nueva página del texto (beta)
Inglés (pdf)
Artículo en XML
Referencias del artículo
Traducción automática
Enviar artículo por email
Citado por SciELO 
Similares en
SciELO 
Permalink
Introduction
The genus Akodon Meyen, 1833, contains ca. 39 extant species and is one of the most diverse South American genus of rodents. The taxonomy of Akodon has been actively changing since it was first established. For instance, the species A. philipmyersi, A. polopi, and A. josemariarguedasi have been recently described (Pardiñas et al. 2005; Jayat et al. 2010; Jiménez et al. 2013); A. aliquantulus, A. oenos, and A. viridescens were recently synonymized (Jayat et al. 2010; Pardiñas et al. 2011; D'Elía et al. 2011); finally, other forms like A. caenosus were elevated to full species (Jayat et al. 2010). In addition, two species: A. latebricola and A. bogotensis, were removed from Akodon and allocated to the new genus Neomicroxus (Alvarado-Serrano and D'Elía 2013). Still, other species, like A. azarae, A. budini, A. pervalens, A. siberiae, and A. mimus, have unclear phylogenetic relationships, suggesting that the radiation of Akodon is currently neither fully resolved nor understood.
Previous studies have suggested a monophyletic Akodon divided into five main groups of species: aerosus, boliviensis, cursor, varius, and dolores (Smith and Patton 2007; Jayat et al. 2010; Coyner et al. 2013), the first two of which occur in Peru. The Akodon aerosus clade was first recovered by Patton and Smith (1992) and includes the species A. aerosus, A. mimus, A. mollis, A. orophilus, and A. torques, all of which are distributed in Peru. Later, this species group was expanded by Smith and Patton (2007) with four additional species (A. cf. budini, A. affinis, A. albiventer, and A. siberiae). Both studies supported the monophyly of the group; however, a recent study by Coyner et al. (2013) suggested that this group is paraphyletic and includes members of the cursor group. This group is minimally composed of A. aerosus, A. albiventer, A. affinis, A. mollis, A. orophilus, A. surdus, and A. torques (Pardiñas et al. 2015). Meanwhile, studies of the aerosus group at the species level are scarce, and issues such as variable chromosome numbers (see Patton and Smith 1992; Smith and Patton 2007), the taxonomic complexity of some species (e. g., A. aerosus Patton and Smith 1992), and the highly divergent mitochondrial DNA between species (see Patton and Smith 1992; Smith and Patton 1991; 1993; 2007) add uncertainty to the taxonomy of the group.
Patton and Smith (1992) and Smith and Patton (1993, 2007), based on a phylogenetic analysis of cytochrome b sequences, suggested that A. orophilus encompasses more than one species. They reported sequence divergence values between 5 and 10 % between samples from Unchog (Huánuco), Puerta del Monte (San Martín), Leymebamba (Amazonas), and Palca (Junín). Later, Pacheco et al. (2012) described the karyotype of A. orophilus based on specimens from Huánuco (2n = 22). Later, Jiménez et al. (2013) assigned that karyomorph to a new entity, A. josemariarguedasi, assigning the diploid complement 2n = 26 to A. orophilus s. s. from Amazonas. As such, Jiménez et al. (2013) corroborated that A. orophilus s. l. is a species group. These authors restricted the distribution of A. orophilus to the eastern Andean slopes in Amazonas and northern San Martín departments, while A. josemariarguedasi was restricted to the south of the Río Huallaga in Huánuco and Junín departments. More recently, Coyner et al. (2013) proposed the recognition of A. orientalis as a distinct species from A. orophilus based on some morphological differences and molecular data; however, Pardiñas et al. (2015) retained it as subspecies following previous authors.
In 2001 and 2002, the junior author (VP) led two expeditions to the montane forests of Cordillera de Carpish (Huánuco), and collected A. orophilus among other species. We review these specimens and found them different from A. orophilus s. s., and other species of Akodon. Herein, we describe these species as belonging to a new species of Akodon based on an assessment of craniodental and external morphological traits and analysis of morphometric data. We compare this new species with other members of the A. aerosus species group based on their close phylogenetic relationships established in previous molecular studies (Smith and Patton 2007; Jayat et al. 2010; Coyner et al. 2013) and summarized by Pardiñas et al. (2015). Finally, we also highlight the need to protect the Cordillera de Carpish given its high species diversity and endemism (Parker and O'Neil 1976; Young and Leon 1999; Beltrán and Salinas 2010).
Material and Methods
A total of 44 specimens allocated to Akodon orophilus and collected in Huánuco department were examined in this study (Appendix). These specimens are housed in the collection of the Departamento de Mastozoología del Museo de Historia Natural, Universidad Nacional Mayor de San Marcos, Lima, Peru (MUSM). Morphological descriptions and comparisons were based on qualitative external and cranial characters using the terminology of Voss (1988), Myers et al. (1990), Pacheco (2003), and Jiménez et al. (2013); molar cusps and enamel fold terminology follows Reig (1977); capitalized color nomenclature follows Smithe (1975). Age classes were estimated following the criteria of Myers (1989). Comparative measurements were taken from Jiménez et al. (2013) and Myers and Patton (1989).
In addition, we examined specimens of A. aerosus, A. josemariarguedasi, A. orophilus s. l., A. torques, and A. surdus (Appendix) for morphological comparisons; these species were chosen for their geographic distribution in eastern montane cloud forest and close phylogenetic relationships (see Smith and Patton 2007; Jayat et al. 2010; Coyner et al. 2013, Pardiñas et al. 2015). Also included in the comparisons were other species that inhabit cloud forests, such as A. kofordi, A. fumeus, A. budini, and A. siberiae; comparisons with these species were based on the descriptions provided by Myers and Patton (1989) and Pardiñas et al. (2015).
For the morphometric characterization, 19 cranial dimensions for 61 specimens of Akodon were recorded to the nearest millimeter using digital calipers, following Luna and Pacheco (2002) and Jiménez et al. (2013). These 19 measurements and their abbreviations include: greatest skull length (GSL), condyloincisive length (CIL), condylomolar length (CML), length of orbital fossa (LOF), length of nasals (LN), diastema length (DL), length of incisive foramina (LIF), length of maxillary toothrow (LM), breadth of incisive foramina (BIF), breadth of rostrum (BR), breadth of palatal bridge (BPB), breadth of first upper molar (BM1), breadth of nasals (BN), least interorbital breadth (LIB), zygomatic breadth (ZB), braincase breadth (BB), breadth of zygomatic plate (BZP), depth of incisor (DI), and height of braincase (HBC). For the multivariate analyses, we used only adult specimens of age classes 4 and 5 and after rejecting the hypothesis of sexual dimorphism by a t-test analysis, pooled specimens of both sexes. All 19 measurements were transformed to their natural logarithm to perform a Principal Component Analysis (PCA) on a correlation matrix, in order to assess morphometric variation. A discriminant function analysis (DA) was also carried out to examine the level of discrimination between samples. These analyses and the univariate descriptive analysis for all measurements were conducted with the program SPSS 13.0 for Windows.
Results
Morphometric analyses. The first three principal components of the PCA explained 60.85% of the total variation in the PCA. Loadings of the variables analyzed along components 1, 2, and 3 are shown in Table 1. All variables showed a positive loading on the first axis (except DI), thus indicating a general variation in size where CML, ZB, GSL, CIL, and BR had relatively large loadings. Meanwhile, LIF, DI, and DL (positives) on the second axis, and BZP (negative) on the third, had relatively large loadings. Individual scores for the components (PC1 and PC2) are plotted in Figure 1. Three main groups of specimens can be identified in the scatterplot of PC1 and PC2, one formed by A. orophilus s. s. from Amazonas and San Martín departments, A. josemariarguedasi and specimens collected in the Cordillera de Carpish (Huánuco department). Specimens of A. orophilus s. s. overlap marginally with specimens from the Cordillera de Carpish currently assigned to A. orophilus; meanwhile A. josemariarguedasi separates completely from A. orophilus s. s. and specimens collected at the Cordillera de Carpish currently assigned to A. orophilus, confirming the difference in size among these populations (Figure 1). A discriminant function analysis revealed a significant variation among these three groups (Wilk ́s lambda = 0.027; p < 0.000) and showed a clear separation among them (Figure 1). The first discriminant function accounted for 68.4% of the variance. The most influential variables for the first and second function were BR and BN, respectively. Percentages of misidentifications were low, only one individual from the Cordillera de Carpish (Huánuco department) and A. orophilus s. s. were mistakenly identified, while 100% were correctly classified as A. josemariarguedasi. These results indicate that the specimens collected in the Cordillera de Carpish, currently assigned to A. orophilus, are morphometrically distinct from typical A. orophilus and A. josemariarguedasi. The mean standard deviation and range of external and cranial measurements of the specimens examined are shown in Table 2. In addition, specimens from Carpish have some morphological peculiarities, such as dark dorsal coloration, narrow parapterygoid fossa, with straight outward edges, long incisive foramina that nonetheless only reaches the protoflexus of M1, and smaller entoconid-hypoconid cusp pair of m1, which differentiates them from A. orophilus s. s. and A. josemariarguedasi.
Table 1 Loadings of the first three components of a Principal Component Analysis of 19 measurements of Akodon orophilus s. s. (n = 18), Akodon josemariarguedasi (n = 21), and Akodon kotosh sp. nov. (n = 22) of tooth-wear classes 4 and 5. Measurements are defined in Material and Methods.
Figure 1 Specimen scores of adult individuals (age classes 4 and 5) of Akodon kotosh sp. nov. (white circle, n = 22), Akodon josemariarguedasi (gray circle, n = 21) and Akodon orophilus s. s. (black circle, n = 18) for principal components 1 and 2 generated from the correlation matrix (top) and for canonical variants extracted from a discriminant function analysis (bottom) of 19 craniodental measurements. The percent of the total variation explained by each axis is indicated.
Discussion
Based on the high level of sequence divergence for the cytochrome b gene, Patton and Smith (1992), suggested that Akodon orophilus is a complex of species. Subsequent studies (Smith and Patton 1993; 2007) that included a larger geographic range, analyzing samples from the Peruvian departments of Amazonas, San Martín, Huánuco, and Junín, supported their hypothesis. During our research we have examined specimens assigned to A. orophilus from Amazonas, Huánuco, and San Martín, and found morphological (see below) and morphometric differences among them that support the initial hypothesis of Patton and Smith (1992). Taking this information into consideration, and based on our combined analyses of morphology and morphometric data, we propose the recognition of populations from the Cordillera de Carpish (Huánuco department) as a new species, which is described as follows.
Akodon kotosh sp. nov.
Holotype. Adult female (age class 4) deposited at the Museo de Historia Natural of the Universidad Mayor de San Marcos (MUSM 18950), skull removed and cleaned, with tissue sample preserved in alcohol, collected on 15 August 2002 by Marina Villalobos, original field number MVC 115.
Type locality. Peru, Huánuco department, Huánuco province, Chinchao district, Caserío de San Pedro de Carpish, -9o 41.70' S, -76o 05.28''W, elevation 2,400 m.a.s.l. (Figure 2).
Figure 2 A: Map of Peru showing the distribution of Akodon orophilus s. s., A. josemariarguedasi, Akodon mollis, and A. kotosh sp. nov. B: Map of central Peruvian Andes displaying a fraction of the distribution of A. josemariarguedasi and the collecting localities of A. kotosh sp. nov., 1) Carpish (Type locality), 2) Huanacaure (based on Jiménez et al. 2013 and Pardiñas et al. 2015).
Paratypes. Four specimens, two males (MUSM 19006, 18998) and two females MUSM (19036, 19023), collected at the type locality in August 2002 and preserved as skulls and skins.
Nomenclature statement. A life science identifier (LSID) number was obtained for the new species Akodon kotosh: urn:lsid:zoobank.org:pub:DE7571D4-5F5A-4D4A-87DD-9E8CC2183FB5
Diagnosis. Akodon kotosh sp. nov. is distinguished from other species of the Akodon aerosus species group by the following combination of characters: large size (TL = 170 to 222 mm; LT = 71.5 to 100 mm), uniform dark olive brown coloration, with dorsal and ventral pelage not countershading, ventral hairs with dark bases; unicolored tail; large and broad skull (GSL= 26.06 to 28.95 mm, ZB = 12.65 to 14.40 mm); broad nasals (BD = 2.95 to 3.72 mm); large incisive foramen (LIF = 5.68 to 6.71 mm); large lacrimals; broad antorbital bridge; wide zygomatic notch being somewhat shallow in depth; conspicuous gnathic process; incisive foramen that extends backward to the protoflexus of the first upper molar M1; broad interorbital region (LIB = 5.13 to 5.68 mm); narrow parapterygoid fossa; small foramen ovale; robust jaw without a prominent capsular process; low-crowned molars; and anterior edge of the masseteric crest placed posterior to procingulum of m1; bilobed m3 with small entoconid/hypoconid cusp pair.
Description. A large member of the genus Akodon (Table 2), the overall color of the dorsum is dark olive brown, fur hairs average about 10.0 mm in length, the base of which is dark gray, the subterminal band is pale brown, and the tip is dark brown. Guard hairs are gray at the base and black at the tip, extending up to 2 mm beyond fur hairs. The venter is olive gray and slightly paler than the dorsum. The chin has a small patch of white hairs (Figure 3). Eyerings are inconspicuous. Forefeet are slightly paler and covered with black or brown-based hairs. The ungual tufts on the manus are white. The hindfeet are covered by dark brown hairs. The ungual tufts of pes are brown or bicolored with white tips. Ears are covered by delicate, short hairs. Mystacial vibrissae are moderately long and reach the ears when bent, some are white and others are black; the submental and interramal vibrissae are white. One to three superciliary vibrissae are present, one of them more conspicuous, and genal-1 is present; both types of vibrissae are black. The tail is short, about 85% of the head-body length, and slightly contrasting in color; the hairs on the dorsal surface are black, while those on the ventral side are white; scales on both sides of the tail are small and black. Dorsal and ventral hairs are over 2 scales in length.
Table 2 Summary statistics (X, standard deviation, range) for 23 external and cranial measurements (in mm) of adult specimens (age classes 4 and 5) of Akodon orophilus s. s., Akodon kotosh sp. nov., and Akodon josemariarguedasi. Variables are described in Material and Methods.
Figure 3 Dorsal and ventral views of museum study-skin of Akodon kotosh sp. nov. (MUSM 18947; left), A. orophilus s. s. (MUSM 36984; center), and A. josemariarguedasi (MUSM 22759; right). Scale bar = 10 mm.
The skull is slightly elongated, with a rounded braincase (Figure 4). The profile of the skull is flat. The rostrum is arched, long, and somewhat broad. The gnathic process is conspicuous. The nasals are long and wide, extending anteriorly beyond the premaxillae, but not expanding beyond the lacrimals posteriorly. The anterior edges of nasals are pointed with posterior margins slightly blunt or somewhat pointed. The premaxilla extends posteriorly slightly behind the nasals. The fronto- parietal suture is somewhat rounded or little rounded in Huanacaure specimens.
Zygomatic notches are wide and moderately deep with rounded borders. Zygomatic arches are robust and anteriorly convergent, and the malar process is thick. Lacrimals are large. Lambdoid ridges are well developed, and the interparietal bone is reduced. The interorbital region is hourglass- shaped without ridges and broad. Incisive foramina extend only to the protoflexus of M1, not reaching the protocone and diverging on the posterior border. The maxillary septum occupies less than half of incisive foramina length. Anterior palatal pits are conspicuous and located at the level of the second upper molar (M2) hypoflexus. Posterior palatal pits are located slightly behind the anterior margin on both sides of the mesopterygoid fossa and are absent in some cases. The mesopterygoid fossa is broader than the parapterygoid fossa at midpoint and has parallel margins, extending anteriorly to the posterior border of the third upper molar (M3) or slightly anterior to it, producing a short palate; its anterior margin is slightly biconcave. The parapterygoid plate is narrow, fenestrated, and divergent with straight external margins. Sphenopalatine vacuities are usually present and slightly developed. Auditory bullae are flask-shaped and intermediate in size, with short and broad Eustachian tubes.
Laterally, the zygomatic plate is moderately broad. The anterior border is generally convex. A distinct masseteric tubercle is present in front of the root of the zygomatic plate. The posterior ascending process of the alisphenoid is always ventral to the squamoso-alisphenoid groove. The tegmen tympani slightly overlaps the posterior process of the squamosal. The dorsal aperture of the ectotympanic ring is either opened or closed. The nasolacrimal foramen is present, its diameter being smaller than or equal in size to M2. The oval foramen and foramen ovale accessorius are usually smaller than M3. The optic foramen is larger than M3. The ethmoid foramen is dorsal to M3. The carotid circulation corresponds to pattern 1 with the sphenofrontal foramen, stapedial foramen, and squamoso-alisphenoid groove present. The postglenoid foramen is large, almost twice the size of the subsquamosal fenestra, and both are rounded. The hamular process is usually thick and curved but thicker and shorter in specimens from Huanacaure. The sphenopalatine foramen is shorter than M2 and mostly closed.
The upper molars are crested, and the main cusps show an alternate arrangement. The anterolabial and anterolingual conules of M1 are approximately equal in size and are divided by a somewhat developed anteromedian flexus. The paraflexus and metaflexus are conspicuous but the anteroflexus and posteroflexus are shallow. The anteroloph, mesoloph, posteroloph, paralophule, parastyle, and mesostyle are present, while the metalophule and mesoflexus are absent. On the lingual side, the protoflexus and hypoflexus are present while the enterostyle is absent. In M2, the protocone, paracone, hypocone, and metacone show a diagonal arrangement; the paraflexus, metaflexus, protoflexus, and hypoflexus are present but the anteroloph, mesoloph, and metalophule are absent; the mesostyle, paralophule, and posteroflexus are present. The posteroloph is poorly defined. M3 has a deep metaflexus, posteroflexus, posteroloph, and hypoflexus. The paracone and protocone are larger than the metacone and hypocone. In some individuals, molars show fossettes that could represent vestigial mesoflexi. In the lower toothrow, the conids of the first lower molar (m1) and second lower molar (m2) are arranged diagonally with the metaconid and entoconid anterior to the protoconid and hypoconid, respectively. In m1, the anteromedian flexid is inconspicuous; the anterolingual conulid is smaller than the anterolabial conulid; the protostylid is well developed and the ectostylid is minute; the posterolophid and posteroflexid are well defined. The second lower molar presents a tiny mesostylid. The third lower molar is eight-shaped and small, with a small posterior lobe corresponding to the entoconid-hypoconid cusp pair (Fig. 5).
Figure 5 Detailed occlusal views of upper (top) and lower (bottom) right molar series of Akodon kotosh sp. nov. (MUSM 18598; left), A. orophilus s. s. (MUSM 36583; center), and A. josemariarguedasi (MUSM 22750; right).
The lower jaw is similar to the one observed in other species of Akodon, with the mandibular rami somewhat robust. The coronoid process is delicate and falciform, and its dorsal margin lies above the condylar process. The lunar notch is deep and wide and the angular process is robust. The condylar process is moderately long, usually placed at the same level as the angular process.
The lower incisor alveolus lacks a distinct capsular process; mandibular and mental foramina are present, although reduced. The masseteric tubercle and the lower masseteric crest are highly noticeable. The anterior edge of the masseteric crest is posterior to the procingulum of m1.
Akodon kotosh sp. nov. has 12 thoracic ribs; the vertebral column includes 7 cervical, 12 thoracic, 6 lumbar, 4 sacral, and 28 caudal vertebrae.
Etymology. The term kotosh derives from a noun in apposition referring to an archaeological site where a pre-Columbian temple called "manos cruzadas" is located. This site is about 5 km from the town of Huánuco. This pre-Columbian culture emerged about 1,800 BC. Kotosh is one of the most important archaeological sites in Huánuco department and is located 30 km southwest of the type locality.
Distribution. Akodon kotosh sp. nov. is currently known from two localities only: Caserío San Pedro de Carpish and Huanacaure forest, Cordillera de Carpish, Huánuco department (Figure 2). Comparisons. Akodon kotosh sp. nov. can be differentiated from A. orophilus by its darker coloration; longer tail (91.10 mm versus 85.19 mm on average); shorter ears (15.69 mm versus 16.55 mm on average); longer (27.48 mm versus 27.04 mm on average) and broader skull (13.51 mm versus 13.37 mm on average); broader nasals (3.31 mm versus 3.09 mm on average) that are more projected anteriorly; longer incisive foramina (6.14 mm versus 5.82 mm on average) reaching the protoflexus of M1; smaller entoconid-hypoconid cusp pair; a conspicuous masseteric tubercle; and a narrower parapterygoid fossa with parallel sides (Figures 5, 6).
Figure 6 Cranial differences between Akodon kotosh sp. nov. (left) and A. orophilus s. s. (right). A) Narrower parapterygoid fossa in Akodon kotosh sp. nov. B) The nasal bone is projected more anteriorly in Akodon kotosh sp. nov. C) Broader nasal bone in Akodon kotosh sp. nov. D) Shorter incisive foramen in Akodon kotosh sp. nov. if = incisive foramen; max = maxillary; na = nasal; pal = palatine; pf = parapterygoid fossa; pre = premaxillary.
Akodon kotosh sp. nov. is distinguishable from A. josemariarguedasi by its darker and uniform body pelage; larger hindfoot (23.96 mm versus 22.84 mm on average); longer (27.48 mm versus 26.81 mm on average) and broader skull (13.51 mm versus 12.92 mm on average); nasal projecting little beyond the premaxilla; incisive foramina reaching only to the protoflexus of M1; broader palate (2.96 mm versus 2.71 mm on average); larger molars (4.47 mm versus 4.22 mm on average); broader mesopterygoid fossa with parallel margins; narrower parapterygoid fossa with straight outward edges; and smaller entoconid-hypoconid cusp pair; inconspicuous anteromedian flexus and flexid (Figure 5).
Akodon kotosh sp. nov. differs from A. torques by its slightly darker coloration. Nasals are longer (10.87 mm versus 10.65 mm on average) and broader than in A. torques (3.31 mm versus 3.07 mm on average); the posterior margin of nasals is less tapered and it does not extend beyond the lacrimals; the incisive foramen extends backward to the protoflexus of M1. Akodon kotosh sp. nov. exhibits a rounded fronto-parietal suture; narrower and longer parapterygoid fossa with straight external borders; and a larger oval foramen and broader zygomatic plate (2.04 mm versus 1.85 mm on average).
Akodon kotosh sp. nov. is distinguishable from A. aerosus by its darker coloration; longer tail (91.10 mm versus 87.37 mm on average); deeper zygomatic notch and narrower antorbital bridge; narrower and rectangular zygomatic plate with a convex anterior border (2.04 mm versus 2.22 mm on average); the incisive foramen extends backward only to the protoflexus of M1; smaller oval foramen and indistinct capsular process.
Akodon kotosh sp. nov. is distinguished from A. surdus by its more olivaceous tinge, paler toe and longer tail (91.10 mm versus 70.00 mm on average); smaller skull (27.48 mm versus 28.93 mm on average) with shorter (10.87 mm versus 11.43 mm on average) and narrower nasals (3.31 mm versus 3.60 mm on average). It has a more pointed anterior border and a less tapered posterior border; incisive foramina extend backward to the protoflexus of M1; narrower interorbital region (5.44 mm versus 5.52 mm on average); narrower antorbital bridge; narrower and slating zygomatic plate (2.04 mm versus 2.31 mm on average); narrower mesopterygoid; and parapterygoid fossa with straight external borders.
Akodon kotosh sp. nov. markedly differs from A. budini by its shorter ears (15.69 mm versus 20.2 mm on average); unicolored tail; less hypsodont molars; shallower zygomatic notch; incisive foramen extending backward to the protoflexus of M1; narrower mesopterygoid fossa; broader interorbital region (5.44 mm versus 4.96 mm on average); longer nasal (10.87 mm versus 10.09 mm on average); shorter molars (4.47 mm versus 4.98 mm on average); narrower zygomatic plate (2.04 mm versus 2.36 mm on average); and inconspicuous anteromedian flexid.
Akodon kotosh sp. nov. can be easily distinguished from A. siberiae by its shorter ears (15.69 mm versus 18.5 mm on average); shorter dorsal hairs (10 mm versus 12 mm on average); broader interorbital region (5.44 mm versus 5.15 mm on average); shorter molars (4.47 mm versus 4.81 mm on average); shorter incisive foramen (6.14 mm versus 6.37 mm on average) that extends backward to the protoflexus of M1; and narrower mesopterygoid fossa with straight lateral margins.
Two other species occur in montane forests: Akodon fumeus and A. kofordi. Both belong to the Akodon boliviensis species group (sensu Pardiñas et al. 2015). In general, they have a different color pattern with a dark brown dorsum with an olivaceous tone and a buff ventral side; they are smaller in most external and cranial dimensions. Both have a narrower mesopterygoid fossa and interorbital region than Akodon kotosh sp. nov. Natural History. Akodon kotosh sp. nov. is sympatric with Microryzomys minutus, Thomasomys notatus, T. ischyrus, and T. kalinowskii. No pregnant individuals were found during fieldwork in February, May, or November.
Akodon kotosh sp. nov. inhabits mountain forests above 2,400 m.a.s.l. Carpish is a dense montane forest with trees not exceeding 20 m. Trees are conspicuously covered by epiphytes, mosses, lichens, and bromeliads. The most abundant plant families at Carpish are Orchidaceae, Asteraceae, Melastomataceae, and Rubiaceae (Beltrán and Salinas 2010).
Remarks. Molecular data for the new species is not available, although sequences are available for specimens previously assigned to A. orophilus by Smith and Patton (1993; 2007) that were collected in the nearby areas of Palca (Junín) and Unchog (Huánuco). The first one corresponds to Akodon josemariarguedasi and the second could belong to an undescribed species. Based on molecular data and some morphological features, Coyner et al. (2013) indicated that the specimen from Unchog (Huánuco department) belongs to Akodon orientalis, a form previously considered a synonym of A. orophilus. However, Jiménez et al. (2013) pointed out that there is no morphological difference between A. orophilus orientalis and A. o. orophilus except that specimens of orientalis from Amazonas and northeastern San Martín department are darker; meanwhile, Pardiñas et al. (2015) considered orientalis as a subspecies of orophilus. For us, this specimen may represent an unnamed species. This specimen exhibits some differences with A. orophilus, including a ventral side with whitish hair tips, larger lacrimals, serrated contour of the posterior end of nasal, and a more delicate mandible. Compared to A. kotosh sp. nov., the specimen from Unchog has a longer incisive foramen that extends to the protocone of M1, rounded anterior border of the nasal, and a more delicate mandible. In addition, this specimen has a more conspicuous gnathic process, a pointed posterior margin of nasal, shallow zygomatic notches, a long incisive foramen, and inconspicuous antero-palate pits, relative to A. josemariarguedasi.
Similarly, populations from five localities in northern Huánuco department (Kenwarajra, Pampa Hermosa, Iscarag, Campamento Provias, and Campamento regional) and two localities in central San Martín (La playa and Los Chochos) were originally allocated to Akodon orophilus; however, our more detailed analysis points to an inconclusive status for them due to their high morphological variability. These specimens do not fully agree with the descriptions of A. josemariarguedasi, A. orophilus s. s., nor A. kotosh sp. nov. For instance, populations from northern Huánuco department have a more yellowish body coloration, square posterior border of the nasal, conspicuous posterior palatal pits, and a larger lacrimal; individuals from San Martín are slightly smaller, with larger lacrimals, a more yellowish rostrum, more conspicuous eyerings, and bicolored hairs on the ventral side of the tail, compared to A. kotosh sp. nov., A. orophilus s. s., and A. josemariarguedasi. We suggest that molecular analyses could contribute to clarify this complex taxonomic scenario.
After the description of Akodon kotosh sp. nov., nine species of Akodon are known from the Peruvian montane cloud forests. In addition, our results show that further studies are still needed to clarify the true diversity of Akodon inhabiting the Peruvian cloud forest, particularly when other member of the aerosus group such as A. mollis and A. aerosus could encompass more than one species. This is also true for other rodent species that inhabit this kind of forest, which is home to a number of endemic mammal species. For that reason, the establishment of more protected areas is necessary because the montane cloud forests suffer one of the highest rates of species loss (Gentry 1992) owing to the conversion of forested land and logging.
Acknowledgments
We are very grateful to colleagues in the Departamento de Mastozoología del Museo de Historia Natural for all their support during fieldwork. We would like to thank Ph. D. T. Gregory and L. Sperling for providing valuable suggestions and grammar corrections that helped to improve this contribution. This work was greatly benefited from the comments of two anonymous reviewers. This work was supported by Consejo Superior de Investigación under the multidisciplinary project "Biodiversidad, conservación, impacto ambiental y macrozonificación de los bosques montanos del departamento de Huánuco"and by CONCYTEC through the grant"Fondo concursable Santiago Antúnez de Mayolo 2001", both granted to VP. Maria Elena Sánchez Salazar contributed to the edition of this manuscript with support of THERYA through a CONACYT grant.
