; for explanation of abbreviations see “Material and methods”; M = male, F = female, sa = subadult. Reg.-No. SAMAR71647 ZMB 91129 ZMB 91130 SAMA R71648 SAMA R65069 SAMA R65070 SAMA R65071 Mean + SD Sex F M M M M M sa F SUL 34.3 36.9 41.0 36.2 37.3 34.5 3931 36.47 + 2.30 Apis 15.0 17.0 18.6 15.3 15.8 15:3 15.4 16.06 + 1.30 TaL 10.0 11.8 12.6 10.2 11.0 10.3 9:9 10.83 + 1.03 T4L 16.9 17.6 19.1 15.2 16.9 eb 21 15.5 16.61 + 1.46 T4D ded 1.6 as) 1.3 1.5 1.4 1.3 1.41 40.12 T1D 0.8 1.0 1.0 0.9 1.0 0.9 0.9 0.93 + 0.076 F3L 6.3 8.1 9.1 7.0 7.8 6.7 7.2 746 + 0.82 F3D 0.8 1.0 Apel 0.8 0.8 0.9 0.9 0.93 + 0.076 F1D 0.7 0.9 1.0 0.6 0.7 0.8 0.7 0.77 + 0.14 HL 9.5 10.4 12.0 9.7 11.2 10.1 10.3 10.46 + 0.87 HW 11.9 14.6 16.0 13 1 L2:f 11.4 13.1 13.26 + 1.58 END Yeu) 3.0 3.6 2.6 3.5 3.4 3.0 3.09 + 0.43 IND 1.7 2.3 2.6 2.2 2.4 2.3 2.2 2.24 + 0.28 SL 4.0 4.3 oa 4.2 5.0 4.7 4.7 4.57 + 0.42 EST 3.6 4.1 5.0 3:5 4.6 4.7 4.2 4.24 + 0.56 ED 2.4 2.7 2.8 2.3 2.4 2.2 Pasa 2.41 40.25 TyD 2.4 2.5 ZL 2.0 2.1 1.8 1.9 2.11 40.25 TL/SUL 0.44 0.46 0.45 0.42 0.42 0.44 0.44 0.44 + 0.015 TaL/SUL 0.29 0.32 0.31 0.28 0.29 0.30 0.28 0.30 + 0.015 T4L/SUL 0.49 0.48 0.47 0.42 0.45 0.44 0.44 0.46 + 0.025 T4D/SUL 0.038 0.043 0.037 0.036 0.040 0.041 0.037 0.039 + 0.003 T1D/SUL 0.023 0.027 0.024 0.025 0.027 0.026 0.026 0.025 + 0.002 F3L/SUL 0.184 0.220 0.222 0.193 0.209 0.194 0.205 0.204 + 0.014 F3D/SUL 0.023 0.027 0.027 0.022 0.021 0.026 0.026 0.025 + 0.003 FID/SUL 0.020 0.024 0.024 0.017 0.019 0.023 0.020 0.021 + 0.003 T4D/F3D 1.63 1.60 1.36 1.63 1.88 1.56 1.44 1.59 + 0.165 T1D/F1D 1.14 1.11 1.00 1.50 1.43 1.13 1.29 1.23 + 0.183 HL/SUL 0.28 0.28 0.29 0.27 0.30 0.29 0.29 0.29 + 0.009 HW/SUL 0.35 0.40 0.39 0.36 0.34 0.33 0.37 0.36 + 0.026 HL/HW 0.80 0.71 0.75 0.74 0.88 0.89 0.79 0.79 + 0.069 END/SUL 0.073 0.081 0.088 0.072 0.094 0.099 0.085 0.085 + 0.010 IND/SUL 0.050 0.062 0.063 0.061 0.064 0.067 0.063 0.061 + 0.005 END/IND 1.47 1.30 1.38 1.18 1.46 1.48 1.36 1.38 + 0.109 ED/SUL 0.070 0.073 0.068 0.064 0.064 0.064 0.060 0.066 + 0.004 TyD/SUL 0.073 0.068 0.051 0.055 0.056 0.052 0.054 0.058 + 0.009 TyD/ED 1.00 0.93 0.75 0.87 0.88 0.82 0.90 0.88 + 0.079 SL/SUL 0.117 0.117 0.124 0.116 0.134 0.136 0.134 0.125 + 0.009 EST/SUL 0.105 0.111 0.122 0.097 0.123 0.136 0.120 0.116 + 0.013 zse.pensoft.net 360 Gunther, R. & Richards, S.: Six new species of Xenorhina Figure 2. Xenorhina lacrimosa sp. nov. paratypes in life. Dorsolateral views: (a) SAMA R65070; (b) ZMB 91129; (c) ZMB 91130; (d) SAMA R71647. Ventral views: (e) SAMA R65070S; (f) ZMB 91129; (g) SAMA R65071; (h) SAMA R71647. zse.pensoft.net Zoosyst. Evol. 97 (2) 2021, 355-382 Vocalisation. One call series from SAMA R71648 (holotype), one from ZMB 91129 (paratype), one from SAMA R65069 (paratype) and four from ZMB 91130 (paratype) recorded at air temperatures of 21.2—25.5 °C were analysed. Each call is a single, unpulsed mournful note that is always produced in a series, during which both volume and pitch increase gradually between first and last call (Fig. 3a). Despite some variation in recording temper- ature, all recorded calls are extremely similar, so they were pooled for analyses. Call series last 26.4—60.4 s (mean 40.0 + 11.8 s, n= 7), with 7-12 calls/series (mean 9.0 + 2.2, n= 7) produced at a rate of 0.20—0.27 calls/s (mean Relative amplitude oe Frequency (kHz) oa Led i 0.2 a4 Time (33 361 0.23 + 0.02). Call length is 141-231 ms (mean 193.5 + 19.1 ms, n= 63) and first and last call in a series are often the shortest; inter-call interval length is 2.8—8.0 s. (mean 4.8+1.0s,n=56). Calls start abruptly at high amplitude that rises quickly to a maximum, then decreases gradual- ly until termination of call (Fig. 3b). All calls have 5—7 harmonics (Fig. 3c and 3d)). Dominant frequency may be carried by a second harmonic (..e. first two calls of series from holotype, with peak at 1.2 kHz) or by first harmon- ic (all other calls, with peaks increasing from 0.7 kHz in third call to 0.8 kHz in last call in the series. Frequency de- clines at end of each call in second half of series (Fig. 3c). Relative amplitude eg 2.0 4.0 40 Frequency (kHz) Figure 3. (a) Oscillogram of a complete advertisement call series with seven calls from the holotype of Xenorhina lacrimosa sp. nov.; (b) oscillogram and (c) spectrogram of the penultimate call of the call series shown on Fig. 3a; (d) amplitude spectrum of the call on (b.) and (c.). zse.pensoft.net 362 Etymology. The specific epithet /acrimosa is a Latin adjective in female gender; translated literally 1t means “tearful”, but it is also translated as “lamentable voice” and refers to the mournful sounding advertisement call of the new species. Comparisons with other species. We compare Xenorhina lacrimosa sp. nov. with all congeners of a similar size (SUL 30-43 mm) that have a single spike on each vomeropalatine bone. Xenorhina fuscigula (Blum & Menzies, 1989) has hind legs shorter (TL/SVL < 0.40 vs. > 0.40 in Xenorhi- na lacrimosa sp. nov.), eye-naris distance shorter (END/ SVL 0.064—0.074 vs. 0.072—0.099), inner metatarsal tu- bercle absent (vs. present), ventral surfaces black (vs. or- ange-red or grey-brown) and call consisting of a single long note (vs. a series of 7-12 notes = calls). Xenorhina huon (Blum & Menzies, 1989) is small- er (SUL to 32 mm vs. 34.3-41.0 mm), with hind legs shorter (TL/SUL < 0.40 vs. > 0.40), internarial distance greater (0.064—0.081 vs. 0.050—0.067), eyes larger (ED/ SVL 0.070-0.091 vs. 0.060—0.073) and ventral surfaces with dark flecking (vs. ventral surfaces with no or sparse brownish reticulation). Xenorhina subcrocea (Menzies & Tyler, 1977) is smaller (SUL 30.5-33.3 mm vs. 34.3-41.0 mm), with hind legs longer (TL/SVL > 0.46 vs. < 0.46 in Xenorhina lacrimosa sp. nov.) and ventral surfaces with dark reticu- lation (vs. without dark reticulation); call length is shorter 64-69 ms (vs. 141—231 ms), with inter-call interval also much shorter (154-285 ms vs. 2.8—8.0 s). Xenorhina zweifeli (Kraus & Allison, 2002) is about the same size and has similar body ratios. It differs by having a conspicuous dark brown supratympanic stripe (vs. absent in Xenorhina lacrimosa sp. nov.) and in several aspects of its advertisement calls. Xenorhina zweifeli utters single calls at irregular intervals, with two or three calls sometimes produced in quick succession (Kraus and Allison 2002), during both day and early evening. In contrast, Xenorhina lacrimosa sp. nov. always produces calls in discrete series of at least seven relatively evenly spaced calls of increas- ing pitch and volume; calls are never produced in quick succession and males always call at night. Other differenc- es include: mean length of calls produced by holotype of X. zweifeli is 310 ms (Kraus and Allison 2002) (vs. mean length of calls from Xenorhina lacrimosa sp. nov. 194 ms); the fundamental frequency of zweifeli calls is at 610 Hz and dominant frequency at 1910 Hz (third harmonic), (vs. fun- damental and dominant frequency of Xenorhina lacrimosa sp. nov., both at 800 Hz); amplitude of X. zweifeli calls rises more slowly than that of /acrimosa calls and all harmonics are frequency modulated, with pitch decreasing during en- tire length of call (vs. frequency modulation only occurring at end of harmonics in Xenorhina lacrimosa sp. nov. calls). Moreover, X. zweifeli occurs only on two mountain ranges in northern Papua New Guinea, while Xenorhina lacrimo- Sa Sp. nov. lives predominantly in the lowlands and foot- hills of southern Papua New Guinea. Therefore, the known zse.pensoft.net Gunther, R. & Richards, S.: Six new species of Xenorhina distributions of the two species are separated by a major biogeographic barrier, New Guinea’s central cordillera. Xenorhina perexigua sp. nov. http://zoobank.org/96A FDF65-2D60-4245-AB15-BDA2728A6A88 Holotype. SAMA R71645 (SJR 10418), adult male, from Camp 2, upper Strickland River basin, Western Province, Papua New Guinea (5.9018°S, 142.4360°E; 950 ma.s.1.), collected by S.J. Richards on 20-02-2008 Diagnosis. This species of Xenorhina is character- ised by the unique combination of: very small body size (SUL of the only adult male 16.7 mm); vomeropalatines each with a single triangular spike; legs moderately long (TL/SUL 0.46); all fingers and first toe without and toes 2—5 with expanded discs; eye-naris distance greater than internarial distance (END/IND 1.27); tympanum small- er than eye (TyD/ED 0.77); dorsal surfaces in life beige brown (RAL 8024) with darker areas on upper flanks, in middle of back and on neck; lower flanks with whitish spots and reticulations and some irregular dark brown flecks; supratympanic area with dark brown fleck; ventral surfaces off-white with extensive blackish-brown reticu- lation. Advertisement calls in series containing about 30 soft “popping” calls of 30-40 ms duration, produced at a rate of 6.8-6.9 calls/s. Description of the holotype. Measurements and ra- tios are presented in Table 2. Body squat (Fig. 4a and b), head broader than long (HL/HW 0.84); snout strongly acuminate from above and below and protruding in pro- file; tongue broad, only its lateral and posterior edges free; prepharyngeal ridge without denticles; a single tri- Table 2. Body measurements and body ratios of the male ho- lotype (SAMA R71645) of Xenorhina perexigua sp. nov.. All measurements in mm; for explanation of abbreviations see “Material and methods”. Reg.-No. SAMA R71645 Reg.-No. SAMA R71645 SUL 16.7 TL/SUL 0.46 TL 7.6 TaL/SUL 0.30 TaL 5.0 TAL/SUL 0.45 TAL 75 TAD/SUL 0.036 TAD 0.6 T1D/SUL 0.018 T1D 0.3 F3L/SUL 0.186 F3L 3.1 F3D/SUL 0.021 F3D 0.35 F1D/SUL 0.012 F1D 0.2 T4D/F3D ewAl HL 46 T1D/F1D 1.50 HW 5.5 HL/SUL 0.28 END 1.4 HW/SUL 0.33 IND 1.1 HL/HW 0.84 SL 20 END/SUL 0.084 EST 2.0 IND/SUL 0.066 ED fe END/IND yy TyD 1.0 ED/SUL 0.078 TyD/SUL 0.060 TyD/ED 0.77 SL/SUL 0.132 EST/SUL 0.120 Zoosyst. Evol. 97 (2) 2021, 355-382 363 Figure 4. Holotype (SAMA R71645) of Xenorhina perexigua sp. nov. in life: (a) Dorsolateral view; (b) Ventral view; (c) Volar view of right hand; (d) Thenar view of right foot. angular spike of moderate size on each vomeropalatine; loreal region oblique, no canthus rostralis; nostrils near tip of snout, directed dorsolaterally, visible from above, but not from below; eye-naris distance greater than internari- al distance (END/IND 1.27); greater part of tympanum visible in life and preservative, its diameter smaller than that of eye (TyD/ED 0.77); supratympanic fold short, not contacting posterior edge of eye and not reaching insertion of fore leg; shank of moderate length (TL/SUL 0.46); fin- gers moderately short, not webbed; tips of fingers not wid- er than penultimate phalanges, all with circum-marginal grooves that extend along entire length of digits, relative lengths of fingers 3 > 4 > 2 = 1 (Fig. 4c); all toe tips with circum-marginal grooves, all tips, except that of toe 1 wid- er than penultimate phalanges; toes not webbed, relative lengths 4 >3 >5 >2> 1 (Fig. 4d); plantar and palmar tu- bercles (with exception of small, but prominent inner met- atarsal tubercle), as well as subarticular tubercles scarcely visible. Dorsal surfaces with only a few tubercles and a raised mid-dorsal ridge, ventral surfaces smooth (Fig. 4b). zse.pensoft.net 364 — = 7 eo - NOUR La m1 , q ‘ ‘ pit ‘adi a = i J O23 o.4 0.6 O28 Time (3) Gunther, R. & Richards, S.: Six new species of Xenorhina Relative am plitude Lo ic Ma te Se eS Frequency (kHz) Figure 5. (a) Oscillogram and (b) Spectrogram of the last six calls of a call series containing 31 calls from the holotype of Xenorhina perexigua sp. nov.; (c) Amplitude spectrum of an advertisement call from the holotype of X. perexigua sp. nov. In life, dorsal surfaces beige brown with darker areas on upper flanks, in middle of back and in scapular region (Fig. 4a); lower flanks with whitish spots and whitish re- ticulations and three (left side of body) or four (right side of body) irregular dark brown flecks which merge with dark brown reticulum of abdomen; supratympanic area with conspicuous dark brown fleck; all ventral surfaces light grey with dense dark brown reticulations (Fig. 4b); lumbar spot absent; vent and adjacent areas of thighs en- closed in large, triangular dark brown patch, borders of which disintegrate ventrally; outer margin of iris black- ish, inner margin golden. In preservative, dorsal surfaces reddish-brown, flanks with dark irregular spots and supratympanic region with large, dark brown fleck; ventral surfaces ivory-white with brown beige (RAL 1011) reticulum; large ivory-white area between eye and insertion of fore-leg present (not evident in life). Distribution and ecological notes. Xenorhina perexi- gua sp. nov. is known only from one locality, in hill forest at an altitude of 950 m a.s.|. in the upper Strickland River basin of south-western Papua New Guinea (Fig. 8). The holotype was calling from within leaf litter on the forest floor at night during rain. Vocalisation. Two call series, produced by the holo- type (SAMA R71645) at an air temperature of 21.2 °C, were analysed. Each call is a single soft, unpulsed “pop” note uttered in rapid succession (Fig. 5a). The two se- ries lasted 4.1 s and 4.5 s and contained 28 and 31 calls produced at a rate of 6.8 and 6.9 calls/s. Call length 29- 42 ms (mean 34.6 + 3.6 ms, n= 59) and inter-call interval 101-195 ms (mean 115.0 + 17.1 ms, n= 57). While calls are of approximately equal length throughout each series, inter-call intervals are slightly longer at the end of call series than at the beginning. Volume of calls increases during course of series, as is typical for many Xenorhina species. Calls start abruptly at high amplitude, which then zse.pensoft.net decreases at an irregular rate until the end of each call (Fig. 5a). The start of each call also has a broad frequen- cy range that drops rapidly to a more narrowly defined, frequency-modulated band (Fig. 5b). Fundamental and dominant frequency peak at 1.4 kHz (Fig. 5c). Etymology. The specific epithet perexigua is a Lat- in adjective of feminine gender, meaning very small (translation of perexiguus, -a, -um in the Dictionarium latino-germanicum means “sehr klein”) and refers to the diminutive size of the new species. Comparisons with other species. Although this spe- cles is represented by only a single specimen, it is an adult male of very small size (16.7 SUL mm) and, given knowledge about the size ranges of congeners, its SUL is unlikely to exceed 25 mm. We, therefore, compare Xenorhina perexigua sp. nov. with all congeners of a similar size (SUL 15—25 mm) that have a single spike on each vomeropalatine. Xenorhina anorbis (Blum & Menzies, 1989) is larger (holotype is an adult male with SVL of 21.3 mm [range of type series 21.3—23.4 mm but sex of other specimens not specified] vs. SUL 16.7 mm in one male), has hind legs shorter (TL/SVL < 0.38 vs. > 0.38) and discs of fingers and toes not wider than penultimate phalanges (vs. discs on toes 2—5 clearly wider than penultimate phalanges in Xenorhina perexigua sp. nov.). Xenorhina brachyrhyncha Kraus, 2011 appears to be larg- er (two adult females with SVL 21.2 and 22.8 mm vs. SUL 16.7 mm tn one male), with snout blunt in dorsal and ventral view (vs. strongly acuminate), head wider and longer (HW/ SVL 0.35—0.38 vs. 0.32 and HL/SVL 0.30-0.32 vs. 0.28) with much lower ratio of eye-naris distance to internarial distance (END/IND 1.06—1.13 in _X. brachyrhyncha vs. 1.27 in Xenorhina perexigua sp. nov.); differences in colour in- clude lack of a dark supratympanic spot in_X. brachyrhyncha (vs. present in Xenorhina perexigua sp. nov.) and less pro- nounced dark reticulation on all ventral surfaces. Zoosyst. Evol. 97 (2) 2021, 355-382 Xenorhina lanthanites is larger SUL 21.3—22.4 mm vs. SUL 16.7 mm), with tips of toes wider than penulti- mate phalanges only on 4" toe (vs. toes 2—5 with expand- ed terminal discs); ratio of END/IND lower (0.94—1.20 vs. 1.27); and advertisement call series much longer, lasting up to more than one minute (vs. < 5s) with aver- age call length of 121 ms (vs. 35 ms in Xenorhina per- exigua sp. nov.), dominant frequency of about 1.0 kHz (vs. 1.4 kHz) and call repetition rate of 1—2 calls/s (vs. 6.8—6.9 calls/s). Although it is known from just one specimen, it 1s an adult male suggesting that Xenorhina perexigua sp. nov. is amongst the smallest known members of the genus. Only one other species, X. bouwensi, may be smaller than Xenorhina perexigua sp. nov., but it can be immediately distinguished from the new species by its lacking odon- toid spikes on the vomeropalatines. Xenorhina pohleorum sp. nov. http://zoobank.org/91F9054A-3CF0-4672-8A 86-F64F6A2BA7AA Holotype. SAMA R71644 (SJR 14202), adult male, from Rentoul River, Western Province, Papua New Guinea (6.4355°S, 142.5615°E; 380 maz.s.l.), collected on 11-08- 2014 by S.J. Richards and K. Aplin. Paratype. SAMA R60217 (SJR 3223), adult male, Darai Plateau, Gulf Province, Papua New Guinea (7.1295°S, 143.6134°E; 435 m a.s.l.), collected on 1-08- 2003 by S.J. Richards. Diagnosis. This species of Xenorhina is characterised by the unique combination of: small size (SUL of two adult males 20.3 and 21.2 mm); vomeropalatines each with a single moderately developed triangular vomerine spike; legs of medium length (TL/SUL 0.44 in both speci- mens); all fingers without and all toes with, expanded ter- minal discs; tips of all fingers and toes with circum-mar- ginal grooves, all grooves extending at least partly along digits; head short (HL/SUL 0.26 in both specimens); eye-naris distance greater than internarial distance (END/ IND 1.33 in both specimens); dorsal surfaces in life brown-beige (RAL 1011) or grey-brown; ventral surfaces ivory-white with extensive pale brown (RAL 8025) retic- ulation; mid-dorsal line and lumbar spots absent; adver- tisement calls uttered in series lasting 4-9 s, containing 10-30 “piping” calls, each 56—93 ms duration with repe- tition rate of 2.5—3.6 calls/s. Description of the holotype. Measurements are sum- marised in Table 3. Body squat (Fig. 6a and b), head broad- er than long (HL/HW 0.83); snout short (HL/SUL 0.26), strongly acuminate from above and below, protruding in profile; tongue broad, only its lateral edges and posterior lobes free; prepharyngeal ridge with few tiny denticles; vomerine spikes triangular and of moderate size; lore- al region oblique, canthus rostralis absent; nostrils near 365 Table 3. Body measurements and body ratios of the male holo- type of Xenorhina pohleorum sp. nov. (SAMA R71644) and the male paratype (SAMA R60217). All measurements in mm; for explanation of abbreviations see “Material and methods”. Reg.-No. SAMA R71644 SAMA R60217 Mean SUL 20.3 21.2 20.75 TL 9.0 9.4 9.20 TaL 59 bef 5.80 T4L 9.0 O/, 9.35 T4D 0.7 0.7 0.70 T1D 0.4 0.4 0.40 F3L 3.6 3.8 3.70 F3D 0.4 0.4 0.40 F1D 0.3 0.3 0.30 HL 5:3 5:5 5.40 HW 6.4 6.5 6.45 END 1.6 1.7 1.65 IND 1.2 1.3 1.25 SL 2.6 3.0 2.80 EST 2.2 2.5 2.40 ED 1.3 1.2 1.25 TyD 1.2 0.9 1.05 TL/SUL 0.44 0.44 0.44 TaL/SUL 0.29 0.27 0.28 T4L/SUL 0.44 0.46 0.45 T4D/SUL 0.035 0.033 0.034 T1D/SUL 0.020 0.019 0.020 F3L/SUL 0.178 0.179 0.179 F3D/SUL 0.020 0.019 0.020 F1ID/SUL 0.015 0.017 0.016 T4D/F3D 1.75 1.75 175 T1D/F1D 1:33 1:33 1733 HL/SUL 0.26 0.26 0.26 HW/SUL 0.32 0.31 0.32 HL/HW 0.83 0.85 0.84 END/SUL 0.079 0.080 0.080 IND/SUL 0.059 0.061 0.060 END/IND In 33 1.31 1.32 ED/SUL 0.064 0.057 0.061 TyD/SUL 0.059 0.042 0.051 TyD/ED 0.92 0.75 0.84 SL/SUL 0.129 0.142 0.136 EST/SUL 0.109 0.118 0.114 tip of snout, directed more laterally than dorsally, visi- ble from above, but not from below; eye-naris distance significantly greater than internarial distance (END/IND 1.33); tympanum nearly as large as eye (TyD/ED 0.92), supratympanic fold weakly expressed, not reaching eye or insertion of fore leg; shank moderately long (TL/SUL 0.44); fingers moderately short, not webbed, all fingers without and all toes with expanded terminal discs; cir- cum-marginal grooves on all fingers and all toes, extend- ing at least partly along most digits; head short (HL/SUL 0.26); eye-naris distance greater than internarial distance (END/IND 1.33); tympanum slightly larger than half the size of eye (TyD/ED 0.59); relative lengths of fingers 3 >4=2=1 (Fig. 6c); toes not webbed, relative lengths 4 >3>5>2> 1 (Fig. 6d); plantar and palmar tubercles, as Well as subarticular tubercles, not clearly demarcated, with the exception of small, but prominently raised in- ner metatarsal tubercle (Fig. 6d). Dorsolateral surfaces of body and dorsal surfaces of shanks with some tubercles, more conspicuous in life than 1n preservative; ventral sur- faces smooth; tip of snout lighter than surrounding skin, with some tiny depressions. zse.pensoft.net 366 Gunther, R. & Richards, S.: Six new species of Xenorhina Figure 6. Holotype (SAMA R71644) of X. pohleorum sp. nov. in life: (a) Dorsolateral view; (b) Ventral view; (c) Volar view of right hand; (d) Thenar view of right foot. In life, dorsal surfaces brown beige (RAL 1011); lum- bar spots and mid-dorsal line absent; tubercles with whit- ish apices concentrated on upper flanks; lower flanks, lateral surfaces of head and anterior hind limbs off-white with conspicuous fawn (RAL 8007) reticulum; snout tip window grey (RAL 7040); iris blackish with few gold- en specks (Fig. 6a); ventral surfaces pearl-white (RAL 1013) with dusky pink (RAL 3014) reticulum and irreg- ular pearl-white spots; throat dusky pink with only a few whitish spots (Fig. 6b). In preservative, ground colour of dorsal surfaces of head, back and hind limbs fawn brown (RAL 8007) zse.pensoft.net with some inconspicuous darker areas; head less dense- ly pigmented than adjacent neck; ground colour of dorsal surfaces of fore limbs and anterior hind limbs beige (RAL 1001) with conspicuous terra-brown strikes and reticula; rear of thighs predominantly terra-brown with a few whitish spots below and small blackish area around vent; ventral surfaces fawn-brown with con- spicuous pearl-white spots; throat and middle of chest least spotted. Morphological variation. Measurements and body ratios of paratype are similar to holotype (Table 3). Dorsal Zoosyst. Evol. 97 (2) 2021, 355-382 surfaces more tubercular in life (Fig. 7), but in preserva- tive, lateral surfaces with fewer tubercles; colour of dor- sal surfaces in life a mixture of indistinct lighter and dark- er grey-brown flecks, lower lateral surfaces of body and upper arms beige-brown (RAL 8024) with off-white spots and ventral surfaces beige-brown with off-white spots. Dorsal surfaces in preservative beige with signal brown (RAL 8002) spots, stripes and reticula; ventral surfaces in preservative paler than holotype, light ivory (RAL 1015) with scarcely visible brownish network. Figure 7. Paratype SAMA R60217 of X. pohleorum sp. nov. in life. Papua New Guinea 367 Distribution and ecological notes. Xenorhina pohle- orum sp. nov. is known from two localities approximate- ly 140 km apart in the lowland rainforests of Gulf and Western Provinces in south-central Papua New Guinea (Fig. 8), where males called from under the litter or with- in the humus layer, at night during rain. Vocalisation. Advertisement call is a single short, un- pulsed and melodic “piping” note and is always uttered in series. Call length and inter-call interval are variable, but call intervals are always much shorter than the inter- val between call series. Due to some differences in call features, we analysed five call series from the holotype (SAMA R71644) recorded at an air temperature of 24 °C separately from seven call series produced by the para- type (SAMA R60217) at an air temperature of 22 °C. Call series produced by the holotype last 3.6—8.8 s (mean 5.8 + 1.8 s) and contain 13—28 calls (mean 18.2 + 5.6) produced at a rate of 2.55—3.61 calls/s (mean 3.22 + 0.41, n = 5). Call length is 56-93 ms (mean 74.5 + 8.5 ms, n = 91) and length of call intervals is 139-528 ms (mean 253.4 + 71.7 ms, n= 86). Calls start abruptly at maximum or almost maximum amplitude which then decreases at an irregular rate until end of call (Fig. 9a). Fundamental and dominant frequencies are at 1.5 kHz and the only upper harmonic (at 3.0 kHz) has much less energy (Fig. 9b and ee ak Bee ae Port Moresby 4 Elevation (m) 4000 2000 0 Figure 8. Map of Papua New Guinea showing the known distributions of X. /acrimosa sp. nov. (blue circles), X. perexigua sp. nov. (yellow triangle) and_X. pohleorum sp. nov. (red squares). Arrows indicate the type localities. zse.pensoft.net 368 Gunther, R. & Richards, S.: Six new species of Xenorhina 3 = | : a 37 z= = a Sis pe, HS | 04 0.8 Tez 1.6 2.0) 24 2.6 ae 0 1.0 2.0) Sal] Time (3) Frequency (kH=) ll ky TEA ia tadiadat h PO Nae tama (yo \ a1 1 ha 1 | | i | 4 ‘pe fi - a " rh Wi | a "I hy i lee os " ‘fy, 4 i Nivaw " CEng Ts hn ame ts 1S a des BANCO He a ag Frequency (kHz) Le Time fs) Figure 9. (a) Oscillogram; (b) Spectrogram and (c) Amplitude spectrum of 10 consecutive advertisement calls from a longer series produced by the holotype of Xenorhina pohleorum sp. nov; (d) Oscillogram and (e) Spectrogram of six advertisement calls from the holotype of X. pohleorum sp. nov. (higher volume) are answered in exact antiphony by an unvouchered male (lower volume). 9c). Frequency of calls is weakly modulated with a slight increase over the duration of the call. A number of calls were uttered in exact antiphony with calls from an un- vouchered specimen (Fig. 9d and 9e). Calls of the paratype (SAMA R60217) are similar to those of the holotype, but call series generally contain fewer calls (10-15, mean 12.9 + 1.77, n = 7, vs. 13-28 mean 18.2; see above) and so are shorter (3.4—5.4 s, mean 46+0.72 s,n=7 vs. 3.6-8.8 s, mean 5.8 + 1.8 s), al- though there is some overlap. Calls of the paratype are also slightly longer (66-98 ms, mean 88.4 + 4.8 ms, n= 89 vs. 56-93 ms, mean 74.5). Other structural parameters of calls from the paratype fall within the range produced by the holotype: inter-call intervals 234-408 ms (mean zse.pensoft.net 290.0 + 31.1 ms, n = 83) and mean repetition rate 2.73- 3.0 calls/s (mean 2.83 + 0.10, range, n = 7). Calls of the holotype do not show the typical increase in volume and pitch that is typical of the series produced by the para- type. However, the holotype was calling within a group of closely adjacent males and exhibited antiphonal calling behaviour (Fig. 9d and 9e). It cannot be discounted that the slight differences noted between calls of holotype and paratype were a result of their different calling situations (alone vs. within a chorus). Etymology. The specific epithet poh/leorum is the Lat- inised patronymic adjective in genitive plural derived from the family name Pohle. It is to recognise a very Zoosyst. Evol. 97 (2) 2021, 355-382 long-lasting friendship of the senior author with Sybille and Claus Pohle from Berlin. Comparisons with other species. We compare Xenorhina pohleorum sp. nov. with all congeners of a similar size (SUL 18—25 mm) that have a single spike on each vomeropalatine. Xenorhina anorbis has hind legs shorter (TL/SVL < 0.38 vs. > 0.38) and fingers and toes without expanded terminal discs (vs. enlarged discs on all toes in Xenorhina pohleorum sp. nov.). Xenorhina brachyrhyncha has legs longer (TL/SVL 0.46—0.49 vs. twice 0.44), head longer (HL/SVL 0.30—0.32 vs. twice 0.26) and broader (HW/SVL 0.35-0.38 vs. 0.31— 0.32), with END/IND ratio lower (1.06—1.13 vs. 1.31—1.33). Xenorhina lanthanites has expanded disc only on 4" toe (vs. on all toes), head broader (HW/SVL 0.35—0.37 vs. 0.31—0.32), eyes larger (ED/SUL 0.071-0.081 vs. 0.057—0.064), END/IND ratio lower (0.94—1.20 vs. 1.31- 1.33) and advertisement call series much longer (up to more than one minute vs. less than 10 seconds). Xenorhina mehelyi appears to be much larger (SVL 20.7—35.2 mm vs. 20.3—21.2 mm); although the sex (or state of maturity) of previously reported specimens 1s unknown, with a male SUL of 20.3—21.2 mm, it is un- likely that Xenorhina pohleorum sp. nov. of either sex will approach the upper size limit reported for X. mehelyi. Xenorhina mehelyi also has eyes larger (ED/SVL 0.067— 0.079 vs. 0.057—0.064) and different advertisement calls. Mean call interval 1.5 s, (vs. 0.25 s) and mean call rate 0.60 calls/s (vs. 3.2 calls/s); calls are also longer (mean 140 ms vs. 74.5 ms) and have a much lower dominant fre- quency (0.88 kHz vs. 1.5 kHz) (Blum and Menzies 1989). Xenorhina perexigua is smaller than Xenorhina pohle- orum sp. nov. (males 16.7 mm vs. 20.3—21.2 mm SUL). Some body ratios also differ (Tables 2 and 3), but sample sizes are too small for robust comparisons. However, sub- stantial differences in advertisement calls support recogni- tion of Xenorhina pohleorum sp. nov. as a distinct species: calls of Xenorhina perexigua sp. nov. are shorter (29-42 ms vs. 56-93 ms), there are more calls/series (28-31 vs. 10-28 calls) and inter-call intervals are shorter (101-195 ms vs. 139-528 ms), so the call rate is twice as fast in Xenorhina perexigua sp. nov. (6.8-6.9 calls/s vs. 2.6—3.6 calls/s). The substantially greater call rate of Xenorhina perexigua sp. nov. (double that of Xenorhina pohleorum sp. nov.) cannot be attributed to differences in temperature because the re- cording temperature for the former was lower than that of latter, which should reduce, not increase, the call rate. Xenorhina schiefenhoeveli (Blum & Menzies, 1989) is larger (SVL 26.7—30.7 mm vs. 20.3—21.2 mm) and its call series lasts more than 100 s (vs. not more than 10 s in Xenorhina pohleorum sp. nov.), with call intervals of more than 700 ms (vs. < 528 ms). Xenorhina tumulus (Blum & Menzies, 1989) 1s larger (male SVL more than 26.0 mm vs. less than 22.0 mm), has ventral surfaces of toes with striped pattern (vs. ab- sent) and abdomen partly pink or red (vs. pearl-white 369 with dusky pink reticulum and irregular pearl-white spots); and supratympanic ridge is absent (vs. present). Advertisement calls of X. tumulus differ in, amongst oth- er characters, having a much lower dominant frequency (0.9 kHz vs. 1.5 kHz). Xenorhina tumulus is known only from an elevation of about 1500 m a.s.l. in the Adelbert Range, an isolated mountain range near the north coast of Papua New Guinea, while Xenorhina pohleorum sp. nov. 1s known only from altitudes of around 400 m on the southern side of New Guinea’s central cordillera. Xenorhina thiekeorum sp. nov. http://zoobank.org/772E7466-6C63-48E7-8CAE-96A | BESF78F5 Holotype. SAMA R71651 (SJR 209047), adult male, from Ok Menga near Tabubil, Western Province, Papua New Guinea (5.3205°S, 141.3049°E; 620 m a.s.l.), col- lected by S.J. Richards, M. Cunningham and A. Dennis on 14-11-1994. Paratypes. ZMB 91131 (SJR 209051), PNGNM (SJR209052), SAMA R71652 (SJR209053), same details as for holotype. Diagnosis. This species of Xenorhina is characterised by the unique combination of: moderately small size ( males 20.7—23.5 mm SUL); vomeropalatines each with one moderately developed triangular vomerine spike; legs moderately short (TL/SUL 0.40-0.44); all fingers and first toe without and toes 2—5 with, expanded termi- nal discs; tips of all fingers and toes with circum-margin- al grooves that extend, at least partially, along most dig- its; head short (HL/SUL 0.26—0.28), eye-naris distance much greater than internarial distance (END/IND 1.36- 1.54); tympanum approximately 2/3 size of eye (TyD/ ED 0.63—0.69). Dorsal surfaces in life reddish-brown, covered extensively with small, white-tipped tubercles, lower flanks with larger off-white spots; back with faint yellowish mid-dorsal line. Advertisement calls uttered in series containing less than 10 short, extremely soft “pip- ing” calls of 133-162 ms duration, produced at a rate of 2.5—3.0 calls/s. Description of the holotype. Measurements are sum- marised in Table 4, a dorsolateral view in life is shown in Fig. 10a and ventral surfaces in life in Fig. 10b. Head broader than long (HL/HW 0.81); snout short (HL/SUL 0.27), strongly acuminate from above and below, pro- truding in profile; tongue long, broad, only its lateral and posterior edges free; prepharyngeal ridge with eight small denticles; vomerine spikes triangular and of moderate size; loreal region oblique, no canthus rostralis; nostrils near tip of snout, directed dorsolaterally, visible from above, but not from below; eye-naris distance significant- ly greater than internarial distance (END/IND 1.46); bor- ders of tympanum poorly defined in life and preservative, its diameter 2/3 that of eye (TyD/ED 0.67); supratympanic zse.pensoft.net Gunther, R. & Richards, S.: Six new species of Xenorhina Figure 10. Holotype (SAMA R71651) of Xenorhina thiekeorum sp. nov. (a) Dorsolateral view in life; (b) Ventral view, (c) Volar view of right hand and (d) Thenar view of right foot of preserved specimen. fold not reaching posterior edge of eye or insertion of fore-leg; shank short (TL/SUL 0.40); fingers moderately short, not webbed; all fingers and first toe without and toes 2—5 with, expanded terminal discs; circum-marginal grooves on tips of all fingers and toes, extending at least partly along most digits; relative lengths of fingers 3 > 4 = 2 = 1 (Fig. 10c); toes not webbed, relative lengths 4 > 3 > 5 >2> 1 (Fig. 10d); plantar and palmar tubercles as well as subarticular tubercles poorly defined. Dorsal surfaces with scattered low tubercles and a slightly raised yellow- ish mid-dorsal ridge, ventral surfaces smooth. In life, dorsal surfaces brown beige (RAL 1011) with irregularly shaped, indistinct lighter markings in lumbar region and narrow, pale mid-dorsal line; dorsum with nu- zse.pensoft.net merous small, white-tipped tubercles; lower flanks and anterior and posterior of tympana with whitish spots; dor- sal surfaces of limbs and dorsal edge of tympana with few dark brown spots and/or streaks; iris blackish with scarce- ly visible golden veins and solid golden inner margin. Colour of ventral surfaces in life was not documented. In preservative, ground colour of dorsal surfaces reddish-brown; dorsolateral surfaces with conspicuous blackish-brown spots, mostly associated with white- tipped tubercles; extremities and anterior back with light- er brown flecks than those on dorsolateral surfaces; solid reddish-brown areas of back merge on lower flanks into ivory-white ground colour of ventral surfaces, which are covered by a dense orange-brown reticulum. Zoosyst. Evol. 97 (2) 2021, 355-382 Morphological variation. Measurements of the type series are summarised in Table 4. Ground colour of dorsal surfaces in preservative is the same in all types (including holotype), except SAMA R71652 which is slightly paler; number and intensity of brown dorsal and lateral spots varies slightly. Ventral surfaces and rear of thighs in all frogs show a more or less dense orange brown reticulum on ivory-white ground. Table 4. Body measurements and body ratios of the type series of Xenorhina thiekeorum sp. nov. SAMA R71651 is the male holotype. ZMB 91131, PNGNM (SJR 209052) and SAMA R71652 are male paratypes. All measurements in mm; for ex- planation of abbreviations see “Material and methods”. Reg.-No. SAMA ZMB PNGNM SAMA Mean + SD R71651 91131 R71652 SUL 23.0 20.7 2315 22.6 22.45 + 1.22 TL 9.2 9.2 10.1 9.7 9.55 + 0.44 Tal 6.4 6.5 AES 6.5 6.68 + 0.42 T4b 9.2 9.7 10.3 10.1 9.83 0.49 T4D 0.7 0.7 0.8 0.7 0.73 £0.05 T1D 0.4 0.4 0.5 0.5 0.45 + 0.06 F3L 4.3 4.1 5.0 4.5 4.48 + 0.39 F3D 0.5 0.5 0.6 0.6 0.55 + 0.06 F1D 0.3 0.3 0.5 0.4 0.38 + 0.09 HL 6.1 5.8 6.4 5:9 6.05 + 0.26 HW 1D 6.9 7.7 53 7.35 £0.34 END 1.9 1.7 2.0 1.9 1.88 + 0.13 IND 1.3 1.2 In 1.4 1.30 + 0.08 SL 2.8 2:5 2h 2.6 2.65 + 0.13 EST 25 2.4 2.6 2.3 2.45 £0.13 ED 1.8 1.6 1.6 1.7 1.68 + 0.09 TyD 1.2 1.0 1.1 Ayer 1.10 + 0.08 TL/SUL 0.40 0.44 0.43 0.43 0.43 + 0.017 TaL/SUL 0.28 0.31 0.31 0.29 0.30 + 0.015 T4L/SUL 0.40 0.47 0.44 0.45 0.44 + 0.029 T4D/SUL 0.030 0.034 0.034 0.031 0.032 + 0.002 T1D/SUL 0.017 0.019 0.021 0.022 0.020 + 0.002 F3L/SUL 0.187 0.198 0.212 0.199 0.199 +0.010 F3D/SUL 0.022 0.024 0.026 0.027 0.025 + 0.002 FID/SUL 0.013 0.014 0.017 0.018 0.016 + 0.002 T4D/F3D 1.40 1.40 1.38 Le 1.33 + 0.108 T1D/F1D 133 1233 1.00 1.25 1.23 0.156 HL/SUL 0.27 0.28 0.27 0.26 0.27 + 0.008 HW/SUL 0.33 0.33 0.33 0.32 0.33 + 0.005 HL/HW 0.81 0.84 0.83 0.81 0.82 + 0.015 END/SUL 0.083 0.082 0.085 0.084 0.084 +0.001 IND/SUL 0.057 0.058 0.055 0.062 0.058 + 0.003 END/IND 1.46 1.42 1.54 1.36 1.45 + 0.075 ED/SUL 0.078 0.077 0.068 0.075 0.075 + 0.005 TyD/SUL 0.052 0.048 0.047 0.049 0.049 + 0.002 TyD/ED 0.67 0.63 0.69 0.65 0.66 + 0.026 SL/SUL 0.122 0.121 0.115 0.115 0.118 + 0.004 EST/SUL 0.109 0.116 0.111 0.102 0.110 + 0.006 Distribution and ecological notes. Xenorhina thieke- orum sp. nov. 1s known only from the type locality ad- jacent to the Ok Menga (“Ok” = River in the local Min language), at an altitude of 620 maz.s.l. in the foothills of the Hindenburg Range, Ok Tedi headwaters in Western Province, Papua New Guinea (Fig. 16). The frogs were calling from 1—3 cm beneath the soil surface at the base of ginger plants after rain at night. Unlike many Xenorhina species, the distribution of calling males was “clumped”; SAL all four frogs were detected by their calls within an area of approximately 4 m? of wet hill forest, while none was heard calling in apparently suitable adjacent forest. Vocalisation. Three call series from the holotype (SAMA R71651) recorded at an air temperature of 22.5 °C were analysed, but due to poor recording quality, the lengths of calls and length of call intervals could not be measured for one of these series. Calls are a single, un- pulsed “piping” notes produced in short series. Calls are extremely soft and were barely audible to the human ear. Call series contain 6—8 calls produced at a rate of 2.5—-3.0 calls/s and last 2.0—-2.9 s (mean 2.3 s) (Fig. lla and b). Call length is 133-162 ms (mean 143.4 + 8.8 ms, n= 14) and length of call intervals is 168—376 ms (mean 250.6 + 51.8 ms, n= 12). There are four harmonics with frequen- cy peaks at 1.1, 2.2, 3.3 and 4.4 kHz; the third harmonic carries the dominant frequency (Fig. 11c). Volume and pitch of calls both increase marginally during the course of call series. Etymology. The specific epithet thiekeorum is the Lat- inised patronymic adjective in genitive plural of the fam- ily name Thieke. It is given to recognise a very long-last- ing friendship of the senior author with Heidi and Ulrich (Uli) Thieke from Berlin. Comparisons with other species. We compare Xenorhina thiekeorum sp. nov. with all congeners of a similar size (males with SUL ~ 18-25 mm) that have a single spike on each vomeropalatine. Xenorhina anorbis has hind legs shorter (TL/SVL < 0.38 vs. > 0.38), digital discs on toes absent (vs. expand- ed discs present on toes 2—5) and END/IND ratio lower (1.26—1.32 vs. 1.36—-1.54). Xenorhina brachyrhyncha has legs longer (TL/SVL 0.46—-0.49 vs. 0.40—-0.44), head longer (HL/SVL 0.30- 0.32 vs. 0.26—-0.28) and broader (HW/SVL 0.35-0.38 vs. 0.32—0.33) and END/IND ratio much lower (1.06-1.13 vs. 1.36-1.54). Xenorhina lanthanites has legs longer (TL/SUL 0.44— 0.46 vs. 0.40—0.44), dilated disc only on 4" toe (vs. dilat- ed discs on toes 2—5), T4D/F3D ratio higher (1.50—2.0 vs. 1.17—1.40), END/IND ratio lower (0.94—1.20 vs. 1.36- 1.54) and advertisement call series much longer (up to more than one minute vs. a few seconds), with call inter- vals longer (397—896 ms vs. 168-376 ms) and repetition rate lower (1.2—1.8 vs. 2.5—3.0 calls/s). Xenorhina mehelyi is probably much larger (SVL to > 35 mm vs. males 20.7—23.5 mm), internarial distance greater (IND/SVL 0.061-0.077 vs. 0.055—0.062) and has different advertisement calls: call series of X. mehelyi contain > 10 calls produced at a rate of 0.60 calls/s (vs. < 10 calls produced at a rate of 2.75 calls/s in Xenorhina thiekeorum sp. nov.); and dominant frequency is 0.88 kHz in X. mehelyi (vs. 3.3 kHz in X. thiekeorum). Xenorhina perexigua is smaller (16.7 mm _ vs. 20.7—23.5 mm SUL) and many body ratios differ from zse.pensoft.net 372 i= a) = = < 3 2 aoe ay YS ; 0 0.4 0.8 a 1.6 Titt € (3) Gunther, R. & Richards, S.: Six new species of Xenorhina Relative am plitude | } 20 10 2 3.0 40 Frequency (kHz) Figure 11. (a) Oscillogram; (b) Spectrogram and (c) amplitude spectrum of a series of six calls from the holotype of Xenorhina thiekeorum sp. nov. Xenorhina thiekeorum sp. nov. (Tables 2 and 4), but small sample sizes preclude robust comparisons of body ratios. Advertisement calls differ as follows: Xenorhi- na perexigua sp. nov. utters calls in series lasting more than 4 s, containing about 30 calls produced at rate of 6.8-6.9 calls/s (vs. call series lasting 2—3 s containing just 6-8 calls produced at rate of 2.5—3.0 calls/s); call length of Xenorhina perexigua sp. nov. is also much shorter (29— 42 ms vs. 133-162 ms in Xenorhina thiekeorum sp. nov.). Xenorhina pohleorum has fingers shorter (F3L/SUL 0.178—-0.179 vs. 0.187—0.212), disc on third finger small- er (F3D/SUL 0.019-0.020 vs. 0.022-—0.027), T4D/F3D ratio higher (1.75 vs. 1.17—1.40), END/IND ratio lower (1.31—1.33 vs. 1.36—1.54), eyes smaller (ED/SUL 0.057-— 0.064 vs. 0.068—0.078) and TyD/ED ratio higher (0.75- 0.92 vs. 0.63-0.69). Moreover, call length of Xenorhi- na pohleorum sp. nov. is much shorter (~ 70-90 ms vs. 130-150 ms). Xenorhina_ schiefenhoeveli is larger (SVL 26.7— 30.7 mm vs. 20.7—23.5 mm), with ratio of END/IND lower (1.04—1.33 vs. 1.36—-1.54) and different calls; call series last > 100 s (vs. 2-3 s in Xenorhina thiekeorum sp. nov.), with call intervals > 700 ms (vs. less than 400 ms). Xenorhina tumulus is larger (SVL > 26.0 mm vs. < 24.0 mm), with internarial distance relatively longer (IND/ SVL 0.063—0.069 vs. 0.055—0.062), distance between eye and naris relatively shorter (END/SVL 0.073-0.081 vs. 0.082-0.085), END/IND ratio lower (1.11—1.28 vs. 1.36- 1.54) and call length shorter (60-70 ms vs. 133—162 ms). Xenorhina wiegankorum sp. nov. http://zoobank.org/DD757B96-EA FD-427B-9844- 1 BD80D13544C Holotype. SAMA R71653 (SJR 10372), adult male, from Baia River, Western Province, Papua New Guinea (6.0205°S, 142.5473°E; 330 m a.s.l.), collected by S.J. Richards on 15-02-2008. zse.pensoft.net Paratypes. PNGNM (FN SJR10373), adult male, same details as for holotype; SAMA R71654 (FN SJR10400), adult male, from Camp 2, upper Strickland River ba- sin, Western Province, Papua New Guinea (5.9018°S, 142.4360°E; 950 m a.s.l.), collected by S.J. Richards on 19-02-2008; ZMB 91132 (FN SJR14220), adult male, Rentoul River, Western Province, Papua New Guinea (6.4355°S, 142.5615°E; 380 ma.s.l.), collected on 14-08- 2014 by S.J. Richards; SAMA R65073 (FN SJR10948), adult male, Gugusu Camp, Muller Range, Western Prov- ince (5.7290°S, 142.2630°E; 515 m a.s.l.), collected by S.J. Richards and C. Dahl on 8-09-2009. Diagnosis. This species of Xenorhina is character- ised by the unique combination of: medium size (males 32.0-35.7 mm SUL); vomeropalatines each with one strongly developed triangular spike; legs moderately long (TL/SUL 0.44-0.47); all fingers tips without and all toe tips with expanded discs; eye-naris distance greater than internarial distance (END/IND 1.19-—1.37); tympanum same size as, or slightly smaller than, eye (TyD/ED 0.80— 1.00). Dorsal surfaces in life different shades of grey or brown; ventral surfaces different shades of red or yellow, throat and chest with some darker flecks. Advertisement calls uttered in series lasting 10—20 s and containing 20- 40 calls; length of calls 60-100 ms, dominant frequency at 0.5 kHz. Description of the holotype. Measurements are sum- marised in Table 5, a dorsolateral view in life 1s shown in Fig. 12a and ventral surfaces in life in Fig. 12b. Head broader than long (HL/HW 0.84); snout acuminate from above and below and distinctly protruding in profile; vomerine spikes strongly developed; prepharyngeal ridge clearly expressed with about 14 denticles; tongue long, broad, not bilobed posteriorly; loreal region oblique, no canthus rostralis; nostrils near tip of snout, positioned dorsolaterally, visible from above, but not from below; Zoosyst. Evol. 97 (2) 2021, 355-382 343 Figure 12. Holotype (SAMA R71653) of Xenorhina wiegankorum sp. nov. in life: (a) Dorsolateral view; (b) Ventral view; (c) Volar view of left hand; (d) Thenar view of left foot. eye-naris distance greater than internarial distance (END/ IND 1.37); tympanic annulus more strongly defined in preservative than in life, its diameter smaller than that of eye (TyD/ED 0.80); well defined supratympanic fold ex- tends from marginally behind eye to insertion of fore leg; shank moderately short (TL/SUL 0.44); fingers moderate- ly short, not webbed, tips of all fingers not wider than pe- nultimate phalanges, but with circum-marginal grooves, relative lengths of fingers 3 > 4 > 2 = 1 (Fig. 12c); all toe tips acuminate, but wider than penultimate phalanges, with circum-marginal grooves; toes not webbed, relative lengths 4>3>5>2> 1 (Fig. 12d); plantar, palmar and subarticular tubercles barely defined. Body laterally and dorsum of legs partly, with scattered small tubercles in life and in preservative; all ventral surfaces smooth; tip of snout (especially ventrally) with several tiny elevations. In life, all dorsal surfaces almost uniformly light ol- ive-brown (RAL 8008); lumbar spot absent; back with yellowish mid-dorsal line that continues along hind legs on to tarsus; tubercles with whitish apices concentrated mainly on lateral surfaces of body; large dark triangu- lar spot on posterior of thighs around vent absent; iris blackish with golden speckles; ventral surfaces of toes predominantly signal-grey (RAL 7004), plantar surfaces zse.pensoft.net 374 Table 5. Body measurements and body ratios of the type series of Xenorhina wiegankorum sp. nov. SAMA R71653 is the male ho- lotype; all others are male paratypes. All measurements in mm; for explanation of abbreviations see “Material and methods”. Reg.-No. SAMA PNGNM ZMB SAMA SAMA Mean + SD R71653 (SJR10373) 91132 R71654 R65073 SUL 32.4 32.0 34.9 feyeull 35.7 33:62°+'1 61 ag 14.4 14.9 16.0 1539) 15:6 15.38 + 0.63 TaL 9° 10.0 LOM, 10.4 10.2 10.16 + 0.45 T4b 14.5 Loz 16.3 15.6 16.8 15.66 + 0.92 T4D le? 1.3 1.2 1.3 1.4 1.28 + 0.08 T1D 0.9 0.9 0.8 1.0 0.9 0.90 + 0.07 F3L 6.1 6.8 6.7 7.0 TA 6.74 + 0.39 F3D 0.8 0.9 0.7 0.8 0.8 0.76 + 0.09 F1D 0.7 0.8 0.7 0.7 0.7 0.72 + 0.05 HL 9.0 9.5 8.6 91 9.7 9.18 + 0.43 HW 10.7 Les 11.9 THES 11.4 11.36 + 0.43 END 2.6 2.7 2.7 2.5 2.5 2.60 + 0.10 IND 1.9 2.1 2.2 Be 2.0 2.06 + 0.11 SL 4.1 4.2 4.5 4.7 4.5 4.40 + 0.24 EST 3.9 3.8 3.9 4.0 4.1 3.94 40.11 ED 2.0 2.1 22 2.2 2.2 2.14 + 0.09 TyD 1.6 2.0 ES Ze] 2.2 L:96nt-0:23 TL/SUL 0.44 0.47 0.46 0.47 0.44 0.46 + 0.015 TaL/SUL 0.29 0.31 0.31 0.31 0.29 0.30 + 0.011 T4L/SUL = 0.45 0.47 0.47 0.47 0.47 0.47 + 0.009 T4D/SUL 0.037 0.041 0.034 0.039 0.039 0.038 + 0.003 T1D/SUL 0.028 0.028 0.023 0.030 0.025 0.027 + 0.003 F3L/SUL 0.188 0.213 0.192 0.211 0.199 0.201 +0.011 F3D/SUL 0.025 0.028 0.020 0.024 0.022 0.024 +0.003 FID/SUL 0.022 0.025 0.020 0.021 0.020 0.022 + 0.002 T4D/F3D ~=:1.50 1.44 1.71 1.63 175 16 le+:O2133 TID/F1D = 1.29 Laks 1.14 1.43 1.29 1.26 + 0.124 HL/SUL 0.28 0.30 0.25 0.27 0.27 0.27 + 0.018 HW/SUL 0.33 0.35 0.34 0.35 0.32 0.34 + 0.013 HL/HW 0.84 0.84 0.72 0.79 0.85 0.81 + 0.054 END/SUL 0.080 0.084 0.077 0.076 0.070 0.078 40.005 IND/SUL 0.059 0.066 0.063 0.069 0.056 0.063 + 0.005 END/IND 1.37 1.24 1223 1.19 1.25 1.26 + 0.068 ED/SUL 0.062 0.066 0.063 0.066 0.062 0.064 + 0.002 TyD/SUL 0.049 0.063 0.054 0.063 0.062 0.058 + 0.006 TyD/ED 0.80 0.95 0.86 0.95 1.00 0.91 + 0.080 SL/SUL 0.127 0.131 0.129 0.142 0.126 0.131 + 0.006 EST/SUL 0.120 0.119 0.112 0.121 O.115 0.117 + 0.004 brown-grey; ventral surfaces of fingers and palms pre- dominantly signal-grey; abdomen and ventral surfaces of thighs, shanks and arms melon-yellow (similar to RAL 1028) with inconspicuous whitish spots; ground colour of throat and chest also melon-yellow, but overlain with dense pattern of beige-grey and off-white spots. In preservative, all dorsal surfaces pastel-violet (RAL 4009), with only few darker areas and inconspicuous whitish tubercle apices. Melon-yellow ventral surfac- es faded to ivory colour in preservative and pattern on chest and throat changed from beige-grey to brown-beige (RAL 1011). Morphological variation. Morphometric data for all paratypes are similar (Table 5). Colour pattern of ZMB 91132 (and probably of PNGNM [SJR 10373]) in life is similar to holotype. Dorsal surfaces of SAMA R71654 are telegrey (RAL 7045) with small whitish spots (Fig. 13) and ventral surfaces predominantly broom-yellow (RAL 1032). Dorsal surfaces of SAMA R65073 are a mixture of stone-grey (RAL 7030) and brown-grey (RAL 7013) zse.pensoft.net Gunther, R. & Richards, S.: Six new species of Xenorhina Figure 13. Xenorhina wiegankorum sp. nov. paratype SAMA R71654 in dorsolateral view. reticula interspersed with whitish spots (mainly on lower flanks) and ventral surfaces predominantly zinc-yellow (RAL 1018). In preservative, ground colour of dorsal surfaces of head and back of all specimens is dark shades of pastel-vi- olet (RAL 4009), with dorsal surfaces of extremities light brown with dark brown stripes and spots. Two paratypes with and two without, light mid-dorsal line. Snout tip grey in all specimens. Part of chest, entire abdomen and ventral surfaces of thighs light ivory; throat and part of chest light ivory overlain by more or less expanded brown-beige ar- eas. Rear of thighs in all type specimens predominantly brown, only a small area around vent blackish. Distribution and ecological notes. Xenorhina wie- gankorum sp. nov. has a known distribution limited to altitudes of 330-950 m a.s.l. in the foothills of the up- per Strickland River catchment in Western Province, south-western Papua New Guinea (Fig. 16). Males called at night from under the litter on the forest floor or from slightly beneath the soil surface, during or immediately after heavy rain. Vocalisation. We analysed one call series from the holotype (SAMA R71653) recorded at an air temperature of 23.7 °C, two call series from paratype SJR 10400 re- corded at 21.0 °C and one call series of paratype ZMB 91132 recorded at 25.0 °C. Calls are rather deep, unpulsed “popping” notes that, as is typical for many Xenorhina species, increase in volume during the course of the call series. Pitch of calls also increases slightly during the course of each series. Although there is some variation in call length and inter-call interval amongst calls of the three animals recorded, there is high overlap in all call pa- rameters and we have no doubt that all represent the same species. We, therefore, combined the calls for analysis Calls are of approximately equal length, but inter-call intervals are somewhat variable. A call starts abruptly at high amplitude, which then decreases gradually until end of call (Fig. 14a). There are 2—7 harmonics, though the second is often missing (Fig. 14b and c); fundamental and dominant frequencies are at 0.55 kHz (Fig. 14c). Length Zoosyst. Evol. 97 (2) 2021, 355-382 5 2 Ms by Hh inn edi | li \ q ~ Wh : ee | i " \ in a 0 . o4 O8 ee 16 Tit & (3) 375 Relative am plitude 210 10 20 30 Frequency (kHz) Figure 14. (a) Oscillogram; (b) Spectrogram and (c) Amplitude spectrum of the last five calls from a call series containing 29 calls, produced by paratype ZMB 91132 of Xenorhina wiegankorum sp. nov. of call series is 13.8-18.1 s (mean 15.3 s, n = 4); with 22-39 calls per series (mean 28.8, n = 4); call length is 60—104 ms (mean 87.1 + 6.7 ms, n= 115); intercall inter- val length is 286-1073 ms (mean 459.6 + 137.6 ms, n= 111) with call repetition rate of 1.71—2.15 calls/s (mean 1.86 calls/s). Etymology. The specific epithet wiegankorum is the Latinised patronymic adjective in genitive plural of the family name Wiegank. It is given to recognise a very long-lasting friendship of the senior author with Ulla and Friedrich-Manfred (Conny) Wiegank from Potsdam. Comparisons with other species. We compare Xenorhina wiegankorum sp. nov. with all congeners of a similar size (SUL ~ 28-38 mm) that have a single spike on each vomeropalatine bone. Xenorhina fuscigula has hind legs shorter (TL/SVL < 0.40 vs. > 0.40), eye-naris distance shorter (END/SVL 0.064—0.074 vs. 0.070-0.084) and fourth toe shorter (T4L/SVL 0.34—0.41 vs. 0.45—-0.47); advertisement calls of X. fuscigula are produced singly (vs. in a long series containing up to 39 calls). Xenorhina huon (Blum & Menzies, 1989) has hind legs shorter (TL/SVL < 0.40 vs. > 0.40), eyes larger (ED/SVL 0.070-0.091 vs. 0.062—0.066) and ventral surfaces with dark flecking (vs. ventral surfaces without dark flecking). Xenorhina huon 1s also known only from mountainous regions 1800-2000 m a.s.l. on the Huon Peninsula, near the north coast of Papua New Guinea (vs. lowlands south of the central cordillera). Xenorhina lacrimosa exhibits considerable overlap in many morphometric characters, but displays extensive variation in dorsal colouration (vs. predominantly brown or grey); vent enclosed in dark brown patch (vs. patch absent) and ventral surfaces deep orange or occasional- ly grey-brown, with white spots (vs. ventral surfaces at least partially yellow) (Figs 1-2 vs. 12—13); dorsal sur- faces also appear less rugose in life (Figs 1—2 vs. 12-13). Advertisement calls are very different: call series of X. lacrimosa much longer (26—60 s vs. 12-18 s), with fewer calls (7-12 vs. 22-39), repetition rate much slower (0.20— 0.27 vs. 1.70—2.15 calls/s), call length longer (141-231 ms vs. 60 to 104 ms) and call interval longer (2.8—-8.0 s vs. 286-1073 ms). Xenorhina subcrocea (Menzies & Tyler, 1977) is smaller (SVL 30.5-—33.3 vs. 32.0-35.7), with hind legs longer (TL/SVL > 0.46 vs. < 0.47), ventral surfaces with dark reticulation in preservative (vs. without dark reticu- lation), call intervals within series shorter (154-285 ms vs. 286-1073 ms), produced at rate of about 4 calls/s (vs. 1.7—2.2 calls/s). Xenorhina zweifeli has similar body size and ratios. It differs from Xenorhina wiegankorum sp. nov. by having a conspicuous dark brown supratympanic stripe (vs. ab- sent) and greatly different advertisement calls: X. zweifeli utters single calls at long and irregular intervals (Kraus and Allison 2002), with 2—3 calls sometimes uttered in quick succession, during the day and early evening (Kraus and Allison 2002); in contrast, Xenorhina wiegankorum Sp. nov. produces calls in discrete series with 22-39 calls produced in rapid succession, only at night. Xenorhina woxvoldi sp. nov. http://zoobank.org/2F2CA28A-5E2F-485C-911B-E3FD35AF7E27 Holotype. SAMA R71646 (SJR10249), adult male, from southern edge of Karius Range, Hela Province, Papua New Guinea (5.9911°S, 142.6707°E; 1,368 m a.s.l.), col- lected on 07-02-2008 by S.J. Richards. Paratype. ZMB 91133 (SJR 10311), adult male, same collection details as for holotype. zse.pensoft.net 376 Gunther, R. & Richards, S.: Six new species of Xenorhina .< 4 Figure 15. Holotype (SAMA R71646) of X. woxvoldi sp. nov. in life: (a) Dorsolateral view; (b) Ventral view; (c) Volar view of right hand; (d) Thenar view of right foot. Diagnosis. This species of Xenorhina is character- ised by the unique combination of: small to medium-size (males 28.7—30.1 mm SUL); vomeropalatines each with one moderate-sized vomerine spike; legs short (TL/SUL 0.36 in two specimens); all fingers and toe 1 without ex- panded discs, toes 2—5 with weakly expanded discs (T4D/ SUL 0.038—0.040); eye-naris distance smaller than inter- narial distance (END/IND 0.80-0.91); tympanum slightly larger than eye (TyD/ED 1.11 in two specimens). Dorsal surfaces bluish-brown in life, ventral surfaces dark orange with irregular whitish and greyish spots. Advertisement calls uttered in series lasting 3—S s, calls per series 13-19, call length 37—84 ms, repetition rate 4.0—4.5 calls/s. Description of the holotype. Measurements are sum- marised in Table 6, a dorsolateral view in life is shown zse.pensoft.net in Fig. 15a and ventral surfaces in life in Fig. 15b. Head broader than long (HL/HW 0.75); snout acuminate from above, protruding in profile; loreal region oblique, no canthus rostralis; nostrils near tip of snout, directed more laterally than dorsally, visible from above, but not from below; eye-naris distance less than internarial dis- tance (END/IND 0.90); tympanum visible in life and preservative, its diameter slightly larger than eye (TyD/ ED 1.11); tongue very broad; vomerine spikes triangular, moderately large; prepharyngeal ridge narrow with four denticles; supratympanic fold well-developed, not reach- ing eye or insertion of fore leg (Fig. 15a); shank short (TL/SUL 0.36); fingers moderately short, not webbed; tips of all fingers with circum-marginal grooves, not or only marginally wider than penultimate phalanges, rela- tive lengths of fingers 3 > 4 >2 > 1 (Fig. 15c); all toe tips Zoosyst. Evol. 97 (2) 2021, 355-382 Table 6. Body measurements and body ratios of the type series of Xenorhina woxvoldi sp. nov. SAMA R71646 is the male holo- type; ZMB 91133 is a male paratype. All measurements in mm; for explanation of abbreviations see “Material and methods”. Reg.-No. SAMA R71646 ZMB 91133 Mean SUL 30.1 28.7 29.40 ile 10.8 10.4 10.60 Tal 7.8 7.0 7.40 T4L 12.2 12.0 12.10 T4D 1.2 ia 1eT5 T1D 0.7 0.6 0.65 F3L 5 5.0 5135 F3D 0.8 0.7 0.75 F1D 0.6 0.6 0.60 HL 7.6 Pal 7.35 HW 10.2 8.7 9.45 END 1.8 1.6 1.70 IND 2.0 2.0 2.00 SL 3.3 3.1 3.20 EST 3.0 2.8 2.90 ED 1.8 1.9 1.85 TyD 2.0 2.1 2.05 TL/SUL 0.36 0.36 0.36 TaL/SUL 0.26 0.24 0.25 T4L/SUL 0.41 0.42 0.415 T4D/SUL 0.040 0.038 0.39 T1D/SUL 0.023 0.021 0.22 F3L/SUL 0.189 0.174 0.182 F3D/SUL 0.027 0.024 0.026 FID/SUL 0.020 0.021 0.021 T4D/F3D 1.50 15:7 1.54 T1D/F1D 1.16 1.00 1.08 HL/SUL 0.25 0.25 0.25 HW/SUL 0.34 0.30 0.32 HL/HW 0.75 0.82 0.79 END/SUL 0.060 0.056 0.058 IND/SUL 0.066 0.070 0.068 END/IND 0.90 0.80 0.85 ED/SUL 0.060 0.066 0.063 TyD/SUL 0.066 0.073 0.070 TyD/ED 1.11 allah 1.11 SL/SUL 0.110 0.108 0.109 EST/SUL 0.100 0.098 0.099 with circum-marginal grooves, those on toes 2—4 clear- ly wider than penultimate phalanges, those on toe 1 and toe 5 scarcely wider than penultimate phalanges; toes not webbed, relative lengths 4 > 3 >5 >2 > 1 (Fig. 15d); plantar, palmar and subarticular tubercles barely visible; body laterally with some distinct tubercles in life, bare- ly visible in preservative; dorsal surfaces of extremities, middle of dorsum and all ventral surfaces smooth; tip of snout with several tiny pimples. In life, dorsal surface of head, body and extremities a mixture of grey-brown and copper-brown (RAL 8004) (Fig. 15a); lower flanks uniform greyish with bluish hue and off-white dots and streaks; semicircular lumbar spot present, but only vaguely defined; a distinct whitish mid-dorsal line extends on to rear of thighs and on shanks and tarsi, then as broken line on to abdomen and chest; dorsal and ventral surfaces of fingers and toes and palmar surfaces orange; plantar surfaces a mixture of irregular light grey, dark grey and orange spots. Ventral surfaces of throat, chest, abdomen and extremities orange-brown (RAL 8023) with irregular light grey spots (Fig. 15b); most tubercles on flanks and extremities with whitish 377 tips; snout tip light grey with tiny dark grey spots; out- er margin of iris blackish and inner margin gold-orange, with some integration of colours at their margins. In preservative, dorsal surfaces changed from cop- per-brown to mahogany-brown (RAL 8016), that of ven- tral surfaces from orange-brown to ivory (RAL 1014). Dorsal surfaces of fingers and toes also become ivory coloured. Lumbar spots no longer visible. Morphological variation. All body measurements and body ratios of holotype and paratype are similar (Ta- ble 6). In life, dorsal surfaces of paratype a mixture of lighter and darker brown and reddish areas, with reddish components more restricted than in holotype. Colours of flanks and dorsal surfaces of fingers and toes and col- our and extent of mid-dorsal line (extending on to hind limbs and abdomen) as for holotype. Ventral surfaces more yellow and light grey spotting more extensive, in paratype. Dorsal surfaces in preservative slightly paler than holotype, ventral surface with more extensive pale brown reticulation. Distribution and ecological notes. Xenorhina wox- voldi sp. nov. is known only from one location at an altitude of 1,368 m a.s.l. on the southern fringe of the Karius Range in Hela Province, Papua New Guinea (Fig. 16), where males called from within the humus layer in lower montane rainforest during late afternoon and early evening. Vocalisation. Two call series from the holotype (SAMA R71646) and one from the paratype (ZMB 91133), record- ed at air temperatures of 18—19.5 °C, were analysed. Call is a Single unpulsed, piping note produced in discrete se- ries. Call series last 2.9-4.8 s (mean 3.7 s, n=3) and con- tain 13-19 calls (mean 5.7 calls, n = 3) produced at a rate of 4.0-4.5 calls/s (mean 4.3 calls/s, n = 3). Call length is 37-84 ms (mean 75.1 + 8.3 ms, n= 47) and call intervals last 137—250 ms (mean 172.4 + 28.4 ms, n= 44). Calls are of approximately equal length throughout a series (first call may be shorter) with approximately equal intervals (intervals between first two and last two calls of a series may be slightly longer). Volume of each call increases during course of call series, but rise in pitch is marginal. Calls start abruptly at maximum amplitude, which then decreases gradually until end of call (Fig. 17a). All calls have harmonic structure with 4-5 harmonics between 0.7 and 3.2 kHz (Fig. 17b). First harmonic clearly dominant, with peak at 0.8 kHz (Fig. 17c). Third harmonic often with more energy than second. Frequency of calls weakly modulated with slight reduction during course of call. Etymology. The specific epithet woxvoldi is the Lat- inised patronymic adjective in genitive singular derived from the family name Woxvold. It is in gratitude of the junior author to Iain Woxvold for the many years of friendship, camaraderie and shared adventures in remotest New Guinea. zse.pensoft.net 378 Gunther, R. & Richards, S.: Six new species of Xenorhina od el 4 teee eae su Papua New Guinea Port Moresby Elevation (m) 4000 2000 6) Figure 16. Map of Papua New Guinea showing the known distributions of X. thiekeorum sp. nov. (blue circle), X. wiegankorum sp. nov. (yellow triangles) and X. woxvoldi sp. nov. (red square). Arrows indicate the type localities. pk) 3 <7 ti = | an 4 a = br a oii| = 37 m | = " = 2 ie a HH mH { y O | ee ee A ee ee ee ee oe ee | 4 Cee ee ee ee oe on eT o4 ; 2 10 20 30 40 Titi € (8) Frequency (kHz) Figure 17. (a) Oscillogram, (b) Spectrogram and (c) Amplitude spectrum of the last five calls of a series containing 15 calls from the holotype of Xenorhina woxvoldi sp. nov. Comparisons with other species. We compare Xenorhina fuscigula differs from Xenorhina woxvoldi Xenorhina woxvoldi sp. nov. with all congeners of a sim- sp. nov. by having an internarial distance shorter (IND/ ilar size (SUL ~ 25-35 mm) that have a single spike on SVL 0.054—0.064 vs. 0.066—0.070), eye-naris distance each vomeropalatine. greater (END/SVL 0.064—0.074 vs. 0.056—0.060), END/ zse.pensoft.net Zoosyst. Evol. 97 (2) 2021, 355-382 IND ratio higher (1.00—1.36 vs. 0.80—0.90), ventral sur- faces pale with dark reticulation (vs. orange with light grey spots) and calls produced singly (vs. produced in rapid series of 13-19 calls). Xenorhina huon has eye-naris distance greater (0.073- 0.103 vs. 0.056—-0.060), END/IND ratio higher (1.00- 1.27 vs. 0.80-0.90), eyes larger (ED/SVL 0.070-0.091 vs. 0.060—0.066), head wider (HW/SVL 0.35-0.47 vs. 0.30—0.34) and ventral surfaces with dark flecking (vs. ventral surfaces with light flecking in life and pale brown reticulation 1n preservative). Xenorhina lacrimosa is larger (SUL 34.3-41.0 mm vs. 28.7—30.1 mm), has shanks longer (TL/SUL 0.42—0.46 vs. 0.36 in both known Xenorhina woxvoldi sp. nov.), fourth toe longer (T4L/SUL 0.42-0.49 vs. 0.41-0.42), head longer (HL/SUL 0.27-0.30 vs. 0.25 in both known Xenorhina woxvoldi sp. nov.), eye-naris distance greater (END/SUL 0.073-0.099 vs. 0.056—0.060) and advertise- ment calls longer (141-231 ms vs. 37-84 ms) with lower repetition rate (0.20-0.27 vs. 4.0-4.5 calls/s). Xenorhina mehelyi has hind legs longer (TL/SVL > 0.42 vs. 0.36), eye-naris distance greater (END/SVL 0.076-0.096 vs. 0.056—0.060), END/IND ratio higher (1.12—1.50 vs. 0.80-0.90), eyes larger (ED/SVL 0.067— 0.079 vs. 0.060—0.066), ventral surfaces with dark mot- tling (vs. no dark mottling) and calls longer (on aver- age140 ms vs. 75 ms) with inter-call intervals also longer (on average 1500 ms vs. 172 ms). Xenorhina_ schiefenhoeveli has eye-naris distance greater (END/SVL 0.077 vs. 0.056—0.060), END/IND ratio higher (1.16—1.21 vs. 0.80—-0.90), eyes larger (ED/ SVL 0.071—0.081 vs. 0.060—0.066), ventrum cream, with reticulated brown (vs. orange-red with whitish flecking); calls longer (~ 100 ms vs. mean of 75 ms), uttered in very long series of more than 100 calls (vs. 13-19 calls) with repetition rate about 2 calls/s (vs. 4.0—4.5 calls/s). Xenorhina subcrocea has hind legs longer (TL/SVL > 0.46 vs. < 0.40), ratio of END/IND much larger (1.26— 1.41 vs. 0.80-0.90), ventral surfaces with dark reticula- tion (vs. with whitish flecking) and mid-dorsal line absent (vs. distinct dorsal line present). Xenorhina tumulus has eye-naris distance greater (0.073- 0.081 vs. 0.056—-0.060), END/IND ratio higher (1.11—1.28 vs. 0.80—0.90), ventral surfaces in life pinkish, mottled with brown (vs. orange-brown with no brown mottling) and call intervals within series 300-400 ms (vs. 137—250 ms). Xenorhina wiegankorum appears to be larger (five males 32.0—-35.7 mm vs. two males 28.7—30.1 mm SUL), has hind legs much longer (TL/SUL 0.44—0.47 vs. 0.36 in two specimens), has toes longer (T4L/SUL 0.45—0.47 vs. 0.41-0.42), fingers longer (F3L/SUL 0.188—0.213 vs. 0.174—0.189), END/IND ratio higher (1.19-1.37 vs. 0.80-0.90) and a different advertisement call (see de- scription of X. wiegankorum, this paper). Xenorhina zweifeli is larger (SVL 33.2-38.0 vs. 28.7— 30.1), with internarial distance smaller (IND/SVL 0.052- 0.063 vs. 0.066—0.070), eye-naris distance larger (END/ SVL 0.071-0.085 vs.0.056—0.060), END/IND ratio higher (1.17-1.47 vs. 0.80—-0.90); ventral colour pattern of dark 379 brown flecks on a cream ground in preservative (vs. pale brown flecks on ivory-coloured ground) and call consisting of a single note (vs. 13—19 calls produced in distinct series. Discussion Recent assessments of anuran faunas on the large tropical islands of Sri Lanka (Meegaskumbura et al. 2002, Pethiya- goda et al. (2014) and Madagascar (Vieites et al. 2009) have revealed vastly underestimated levels of diversity. A similar pattern is emerging for New Guinea, the world’s largest and highest tropical island. New Guinea has the most diverse insular anuran fauna globally, with more than 400 species currently recognised (Frost 2021). Furthermore, field-based Species inventories across the Island during the past 2—3 de- cades have rapidly increased the rate of species discovery and description, a trend that shows no sign of approaching an asymptote (Allison 2014). This rapid advance in taxo- nomic knowledge of the amphibian fauna has been generat- ed substantially by studies of morphological and bioacoustic variation (e.g. Gunther 2001, Richards and Gunther 2018, Kraus 2019), while molecular assessments of New Guinea anuran diversity remain relatively rare (Oliver et al. 2013). Anuran advertisement calls are useful for taxonom- ic studies because they are mate recognition signals that are generally species-specific, exhibit limited variation amongst individuals and populations (although some features can be influenced in partially predictable ways by environmental factors, such as temperature) and like- ly have a genetic basis (Hoskin 2005, Kohler et al. 2017, Emmnrrich et al. 2020). We, therefore, consider the unique bioacoustics traits of each new species described here to be a strong indicator of species level divergence. Known calls of Xenorhina species reflect the acoustic constraints imposed by a fossorial existence. They comprise short, precise and melodious “hooting” or “piping” notes with a low fundamental frequency and well-defined harmonics that are normally produced in regular call series (Menzies and Tyler 1977, Blum and Menzies 1989). The six new species described here each produce advertisement calls of this type, but each is distinct from the known calls of congeners and these differences are concordant with the patterns of morphological variation documented. These calls meet the criteria for Call Guild A: “non-frequency modulated, non-pulsed simple call,” or Call Guild B: “fre- quency modulated, non-pulsed simple call” of Emmrich et al. (2020), depending on the extent of modulation exhibit- ed amongst species (defined as “with significant change” vs. “without significant change” by Emmrich et al. (2020). The description of Xenorhina perexigua sp. nov. on the basis of a single specimen reflects the difficulty of detect- ing and capturing small, nocturnal, fossorial frogs in an inaccessible terrain, that furthermore call most frequently during torrential rain. Thus, we are unable to determine whether this species is genuinely rare or merely difficult to detect. However, it is notable that the holotype was the only individual encountered during nearly one week of survey effort at the type locality. Numerous species zse.pensoft.net 380 of microhylid frogs have been described from the New Guinea region on the basis of “singletons” (Allison and Kraus 2000, Gunther et al. 2016). Lim et al. (2012) not- ed that “rare” species are common in taxonomic treatises and that additional sampling often leads only to single- tons becoming “doubles,” accompanied by detection of additional new species, based on singletons. The high-rainfall belt that extends across the southern slopes and adjacent lowlands of Papua New Guinea’s Central Cordillera (McAlpine et al. 1983) is proving to be a hotspot of anuran diversity (Gunther and Richards 2016, 2017, 2018, 2019, 2020, Richards and Gunther 2019) and our documentation of six previously undescribed frogs in the genus Xenorhina adds substantially to this already exceptional known diversity. Further studies are required to better document the distributions of these Xenorhina species. Several of them are known from only one or a handful of locations, but it 1s unclear whether they are genuinely range-restricted or whether their apparent rari- ty reflects the difficulty of conducting surveys throughout much of the region’s remote and rugged landscape. Acknowledgements We express our gratitude to The PNG National Research Institute who assisted with the Research Visa and the PNG Department of Environment and Conservation (now Conservation and Environment Protection Authority) for approving export of specimens. Ralph Foster, Sally South, Carolyn Kovach and Mark Hutchinson provided access to material, registration numbers and numerous other courtesies at the South Australian Museum and Lisa Capon kindly produced the maps. Ken Aplin, Chris Dahl and Demas Ama assisted with specimen collections in the field. Frank Tillack (ZMB) and Elke Gunther provid- ed various technical help. Fieldwork was supported by ExxonMobil PNG Limited (EMPNG), World Wildlife Fund, Wildlife Conservation Society and Ok Tedi Min- ing Limited. Linda Ford (then AMNH), Carla Kishinami (BPBM), George Lenglet (then IRSNB), José Rosado (MCZ), Giuliano Doria (MSNG), Hellen Kurniati and Mumpuni (MZB), Pim Arntzen (RMNH), Miguel Vences (then ZMA) and Frank Tillack and Mark-Oliver Rodel (ZMB) kindly provided access to specimens in their care. E. Lehr, N. Poyarkov, R. Brown and an anonymous re- viewer provided valuable comments that greatly im- proved the quality of this manuscript. We are most grate- ful to all of them for their support. References Allison A (2014) Status and diversity of the frogs of New Guinea. In: Heat- wole H, Das I (Eds) Conservation biology of amphibians of Asia. Sta- tus of conservation and declines of amphibians: Eastern Hemisphere. Natural History Publications (Borneo), Kota Kinabalu, 362-382. zse.pensoft.net Gunther, R. & Richards, S.: Six new species of Xenorhina Allison A, Kraus F (2000) A new species of frog of the genus Xenorhina (Anura: Microhylidae) from the north coast ranges of Papua New Guinea. Herpetologica 56(3): 285—294. Blum JP, Menzies JI (1989 [“1988”]) Notes on Xenobatrachus and Xe- norhina (Amphibia: Microhylidae) from New Guinea with descrip- tion of nine new species. Alytes 7(4): 125-163. Clulow S, Swan M (2018) A complete guide to frogs of Australia. Aus- tralian Geographic, Sydney, 336 pp. de Sa RO, Streicher JW, Selenkoyela R, Forlani MC, Loader SP, Green- baum E, Richards SJ, Haddad CFB (2012) Molecular phylogeny of microhylid frogs (Anura: Microhylidae) with emphasis on relation- ships among New World genera. BMC Evolutionary Biology 12: e241. https://doi.org/10.1186/1471-2148-12-241 Emmrich M, Vences M, Ernst R, Kohler J, Bare; MF, Glaw F, Jansen M, Rodel M-O (2020) A guild classification system proposed for anu- ran advertisement calls. Zoosystematics and Evolution 96: 515-525. https://do1.org/10.3897/zse.96.38770 Frost DR (2021) Amphibian Species of the World: an Online Refer- ence. Version 6.1. American Museum of Natural History, New York. [Electronic Database] http://research.amnh.org/herpetology/am- phibia/index.html [Accessed 19 February 2021] Frost DR, Grant T, Faivovich J, Bain RH, Haas A, Haddad CFB, de Sa RO, Channing A, Wilkinson M, Donnellan SC, Raxworthy CJ, Campbell JA, Blotto BL, Moler PE, Drewes RC, Nussbaum RA, Lynch JD, Green DM, Wheeler WC (2006) The amphibian tree of life. Bulletin of the American Museum of Natural History 297: 1-370. https://doi.org/10.1206/0003-0090(2006)297[0001:TA- TOL 20.602 Gamble T (2014) Collecting and preserving genetic material for herpe- tological research. Herpetological Circular 41. Society for the Study of Amphibian and Reptiles, Salt Lake City. Gunther R (2001) The Papuan frog genus Hy/ophorbus (Anura: Micro- hylidae) is not monospecific: descriptions of six new species. Rus- sian Journal of Herpetology 8(2): 81-104. Gunther R, Knop R (2006) A new species of Xenobatrachus (Anura, Microhylidae) with a striking resemblance to Xenorhina bouwensi. Zootaxa 1268: 39-57. https://doi.org/10.11646/zootaxa.1268.1.2 Gunther R, Richards SJ (2005) Two new tree-dwelling species of the genus Xenorhina from New Guinea (Anura, Microhylidae). Mit- teilungen aus dem Museum fiir Naturkunde in Berlin. Zoologische Reihe 81(2): 167-176. https://doi.org/10.1002/mmnz.200510010 Gunther R, Richards SJ (2016) Description of two new species of the microhylid genus Oreophryne (Amphibia: Anura: Microhylidae) from southern Papua New Guinea. Vertebrate Zoology 66: 157-168. Gunther R, Richards SJ (2017) Three new species of the microhylid frog genus Choerophryne (Amphibia, Anura, Microhylidae) from Papua New Guinea. Zoosystematics and Evolution 93(2): 265-279. https://doi.org/10.3897/zse.93.11576 Gunther R, Richards SJ (2018) A new species of the microhylid frog genus Choerophryne from Papua New Guinea. Alytes 36: 159-169. Gunther R, Richards SJ (2019) Three new species of Austrochaperina from southern Papua New Guinea (Anura, Microhylidae). Vertebrate Zoology 69: 327-344. Gunther R, Richards SJ (2020) Two new frog species of the genus Copiula Mehely, 1901 (Anura, Microhylidae, Asterophryinae) from southern Papua New Guinea. Russian Journal of Herpetology 27: 41-53. https://doi.org/10.30906/1026-2296-2020-27-1-41-53 Zoosyst. Evol. 97 (2) 2021, 355-382 Gunther R, Richards SJ, Dahl C (2014) Nine new species of microhylid frogs from the Muller Range in western Papua New Guinea (Anura, Microhylidae). Vertebrate Zoology 64(1): 59-94. Gunther R, Richards SJ, Tjaturadi B (2016) A new species of the frog genus Pseudocallulops from the Foja Mountains in northwestern New Guinea (Amphibia, Microhylidae). Russian Journal of Herpe- tology 23(1): 63-69. Gunther R, Richards SJ, Tjaturadi B, Krey K (2020) Two new micro- hylid frog species of the genus Xenorhina Peters, 1863 from the Raja Ampat Islands, Indonesia. Vertebrate Zoology 70(3): 333-347. https://do1.org/10.26049/VZ70-3-2020-06 Gunther R, Stelbrink B, von Rintelen T (2010) Oninia senglaubi, anoth- er new genus and species of frog (Amphibia, Anura, Microhylidae) from New Guinea. Zoosystematics and Evolution 86(2): 245-256. https://doi.org/10.1002/zoos.201000007 Hoskin CJ, Higgie M, McDonald KR, Moritz C (2005) Reinforcement drives rapid allopatric speciation. Nature 437: 1353-1356. https:// doi.org/10.1038/nature04004 Hyndman D, Menzies JI (1990) Rain Forests of the Ok Tedi headwaters, New Guinea: an ecological analysis. Journal of Biogeography 17(3): 241-273. https://doi.org/10.2307/2845 122 Kohler F, Gunther R (2008) The radiation of microhylid frogs (Am- phibia: Anura) on New Guinea: A mitochondrial phylogeny reveals parallel evolution of morphological and life history traits and dis- proves the current morphology-based classification. Molecular Phy- logenetics and Evolution 47: 353-365. https://doi.org/10.1016/). ympev.2007. 11.032 Kohler J, Jansen M, Rodriguez A, Kok PJR, Toledo LF, Emmrich M, Glaw F, Haddad CFB, Rédel M-O, Vences M (2017) The use of bioacoustics in anuran taxonomy: theory, terminology, methods and recommendations for best practice. Zootaxa 4251(1): 1-124. https:// doi.org/10.11646/zootaxa.4251.1.1 Kraus F (2011) New frogs (Anura: Microhylidae) from the mountains of western Papua New Guinea. Records of the Australian Museum 63(1): 53-60. https://doi.org/10.3853/j.0067-1975.63.2011.1584 Kraus F (2019) A revision of Callulops doriae (Anura: Microhylidae), with descriptions of four new species. Zootaxa 4612: 1—28. https:// doi.org/10.11646/zootaxa.4612.1.1 Kraus F, Allison A (2000) Two new species of Cophixalus from New Guinea. Journal of Herpetology 34: 535-545. https://doi. org/10.2307/1565268 Kraus F, Allison A (2002) A new species of Xenobatrachus (Anura: Microhylidae) from northern Papua New Guinea. Herpetologica 58(1): 56-66. https://doi.org/10.1655/0018-083 1(2002)058[0056 “ANSOXA]2.0.CO;2 Lim GS, Balke M, Meier R (2012) Determining species boundaries in a world full of rarity: singletons, species delimitation methods. System- atic Biology 61(1): 165-169. https://doi.org/10.1093/sysbio/syr030 Mayr E (1963) Animal species and evolution. The Belknap Press of Harvard University Press, Cambridge, Massachusetts, 797 pp. McAlpine JR, Keig G, Falls R (1983) Climate of Papua New Guinea. Australian National University Press, Canberra, 208 pp. 381 Meegaskumbura M, Bossuyt F, Pethiyagoda R, Manamendra-Arachchi K, Bahir M, Milinkovitch MC, Schneider CJ (2002) Sri Lanka: an amphibian hot spot. Science 298: 379-379. https://doi.org/10.1126/ science.298.5592.379 Menzies JI (2006) The frogs of New Guinea and the Solomon Islands. Pensoft, Sofia-Moscow, 345 pp. Menzies JI, Tyler MJ (1977) Systematics and adaptations of some Papuan microhylid frogs which live underground. Journal of Zoology, London 183: 431-464. https://doi.org/10.1111/j.1469-7998.1977.tb04198.x Oliver LA, Rittmeyer EN, Kraus F, Richards SJ, Austin CC (2013) Phylogeny and phylogeography of Mantophryne (Anura: Micro- hylidae) reveals cryptic diversity in New Guinea. Molecular Phy- logenetics and Evolution 67: 600-607. https://doi.org/10.1016/j. ympev.2013.02.023 Peloso PLV, Frost DR, Richards SJ, Rodrigues M, Matsui M, Raxwor- thy CJ, Donnellan SC, Byyu SD, Lemmon EM, Lemmon AR, Wheel- er WC (2015) The impact of anchored phylogenomics and taxon sampling on phylogenetic inference in Narrow-mouthed Frogs (An- ura, Microhylidae). Cladistics (2015): 1-28. https://doi.org/10.1111/ cla.12118 Pethryagoda R, Meegaskumbura M, Manamendra-Arachchi K, Schnei- der CJ (2014) Essay 1.2. Amphibian diversity and the case of Sri Lanka’s burgeoning inventory. In: Stuart SN, Hoffmann,M, Chanson JS, Cox NA, Berridge RJ, Ramani P, Young BE (Eds) Threatened Amphibians of the World. Lynx Edicions, IUCN and Conservation International, Barcelona, 18—19. Pyron AR, Wiens JJ (2011) A large-scale phylogeny of Amphibia in- cluding over 2800 species, and a revised classification of extant frogs, salamander, and caecilins. Molecular Phylogenetics and Evo- lution 61: 543-583. https://do1.org/10.1016/j.ympev.2011.06.012 Richards SJ, Gunther R (2019) Three new scansorial species of micro- hylid frogs (Anura: Cophixalus, Oreophryne) from Papua New Guin- ea. Salamandra 55(2): 55-72. http://www.salamandra-journal.com Rivera JA, Kraus F, Allison A, Butler MA (2017) Molecular phyloge- netics and dating of the problematic New Guinea microhylid frogs (Amphibia: Anura) reveals elevated speciation rates and need for taxonomic reclassification. Molecular Phylogenetics and Evolution 112: 1-11. https://doi.org/10.1016/j.ympev.2017.04.008 Suwannapoom C, Sumontha M, Tunprasert J, Ruangsuwan_ T, Pawangkhanant P, Korost DV, Poyarkov Jr NA (2018) A striking new genus and species of cave-dwelling frog (Amphibia: Anura: Microhylidae: Asterophryinae) from Thailand. PeerJ 4422: 1-42. https://doi.org/10.7717/peer).4422 Vieites DR, Wollenberg, EKC, Andreone F, Kohler J, Glaw F, Vences M (2009) Vast underestimation of Madagascar’s biodiversity evi- denced by an integrative amphibian inventory. Proceedings of the National Academy of Sciences of the United States of America 106: 8267-8272. https://doi.org/10.1073/pnas.0810821106 Zweifel RG (1972) Results of the Archbold Expeditions. No. 97. A revi- sion of the frogs of the subfamily Asterophryinae, Family Microhyl- idae. Bulletin of the American Museum of Natural History 148(3): 411-546. zse.pensoft.net 382 Appendix 1. Table Al. Specimens examined. Gunther, R. & Richards, S.: Six new species of Xenorhina Species Location Registration numbers Xenorhina adisca Kraus & Allison, Indonesia: Papua Province: Tembagapura MZB Amph.8403 (holotype) 2003 Xenorhina arboricola Allison & Kraus, Papua New Guinea: West Sepik Province: Mt Menawa BPBM 13747 (paratype) 2000 Xenorhina arboricola Allison & Kraus, | Papua New Guinea: West Sepik Province: Mt Hunstein BPBM 13745 (paratype) 2000 Xenorhina arndti Gunther, 2010 Indonesia: Papua Province: Bomberai Peninsula ZMB 74629-31 (type series) Xenorhina bidens van Kampen, 1909 Xenorhina bouwensi (De Witte, 1930) Indonesia: Papua Province: “Digul-Fluss” Indonesia: West Papua Province: Arfak Mountains ZMA 5705 (holotype) IRSNB 1019 (holotype), plus several specimens collect- ed by R. GUnther between 1998-2008 and stored in the ZMB collection Xenorhina eiponis Blum & Menzies, Indonesia: Papua Province: Eipomek Valley 1989 Xenorhina gigantea van Kampen, Indonesia: Papua Province: Snow Mountains 1915 Xenorhina lanthanites (GUnther & Indonesia: Papua Province: Yapen Island Knop, 2006) AMNH 128234 (paratype) ZMA 5702 (lectotype), ZMA 5703 (paralectotype) ZMB 69557-61 (type series) Xenorhina macrodisca Xenorhina macrops van Kampen, 1913 Indonesia: Papua Province: Wapoga River Headwaters Indonesia: Papua Province: Hellwig Mountains MZB Amph.10916 (holotype) ZMA 5725 (lectotype), ZMA 5726-5728 (paralecto- types) Xenorhina mehelyi (Boulenger, 1898) Papua New Guinea: Central Province: “Vikaiku”, Ang- abunga River Indonesia: Papua Province: Went Mountains Indonesia: Papua Province: Hellwig Mountains Xenorhina minima (Parker, 1934) Xenorhina ocellata van Kampen, 1913 MSNG 29112 (holotype) ZMA 5818 (holotype), ZMA 5817 (paratype) ZMA 5815-16 (syntypes) Xenorhina ophiodon (Peters & Doria, 1878) Xenorhina oxycephala Schlegel, 1858 Indonesia: Papua Province: Hatam, Arfak Mountains Indonesia: Papua Province: Triton Bay MSNG 29129 (lectotype) RMNH 2280A and 2280B (syntypes) (plus several spec- imens collected by R. Gunther between 1998-2008 and stored in the ZMB collection) Xenorhina parkerorum Zweifel, 1972 Xenorhina parkerorum Zweifel, 1972 Xenorhina salawati Gunther, Richards, Tjaturadi & Krey, 2020 Papua New Guinea: Western Province: Imigabip Indonesia: Papua Province: Tenmasigin, Star Mountains Indonesia: West Papua Province: Salawati Island MCZ 81678 (holotype), RMNH 16619 (paratype) MZB Amph.12121-22, 12124-26, 12132, 12134, (type series) Xenorhina tillacki Gunther, Richards & Dahl, 2014 Xenorhina varia Papua New Guinea: Western Province: Muller Range Indonesia: Papua Province: Yapen Island SAMA R65067-68, ZMB 79532 (type series) ZMB 65133-37 (type series) Xenorhina waigeo Gunther, Richards, Tjaturadi & Krey, 2020 Indonesia: Papua Province: Waigeo Island zse.pensoft.net MZB Amph. 12119-20, 12123, 12127-31, 12133, 12155 (type series)