Rediscovery of Rhipidomys ochrogaster JA Allen ...

Zootaxa 3106: 42–59 (2011) www.mapress.com / zootaxa/ Copyright © 2011 · Magnolia Press

ISSN 1175-5326 (print edition)

Article

ZOOTAXA ISSN 1175-5334 (online edition)

Rediscovery of Rhipidomys ochrogaster J.A. Allen, 1901 (Cricetidae: Sigmodontinae) with a redescription of the species VÍCTOR PACHECO1,2,3 & MARÍA PERALTA1 1

Museo de Historia Natural, Universidad Nacional Mayor de San Marcos, Apartado 14-0434, Lima-14, Peru Instituto de Ciencias Biológicas “Antonio Raimondi”, Facultad de Ciencias Biológicas, Universidad Nacional Mayor de San Marcos, Apartado 14-0434, Lima-14, Peru 3 Corresponding author. E-mail: [email protected] 2

Abstract Rhipidomys ochrogaster is a poorly known sigmodontine rodent from the Yungas of southeastern Peru, known only from a few specimens collected more than 110 years ago. This work reports a new specimen from Yanacocha, Puno, Peru, collected on 22 May 2010; provides a morphological description and comparisons with R. leucodactylus and R. gardneri; characterizes the habitat; and presents data on the diet of the species. This discovery stresses the urgent need for additional inventory work in the region, in order to document its biodiversity. Key words: Rodentia, Cricetidae, Rhipidomys, Yungas, Peru

Introduction Among the more than 508 mammal species recognized for Peru (Pacheco et al., 2009), one of the least known is Rhipidomys ochrogaster J.A. Allen, 1901. According to Musser and Carleton (2005), this species was described from Peru, Puno Dept., valley of Río Inambari, Inca Mines (= Santo Domingo), 6000 ft (1830 m), based on two female specimens, the holotype (AMNH 16481), collected on 14 September 1900, and the paratype (AMNH 16482), collected on 9 October 1900, both by H. H. Keays. For almost a century, these were the only specimens known in the literature and all citations (Cabrera 1961, Musser and Carleton 1993, 2005) are apparently based upon them alone. More recently, Tribe (1996) examined two additional specimens housed at the Natural History Museum, London. These specimens have the following information (P. Jenkins, pers. comm.): BM(NH) 1902.7.27.2 collected on 14 December 1901, and BM(NH) 1902.7.27.3 collected on 29 December 1901. Both specimens are male, skins and skulls, collected at Santo Domingo, Carabaya, Peru, 5500 feet [1676 m], by G. Ockenden (collector numbers 217 and 218 respectively), and purchased from W. Rosenberg. Rhipidomys ochrogaster was retained as a species until Cabrera (1961) synonymized it under Rhipidomys leucodactylus (Tschudi) without providing an argument. Later, Musser and Carleton (1993, 2005) and Tribe (1996) considered R. ochrogaster as a valid species. Tribe (1996) divided the genus Rhipidomys in three sections: "leucodactylus", "fulviventer", and "macconnelli" mainly based on the type of carotid circulatory pattern; and placed R. ochrogaster in the "fulviventer" section (not in the "leucodactylus" section as stated by Musser and Carleton, 2005: 1171). The scarcity of specimens of Rhipidomys ochrogaster led to its being classed as Vulnerable in the Peruvian official list of threatened species (Ministerio de Agricultura, 2004), but only as Data Deficient, ver. 3.1, in the IUCN Red List (Zeballos et al., 2008). We herein report a new record of Rhipidomys ochrogaster, provide a full morphological description including several unreported characters, compare with potential sympatric species, discuss the current distribution, and present new information on the habitat and diet of the species. The conservation status of the species is also addressed.

42

Accepted by M. Weksler: 20 Oct. 2011; published: 22 Nov. 2011

Material and methods Trapping was carried out at Yanacocha, Alto Inambari District, Sandia Province, Puno Department, Peru (14º 11' 37.5" S; 69º 15' 20.4'' W, 1942 m; Fig. 1), close to the Bolivian border, as part of a faunal survey of the headwaters of the Río Tambopata. Some 240 traps were set for seven consecutive nights, totaling 1680 trap-nights. One snap trap (Victor trap or Museum Special) and one Sherman trap were set together at each trap-station on the ground, but at some trap-stations one of the traps was placed on a branch. The morphological redescription of Rhipidomys ochrogaster is based on the holotype (AMNH 16481), the paratype (AMNH 16482), and the specimen herein presented; the type specimens were examined by the senior author. Five external and 19 cranial measurements were taken with digital calipers to the nearest 0.01 mm following Voss (1993) (see Table 1). Age criteria, and descriptions of external, dental and cranial morphological features are according to Voss (1988), Tribe (1996), and Pacheco (2003). Dental nomenclature follows Carleton and Musser (1989), and stomach morphology was examined according to the description of Carleton (1973). Color nomenclature follows Smithe (1975). Rhipidomys ochrogaster was compared with 12 specimens of R. leucodactylus, and 18 specimens of R. gardneri Patton, da Silva, and Malcolm, species of similar size and members of the "leucodactylus" section (sensu Tribe, 1996) (Appendix 1).

Results Among the material collected in the Río Tambopata drainage, a new specimen of Rhipidomys ochrogaster, a subadult male (age 2), was collected at Yanacocha in a Victor snap trap at ground level, by María Peralta (MCP 857) on 22 May 2010. The external measurements and weight are: Total length 350.6, Length of tail 198.5, Length of foot 34.5, Length of ear 21; and Weight 129 g. The exemplar was prepared as a fluid specimen with skull removed and was deposited in the mammal collection of Museo de Historia Natural, Universidad Nacional Mayor de San Marcos, under catalog number MUSM 35095. The specimen was identified by comparisons with the holotype description (Allen, 1901), and the morphological descriptions of external, cranial and teeth characters provided by Tribe (1996) and Pacheco (2003); as well as the use of unpublished measurements of the type series taken by the senior author. The examination of this specimen and the type series led us to present an updated description of the morphology of the species.

Rhipidomys ochrogaster J.A. Allen, 1901 Type locality. “Inca Mines [= Santo Domingo Mine], about 200 miles northeast of Juliaca, on the east side of the Andes, on the Inambary River [Río Inambari].” Department of Puno, Peru. Distribution. Only known from the Yungas of southeastern Peru, Department of Puno, in the Río Inambari drainage (Allen, 1901; Tribe, 1996), and the Río Tambopata drainage (this work), from 1220 m to 1942 m (Fig. 1). Revised diagnosis. A large-sized, long-tailed species of Rhipidomys with orange-brown dorsal fur and pale orange or melon ventral fur; tail dark, considerably longer than head and body, and with long terminal pencil; broad hind foot with metatarsal patch extending onto first phalanges of toes. Skull moderately large, with rather narrow and pointed rostrum, broad interorbital region and broad, rounded braincase; supraorbital shelf well developed and divergent backward; postorbital ridges absent or weak; primitive (pattern 1) carotid circulation (large stapedial foramen, squamosal-alisphenoid groove visible but not strong, and sphenofrontal foramen); and narrow zygomatic plate. Description. External morphology. A large-sized Rhipidomys with orange-brown dorsal pelage and relatively long hairs (cover hairs: 11–13 mm, guard hairs: 14–17 mm), pale-orange ventral pelage moderately countershaded (Fig. 2). The dorsal coloration is Color 26 Clay (color codes follow Smithe, 1975), washed with Color 28 Olive Brown; ventrally the pelage is Color 24 Buff, and the margins are Color 39 Cinnamon. The chin is whitish. Ventral hairs lack a gray-based band except on the chest. A middorsal stripe and a pectoral streak are lacking. Ears are small and dark with a very narrow antitragus; an auricular patch behind the ears is absent; the orbicular ring is dark but narrow. The anus is not protruded. RHIPIDOMYS FROM PERU

Zootaxa 3106 © 2011 Magnolia Press ·

43

FIGURE 1. Collecting localities of Rhipidomys ochrogaster in Peru, Puno department: (1) Inca Mines (= Santo Domingo) in the valley of Río Inambari; 69° 41' W, 13° 51' S; (2) Yanacocha, in the valley of Río Tambopata; 69º 15' 20.4'' W, 14º 11' 37.5" S; 1942 m.

44 · Zootaxa 3106 © 2011 Magnolia Press

PACHECO & PERALTA

FIGURE 2. External appearance of Rhipidomys ochrogaster (MUSM 35095): (A) lateral view, (B) dorsal view, and (C) ventral view. Scale is 20 mm.

RHIPIDOMYS FROM PERU


45

FIGURE 3. Hind feet and tail of Rhipidomys ochrogaster (MUSM 35095): (A) dorsal view of left foot, (B) ventral view of left foot, and (C) terminal part of the tail showing a well developed pencil. Scale is 10 mm.

Genal, superciliary, and mystacial vibrissae are present. Mystacial vibrissae are very long, extending back far beyond the pinna. Genal and superciliary vibrissae are long and also projected backward far beyond the pinna when appressed to the body. Ungual tufts on the manus cover the claws. The metacarpal region is dark brown flecked with white hairs. The claws of the manus are curved with open bases; the thenar and hypothenar pads are both large and similar in size; digit II is longer than digit V. Pes long and relatively broad, with long ungual tufts that cover the claws; the metatarsal patches are dark brown (Color 27 Drab), flecked with golden hairs, and extending to first phalanges of toes; the terminal portion of toes and sides of the feet are pale orange to cream (Fig. 3a). Six plantar pads are present; the thenar and hypothenar are well developed, the first is larger than the second; the hypothenar is located directly behind the pad IV, and overlaps the thenar and pad I (Fig. 3b). Pad region of the plantar surfaces are squamated. Digits on the pes are long, digit I extends to halfway along the second phalange of digit II, and digit V to the base of the claw of digit IV (Fig. 3a, b). The tail is uniformly dark, longer than head and body length (130% in our specimen and 148% in the holotype), and terminates in a long pencil of hairs of 17 to 23 mm in length (Fig. 3c). The base of the tail is not densely haired. Hairs along the shaft extends over 2.5 rows of scales; and one centimeter covers 13 or 14 rows of the tail. There are three pairs of mammae in inguinal, abdominal and pectoral position (sensu Pacheco, 2003). Cranial morphology. Skull is relatively large (greatest length 36.61 and 38.16 mm in the holotype and our specimen respectively) with a convex braincase profile (Fig. 4). Rostrum relatively narrow and pointed with reduced gnathic process; nasals are long, narrow, and tapering posteriorly to a point posterior to premaxillary-maxillary-frontal joint. Zygomatic notches are shallow; interorbital region is moderately broad, with base of molars visible from dorsal view; supraorbital shelf is prominent and divergent backward, the anterior half is also expanded; postorbital ridge is absent or weak; braincase broad and rounded. Premaxillary-maxillary-frontal joint placed behind zygomatic notch. Zygomatic plate is vertical and narrow, almost as broad as the length of M1. Zygomatic arches with rounded profile but narrow anteriorly, and do not dip to the plane of molar ridges. Anterior orbital bridge is moderately high. Lacrimals are small, elongated, and without posterior process; frontals are broad, flat, and without a medial depression; fronto-parietal suture is rounded or triangular; a rhomboid interparietal is longitu-


PACHECO & PERALTA

dinally longer than half the length of parietals. Parieto-occipital suture is about a third of the parietal breadth. Incisive foramina is long and broad, with posterior margins rounded, and extend backward to a place between M1 procingulum; septum is long and narrow, close to half the length of incisive foramina. Diastema shape is concave; M1 are placed posterior to the base of zygomatic plate. Palate length is short and broad, extending to M3 but not reaching the hypoflexus of M3. Posterolateral palatal pits are small and placed at both sides of mesopterygoid fossa. Maxillary pits are large and placed at level of M1 procingulum. Sphenopalatine vacuities are closed or represented by small slits; maxillary base of molars are rectangular. Mesopterygoid fossa is broad with sides parallel; medial posterior process of palate is very small. Parapterygoid fossa is triangular, deep, and without vacuities; posterior alisphenoid channel and foramen ovale are large; middle lacerate foramen is conspicuous, extends from basisphenoid to near the tegmen tympani; posterior margin of squamosals lacks a notch; alisphenoid strut is present and robust; anterior alisphenoid channel is small. Carotid circulation pattern 1 as determined by Voss (1988): large stapedial foramen, squamosal-alisphenoid groove visible but not strong, and sphenofrontal foramen. Internal carotid canal is relatively small and bounded by basioccipital and Eustachian tube, but not the periotic. Tegmen tympani are large and robust and sinus groove are present. Jugal is robust, maxillary and squamosal endings do not overlap. Ectotympanic ring is open, dorsal margin of ectotympanic do not reach the petrosal; ventral margin of ectotympanic has a small leaf-process. Postglenoid foramen is very small and hidden by the hamular process; subsquamosal fenestra is placed posterior or dorsal to postglenoid foramen, and with its posterior margin opened. Lambdoid ridge is absent, the occipital condyle is at the same level of braincase; paraoccipital process is conspicuous and not bifurcated; ethmoid foramen is placed dorsal to M2; ethmoturbinals are moderate in size; sphenopalatine foramen is bordered by palatine bones; optic foramen is relatively large and dorsal to M3. Bullae are small and not globose; Eustachian tube is long and tubular; anterior process is broad and short, with two spines that may or may not reach the alisphenoids; basioccipital is very broad; mastoids lack perforation. Lamina of malleus is very narrow and not deep, processus brevis of incus is long, conical, and relatively robust; orbicular apophysis is short and massive. Dental morphology. Incisors deep, moderately opisthodont, with wearing surface facing caudally; upper and lower molar cusps are slightly alternated; tooth design and topography are bunodont and brachyodont; molar hypoflexus and paraflexus do not overlap (Fig. 5). Labial cingulum on upper molars are well developed. Procingulum breadth of M1 is less than the breadth across the protocone and paracone. Anteromedian flexus is well defined and separates a smaller anterolingual conule from the anterolabial conule. M1 anteroloph is present and perpendicular to the mure; anterolophule is present and fills the anteroflexus; paraloph is perpendicular to the mure; mesoloph is perpendicular to the cingulum; hypoflexus is distinct and triangular in shape; protostyle and enterostyle are present; paralophule well developed reaching mesoloph. M2 protoflexus is present and small, paraflexus and anteroloph are complete; M2 paraloph obliquely oriented to the mure; hypoflexus and posteroflexus of M2 distinct: paralophule also distinct. M3 topography is crenulated, with a distinct hypoflexus and a low metacone. The procingulid of m1 is divided by the anteromedian flexid into two subequal conules. The m1 presents a well developed lingual cingulid, and a cingulid shelf on the anterolabial margin of the procingulid; protolophid and anterolophid are present; mesolophid and entolophid are perpendicular to murid; posterolophid and posteroflexid are long and narrow; ectolophids are absent; anterolabial cingulum of m2 and m3 are conspicuous. Mandible and other features. Masseteric crests level with the anterior edge of m1; mental foramen large and placed close to the diastema; capsular process not developed and forming a shelf; mandibular bone not deep, with the ventral margin shallow. Coronoid process short, triangular, slightly ventral to the condylar process; condylar process long, robust, and extending posterior to the angular process. Hyoid basihyal lacks medial process. The morphology of the stomach conforms to the unilocular-hemiglandular pattern sensu Carleton (1973); and it is similar to the illustration of Thomasomys cinereus in Carleton (1973: fig 4b). The stomach content analysis revealed a great abundance of seed and insects. Seeds belonged to the species Piper sp. 1 and Piper sp. 2 (Arias, E., pers. comm.); while insects were represented by only one species of army ants of the genus Labidus (Azorsa, F., pers. comm.). Seeds represented about 30% and the army ants 70% of items in 10 randomly selected fields (Arias, E., pers. comm.). The gall bladder is absent. No endoparasites were observed. Comparisons. This section is based on characteristics presented by Tribe (1996), Patton et al. (2000) and by our own examination. Compared to Rhipidomys leucodactylus, R. ochrogaster may be distinguished externally by the bright orange-brown dorsal pelage coloration instead of a dark-brown; ventral fur is pale orange instead of white or light yellowish, and with gray-based hairs only on the chest versus throughout all venter; dorsal hairs are



47

FIGURE 4. Dorsal, ventral and lateral views of cranium and mandible of Rhipidomys ochrogaster (MUSM 35095). Scale is 10 mm.


PACHECO & PERALTA

FIGURE 5. Occlusal view of molar rows of Rhipidomys ochrogaster (MUSM 35095): (A) left upper molars, (B) right lower molars. Scale is 1 mm.



49

FIGURE 6. Dorsal and ventral view of the skull of three species of Rhipidomys: (A) R. gardneri (MUSM 14472), (B) R. leucodactylus (MUSM 12987), and (C) R. ochrogaster (MUSM 35095). Scale is 10 mm.


PACHECO & PERALTA

FIGURE 7. Lateral view of the skull and mandible of three species of Rhipidomys: (A) R. gardneri (MUSM 14472), (B) R. leucodactylus (MUSM 12987), and (C) R. ochrogaster (MUSM 35095). Scale is 10 mm.



51

FIGURE 8. Occlusal view of upper and lower molar rows of three species of Rhipidomys: (A–D) R. gardneri (MUSM 14472), (B–E) R. leucodactylus (MUSM 12987), and (C–F) R. ochrogaster (MUSM 35095). Scale is 1 mm.

longer, 11–13 mm versus 6–10 mm; metatarsal patch coloration extends into the first phalanges of toes versus extending the entire toes (Patton et al., 2000); tail with a terminal tuft of hairs from 17 to 23 mm long versus about 25 mm in our samples of R. leucodactylus, although it might reach 40 mm in the last taxon (Tribe, 1996); scales cover 13 or 14 rows per 10 mm longitudinally versus 12 rows. R. ochrogaster is also distinguished from R. leucodactylus in several cranio-dental characteristics (Figs. 6, 7). The braincase is broad and rounded versus narrow and more elongated, and the rostrum is narrow versus broad (Tribe, 1996). The mesopterygoid fossa is narrow versus broad. The carotid circulatory pattern is primitive (stapedial foramen, internal groove on the squamosal and alisphenoid, and sphenofrontal foramen present) versus the derived condition (Tribe, 1996); and orbicular apophysis is short versus long and digit-like (Fig. 9). Rhipidomys ochrogaster is also generally smaller in several skull measurements (Table 1): Greatest length of skull, Condyloincisive length, Condylomolar length, Length of orbital fossa, Diastema length, Length of incisive foramina, and Length of maxillary toothrow; and narrower in Breadth of inci-


PACHECO & PERALTA

sive foramina, Breadth of rostrum, Breadth of palatal bridge, Breadth of nasals, and Zygomatic breadth. It is also shorter in Breadth of zygomatic plate and Depth of incisors. Rhipidomys ochrogaster is also distinct from R. gardneri by the bright orange-brown dorsal fur coloration instead of yellow brown; ventral fur is pale orange instead of whitish, and with gray-based hairs only on chest versus on the throat and chest (Patton et al. 2000) or throughout all venter (this work); dorsal hairs are longer, 11–13 mm versus 5–10 mm; dark metatarsal patch extends into the first phalanges of toes versus not extending onto the toes (Patton et al., 2000); tail with moderate scales, 13 or 14 per 10 mm longitudinally versus 16 on average (Patton et al. 2000), and the terminal tuft of hairs in the tail is longer, 17–23 mm versus 6–14 mm. In the skull, the braincase in R. ochrogaster is broad and rounded versus narrow and more elongated, and the rostrum is narrow versus broad (Tribe, 1996); the zygomatic notch is shallow versus deep, and the mesopterygoid fossa lacks a medial spine on the anterior border versus present (Patton et al., 2000); the carotid circulatory pattern is primitive (stapedial foramen, internal groove on the squamosal, and sphenofrontal foramen present) versus the derived condition (Tribe, 1996); processus brevis of incus is robust and conspicuous versus weak and slender (Fig. 9). Also, R. ochrogaster exhibit a distinctly narrow nasal bone (4.18 versus 4.62 in average) and a narrow zygomatic plate (2.93 versus 3.58 in average) in comparisons to R. gardneri (Table 1). TABLE 1. External and cranial measurements (in millimeters) of Rhipidomys leucodactylus, R. gardneri, and R. ochrogaster from Peru. Measurements are given in mean ± standard deviation, with range and sample size. Characters

R. leucodactylus

R. gardneri

R. ochrogaster AMNH 16481a MUSM 35095

Head and body length

176 ± 11.53 (163–185) 3

171 ± 9.165 (161–179) 3

154

152.1

Total length

404.7 ± 11.68 (392–415) 3

356.33 ± 7.09 (350–364) 3

382

350.6

Length of tail

228.67 ± 6.50 (222–235) 3

185.33 ± 8.50 (177–194) 3

228

198.5

Length of foot

36.67± 2.52 (34–39) 3

34.66 ± 3.51 (31–38) 3

35

34.5

Length of ear

19.7 ± 5.09 (15–25.1) 3

19.33 ± 0.577 (19–20) 3

19

21

Greatest length of skull

41.4 ± 1.2 (40–43) 10

40.4 ± 1.41 (38.2–41.77) 6

36.61

38.16

b

Condyloincisive length

38.5 ± 1.3 (37–41) 10

36.77 ± 1.31 (34.92–38.33) 6

34.34

34.51

Condylomolar length

25.1 ± 0.7 (23.9–26.3) 10

24.04 ± 0.72 (23.22–25.07) 6

22.72

23.15

Length of orbital fossa

13.7 ± 0.4 (13.1–14.2) 10

13.38 ± 0.52 (12.7–14.06) 6

11.99

12.66

Length of nasals

14.9 ± 0.7 (14–15.7) 10

14.00 ± 0.61 (13.05–14.83) 6

14.23

14

Diastema length

11.2 ± 0.7 (10.5–11.6) 10

10.62 ± 0.49 (9.8–10.93) 6

9.63

9.46

Length of incisive foramina

8.5 ± 0.3 (8.2–9.4) 10

7.57 ± 0.79 (6.1–8.35) 6

7.95

7.62

Length of maxillary toothrow

6.7 ± 0.3 (6.3–7.1) 10

6.21 ± 0.17 (5.93–6.31) 6

6.31

6.13

Breadth of incisive foramina

3.44 ± 0.2 (3.1–4) 10

3.32 ± 0.16 (3.19–3.62) 6

2.88

3.13

Breadth of rostrum

7.4 ± 0.4 (7–8.1) 10

6.17 ± 0.54 (5.57–6.9) 6

5.89

6.25

Breadth of palatal bridge

4.4 ± 0.3 (4–4.7) 10

3.75 ± 0.36 (3.26–4.31) 6

3.72

4.03

Breadth of first upper molar

1.8 ± 0.1 (1.6–1.9) 10

1.67 ± 0.05 (1.61–1.75) 6

1.77

1.81

Breadth of nasals

4.9 ± 0.4 (4.4–5.6) 10

4.62 ± 0.24 (4.35–4.96) 6

4.23

4.12

Least interorbital breadth

6.2 ± 0.5 (5.6–6.9) 10

6.26 ± 0.21 (6.03–6.61) 6

5.81

6.1

Zygomatic breadth

22 ± 0.8 (20.6–22.9) 10

20.94 ± 0.68 (20.2–21.72) 6

19.68

20.48

Braincase breadth

16.8 ± 0.5 (16.3–17.4) 10

16.49 ± 0.42 (15.78–16.92) 6

16.41

17.02

Breadth of zygomatic plate

3.7 ± 0.2 (2.9–4) 10

3.58 ± 0.18 (3.23–3.73) 6

2.92

2.94

Depth of incisor

2.3 ± 0.1 (2.1–2.4) 10

2.22 ± 0.18 (1.95–2.41) 6

1.95

2.03

Height of braincase

11.5 ± 0.4 (10.4–12) 9

11.05 ± 0.26 (10.73–11.37) 6

11.38

11.3

Weight

159 ± 3.61 (155–162) 3

144 ± 12.86 (126–155) 5

-

129

a

External measurements of the holotype taken from Allen (1901). Cranial measurements of the holotype taken by the senior author; b measurement taken from dry skin



53

FIGURE 9. Lateral view of braincase and an enlargement of the auditory bullae region of three species of Rhipidomys: (A–B) Rhipidomys gardneri (MUSM 14472), (C–D) R. leucodactylus (MUSM 12987), and (E–F) R. ochrogaster (MUSM 35095). Scale is 10 mm. pop= paraoccipital process, pb=processus brevis of incus, oa=orbicular apophysis.


PACHECO & PERALTA

FIGURE 10. Mountain forest habitat near Yanacocha town where Rhipidomys ochrogaster was collected in our study. Photograph by M. Peralta.



55

Upper and lower molars of R. ochrogaster are very similar to those of R. leucodactylus and R. gardneri (Fig. 8), except that anteromedian flexus and flexid are deeper in R. leucodactylus and R. gardneri than in R. ochrogaster (Tribe, 1996). This author also mentioned that R. ochrogaster have upper molars with well developed paralophule/ mesostyle, but these structures are also present in the other two species. R. ochrogaster and R. leucodactylus lack enteroloph and ectolophids on upper and lower molars respectively, but these structures are frequently seen in R. gardneri (Fig. 8). Natural history. The habitat of the collection site is characterized by a forest of slender trees that reach approximately 15 m in height, a dense understory, and a forest floor covered to a depth of at least 3 cm with soil litter (Fig. 10). The trap line was set on a steep slope, dominated by woody plants and herbaceous patches of fern, grass and sedge. Plant species present in this habitat are Ilex sp. (Aquifoliaceae), Viburnum sp. (Adoxaceae), Miconia sp. (Melastomataceae), Myrsine sp. (Myrsinaceae), Myrcia sp. (Myrtaceae), Clusia sp. (Clusiaceae), Cecropia sp. (Urticaceae), and Clethra sp. (Clethraceae). This habitat corresponds to the Yungas (Tovar Narváez et al., 2010) or the humid eastern montane forests (Young and Leon, 1999). The capture took place at the beginning of the dry season (May–November). During the fieldwork the weather was sunny and clear with sporadic rains. Rhipidomys ochrogaster was collected in the same trap line with the species Oligoryzomys destructor (Tschudi), Neacomys spinosus (Thomas), Lenoxus apicalis J. A. Allen, Oxymycterus juliacae J. A. Allen, and Akodon baliolus Osgood. The recognition of the last two taxa follow Pacheco et al. (2011) who consider them as valid species.

Discussion Rhipidomys ochrogaster is geographically isolated from other species of the "fulviventer" section (sensu Tribe, 1996), whose distribution is restricted to moderately high altitudes of Colombia and Venezuela (Tribe, 1996). The extensive gap of more than 2000 km between these regions complicates the affinities and membership of R. ochrogaster to the "fulviventer" section. However, this scenario appears to be a sampling artifact, since some specimens allied to the "fulviventer" section have recently been collected in Amazonas Department in northern Peru (Jiménez, C., pers. comm.), filling partially the mentioned gap. In addition, and besides the disjunct distribution of species member of the "fulviventer" section, no phylogenetic analysis has proved yet that this group is monophyletic. A phylogenetic analysis of thomasomyine rodents based on morphological characters, which included R. ochrogaster and six other congeners, found support for the monophyly of Rhipidomys but no internal grouping except for a weak relationship between R. fulviventer Thomas and R. caucensis J. A. Allen (Pacheco, 2003). More recently, de la Sancha et al. (2011) developed a phylogenetic analyses of Rhipidomys based on an 801 bp fragment of the cytochrome b gene, and recovered a monophyletic Rhipidomys with strong support; and an arrangement congruent with the three sections within Rhipidomys proposed by Tribe (1996); unfortunately, R. ochrogaster was not included in the analysis. The affinities of R. ochrogaster to other congeners and an interpretation of its restricted distribution must await a comprehensive phylogenetic analysis of the genus Rhipidomys. Rhipidomys ochrogaster may be sympatric with R. gardneri, which is distributed in the same region of southeastern Peru, from 200 to 1828 m elevation. However, no record of sympatry between these two species is currently known, and it needs to be determined whether the ranges of these two species overlap at all. The distributional range of R. leucodactylus, from 120 to 1580 m, is similar to R. gardneri; however, R. leucodactylus ranges in Peru only from Tumbes to Ayacucho Departments (Appendix 1), in north and central Peru, and it is apparently replaced by R. gardneri further south. Therefore, it is unlikely that R. leucodactylus and R. ochrogaster are to be found sympatric. The current distribution of Rhipidomys ochrogaster indicates the species is restricted to the Yungas of southeastern Peru; however, a Maxent distribution map (not included) predicts that the species might range from central Peru (around the Río Apurímac valley) to all along the Bolivian Yungas (H. Quintana and V. Pacheco, pers. comm.). Although additional studies will be needed to support this hypothesis, this distributional pattern is similar to that of other sigmodontine rodents such as Thomasomys oreas Anthony, T. daphne Thomas, Oxymycterus hiska Hinojosa, Anderson, and Patton, and Lenoxus apicalis, as presented by Pacheco et al. (2007). Rhipidomys is considered to be an arboreal species (Reig, 1984; Emmons, 1999; Alho et al. 2002; Ribeiro et al., 2004; Bonvicino et al., 2008; Asfora and Pontes, 2009), and it is associated with lower montane forest and tran-


PACHECO & PERALTA

sitional forest (Reig, 1984; Eisenberg and Redford, 1999; Ribeiro et al., 2004). Some morphological features are associated with arboreality, such as well-developed pads on the feet, broad feet and a long tail; however, our specimen of R. ochrogaster was trapped on the ground, indicating that this species may also forage at ground level. Other arboreal species of sigmodontine rodents have also been collected on the ground (Pinheiro et al., 2004). The analysis of the stomach contents of Rhipidomys ochrogaster found that the species forage on seeds and ants, suggesting an omnivorous diet; in concordance with the omnivorous-insectivorous diet described for Rhipidomys latimanus (Tomes) in Eisenberg and Redford (1999), and Ribeiro et al. (2004) who considered the genus as omnivorous. However, additional samples are needed to fully determine the diet category of the species. The unilocular-hemiglandular stomach (sensu Carleton, 1973) found in Rhipidomys ochrogaster is similar to the stomach morphology found in other congeners. Pacheco (2003) reported that the unilocular-hemiglandular stomach was also exhibited by R. fulviventer, R. latimanus, R. leucodactylus, R. gardneri, and R. macconnelli de Winton. This type of stomach is probably characteristic for the genus. The rarity of Rhipidomys ochrogaster in collections is probably due to the scarce sampling effort in the Yungas of southeastern Peru. Since the visits by H.H. Keays in 1900 (Allen, 1901) and G. Ockenden in 1901 (Tribe, 1996), it appears that only J.L. Patton and colleagues had explored the region before our expedition (Cadle and Patton, 1988, Patton and Smith, 1992). Rhipidomys ochrogaster is listed as Data Deficient by the IUCN (Zeballos et al., 2008) but we propose the species qualifies for Endangered status, with the criteria B1ab (i, iii, iv), because the area occupied by the species is apparently less than 150 km2, severely fragmented and targeted as a potential center for development. The increasing deforestation of Yungas habitats in Puno Department poses a serious threat to R. ochrogaster and other cloud forest species, but the Yungas of southeastern Peru are not included in any National protected area, despite the high diversity and uniqueness of the region and the fact that it is a target for development. The Inambari hydroelectric dam and the Interoceanic highway are two major man-made infrastructures that may threaten the unique diversity of the region.

Acknowledgements We wish to thank our friends and companions in the field: Carlos Jimenez, Oscar Centty, Edith Briceño, and Jorge Veliz; Carlos for his work as assistant coordinator, and Edith for the botanical support. Paulina Jenkins kindly provided information on the two specimens of Rhipidomys ochrogaster housed at the BM(NH). Our gratitude is extended to Edith Arias for the analysis of stomach contents, and to Frank Azorsa for the identification of ant samples. Christopher Tribe and one anonymous reviewer provided substantial comments that helped to improve the manuscript. The Consejo Superior de Investigaciones UNMSM awarded grant 101001181: “Endemic mammals of the Yungas of Puno department, Peru” to the senior author.

References Alho, C.J.R., Schneider, A. C. & Vasconcellos, L.A. (2002) Degree of threat to the biological diversity in the Ilha Grande State Park (RJ) and guidelines for conservation. Brazilian Journal of Biology, 62(3): 375–385. Allen, J.A. (1901) On a further collection of mammals from southeastern Peru, collected by Mr. H.H. Keays, with descriptions of new species. Bulletin of the American Museum of Natural History, 14: 41–46. Asfora, P.H. & Pontes, A.R.M. (2009) The small mammals of the highly impacted North-eastern Atlantic Forest of Brazil, Pernambuco Endemism Center. Biota Neotropical, 9(1), 31–35. Bonvicino, C.R., Oliveira, J.A. & D’Andrea, P.S. (2008) Guia dos Roedores do Brasil. Com chaves para gêneros baseadas em caracteres externos. Rio de Janeiro: Centro Pan-Americano de Febre Aftosa - OPAS/OMS, 120 pp.: il. (Série de Manuais Técnicos, 11. Cabrera A. (1961) Catálogo de los mamíferos de América del Sur. Parte II. Revista del Museo Argentino de Ciencias Naturales “Bernardino Rivadavia” e Instituto Nacional de Investigación de las Ciencias Naturales. Ciencias Zoológicas, 4(2): 309– 732. Cadle, J.E. & Patton, J.L. (1988) Distribution patterns of some amphibians, reptiles, and mammals of the eastern Andean slope of southern Peru. In: W.R. Heyer & P.E. Vanzolini (Eds), Proceedings of a workshop on Neotropical distribution patterns. Rio de Janeiro: Academia Brasileira de Ciências, pp. 225–244. Carleton, M.D. (1973) A survey of gross stomach morphology in New World Cricetinae (Rodentia, Muroidea), with comments



57

on functional interpretations. Miscellaneous Publications, Museum of Zoology, 146, 1–43. Carleton, M.D. & Musser, G.G. (1989) Systematic studies of oryzomyine rodents (Muridae: Sigmodontinae): Synopsis of Microryzomys. Bulletin of the American Museum of Natural History, 191, 1–83. de la Sancha N., D'Elía, G., Tribe, C.J., Perez, P.E., Valdez, L. & Pine, R.H. (2011) Rhipidomys (Rodentia, Cricetidae) from Paraguay: noteworthy new records and identity of the Paraguayan species. Mammalia 75, 269–276. Eisenberg, J. & Redford, K. (1999) Mammals of the Neotropics. The Central Neotropics. Vol. 3. Ecuador, Peru, Bolivia, Brasil, 609 pp. Emmons, L. H. (1999) Mamíferos de los bosques húmedos de América tropical. F.A.N. (Ed.). Santa Cruz de la Sierra, Bolivia, 299 pp. Ministerio de Agricultura (2004) Decreto Supremo No. 034-2004- AG. El Peruano. pp. 276853–276855. Musser, G.G. & Carleton, M.D. (1993) Family Muridae. In: Wilson, D.E. & Reeder, D.M. (Eds.). Mammals species of the world: a taxonomic and geographic reference, 2nd. ed. Smithsonian Institution Press, Washington DC. Musser, G.G. & Carleton, M.D. (2005) Superfamily Muroidea. 894–1531, In: Wilson, D.E. & Reeder, D.M. (Eds.), Mammal Species of the World, 3rd ed. Vol. 2. Johns Hopkins University Press, Baltimore, MD., pp. 2141. Pacheco, V., Patterson B.D., Patton, J.L., Emmons L.H., Solari, S. & Ascorra C.F. (1993) List of mammal species known to occur in Manu Biosphere Reserve, Peru. Publicaciones del Museo de Historia Natural, UNMSM, Zoología (A), 44, 1–12. Pacheco, V. (2003) Phylogenetic analyses of the Thomasomyini (Muroidea: Sigmodontinae) based on morphological data. Ph D. dissertation. The City University New York, New York, USA. Pacheco, V., Quintana, H.L. Hernandez, P.A., Paniagua, L., Vargas J. & Young, B.E. (2007) Mamíferos. In: Young B.E. (Ed.). Distribución de las especies endémicas en la vertiente oriental de los Andes en Perú y Bolivia. NatureServe, Arlington, Virginia, pp. 40–45. Pacheco, V., Cadenillas R., Salas E., Tello C. & Zeballos H. (2009) Diversidad y endemismo de los mamíferos del Perú. Revista Peruana de Biología, 16(1), 5–032. Pacheco V., Márquez G., Salas E. & Centty O. (2011) Diversidad de mamíferos en la cuenca media del río Tambopata, Puno, Perú. Revista Peruana de Biología 18(2), 231 – 244. Patton, J. L. & Smith M.F. (1992) Evolution and systematics of akodontine rodents (Muridae: Sigmodontinae) of Peru, with emphasis on the genus Akodon. Memorias del Museo de Historia Natural, U.N.M.S.M, 21, 83–103. Patton J.L., da Silva M.N.F. & Malcolm J.R. (2000) Mammals of the Rio Juruá and the evolutionary and ecological diversification of Amazonia. Bulletin of the American Museum of Natural History, 244, 1–306 Pinheiro P., Hartmann, P. & Geise, L. (2004) New record of Rhagomys rufescens (Thomas 1886) (Rodentia: Muridae). Zootaxa, 431, 1–11. Reig, O.A. (1984) Distribuição geográfica e história evolutiva dos roedores muroideos sulamericanos (Cricetidae: Sigmodontinae). Revista Brasileira de Genética, 2, 333–365. Ribeiro, M.F.S., Rocha, P.L.B., Mendes, L.A.F., Perry, S.F. & Oliviera, E.S. (2004) Physiological effects of short term water deprivation in the South American sigmodontine rice rat Oligoryzomys nigripes and water rat Nectomys squamipes within a phylogenetic context. Canadian Journal of Zoology, 82, 1326–1335. Smithe, F.B. (1975) Naturalist's color guide. New York: American Museum of Natural History. Solari, S., Pacheco, V., Luna, L., Velazco, P.M. & Patterson, B.D. (2006) Mammals of the Manu Biosphere Reserve. In: Patterson, B.D., Stotz, D.F. & Solari, S. (Eds.), Mammals and Birds of the Manu Biosphere Reserve, Peru. Fieldiana, Zoology, 10, pp. 13–23. Tovar Narváez, A., Tovar Ingar, C., Saito Díaz, J., Soto Hurtado, A., Regal Gastelumendi, F., Cruz Burga, Z., Veliz Rosas, C., Vásquez Ruesta, P. & Rivera Campos, G. (2010) Yungas Peruanas – Bosques montanos de la vertiente oriental de los Andes del Perú: Una perspectiva ecorregional de conservación. CDC – UNALM, Lima, Peru, 150 pp. Tribe, C.J. (1996) The Neotropical rodent genus Rhipidomys (Cricetidae: Sigmodontinae) – a taxonomic revision. Unpublished Ph.D. dissertation, University College London. Voss, R.S. (1988) Ichthyomyine rodents (Muroidea): Patterns of morphological evolution in a small adaptive radiation. Bulletin of the American Museum of Natural History, 188, 259–493. Voss, R.S. (1993) A revision of the Brazilian muroid rodent genus Delomys with remarks on "Thomasomyine'' characters. American Museum Novitates, 3073, 1–44. Young, K.R. & León B. (1999) Perú´s humid eastern montane forests: an overview of their physical settings, biological diversity, human use and settlement, and conservation needs. DIVA Technical Report, 5, 1–97. Zeballos, H., Vivar, E., Weksler, M. & Patterson, B.D. (2008) Rhipidomys ochrogaster. IUCN 2010. IUCN Red List of Threatened Species. Version 2010. 2. Available from http://www.iucnredlist.org (accessed 27 July 2010).


PACHECO & PERALTA

APPENDIX 1. Specimens examined. The following list summarizes the information of taxa and specimens examined for this article. Specimens measured are indicated with asterisks. Rhipidomys gardneri—Peru: Madre de Dios, Tambopata, Las Piedras, Puerto Maldonado (MUSM 7152*—7153*); Tambopata, Las Piedras, Albergue a 12 km Este de Puerto Maldonado (MUSM 2506*); Tambopata, Las Piedras, Reserva de Cusco Amazónico, 15km NE de Puerto Maldonado (MUSM 8303—8306); Río Manu, Parque Nacional del Manu, Puesto de Vigilancia de Pakitza (MUSM 9344—9345). Cuzco, Quispicanchi, Camanti, Marcapata, Hacienda Cadena (MUSM 2507*); Paucartambo, Kosñipata, Paucartambo, Estación Biológica Bosque Nublado, San Pedro (MUSM 11997*); La Convención, Echarate, La Convención, Camisea, San Martín (MUSM 14469—14471, 14472*, 14473—14475). Rhipidomys leucodactylus— Ecuador, Napo, San José Nuevo (AMNH 68189*). Peru: Ayacucho, Huanhuachayo (AMNH 241643). Piura, Sullana, Marcavelica, El Angolo (MUSM 1286*). Pasco, Oxapampa, Pozuzo, Delfín (MUSM 12152*, MUSM 12713*, MUSM 12986*—12987*). Junín, Utcuyacu (AMNH 63858). Loreto, Maynas, San Juan Bautista, 1 km E km 25.3 de la carretera Iquitos-Nauta, Fundo San Martín (MUSM 30425*); Boca río Curaray (AMNH 71906*—71907*). Tumbes, Pampas de Hospital, Parque Nacional Cerros de Amotape, Cochas, [uncatalogued MUSM-RCO 841*]. Rhipidomys ochrogaster—Peru: Puno, Santo Domingo, Inca Mines (= Santo Domingo) (AMNH 16481*--16482); Sandia, Alto Inambari, Yanacocha (MUSM 35095*).



59