Bioacoustics of Rhinolophus thailandensis

Echolocation: The constant frequency ( CF ) component of echolocation calls emitted by our bats (53.1–54.9 kHz) is comparable to that of Cambodian R. thailandensis (53.6 kHz: Chheang et al . 2013 ) ( Table 2 ; Appendix 4 ). Genetics: Consistent with previous studies ( Francis et al. 2010 ; Mao et al. 2010 ; Zhou et al. 2023 ), our COI sequence analyses support the monophyly of “ pearsonii group”. Within this species group, multiple monophyletic lineages (PP=0.75–1.0)—differing from each other by ≥ 2.7% (p-distance)—can be identified ( Figure 4 ). Among these, our two tentatively identified R. thailandensis individuals, one R. yunanensis ( = R. thailandensis sensu Francis 2019 ) from southern Thailand , and three R. chiewkweeae from Peninsular Malaysia form a monophyletic lineage (PP=1), which diverged by ≥9.3% p-distance from the other lineages including samples from China to central Vietnam , Laos and northeastern Thailand . Within this group, our bats occupy a basal position and are divergent from the south Thailand and Malayan counterparts by 1.5% and 2% p-distances, respectively ( Appendix 5 ).

Morphology: Our specimens were identified as members of the R. pearsonii group, which includes four strikingly similar species— R. pearsonii Horsfield, 1851 , R. yunanensis Dobson, 1872 , R. thailandensis Wu, Harada & Motokawa, 2009 , and R. chiewkweeae Yoshiyuki & Lim Boo Liat, 2005 —based on the following morphological characters: Externally, they are medium-sized horseshose bats with an FA of 51.7–52.8 mm ( Table 2 ). The pelage is long and woolly, dorsally orangish-brown and ventrally lighter. The noseleaf structure includes a lancet that is broadly triangular and long, with nearly parallel, straight sides; the connecting process is low and broadly rounded in lateral view. The sella is relatively blunt at the tip, broader at the base, with a characteristic medial constriction, and lacks basal lappets. The horseshoe is wide, covering the muzzle, and the ears are moderately large ( Figure 2B ). Characters R. thailandensis R. cf. macrotis R. marshalli H. kingstonae H. galeritus FA 52.1±0.4; 10 51.7–52.8 42.9±1.1; 6 41.8–44.5 44.2±1.7; 3 42.5–45.9 38.2±1.0; 14 37.0–40.3 48.1±1.0; 14 46.5–50.0 T 20.9±1.7; 10 18.4–23.2 20.7±1.8; 6 19.5–23.8 21.1±0.1; 3 21.0–21.1 26.3±1.3; 14 24.0–28.0 36.5 ±0.5; 3 36.0–37.0 E 26.5±0.8; 10 25.3–27.5 23.5±0.3; 6 23.3–23.8 25.3±0.7; 3 24.8–25.8 15.9±0.7; 14 15.1–17.5 15.9±0.8; 5 15.3–16.5 Tib 25.4±0.4; 10 25.0–26.0 18.2±0.7; 6 17.3–18.9 19.1±0.5; 3 18.7–19.3 16.6±0.7; 14 15.6–17.7 19.7±0.4; 7 19.1–20.2 GLS 25.39±0.47; 5 24.89–26.00 17.62±0.13; 3 17.48–17.73 17.61 15.75±0.07; 5 15.64–15.82 17.66 CCL 21.79±0.22; 5 21.60–22.13 15.57±0.16; 3 15.38–15.69 15.59 13.26±0.11; 5 13.11–13.41 14.96 ZB 12.55±0.19; 5 12.26–12.75 8.22±0.18; 3 8.05–8.40 8.23 7.63±0.40; 5 7.33–8.33 8.38 MB 11.00±0.19; 5 10.72–11.20 8.70±0.08; 3 8.63–8.75 8.96 8.06±0.30; 5 7.54–8.28 9.08 CC 6.64±0.23; 5 6.42–7.03 3.88±0.07; 3 3.81–3.95 3.78 3.20±0.16; 5 2.98–3.40 3.33 MM 9.06±0.03; 5 9.04–9.12 5.64±0.08; 3 5.56–5.71 5.48 5.23±0.11; 5 5.04–5.30 5.60 UCM 9.81±0.16; 5 9.65–10.08 6.19±0.09; 3 6.09–6.23 6.39 4.94±0.12; 5 4.74–5.07 5.62 ML 17.23±0.39; 5 16.85–17.86 11.50±0.04; 3 11.46–11.54 11.27 8.96±0.10; 5 8.88–9.12 10.17 LCM 10.78±0.22; 5 10.61–11.15 6.60±0.04; 3 6.57–6.64 6.45 5.33±0.08; 5 5.22–5.45 5.96 FmaxE (kHz) 53.1–54.9 56.2–57.6 45.4–46.2 142–144 84–85 Craniodentally, the skull is large with a GLS of 25.39±0.47 (24.89–26.00) mm and CCL of 21.79±0.22 (21.60– 22.13) mm and robust; the zygomatic width exceeds mastoid width. The rostrum profile is relatively low, and the frontal depression is shallow. Both the sagittal and supraorbital crests are well developed. The anterior median swellings are prominent, whereas the lateral swellings are less pronounced and the posterior swellings are relatively concave. The upper canine (C¹) is strong and well developed. The second upper premolar (P²) is small and situated within the tooth row. The second lower premolar (P 3 ) is small, slightly extruded from the toothrow ( Figure 3A ; Table 2 ). Within the R. pearsonii group, R. pearsonii includes two recognized subspecies: R . p. pearsonii occurring across South Asia, China , and mainland Southeast Asia (excluding Cambodia and the VMD ), whereas R . p. chinensis K. Andersen 1905 is endemic to eastern China ( Bates et al. 2019 ; Hutson et al. 2019 ). Rhinolophus yunanensis is found widely across China , India , Myanmar , and Thailand (excluding southern Thailand ) ( Bates et al. 2019 ), whereas R. thailandensis has been recorded from northwest and southern Thailand (close to the border between Thailand and Malaysia ) to southern Cambodia (Bokor National Park [ 10.695841ºN , 104.050246ºE ]) ( Bouillard 2020 ; Zhou et al. 2023 ) and R. chiewkweeae is endemic to the south of the Isthmus of Kra ( Waldien 2019 ). Given this distribution, our bats from An Giang province (southern Vietnam ) are tentatively conspecific with R. thailandensis in Bokor National Park, Cambodia . This preliminary identification is supported by: 1) geographic proximity (~ 100 km distance) and lack of physical barriers between the two sites; 2) high overlap with published measurements for Cambodian R. thailandensis (= R. pearsonii in Chheang et al . 2013 ) ( FA : 51.7–52.8 mm vs. 51.5–53.6 mm ; CCL: 21.60–22.13 mm vs. 21.30–22.02 mm ; ( Table 2 ).

Taxonomic notes: While chromosomal differences among Rhinolophus pearsonii , R. yunanensis , and R. thailandensis (species with overlapping distributions) were reported by Zhou et al . (2023) , no cytogenetic data exist for R. chiewkweeae . Examination of additional specimens reveals extensive overlap in mensural data between the geographically adjacent sister species R. thailandensis and R. chiewkweeae (e.g., FA: 51.5–61.0 mm vs. 51.9–56.9 mm ; CCL: 21.30–24.80 mm vs. 21.30–23.50 mm ) and echolocation frequencies (FmaxE: 49.0–53.6 kHz vs. 53.6– 54.7 kHz) ( Chheang et al. 2013 ; Francis 2019 ; Hutson et al. 2019 ; Morni et al. 2016 ; Wu et al. 2009 ; Yoshiyuki & Lim 2005 ). This phenetic overlap undermines the reliability of morphological and acoustic characters proposed to distinguish the two species; a situation likely attributable to limited specimen availability at the time of initial description, as observed in other rhinolophid pairs such as R. rex and R. paradoxolophus ( Tu et al. 2023 ) . Consistent with this, our COI sequence analyses—while supporting previous indications of higher cryptic diversity within the pearsonii group ( Francis et al. 2010 ; Zhou et al. 2023 )—found low genetic divergence ( p -distance ≤ 2%) between allopatric populations currently classified as R. thailandensis (e.g., VMD, southern Thailand ) and R. chiewkweeae (W Malaysia ). This low genetic divergence suggests that R. thailandensis would represent a junior synonym of the older name R. chiewkweeae , should further evidence confirm their conspecificity.