Glischropus aquilus, Csorba, Gábor, Görföl, Tamás, Wiantoro, Sigit, Kingston, Tigga, Bates, Paul J. J. & Huang, Joe Chun-Chia, 2015

Csorba, Gábor, Görföl, Tamás, Wiantoro, Sigit, Kingston, Tigga, Bates, Paul J. J. & Huang, Joe Chun-Chia, 2015, Thumb-pads up — a new species of thick-thumbed bat from Sumatra (Chiroptera: Vespertilionidae: Glischropus), Zootaxa 3980 (2), pp. 267-278 : 269-273

Glischropus aquilus n. sp.

Figures 1–4 ; Tables 1 .

Synonymy. Glischropus sp.: Huang et al. (2014).

Holotype. MZB 35030 (field number # JH20110815.2), adult male, in alcohol, skull removed. Collected by Joe Chun-Chia Huang on 15 August 2011.

Type locality. Sukabanjar village, Lampung, Sumatra, Indonesia, 4°56’S, 103°52’E, 768 m a.s.l.

External measurements of the holotype (in mm): FA= 32.1, EAR = 11.0, HB= 35.4, TAIL = 38.8, TIB= 15.1, TH= 5.6, HF= 6.2, body mass (g)= 4.8.

Etymology. The specific epithet /a.kvi.lus/ (meaning dark-coloured in English) refers to the blackish ears and generally darker pelage of the new species relative to its congeners. The proposed English name is Dark Thickthumbed bat.

Diagnosis. A rather large representative of the genus (FA over 32 mm), with dark brown fur, elevated frontal part, globose braincase and gradually narrowing interorbital region.

Description. Forearm length= 32.1 mm ( Table 1 ); ears are blackish, wide, and broadly rounded; tragus Pipistrellus -like, relatively narrow and bluntly pointed ( Fig. 1 ). The dorsal pelage is dark brown with no reddish hints; ventrally lighter, medium brown. The individual hairs on the back are mostly unicoloured dark brown with only the tips a lighter brown; the hairs on the belly have two bands, the basal 2/3 being dark brown, the distal portion medium brown. The plagiopatagium is attached to the base of the toe. The calcar is short, extends to less than half of the free edge of the uropatagium; the lobe on the calcar is well-developed, elongated, and with a supporting median cartilage. The thumb has a large pinkish pad, characteristic of the genus, oval in outline and 3 mm in length. The sole of the foot is pink and fleshy. The penis is dorsoventrally flattened, 5.2 mm in length; the proximal half is practically naked, whereas the distal half is strongly pilose, with stiff, whitish hairs on the dorsal surface and around the glans ( Fig. 2 ).

The skull has an elevated frontal region and a relatively globose braincase ( Fig. 3 ). The narial emargination is wide and U-shaped; the sagittal crest obsolete, the lambdoid crests are moderately developed. The zygoma is thin and nearly straight with no dorsal eminence. The basioccipital pits are almost imperceptible.

The tips of the four upper incisors are situated in a nearly straight line; the concavity of the second upper incisor (I3) is turned outwards. The first incisor (I2) is bifid, I3 reaches half the height of I2. The main cusp of I3 is much longer than the faint secondary cusplet of the same tooth and its tip is directed downwards. Basal dimension of I3 equals that of I2. The first upper premolar (P2) is basally as large as I2, partly intruded from the toothrow and visible in the lateral view; its cusp reaches well beyond the cingulum of the posterior premolar. The upper and lower molars show no specific modifications and the lower molars are nyctalodont ( Fig. 4 ).

Comparisons. Glischropus aquilus n. sp. is readily distinguishable externally from all other Glischropus species by the general impression of the dorsal fur being dark brown instead of reddish-yellow and the nearly black ear and tragus (vs. brown in all other species of the genus) ( Fig. 1 ).

Beside the external features, G. aquilus n. sp. is also clearly different from both G. tylopus and G. javanus in skull proportions, having a deeper rostrum, more elevated frontal part and higher occipital region ( Fig. 3 ). G. aquilus n. sp. is larger in FA, M3M3W, RW and IOW than any G. tylopus specimen investigated. G. aquilus n. sp. was also different from all investigated Sumatran G. tylopus (see Comparative material) having a wide ’U’ shaped narial emargination (vs. elongated and posteriorly narrowing).

Although in skull shape G. aquilus n. sp. is more similar to G. bucephalus and except the PDW value (which is smaller in all investigated G. bucephalus specimens) its craniodental measurements fall within the range of the latter species, its braincase in general is less swollen, interorbital region gradually narrowing (vs. abruptly narrowing in G. bucephalus ); and I3 nearly equals I 2 in basal dimensions (vs. much smaller).

Multivariate analyses. In addition to the morphological comparisons a Principal Component Analysis was also performed. The 49 specimens clustered into two main groups: G. tylopus on the left and G. bucephalus on the right of the PC1 axis ( Fig. 5 ). The single G. j a v a n u s type fell near the group of G. t y l o p u s, whereas G. aquilus n. sp. is far from all other Glischropus specimens. The PC1 axis is a size axis and shows the wideness of the skull; it accounted for a relatively large (71.49%) proportion of the total variation, whereas PC2 is responsible for 19.40%of variation and indicates a shape factor dominated by PDW ( Table 2 ).

Phylogenetic reconstruction. Because G. javanus is represented only by the holotype, the phylogenetic analysis did not include this species. The analyzed Glischropus cytb sequences grouped into a monophyletic clade and G. aquilus n. sp. is clearly separated from other congeners ( Fig. 6 ). The genetic distance between G. aquilus n. sp. and other Glischropus species—including G. bucephalus paratypes and a G. tylopus specimen collected close to the type locality—is between 12.1–14.6%, which clearly supports that G. aquilus n. sp. is a separate species. The two G. bucephalus paratypes differed only in a few nucleotides, but the two G. tylopus differed considerably, by 5.2% from each other ( Table 3 ).

Ecological notes. The only known specimen was caught along a trail in a secondary forest with a four-bank harp-trap which was set near a bamboo stand and a small stream. Another 14 species were recorded in the forest (including Tylonycteris robustula another bamboo specialist) and an additional eight species from a plantation nearby; hence, 23 bat species were confirmed from the site ( Huang et al. 2014).

Figures

FIGURE 1. Portraits of live specimens of a) G. aquilus n. sp. holotype from Sumatra (MZB 35030), b) G. bucephalus paratype from Cambodia (HNHM 2006.34.37.), c) G. tylopus from Thailand (HNHM 2009.52.1.). Not to scale.

FIGURE 2. Lateral view of penis of G. aquilus n. sp. holotype from Sumatra (MZB 35030).

FIGURE 3. Lateral views of skulls of a) G. aquilus n. sp. holotype from Sumatra (MZB 35030), b) G. bucephalus holotype from Cambodia (HNHM 2006.94.49.), c) G. tylopus holotype from Borneo (BM (NH) 70.2.10.2), d) G. batjanus holotype from Bacan I. (ZMB 85513), e) G. javanus holotype from Java (RMNH 15323). Scale bar represents 5 mm.

FIGURE 4. a) Lateral view of the mandible and occlusal view of b) left upper and c) right lower dentition of G. aquilus n. sp. holotype (MZB 35030). Scale bars represent 3 mm.

FIGURE 5. Principal Component Analysis based on one external and three craniodental measurements of G. bucephalus (white circles), G. aquilus n. sp. black squares (singular), G. javanus white squares (singular) and G. tylopus (black circles). Each symbol represents the position of an individual skull projected on the first two principal components, which together accounted for 91 % of the total variation.

FIGURE 6. Maximum likelihood tree based on 404 bp fragment of cytb gene of G. aquilus n. sp. and selected Glischropus, Pipistrellus and Nyctalus species. Tylonycteris robustula was used as an outgroup taxon.

Tables

TABLE 1. Selected external and craniodental measurements (in mm) of G. bucephalus, G. aquilus n. sp., G. javanus, and G. tylopus. Values are given as mean, ± SD, min-max, (n). Acronyms and definitions for measurements are given in the text.

character	G. bucephalus	G. aquilus n. sp.	G. javanus	G. tylopus
FA	34.03 ± 0.99 32.1–35.7 (15)	32.2 (1)	32.0 (1)	29.17 ± 0.93 27.0–30.9 (29)
STL	12.49 ± 0.24 12.14–13.05 (14)	12.56 (1)	12.03 (1)	11.83 ± 0.32 11.16–12.72 (31)
CCL	11.90 ± 0.22 11.61–12.39 (14)	11.95 (1)	11.40 (1)	11.25 ± 0.32 10.68–12.06 (32)
CCW	4.05 ± 0.11 3.75–4.24 (13)	4.07 (1)	3.95 (1)	3.81 ± 0.15 3.46–4.20 (31)
M3M3W	5.75 ± 0.14 5.47–5.98 (13)	5.68 (1)	5.32 (1)	5.23 ± 0.10 4.97–5.44 (32)
PDW	2.27 ± 0.08 2.19–2.40 (7)	2.66 (1)	2.81 (1)	2.64 ± 0.14 2.39–2.78 (10)
RW	4.99 ± 0.11 4.82–5.17 (13)	4.84 (1)	4.63 (1)	4.47 ± 0.13 4.16–4.77 (32)
IOW	3.68 ± 0.11 3.53–3.93 (14)	3.79 (1)	3.72 (1)	3.49 ± 0.11 3.26–3.68 (32)
CM3L	4.74 ± 0.07 4.63–4.85 (13)	4.65 (1)	4.67 (1)	4.45 ± 0.12 4.20–4.74 (36)
ML	9.03 ± 0.20 8.77–9.54 (13)	8.86 (1)	8.54 (1)	8.31 ± 0.26 7.89–8.98 (35)
CM3L	5.03 ± 0.10 4.86–5.17 (14)	4.85 (1)	4.91 (1)	4.64 ± 0.11 4.43–4.94 (36)

TABLE 2. Factor loadings of selected external and craniodental characters obtained by the PCA.

Character	PC1	PC2
FA	0.876	0.200
M3M3W	0.944	0.189
PDW	-0.559	0.829
RW	0.943	0.115
Eigenvalue	2.860	0.776
Percentage of variance	71.49	19.40