Voucher Material (total = 6): Jenaro Herrera (AMNH 276706), Nuevo San Juan (AMNH 273038, 273059; MUSM 11024, 15290, 15291).

Other Interfluvial Records: Jenaro Herrera (Fleck and Harder, 1995), San Pedro (Valqui, 1999).

Identification: Caluromys lanatus has never been revised, and several subspecies are currently recognized as valid (Gardner, 2008). To assess the taxonomic significance of the current trinomial classification, we analyzed sequence data from the mitochondrial gene encoding cytochrome b, and we examined specimens from throughout western Amazonia, where no fewer than three subspecies—C. l. nattereri (Matschie, 1917), C. l. ochropus (Wagner, 1842), and C. l. ornatus (Tschudi, 1845)—are said to occur. Unfortunately, representative sequence data are currently unavailable from the nominotypical subspecies (in Paraguay), C. l. cicur (Bangs, 1898; in transAndean Colombia), and C. l. vitalinus (Miranda-Ribeiro, 1936; in southeastern Brazil), so the materials at hand are insufficient for a comprehensive revision of the species.

The results of our molecular analysis (figs. 7, 8; table 1; appendix 3) provide robust support for the reciprocal monophyly of all three currently recognized species of Caluromys, but only weak support for monophyly of the subgenus Mallodelphys (= C. derbianus + C. lanatus). Remarkably, we recovered only shallow phylogeographic structure in C. derbianus and none at all in C. lanatus, both of which exhibit negligible sequence variation among samples collected at widely scattered localities. By contrast, our samples of C. philander reveal noteworthy genetic heterogeneity in that species, which clearly merits revisionary attention.²

The absence of phylogeographic structure among genetic samples of Caluromys lanatus from widely scattered populations in western Amazonia and central Brazil suggests that the subspecies currently recognized from these regions are based on nothing more than coat-color variants. In fact, western Amazonian specimens of C. lanatus exhibit considerable variation in pelage coloration. Whereas some individuals are almost uniformly reddish brown dorsally (e.g., MVZ 157611), others have distinctly grayish fore- and hind limbs (e.g., MVZ 190249, USNM 546177), and others are almost uniformly brownish (e.g., MUSM 15291). The base of the tail is reddish in some specimens, but grayish in others. A patch of grayish hair between the shoulders (said to be diagnostic of C. derbianus; Cáceres and Carmignotto, 2006) is present in several specimens among those we examined (e.g., MVZ 157608, 190249, 190250). Most specimens have uniformly gray-based fur on the throat, chest, and upper abdomen as well as on the ventral surfaces of the forelimbs, but others have irregular patches of self-cream fur on the throat and in the forelimb axillae; the lower abdomen and groin are seemingly always self-cream, but parous females have orange-stained fur surrounding the pouch. Although some of this pelage variation might be geographic, there are substantial differences in coloration among sympatrically collected skins, and intermediates exist among most coat-color phenotypes. Cranial size and shape differences among the specimens we examined do not suggest anything more than individual and ontogenetic variation. In effect, we concur with Fonseca and Astúa's (2015) suggestion that only a single phenotypically distinguishable taxon, which they called C. lanatus ochropus, occurs in the Amazon and Cerrado.

The holotype of Caluromys lanatus, a juvenile specimen from Paraguay collected in the late 18th century, was originally preserved in alcohol; its pelage is discolored, and few useful measurements can be obtained from the incompletely erupted dentition (Voss et al., 2009a). However, recently collected Paraguayan adults (AMNH 66780, UMMZ 134007) have very pale, almost entirely grayish fur that lacks the rich reddish-brown pigments seen in most western Amazonian specimens; additionally, the naked caudal skin (distal to the furry basal portion) is completely unpigmented, whereas most western Amazonian specimens have dark-spotted tails. Metrical comparisons (table 2) indicate that these Paraguayan specimens are smaller than western Amazonian material in several dimensions (e.g., condylobasal length), but both of them are young adults, and measurements of age-invariant structures (e.g., LM, WM3) are within the observed range of western Amazonian variation. With just two Paraguayan examples at hand, and in the absence of any relevant molecular data, it is hard to assess the taxonomic significance of such comparisons, or to properly evaluate Fonseca and Astúa's (2015) suggestion that C. l. ochropus is really distinct from the nominotypical form.

Ethnobiology: The Matses name for the woolly opossum is abuk cheka, which literally means “up opossum,” in reference to its arboreal habits. The Matses do not eat or use the woolly opossum for any purpose, nor do they seem to have any particular beliefs about it.

Matses Natural History: The woolly opossum is reddish. It has a stripe going down its rostrum. Its tail is partly bare. Its ears are large.

The woolly opossum is arboreal, but it sometimes descends to the ground to forage. It descends by climbing down a vine. It is found in all rainforest habitat types. It makes its nest in tree cavities that it lines with leaves. It also makes nests in the branches of dicot trees, or in leaf litter that collects in palm crowns.

The woolly opossum is nocturnal. After foraging for dicot tree fruits and insects during the night, it returns to its nest at dawn. It climbs quickly on vines.

The woolly opossum is solitary. It gives birth to many young that it keeps in its pouch. The young suck milk inside the pouch. When the young get big, they leave their mother, one by one.

Margays and snakes prey on the western woolly opossum.

When the wooly opossum sees people, it opens its mouth and hisses, wanting to bite them.

The woolly opossum eats crickets/katydids (the Matses do not distinguish lexically between crickets and katydids), cockroaches that live up in trees, armored millipedes (Barydesmus) that are on the ground, dicot tree fruits including those of the bata (Pseudolmedia spp. and/or ?Maquira spp.; Fam. Moraceae), kuëte bata trees (?Pseudolmedia; Fam. Moraceae), kuëte mëdiad (unidentified tree with starchy fruits that the Matses cook before eating), and vine fruits. It also eats the mesocarp of ripe isan palms (Oenocarpus bataua) that have not fallen to the ground yet.

Remarks: Four of our six vouchers were shot from trees at night by Matses hunters, three in secondary forest (abandoned swiddens) and one in hilltop primary forest; heights were not recorded, but the specimen shot in primary forest was said to be “very high up in a big tree” (English translation from Matses field notes). One specimen was captured by hand in the late afternoon by several boys, who shook it from the crown of a small guaba tree (Fabaceae: Inga edulis) on the outskirts of the village. The specimen from Jenaro Herrera was trapped on a liana 18 m above the ground in disturbed primary forest.

Other Specimens Examined (total = 29): Bolivia—Pando, Isla Gargantua (MSB 56998). Brazil—Acre, Nova Vida on right bank Rio Juruá (MVZ 190250, 190251), opposite Igarapé Porongaba on left bank Rio Juruá (MVZ 190249), Sena Madureira (USNM 546177); Amazonas, opposite Altamira on left bank Rio Juruá (MVZ 190248), Rosarinho (AMNH 92760), Sacado on right bank Rio Juruá (MVZ 190247). Ecuador—Napo, San Jose Abajo (AMNH 68282); Orellana, Parque Nacional Yasuní (ROM 104570). Paraguay—Caazapá, Caazapá (MNCN-M2630 [holotype]; Canendiyu, 13.3 km N Curuguaty (UMMZ 134007); Guairá, Villarrica (AMNH 66780). Peru—Amazonas, La Poza on Río Santiago (MVZ 157608, 157611, 157612); Huánuco, Tingo Maria (MVZ 140041); Loreto, Boca Río Curaray (AMNH 71979, 71983, 71984), Estación Biológica Allpahuayo (TTU 99025, 101044); Madre de Dios, Reserva Cusco Amazónico (KU 144058, MVZ 168852); Pasco, Nevati Mission (AMNH 230001), San Juan (USNM 364160); Ucayali, Boca Río Urubamba (AMNH 75911), Lagarto (AMNH 78951). Venezuela—Amazonas, Esmeralda (AMNH 76970).

Voucher Material (total = 1): Quebrada Vainilla (LSUMZ 28421).

Other Interfluvial Records: None.

Identification: As currently understood, Glironia venusta ranges across much of Amazonia but remains known from only a few specimens (Barkley, 2008; Ardente et al., 2013). Among the handful that we were able to examine for this report were the holotypes of two nominal species—aequatorialis Anthony, 1926, and criniger Anthony, 1926—that have long been regarded as junior synonyms of venusta. Whereas the type localities of aequatorialis (Boca Río Lagartococha, on the Peruvian-Ecuadorian frontier) and criniger (Boca Río Curaray, in Loreto department) are both north of the Amazon,³ the type locality of venusta (Pozuzo, in Pasco department) is south of the Amazon. In addition to these and our single specimen from Quebrada Vanilla, we examined a fourth specimen, from eastern Ecuador.

Although Barkley (2008) said that LSUMZ 28421 was an adult, it is really a subadult with a still incompletely erupted P3 on each side. The immaturity of this specimen plausibly explains why it is smaller than the others we measured (table 3) in several cranial dimensions subject to postweaning growth (e.g., CBL, ZB, PB), but not in age-invariant molar dimensions (LM, M1–3, WM3). We did not note any conspicuous craniodental differences among these specimens, but the tip of the tail is abruptly white for about 2 cm in LSUMZ 28421, a marking that is also present on BMNH 12.1.15.7 (Thomas, 1912b), but not in any of the specimens we examined from the left (“north”) bank of the Amazon, which all have grayish or brownish tail tips.

Preliminary genetic data suggest that distinct haplotypes of Glironia venusta occupy opposite banks of the Amazon. Patton et al. (1996) obtained 1140 bp of cytochrome b from a single specimen (INPA 2570) collected on the upper Rio Urucu, a right-bank tributary of the upper Amazon (Solimões) in western Brazil. This sequence differs by about 6% (uncorrected) from two large fragments (both >500 bp) that we obtained from the left-bank holotypes of criniger (AMNH 71394) and aequatorialis (AMNH 71395); by contrast, the latter two sequences differ from one another by just a single base-pair substitution (an uncorrected distance of just 0.2%). Although it might make some sense in the light of these results to recognize two subspecies, G. v. criniger (including aequatorialis) on the north bank and G. v. venusta on the south bank, we are reluctant to formalize any taxonomic conclusions on such an inadequate basis. Nevertheless, the trivial genetic distance between the two left-bank specimens does tend to support the conclusions of previous taxonomists that the allegedly diagnostic morphological differences between criniger and aequatorialis reported by Anthony (1926) are nothing more than intraspecific variation.

Ethnobiology: The Matses have no definite knowledge of this species, and therefore do not have a name for it or any particular beliefs about it.

Matses Natural History: The Matses have no knowledge of the appearance or behavior of this species.

Remarks: According to Barkley (2008), LSU 28421 was captured at night in a mist net set for bats in primary forest. Local habitats at the capture site, which is not subject to seasonal inundation, were described by Robbins et al. (1991).

Other Specimens Examined (total = 3): Ecuador—Pastaza, Montalvo (FMNH 41440). Peru—Loreto, Boca Río Curaray (AMNH 71394). Peru (Loreto) or Ecuador (Orellana)—Boca Río Lagartococha (AMNH 71395).

Voucher Material (total = 2): Nuevo San Juan (AMNH 276725; MUSM 11031).

Other Interfluvial Records: None.

Identification: This tiny marsupial, long unknown to science and only recently recognized as the sole survivor of an ancient lineage, is still represented by fewer than two dozen museum specimens despite its wide Amazonian distribution and apparent lack of habitat specificity (Hershkovitz, 1992; Voss et al., 2001; Jansa and Voss, 2005; Astúa, 2006; Hice and Velazco, 2012; Catzeflis, 2018). Our two specimens consist of the damaged skull of a juvenile that retains the diagnostically reduced milk dentition (Voss et al., 2001: figs. 17, 18), and the fluid-preserved carcass and extracted skull of an adult female. Díaz (2014) erroneously reported that the latter specimen (AMNH 276725) was collected at Jenaro Herrera.

Hyladelphys kalinowskii is one of three superficially similar species in our region, all of which are very small (<30 g) opossums with long tails, black masks, and reddish-brown dorsal fur. Despite these resemblances, they are only distantly related to one another, and specimens in hand are easily identified by numerous integumental and craniodental differences (figs. 9, 10; table 4). Traits unique to H. kalinowskii among this trio of tiny didelphids include the posterior extent of its blackish facial markings, possession of just four mammae, an indistinctly banded tail (due to paler skin over the vertebral articulations), lack of a premaxillary rostral process, absence of a posterior accessory cusp on C1, a third upper premolar (P3) that is conspicuously smaller than P2, and an exceptionally short upper molar row (LM ≤4.6 mm). Hyladelphys kalinowskii additionally differs from Marmosa lepida by its exclusively self-white ventral fur and lack of postorbital processes. Hyladelphys kalinowskii additionally differs from Gracilinanus emiliae by having a much shorter tail, and by lacking a gular gland, palatine fenestrae, and secondary foramina ovales.

Both of our specimens conform in all qualitative traits to the emended description of Hyladelphys kalinowskii provided by Jansa and Voss (2005). Measurements of our adult female specimen are a bit smaller than those of the adult female holotype from Cuzco department and the adult female paratype from Junín (table 5), but without any adequate population sample to assess individual (nongeographic) variation in this species, such differences are hard to interpret. Although Jansa and Voss (2005) discussed the possibility that multiple species of Hyladelphys might be represented among the specimens currently referred to H. kalinowskii, ours are so similar to Hershkovitz's (1992) original material that this identification would seem to be beyond dispute.

Ethnobiology: The Matses do not distinguish this species from other pouchless, long-tailed, black-masked species of small opossums, all of which are known indiscriminately as chekampi (see the account for Marmosa, below). Therefore, they have no particular beliefs about it, nor is it of any cultural importance.

Matses Natural History: The Matses have no definite knowledge of this species.

Remarks: Our juvenile specimen was captured during the day by a Matses boy when it ran out of some leaf litter.⁴ Another was captured by a Matses hunter in unknown circumstances.

Other Specimens Examined (total = 5): French Guiana—Paracou (AMNH 267003, 267338, 267339). Peru—Cuzco, Hacienda Cadena (FMNH 89991 [holotype]; Junín, Chanchamayo (FMNH 65754).

Members of the genus Marmosa are among the most abundant species of nocturnal-arboreal insectivorous-frugivorous small mammals throughout lowland Amazonia. By contrast with previous classifications (e.g., in Gardner, 2005, 2008), the species referred to Marmosa are now allocated among five subgenera (Voss et al., 2014). In the Yavarí-Ucayali interfluve, the nominotypical subgenus is represented by M. macrotarsus, the subgenus Micoureus by M. constantiae and M. rutteri, and the subgenus Stegomarmosa by M. lepida. In addition to these four species, M. (Eomarmosa. rubra might also be expected to occur in our region (appendix 2).

The Matses do not lexically distinguish the various species of small, long-tailed, black-masked didelphids that occur in their tribal territory (Hyladelphys kalinowskii. Marmosa spp., Gracilinanus emiliae. Marmosops spp.), but this seems as good a place as any to summarize relevant ethnological information.

Ethnobiology: The opossums that the Matses call chekampi (“little opossums”) are superficially similar (small, black-masked, long-tailed, and pouchless), although some of our interviewees acknowledged that chekampi come in different sizes and in different shades of gray and brown. Some Matses use mapiokosëmpi (the name of the common opossum with a diminutive suffix) as an alternative name.

Mouse opossums come into Matses houses and eat their food. Sometimes they make nests in Matses houses, but they seldom stay long. The Matses say that mouse opossums usually enter their houses during heavy rains.

Matses Natural History: Mouse opossums have long tails and large ears. Some are gray, while others are reddish, and others are dark-colored. They are similar to four-eyed opossums, but much smaller.

Mouse opossums are arboreal and terrestrial. They are abundant in the forest and sometimes come into Matses swiddens. They make nests of dry leaves in thick vegetation up in trees, in cavities in tree branches, or in leaf litter that accumulates in the crowns of palms. In Matses swiddens they make nests among plantain plants using dead plantain leaves. In houses they make nests in containers where clothes are kept.

Mouse opossums are strictly nocturnal. They give birth to many young. They always carry their young with them. Otherwise they are solitary.

Mouse opossums are eaten by margays and snakes. They make a high-pitched hiss when they are threatened.

Mouse opossums eat all sorts of things. They eat crickets/katydids; large cockroaches; and small birds, eggs, and hatchlings that they find at night in nests. They also eat dicot tree fruits, including those of the mannan tsipuis tree (Inga spp.; Mimosoideae) and the mesocarp of swamp-palm (Mauritia flexuosa) fruits.

Voucher Material (total = 20): Nuevo San Juan (268214, 272816, 272870, 273062, 273063, 273140, 273178, 273188, 276726; MUSM 11028, 11029, 13283, 15293–15297, 23798), Orosa (AMNH 73855, 73856).

Other Interfluvial Records: Jenaro Herrera (as Marmosa murina; Fleck and Harder, 1999), San Pedro (as M. murina; Valqui, 1999).

Identification: The species currently recognized as Marmosa macrotarsus and M. waterhousei (Tomes, 1860), both of which occur in eastern Peru, were long treated as subspecies or synonyms of M. murina (Linnaeus, 1758), but Rossi (2005) summarized evidence that these three taxa are phenotypically distinct, and Gutiérrez et al. (2010) subsequently recovered them as reciprocally monophyletic mtDNA clades with average sequence divergence values of about 9% in all pairwise comparisons. According to Rossi's (2005) still incompletely published revision, the species belonging to this complex are allopatric. Whereas Marmosa murina occurs in eastern Amazonia and the Atlantic Forest, M. waterhousei is in northwestern Amazonia, and M. macrotarsus is in southwestern Amazonia. In eastern Peru, M. waterhousei is said to occur north of the Amazon-Marañón (in Amazonas and northern Loreto departments), whereas M. macrotarsus is distributed south of the river (e.g., in Cuzco, Huánuco, Junín, Madre de Dios, Pasco, Ucayali, and southern Loreto).⁵

To assess the diagnostic morphological traits attributed to Marmosa macrotarsus and M. waterhousei, we examined Peruvian specimens of the nominotypical subgenus from both sides of the Amazon. According to Rossi (2005) and Díaz (2015), M. macrotarsus can be distinguished from M. waterhousei by the absence of a gular gland (said to be present in M. waterhousei) and by the form of the base of the upper canine (said to be less curved than in M. waterhousei), but neither character seems to be taxonomically useful. None of the Peruvian specimens we examined from south of the Amazon (nominally M. macrotarsus) showed any trace of a gular gland, but of the seven adult males we examined from north of the Amazon (nominally M. waterhousei), only three showed any sign of glandular activity on the chest or throat. We observed considerable variability in canine curvature in specimens from both sides of the Amazon, and we were unable to consistently distinguish north-bank from south-bank material by this criterion. Instead, the only trenchant difference we observed between Peruvian specimens of Marmosa (Marmosa) from opposite banks of the Amazon is size: specimens from the north side of the river are substantially larger, on average, than specimens from the south bank (table 6), and although our samples are inadequate to show the full range of intraspecific morphometric variation that might be expected, the lack of overlap in same-sex comparisons for several dimensions (e.g., HBL, LT, CBL, LIB, PL, PB, and MTR for males) seems noteworthy. Altogether, the morphometric evidence for genetically distinct populations on opposite banks of the upper Amazon seems reasonably compelling (Rossi, 2005), although additional phenotypic criteria for species recognition would be very welcome.

Ethnobiology: The Matses do not distinguish this species from other pouchless, long-tailed, black-masked species of small opossums (all known as chekampi; see the account for Marmosa, above) and therefore have no particular beliefs about it.

Matses Natural History: The Matses have no definite knowledge of this species.

Remarks: Of the 13 specimens from Nuevo San Juan for which we have habitat information, 12 were trapped, shot, or captured by hand in secondary vegetation; only one specimen is known to have been taken in primary forest. Three additional specimens trapped by children in unrecorded circumstances at Nuevo San Juan were probably also taken in secondary growth (children seldom enter primary forest except in the company of adults). Most specimens were trapped, shot, or caught by hand at night, but two were taken in the daytime by men weeding gardens or clearing trash piles, who presumably disturbed sleeping individuals in their daytime refugia. Most specimens were taken at or near ground level (on the ground or on fallen logs), but one was caught by hand at a height of 2 m above the ground in a small tree, and another was shot from a tree at an unrecorded height above the ground.

Other Specimens Examined (total = 8): Peru—Huánuco, Pozuzo (FMNH 24754, 24755); Pasco, Nevati (AMNH 254508); San Martín, Moyobamba (FMNH 19352, 19357); Ucayali, Balta (MVZ 136368–136370).

Specimens of Marmosa waterhousei Examined (total = 14): Peru—Amazonas, Huampami on Río Cenepa (MVZ 154754, 154761), Kayamas on Río Cenepa (MVZ 153286), La Poza on Río Santiago (MVZ 157632), 12 mi SSW Nazareth (MVZ 139955); Loreto, 25 km S Iquitos (LACM 96112; TTU 98654, 98716, 98717, 98934, 100922, 101153), 61.5 km S Iquitos (TTU 124790, 124796).

Voucher Material (total = 21): Nuevo San Juan (AMNH 268219, 268220, 272667, 272832, 273052, 273079, 273113, 273118; MUSM 11060, 11062, 11064, 13294–13296, 15317, 15310–15313), Jenaro Herrera (MUSM 23806), Santa Cecilia (FMNH 87117).

Other Interfluvial Records: None that can be confidently associated with this species (see below).

Identification: Two species of the subgenus Micoureus (“woolly mouse opossums,” formerly recognized as a distinct genus; Gardner and Creighton, 2008) are sympatric at Nuevo San Juan, and the same two species occur sympatrically at several other localities south of the Amazon in eastern Peru and western Brazil. Specimens of both species collected sympatrically in western Brazil were identified as Micoureus demerarae (Thomas, 1905) and Mic. regina (Thomas, 1898) by Patton et al. (2000), who, however, cautioned that different names might apply to their material. Among other possibilities, they (Patton et al., 2000: 72) suggested that “constantiae might best be considered a junior synonym of demerarae,” but the former name has priority. Therefore, Marmosa constantiae is the binomen used in this report for the demeraraelike woolly mouse opossum of southwestern Amazonia.⁶ For reasons explained in the next account, we use the name Mar. rutteri Thomas, 1924, for the woolly mouse opossum that Patton et al. (2000) called Mic. regina.

Marmosa constantiae and M. rutteri are large mouse opossums with broadly overlapping external and craniodental measurements (table 7). Where they occur sympatrically in eastern Peru and western Brazil, both have drab, somewhat woolly dorsal fur; long, all-dark tails with rhomboidal scales arranged in spiral series; almost completely ossified palates (lacking palatine fenestrae, and with short-narrow maxillopalatine openings); well-developed postorbital processes (in most mature adults); and small auditory bullae. As in other members of the subgenus Micoureus, gular glands are absent in both sexes, and large adult males have well-developed medial and lateral carpal tubercles (Voss et al., 2014). Although there are several pelage and craniodental differences that allow confident identification of specimens in hand (table 8), M. constantiae and M. rutteri are otherwise so alike that unvouchered records (e.g., of “Micoureus cinereus” at Jenaro Herrera; Fleck and Harder, 1995) cannot be confidently associated with one or the other species.

As described and illustrated by Patton et al. (2000: 65–77), the ventral pelage is mostly gray-based in Marmosa constantiae, but it is mostly self-yellowish in M. rutteri. In the former species, self-yellowish fur is restricted to the throat, groin, and (sometimes) to a narrow streak along the midline of the chest and abdomen; the fur on the insides of the fore- and hind limbs is always gray-based. In M. rutteri, by contrast, self-yellowish fur extends over the chin, throat, chest, and groin, and along the insides of the fore- and hind limbs; a median abdominal band of self-yellowish fur is always present, and it is at least as broad as or broader than the flanking lateral zones of gray-based fur.

Another taxonomically informative external trait is the length of the dorsal fur, which appears to be consistently longer in Marmosa constantiae than in sympatric M. rutteri. In our material from the Yavarí-Ucayali interfluve, the middorsal fur of adult M. constantiae is about 12–15 mm long, whereas the dorsal fur of M. rutteri from our region and adjacent lowlands is only about 8–11 mm. Although these observed ranges almost overlap, the modal fur lengths for these species (about 13 mm and 9 mm, respectively) correspond to visibly and tactilely distinct phenotypes.

A third diagnostically useful external trait (illustrated by Patton et al., 2000: fig. 53) concerns the tail, which has a more conspicuously furry base in Marmosa constantiae than in M. rutteri. Not only does the caudal fur (identical in color and texture to the fur of the rump) extend distally further along the tail in constantiae, but the hairs that comprise the caudal fur are longer in this species than in rutteri. Measured from skins in our material of constantiae, this fluffy-furred tail base extends distally for about 28 mm (on average) with hairs that are about 10 mm long, whereas the homologous average values for rutteri are about 16 mm and 5 mm.

Although Patton et al. (2000: table 13) found statistically significant differences for most craniodental measurements in same-sex comparisons of Marmosa constantiae (“Micoureus demerarae”) and M. rutteri (“Micoureus regina”), these species have broadly overlapping morphometric distributions, such that no measurement (nor any ratio of measurements that we computed) is diagnostic. Unmagnified comparisons of cranial morphology are likewise uninformative, but two dental features (visible under low magnification) are useful for species recognition. In constantiae the postprotocrista (the posteriormost of the two enamel crests that pass labially from the apex of the protocone; Voss and Jansa, 2009: fig. 20) is short, because it terminates at or near the base of the metacone, whereas the postprotocrista is much longer in rutteri, extending labially well beyond the base of the metacone on the posterior surface of the tooth. On the lower molars, a posterior cingulid (a small enamel shelf at the posterolabial base of the hypoconid) is consistently absent in constantiae, whereas a small but persistent posterior cingulid is usually present, at least on m2, in rutteri.

Unlike the Peruvian material described above, the holotype of Marmosa constantiae (BMNH 3.7.7.157, from Chapada dos Guimarães, Mato Grosso, Brazil; Thomas, 1904a), other specimens from Mato Grosso, and many specimens from eastern Bolivia (Beni and Santa Cruz departments, see below) have tails that are one-third to one-half whitish distally. Nevertheless, all the material that we refer to M. constantiae is morphologically similar in other respects, and sequenced specimens from Mato Grosso and eastern Peru belong to the same well-defined mtDNA haplogroup (the “SW-S” clade of Patton and Costa, 2003: fig. 9). Therefore, we interpret geographic differences in caudal markings as intraspecific variation.

Unfortunately, the epithet constantiae has often been misapplied (e.g., by Tate, 1933; Flores et al., 2007; Gardner and Creighton, 2008; Gutiérrez et al., 2010; de la Sancha et al., 2012; Voss et al., 2014) to another widespread species of woolly mouse opossum, for which the oldest available name is Marmosa rapposa Thomas, 1899. As recognized in this report, M. rapposa (with type locality at Huadquiña, near Cusco, Peru; Ceballos-Bendezu, 1981) ranges at middle elevations in Andean cloud forests from Junín department, Peru, throughout the Bolivian Yungas to northeastern Argentina (budini Thomas, 1920, from Jujuy is a junior synonym); at lower elevations, this species occurs in Cerrado woodlands from eastern Bolivia across central Brazil to eastern Paraguay. Although specimens of M. rapposa resemble the holotype of M. constantiae in having white-tipped tails and yellowish underparts, they differ in other respects (table 8), most notably in having well-developed palatine fenestrae (fig. 11B), long postprotocristae, well-developed posterior cingulids on m1–m3, and highly divergent mtDNA sequences (Voss et al., in prep.).

Ethnobiology: The Matses do not distinguish this species from other pouchless, long-tailed, black-masked species of small opossums (all known as chekampi; see the account for Marmosa, above) and therefore have no particular beliefs about it.

Matses Natural History: The Matses have no definite knowledge of this species.

Remarks: Of the 19 specimens of Marmosa constantiae collected at Nuevo San Juan, 11 were trapped or shot in primary upland forest; 2 were trapped in primary floodplain forest; 4 were trapped, shot, or caught by hand in secondary growth (abandoned swiddens); and 2 were caught by hand in Matses houses. Of 10 specimens accompanied by information about capture height, 2 were trapped on the ground, 1 was trapped on a fallen log, and 7 were trapped or shot in trees or on lianas at estimated heights ranging from 0.4 to 4 m above the ground. Our single specimen from Jenaro Herrera was trapped at a height of 17 m in a tree at the edge of a clearing.

Other Specimens Examined (total = 62): Bolivia—Beni, Río Iténez frente Costa Marques (AMNH 209158–209162), Río Mamoré lado este frente Cascajal (AMNH 210397); Pando, La Cruz (MSB 57001). Brazil—Amazonas, Altamira on right bank of Rio Juruá (MVZ 190309–190312), Coleção Vira-Volta on left bank of Rio Juruá (MVZ 190316–190318), Penedo on right bank of Rio Juruá (MVZ 190301–190303), Seringal Condor on right bank of Rio Juruá (MVZ 190305–190307), Ilhazinha on left bank of Rio Juruá (MVZ 190313, 190314); Mato Grosso, Chapada dos Guimarães (AMNH 384, BMNH 3.7.7.157 [holotype], OMNH 37209), Fazenda Noirumbá (MVZ 197415), Reserva Ecológica Cristalino (MVZ 197407–197414). Peru—Cuzco, 15.9 km SW Pilcopata (FMNH 174443), 2 km SW Tangoshiari (MUSM 13412, 13414, 13415), Pagoreni (MUSM 14139, 14141); Madre de Dios, Altamira (FMNH 98029), 13.5 km NW Atalaya (MUSM 19830), Hacienda Amazonía (FMNH 138852), Itahuanía (FMNH 84254, 84255), Reserva Cuzco Amazónico (KU 144092, 144094, 144096–144099, 144103); Pasco, Pozuzo (BMNH 12.1.15.18); San Martín, Área de Conservación Municipal Mishquiyacu Rumiyacu-Almendra (FMNH 203510), Moyobamba (FMNH 19348), Yurac Yacu (BMNH 27.1.1.174–27.1.1.176); Ucayali, Balta (MVZ 136374–136377).

Voucher Material (total = 6): Nuevo San Juan (AMNH 273164; MUSM 11055, 11063, 15315, 15316), Orosa (AMNH 74087).

Other Interfluvial Records: None that can be confidently associated with this species.

Identification: As explained in the preceding account, this is the species that Patton et al. (2000) called Micoureus regina, but their usage is taxonomically indefensible. As those authors noted, regina is based on a specimen collected in the Magdalena Valley of Colombia, so it seems biogeographically improbable that the name could validly apply to an Amazonian taxon. Additionally, the type of regina (BMNH 98.5.15.4, an old adult male) does not closely resemble any Amazonian woolly mouse opossum that we have seen. Among other traits that do not fit the present species, the dorsal fur of BMNH 98.5.15.4 is redder (near Prout's Brown according to Tate [1933: 83], who saw the type when it was less than 40 years old), the upper molars are smaller (LM = 8.1 mm), and the postprotocristae are short. Like Tate (1933), we restrict the application of Marmosa regina to the holotype, and we reject the application of this name to any of the Amazonian nominal taxa treated as subjective junior synonyms or subspecies by Gardner (1993, 2005).

According to Patton et al. (2000), the next-oldest name for this species is Marmosa germana Thomas, 1904, which was treated as a synonym of M. regina by Gardner (1993), but as a valid subspecies of M. regina by Gardner (2005) and by Gardner and Creighton (2008). However, the type of germana (BMNH 80.5.6.77, an adult female from eastern Ecuador; Thomas, 1904b) more closely resembles the species that Patton and his colleagues called M. demerarae than the species that they called M. regina. Among other traits that do not match the present species, BMNH 80.5.6.77 has completely gray-based ventral fur; a long (30 mm), fluffy-furred tail base; and short postprotocristae. Unpublished DNA sequence data (Voss et al., in preparation) suggest that M. germana is a distinct species, restricted to the left (north) bank of the upper Amazon.

Thomas (1924) described Marmosa rutteri as a valid species, but Tate (1933: 81) considered rutteri to be a subspecies of M. germana, and most subsequent authors have considered rutteri to be a junior synonym of germana (now usually ranked as a subspecies of regina; e.g., by Gardner, 2005; Gardner and Creighton, 2008). In fact, M. rutteri is a distinct species with the diagnostic traits listed in table 8, and analyses of DNA sequence data (Voss et al., in prep.) suggest that it is more closely related to M. rapposa and other extralimital species than to M. germana. The holotype (BMNH 24.2.22.67, an adult male) is from Tushemo, on the right bank of the upper Ucayali. As recognized in this report, the species is widely distributed in western Amazonia, but different haplogroups are found north and south of the Amazon; the specimens listed below, which include all of those with measurement data summarized in table 7, are from south of the Amazon.

Ethnobiology: The Matses do not distinguish this species from other pouchless, long-tailed, black-masked species of small opossums (all known as chekampi; see the account for Marmosa, above) and therefore have no particular beliefs about it.

Matses Natural History: The Matses have no definite knowledge of this species.

Remarks: Of our four specimens from Nuevo San Juan, one was captured by hand inside a Matses house, one was trapped on top of a fallen log in primary upland forest, one was shot from a tree (at an unrecorded height) in primary upland forest, and one was taken by hand inside a hollow tree at the edge of the Río Gálvez.

Other Specimens Examined (total = 84): Brazil—Acre, Igarapé Porongaba (MVZ 190332), Nova Vida (MVZ 190333); Amazonas, Boa Esperança (MVZ 190330, 190331), Igarapé Nova Empresa (MVZ 190321, 190323–190325), opposite Altamira (MVZ 190328, 190329), Penedo (MVZ 190319, 190320), Seringal Condor (MVZ 190326). Colombia—Caquetá, Tres Troncos (FMNH 70964–70966). Ecuador—Napo, San José de Payamino (FMNH 124613). Peru—Amazonas, La Poza (MVZ 157629), mouth of Río Cenepa (AMNH 98712), vicinity of Huampami (MVZ 153278, 154749, 154751, 154755, 154758, 154762, 154764, 154766, 157628, 157630, 157631), vicinity of Kayamas (MVZ 153281); Ayacucho, Santa Rosa on Río Santa Rosa (LSUMZ 15674); Cuzco, Quincemil (FMNH 75100); Loreto, Boca Río Curaray (AMNH 71951, 71956, 71958, 71964, 71966, 71968, 71975, 72010), Otorongo (MUSM 33443), Pampa Chica (FMNH 87118), San Antonio (AMNH 98655), San Jerónimo (BMNH 28.5.2.231–28.5.2.241; FMNH 46110, 46111), Sarayacu (AMNH 76302, 76303), El Triunfo (TTU 124799), Yurimaguas (FMNH 19635); Madre de Dios, Blanquillo (MUSM 8399), Reserva Cuzco Amazónico (KU 144091, 144093, 144095, 144100, 144102, 144107, 144110, 144111; MUSM 6083, 6086–6088, 6090–6092, 6100, 6101); Pasco, San Pablo (AMNH 230019, 230021); Ucayali, Tushemo (BMNH 24.2.22.67 [holotype]), Yarinacocha (FMNH 55467). Peru (Loreto) or Ecuador (Orellana)—Boca Río Lagartococha (AMNH 72008, 72009)

Voucher Material (total = 1): Nuevo San Juan (AMNH 273186).

Other Interfluvial Records: None.

Identification: Our unique voucher, an adult male preserved in fluid with the skull extracted and cleaned, exhibits all the diagnostic traits of the nominotypical form (Marmosa lepida lepida), including small size, reddish dorsal fur, a midventral zone of self-whitish fur flanked by lateral zones of gray-based fur, a long tail with dense fringes of silvery hairs flanking the distal prehensile surface, a very long rostral process of the premaxillae, large postorbital processes, and small posterior accessory cusps on C1 and c1. Additionally, our specimen lacks any trace of a gular gland, and palatine fenestrae are absent. Morphometric comparisons with other specimens from western Amazonia, including the holotype, suggest striking uniformity in most craniodental dimensions, especially of the molar dentition (table 9). A cytochrome b sequence obtained from AMNH 273186 was analyzed by Gutiérrez et al. (2010), who reported that it formed a strongly supported clade with sequences from other specimens that also exhibit the distinctive morphological traits of M. lepida.

The only local taxa with which Marmosa lepida could possibly be confused in the field are Hyladelphys kalinowskii and Gracilinanus emiliae, both of which are also small, reddish, black-masked, and long-tailed. Comparisons with Hyladelphys kalinowskii have already been described, but it remains to compare M. lepida with G. emiliae. The chief external character distinguishing these taxa (table 4) is the ventral coloration (self-whitish and gray-based in M. lepida versus entirely self-white in G. emiliae). Crania of these species (figs. 9, 10) are most easily distinguished by the occurrence of postorbital processes (present in M. lepida, absent in G. emiliae), palatine fenestrae (absent in M. lepida, present in G. emiliae), and secondary foramina ovales (usually absent in M. lepida, consistently present in G. emiliae); additionally, the upper molar series is much longer in M. lepida (LM = 5.6–6.1 mm) than in G. emiliae (LM = 4.8–5.1 mm).

Two subspecies of Marmosa lepida were recognized by Tate (1933), of which our material is unambiguously referable to the nominotypical form. Marmosa l. lepida occurs on both sides of the Amazon, from the base of the Andes to the Atlantic coast, with remarkably little evidence of geographic variation in either morphology or cytochrome b sequences (Tate, 1933; Gutiérrez et al., 2010; Guimarães et al., 2018). By contrast, M. l. grandis Tate, 1931, is known only from Buenavista in Santa Cruz province, Bolivia. We examined the young adult female holotype of grandis (BMNH 26.12.4.94), which differs from other specimens of M. lepida in ventral pelage color (entirely gray-based buffy) and in lacking posterior accessory cusps on C1 and c1. Given the morphological uniformity exhibited by Amazonian specimens of M. lepida, we are not persuaded that grandis is really conspecific. Although Tate (1933: 205) listed two specimens that he identified as M. l. lepida from Buenavista (paradoxically implying sympatry of two subspecies), the specimens in question are offspring of the holotype of grandis (as documented by notes on the skin tags). We have not examined any Bolivian specimen definitely referable to typical M. lepida as diagnosed by the suite of traits listed in the first paragraph of this account.

Ethnobiology: The Matses do not distinguish this species from other pouchless, long-tailed, black-masked species of small opossums (all known as chekampi; see the account for Marmosa, above) and therefore have no particular beliefs about it.

Matses Natural History: The Matses have no definite knowledge of this species.

Remarks: Our single specimen was shot at night by a Matses hunter as it perched in the subcanopy of tall primary forest.

Other Specimens Examined (total = 13): Brazil—Amazonas, Igarapé Grande on Rio Juruá (FMNH 140824). Colombia—Amazonas, 20 km downstream from La Chorrera on Río Igaraparaná (MNHN 1982-653). French Guiana—Les Nouragues (MNHN 1998-306). Guyana—Potaro-Siparuni, 42 km WNW Siparuni (ROM 107034). Ecuador—Pastaza, Río Tigre (AMNH 182937); Sucumbíos, Lago Agrio (KU 135118). Peru—Amazonas, Huampami on Río Cenepa (MVZ 154750, 154752, 155245); Loreto, Río Pisqui (AMNH 98656), Santa Cruz on Río Huallaga below Yurimaguas (BMNH 69.3.31.4 [holotype]); Ucayali, Lagarto (AMNH 78001). Venezuela—Amazonas, El Platanal (EBD, uncataloged).

Three species of Monodelphis are known from the Yavarí-Ucayali interfluve, and no others seem likely to occur in the region. Two of the five subgenera recognized by Pavan and Voss (2016) are represented locally, Mygalodelphys with two species and Pyrodelphys with one. All three species are readily identified by external, craniodental, and morphometric traits (table 10). Because the Matses do not distinguish different species of Monodelphis, we summarize ethnobiological observations and Matses natural history under this generic heading.

Ethnobiology: The Matses name for short-tailed opossums is yama, a term that is not linguistically analyzable. Curiously, yama is also used to refer to the turnip-tailed gecko (Gekkonidae: Thecadactylus solimoensis) and to an unidentified arboreal rat that was formerly used for black magic.⁷ The Matses do not consider yama to be a type of cheka.

The Matses do not eat short-tailed opossums and have no other use for them. Not all Matses are familiar with them.

Matses Natural History: Short-tailed opossums are similar to mouse opossums, but they have very short tails. They are terrestrial, diurnal, solitary, and make nests in cavities among the roots of bottle palms (Iriartea deltoidea) and in hollows at the bases of trees. They are eaten by jaguarundi cats.

Voucher Material (total = 7): Jenaro Herrera (AMNH 276698, 276704, 276709; MUSM 15991, 23808–23810).

Other Interfluvial Records: None.

Identification: Our voucher material was examined by Solari (2007) and served as the basis for his description of the species, which remains known only from Jenaro Herrera. Among other congeners, Monodelphis handleyi most closely resembles M. ronaldi (described by Solari, 2004), formerly known only from the holotype collected at Pakitza (11°56′S, 71°17′W) in Madre de Dios department, about 800 km southeast of Jenaro Herrera. Solari (2007) compared M. handleyi with M. ronaldi, which, among other differences, is substantially larger, but recently collected new material of M. ronaldi should allow a more complete differential diagnosis of these apparently closely related species (D. Ruelas, personal commun.). With the exception of a single juvenile female (AMNH 276704), all specimens of M. handleyi collected to date are male.

Several aspects of Solari's (2007) original description merit comment. The ventral pelage was described as entirely self-cream from chin to anus with a somewhat paler streak along the midline, but in one topotypical skin (MUSM 23810) most of the ventral fur is distinctly brownish and the hairs have pale-gray bases except on the center of the chest and upper abdomen, which is marked by a broad self-cream blaze. Whereas the central hair of each caudal-scale triplet was described as “thicker (petiolate) than the lateral hairs” (Solari, 2007: 324), the scale triplets we examined on several specimens had subequal lateral and central hairs. Both of the adult male skulls at hand (AMNH 276709, MUSM 23809) have distinct sagittal crests, which are also present but less distinct on all three sub-adults (AMNH 276698; MUSM 23808, 23810). The upper and lower canines lack accessory cusps in both sexes.

Monodelphis handleyi is much the larger of the two local members of the subgenus Mygalodelphys. Although only two fully adult specimens of M. peruviana from the Yavarí-Ucayali interfluve are available for comparisons, measurements of other material referable to M. peruviana suggest that these species have nonoverlapping craniodental measurements (tables 11, 12). Additionally, several qualitative differences inferred from our small samples merit evaluation when more specimens of both species become available: (1) The lacrimal foramina are concealed from lateral view inside the anterior margin of the orbit in M. handleyi, whereas these foramina are laterally exposed just anterior to the orbit in M. peruviana. (2) Most subadult and adult specimens of M. handleyi have a sagittal crest, but the left and right temporalis scars are separated and a sagittal crest is absent in all examined specimens of M. peruviana. (3) The upper and lower canines lack accessory cusps in both sexes of M. handleyi, whereas both upper and lower canines have small but distinct accessory cusps in female specimens of M. peruviana.

Ethnobiology: The Matses do not distinguish this species from other short-tailed opossums (all known as yama; see the account for Monodelphis, above) and therefore have no particular beliefs about it.

Matses Natural History: The Matses have no definite knowledge of this species.

Remarks: Of the six specimens for which we have capture information, five were taken in pitfalls along a single trapline flanked by swampy primary forest. The sixth specimen was taken in a pitfall trapline sited in a small patch of white-sand forest.

Voucher Material (total = 4): Nuevo San Juan (AMNH 272695, 272781; MUSM 13297, 15318).

Other Interfluvial Records: None.

Identification: This taxon was long treated as a synonym or subspecies of Monodelphis adusta (Thomas, 1897)—for example, by Cabrera (1958), Gardner (2005), and Pine and Handley (2008)—but it was recognized as a valid species by Solari (2007), largely on the basis of mtDNA sequence analyses. In fact, phylogenetic analyses of multiple genes provide compelling evidence that M. adusta and M. peruviana are not sister taxa: whereas M. peruviana belongs to a robustly supported clade that includes M. handleyi. M. osgoodi Doutt, 1938, and M. saci Pavan et al., 2017, the sister taxon of M. adusta (as currently recognized; see below) is M. reigi Lew and Pérez-Hernández, 2004 (Pavan et al., 2014, 2017). In light of those results, it seems reasonable to treat M. adusta and M. peruviana as distinct species, but the absence of unambiguously diagnostic morphological characters (Solari, 2004) is a relevant problem. In effect, M. adusta and M. peruviana are names applied to haplogroups vouchered by morphologically similar material collected north and south of the Amazon, respectively. Although the application of peruviana to material from the Yavarí-Ucayali interfluve seems adequately justified by proximity to the type locality (Moyobamba, only about 200 km west of our region; Osgood, 1913), the application of adusta to the haplogroup that occurs on the north bank of the Amazon is more problematic.⁸ Nevertheless, we maintain current conventions for applying these epithets to small specimens of Mygalodelphys from western Amazonia.

To assess the problem of species diagnosis, we compared our voucher material and other referred specimens of Monodelphis peruviana with cis-Andean specimens of M. “adusta,” including several from the Reserva Nacional Allpahuayo-Mishana (RNAP), a north-bank locality adjacent to our region (fig. 1). Sequence data from one of our vouchers (AMNH 272695 [= RSV 2086]) and from specimens collected at RNAP (TTU 101019 [= TK 73496], TTU 101164 [= TK 73868]) document their membership in the haplogroups currently associated with these binomina (e.g., by Solari, 2007; Pavan et al., 2014). Contra Solari (2004: 150), we found no external morphological differences between these taxa: whereas he reported that adusta has shorter dorsal fur than peruviana, our measurements indicate that both species have dorsal fur that ranges from 3 to 4 mm long, and we were unable to distinguish the “ill-defined blackish area on the posterior dorsum and the rump” that was said to be present in adusta, but absent in peruviana. Likewise, we observed no consistent qualitative craniodental differences between specimens collected on opposite banks of the Amazon. Sample sizes are too small for confident statistical comparisons of craniodental measurements, but broadly overlapping ranges for most dimensions (table 12) suggest that morphometric differences, if any, are unlikely to provide a secure basis for species separation. Although we are currently unable to provide diagnoses of the taxa currently associated with these names, it seems prudent to maintain current usage pending a comprehensive revision of the subgenus Mygalodelphys.

Ethnobiology: The Matses do not distinguish this species from other short-tailed opossums (all known as yama; see the account for Monodelphis, above) and therefore have no particular beliefs about it.

Matses Natural History: The Matses have no definite knowledge of this species.

Remarks: One of our four vouchers (AMNH 272695) was trapped on the ground in well-drained (hill-slope) primary forest; another (AMNH 272781) was trapped under a log in what was said to be primary forest by the Matses boy who caught it; a third (MUSM 13297) was trapped by another Matses child at an undisclosed location near the village; and the fourth (MUSM 15318) was caught by hand by a Matses man at the edge of his swidden.

Other Specimens Examined (total = 9): Peru—Cuzco, Camisea (USNM 582782), 2 km SW Tangoshiari (USNM 588019); Huánuco, Hacienda Exito (FMNH 23772), Hacienda San Antonio (FMNH 23774, 23775; USNM 259433); Madre de Dios, Reserva Cuzco Amazónico (MUSM 7157); San Martín, Moyobamba (FMNH 19361, 19362 [holotype]).

Specimens of Monodelphis adusta Examined (total = 9): Ecuador—Pastaza, Mera (AMNH 67274); Sucumbíos, km 61 on Lago Agrio-Quito highway (USNM 534286); Tungurahua, Palmira (AMNH 67275); Zamora-Chinchipe, Zamora (AMNH 47189). Peru—Loreto, 25 km S Iquitos (TTU 98686, 98923, 101019, 101164).

Voucher Material (total = 7): Jenaro Herrera (AMNH 276721, MUSM 23807), Nuevo San Juan (AMNH 268221; MUSM 11065, 13298), Río Aucayo (FMNH 58955), San Pedro (MUSM 22333).

Other Interfluvial Records: Quebrada Curacinha (photograph, fig. 12).

Identification: Our voucher material exhibits all of the distinctive phenotypic traits of the monotypic subgenus Pyrodelphys, including disruptively colored dorsal pelage (a reddish head and rump separated by grizzled-grayish middorsal fur), contact or fusion of the thenar and first interdigital pads of the hind foot, a rounded but distinct frontal process of the jugal, parietal-mastoid contact, exposure of the basioccipital in the rear of the orbit, a broadly rounded rostral tympanic process of the petrosal, broad lower-molar anterior cingulids, and distinct entoconids on m1–m3 (Pavan and Voss, 2016).

A cytochrome b sequence from MUSM 13298 was analyzed by Pavan et al. (2014), who recovered it as part of a strongly supported haplogroup with other specimens that exhibit the diagnostic morphological traits of Monodelphis emiliae. Although substantial phylogeographic structure was reported among the sequences that Pavan et al. (2014) identified as M. emiliae, morphological comparisons of our Peruvian specimens with the holotype and other material from eastern Amazonia have not revealed any consistent differences. At least for the time being, we recognize M. emiliae as a widespread and morphologically uniform species.

According to Pine and Handley (1984: 242), Monodelphis emiliae has “well-defined postorbital processes on the frontals,” but among the specimens we examined this trait is developed in just a few old adult males (e.g., BMNH 20.7.14.44, MUSM 20452). By contrast, females and other adult males (some of which have well-worn teeth; e.g., FMNH 58955) lack any trace of frontal processes. The same authors (op. cit.: 241) described the tail of M. emiliae as “furred throughout its length,” but only the base of the tail bears soft fur on the specimens we examined, the remainder of that organ being covered with the usual caudal bristles (albeit thicker, longer, and more densely crowded than in many other congeneric species). The measurements of M. emiliae reported by Patton et al. (2000: table 15) include several values that are smaller than any obtained by us (table 13) or by Pine and Handley (1984); although we have not examined all of Patton et al.'s material, at least one of their measured specimens (MVZ 190335) is a subadult. This species is one of the few didelphid species that is polymorphic for presence/absence of secondary foramina ovales, which are formed by stout medial bullar laminae in some specimens (e.g., AMNH 96810, MUSM 20452), but not in others (e.g., AMNH 268221, BMNH 20.7.14.44). The fugitive ventral coloration originally reported by Thomas (1912a) and subsequently noted by Pine and Handley (1984), Patton et al. (2000), and Pavan and Voss (2016: fig. 14) was also seen in our fresh material. The ventral fur of MUSM 2083 (an adult female), for example, was bright pink in life (on 27 May 1998), but it soon faded to a dull beige on the dried skin. Although sample sizes are small, it seems noteworthy that adult males and adult females in our Peruvian material have nonoverlapping measurements for many craniodental dimensions (table 13).

Ethnobiology: The Matses do not distinguish this species from other short-tailed opossums (all known as yama; see the account for Monodelphis, above) and therefore have no particular beliefs about it.

Matses Natural History: The Matses have no definite knowledge of this species.

Remarks: Of the five specimens we collected, one (AMNH 268221) was captured in a Matses house surrounded by secondary vegetation; another (MUSM 11065) was captured by hand in old secondary vegetation near a Matses swidden; another (MUSM 13298) was trapped on the ground in valley-bottom primary forest; a third (MUSM 23807) was trapped on the ground in primary forest growing on white sand; and the last (AMNH 276721) was taken in a pitfall in swampy primary forest. The specimen from San Pedro was trapped “on a hilltop in a primary terra firme forest” (Valqui, 2001). Based on these observations, it would seem that Monodelphis emiliae is a habitat generalist, although Patton et al.'s (2000) trapping results suggest that it might not occur in seasonally flooded forests. One of our specimens (MUSM 13298) was taken on the same date, along the same trapline, and in the same habitat (primary upland rainforest) as a specimen of M. peruviana (AMNH 272695). Consistent with the Matses' observation that short-tailed opossums are diurnal, MUSM 11065 was captured by hand at 07: 30 hrs, and MUSM 13298 was found at 17: 00 hrs in a trap that had previously been checked and found empty at dawn.

Although Monodelphis emiliae was reported from the “Iquitos area” by Patton et al. (2000), all the specimens known to have been collected in northeastern Peru are from the right (“south”) bank of the Amazon.⁹ Despite many decades of collecting near Iquitos—on the left (“north”) bank of the river—including the impressive faunal survey effort monographed by Hice and Velazco (2012), no specimens or sightings of M. emiliae have been reported from there. Nor has M. emiliae, which is very widely distributed along the right bank of the lower Amazon (Pine and Handley, 2008: map 41), been reported to occur at any other left-bank locality. Thus, it seems reasonable to conclude that the Amazon effectively limits the northward distribution of this species.

Other Specimens Examined (total = 12): Brazil—Amazonas, Igarapé Porongaba (MVZ 190335), Seringal Condor (MVZ 190334); Pará, Baião (AMNH 96810), Boim (AMNH 37491, BMNH 11.12.22.16 [holotype]), Vila Braga (BMNH 20.7.14.44). Peru—Cusco, Cashiriari (MUSM 14146, 14148), Pagoreni A (MUSM 36686), San Martín (MUSM 14149); Loreto, Cerros de Canchaguaya (MUSM 18028), Cerros de Contaya (MUSM 20452).

Voucher Material (total = 12): Nuevo San Juan (AMNH 268217, 268218, 272780, 273057; MUSM 11049, 11051, 11053, 11054, 13293, 15308, 15309), San Fernando (FMNH 89001).

Other Interfluvial Records: Jenaro Herrera (as Metachirus nudicaudatus; Pacheco, 1991; Pavlinov, 1994; Fleck and Harder, 1995), San Pedro (as M. nudicaudatus; Valqui, 1999, 2001).

Identification: Same-sex morphometric comparisons of our voucher material and other specimens from southwestern Amazonia with typical examples of Metachirus nudicaudatus (from northeastern Amazonia) document broad overlap in most external and craniodental dimensions (table 14). However, side-by-side visual comparisons of skulls reveal consistent differences in other aspects of craniodental morphology (table 15). Most conspicuously, the left and right scars that mark the dorsalmost origin of the temporalis muscle on each side of the skull are only weakly convergent posteriorly in most specimens of typical M. nudicaudatus, usually remaining widely separated over the parietals and interparietal (fig. 13A); if a sagittal crest is formed, usually in large adult male specimens, it is restricted to the midline of the interparietal. By contrast, the temporalis scars are more strongly convergent posteriorly in fully adult specimens from southwestern Amazonia, often uniting on the midline of the posterior brain-case to form a low sagittal crest over the interparietal and along the mid-parietal suture (fig. 13B). Additionally, the rostrum tends to be more robust and the zygomatic arches tend to be more rounded laterally in southwestern Amazonian material than in M. nudicaudatus, which typically has a longer, narrower rostrum and almost parallel-sided zygomatic arches. In ventral view, the maxillopalatine fenestrae are much shorter and narrower in southwestern Amazonian specimens than in typical M. nudicaudatus (fig. 14), and statistical comparisons of fenestral length and width confirm that these visually obvious differences are highly significant (p < 0.0001 by two-tailed Student's t tests). Although subtle differences in the dentition also appear to distinguish southwestern Amazonian specimens from typical M. nudicaudatus, the only dental character that we were able to score with confidence was size of the entoconid. This is a small cusp, subequal to or sometimes smaller than the paraconid on unworn m1–m3 in M. nudicaudatus (fig. 15A), but it is a large cusp, substantially bulkier than the paraconid and almost equal in size to the hypoconid in southwestern Amazonian specimens (fig. 15B).

To supplement our morphological comparisons, we sequenced the entire mitochondrial gene encoding cytochrome b from ethanol-preserved tissues of 29 specimens of Metachirus from 27 localities in Central and South America representing most of the nominal taxa currently treated as synonyms or subspecies of Metachirus nudicaudatus (fig. 16, appendix 4). The best-fitting topology obtained from maximum-likelihood analysis of this dataset (fig. 17) recovered a strongly supported basal dichotomy between sequences from northeastern Amazonia (French Guiana, Guyana, and Surinam) and those from elsewhere in the geographic range of the genus; uncorrected pairwise distances across this basal split are about 11.2%, on average, and are much larger than any others we obtained from these data (table 16). The mtDNA haplogroup that includes our sequenced vouchers (“SW Amazonia”) also includes sequences from southern Peru (Cuzco and Madre de Dios), eastern Bolivia (Cochabamba and La Paz), and western Brazil (Acre and Amazonas). Uncorrected pairwise distances within this group are uniformly small (about 1.4% on average), and visual inspection of the gene tree suggests that there is little (if any) internal phylogeographic structure. Sister to this clade is a group of sequences from the Atlantic Forest of southeastern Brazil, and these two haplogroups are sister to a clade that includes sequences from northwestern Amazonia and Central America.

Although a comprehensive revision of Metachirus is beyond the scope of this report, the results described in the preceding paragraphs clearly suggest that our southwestern Amazonian material is specifically distinct from M. nudicaudatus which, based on specimens we personally examined (see below), seems to be endemic to northeastern Amazonia (French Guiana, Guyana, Surinam, and the Brazilian state of Amapá). Of the many nominal taxa currently referred to the genus (table 17), the oldest that applies to our material is M. myosuros, which is based on a lectotype from southeastern Brazil. Specimens that we examined from southeastern Brazil (mostly from Rio de Janeiro and São Paulo at the AMNH, FMNH, and USNM; not listed below) closely resemble our material from southwestern Amazonia in all the traits previously discussed, which we regard as diagnostic of the species. Because types and other material of the remaining nominal taxa currently referred to Metachirus more closely resemble M. myosuros than M. nudicaudatus, we provisionally treat them as junior synonyms of the former species; all are based on seemingly minor coat-color differences, and none appears to exhibit morphometric traits that exceed the range of variation in our southwestern Amazonian series.

The absence of any sequence data from southeastern Amazonia (south of the Amazon and east of the Rio Madeira) is a major hiatus in our genetic sampling, so only phenotypic traits are available for identification of the populations that occur there. Specimens from localities on the right (east) bank of the Tocantins (e.g., Baião, Belém, Bragança, and Capim; at the AMNH and USNM) exhibit all the diagnostic craniodental traits of Metachirus myosuros, whereas those from the Tocantins-Xingu interfluve are morphologically distinctive (e.g., with large maxillopalatine fenestrae and small entoconids like M. nudicaudatus, but with strongly convergent temporalis scars like M. myosuros). Patton et al. (2000) reported a highly divergent cytochrome b sequence from this region, but we have not seen the voucher specimen (in Manaus), nor was this sequence included in a subsequent analyses of Metachirus mitochondrial DNA (Costa and Patton, 2006). Additional sequence data from this region would be very welcome, as would a careful assessment of associated morphological material.

Ethnobiology: The Matses do not distinguish this species from other “four-eyed” opossums (all known as cheka bëbëdi; see the account for Philander, below) and therefore have no particular beliefs about it.

Matses Natural History: The Matses have no definite knowledge of this species.

Remarks: Of the 11 specimens of Metachirus myosuros for which we have capture information, seven were trapped on the ground or near ground level (one was trapped on a fallen log), two were shot on the ground at night, one was caught by hand in the daytime as it ran out of a leaf nest on the ground, and another was caught by hand on the ground in the late afternoon as it was eating fallen guavas. One trapped specimen is also known to have been captured in the daytime (between 06: 30 and 15: 00 hours). Four of these specimens were taken in primary upland forest, whereas seven were in secondary growth (abandoned swiddens).

Other Specimens Examined from SW Amazonia (total = 39): Bolivia—Cochabamba, Cavernas de Repechón (MSB 70283); La Paz, Chulumani (BMNH 1.6.7.69–1.6.7.74 [type series of bolivianus]). Brazil—Acre, Nova Vida on right bank Rio Juruá (MVZ 190296, 190297), Sena Madureira (USNM 546190, 546191); Amazonas, Condor on left bank Rio Juruá (MVZ 190288), Penedo on right bank Rio Juruá (MVZ 190287), Vira Volta on left bank Rio Juruá (MVZ 190299); Rondônia, 8 km N Porto Velho (USNM 390034, 390035). Peru—Cuzco, Consuelo (FMNH 174437, 174441), Huajyumbe (FMNH 84248), Kiteni on Río Urubamba (MVZ 166505), Quincemil (FMNH 75097, 75098); Junín, Chanchamayo (FMNH 18206), San Ramón (FMNH 20795, 20798); Loreto, Santa Elena on Río Samiria (FMNH 87128); Madre de Dios, Hacienda Érica (MVZ 166501, 166504), “Inambari River” (BMNH 2.7.27.8 [holotype of infuscus]), Lago Sandoval (MVZ 157634), Puerto Maldonado (USNM 390040), 4 km S Puerto Maldonado (USNM 390036), 4 km W Puerto Maldonado (USNM 390039), 2.75 km E Shintuya (FMNH 169803); Pasco, Pozuzo (FMNH 24788), San Juan (USNM 364161); Ucayali, Balta (MVZ 136383, 136384), 59 km SW Pucallpa (USNM 499009).

Specimens of Metachirus nudicaudatus Examined (total = 33): Brazil—Amapá, Serra do Navio (USNM 393544–393553, 461456, 461457, 545544, 546189). French Guiana—Paracou (AMNH 266435, 266440, 266449, 266450, 266455, 267009, 267010; MNHN 1995.906, 1995.907), “River Arataye” (USNM 548440). Guyana—East Berbice-Corentyne, Amuku Creek (ROM 34252); Potaro-Siparuni, Iwokrama Forest (ROM 111938); Rupununi, 30 mi E Dadanawa (ROM 32459), Kwitaro River (ROM 33051). Surinam—Brokopondo, Brownsberg Nature Park (ROM 114155); Sipaliwini, Bakhuis Transect 9 (ROM 117030), Bakhuis Transect 11 (ROM 117525), Iconja Landing (ROM 120317), Tafelberg (ROM 119571).

Voucher Material: Centro Unión (FMNH 106721).

Other Interfluvial Records: Nuevo San Juan (this report), Quebrada Pobreza (Pitman et al., 2015), San Pedro (Valqui, 1999), Santa Rosa (this report).

Identification: With its boldly gray-and-black banded dorsal pelage, webbed hind feet, and semiaquatic habits, the water opossum cannot be confused with any other species of Neotropical mammal. The single specimen we examined from our region (FMNH 106721) is mounted in a crudely lifelike pose with the skull inside; stuffed with sawdust and with painted seeds for eyes, it was probably manufactured for the tourist trade. A penciled cardboard tag attached to the left hind foot indicates the locality (Río Aucayo), the collector (Pekka Soini), and the date (12 February 1972). An inked FMNH label tied to the right hind foot further resolves the locality as “Río Aucayo, Centro Unión.” We infer that the specimen was probably purchased from a native taxidermist by Pekka Soini in the course of his fieldwork at Centro Unión in the early 1970s (appendix 1), but there is no way to be certain about exactly where the animal was collected. Fortunately, there are other interfluvial records of Chironectes, one of which is documented by a photograph (fig. 18).

Although central Amazonia (including the Yavarí-Ucayali interfluve) falls outside the mapped geographic range of Chironectes minimus in most standard references (e.g., Stein and Patton, 2008), recent observations and gene-sequencing results (Ardente et al., 2013; Brandão et al., 2015; Oliveira et al., 2016; Voss and Jansa, 2018) suggest that the water opossum is continuously distributed from Central America to southeastern Brazil. Despite geographic variation in craniodental morphology reported by Damasceno and Astúa (2016) and Cerqueira and Weber (2017), this species seems to be genetically undifferentiated across most, if not all, of cis-Andean South America, and it shows minimal genetic divergence even across the Andes (Voss and Jansa, 2018). Therefore, the “subspecies” of C. minimus currently recognized by authors (Gardner, 2005; Stein and Patton, 2008) are unlikely to represent valid taxa.

Ethnobiology: Because the water opossum is unknown to most of the Matses, they have no particular beliefs about it.

Matses Natural History: The Matses have no definite knowledge of this species.

Remarks: One of the previously unpublished records from our region is based on an unambiguous sighting made by Matses hunters near Nuevo San Juan and reported by them to D.W.F. in 2006. The second is based on a specimen killed by a Matses hunter on 27 January 2017 near the village of Santa Rosa and photographed by D.W.F. at Estirón on the following day; in the absence of collecting permits, it was not preserved.

Voucher Material (total = 7): Jenaro Herrera (AMNH 276703; MUSM 23797), Nuevo San Juan (AMNH 268213, 272836; MUSM 11025, 11027, 13282).

Other Interfluvial Records: Actiamë (Vriesendorp et al., 2006a), Divisor (Vriesendorp et al., 2006b), Jenaro Herrera (Fleck and Harder, 1995), Nuevo San Juan (three unvouchered sightings), San Pedro (Valqui, 1999).

Identification: Didelphis marsupialis, the so-called common or black-eared opossum of Amazonia and Central America, is externally unmistakable (Husson, 1978; Emmons, 1997), and although similar in most qualitative aspects of craniodental morphology to species of Philander (Voss and Jansa, 2009), it is so much larger in all dimensions that skulls, and even isolated teeth, are easily identified. There is remarkably little genetic variation in this species throughout its geographic range (e.g., <1.6% mean sequence divergence at the cytochrome b locus among samples from Costa Rica to central Amazonia; Patton et al., 2000: fig. 39). Measurements of the few adult specimens from our region (table 18) overlap the range of variation reported by Patton et al. (2000) for material collected along the Rio Juruá in western Brazil, but our specimens are large by comparison with their tabulated sample means.

Although the name Didelphis marsupialis has been used consistently for many years to refer to the black-eared Amazonian opossum, Gurgel-Filho et al. (2015) resurrected the obsolete synonym D. karkinophaga Zimmermann, 1780, for this species and proposed that Linnaeus's (1758) epithet be used for the white-eared species that has long been known as D. albiventris Lund, 1841. Their justification for thus disrupting current usage was discussed by Dias et al. (2018), who thought that a specimen in Uppsala might be the lectotype designated by Thomas (1911). Feijó and Voss (2019) disagreed, established that the lectotype is almost certainly lost, and designated a neotype for D. marsupialis to conserve prevailing usage of this name for the black-eared opossum of Amazonia.

Ethnobiology: The common opossum is called mapiokos by the Matses, a term that is not analyzable and is not found in other Panoan languages. The Matses recognize its similarity to the opossums they call cheka, but they do not classify it as a type of cheka.

Common opossums are considered pests because they eat chickens at night. The Matses kill them when they find them near chicken coops. The Matses traditionally did not eat common opossums, but local nontribal Peruvians do, so a few Matses eat them now, but most do not because of their foul smell.

The Matses believe that when the common opossum vocalizes, it is a death omen. If it calls from the trees, a man will die, and if it calls from the ground a woman will die.

Matses Natural History: The common opossum has a naked tail. It has dark hairs sprinkled with white hairs, as if its coat were turning gray. It has a white snout and large ears. It has a strong, foul odor that can be smelled from far away. One can readily detect its scent where it has passed by hours ago.

The common opossum is terrestrial and arboreal, but forages mostly on the ground. It is found in upland forest, floodplain forest, and along streams. It occurs in both primary and secondary forest. It comes to Matses swiddens to eat plantains and papayas. It makes its nest in hollow trees and in the crowns of isan palm trees (Oenocarpus bataua).

The common opossum is nocturnal and solitary. The female carries its young in its pouch and suckles them inside the pouch. It gives birth to many young.

The common opossum is eaten by jaguars, ocelots, and margays. It vocalizes saying “chocod chocod.”

The common opossum eats spiny rats and birds, including tinamous, that it finds nesting on or near the ground at night. It enters coops to eat chickens, and it also eats pet guans, crickets/ katydids, cockroaches, and rotten meat.

Remarks: Of our seven specimens, two were trapped on the ground in a swamp-palm (Mauritia flexuosa) swamp, one was trapped on the ground in primary upland forest, one was trapped at a height of 9 m in a tree in primary upland forest, one was caught by hand 2 m above the ground on a sapling in primary upland forest, one was trapped by a Matses boy (probably on the ground in secondary forest), and one was shot by a mestizo hunter in unrecorded circumstances (near Jenaro Herrera). The “common” opossum was far from common at Nuevo San Juan in 1999, when Matses hunters recorded only three sightings in 409 hours of night hunting: once perched at an unrecorded height on a vine in old secondary growth, once on a fallen tree next to a stream (habitat unrecorded), and once on the ground in hilltop primary forest.

Two species of gray four-eyed opossums, Philander mcilhennyi and P. pebas, are definitely known to occur in the Yavarí-Ucayali interfluve, and a third species (P. canus) could be expected to occur there based on geographic range data (appendix 2). Specimens of both P. pebas and P. canus have long been misidentified as P. opossum (an eastern-Amazonian taxon; Voss et al., 2018), so the identity of animals previously reported as P. opossum from the Yavarí-Ucayali interfluve (Pavlinov, 1994; Fleck and Harder, 1995) is unclear. To facilitate the identification of specimens collected in the course of future fieldwork, we summarize qualitative traits and morphometric variation for all three species (tables 19–21).

The Matses do not consistently recognize more than a single kind of opossum with pale supra-ocular spots, so the name they apply to such animals, cheka bëbëdi, could refer either to gray four-eyed opossums or to the superficially similar brown four-eyed opossum (Metachirus myosuros). Because species of Philander are more commonly encountered than M. myosuros, we summarize relevant ethnographic information here.

Ethnobiology: Cheka bëbëdi, the most frequently used name for these species, literally means “spotted-forehead opossums,” but in some Matses villages they are called cheka dëwisak “long-snouted opossums.” They are sometimes also referred to as chekadapa “big opossum,” but this is not considered a real name; rather it is a descriptive phrase used to distinguish these large species from mouse opossums. Some Matses recognize that there is more than one local type of cheka bëbëdi, but they do not differentiate them linguistically.

The Matses do not eat these opossums, although children sometimes shoot them with arrows. One informant mentioned that a four-eyed opossum ate his pet dove.

Matses Natural History: Some four-eyed opossums are black, while others are gray, and others are reddish/yellowish. They have a very long tail, a long snout, spots on their foreheads, and long whiskers.

They are arboreal and terrestrial. They are common in all rainforest habitats but seem to be more common along streams.

Four-eyed opossums make nests in different places: in the leaf litter that collects at the base of large stemless palms; in hollow logs; in vine tangles up in the trees; on the ground at the base of two trees that are growing right next to each other; and in the crown of plantain plants. These they line with dry leaves.

Four-eyed opossums are nocturnal.

Four-eyed opossums have many young, which they carry around and suckle inside their pouches. The young leave the mother when they get about half the size of the mother. Other than mothers with young, they are solitary.

Four-eyed opossums are eaten by margays and ocelots.

Four-eyed opossums open their mouth very wide and hiss aggressively when found by people, and try to bite them if they get close.

Four-eyed opossums eat anything. They eat sweet tree fruits, the mesocarp of swamp-palm (Mauritia flexuosa) fruits, crickets/katydids, spiders, and armored millipedes. They eat birds that they find nesting up in the trees or on the ground. If the bird flies off or is too large, they eat the eggs or hatchlings. Among the birds they prey upon are tinamous, ground doves, and rusty-belted tapaculos.

Voucher Material (total = 18): Nuevo San Juan (AMNH 268223, 268224, 272818, 273039, 273040, 273054, 273055, 273089; MUSM 11068, 11070, 11071, 11073, 13299, 15319–15323), Orosa (AMNH 74088).

Other Interfluvial Records: Choncó (Vriesendorp et al., 2006a), Divisor (Vriesendorp et al., 2006b), Jenaro Herrera (as Philander andersoni; Fleck and Harder, 1995).

Identification: Amazonian specimens of Philander with distinctly blackish middorsal fur belong to two strongly supported cytochrome b haplogroups (Patton and da Silva, 1997; Patton et al., 2000; Voss et al., 2018). One haplogroup is represented by specimens collected north of the Amazon (in northern Loreto, eastern Ecuador, southeastern Colombia, southern Venezuela, and northwestern Brazil), but it also includes a sequence obtained from a paratype of P. andersoni (Osgood, 1913), which was collected south of the upper Amazon at Yurimaguas, in western Loreto department. The other haplogroup, represented by specimens collected south of the Amazon in Peru and western Brazil, includes specimens that closely resemble the type series of P. mcilhennyi (from Balta in Ucayali department). Both haplogroups are currently recognized as valid species (Patton and da Silva, 2008), but Voss et al. (2018) noted that the material they referred to P. andersoni and P. mcilhennyi on the basis of cytochrome b sequences could not be consistently distinguished by morphological criteria.

Most specimens of Philander mcilhennyi (e.g., the type series from Balta, Peru, and material from western Brazil) are very dark: completely blackish middorsally with dark-grayish flanks and underparts. Although the lateral and ventral fur is frosted with gray, the tonal contrast with the middorsal fur is not abrupt. Additionally, the middorsal fur is much longer than the lateral fur, giving some skins a distinctly shaggy appearance. By contrast, typical material of P. andersoni has paler-grayish flanks, such that the black middorsal stripe is more obvious; additionally, the middorsal fur of andersoni is said to be shorter than that of mcilhennyi, and the ventral fur is either self-whitish or gray-based whitish (rather than dark gray). Lastly, P. andersoni is said to have a medial patch of pale fur at the base of the ear that is absent from the entirely blackish crown of P. mcilhennyi (see Patton and da Silva, 1997, 2008).

Patton et al. (2000) remarked that Philander andersoni and P. mcilhennyi might be sympatric on the “lower Río Javarí in northeastern Peru,” citing a personal communication with D.W.F., who had observed both phenotypes at Nuevo San Juan (on the Río Gálvez). In fact, our voucher material includes specimens that distinctly resemble P. mcilhennyi (e.g., AMNH 272818, MUSM 13299) and others that closely resemble P. andersoni (e.g., AMNH 273055; fig. 19). Cytochrome b sequences that we obtained from specimens of both phenotypes, however, all belong to the mcilhennyi haplogroup (Voss et al., 2018), so introgression rather than sympatry might be the more appropriate interpretation of pelage variation at this locality. Nevertheless, we maintain current binomial usage pending further genetic analysis of our material (Jansa and Voss, in prep.).

A curious aspect of intraspecific variation in both Philander mcilhennyi and P. andersoni is sexual dimorphism in the middorsal fur, which is significantly longer, on average, in females than in males (table 22). Same-sex comparisons, however, support Patton and da Silva's (1997, 2008) description of P. mcilhennyi as longer-furred than P. andersoni, although there is species overlap in our measurements of fur length even when comparing males with males and females with females. To our knowledge, sexual dimorphism in fur length has not been reported from other didelphid species.

Ethnobiology: The Matses do not distinguish this species from other “four-eyed” opossums (all known as cheka bëbëdi; see the generic account for Philander, above) and therefore have no particular beliefs about it.

Matses Natural History: The Matses have no definite knowledge of this species.

Remarks: Of the 17 specimens for which we have definite habitat data, seven (41%) were trapped or shot in primary forest (usually at well-drained sites, but once in the narrow floodplain of the Río Gálvez), another nine (53%) were taken in secondary forest (regenerating abandoned swiddens), and one was taken in an active swidden. Of the 14 specimens with recorded capture heights, 10 (71%) were trapped or shot on the ground, whereas four (29%) were shot or trapped on elevated substrates. Of the latter, one was trapped on a log 1 m above the ground, another was shot 3 m above the ground in a tree, and two others were shot at unrecorded heights in trees (one said to be “not very high” and the other “high up” [English translations from Matses field notes]).

Other Specimens Examined (total = 17): Brazil—Acre, Sena Madureira (USNM 546220–546222), Sobral on left bank Rio Juruá (MVZ 190337). Peru—Loreto, Santa Elena on Río Samiria (FMNH 87125). Ucayali, Balta (LSUMZ 14014, 14015, 16393, 16394; MVZ 136379–136381), 59 km SW Pucallpa (USNM 461133, 499003, 499005–499007).

Voucher Material (total = 1): Orosa (AMNH 73852).

Other Interfluvial Records: See Remarks (below).

Identification: Our single voucher specimen, the skin and skull of a juvenile female collected by the Olallas in 1926, is in poor shape, but it preserves all of the diagnostic traits of the species (table 19). In particular, it has uniformly grayish dorsal pelage (only indistinctly darker middorsally than on the flanks), extensively gray-based ventral pelage (without any median streak of self-whitish fur), a short white tail tip (less than 1/5 of the unfurred part of the tail is unpigmented), upper molars with crenulated protocones and pre- and postcingula, and lower molars with small but distinct postcingulids on m1 and m2 (Voss et al., 2018).

Ethnobiology: The Matses do not distinguish this species from other “four-eyed” opossums (all known as cheka bëbëdi; see the generic account for Philander, above) and therefore have no particular beliefs about it.

Matses Natural History: The Matses have no definite knowledge of this species.

Remarks: The grayish four-eyed opossums reported as Philander “opossum” from the Yavarí-Ucayali interfluve by Pavlinov (1994) and Fleck and Harder (1995) might have been P. pebas or they might have been P. canus. Unfortunately, the specimens collected as vouchers for Fleck and Harder's ecological study, which were deposited at the Instituto de Investigaciones de la Amazonía Peruana, appear to have been lost or stolen, and we have not examined Pavlinov's material (in Moscow).

Other Specimens Examined (total = 57): Brazil—Acre, Fazenda Santa Fé on Rio Juruá (MVZ 190345), opposite Ocidente on Rio Juruá (MVZ 190346); Amazonas, Igarapé Nova Empresa on Rio Juruá (MVZ 190343), Lago do Baptista on S bank of Amazon (FMNH 51095), Sacado on Rio Juruá (MVZ 190344), Santo Isidoro [near] Tefé on S bank of Amazon (AMNH 78954), Parintins (“Villa Bella Imperatriz”) on S bank of Amazon (AMNH 92880, 92881, 93526–93528, 93968), Tapauá on Rio Purus (USNM 461374). Ecuador—Orellana, 42 km S Pompeya Sur (ROM 106101, 106139). Peru—Loreto, Apayacu (AMNH 74388), Avícola San Miguel (MUSM 33590, 33592, 33593), Cabo López (MUSM 33566, 33567, 33569, 33570, 33572), Carretera Iquitos-Nauta km 28.8 (MUSM 34892), Caserio Cahuide (MUSM 33564, 33574, 33576), El Paujil (MUSM 33580), El Triunfo (MUSM 33586, 33587, 33583), Iquitos (AMNH 98642), 19.7 km SW Iquitos (MUSM 33588), Mishana (MUSM 33597), Otorongo Army Base (LACM 91621, 91622), Peña Negra (MUSM 33598), Picuro Yacu (MUSM 33594), Quistococha (FMNH 122745–122748; MUSM 33599, 33600), San Gerardo (MUSM 33602), Santo Tomas (MUSM 33603), Sarayacu on Río Ucayali (AMNH 76448–76450); Madre de Dios, Cusco Amazónico (KU 144120, 144121; MUSM 6074); Ucayali, Balta (LSUMZ 12007, 12010, 14011), Yarinacocha (FMNH 55411).

Voucher Material (total = 1): Nuevo San Juan (MUSM 15292).

Other Interfluvial Records: None.

Identification: Our single voucher conforms in all respects to the morphological diagnosis of Gracilinanus emiliae provided by Voss et al. (2009b), who also tabulated external and craniodental measurements of MUSM 15292 and compared them with homologous values from other specimens, including Thomas's (1909) subadult holotype. Within the Yavarí-Ucayali interfluve G. emiliae might be confused with two superficially similar sympatric species, Hyladelphys kalinowskii and Marmosa lepida, but numerous external and craniodental differences are sufficient for unambiguous identifications of specimens in hand (table 4).

Gracilinanus emilae is the only species of Gracilinanus expected to occur in the Yavarí-Ucayali interfluve, although G. peruanus—formerly synonymized with G. agilis (see Semedo et al., 2015)—is known from scattered localities in the lowlands and foothills of central and southern Peru. Based on material we examined, G. peruanus occurs in the departments of Cusco (e.g., at La Convención, Camisea, San Martín: MUSM 14086), Huánuco (Tingo Maria: BMNH 27.11.1.268, 27.11.1.269), Madre de Dios (Pakitza: MUSM 8922), Pasco (near Villa Rica: AMNH 67242), and Ucayali (Balta: LSUMZ 16378).¹⁰ The only other Peruvian congener, G. aceramarcae, appears to be a strictly montane taxon. Semedo et al. (2015: table 10) provided diagnostic comparisons among the three known Peruvian species of Gracilinanus.

Recently reported specimens of Gracilinanus emiliae from scattered localities in Brazil (Silva et al., 2013; Brandão et al., 2014; Rocha et al., 2015) confirm that this species is very widely distributed in Amazonia. Despite substantial (ca. 5%) divergence at the cytochrome b locus between our Peruvian voucher and the eastern Amazonian material analyzed by Rocha et al. (2015), this appears to be a phenotypically homogeneous taxon, without any obvious morphological differences among the specimens we examined.

Ethnobiology: The Matses do not distinguish this species from other pouchless, long-tailed, black-masked species of small opossums (all known as chekampi; see the account for Marmosa, above) and therefore have no particular beliefs about it.

Matses Natural History: The Matses have no definite knowledge of this species.

Remarks: Our single specimen was shot at night by a Matses hunter in hilltop primary forest as it perched low in the crown of an understory tree.

Other Specimens Examined (total = 11): Brazil—Pará, Belém (BMNH 9.3.9.10 [holotype]), Capím (AMNH 203363). Colombia—Meta, Los Micos (FMNH 87924). French Guiana—Paracou (AMNH 267006). Guyana—Upper Takutu-Upper Essequibo, 20 mi E Comiwariwau Head (ROM 33103), 12 km E Dadanawa (ROM 35465, 35466), 40 mi E Dadanawa (ROM 55519). Peru—Loreto, Zona Reservada Pucacuro (MUSM 24430). Surinam—Marowijne, Langamankondre (RMNH 18231). Venezuela—Monagas, 47 km SE Maturín (USNM 385066).

Species of Marmosops superficially resemble species of Marmosa, but external and craniodental traits that distinguish these distantly related taxa—first recognized as separate genera by Gardner and Creighton (1989)—were described and illustrated by Voss et al. (2004). Subsequent phylogenetic research revealed a deep dichotomy in the genus that was recently formalized by subgeneric nomenclature (Díaz-Nieto et al., 2016). Of the three species definitely known to occur in the Yavarí-Ucayali interfluve, two belong to the nominotypical subgenus and the other to the subgenus Sciophanes; each is easily distinguished from the others by size and qualitative characters (table 23). A fourth species, apparently associated elsewhere with várzea habitats, might also occur in our region (appendix 2).

Voucher Material (total = 38): Nuevo San Juan (AMNH 272704, 272715, 272775, 272782, 272809, 273034, 273051, 273058, 273060, 273061, 273131; MUSM 11032–11036, 11038, 11041, 11042, 11044, 11048, 13288–13292, 15301–15305, 15314), Orosa (AMNH 73853, 73854), San Pedro (UF 30451–30454), Santa Cecilia (FMNH 87122).

Other Interfluvial Records: Jenaro Herrera (Fleck and Harder, 1995).

Identification: Like other material traditionally referred to Marmosops noctivagus (e.g., by Tate, 1933; Patton et al., 2000; Voss et al., 2004; Hice and Velazco, 2012), specimens from the Yavarí-Ucayali interfluve are rather large mouse opossums (table 24) with mostly self-whitish ventral fur and conspicuously beaded interorbital regions, but without distinct postorbital processes at any age or in either sex.¹¹ As reported by Díaz-Nieto et al. (2016), DNA sequences from specimens with these phenotypic traits form a geographically widespread clade that extends across much of western Amazonia and along the adjacent eastern foothills and lower slopes of the Andes. However, this clade includes several allopatric haplogroups, which differ among themselves by about 3%–5% in mean uncorrected sequence comparisons at the cytochrome b locus. Sequenced specimens from the Yavarí-Ucayali interfluve belong to the haplogroup that Díaz-Nieto et al. (2016) called “noctivagus C,” which (in their analyses) also included other sequences from south of the Amazon in western Brazil, eastern Peru, and northeastern Bolivia. Although the “noctivagus C” haplogroup is widespread in lowland habitats, it is also known to extend to at least 2200 m above sea level in southeastern Peru (e.g., near Amay-bamba in Cuzco department; Díaz-Nieto et al., 2016: appendix, locality 98).

Eleven nominal taxa are currently regarded as synonyms of Marmosops noctivagus (e.g., by Voss and Jansa, 2009), but it is currently difficult to determine the application of names to most of the haplogroups discovered by Díaz-Nieto et al. (2016), a problem that those authors discussed at some length. The difficulty arises both from the lack of unambiguously diagnostic characters by which to distinguish taxa in the M. noctivagus complex and from the lack of sequence data from many type localities. Although geographically distant samples of M. noctivagus often exhibit modest morphological differences, such differences are hard to interpret in the absence of data from geographically intermediate sites and in the absence of relevant genetic information. Our material from the Yavarí-Ucayali interfluve, for example, averages larger in most measured dimensions than specimens collected near the type locality of M. noctivagus (in the Andean foothills of Junín department; Gardner, 2008),¹² but the differences we observed are insufficient as a basis for taxonomic distinction, and we have no sequence data from the vicinity of the type locality, which is not, in fact, far outside the known geographic range of the “noctivagus C” haplogroup.

In the absence of a comprehensive revision of the Marmosops noctivagus complex based on better morphological samples and denser molecular sequencing than those available to us at present, it seems profitless to speculate about trinomial nomenclature. However, it is relevant for the purposes of this report to note that sequences obtained from specimens collected on the left (“north”) bank of the Amazon (e.g., at the Estación Biológica Allpahuayo; fig. 1) belong to a different mtDNA haplogroup than those from the Yavarí-Ucayali interfluve, from which they differ by about 3.9% at the cytochrome b locus (Díaz-Nieto et al., 2015: table S4). Because the left-bank haplogroup (“noctivagus A”) also occurs in eastern Ecuador, the type locality of the nominal taxon politus Cabrera, 1913, that epithet would seem to apply to it if there were any point in recognizing subspecies. However, measurements and side-by-side comparisons of our vouchers with left-bank specimens (e.g., the LACM and TTU specimens reported by Hice and Velazco, 2012) did not reveal any consistent morphological differences between them.

Ethnobiology: The Matses do not distinguish this species from other pouchless, long-tailed, black-masked species of small opossums (all known as chekampi; see the account for Marmosa, above) and therefore have no particular beliefs about it.

Matses Natural History: The Matses have no definite knowledge of this species.

Remarks: Of 25 specimens accompanied by definite habitat information, 13 (52%) were taken in primary upland (unflooded) forest, 10 (40%) were taken in secondary growth (abandoned swiddens), and 2 (8%) were captured in houses. Of 12 specimens accompanied by substrate information, 9 (75%) were trapped on the ground or on fallen logs, 2 (17%) were trapped on lianas between 1.5 and 1.8 m above the ground, and one was taken by hand as it perched close to the ground on a small tree. A single specimen was taken in the daytime from a leaf nest in the crown of an Astrocaryum palm at an unrecorded height above the ground, and another specimen was found dead, but all the specimens shot or captured by hand while active were taken at night, and all of the specimens trapped by D.W.F. and R.S.V. were found at dawn in traps that had been baited in the late afternoon of the previous day.

Holotype: MUSM 13284, an adult male specimen collected by the first author (original number RSV 2114) on 1 June 1998 at Nuevo San Juan on the right bank of the Río Gálvez, Loreto department, Peru. The entire specimen was originally preserved in ethanol, but the skull was later extracted and cleaned. The holotype is additionally represented by frozen tissues in the Ambrose Monell Cryo Collection (AMNH), from which Díaz-Nieto et al. (2016) obtained a nearly complete (1148 bp) cytochrome b sequence (GenBank: KT437848) and a shorter (882 bp) fragment from exon 11 of the nuclear BRCA1 gene (KT454025).

Other Voucher Material (total = 24): Jenaro Herrera (AMNH 276714; MUSM 23804, 23805), Nuevo San Juan (AMNH 268215, 268216, 272709, 272760, 273050, 273078, 273151, 273189; MUSM 11040, 11046, 11047, 13285, 13286, 15298–15300, 15306, 15307), San Pedro (MUSM 22331; UF 30449, 30450).

Other Interfluvial Records: None.

Distribution: Known only from the Yavarí-Ucayali interfluve.

Description: A species of Marmosops conforming in all respects to the generic description provided by Díaz-Nieto and Voss (2016: 14–15), but distinguished from other congeneric taxa by the following combination of traits: Dorsal pelage (fig. 20) dull reddish-brown, somewhat resembling Ridgway's (1912) Natal Brown or Olive Brown, but slightly paler. Ventral pelage (fig. 21) continuously self-cream from chin to groin, but this median color much narrowed between the fore- and hind limbs by broad lateral zones of gray-based fur. Tail very long (>140%, on average) and uniformly dark from base to tip dorsally, but indistinctly paler ventrally in some specimens. Hands and feet covered dorsally with mostly pale hairs, but sometimes indistinctly darker over the metapodials. Gular gland present in adult males and adult females. Lateral carpal tubercle present as an anteroposteriorly elongated knob in adult males. Forearm with one antebrachial vibrissa. Mammae 3–1–3 = 7 (e.g., AMNH 272709, 273189; MUSM 15300) or 4–1–4 = 9 (e.g., AMNH 273151), all abdominal/inguinal. Scrotum unpigmented, white.

Nasal bones conspicuously wider posteriorly (near the maxillary-frontal suture) than anteri-orly, and long (extending posteriorly beyond the lacrimals). Lacrimal foramina concealed inside the orbit in some specimens, partially exposed to lateral view on the anterior orbital margin in others. Supraorbital margins smoothly rounded or with inconspicuous beading in juveniles, sub-adults, and young adult females (e.g., MUSM 15298, 15299), but distinctly beaded in older and larger adults, especially males (e.g., AMNH 272760, 276714; MUSM 11046), which also develop small postorbital processes (fig. 23A). Incisive foramina short, not extending posteriorly beyond the canine alveoli; maxillopalatine fenestrae long and narrow, usually extending from between the upper third premolars (P3s) to between the third molars (M3s); palatine fenestrae always present, usually as multiple holes on each side. Subsquamosal foramen short (as in other members of the nominotypical subgenus), not exposing the lateral surface of the petrosal behind the sulcus for the prootic sinus (Díaz-Nieto and Voss, 2016: fig. 9).

Upper canine (C1) simple, without accessory cusps in either sex. M3 anterior cingulum incomplete (preprotocrista terminates at base of paracone, not continuous with anterolabial cingulum). Unworn lower canine (c1) distinctly taller than first lower premolar (p1), sometimes with a small but distinct posterior accessory cusp. Entoconids small, neither as tall nor as bulky as adjacent paraconids.

Comparisons: This species, previously referred to as Marmosops “Gálvez,” was included in molecular analyses reported by Díaz-Nieto et al. (2016), who recovered it within a strongly supported monophyletic group (“Clade E”) that also included M. ocellatus Tate, 1931, and M. “Juruá” (another unnamed form). Marmosops ocellatus—resurrected from synonymy and rede-scribed by Voss et al. (2004)—occupies deciduous, semideciduous, and riparian forests in the Cerrado landscapes of eastern Bolivia and southwestern Brazil (Emmons et al., 2006; Cáceres et al., 2007; Semedo et al., 2013), whereas M. “Juruá” corresponds to the Amazonian species that Patton et al. (2000) called M. impavidus Tschudi, 1845 (a nomen dubium).¹³ Comparisons with M. “Juruá” will be provided when that species is formally named in a subsequent publication. The following comparisons serve to distinguish M. soinii from M. ocellatus, its closest named relative as determined by Díaz-Nieto et al.'s (2016) results.

Marmosops soinii and M. ocellatus are morphometrically similar (table 25), with broad overlap in most measured dimensions (except male tail length). In qualitative comparisons (table 26), M. soinii differs from M. ocellatus by its reddish-brown dorsal pelage, which appears somewhat darker and more richly pigmented in side-by-side comparisons with the somewhat paler dorsal fur of the latter species (fig. 20). The wider lateral zones of gray-based fur in M. soinii are conspicuous in ventral pelage comparisons (fig. 21). Marmosops soinii has a relatively longer tail (about 147% of head-and-body length, on average) that is uniformly dark from base to tip, whereas the relatively shorter tail of M. ocellatus (about 136% of head-and-body length) is bicolored at its base (distinctly paler ventrally than dorsally) and particolored (paler distally than proximally, becoming completely unpigmented toward the tip in most examined specimens). Less conspicuously, a gular gland is present in adult males and at least some adult female specimens of M. soinii, whereas most examined adult specimens of M. ocellatus—the holotype (BMNH 28.2.9.87) is the unique exception—show no trace of glandular activity on the throat or upper chest. Lastly, adult female specimens of M. soinii have either 3–1–3 = 7 or 4–1–4 = 9 abdominal-inguinal mammae, whereas M. ocellatus seems to consistently have 6–1–6 = 13 mammae (L.H. Emmons, personal commun.), of which the two anteriormost pairs are “thoracic” (sensu Tate, 1933), extending toward the lower chest from the abdominal/inguinal positions occupied by the more posterior teats.

These species are craniodentally similar, but specimens of M. ocellatus seem to lack any trace of supraorbital beading, and none that we examined has postorbital processes (fig. 23B). In almost all examined specimens of M. soinii the basisphenoid is concealed from lateral view in the rear of the orbit (AMNH 272760 is the unique exception), but the basisphenoid is laterally exposed through an expanded sphenorbital fissure in the rear of the orbit in M. ocellatus.¹⁴

Etymology: For Pekka Soini (fig. 24), legendary field biologist and visionary conservationist, who lived and worked in Loreto for many years and made important contributions to the herpetology and mammalogy of northeastern Peru (Mittermeier et al., 2004).

Ethnobiology: The Matses do not distinguish this species from other pouchless, long-tailed, black-masked species of small opossums (all known as chekampi; see the account for Marmosa, above) and therefore have no particular beliefs about it.

Matses Natural History: The Matses have no definite knowledge of this species.

Remarks: Of the 23 specimens for which we have capture information, 13 were trapped, shot, or captured by hand while climbing on saplings, lianas, logs, or fallen branches at estimated heights from 0.2 to 1 m above the ground; only five are definitely known to have been captured on the ground (of which three were taken in pitfalls). Most (15) specimens were taken in primary upland forest, but four were in secondary forest (abandoned swiddens), two were in swampy primary forest, one was in primary floodplain forest, and one was captured in a house. Ten specimens were shot or captured by hand while active at night, and another 10 were found at dawn in traps that had been baited in the late afternoon of the previous day (three specimens found at dawn in pitfalls might have been taken at any time in the previous 24 hours).

Specimens of Marmosops ocellatus Examined (total = 23): Bolivia—Santa Cruz, 7 km SE Ariruma (AMNH 275461), 6 km W Ascención (AMNH 261265), Ayacucho (USNM 390571, 390572), Buenavista (BMNH 26.1.5.25 [holotype], 28.2.9.87, 28.2.9.90), El Refugio (USNM 584466, 584467, 584469), 3.5 km W Estación Pailón (AMNH 260026–260028), 7 km E and 3 km N Ingeniero Mora (AMNH 247652), Lago Caimán (USNM 581979), 2 km S Las Cruces (AMNH 263549), 10 km N San Ramón (AMNH 261266, 261267), 15 km S Santa Cruz (MSB 58512, 58513), Warnes (USNM 390022), 13 km N Zanja Honda (AMNH 275462). Brazil—Mato Grosso, 3 km W Cáceres on BR 30 (USNM 390026).

Voucher Material (total = 12): Jenaro Herrera (AMNH 276697, 276700, 276705, 276718, 276723; MUSM 23799–23803), Nuevo San Juan (MUSM 13287), San Pedro (UF 30454).

Other Interfluvial Records: None.

Identification: The small species of Marmosops formerly associated with the name M. parvidens (e.g., by Pine, 1981) were referred to the subgenus Sciophanes by Díaz-Nieto et al. (2016). Specimens from the Yavarí-Ucayali interfluve were subsequently examined by Díaz-Nieto and Voss (2016), who referred them to M. bishopi, a taxon originally described as a subspecies of M. parvidens on the basis of a female holotype collected in central Brazil. As recognized by Díaz-Nieto and Voss (2016), M. bishopi is a widespread species that includes several morphologically indistinguishable mtDNA haplogroups. Unfortunately, no sequence data is available from the holotype, nor did Díaz-Nieto et al. (2016) obtain sequences from specimens collected in the Yavarí-Ucayali interfluve. The holotype (USNM 393535) has paler fur and a much shorter tail (116 mm) than the single adult female from our region (table 27), but other measurement differences between these specimens seem unremarkable. Our material agrees with the emended description of this species by Díaz-Nieto and Voss (2016: 43–49), who provided cranial photographs of one of our vouchers (MUSM 23803).

Ethnobiology: The Matses do not distinguish this species from other pouchless, long-tailed, black-masked species of small opossums (all known as chekampi; see the account for Marmosa, above) and therefore have no particular beliefs about it.

Matses Natural History: The Matses have no definite knowledge of this species.

Remarks: Our single specimen from Nuevo San Juan was caught by hand at night by a Matses hunter, but no information was recorded about the habitat in which it was taken. The remaining 10 specimens (all from Jenaro Herrera) were taken in pitfall traplines: five in well-drained secondary growth, three in swampy primary forest, and two in white-sand forest.