Additional records of small mammals collected from Nepal

Small mammals were surveyed in areas of the Annapurna Conservation Area, Shey Phoksundo National Park, and the Manaslu Conservation Area between 2013 and 2015. We collected 117 specimens representing 10 species of small mammals from two orders, Soricomorpha and Rodentia. We report on localities sampled from Central Nepal expanding known distributions and reporting new sampling localities for Apodemus gurkha Thomas, 1924; Apodemus pallipes (Barrett-Hamilton, 1900); Mus musculus Linnaeus, 1758; Niviventer fulvescens (Gray, 1847); Rattus rattus (Linnaeus, 1758); Rattus pyctoris (Hodgson, 1845); Rattus nitidus (Hodgson, 1845); and Soriculus nigrescens (Gray, 1842). Specimens collected from the Manaslu Conservation Area, which had not been previously surveyed, docu-mented the occurrence of four species between previously known collection localities in the Kali Gandaki Valley and the Langtang and Khumbu valleys. Distribution records of Episoriculus soluensis Gruber, 1969 in Nepal are clarified for this recently recognized species.


Introduction
provided a history of the mammalian surveys conducted in Nepal between the early 1820s and 2001. Brian Houghton conducted some of the earliest surveys of mammalian biodiversity in Nepal (Hodgson 1832(Hodgson , 1835(Hodgson , 1836a(Hodgson , 1836b(Hodgson , 1840(Hodgson , 1841a(Hodgson , 1841b(Hodgson , 1841c(Hodgson , 1841d(Hodgson , 1842(Hodgson , 1845(Hodgson , 1846, collecting more than 370 specimens representing 70 genera and 114 species; of these specimens 40 are type material (Mitchell 1975). Hinton and Fry (1923) and Fry (1925) cataloged the specimens collected on three expeditions organized by the Bombay Natural History Society in the early 1920s. The results of later surveys by the Smithsonian Institution and the Zoological Survey of India were reported by Johnson et al. (1980) and Biswas et al. (1955Biswas et al. ( , 1957, respectively. A number of surveys conducted in the mid-1900s focused on ectoparasites associated with small mammals (Gruber 1969;Weigel 1969;Martens and Niethammer 1972;Mitchell and Derksen 1976;Mitchell 1979Mitchell , 1980. Abe (1971Abe ( , 1977Abe ( , 1982 surveyed small mammals in the Langtang andAnnapurna regions between 1968 and1975, and several smaller surveys for small mammals were conducted in the late 1900s and the early 21 st century by Gregori and Petrov (1976), Daniel and Hanzák (1985), Newton et al. (1990), and Mekada et al. (2021). Between 2013 and 2015, Poulton (2019) live-trapped small mammals (n = 792) from four geographic localities and confirmed the identities of 18 taxa with cytochromeb sequences, and between 2014 and 2016, Adhikari et al. (2016) live-trapped small mammals (n = 221) from urban and wildlife habitats in Kathmandu, Pokhara, and Lumbini.
Three species of mammals are endemic to Nepal: Myotis csorbai Topál, 1997 (Csorba's Mouse-eared Bat); Apodemus gurkha Thomas, 1924 (Nepalese Field Mouse); and Neodon nepalensis Pradhan, Sharma, Sherchan, Chhetri, Shrestha & Kilpatrick, 2019 (Nepalese Mountain Vole). Apodemus gurkha is known from a small area in Central Nepal where it has been collected from a few localities in both the eastern and western temperate and coniferous forest (Mitchell 1977;Musser et al. 1996;Pearch 2011). The Nepalese Field Mouse has been classified as Endangered by the IUCN due to its small range and habitat loss and fragmentation (Molur et al. 2005).
Fieldwork was conducted in Central Nepal in the summers of 2013-2015 with the primary focus of collecting additional material from Nepalese Feld Mice for phylogenetic and phylogeographic analyses of molecular data generated from this new material. During this fieldwork, nine species of non-volant small mammals were collected in addition to 48 specimens of the Nepalese Field Mouse. Examination of the collected voles has resulted in the detection of a previously undescribed mountain vole, Neodon nepalensis, from western Nepal (Pradhan et al. 2019). Both volant and terrestrial small mammals are poorly sampled from Nepal, and Jnawali et al. (2011) indicated that 48% of Nepalese small mammals have an IUCN status of Data Deficient. Thus, our objective here is to report the distributional data obtained from this fieldwork.

Study Area
Annapurna Conservation Area (ACA). This is the largest protected area in Nepal (7,629 km 2 ) and is located in the Annapurna range of the Himalayas at elevations ranging from 790 to 8,091 m a.s.l. Three areas were surveyed ( Fig. 1), including: (1) Banthanti, Kaski District (28.3682°N, 083.7315°E, 2330 m a.s.l.)-a village with terraced vegetable fields and forests with rhododendron (Rhododendron L. spp.), oak (Quercus L. spp.), other broad-leaved trees, and some conifers. The forested area had moderate numbers of moss-covered rocks, dense undergrowth, and several streams, likely due to monsoon rains.
(8) Lho, Gorkha District (28.5741°N, 084.7031°E, 3180 m a.s.l.)-an area of mixed coniferous-deciduous forest (Tsuga, Pinus, Quercus, and Betula spp.) located on a steep slope near the village. Leaf litter was present at this site with moderate herbaceous ground cover.

Shey Phoksundo National Park (SPNP)
. This is the only National Park in Nepal. It encompasses 3,555 km 2 at elevations between 2,130 and 6,885 m a.s.l. in the Trans-Himalayas of northwestern Nepal. Three areas were surveyed ( Fig. 1), including: (10) Renchi, Dolpo District (29.1173°N, 082.8873°E, 3030 m a.s.l)-an area predominantly with coniferous forest (Pinus spp. and Picea L. spp.) and a few hardwoods. Understory was sparsely vegetated with ferns and other herbaceous plants.

Methods
Field methods. Small terrestrial mammals were collected following methods approved by the American Society of Mammalogists (Sikes and The Animal Care and Use Committee of the American Society of Mammalogists 2016) using Sherman live traps (7.62 × 8.89 × 22.86 cm) baited with oats. Traps were set in a line, selecting sites with traces of animal sign or that appear to provide cover in the late afternoon and checked early the following morning. Collecting was conducted in 2013 under permit no. 10/070/071 and in 2014 and 2015 under permit no. 1/2071/72.
Captured individuals were identified with dichotomous keys (see Appendix, Dichotomous keys used to identify small mammal species in Central Nepal). Tissue (ear punches) was collected from live specimens as a tissue source for DNA sequencing and assigned identification numbers associated with collecting locality. Specimens were measured, photographed, and examined for determination of gender and the presence of ectoparasites before being released. Specimens found dead in the live traps were prepared as voucher specimens and deposited in the Central Department Zoology Museum of Tribhuvan University in Katmandu under the accession numbers CDZ MTU-MRO51 to CDZ MTU-MRO58 and CDZ MTU-MEU31 to CDZ MTU-MEU36. Liver tissue was collected from these voucher specimens for DNA sequencing. Molecular methods. DNA extraction and sequencing were carried out at the Center for Molecular Dynamics, Katmandu, Nepal (CMDN). Approximately 25 mg of ear or liver tissue stored in 95% ethanol was used for DNA extractions which were conducted using the DNeasy Blood and Tissue Kit (Qiagen, Germany). Tissue samples were air dried for 30 min prior to extraction. Each tissue sample was cut into small pieces (~20) and incubated at 56 °C in 180 μl of ATL (Tissue Lysis Buffer) and Proteinase K in a shaking incubator until the tissue samples were completely lysed. DNA was finally eluted in 50 μl of elution buffer and the quantity and quality of the extracted DNA was assessed on a NanoVue Plus (GE Healthcare Life Sciences) spectrophotometer.
Chromatograms outputted from the sequencers were visualized and edited in Chromas v. 2.6.2 (Technelysium: http://technelysium.com.au/wp/chromas/). Sequences of the mitochondrial gene Cytb (1140 bp) was used for species verification and identification by use of MegaBLAST (Morgulis et al. 2008) comparisons with reference sequences in GenBank (https://www.ncbi.nim.nih.gov/gen bank/). Sequences of Cytb of representative taxa from Nepal and the surrounding area (Appendix , Table A1) were used to conduct a bootstrapped (1000 replications) maximum parsimony analysis with MEGA v. 11 (Tamura et al. 2021) to determine the phylogenetic affinities of the specimens sequenced in this study.

Results
We collected 117 specimens representing 10 species of small mammals from two orders, Soricomorpha and Rodentia. All specimens not designated as vouchers were released at the site where they were captured. We present the distribution information for all specimens captured below providing comments on how taxa were identified and remarks on their current taxonomy.

Episoriculus soluensis Gruber, 1969
Nepalese Brown-toothed Shrew Figure  Remarks. This species was previously reported from Ghorepani, Annapurna Conservation Area as E. caudatus soluensis. Its distribution is poorly known because of only recently has it been elevated to a full species. Identification. This is a small, red-toothed shrew possessing small fore claws (1-2 mm) and a tail shorter than head and body. Episoriculus soluensis may be distinguished from other species of Episoriculus that occur in Nepal by the length of tail relative to head and body (~81%). Episoriculus leucops (Horsfield, 1855) has a tail equal to or slightly longer (~105%) than head and body, E. macrurus (Blanford, 1888) has a tail significantly longer (~130%) than head and body, and E. caudatus (Horsfield, 1851) has a tail length shorter (~90%) than the head and body but longer than that of E. soluensis (tail ~80% of head and body). Abe (1977) reported a shorter tail (48-58 mm) in E. soluensis compared with a range of tail length of 62-66 mm in E. caudatus. The tail length of our specimen was 51 mm. Distribution. This species has been reported to occur in Nepal and Sikkim. In Nepal, reported from elevations between 2290 and 4300 m in the Kali Gandaki Valley and Langtang area of northcentral Nepal and the Solukhumbu and Sankhuwasabha districts in northeastern Nepal. Abe (1977) indicated that this taxon was usually found at higher elevations than E. caudatus. Taxonomic remarks. Episoriculus soluensis was considered a synonym of E. caudatus by Hutterer (2005) and Hoffmann and Lunde (2008), but it was recognized as a distinct species by Jenkins (2013) following the recommendation of Motokawa et al. (2009). See Jenkins (2013: 164) for a discussion of the taxonomic history of E.
soluensis. The specific status was supported by Abramov et al. (2017) and recognized by Burgin and He (2018).

Soriculus nigrescens (Gray, 1842)
Himalayan Shrew Figure   A B (Barrett-Hamilton, 1900) has yellowish-brown dorsal and lighter ventral pelage and hind feet that are usually smaller than 23 mm. The Cytb sequence obtained in this study (OK257681) was found to have a 98-97% identity to Cytb sequences from A. gurkha in GeneBank and our maximum parsimony analysis recovered our specimen in a A. gurkha clade with high bootstrap support (Fig. 4). Remarks. Phylogeographic examination of Cytb sequences from the specimens reported above indicate that more than one evolutionary significant unit may be represented among populations of Nepalese Field Mice (Pradhan 2018). Distribution. This species lives in mid-elevation (2,200-3600 m) broadleaf and conifer forests in Central Nepal. The area of its geographic range was estimated to be 11,636 km 2 by Pearch (2011).

Niviventer fulvescens(Gray, 1847)
Indochinese White-bellied Rat Figure 5 Material examined (from new localities) NEPAL - Identification. This species has soft, dense pelage with flattened spines; yellowish dorsal pelage, uniformly white venter, and a long tail that is usually less than 135% of the head and body combined. The tail is moderately well haired towards the tip. Niviventer niviventer (Hodgson, 1836) has greyish pelage and a shorter tail (115% of head and body length) with a discrete white tip. Niviventer fulvescens can be distinguished from N. eha (Wroughton, 1916) which has a smaller head and body length, longer tails (more than 140% of the head and body), and dorsal pelage that is brown and ventral pelage that is grey made up of hairs with white tips. Remarks. The number of recognized species of Niviventer has increased from 17 (Musser and Carleton 2005) to 25 due to molecular analyses (Li et al. 2008;Balakirev and Rozhnov 2010;Balakirev et al. 2014;Ge et al. 2018Ge et al. , 2020. Balakirev and Rozhnov (2010) split N. fulvescens into three species, including N.  , Table A1) analyzed with the sequences obtained in this study.
bukit (Bonhote, 1903) and N. huang (Bonhote, 1905). The distributional limits and presence of additional cryptic species within N. fulvescens require further assessment. The Cytb sequence (OK257679) from our specimen from Banthanti demonstrates no differentiation from a sequence of Niviventer sp. (Appendix , Table A1) from Nepal (KU214574) but only a 92% identity from a specimen of N. fulvescens (KY587417) from Nepal. In addition, this sequence demonstrated >99% identity with sequences from N. fulvescens from Tibet reported by Ge et al. (2021a) and from Niviventer sp. 1 from Tibet reported by Lu et al. (2015). However, the low nodal support for a N. fulvescens clade (Fig. 4) suggests that more than one taxon may be represented among the N. fulvescens sequences obtained from these specimens from Nepal. Identification. Rattus rattus does not have a bicolored tail like R. pyctoris, and it possesses broad, flat spines on the dorsum and a tail which is longer than the head and body combined, unlike R. nitidus (Hodgson, 1845) and R. tanezumi Temminck, 1844. Distribution. This species is native to the Indian subcontinent and introduced worldwide in the temperate, tropical, and subantarctic zones.  Identification. This is a medium-sized rat very similar to R. norvegicus (Berkenhout, 1769) in having a similarly robust body and broad nose, but the tail is relatively shorter (90%) with respect to the head and body ratio. Rattus nitidus is very difficult to differentiate from Brown Rat, R. norvegicus, its sister taxon. The former has proportionally shorter hind feet (32-36 mm) and an overall smaller body (Burgin 2017). Rattus nitidus is differentiated from other species of Rattus that occur in Nepal by its soft dorsal and grayish ventral pelage. The Cytb sequence from our specimen (OK257680) is >95% in agreement with Cytb sequences from specimens of R. nitidus from Nepal and India and form a strongly supported R. nitidus clade (Fig. 4). Remarks. This species was initially identified in the field as R. norvegicus, but Pearch (2011) failed to find any verified records of R. norvegicus from Nepal. Musser and Carleton (2005) concluded that given the wide distribution of R. norvegicus it would seem improbable for it not to occur at least in the north-central areas of Nepal. Distribution. Northern and northeastern India, eastern Nepal, Bhutan, Myanmar, southern and central China, northern Thailand, Laos, and Vietnam. Introduced to the Philippines, central Sulawesi, northwestern New Guinea, and Palau, although it may prove to have been more widely introduced.

Mus musculus Linnaeus, 1758
House Mouse  (Pearch 2011); however, many of these reports are under older names (dubius Hodgson 1845, homourus Hodgson, 1845, urbanus Hodgson, 1845 that are now considered junior synonyms. Adhikari et al. (2017) verified the occurrence of both M. m castaneus and M. m. bactrianus Blyth, 1846 in Nepal from Cytb sequence data. Unfortunately, we did not sequence specimens of Mus captured in this study. Distribution. This species has been spread by association with humans to many islands and all continents except for Antarctica. It is restricted to human dwellings and human-maintained habitats in some areas, but wild populations have brcome established in other regions. Identification. Neodon nepalensis is smaller than N. sikimensis (Horsfield, 1841) and larger than N. irene (Thomas, 1911). It has a shorter tail but a higher head and body-to-tail ratio compared to N. sikimensis. Compared to N. leucurus (Blyth, 1863), N. nepalensis is much smaller but with a longer tail. The ventral pelage of N. nepalensis is brown in contrast to the lighter venters of N. sikimensis and N. irene. The molar patterns of the m1 and M3 observed in N. nepalensis are simpler than those observed in N. sikimensis and are more similar to those observed in N. irene. Distribution. West of the Trisuli River in western Nepal at elevations between 2,400-4,200 m.

Discussion
Small mammals collected during fieldwork in Nepal between 2013 and 2015 have detected a previously undescribed mountain vole, Neodon nepalensis, from western Nepal (Pradhan et al. 2019). This new species was reported to occur west of the Trisuli River at elevations between 2400 and 4200 m a.s.l., but the limits of its distribution remain unresolved. Poulton (2019) reported specimens from Langtang Conservation Area with Cytb sequences similar to specimens collected in Pipar, Annapurna Conservation Area, that likely represent N. nepalensis and would expand it range eastward to the slopes of the Langtang Valley.
New records from the Manaslu Conservation Area have identified the occurrences of four species (Soriculus nigrescens, Niviventer fulvescens, Rattus pyctoris, and Rattus rattus) in areas located between previously known records for these taxa from the Kali Gandaki Valley and the Langtang and Khumbu Valleys. The known range of the Nepalese endemic Apodemus gurkha is expanded to the east with new records from the Manaslu Conservation Area and slightly to the south with new records from the Annapurna Conservation Area. Samdo, in Manaslu Conservation Area, is the highest elevation at which A. gurkha has been recorded (3,825 m a.s.l.). The previous elevation record for A. gurkha was 3,750 m a.s.l. at Phoksundo Lake (ZFMK 84.1112). Specimens of Apodemus pallipes were collected in Shey Phoksundo National Park in locations between the westernmost record at Rara Lake reported by Pearch (2011) and more easterly samples from the Suli Gad Valley. Specimens of Rattus nitidus from the Annapurna Conservation Area expand the known range well to the west of the range reported by Pearch (2011), who gave its previously reported western limit in the Langtang and Kathmandu valleys.
Episoriculus caudatus was originally described as Soriculus caudatus by Horsfield (1851) who failed to designate a type specimen. A lectotype and a paralectotype were designated by Thomas (1918) and the specimen tags of these types indicate the collecting locality was Nepal (Motokawa and Lin 2005). Hinton (1922) concluded that this locality was probably inaccurate, and the type locality has been estimated to be Darjeeling, Sikkim.
In his description of E. c. soluensis, Gruber (1969) noted the smaller size of these specimens from Nepal (n = 28) compared to specimens of E. c. caudatus from Sikkim (n = 14), but Motokawa and Lin (2005) suggested that his samples of E. c. caudatus may have included some specimens of E. leucops baileyi (Thomas 1914), Abe (1977) demonstrated differences in tail, combined head and body, and condylobasal skull lengths that differentiated E. c. caudatus from E c. soluensis from Nepal. However, the sample size of E. c. caudatus was very small (n = 3). He (2018a) indicated that the morphological boundaries are not entirely clear between E. caudatus and E. soluensis and Motokawa and Lin (2005) reported no differences in cranial morphology, except for the larger skull size of the former. Motokawa and Lin (2005) pointed out that the three E. caudatus examined by Abe (1977) from Nepal and the lectotype from Sikkim are among the larger individuals they examined.
The distribution limits of Episoriculus soluensis are unclear because it has recently been elevated to a full species (Jenkins 2013). Whether recognized as a subspecies or more recently as a species, the distributional range of E. soluensis is reported to include Nepal and Sikkim (Motokawa et al. , 2009Abramov et al. 2017;He 2018b). In Nepal, specimens of E. soluensis (Table 1) have been reported from locations in the Annapurna Conservation Area and the Langtang Valley in northcentral Nepal by Abe (1977) and 10 localities in northeastern Nepal by Gruber (1969), Gregori and Petrov (1976), and Daniel and Hanzák (1985). Our review of the literature could not verify any reports of a "smaller" E. caudatus from Sikkim that would be assignable to E. soluensis. In the absence of verifiable records of E. soluensis in Sikkim, it should be considered endemic to Nepal. The suggestion by Abramov et al. (2017), citing Motokawa et al. (2009) as the source, that E. caudatus only occurs in Nepal, cannot be correct given the type locality of E. caudatus is Sikkim (Hinton 1922).

Acknowledgements
The Department of National Parks and Wildlife Conservation, Nepal provided collection permits. Special thanks are extended to Saurav Chhetri, and Paliza Shrestha for their assistance with the fieldwork and to Lily Pradhan for assistance with obtaining permits. We are grateful to the staff at the Center for Molecular Dynamics Nepal, especially Ajay Sharma, Adarsh Sherchan, Jyoti Joshi, and Priya Joshi for providing high-quality sequences. Funding was provided to NP by a Grant in Aid of Research from the American Society of Mammalogists and a grant from the Carolyn M. Glass Fund of the University of Vermont. We thank Sanjan Thapa for his constructive review of our paper and for his guidance and assistance in transferring our vouchers specimens into a publicly accessible collection in Nepal. Dr. Tej Thapa generously accessioned our vouchers into the collections at the Central Department of Zoology Museum of Tribhuvan University.