TEXAS TECH UNIVERSITY Natural Science Research Laboratory Occasional Papers Museum of Texas Tech University Number 325 18 June 2014 Defining Species and Species Boundaries in Uroderma (Chiroptera: Phyllostomidae) with a Description of a New Species Front cover: Uroderma bakeri , new species. Illustration by Hugo Mantilla-Meluk, based upon museum voucher specimens. Defining Species and Species Boundaries in Uroderma (Chiroptera: Phyllostomidae) with a Description of a New Species Hugo Mantilla-Meluk Abstract Anew species of Peter’s tent-making bat, genus Uroderma (Chiroptera: Phyllostomidae), is described from Venezuela’s Cordillera del Caribe and piedmonts of the Eastern Cordillera of the Colombian Andes. Cranial morphology of the new species approaches that of U. magnirostrum, with a notable enlargement of the nasals in the post-orbital region, giving the skull a flatter ap¬ pearance in lateral profile than typical specimens of U. bilobatum. The new species of Uroderma can be distinguished easily from U. magnirostrum by several discrete skin and cranial charac¬ ters. Taxonomic affinities among geographic variants of U. bilobatum were assessed through analyses of morphologic, karyotypic, and molecular variation. Although statistical support for well-defined groups was low in morphological assessments, four morphotypes were recognized based on geographic structure and previously reported karyotypic and molecular data. In addition to the newly recognized species, U. convexum Lyon 1902 is elevated to species status, and U. c. molaris is treated as a subspecies of U. convexum. Further, based on reciprocal monophyly of supported clades present in analyses of mitochondrial DNA data, assortment of karyotypes distinguished by three rearrangements, and reduced fitness in hybrids, U. davisi Baker and Mc¬ Daniel 1972 is treated as a species. These taxonomic revisions result in a genus comprised of five species (U. bilobatum , V convexum , U. davisi , U. magnirostrum , and U. bakeri sp. nov.). Keywords: Colombia, taxonomic revision, Uroderma , U. bakeri sp. nov., U. bilobatum , U. convexum , U. c. molaris , U. davisi , Venezuela Resumen Se describe una nueva especie de murcielago constructor de tiendas de Peter Uroderma (Chiroptera: Phyllostomidae) de la Cordillera del Caribe en Venezuela y el piedemonte oriental de la Cordillera Oriental de los Andes de Colombia. La morfologia craneal de la nueva especie es cercana a aquella de U. magnirostrum , con un notable ensanchamiento de los nasales en la region postorbital, dandole al craneo una apariencia mas plana en perfil lateral en comparacion con craneos tipicos de U. bilobatum. La nueva especie puede ser facilmente diferenciada de U. magnirostrum por varios caracteres discretos de piel y craneo. Las afinidades taxonomicas entre las variantes geograficas descritas para U. bilobatum fueron re-evaluadas teniendo en cuenta su variacion morfologica, cariotipica y molecular previamente descrita. A pesar de un bajo soporte estadistico en los analisis morfologicos, cuatro grupos fueron identificados de acuerdo a su estructura geografica e informacion cariotipica y molecular previamente descrita. Adicionalmente, U. convexum Lyon 1902, es elevada a estatus especifico, y U. c. molaris es considerada como subespecie de U. convexum. De otra parte, teniendo en cuenta monofilia reciproca en clados estructurados presentes en datos de ADN mitocondrial, asi como, cariotipos identificados por tres arreglos cromosomicos y exito reproductivo reducido en hibridos, U. davisi Baker and McDaniel 1972 es reconocido como una especie valida. Esta revision taxonomica resulta en un genero compuesto por cinco especies ( U. bilobatum , U. convexum , U. davisi , U. magnirostrum , y U. bakeri sp. nov.). Palabras clave: Colombia, revision taxonomica, Uroderma, U. bakeri sp. nov., U. bilo¬ batum , U. convexum , U. c. molaris , U. davisi , Venezuela 2 Occasional Papers, Museum of Texas Tech University Introduction Tent-making bats, genus Uroderma (Peters 1886), are distributed from Mexico to Brazil (Simmons 2005; Gardner 2007). Presently, only two species of Uroder¬ ma , U. bilobatum and U. magnirostrum (Davis 1968; Baker et al. 2003; Simmons 2005), are recognized. Although U. bilobatum exhibits substantial karyotypic and DNA sequence variation, the use of morphol¬ ogy to separate karyotypic and genetic phylogroups has generated conflicting interpretations and debate (Baker and Lopez 1970; Baker 1981; Greenbaum 1981; Barton 1982; Hafner 1982; Lessa 1990; Owen and Baker 2001; Hoffmann et al. 2003). To date, six subspecies of U. bilobatum have been described: U. b. bilobatum (Peters 1866), U. b. convexum (Lyon 1902), U. b. thomasi (Andersen 1906), U. b. trinitatum (Davis 1968), U. b. molaris (Davis 1968), and U. b. davisi (Baker and McDaniel 1972). Simmons (2005) recognized three phylogroups associated with the chromosomal lineages identified by Baker (1981) and molecular clades reported in Hoffmann et al. (2003). Maximum cytochrome-# genetic distances (< 3.7%) among phylogroups of U. bilobatum identified in Hoff¬ mann et al. (2003) were lower than average percentages identified among the majority of recognized phyllosto- mid sister species (Bradley and Baker 2001; Baker and Bradley 2006). However, at a hybrid zone in Pacific Nicaragua, Honduras, and El Salvador between karyo¬ typic races of U. b. convexum and U. b. davisi (Baker 1981; Owen and Baker 2001; Hoffmann et. al. 2003), the concordance of mtDNAand karyotypic geographic boundaries revealed defined geographic structure and geographic isolation maintained by negative hetero¬ sis (Barton 1982). Based on these results, Simmons (2005) encouraged further systematic revisions of U. bilobatum subspecies to clarify their taxonomic status. In this paper, morphologic variation in U. bilobatum and U. magnirostrum is reviewed and the evolutionary concordance of molecular and karyotypic phylogroups assessed by Hoffmann et al. (2003) is evaluated. Materials and Methods Variation in skull morphology within the genus Uroderma was analyzed from 433 adult specimens (222 males and 211 females) representing U. magnirostrum and all currently recognized subspecies of U. bilobatum (.sensu Gardner 2007) from most of the distribution of the genus (Appendix I). Age of specimens was deter¬ mined based on the degree of ossification of phalangeal epiphyses and completeness of basisphenoid suture ossification. Specimens examined were archived in the Instituto de Ciencias Naturales, Universidad Na- cional de Colombia (ICN), the National Museum of Natural History (NMNH), and the Museum of Texas Tech University (TTU). Museum abbreviations follow Hafner et al. (1997). Taxonomic identification. —All specimens were evaluated for each diagnostic character mentioned in the original descriptions of each recognized species and subspecies for the genus (Peters 1866 - U. b. bilobatum ; Lyon 1902 - U. b. convexum ; Andersen 1906 - U. b. thomasi ; Davis 1968 - U. b. molaris, U. b. trinitatum , and U. magnirostrum ), as well as geographic criteria depicted in Davis (1968), Baker and McDaniel (1972), and Baker and Clark (1987). Measurements. —Originally, 16 craniodental measurements were recorded including: greatest length of skull (GLS); condylo-basal length (CBL); palatal length (PAL); zygomatic width (ZW); mastoid width (MW); brain-case width (BCW); interorbital constric¬ tion width (IOC); maxillary width (MxW); maxillary height (MxH); rostral depth (RD); distance across third upper molars (M-M); internal width between third upper molars (IMW); distance across upper canines (C-C); canine-maxillary second molar length (CM2); mandibular length (MDL); and mandibular tooth row length (cm3). All measurements were taken in millime¬ ters to the nearest 0.1 mm with a dial caliper (Mitutoyo Absolute Snap series 573). Selection of informative measurements. —In order to avoid redundancy among the 16 recorded measurements and to determine the minimum number of variables to be used in analyses, a preliminary prin¬ cipal components analysis (PCA) was performed and correlated variables were eliminated through a PCA Cartel Scree plot test in the Statgraphics 15 package (Statgrahics 2009). Multiple correlation coefficients for the analyzed variables were calculated and ordered Mantilla-Meluk—New Uroderma from Colombian-Venezuelan Andean Piedmont 3 (low to high) for the number of variables suggested in the Scree plot. Partial correlation matrices were evaluated and those variables possessing significant differences were eliminated. In addition, values of a variance/covariance matrix from standardized data were calculated and the minimum number of variables to be included was determined based on the multiple correlation coefficient value. Sexual dimorphism. —Sexual dimorphism was evaluated for each recognized taxon using a Hotelling’s T 2 test in the statistical package PAST (Hammer et al. 2001). The 16 recorded craniodental measurements from 222 adult males and 211 adult females were used in this analysis. Assessment of skull morphometric variation within Uroderma.—To analyze the phenetic similarities within the genus Uroderma , a PC A and a discriminant function analysis (DFA) were performed in the statis¬ tical packages PAST (Hammer et al. 2001) and SPSS 9.0 (SPSS Inc. 1999), respectively. The selected, informative craniodental measurements were used in these analyses. Evaluation of an unidentified morphogroup .— Morphological differences between U. magnirostrum and an unrecognized morphotype, identified herein as ECNV (for eastern Colombia and northern Venezuela), also were assessed statistically using a PC A and a DFA performed on the selected craniodental variables. Both PC A and DFA were performed separately for males and females in PAST (Hammer et al. 2001) and SPSS 9.0 (SPSS Inc. 1999) softwares, respectively. Results Selection of informative measurements. —Six craniodental measurements were selected for the morphometric analyses, as follows: GLS; MW; BCW; IOC; M-M; and MDL. Selected variables were tested for normality by the application of an Energy test in R mvnorm.etest for independent variables. Sexual dimorphism. —Although statistically sig¬ nificant differences were found between measurements of males and females in three of the analyzed taxa (U. b. bilobatum, U. b. convexum, and U. b. davisi ), all groups possessed percentages of correctly assigned individu¬ als greater than 70% in the Hotelling’s T 2 -test: U. b. bilobatum (85%; Hotelling=58.1; F=2.03; p=0.03); U. b. thomasi (95%; F=2.19; p=0.06); U. b. convexum (71.77%; Hotelling=54.5; F=2.98; p>0.01); U. b. davisi (85.21%; Hotelling= 124.42; F=5.7; p>0.01); U. b. mo- laris (91.67%; Hotelling=67.38; F=1.66; p=0.15); and U. magnirostrum (95.24%; Hotelling=80.55; F=1.81; p=0.2). The limited sample size for U. b. trinitatum (N=8) prevented a test for sexual dimorphism in that taxon. Assessment of skull morphometric variation within Uroderma.—Because sexual dimorphism was depicted for some taxa, PC A and DFA were performed separately for males and females. The PC A including all taxa revealed low levels of morphometric variation within the genus Uroderma for the six selected vari¬ ables (Fig. 1). The first principal component accounted for 80.44% and 62.52% of the variation among males and females, respectively. Loadings on the first compo¬ nent were all positive and relatively uniform in value for both males and females, indicating variation in general skull size for both sexes. The remaining components were variable in magnitude and sign in both datasets, indicating variation in shape (Table 1). Similarly, DFA revealed a low overall morphometric differentiation within the genus. However, U. magnirostrum and representatives of the unrecognized morphotype ECNV were discriminated from U. bilobatum sensu Davis (1968) (Fig. 2, Table 2). Although not statistically sup¬ ported, the DFA revealed some differentiation between Central and South American U. bilobatum samples, particularly among male specimens (Fig. 2, Table 2). Evaluation of an unidentified morphogroup .— Uroderma magnirostrum and the unrecognized mor¬ photype ECNV proved to have statistically supported different skull morphologies in both male and female datasets. Skulls of U. magnirostrum were smaller, had wider interorbital constrictions, and possessed deeper rostri than those of the ECNV morphotype (Table 3, Fig. 3). The first principal component accounted for U b. magnirostrum Occasional Papers, Museum of Texas Tech University {% ZZ'ZZ) Z lusuoduioo ledpuud CM Q S-gP o brain-case width (BCW); distance across third upper molars (M-M); and mandible length (MDL). Mantilla-Meluk—New Uroderma from Colombian-Venezuelan Andean Piedmont 5 88.95% and 76.59% of the variation among male and female datasets, respectively (Fig. 3). All analyzed individuals of both U. magnirostrum and ECNV were assigned correctly in the DFA (Wilks’ Lambda=0.002; Chi-square=46.30; df=15; P=4.7' 5 for male dataset; and Wilks’ Lambda=0.005; Chi-square=31.14; df=12; P=0.0018 for female dataset; Table 4). Table 1. Loading of vectors, Eigen values, and percentage of variance explained by each component in a principal component analyses (PCA) applied to six selected craniodental variables of 222 adult male and 211 adult female specimens representing eight identifiedphylogroups across the range of distribution o/Uroderma. Abbreviations of measurements: GLS: greatest length of skull; MW: mastoid width; IOC: interorbital constriction; BCW: braincase width; M-M: distance across first upper molars; and MDL: mandible length. Variable Sex Axis 1 Axis 2 Axis 3 Axis 4 Axis 5 Axis 6 -0.6112 0.04605 0.1813 -0.2823 -0.6897 -0.1898 GLS ¥ 0.4250 -0.3985 0.4996 0.1339 -0.2538 -0.5733 -0.6652 0.3928 -0.0055 0.4954 0.3879 0.08597 MW ¥ 0.3990 0.3948 0.0700 -0.6862 0.3842 -0.2480 -0.0950 0.0564 0.6838 -0.5368 0.4251 0.2266 IOC ¥ 0.1138 -0.7334 -0.5208 -0.4200 0.0059 0.0396 S -0.1309 -0.7040 0.3716 0.3488 0.1486 -0.4532 BCW ¥ 0.4372 0.0727 -0.4801 0.5734 0.4479 -0.2091 C? -0.1114 -0.4222 0.0642 0.2244 -0.2346 0.8365 M-M ¥ 0.4463 0.3362 -0.3409 -0.0424 -0.7453 0.1201 S -0.3812 -0.4080 -0.5978 -0.4633 0.3407 0.0090 MDL ¥ 0.5068 -0.1708 0.3572 0.06497 0.1789 0.7417 1.0026 0.1325 0.0390 0.0349 0.0224 0.0148 Eigen Value ¥ 0.6680 0.2375 0.0749 0.0363 0.0307 0.0211 A 80.4480 10.6330 3.1305 2.8030 1.7945 1.1911 Variance ¥ 62.5210 22.2310 7.0068 3.3962 2.8703 1.9740 6 Occasional Papers, Museum of Texas Tech University (%8‘CO Z uoipunj lUBuiujuosi.a CM (%9'SL) Z uojpunj jueuiLuuosia d CD g O d Oh I I A Hb CD a ,CD CM id r»- c _o o c p CM T3 d cd as c CM CM CM C+H O a ^ .a - Uh c3 > N- n- c o 1 c 3 CD TD O 'd ^ a £ o u T3 H £ CD O Ph £ & CD o 03 £ & o a X d o V , - £2 N T3 D P 'g O O & a a § an 1-1 I = O X 03 O a ^ d o e a -E ° 'd o .22 ^ j§ o ,| CM d sa s d a sp & s a Mantilla-Meluk—New Uroderma from Colombian-Venezuelan Andean Piedmont 7 Table 2. Wilks ’ lambda values of the discriminant function analyses performed on six cranial measurements of222 adult male and 211 adult female specimens representing eight identified phylogroups across the range of distribution o/Uroderma. Groups were discriminated a priori according to their morphology. Test of Function(s) Wilks’ Lambda Chi-square df Sig. Males 1 through 5 0.201 344.102 35 0.000 2 through 5 0.524 138.499 24 0.000 3 through 5 0.821 42.264 15 0.000 4 through 5 0.929 15.875 8 0.044 5 0.985 3.199 3 0.362 Females 1 through 6 0.052 583.546 42 0.000 2 through 6 0.298 239.891 30 0.000 3 through 6 0.554 117.092 20 0.000 4 through 6 0.795 45.326 12 0.000 5 through 6 0.925 15.342 6 0.018 6 0.991 1.885 2 0.390 Table 3. Loadings of principal components analysis performed on the six selected variables for U. magnirostrum andECNV. Abbreviations of measurements: GLS: greatest length of skull; MW: mastoid width; BCW: braincase width; IOC: interorbital constriction; M-M: distance across first upper molars; and MDL: mandible length. Variable PC 1 PC2 PC 3 PC 4 PC 5 PC 6 Males GLS 0.4156 0.4166 0.2727 0.0861 -0.1777 0.3455 MW 0.0676 -0.1901 -0.1377 -0.2697 -0.1112 0.5423 BCW 0.2183 0.1001 -0.2537 0.0100 -0.0444 -0.1512 IOC -0.1400 0.2640 -0.1625 0.0899 0.0219 -0.1794 M-M -0.3550 0.0406 0.0417 -0.6275 0.0986 -0.0646 MDL -0.3648 -0.0597 0.1378 0.5572 0.0227 0.4179 Females GLS 0.2682 0.2826 0.2731 0.0141 -0.0414 0.1156 CBL 0.2982 0.2062 0.4585 0.0870 0.1171 -0.1497 IOC -0.3737 0.0956 0.0997 -0.4149 0.0683 -0.1059 MW -0.3396 0.2179 0.2096 0.1943 0.2922 -0.3319 BCW -0.1701 -0.1986 0.2046 -0.3850 -0.0059 0.2048 M-M -0.0256 -0.1436 -0.2300 -0.1587 -0.2482 -0.0989 Occasional Papers, Museum of Texas Tech University c z U B W CD CCS U U on ^ x 8 55 S 2a > 0> GO 1 T3 bfl CD G X 3 tn Xi § a X 5 bD o :c u jd is a g cb xn X a 3 on G o c3 bD G CD T3 +a o G T3 G O o G CD a g Oh bD u o Q -c fc. i T3 § cb § S G O G a o CD G G a £ o U x m o w x s '5 S QJ £* CO CO S G g & 'C S ' G § (%9S'3t) 3 iueuoduuoQ p J> B ^ .£P k§ G Z) Q B S x> cb cb > G O Mantilla-Meluk—New Uroderma from Colombian-Venezuelan Andean Piedmont 9 Table 4. Casewise results of the classification obtained from the discriminant function analyses between the unrecognized morphotype ECNV (Group 1) (N=12) and U. magnirostrum (Group 2) (N=26). For both male andfemale datasets, 100% of the individuals were classified within the hypothesized groups. In column three, P (D>d | G=g) is the significance level of such ay 2 )- Actual Group Predicted Group P(D>d|G=g) P Squared Mahalanobis Distance to Centroid Discriminant Scores Function 1 it 1 0.79 0.07 24.84 it 1 0.52 0.41 23.93 id? 1 0.76 0.09 24.88 i 8 1 0.30 1.08 23.53 i 8 1 0.97 0.00 24.53 it 1 0.20 1.65 25.85 1 0.89 0.02 24.43 21 2 0.97 0.00 -17.24 2 8 2 0.21 1.57 -15.95 2 S 2 0.01 7.82 -20.00 28 2 0.62 0.25 -16.70 2 8 2 0.63 0.23 -16.72 2 8 2 0.76 0.09 -17.51 2 1 2 0.53 0.39 -17.82 28 2 0.91 0.01 -17.09 28 2 0.27 1.19 -16.11 2 8 2 0.01 7.82 -20.00 2 1 2 0.62 0.25 -16.70 28 2 0.63 0.23 -16.72 2 8 2 0.76 0.09 -17.51 28 2 0.27 1.19 -16.11 28 2 0.74 0.11 -16.87 1? 1 0.58 0.31 -23.14 1? 1 0.98 0.00 -23.72 1 0.59 0.28 -24.22 1? 1 0.98 0.00 -23.72 1? 1 0.59 0.28 -24.22 2? 2 0.93 0.01 6.37 2? 2 0.93 0.01 6.37 2? 2 0.93 0.01 6.37 2? 2 0.93 0.01 6.37 2? 2 0.93 0.01 6.37 2? 2 0.93 0.01 6.37 2? 2 0.20 1.65 5.18 2? 2 0.13 2.25 4.96 2? 2 0.16 1.95 7.86 2? 2 0.79 0.07 6.20 29 2 0.87 0.03 6.62 10 Occasional Papers, Museum of Texas Tech University Discussion Bats within the subfamily Stenodermatinae are part of the most recent radiation among phyllostomids and apparently have undergone a rapid process of diversification (Hoofer and Baker 2006; Larsen et al. 2007; Velazco and Patterson 2008). Velazco and Pat¬ terson (2008) suggested that the evolution of species in the subfamily Stenodermatinae is a recent event, characterized by low genetic divergence among species relative to older lineages. Velazco et al. (2010) reported genetic differences of less than 3.0% for Cytb sequences between recognized species of the vampyrissine genus Platyrrhinus (e.g., P. angustirostris \ s. P. fusciventris, 2.03%; P. incarum vs. P. angustirostris, 2.68%). Similarly, molecular data support a rapid disper¬ sion accompanied by low genetic differentiation for vampyrissine bats in the genus Uroderma (maximum Cytb intra-racial differences of 1.7% and overall low morphological differentiation). Interestingly, the only sequence available for a member of the ECNV morphotype possessed the highest divergence in Cytb sequences (3.7%) and clearly was differentiated from both U. bilobatum sensu Davis (1968) and U. mag- nirostrum (Hoffmann et al. 2003). It also has been suggested that recent, rapid evolutionary events, such as the one exemplified by vampyrissine bats, usually result in low or incomplete morphologic divergence. Although overall morpho¬ metric resolution within Uroderma in this study was low, morphometric analyses revealed: 1) a greater affinity between U. magnirostrum and morphotype ECNV; 2) no morphometric overlap between these two forms and U. bilobatum sensu Davis (1968); and 3) 100% morphometric discriminations between the unrecognized morphotype ECNV and U. magnirostrum in both male and female data sets. A new Uroderma from the eastern piedmonts of the northern Andes. —The ECNV morphotype, repre¬ senting individuals from Venezuela’s Cordillera del Caribe and the eastern piedmonts of Colombia’s Eastern Cordillera, appears to represent an undescribed species of Uroderma. Recognition of this species is supported by statistically significant differences and a unique combination of discrete morphological characters, as described below. Family Phyllostomidae Gray 1825 Stenodermatinae Gervais 1856 Uroderma Peters 1865 Uroderma bakeri , new species Holotype. —Adult male, TTU 33358, body preserved in alcohol, with extracted cranium and mandibles in excellent condition (Fig. 4) and tissues (TK 15288), collected on 25 April 1978 by Margaret O’Connell and Robert J. Baker. Measurements of the holotype and specimens in the type series are presented in Appendix II. Type locality. —Santa Crucita, in Parque Nacional Guatopo, Miranda, Venezuela, 10°5'N, 66°33'W at an elevation of 2,480 m (Fig. 5). Type series. —Specimens included in the type series (N=12) consist of seven males (including holo¬ type) and five females from Colombia: adult male ICN 12917 and female ICN 12918, collected on 28 Febru¬ ary 1993 by Juan Manuel Rengifo, in Campo Cano Limon, near Yuca oil well, inside the forest, Arauca; adult male ICN 10881 and adult female ICN 10882, collected on 22 November 1989 by a field party as part of the Introductory Mammalogy course, Universidad Nacional de Colombia, in Medina, Vereda La Sarza, Quebrada La Sarza, Cundinamarca; adult female ICN 15128, collected on 2 July 1998 by Cecilia Ramirez, Herly Zuniga, and Hector Lancheros, in Vereda Soya, margen derecho del Rio Zaguea, Ubala, Cundinamarca; adult male ICN 9456, collected on 29 March 1983 by Group of Advanced Systematics, Universidad Nacional de Colombia, in Fuente de Oro, Km. 9 Road Puerto Limon - Puerto Lleras, Vereda La Esperanza, Finca La Virginia, Meta; adult males ICN 10732 and ICN 10733 and an adult female ICN 10734, collected on 23 April 1988 by Maria del Pilar Rivas and Pedro Sanchez Palomino, in San Juan de Arama, Cano La Curia, Meta; adult male ICN 6882 and female ICN 6884, collected on 7 May 1977 by a field party from the Universidad Nacional de Colombia, in Villavicencio, Finca El Buque, Meta. Two additional specimens of U. bakeri collected on the same night as the U. bakeri holotype were deposited with the Venezuelan authorities that provided the permits, but the museum in which they were archived has not been established. Mantilla-Meluk—New Uroderma from Colombian-Venezuelan Andean Piedmont 11 Figure 4. Ventral (a), dorsal (b), and lateral (c) profile of the holotype (TTU 33358 $) of Uroderma bakeri. Abbreviations: interorbital constriction (IOC); nasal angle (NA); and sutura frontomaxillaris (SFM). 12 Occasional Papers, Museum of Texas Tech University Figure 5. Type locality of Uroderma bakeri in Santa Crucita (red star), Parque Nacional Guatopo, Miranda, Venezuela 10°5'N, 66°33'W, at an elevation of2,480 m. Yellow stars indicate the collecting localities associated with U. bakeri specimens: 1) Arauca, Campo Cano Limon, near oil drill Yuca, inside the forest (ICN 12917-18); 2) Cundinamarca, Medina, Vereda La Sarza, Quebrada La Sarza (ICN 10881-82); 3) Vereda Soya, margen derecho del Rio Zaguea, Ubala; 4) Meta, Fuente de Oro, Km. 9 Road Puerto Limon - Puerto Lleras, Vereda La Esperanza, Finca La Virginia (ICN 9456); 5) San Juan de Arama, Cano La Curia (ICN 10732-34); 6) Villavicencio, Finca El Buque (ICN 6882-84). The area highlighted in yellow represents the potential distribution of U. bakeri. Inset: Photo of the collecting site, taken by Robert J. Baker the day of the capture of the holotype (25 April 1978). Mantilla-Meluk—New Uroderma from Colombian-Venezuelan Andean Piedmont 13 Distribution .—The species is known from the piedmonts associated with the northwestern Colom¬ bian-Venezuelan Orinoco River basin of the Eastern Cordillera of the Colombian Andes and the Cordillera de Merida and Cordillera del Caribe. This species has been recorded in an elevational range of 500 to 2,500 m (Fig. 5). Description of the holotype .—Holotype is a general grey color. Dorsal hairs are three-quarters grey with a paler band at base, and venter is paler than dorsum. Uroderma bakeri has a thin, white line along the dorsal midline, well-defined white stripes across the cheek and above the eye, and well-defined black eyebrows on top of the anterior portion of the eyes. Ears have scattered hairs. Borders of the ears, horseshoe, and nose leaf possess a distinctive yellow coloration. Pelage reaches the middle of forearm, with tibia and middle part of the thigh naked. Interfemoral membrane is notched and naked with no hairs on edge. Holotype has a large and massive skull, characterized by a flat lateral profile (Figs. 4 and 6); a broader rostrum; swol¬ len interorbital area at the sutura frontomaxillaris; and a massive dentition, with the typical formula for genus: i 2/2, c 1/1, p 2/2, m 3/3, total 32. Four upper inci¬ sors are bilobed and not in contact, with inner incisors larger and located anterior to the lateral ones. Lobes of medial incisors are subequal. Both central and lateral incisors are tilted with tips pointed bucally. Canines are massive, particularly at base, and appear straight from a lateral view. Base of first premolar is triangu¬ lar. Anterior surface of mesiostyle (Osborn 1907), or anterior cingular style (Vandebroek 1961), has a flat appearance in ventral view, and half of its length is in contact with canine. Main cone (paracone) of first premolar is elongated and curved toward labial side. Curvature of first premolar, paracone, allows it to be visible behind canines in a rostral view. First upper premolar is shorter than second premolar and barely surpasses half of length of canine. From a ventral view, tip of posterior cingular style of first premolar is overlapped by mesiostyle of the paraconid of second premolar; however, teeth are not in contact as seen from a lateral view. U. bakeri holotype has an acces¬ sory style on first premolar and a wide contact region between second and third molar including the second molar metacone, metaconid, metastyle and its associ¬ ated crista (teeth nomenclature adapted from Van Valen [1994] and Swindler [2002]). Diagnosis.—Uroderma bakeri is diagnosed by the following characteristics: large skull (GLS > 23.00 mm); swollen interorbital area at the sutura frontomax- illaris ; massive dentition; maxillae deflected into plane of nasals; small crest along suture joining the parietals {sutura sagittalis) that terminates where parietals join the temporal bones {sutura parietointerparietalis)', projected edge of squamosal visible from a dorsal view and allowing only a small portion of the tympanic bullae to be viewed; inside the nasal aperture, ventral portion of the nasal maxilloturbinates visible, as well as the septum nasi and part of vomer septum; maxil¬ loturbinates and superior and inferior conchas massive; joining of the nasal bones occurs caudally with respect to the maxillae, nearly forming a straight nasal angle in a lateral view (Figs. 6 and 7). Comparisons. —The new species can be distin¬ guished from Central American representatives of the U. bilobatum complex {U. b. convexum, U. b. molaris, and U. b. davisi ) by its flat lateral profile (Figs. 4 and 7). Uroderma bakeri lacks the typical deflection of the nasal bones at the interorbital area, which are present in U. b. convexum , U. b. molaris and U. davisi. Uroderma bakeri has a broader rostrum, swollen interorbital area at the sutura frontomaxillaris , a more massive denti¬ tion, and a less arched tooth-row than U. b. convexum and U. b. davisi. In contrast with the rostrum of U. bakeri , rostra of U. b. convexum , U. b. molaris, and U. b. davisi are substantially shorter and expand abruptly from the apex to the lachrymal bone edges; therefore, the lachrymal bones are more visible from a rostral view than in U. bakeri. From the eastern South American forms in the U. bilobatum complex {U. b. bilobatum , U. b. thomasi , and U. b. trinitatum ), U. bakeri also can be differentiated by its flat lateral profile, swollen interorbital constriction at the sutura frontomaxillaris , and by having a more massive dentition. In addition, in U. bakeri the union of the nasal bones with respect to the maxillae forms a straight nasal angle in a lateral view, whereas in U. b. bilobatum , U. b. thomasi , and U. b. trinitatum , the nasal angle is obtuse (Fig. 4c). In U. b. bilobatum , U. b. thomasi , and U. b. trinitatum , the edge of the maxillae (that forms the eye socket) is terminated in a rim that is less distinct in U. bakeri. All museum voucher specimens in the U. bakeri type series had been misidentified as U. magnirostrum. However, these two species easily can be distinguished 14 Occasional Papers, Museum of Texas Tech University Figure 6. Comparative images of skin (a and b) and frontal view of skulls (c and d) of Uroderma magnirostrum (left) and U. bakeri (right). Abbreviations: conchas (Co); NS (septum nasi); maxilla bone (Mx); interorbital constriction (IOC); deep rostrum (DR); shallow rostrum (SR); face lines (FL); and hairy uropatagium (HU). Mantilla-Meluk—New Uroderma from Colombian-Venezuelan Andean Piedmont 15 Flat brain-case Globoid brain-case 1 A V Depressed Post-orbital region Figure 7. On the left: comparisons of cranial external characters between Uroderma bakeri (top) and U. convexum (center); detail of nasal aperture of U. bakeri showing the intrusion of the maxilla (Mx) into the plane of the nasals (N) (bottom). On the right: skull profiles of U. magnirostrum (TTU 28496); U. bakeri (TTU 33358); U. bilobatum thomasi (TTU 341912); and U. convexum (TTU 39139) (top to bottom). 16 Occasional Papers, Museum of Texas Tech University by a suite of external and internal characters. Exter¬ nally, U. bakeri has a dark grey general coat color contrasting the brownish coat of U. magnirostrum. In U. bakeri, rostral stripes are well-marked, whereas in U. magnirostrum rostral stripes are absent or only insinuated. Uroderma bakeri has scattered hairs on the border of the ears, contrasting the naked border of the ears of U. magnirostrum. In addition, the borders of the ears, the horseshoe, and the nose leaf in U. bakeri are distinctive yellow, contrasting the less distinctive yellow borders of the ears and nose leaf of U. mag¬ nirostrum. Uroderma bakeri has a wide uropatagium, lacking hair on both dorsal and ventral surfaces, thus differing from the uropatagium of U. magnirostrum, which has scattered hairs on both dorsal and ventral surfaces (Fig. 6). All specimens in the U. bakeri series have long, black hairs on the anterior portion of the eyes forming a brow line. This character was present in only three of 77 U. magnirostrum analyzed (identified based on skull characters, particularly rostrum depth [Andersen 1908]). These three specimens were col¬ lected in two localities from the northern Colombian Amazon at the department of Caqueta (ICN 11342-43 from Montanita, and ICN 12771 from Chiribiquete, department of Caqueta in Colombia). Further analysis is necessary to determine the taxonomic status of these specimens. Internally, U. bakeri and U. magnirostrum share a flattened skull profile. However, the species can be differentiated by the deeper rostrum typical of U. magnirostrum (Andersen 1908). In addition, in U. bakeri the nasal aperture has a broader appear¬ ance than in U. magnirostrum. The maxilloturbinates and the superior and inferior conchas of U. bakeri are more massive than in U. bilobatum, but less massive than those of U. magnirostrum. In U. bakeri, the nasal septum and the mesethmoid is formed by a thin sheet, not terminated in a shield-like structure, contrasting the wide and massive nasal septum and the shield-like structure associated with the mesethmoid of U. mag¬ nirostrum (Fig. 6). In U. magnirostrum the skull is laterally compressed and from a dorsal view, the edges of the squamosal and the tympanic bullae are hidden beneath the parietal. As in other representatives of the genus, the junction of the nasal bones occurs anterior to the junction of the maxilla in U. bakeri. However, in U. bakeri the nasal area, from a lateral view, is divided into two perpendicular planes forming an angle with the vertex marked by the tip of the septum nasi. This angle is more acute in U. bakeri than in other Uroderma (Fig. 4c). In U. bakeri, the protrusion of the maxillae allows the roots of the upper incisor to be more notice¬ able than in other Uroderma. The nasal aperture in U. bakeri is piriform, with the two upper sides deflected internally by a convergent intrusion of the maxillae into the plane of the nasal bones (Figs. 6 and 7); these intrusions are produced by the lateral projections of the nasal septi and are absent in U. magnirostrum where the maxillas are parallel to the nasals (Figs. 6 and 7). In U. bakeri the infraorbital foramen, located above the root of the second premolar, marks the inflexion point of the concave surface of the maxilla bone; in contrast, this inflexion is not noticeable in U. magnirostrum. In U. bakeri, from a rostral view, the lateral sides of the maxillae are deflected or concave and expand into the area of the anterior part of the zygomatic arches; contrasting the flat appearance of the rostrum of U. magnirostrum which extends within a single plane. Genetic data. —Investigation of the molecular phylogenetic affinities between the new species and other representatives of the genus is beyond the scope of the present work. Hoffmann et al. (2003) reported that the holotype specimen TTU 33358 differed in the mtDNA Cytb gene (genetic distance of 3.7%) relative to other representatives of the genus. Karyotype. —The karyotype of U. bakeri is unknown. Ecology. —The holotype of U. bakeri was col¬ lected at the Parque Nacional Guatopo (Fig. 5), located in northern Venezuela, southeast of the city of Caracas. The area is part of the mountainous system of Serrania del Interior. The vegetation corresponds to a tropical hyper-humid pre-montane forest. It is dominated by tree species such as Ochoroma lagopus, Erythrina poeppigiana, Pterocarpus acapulcensis, Tabebuia chrysantha, Bursera simaruba, and Cecropia peltata, and palm trees such as Oenocarpus bataua, Badris sp. and Asterogyne spicata. The understory also presents a high diversity of plants including Calathea and Heli- conia. Epiphytic plants also are well-represented by Araceae, Bromeliaceae, Orquidaceae, and Piperaceae (Weidmann et al. 2003). Morphological characteristics such as larger body size and enlargement of the rostrum at the nasal area, as well as enlarged nasal conchas, may be adaptations Mantilla-Meluk—New Uroderma from Colombian-Venezuelan Andean Piedmont 17 of U. bakeri to highland ecosystems. Ecological dif¬ ferentiation also may suggest that the divergence of U. bakeri can be explained by selection and adaptation (Nosil et al. 2009). Although U. bakeri exists across a wide range of elevational gradients (500-2,500 m), the species inhabits higher elevations than the typical maximum elevational limits of other representatives of the genus. Etymology.—Uroderma bakeri is named for Dr. Robert J. Baker, who has dedicated his life to the study of a wide variety of aspects of the natural history and evolution of the Neotropical mammalian fauna, in particular to the study of phyllostomid bats. Dr. Baker has used Uroderma as a natural model in several studies since the 1970’s. His ongoing research on Uroderma includes hybrid zones, karyotypic evolution, system- atics, population genetics, and niche modeling. The proposed common name is Baker’s tent-making bat. Final Taxonomic Considerations Taxonomic status of Central American Uro¬ derma.—Trans-Andean U. b. convexum possess a different karyotype from typical cis-Andean U. bilo- batum (2N=38) and are found along the Pacific coast of Panama west to Honduras and the Atlantic versant of Mexico (Fig. 8). The easternmost karyotyped re¬ cord for U. b. convexum (2N=38) is located in Melgar, department of Tolima, in the Colombian Inter-Andean valley of the Magdalena River (Baker and Lopez 1970). There is no evidence of cis-Andean representatives of the 2N=38 race. Although U. b. bilobatum (2N=42) and U. b. convexum could not be statistically differ¬ entiated in the current analyses, discrete morphologic characters identified in Lyon’s (1902) description of U. b. convexum allowed discrimination of U. b. bilo¬ batum from U. b. convexum. The molar tooth-row of U. b. convexum is decidedly convex and less nearly parallel compared to U. b. bilobatum. Teeth in U. b. convexum are slightly larger than corresponding teeth in U. b. bilobatum ; most conspicuous are the greater widths of upper premolars and molars. The portion of the palate posterior to the last molar is decidedly shorter and narrower in U. b. convexum than in U. b. bilobatum (Lyon 1902:84). Lyon (1902) also reported that the rostrum of U. b. convexum from Panama is shorter and broader, and the nasals are more flattened compared to Brazilian specimens of U. bilobatum. In addition to the characters described by Lyon (1902), U. b. convexum has, from a lateral view, a more markedly accentuated deflection of nasal bones at the interorbital region. The more deflected appearance of the skull in U. b. convexum is associated with its shorter rostrum compared to other representatives of the genus (Fig. 7). Deflection of the nasal bones in U. b. convexum is marked by the basineurocranial plane, which is de¬ limited by the pterigoid and the anterior portion of the basioccipital. Uroderma b. convexum is characterized by a higher braincase relative to U. b. bilobatum. Although analyses are in agreement with the previously documented statistical support differen¬ tiating U. b. convexum from U. b. davisi (Baker and McDaniel 1972; Baker et al. 1972), the strongest evi¬ dence for specific status of U. convexum and U. davisi is defined by three unique euchromatic chromosomal rearrangements in U. davisi (Baker 1981) that delin¬ eate a well-defined, narrow hybrid zone thought to be maintained by negative heterosis (Barton 1982). Two chromosomally identified phylogroups, concordant with reciprocal monophyly in mtDNA, a narrow hybrid zone, and evidence of negative heterosis in hybrids merits recognition of convexum and davisi as species following the reasoning of Patton and Dingman (1968). Similarly, based on the operational criteria of statistically supported reciprocal monophyly (Da Silva and Patton 2005), U. davisi is recognized as specifically distinct from U. convexum in both nuclear and mito¬ chondrial markers. Uroderma davisi (2N=44) and U. convexum (2N=38) are distinguished by three unique rearrangements (Baker 1981), as well as consistent clades in the analyses of Cytb sequences (Hoffmann et al. 2003). Due to the recent and rapid fixation of the observed chromosomal rearrangements in U. con¬ vexum and U. davisi , clades identified in Hoffmann et al. (2003) have low support values. In addition, 18 Occasional Papers, Museum of Texas Tech University U, bakeri U. b. bilobatum U. b. thomasi U. b. tnnitatum U. convexum U. c. motaris U. davisi Sampling localities -10°0'N 60°b'W Figure 8. Distribution of taxa in the genus Uroderma , as recognized in this study: U. bakeri (yellow); U. bilobatum bilobatum (blue); U. b. trinitatum (pink, identified with label, restricted to the island of Trinidad); U. b. thomasi (red); U. convexum (light green); U. c. molaris (green mustard); and U. davisi (orange). Localities of specimens used in this study are represented by black squares. Mantilla-Meluk—New Uroderma from Colombian-Venezuelan Andean Piedmont 19 unique allozymic alleles in U. davisi were proposed by Greenbaum (1981). Concluding remarks. —The history of the tax¬ onomy of the genus Uroderma is not particularly different than that of other Neotropical mammals. To discern natural diversity is not an easy task, and sometimes the complexity of the biological processes explaining diversity requires the application of more than one technique or conceptual approach. As asserted by Baker (1984) for Rhogeesa genowaysi and Baker et al. (2009) for Eumops wilsoni , there are speciation events in bats that are unaccompanied by obvious morphological changes, or in some cases, significant morphological changes are incongruent with historical patterns recovered from molecular markers. Among Neotropical bats, Uroderma has been one of the most intensively studied genera, including studies of mul¬ tiple datasets documenting its karyotypic, molecular, and morphological variation. Hoffman et al. (2003) I sincerely thank R. J. Baker, F. Hoffmann, R. M. Fonseca, and R. D. Bradley for discussions and com¬ ments on their genetic and morphometric analyses of Uroderma populations that contributed to understand¬ ing taxonomic affinities of the new species. I appreciate the logistical support provided by Y. Munoz-Saba and the Institute de Ciencias Naturales of the Universidad Nacional de Colombia. I am indebted to A. Gardner, K. Helgen, and D. Wilson for assistance at the collec¬ tions of the National Museum of Natural History of the Smithsonian Institution, as well as for the loan of speci¬ mens that, due to the complexity of the group, required proposed that the divergence among the three chromo¬ somal races would have occurred between 0.9 and 0.2 Mya. The recent origin of the Uroderma chromosomal races is a potential explanation for the high intraspecific variation and insipient geographic structure observed in some of the regional morphotypes. Karyotypic, molecular, and morphological geographic partition¬ ing observed in Uroderma support the hypothesis of the northern Andes as an effective barrier preventing population gene flow. Uroderma bakeri appears to have one of the widest elevational ranges in the genus, and its distribution is associated with the piedmonts and mountainous ecosystems in northern South America. Some of the morphologically distinctive characters of V bakeri can be interpreted as adaptations to highland ecosystems, such as the enlargement of the nasal area (Cortes et. al. 2003). The presence of adaptations to highland ecosystems in Uroderma provides new alter¬ natives for the reconstruction of potential evolutionary scenarios for the genus. multiple evaluations. I thank D. Parish for karyotypic preparation, E. Lessa for his valuable comments and assistance in statistical analyses, A. Daugherty and L. Bradley for editorial assistance, E. Mantilla-Meluk and M. Mantilla-Meluk for assistance in data collection at the Institute de Ciencias Naturales of the Universidad Nacional de Colombia, and S. Fernandez-Medina for her logistical support. Finally, I especially thank B. D. Patterson for his critical review of the manuscript. His accurate comments as well as comments from an anonymous reviewer greatly contributed to improve an earlier version of this manuscript. Literature Cited Andersen, K. 1906. Brief diagnoses of a new genus and ten new forms of stenodermatous bats. Annals Maga¬ zine of Natural History 18:419-423. Andersen, K. 1908. A monograph of the chiropteran genera Uroderma, Enchistenes, mdArtibeus. Proceedings of the Zoological Society ofLondon 1908:204—319. Baker, R. J. 1981. Chromosome flow between chromosom- ally characterized taxa of volant mammal, Uro¬ derma bilobatum (Chiroptera: Phyllostomidae). Evolution 35:296-305. Baker, R. J. 1984. A sympatric species of mammal: A new species of Rhogeesa (Chiroptera: Vespertilionidae). Systematic Zoology 33:178-183. Baker, R. J., and R. D. Bradley. 2006. Speciation in mam¬ mals and the genetic species concept. Journal of Mammalogy 87:643-662. Baker, R. J., and C. L. Clark. 1987. Uroderma bilobatum. Mammalian Species 279:1-4. 20 Occasional Papers, Museum of Texas Tech University Baker, R. J., S. Hoofer, C. A. Porter, and R. A. Van Den Bussche. 2003. Evolutionary relationships and classification of New World leaf-nosed bats inferred from DNA sequence. Occasional Papers, Museum of Texas Tech University 230:1-32. Baker, R. J., and G. Lopez. 1970. Chromosomal variation in bats of the genus Uroderma (Phyllostomidae). Journal of Mammalogy 51:786-789. Baker, R. J., and V. R. McDaniel. 1972. Anew subspecies of Uroderma bilobatum (Chiroptera, Phyllostomidae) from Middle America. Occasional Papers, Museum of Texas Tech University 7:1-4. Baker, R. J., W. R. Atchley, and V. R. McDaniel. 1972. Karyology and morphometries of Peters’ tent mak¬ ing bat, Uroderma bilobatum Peters (Chiroptera: Phyllostomidae). Systematic Zoology 21:414-429. Baker, R. J., M. M.Mcdonough, V. Swier, P. A. Larsen, J. P. Carrera, and L. K. Ammerman. 2009. New species of bonneted bat, genus Eumops (Chiroptera: Molos- sidae) from lowlands of western Ecuador and Peru. Acta Chiropterologica, 11:1-13. Barton, N. H. 1982. The structure of the hybrid zone in Uroderma bilobatum (Chiroptera: Phyllostomidae). Evolution 36:863-866. Bradley R. D., and R. J. Baker. 2001. A test for the genetic species concept: cytochrome-/) sequences in mam¬ mals. Journal of Mammalogy 82:960-973. Cortes,A., C. Tirado, and M. Resenmann. 2003. Energy, me¬ tabolism, and thermoregulation in Chinchilla brevi- caudata. Journal of Thermal Biology 28:489-495. Da Silva, M. N., and J. L. Patton. 2005. Molecular phylogeog- raphy and the evolution and conservation of Ama¬ zonian mammals. Molecular Ecology 7:475^186. Davis, W. B. 1968. Review of the genus Uroderma (Chirop¬ tera). Journal of Mammalogy 49:676-698. Gardner, A. L. 2007. Mammals of South America, Volume E Marsupials, xenarthrans, shrews, and bats. Uni¬ versity of Chicago Press, xx + 690 pp. Greenbaum, I. F. 1981. Genetic interactions between hybrid¬ izing cytotypes of the tent-making bat ( Uroderma bilobatum). Evolution 35:305-320. Hafner, J. C. 1982. Genetic considerations at a contact zone of Uroderma bilobatum (Chiroptera: Phyllostomidae). Evolution 36:852-862. Hafner, M. S., W. L. Gannon, J. Salazar-Bravo, and S. T Alvarez-Castaneda. 1997. Mammal collections in the western hemisphere: A survey and directory of existing collections. American Society of Mam- malogists, Lawrence, Kansas. 93pp. Hammer, 0., D.A.T. Harper, and P. D. Ryan. 2001. PAST: Paleontological Statistics Software Package for Education and Data Analysis. Palaeontologia Electronica 4:1-9. Hoffmann, F. G., J. Owen, and R. J. Baker. 2003. mtDNA perspective of chromosomal diversification and hybridization in Peter’s tent making bat ( Uroderma bilobatum. Phyllostomidae). Molecular Ecology 12:2981-2993. Hoofer, S., and R. J. Baker. 2006. Molecular systematics of vampyressine bats (Phyllostomidae: Stenoder- matinae) with comparison of direct and indirect surveys of mitochondrial DNA variation. Molecular Phylogenetics and Evolution 39:424-438. Larsen, P. A., S. Hoofer, M. C. Bozeman, S. Pedersen, H. Genoways, C. J. Phillips, and R. J. Baker. 2007. Phylogenetics and phylogeography of th zArtibeus jamaicensis complex based on citochrome-Z) DNA sequences. Journal of Mammalogy 88:712-727. Lessa, E. P. 1990. Multidimensional analysis of geographic genetic structure. Systematic Zoology 39:242-252. Lyon, M. W. 1902. Description of a new phyllostome bat from the Isthmus of Panama. Proceedings of the Biological Society of Washington 15:83-84. Marin-Vasquez, A., and A. V. Aguilar-Gonzalez. 2005. Mur- cielagos (Chiroptera) del departamento de Caqueta, Colombia. Biota Colombiana 6:211-218. Nosil, P, D. J. Funk, and D. Ortiz-Barrientos. 2009. Divergent selection and heterogeneous genomic divergence. Molecular Ecology 18:375-402. Osborn, H. F. 1907. Evolution of mammalian molar teeth. Macmillan, New York. 250 pp. Owen, J. G., and R. J. Baker. 2001. The Uroderma biloba¬ tum (Chiroptera: Phyllostomidae) cline revisited. Journal of Mammalogy 82:1102-1113. Patton, J. L., and R. E. Dingman. 1968. Chromosome studies of pocket gophers, genus Thomomys. I. The specific status of Thomomys umbrinus (Richardson) in Ari¬ zona. Journal of Mammalogy 49:1-13. Peters, W. C. 1866. Machte eine Mittheilung uber neue oder ungenugend bekannte Flerthiere ( Vampyrops , Uroderma , Ametrida, Tylosoma, Vespertilio, Ves¬ per ugo) und Nager ( Tylomys , Lasiomys). Monats- berichtede Kurftirstlich-Brandenburgische Societat der Wissenschaften 1866:392-397. Simmons, N. B. 2005. Order Chiroptera. Pp. 312-529 in Mammals species of the World, a taxonomic and geographic reference (D. E. Wilson and D. M. Reeder, eds). 3 rd ed. John Hopkins University Press, Baltimore, Maryland. Mantilla-Meluk—New Uroderma from Colombian-Venezuelan Andean Piedmont 21 SPSS Incorporated. 1999. SPSS Base 9.0 for Windows User’s Guide. SPSS Inc., Chicago IL. Statgraphics. 2009. Statgraphics XV for Windows. Statisti¬ cal Graphic Corporation online manuals. StatPoint Technologies, Herndon, Virginia. 305 pp. Swindler, D. R. 2002. Primate dentition, an introduction to the teeth of non-human primates. Cambridge Uni¬ versity Press, United Kingdom. 296 pp. Vandebroek, G. 1961. The comparative anatomy of the teeth of lower and non-specialized mammals. Pp. 215-320 in International colloquium on the evo¬ lution of lower and nonspecialized mammals (G. Vandebroek, ed.). Koninklijke VlaamseAcademie voor Wetenschappen, Letteren en Schone Kunsten van Belgie, Brussels. Van Valen, L. M. 1994. Serial homology: The crests and cusps of mammalian teeth. Acta Palaeontologica Polonica 38:1451-1458. Velazco, P. M., andB. D. Patterson. 2008. Phylogenetics and biogeography of the broad-nosed bats, Genus Plat- yrrhinus (Chiroptera: Phyllostomidae). Molecular Phylogenetics and Evolution 49:749-759. Velazco, P. M., A. L. Gardner, and B. D. Patterson. 2010. Systematics of the Platyrrhinus helleri species complex (Chiroptera:Phyllostomidae), with descriptions of two new species. Zoological Journal of the Linnean Society 159:785-812. Weidmann, K., R. Rangel, C. Todmann, and A. Reig. 2003. Parques Nacionales de Venenezuela. Oscar Tod¬ mann Eds. Editorial Arte, Caracas. Address of author: Hugo Mantilla-Meluk Department of Biological Sciences Texas Tech University Lubbock, TX 79409 Current address: Program of Biology Universidad del Quindlo Armenia, Quindlo, Colombia Editor for this manuscript was Robert D. Bradley. 22 Occasional Papers, Museum of Texas Tech University Appendix I Specimens examined in morphometric and morphologic analyses. In capital letters are the names of the coun¬ tries, followed by the departments or provinces, and the collecting localities. In parentheses is the acronym of the institution housing the specimens, followed by the museum catalogue number and sex of each individual. Acronyms follow those used in the Materials and Methods. Urodermabilobatum bilobatum (62: 29c?: 33$).— BRASIL (15: 9?: 6$): Mato Grosso: (USNM 393686c?, USNM 393687c ?, USNM 393688$, USNM 393689c?, USNM 393690c?, USNM 393691$, USNM 393692c?, USNM 393693c?, USNM 393694$)); Para: (USNM 361663c?, USNM 361665$, USNM 361667$, USNM 361670$, USNM 361673c?, USNM 361675c?); COLOMBIA (21: 9n ov m VD CM d- °v. £ rn CM CM Z ov in’ OV 00 in’ in OV d- csv oo ov rn ov o t-~ Cl VD in c- d- 00 oo o CSV Ov CM ov >n d- c- o' d- rn CM CM rn ov m vo in 00 in c- o in vq ov in Z rn ci cm ci ov in' ov c-' d- in’ 00 d- Q o vq 00 r~~ 00 d- c- vo rn ov 00 m CM CM CM p ov ov c~ 00 £ rn ci Cl ci ov CM in 00 ov in in in 00 d; (A d- d- cm ci Cl rn in’ ov 00 vd in’ ov in Sex FA GLS CBL zw MW IOC BCW CM2 c-c RD M-M MDL cm3 8.35 ND 8.58 8.91 8.36 8.88 8.99 8.72 8.67 8.77 8.74 8.41 Publications of the Museum of Texas Tech University Institutional subscriptions are available through the Museum of Texas Tech University, attn. NSRL Publica¬ tions Secretary, Box 43191, Lubbock, TX 79409-3191. Individuals may also purchase separate numbers of the Occasional Papers directly from the Museum of Texas Tech University. The Museum of Texas Tech University has a catalog of Occasional Papers which may be viewed online at nsrl.ttu.edu. To do so, you must have Adobe Acrobat installed on your computer. If you have difficulty downloading Occasional Papers, please contact the Webmaster. If there is continued difficulty, contact the Webmaster and a single hard copy can be provided to you via mail at no charge. Layout and Design: Lisa Bradley Cover Design: Hugo Mantilla-Meluk Production Editor: Lisa Bradley Copyright 2014, Museum of Texas Tech University This publication is available free of charge in PDF format from the website of the Natural Science Research Laboratory, Museum of Texas Tech University (nsrl.ttu.edu). The authors and the Museum of Texas Tech University hereby grant permission to interested parties to download or print this publication for personal or educational (not for profit) use. Re-publication of any part of this paper in other works is not pennitted without prior written permission of the Museum of Texas Tech University. This book was set in Times New Roman and printed on acid-free paper that meets the guidelines for permanence and durability of the Committee on Production Guidelines for Book Longevity of the Council on Library Resources. Printed: 18 June 2014 Library of Congress Cataloging-in-Publication Data Occasional Papers of the Museum of Texas Tech University, Number 325 Series Editor: Robert J. Baker Defining Species and Species Boundaries in Uroderma (Chiroptera: Phyllostomidae) with a Description of a New Species Hugo Mantilla-Meluk ISSN 0149-175X Museum of Texas Tech University Lubbock, TX 79409-3191 USA (806) 742-2442