Bioacoustics of Hipposideros galeritus

Echolocation : The constant frequency ( CF ) component of echolocation calls emitted by our bats ranged between 84–85 kHz ( Table 2 ; Appendix 4 ). As tabulated in Srinivasulu et al . (2025) and our own data, only populations of H. galeritus found in the Thai-Malay peninsula and Bokor National Park, Cambodia (close to the AG1) are known to emits calls at similar frequencies (80.7–92.0 kHz and 86–87 kHz respectively), whereas populations elsewhere in mainland Southeast Asia and on Borneo produce significantly higher constant frequency components (typically ≥ 98–110 kHz). Genetics: Consistent with previous studies ( Francis et al. 2010 ; Murray et al. 2012 ; Srinivasulu et al. 2025 ), our Bayesian trees based on COI and Cytb datasets strongly support the monophyly of Southeast Asian H. galeritus s.l. (PP=1). In our COI tree, the H. galeritus s.l clade is subdivided into three well-supported geographic subclades (PP=1): A– insular Sundaic (Borneo ( Kalimantan , Indonesia and Sarawak , Malaysia ) + Java), B– mainland Sundaic or the Thai-Malay Peninsula, and C—Indochinese comprising Thailand excluding the peninsula, Cambodia , Laos , Vietnam , including our specimens in the VMD. In our Cytb tree, bats from insular Sundaic and Indochinese subregions appeared in two divergent and highly supported lineages (PP=1) ( Figure 6 ). Genetic divergences based on COI / Cytb sequences within three recognized subclades A, B, C Southeast Asian H. galeritus s.l. are between 0– 4.9% / 0.3–2.8%, 0.3/–, and 0–2.7 / 0.2–0.9%, respectively; distances between these clades are between 11.4–14.6% / 10.7–11.5% and comparable with interspecific variations in the Hipposideros genus (Appendix 7).

Taxonomic notes: Hipposideros galeritus was historically regarded as a widespread species across the Indomalayan Region, comprising four geographic subspecies: H. g. galeritus (distributed in the Thai-Malay peninsula, including Tarutao, Penang, and Bangka islands, Sumatra, SE Thailand , S. Laos , Cambodia , and C-S. Vietnam ), H. g. brachyotus Dobson, 1874 ( India , Bangladesh , Sri Lanka ), H. g. insolens Lyon, 1911 (endemic to Borneo), and H. g. longicauda Peters, 1861 (restricted to Java) ( Corbet & Hill 1992 ; Monadjem et al. 2019 ; Simmons 2005 ). Based on analyses of COI gene sequences, morphological and acoustic data, Srinivasulu et al . (2025) revised this classification by: (1) re-elevating brachyotus to species rank for Indian populations, (2) describing the Sri Lankan population as H. srilankaensis , and (3) concluding that Southeast Asian H. galeritus s.l. represents several distinct species. This conclusion aligns with earlier studies ( Francis et al. 2010 ; Murray et al. 2012 ) which found substantial genetic divergence and distinct echolocation call frequencies among allopatric populations (e.g., ~115 kHz in Borneo, 89.1 ± 1.5 kHz in Peninsular Malaysia , ~108 kHz in Laos ), supporting their status as evolutionarily and ecologically distinct species. However, Srinivasulu et al . (2025) interpreted the observed call frequency variation across Southeast Asian COI lineages as vocal plasticity. Our genetic results confirm deep geographic structuring within Southeast Asian H. galeritus s.l. but delimiting these taxa and elucidating their evolutionary history remains challenging. For instance, despite similar isolation by marine barriers, COI p-distances between one Javan and two Bornean bats were unexpectedly low (4.7–4.9%), contrasting sharply with higher divergences between insular and continental Southeast Asian populations (11.4– 14.6%; Appendix 7). These distances exceed maximum intraspecific variations in the Indochinese H. galeritus ( pdistances ≤2.7%) and interspecific variations between certain pairs of sister species such as 3.3% between H. kunzi and H. kingstonae or 3.5% between H. rotalis and H. khaokhouayensis . However, they are lower than the intraspecific variation between Bornean and Peninsular Malaysian populations of H. cervinus (5.9–6.2%; Appendix 7). This pattern suggests allopatric populations in insular Southeast Asia either represent deeply structured populations of a single species or recently diverged sister species ( Francis et al. 2010 ; Murray et al. 2012 ). Likewise, our low-frequency emitters in the VMD (AG1: 84–85 kHz) are phylogenetically nested within the high-frequency Indochinese clade (~110 kHz), yet genetically divergent from the low-frequency emitters of the Thai-Malay Peninsula (mainland Sundaic subregion). While call plasticity may be a plausible explanation for frequency variation in the Indochinese subregion clade, these mtDNA phylogenetic patterns can stem from either incomplete lineage sorting of ancestral polymorphisms after diversification or introgression following secondary contact ( Ballard & Whitlock 2004 ; Funk & Omland 2003 ). Long branches in the mtDNA phylogenetic tree, especially among the two phonic forms (excluding individuals showing potential introgression) indicate that these lineages have had substantial time to evolve independently after their divergence from a common ancestor. Instead, the “low-frequency” bats carrying mtDNA haplotypes of “high-frequency” congeners was not found randomly but specifically in parapatric zones (including AG1 and KG1-3 [ Figure 1 ] and in Bakor National Park, Cambodia ), suggesting historical or ongoing introgression between previously isolated incipient species following secondary contact(s) ( Mao et al. 2010 , 2013 ; Zhou et al. 2023 ). As proposed by Srinivasulu et al. (2025) and supported by regional phylogeographic studies ( Khan et al. 2010 ; Tu et al. 2017a , 2018b ), the evolutionary history of taxa within the H. galeritus s.l. complex in mainland Southeast Asia was likely shaped by climatic fluctuations and associated changes in forest range and sea levels during the Pliocene–Pleistocene. In line with this hypothesis, cycles of forest contraction and expansion, along with the submersion and re-emergence of the Isthmus of Kra due to sea-level oscillations, may have facilitated allopatric speciation, followed by secondary contact and asymmetric introgressive hybridization. Given the limitations of our study (incomplete geographic sampling, reliance solely on mtDNA markers), future research incorporating nuclear DNA data is needed to (1) test gene flow hypotheses and clarify species boundaries within mainland Southeast Asia; and (2) resolve the taxonomy and phylogeography of insular Sundaic populations (Borneo, Sumatra, Java, Sulawesi ).