A new species in the Hipposideros bicolor group (Chiroptera: Hipposideridae) from Peninsular Malaysia Author Murray, Susan W. Center for Ecology and Conservation Biology Department of Biology Boston University Boston MA 02215 USA & Corresponding author: E-mail: susan. m 88 @ gmail. com susan.m88@gmail.com Author Khan, Faisal A. A. Faculty of Resource Science and Technology Universiti Malaysia Sarawak 94300 Kota Samarahan Sarawak Malaysia Author Kingston, Tigga Department of Biological Sciences Texas Tech University Lubbock TX 79409 USA Author Zubaid, Akbar Faculty of Science and Technology Universiti Kebangsaan Malaysia Bangi Selangor Malaysia Author Campbell, Polly Department of Integrative Biology Oklahoma State University Stillwater OK 74078 USA text Acta Chiropterologica 2018 20 1 1 29 journal article 21482 10.3161/15081109ACC2018.20.1.001 e95dc0f3-4897-4a17-aaf4-ce56c28f49f0 1733-5329 3944846 Hipposideros kunzi sp. nov. Murray, Khan, Kingston, Akbar, and Campbell Kunz’s bicolored leaf-nosed bat Hipposideros bicolor ( Temminck, 1834 ) , part. Hipposideros atrox ( Andersen, 1918 ) , part. Hipposideros bicolor atrox ( Kitchener et al ., 1996 ) , part. Hipposideros bicolor- 142 ( Kingston et al ., 2001 ) Hipposideros atrox (Douangboupha et al ., 2010 ) Etymology The species is named after Thomas H. Kunz in recognition of his many contributions to the ecology and conservation of bats, and his dedication to the promotion of bat research in Malaysia . Holotype Texas Tech University TTU 108222 (tissue and karyotype TK 152065 ; field number VJS 155), adult ♂ , body in alcohol, skull extracted, collected and photographed by Robert J. Baker on 6 August 2006 during TTU-UNIMAS Sowell Expedition ( Khan et al ., 2008 ). Although the echolocation calls were not recorded for the holotype and the paratypes described here, all of the type specimens had mtDNA haplotypes consistent with the 142 kHz phonic group. This was further supported through comparisons of the noseleaf morphology with that of individuals for which the echolocation call frequency was known. Measurements (in mm) — forearm length: 43.31; fifth, fourth, and third metacarpals lengths, respectively: 32.20, 33.87, 32.88. Length of first and second phalanges of third digits, respectively: 17.47, 16.46; tail length: 25.0; hind-foot length: 7.0; tibia length: 19.70; ear height: 17; body mass: 6.5 g ; anterior noseleaf width: 4.66. Skull measurements are provided in Table 3 . Type locality Bukit Rengit, Krau Wildlife Reserve, Pahang , Peninsular Malaysia (WGS84 03°35’45.6”N , 102°10’ 59.0”E — approximate elevation 72 m ). The specimen was collected using a harp trap set across a trail near the Institute of Biological Diversity at Bukit Rengit. Paratypes Texas Tech University TTU 108417 (tissue and karyotype number TK 152001), adult ♂ ( 4 August 2006 ), dry skin and skull with slight crack in brain case ; TTU 108209 (tissue number TK 152051), adult ♀ ( 6 August 2006 ), dry skin (housed at the Universiti Malaysia Sarawak , but missing) and skull (housed at the Texas Tech University ). Both TTU 108417 and TTU 108209 were captured in Krau Wildlife Reserve ( 03°35’45.6”N , 102°10’59.0”E — elevation 72 m ). Specimen TK 152992, adult ♀ ( 17 May 2008 ), dry skin and skull in Department of Wildlife and National Park ( DWNP ), Malaysia ; specimen TK 153519, adult ♀ ( 20 May 2008 ), alcohol preserved specimen at Universiti Malaysia Sarawak . Both TK152992 and 153519 were collected by FAAK during DWNP biodiversity inventory at Kuala Atok , Pahang , peninsular Malaysia ( 04°16.281’N 102°22.316’E — approximate elevation 85 m ) . Taxonomic notes All specimens previously referred to H. atrox ( Douangboubpha et al ., 2010 ) and H. bicolor -142 are here referred to H. kunzi sp. nov. Based on length of forearm, Hill (1963) likely included both H. bicolor and H. kunzi as H. bicolor atrox , although the majority of these individuals are probably H. kunzi based on length of forearm (p. 29, Fig. 4 ). We cautiously assign the individuals of H. bicolor atrox from both Hill et al . (1986) and Zubaid and Davison (1987) to H. kunzi . It is unclear where the bats were collected, but it is suggested they were captured in Northern peninsular Malaysia , which would suggest that they are indeed H. kunzi . In his description of the new species Hipposideros gentilis , Andersen (1918) described the new subspecies H. g. atrox as having a wide range of forearm lengths that span both H. bicolor and H. kunzi : 42–46.2 mm ( Andersen, 1918: 380 ). Thus he likely measured both individuals of H. bicolor and H. kunzi for the subspecies description. Description This is a small to medium-sized hipposiderid bat in the H. bicolor group with a forearm length ranging from 38.8 to 45.6 mm (mean = 42.9 mm ± 0.9), tibia length of 17.1 to 20.6 mm (mean = 18.8 mm ± 0.5), and mass varying from 6.0 to 12.0 g (mean = 8.5 g ± 0.9 — Table 2 ). The dorsal pelage varies from medium or dark brown to bright orange, but is always bicolored with a white base. The ventral pelage ranges from buff or golden, to bright orange ( Fig. 9 ). The wing and tail membranes are dark brown, as are the ears. The ears are large (mean = 17.6 mm ± 0.6) and rounded with a bluntly pointed tip. The noseleaf lacks supplementary lateral leaflets and has an internarial septum that is generally triangular in shape (wider at the base — Fig. 9 ). The posterior and anterior portions of the nose are dark brown-grey in color, while the central part of the noseleaf is more flesh colored. The tail is long (mean = 28.7 mm ± 1.8), extending the full length of the uropatagium. The fifth metacarpal is about 74% of forearm length and the first phalanx of the third digit is about 53% of third metacarpal. Echolocation call frequency of the CF component ranges from 133.2 to 147.5 kHz, with a mean call frequency of 143.1 ± 2.0 kHz ( Fig. 5 and Table 2 ). Hipposideros kunzi has a small and elongate skull with the greatest length of skull (GSL) ranging from 17.69 to 19.13 mm (mean = 18.31 ± 0.33 mm ). The skull is slightly wider across the zygomata (mean = 9.2 ± 0.2 mm ) compared to across the mastoids (mean = 9.2 ± 0.2 mm — Table 3 ). The distal process of the jugal bone is low and not well defined ( Fig. 6 ). The rostrum is well developed with six nasal inflations. The sagittal crest is well developed and is taller more anteriorly. The constriction behind the orbits is well defined and narrower than the rostrum. The upper toothrow is shorter (CM 3 mean = 6.3 ± 0.1 mm ) than the lower (CM 3 mean = 6.8 ± 0.1 mm ). The upper incisor is small and both the upper and lower canines are of moderate size. The upper premolar (P 2 ) is minute and extruded from the toothrow, while the lower premolar (P 2 ) is about half the height of the second premolar (P 4 ). The species is sexually dimorphic with respect to magnitude of certain skull measurements: despite being smaller than females, males have longer and taller skulls and longer canines. Comparisons with similar species Hipposideros kunzi is one of several Hipposideros species described from the Indo-Malayan region, which superficially resemble H. bicolor and lack supplementary leaflets adjacent to the noseleaf. In peninsular Malaysia and southern Thailand , H. kunzi most closely resembles, and is easily confused with, both H. bicolor and H. pomona . Compared to H. bicolor , H. kunzi has a higher echolocation call frequency ( Table 2 ), is generally smaller in body size ( Table 2 ), and has a shorter but wider skull ( Table 3 and Fig. 9 ). In addition, H. kunzi has a narrower anterior noseleaf ( Holotype : 4.66 mm ) that is slightly curved upwards compared to H. bicolor , which has a wider anterior noseleaf ( 4.94−5.46 mm , n = 5) that is flattened and square in appearance ( Kingston et al ., 2006 ), lighter in color, and has rudimentary supplementary lateral leaflets ( Fig. 9 ). The noseleaf characters, however, are only useful if both species are available for comparison in the field. Based on appearance ( Murray et al ., 2012 : figure S1), echolocation call frequency ( H. pomona : 136.4 –139.4 kHz, n = 3), overall size ( H. pomona length of forearm: 42.7–44.8 mm , n = 3), and skull size and shape ( Fig. 7 ), it is very difficult to distinguish H. kunzi from H. pomona . The main morphological difference between these species is ear height, with H. pomona having a much larger ear compared to H. kunzi : 20.0– 21.5 mm ( n = 3) versus 15.0– 19.5 mm (mean = 17.6 mm — Table 2 ), respectively. Hipposideros pomona and H. kunzi , however, are not closely related based on both mitochondrial and nuclear DNA ( Murray et al ., 2012 ; this study). Despite being sister taxa ( Fig. 2 ), having similar appearance, and overlapping in echolocation call frequencies ( Kingston et al ., 2000 ), individuals of H. kunzi and H. cineraceus -B are easily distinguished using body size ( H. kunzi being larger; Table 2 ) and nose morphology: H. cineraceus -B has a small swelling in its internarial septum ( Fig. 9 ). Reproduction In both 2003 and 2004 in peninsular Malaysia , palpably pregnant females were captured in February and March, and lactating individuals were captured from April through September. Similarly, Nurul-Ain et al . (2017) found females from Krau Wildlife Reserve and Samad Cave (ca. 10 Km from Krau) to be seasonally monestrous, with a peak in pregnancy in March, and lactation in June, although lactating females were captured from April through October. Distribution , ecological notes , and conservation status Currently, H. kunzi has only been documented on the Malay Peninsula, between 3°12’N in peninsular Malaysia ( Fig. 1 , site 12) and the Isthmus of Kra at 10°41’N in Southern Thailand (this study; Douangboubpha et al ., 2010 ). Despite extensive sampling, Douangboubpha and colleagues did not capture H. kunzi in Central or Northern Thailand , suggesting that the northern limit of this species’ range is restricted to the Sundaic biogeographical region, as delimited by the Isthmus of Kra ( Douangboubpha et al ., 2010 ). While we did not sample bats in the southern tip of peninsular Malaysia , we expect that H. kunzi should occur throughout the peninsula where suitable habitat exists. Lim et al . (2014) reported a positive correlation between the abundance of H. kunzi (as H. bicolor -142) and latitude across 15 forest sites in peninsular Malaysia , with few or no captures at sites in the southern third of the Peninsula (which may be attributable to the lack of karst). In Singapore , H. bicolor (= H. kunzi ) is considered locally extinct due to habitat loss ( Pottie et al ., 2005 ). Douangboubpha et al . (2010) included Sumatra in the distribution of H. atrox (= H. kunzi ), but because of the high level of cryptic diversity within this group it is impossible to determine whether individuals from Sumatra are conspecific with H. kunzi without genetic data. Based on limited sampling in Borneo ( Fig. 1 ), there is currently no evidence that H. kunzi occurs in Borneo. In peninsular Malaysia , individuals of H. kunzi were captured at all sampling sites ( Fig. 1 ) and were relatively common and widespread in karst regions, but were also common in some non-karst areas (e.g., Krau Wildlife Reserve). Colonies ranged in size from a few individuals to several hundred and were found in caves, mines, and rock crevices. Colonies of H. kunzi were almost always found in caves housing other bat species; these included H. cervinus , H. larvatus , H. armiger , Rhinolophus malayanus , R. stheno , Myotis siligorensis , M. ater , Miniopterus medius , and Taphozous melanopogon . Based on captures and wing morphology, H. kunzi is believed to forage in forested habitats; Douangboupha et al . (2010) suggested that H. kunzi forages in diverse forest types and may be somewhat tolerant of anthropogenically modified landscapes that retain vegetative structure (e.g., secondary forest, rubber and orchard plantations). Given the species’ distribution across the Malay peninsula into Southern Thailand , widespread occurrence and local abundance, we currently recommend H. kunzi be evaluated as a species of Least Concern, following IUCN Red List Categories and Criteria v. 3.1 ( IUCN, 2012 ). Loss and disturbance of caves and foraging habitats would support a higher category of risk. We and others have noted the high levels of cryptic diversity in Hipposideros (e.g., Esselstyn et al ., 2012 ; Murray et al ., 2012 ; Foley et al ., 2017 ). We hope that our taxonomic delineation of a new member of the bicolor species group, H. kunzi , will motivate further efforts to resolve the taxonomy of remaining cryptic lineages. Such efforts are essential to the conservation of the remarkable diversity that exists within this already speciose genus.