Neoromicia isabella, Decher & Hoffmann & Schaer & N Orris & Kadjo & Astrin & Monadjem & Hutterer, 2015

Decher, Jan, Hoffmann, Anke, Schaer, Juliane, N Orris, Ryan W., Kadjo, Blaise, Astrin, Jonas, Monadjem, Ara & Hutterer, Rainer, 2015, Bat diversity in the Simandou Mountain Range of Guinea, with the description of a new white-winged vespertilionid, Acta Chiropterologica 17 (2), pp. 255-282 : 267-273

Neoromicia isabella sp. nov. Decher, Hutterer and Monadjem

Isabelline White-winged Serotine

Neoromicia cf. rendalli ; Monadjem et al., 2013 b

Holotype

ZFMK 2008.0292 , field number JD 650, collected by Jan Decher on 7 March 2008. Adult male in reproductive condition (testes 5 × 4 mm), preserved in 70% ethanol, skull extracted. Tissue preserved (COI sequence GenBank no. KT598187 ).

Type locality

Guinea, Province Macenta, Simandou Range, Foko Confluence, 8°29’48.62’’N, 8°54’48.22’’W, ca. 765 m a.s.l., captured in net set across creek in rainforest.

Paratype

ZFMK 2008.0291 , field number JD 647 ; adult female (pregnant with 1 embryo), preserved in ethanol. Netted same day and place as holotype.

Referred specimen

DM 12619 , Mount Nimba, Liberia ; reported as Neoromicia cf. rendalli by Monadjem et al. (2013 b); COI sequence deposited under GenBank no. JX508832 .

Diagnosis

A medium-sized pipistrelloid bat of the genus Neoromicia , based on the presence of a single upper premolar (Hill and Harrison, 1987) with white wings; only likely to be confused with N. tenuipinnis (slightly smaller and different pelage colour, see below) and N. rendalli (larger). Dorsal and ventral surface of wings pure white (as in N. tenuipinnis ), tips of dorsal pelage light rusty orange-brown (‘isabelline’), not dark brown (as in N. tenuipinnis ). Baculum similar in shape to N. rendalli but smaller. COI sequence different from that of both N. tenuipinnis and N. rendalli ( Fig. 3 ).

Etymology

The species epithet refers (as a noun in apposition) to the ‘Isabella’ colouration (sensu Ridgeway, 1912) of the dorsal fur of this bat, and also recalls the name of Jan Decher’s daughter Isabelle.

Description

Neoromicia isabella sp. nov. is similar in size and general colour pattern to N. tenuipinnis ( Table 3 , Figs. 4 and 5 ) but slightly larger in total length, forearm length, and body mass ( Table 3 ). Tail length is 34% of the total length in the holotype, and 37% on average. This value in the holotype of N. rendalli ( Thomas, 1889) is 49%, and in the holotype of N. tenuipinnis 39% ( Peters, 1872). The tips of the dorsal hairs are of a light isabelline colour ( Ridgeway, 1912), much lighter than in N. tenuipinnis (compare Figs. 4 and 5 ). The dorsal hairs are 4.5 mm long. Ears, lips, wings, uropatagium, and hindfeet are almost white and covered by very short hairs; only the ear tips and dorsal surfaces of forearm, hindfeet and tail have a pale brownish hue.

Ear length is 10 mm in the holotype of N. isabella (10–13 mm, n = 3), and its tip is rounded ( Fig. 4 ). The tragus of the holotype ( Fig. 6 ) is 4.2 mm long, and represents 42% of the ear length. A small notch is present at the basis of the anterior surface. Compared to illustrations of N. tenuipinnis and N. rendalli in Rosevear (1965) and specimens in ZFMK and DM, the tragus of N. isabella is similar to that of N. tenuipinnis (tragus length estimated from figure: 3.9 mm) but the shape of the notch is different ( Fig. 6 ).

The penis of the holotype specimen is 6.4 mm long and covered by short whitish hairs. A digital x -ray of its baculum is shown in Fig. 7A in a semilateral view. It is slightly longer (3.5 mm) than in the N. isabella specimen DM12619 (2.8 mm; Fig. 7B ) reported by Monadjem et al. (2013 b: Fig. 4C ), but very similar in shape, and both are still longer than that of N. tenuipinnis (in Monadjem et al., 2013 b: Fig. 4D ), which is 1.8 mm long and strongly bent ( Fig. 7C ). A baculum figured as N. rendalli by Kearney et al. (2002: Fig. 1F ) is about 4 mm long.

The cranium of N. isabella is illustrated in Fig 8 . Neoromicia isabella is distinguished from N. tenuipinnis (skull figured in Monadjem et al., 2010: 479) by slightly larger cranial and dental measurements ( Table 4 ), particularly in the greatest skull length and the upper and lower molar rows. Greatest skull length in N. rendalli (skull figured in Monadjem et al., 2010: 475) is similar to that of N. isabella , but its cranium is wider, as expressed by zygomatic width, greatest width of braincase, and mastoid width ( Table 4 ).

Within the genus Neoromicia the greatest skull length of N. isabella (GSKL 12.5–13.0 mm) is of medium size and less robust than in N. rendalli , with a narrow maxillary and braincase and a fragile zygomatic arch. Skulls of 15 other species assignable to the genus Neoromicia (see accounts in Monadjem et al., 2010, 2013 b; Goodman et al., 2012; Happold and Happold, 2013) are either larger or smaller in length. The dorsal profile of the skull is slightly concave at the rostrum and slopes gradually upwards to the occiput. There is no sagittal crest, but a faint lambdoid crest. The dental formula is I2/3, C1/1, P1/2, M3/3 = 32, as in the other Neoromicia . Several observers have reported the presence of a tiny anterior upper premolar present on either or both sides of the upper jaw in N. tenuipinnis ( Hayman, 1954; Schlitter et al., 1982; Thorn and Kerbis Peterhans, 2009). The occasional presence of this tiny anterior upper premolar is not unusual in Neoromicia species (e.g. see Rosevear, 1965; Happold and Happold, 2013); however, not one of the specimens that we examined displayed such a tooth. The anterior upper incisor is more than twice as long as the posterior (about twice in N. rendalli and N. tenuipinnis — Monadjem et al., 2010) and unicuspid. The upper canine is long; 1.6-times as long as the anterior upper incisor, and 2.5-times as long as the subsequent premolar. The upper molar rows (C–M 3) are almost parallel; the molars are shorter than wide, and all teeth from the canine to the third molar are in contact. The mandible length is intermediate between that of N. tenuipinnis and N. rendalli ( Table 4 ). The lower incisors are all trifid.

Comparisons and Designation of a Neotype for Neoromicia tenuipinnis

Our comparisons are hampered because the holotype of N. tenuipinnis is lost (Turni and Kock, 2008), and the original description of Peters (1872) does not include skull measurements. Likewise, the description of N. rendalli by Thomas (1889) does not contain skull measurements; however, the holotype skull (BMNH 1889.3.2.3) is figured in Monadjem et al. (2010) and in the ACR (2015: Appendix), and was examined by AM. External measurements of the types of both taxa, however, are congruent with the current treatment of the species ( Table 3 ).

Externally the new species is quite distinct. Our images ( Fig. 4 ) may be compared with those of N. tenuipinnis from Liberia ( Fig. 5 ) and another one from Comoe National Park, Côte d’Ivoire (http:// www.inaturalist.org/photos/189398). The latter species has much darker body coloration. Neoromica rendalli is also darker, and the dark-brown coloration extends to the ventral surface of the body (specimen ZFMK 1977.1015 from South Sudan). Neoromicia isabella is immediately distinguishable from its sister species N. roseveari , itself only recently described, by its white wings and the rich colour of its fur. In N. roseveari , the wings are dark and the fur is dark brown ( Monadjem et al., 2013 b). Furthermore, N. roseveari is far larger (for the holotype, FA is 37.1 mm and GSKL is 14.36 mm).

The whole group, however, may include further unrecognized taxa if one looks at the genetic and morphological variation throughout Africa, which is beyond the scope of our paper. Specimens from Nigeria to Central Africa mentioned by Bergmans (1977), Viellard (1974), Thorn and Kerbis Peterhans (2009), Bates et al. (2013), and other authors should be re-examined in the context of a revision of the genus. It should be also noted that the most current range map for N. tenuipinnis sensu lato ( Fahr, 2013 e) may contain records of both N. tenuipinnis sensu stricto (see below) and the newly described N. isabella .

Under these circumstances, however, we consider it essential to fix the morphological and genetic characters of N. tenuipinnis and to designate a neotype for Vesperus tenuipinnis Peters, 1872 (now Neoromicia tenuipinnis ), because the former holotype, an adult female from ‘Guinea’ ( Peters, 1872: 264), in the Berlin Museum (ZMB) is lost (Turni and Kock, 2008). Monadjem et al. (2013 b) listed a specimen in the Natural History Museum, London (BMNH 1889.5.1.3) as a ‘syntype’ of N. tenuipinnis , which however is incorrect, as Peters (1872: 264) only mentioned a single specimen in the Berlin Museum at hand; the one, which is now lost.

Here we designate specimen DM 13235 in the Durban Natural Science Museum as the neotype. It consists of a male specimen collected by A. Monadjem at Bonlah village , near Yekepa at the base of Mount Nimba, Liberia on 7 January 2012; external and cranial measurements are included in Tables 3 and 4 . The body is preserved in formalin with the skull extracted. The tragus and the baculum (also Fig. 7C ) are documented in Monadjem et al. (2013 b: Figs. 4D and 6D ). A COI sequence is deposited in GenBank ( JX508831 ). The external measurements of the neotype are very similar to those of Peters’ (1872) lost holotype ( Table 3 ).

In light of our new genetic analysis ( Fig. 3 ), the genus Neoromicia , at least in West Africa, no longer appears to be paraphyletic (as suggested by Monadjem et al., 2013 b) but consists of a monophyletic clade. The type species of the genus, N. zuluensis (Roberts, 1924) , was studied genetically by Goodman et al. (2012) showing that it is sister to N. somalica . We are therefore confident that our new species has been assigned to the correct genus. In our COI tree N. isabella is sister to N. roseveari , and both taxa sister to N. brunnea , while N. tenuipinnis and N. rendalli group more at the root of this clade ( Fig. 3 ). Hence, whereas morphological characters suggest a closer relationship between N. tenuipinnis and N. isabella , this is not supported by the genetic analysis.

Remarks

This is the first record of a white-winged bat from the Simandou Range, initially identified as N. tenuipinnis . Eleven individuals were captured, all between 7 and 9 March 2008 in the canopy net over the creek at FC, two of which were kept as vouchers. They group with specimen DM12619 from Mount Nimba, Liberia, that was labelled Neoromicia cf. rendalli by Monadjem et al. (2013 b). Neoromica isabella was not recorded during the 2002 RAP, nor the 2003 RAP. The individuals captured in the context of this study over the forest creek at FC suggests that N. isabella uses drainages lines as travel routes probably emerging from hollow trees or rock crevices upstream. The specimen from the Liberian portion of Mount Nimba was captured emerging from the roof of a hut in a small village close to the larger settlement of Yekepa ( Monadjem et al., 2013 b).

Records from adjacent regions ( Aellen, 1963; De Vree, 1971; Bützler, 1994; Fahr et al., 2006; Monadjem and Fahr, 2007) may refer either to N. tenuipinnis or to the new species. Monadjem et al. (2013 b) collected both species on Mount Nimba. A specimen mentioned by Roche (1971) as N. tenuipinnis deviates by very small external and cranial measurements, and until a re-examination of the voucher specimen we regard this record with caution.

Conservation status

The conservation status of this species has not yet been evaluated ( IUCN, 2015).

Figures

FIG. 3. Bayesian consensus topology based on the cytochrome oxidase subunit I (COI) gene. Tadarida brasiliensis and T. teniotis were designated as outgroup taxa. See Table 3 for complete list of localities and museum and GenBank numbers. Values above branches are Bayesian posterior probabilities

FIG. 4. Views of the head and body of N. isabella sp. nov. (holotype) after capture. Note the isabelline-colored dorsal fur and the white wings (Photographs by J. Decher)

FIG. 5. Dorsal and ventral view of a live-captured N. tenuipinnis from Liberia. Note the dark dorsal fur (Photographs by A. Monadjem)

FIG. 6. Outlines of the left tragus in N. rendalli (left), N. tenuipinnis (middle; both redrawn after Rosevear 1965), and N. isabella sp. nov. (holotype)

FIG. 7. Comparison of bacula. A, N. rendalli, dorsal and lateral views (adopted from Kearney et al., 2002); B, C, N. isabella sp. nov., digital x-ray in semi-lateral view (ZFMK 2008.0292, holotype) and dorsal and lateral views of DM12619; D, N. tenuipinnis (DM13235, Neotype). Scale bar = 1 mm (C, D modified from Monadjem et al., 2013a)

FIG. 1. Study area along the Simandou Ridge in the Guinée Forestière Region of Guinea. Blue triangles = 2002 RAP, red dots = 2008 baseline study sites. Green line = Forêt Classée du Pic de Fon boundary, black line = mining concession boundary, orange line = préfecture boundary. Red triange in insert map shows the location of study area in Guinea. For site abbreviations see Table 1

FIG. 8. Dorsal, ventral and lateral views of the cranium and lateral view of the mandible of N. isabella sp. nov. (ZFMK 2008.0292, JD 650, the holotype) from the Foko Confluence, Simandou Range, Guinea. Greatest length of skull is 12.98 mm. (Photographs by R. Hutterer)