Guibemantis sioka Hutter, Andriampenomanana, Basham, Glaw, Masotti, Lambert & Vences, 2025, sp. nov.

Guibemantis sioka sp. nov. Figs 2, 3, 4; Table 1 Holotype. Adult male, KU 336913 (field number JJW2366), collected on 24 January 2013 at Maharira, Ranomafana National Park (geographical coordinates, 21.3312°S, 47.4106°E, 1330 m a.s.l.), south-eastern Madagascar, by C.R. Hutter, Z.F. Andriampenomanana, and S.M. Lambert. Paratype. A male, probably subadult or not fully grown adult, ZSM 319/2005 (field number FGZC 2434), collected on 28 January 2005 in Andohahela National Park (geographical coordinates ca. 24.5440°S, 46.7141°E, ca. 1650 m a.s.l.), south-eastern Madagascar, by F. Glaw, M. Vences and P. Bora. Additional material. One probable male (not collected; field observation number MA130), captured, photographed, and swabbed by E. W. Basham and M. D. Masotti on 29 October 2023 at Analamary, Ivohiboro Reserve (22.6176°S, 46.7202°E, 1210 m a.s.l.), south-eastern Madagascar. Available names. As discussed by Vences & Glaw (2005) and Vences et al. (2015), there are two available earlier names that need to be considered for new species in the subgenus Guibemantis: Rhacophorus mocquardii Boulenger, 1896, and Mantidactylus acuticeps Ahl, 1929. Both these nomina are currently seen as junior synonyms of Guibemantis depressiceps, and we continue with this taxonomic interpretation based mainly on biogeographic arguments. The nomen acuticeps was described by Ahl (1929) from Central Madagascar, and mocquardii was collected at Sahembendrana, a site likely close to Betampona in the Northern Central East (see Vences et al. 2022b). The flank color pattern of mocquardii (“flanks brown, with three white spots”; Boulenger 1896) according to the original description does not convincingly agree with the distinct vermiculation of the new species whereas the flank color pattern of acuticeps may be similar to that of the new species (“Körperseiten weiss und schwarz marmoiert [sic]” = sides of the body marbled black and white; Ahl 1929). Also, the new species has a very large relative hand size (ratio HAL/ SVL 0.365 in the holotype; 0.327 in the probably not fully grown paratype from Andohahela) which differs from the adult type specimen of mocquardii (0.315). The adult type of acuticeps (HAL/ SVL 0.333) also has a shorter hand than the holotype of the new species, although this value is similar to that of the paratype of the new species. Both acuticeps and mocquardii have relative hand sizes in the range typical for G. depressiceps (see Diagnosis below; morphometric data from Vences & Glaw 2005; Vences et al. 2015). The new species also differs from the type of mocquardii by an almost complete lack of webbing on the hands (vs. presence of distinct traces in mocquardii, and traces also weakly recognizable in acuticeps; see Vences et al. 2015). Furthermore, the new species has not been found in the Northern Central part of Madagascar from where the types of acuticeps and mocquardii were likely collected, and the new species also appears to be much rarer than G. depressiceps (which is widespread and common in the Northern Central East and Southern Central East). This combined evidence suggests that neither of the two names, acuticeps and mocquardii, can readily be assigned to the new species described herein and should rather continue being considered junior synonyms of G. depressiceps. Etymology. The specific epithet sioka is a Malagasy word meaning “chirp”. The name was chosen to describe the distinct call this species makes, which reminds one of the sound of a bird chirping or emitting trills. The name is to be treated as an invariable noun in apposition. Diagnosis. Guibemantis sioka sp. nov. is allocated to the subgenus Guibemantis in the mantellid genus Guibemantis based on presence of intercalary elements between ultimate and penultimate phalanges of fingers and toes (verified by external inspection), the structure of femoral glands in males (of type 1 sensu Glaw et al. 2000; Vences et al. 2007), arboreal habits, partly or fully white subgular vocal sac, dorsal skin not tubercular and without dorsolateral ridges, enlarged terminal disks of fingers and toes and body size of males up to 43 mm, and phylogenetic relationships. The new species is distinguished from all species of the subgenus Pandanusicola in the genus Guibemantis by larger body size (adult male SVL 36–43 mm vs. <34 mm in males of Pandanusicola, and <31 mm in females of Pandanusicola except G. flavobrunneus; Glaw & Vences 2007; Lehtinen et al. 2011, 2012; Vences et al. 2013; Koppetsch et al. 2023; Gabriel et al. 2024), as well as by numerous details of coloration, natural history, and bioacoustics (where advertisement calls are known). Morphological distinction of the new species within the subgenus Guibemantis is more difficult. It differs from G. diphonus by larger body size (male SVL 36–43 mm vs. 34−36 mm) and relatively larger hands (ratio HAL/SVL in the adult holotype 0.365 vs. <0.33); from G. kathrinae by smaller body size (SVL 36–43 mm vs. 57−59 mm in verified, topotypical specimens of G. kathrinae); from most G. tornieri by smaller body size (36−43 mm vs. 42−58 mm); from G. timidus by relatively larger hands (ratio HAL/SVL 0.327 –0.365 vs. <0.32); and from all these species by a vermiculated and contrasted whitish / blackish pattern covering the entire flanks, without yellow elements (vs. lack of vermiculated pattern and often yellow color on posterior flanks in G. diphonus, G. timidus and G. tornieri; and blackish flanks with white spots in G. kathrinae). The new species is most similar to G. depressiceps from which it differs by a more contrasted (vs. often poorly contrasted) vermiculated flank pattern, and by larger hands (ratio HAL/SVL of the adult holotype 0.365 vs. 0.298 –0.351 in 19 adult specimens of G. depressiceps measured by Vences & Glaw 2005; note however that the relative hand length of the not fully-grown paratype of the new species, 0.327, is within the range of G. depressiceps). The advertisement call of the new species (Fig. 5–6) differs from all other species in the subgenus Guibemantis by the distinct differences in pulse duration between first and last pulse, and by the longer pulse duration of the last pulse (up to 55 ms). The last pulse is usually much shorter in other species of the subgenus and especially in G. depressiceps as can be seen on the comparative oscillograms (pulse duration of other species have however not been measured in detail due to difficulties in clearly separating pulses from silent intervals). Furthermore, the call differs from those of G. tornieri and G. kathrinae by a much faster repetition rate of 23–29/s (vs. 12–21/s), from that of G. diphonus by absence of a second note type (vs. presence), and from all other species by the rather tonal “trill” structure with multiple distinct (harmonic) frequency bands (vs. more pulsatile without distinct frequency bands) and a higher dominant frequency at 2497 Hz (vs. <2000 in all other species except G. diphonus; Fig. 6). Description of the holotype. Specimen in good state of preservation, with right thigh muscle removed for tissue sample (Fig. 4). For full morphometric measurements see Table 1. Adult male, SVL 43.0 mm. Body moderately slender; head longer than wide, slightly wider than body; snout rounded in dorsal, ventral, and lateral views; nostrils much nearer to tip of snout than to eye, slightly protuberant and pointed anterolaterally; canthus rostralis distinct, straight; loreal region slightly concave; tympanum distinct, its diameter 47% of eye diameter; distinct supratympanic fold; tongue ovoid, strongly bifid posteriorly; posterior tongue extensions slightly serrated; vomerine teeth quite distinct, forming an elongated aggregation posterolateral of each choana; choanae rounded. Forelimbs slender; subarticular tubercles distinct and single; central metacarpal tubercle rounded but not very prominent, outer metacarpal tubercle more ovoid than the central one but of similar length; a small and distinct prepollex (which could also be considered as an inner metacarpal tubercle) at base of first finger. Fingers largely without webbing, only a minimal trace of webbing recognizable between fingers I and II; relative finger length I<II<IV<III; finger discs distinctly enlarged; nuptial pads absent. Hind limbs long and slender; when adpressed along body, the tibiotarsal articulation reaches between eye and nostril; lateral metatarsalia separated by webbing; inner metatarsal tubercle distinct, ovoid, larger than outer; outer metatarsal tubercle distinct, small and rounded; webbing formula of the foot 1(1), 2i(3), 2e(1), 3i(2), 3e(1), 4i(2.25), 4e(2.25), 5(1); relative toe length I<II<III <V<IV. Skin dorsally smooth with many small spicules across the dorsum; ventral skin smooth with some granulations on posterior belly and in inguinal region. Femoral glands recognizable from external view, distinct and coarsely granular, consisting of many small gland granules in a relatively diffuse field covering most of thigh ventrally, thus of type 1 as defined by Glaw et al. (2000). In life (Fig. 2), the specimen had a dark brown color on most of the dorsal surface, with a broad and sharply delimited yellow-beige vertebral stripe, darker brown color in the tympanic region, and a narrow yellowish-beige stripe running from the snout tip over the nostrils to the eyes, and along the supratympanic fold. Five slightly darker brown incomplete crossbands dorsally on thigh and three on shank; irregular darker brown markings also dorsally on forelimbs. Flank with rather sharply delimited pattern of gray-whitish vermiculation on a dark brown to blackish color. Upper part of the iris yellowish-beige, lower part of the iris dark brown. Ventrally white on throat changing to a darker beige towards the posterior part of the body, with dense irregular dark spotting especially on chest and belly. Limbs ventrally dark brown. After 11 years in preservative (Fig. 4), the pattern remains identical, with some e