Marmosa (Micoureus) germana Thomas, 1904

Voss, Robert S. & Giarla, Thomas C., 2021, A Revision of the Didelphid Marsupial Genus Marmosa Part 3. A New Species from Western Amazonia, with Redescriptions of M. perplexa Anthony, 1922, and M. germana Thomas, 1904, American Museum Novitates 2021 (3969), pp. 1-28 : 18-20

Marmosa (Micoureus) germana Thomas, 1904

Marmosa germana Thomas, 1904: 143 (original description).

Marmosa (Marmosa) waterhousei: Cabrera, 1919: 39 (part, germana treated as synonym).

Marmosa germana germana: Tate, 1933: 80 (name combination).

Marmosa (Marmosa) germana germana: Cabrera, 1958: 15 (name combination).

Micoureus regina: Gardner, 1993: 20 (part, germana treated as synonym).

Micoureus regina germanus: Gardner, 2005: 13 (name combination).

Marmosa (Micoureus) regina: Voss and Jansa, 2009: 101 (part, germana treated as synonym).

TYPE MATERIAL AND TYPE LOCALITY: The holotype (by original designation, BMNH 80.5.6.77) consists of the skin and skull of a subadult female collected by Clarence Buckley at Sarayacu, a village on the Río Bobonaza in Pastaza province, Ecuador. No date of collection accompanies the holotype, nor was any mentioned by Thomas (1904), but Buckley’s mammals from eastern Ecuador are thought to have been collected between July 1877 and September 1878 ( Vane-Wright, 1991). An adult male paratype ( BMNH 80.5.6.76), consisting of a skin and partial skull, was also collected by Buckley at Sarayacu.

DISTRIBUTION AND SYMPATRY: Examined specimens of Marmosa germana are from the Amazonian lowlands of southeastern Colombia (Caquetá), eastern Ecuador (Orellana, Pastaza), and northeastern Peru (Loreto); all the localities in the Peruvian department of Loreto are on the left (north) bank of the Amazon/Marañón ( fig. 2 ). Marmosa germana has been collected sympatrically with several congeners including M. (Eomarmosa) rubra , M. (Marmosa) waterhousei , M. (Micoureus) jansae , M. (Micoureus) rutteri , and M. (Stegomarmosa) lepida (appendix 2).

DESCRIPTION: The dorsal fur of Marmosa germana ( fig. 3 ) is distinctly browner than that of most other congeners; Thomas (1904) described the dorsal pelage color of M. germana as “rather paler” than Ridgway’s (1912) Mummy Brown and “rather less yellow” than Burnt Umber, whereas Tate (1933: 81) described it as “between Mummy Brown and Bistre.” With many more specimens at hand than either Thomas or Tate examined, including some reasonably fresh skins (<25 years old), these colors do not seem like good matches: all of them are far too dark. Better matches are to be found among paler shades in the same dilution series (e.g., Dresden Brown rather than Mummy Brown; Snuff Brown rather than Bistre), but somewhat less saturated colors (like Buffy Brown) are even closer matches. The ventral pelage is uniformly gray-based over the throat, chest, the insides of the forelegs, and the upper abdomen; self-colored (buffy) fur is present only on the chin and groin of most specimens, but one (LSUMZ 28018) has a very narrow and apparently continuous midventral streak of self-buffy fur. The dorsal surface of the hind feet is covered with short, pale fur, but the fur over the metacarpals and the proximal phalanges of the forefeet is usually dark, such that only the distal phalanges of the hand are pale. The tail is about 143% of head-and-body length on average, and about 30 mm or more of the base of the tail is covered with moderately long (ca. 10 mm) fur. The naked (scaly) part of the tail is uniformly dark (brownish), without any pale markings.

Skulls of Marmosa germana are large in all measured dimensions ( figs. 4C, 4F ; table 4) with broad rostrums, wide zygomatic arches, and well-developed postorbital processes. The palate is almost completely ossified, with short (3.3 ± 0.8 mm) maxillopalatine fenestrae that extend only from M1 to M 2 in most (10 out of 14) specimens scored for this trait, and palatine fenestrae are consistently absent. The auditory bullae are very small and lack well-developed anteromedial laminae; although some specimens have conventionally “globular” (roughly hemispherical) bullae, a few large specimens from eastern Ecuador (e.g., AMNH 67292; FMNH 43179, 124606) have “subconical” bullae ( Tate, 1933: 80), so called because they have an apical process containing a vascular sinus. The second upper premolar (P2) has an incomplete lingual cingulum, the postprotocrista of M3 terminates at or near the base of the metacone, and the lower molars lack posterior cingulids.

COMPARISONS: In our phylogenetic analyses of molecular data ( fig. 1 ), Marmosa germana is consistently recovered as the sister taxon of M. alstoni , from which it differs on average by about 9% in uncorrected pairwise sequence comparisons at the cytochrome b locus. Marmosa alstoni is the oldest name that applies to a putative species represented by nine sequences from Central America, western Colombia, and northwestern Ecuador (Voss et al., 2020: table 2, fig. 1 ). It is our belief that these sequences represent at least four distinct species, two of which are undescribed (Voss et al., in prep.). For simplicity, the following comparisons concern M. alstoni sensu stricto, which occurs in the volcanic cordilleras of Costa Rica.

Although Marmosa germana and M. alstoni are similar in size, with broadly overlapping external dimensions, they are easily distinguished by qualitative external features. In side-byside comparisons, the dorsal pelage of M. germana appears browner, and it is coarser and less cohesive than that of M. alstoni , which is grayer, softer, and woollier. Less subjectively, the tail of M. germana has a shorter furry base (only about 25–35 mm of the base of the tail is furred), and the basal fur is shorter (usually about 10–11 mm long); by contrast, about 40–60 mm of the base of the tail is furred in M. alstoni , and the basal fur is about 20 mm long. Additionally, the naked (scaly) part of the tail is completely dark in M. germana whereas the distal two-thirds or more of the naked caudal epithelium is unpigmented (whitish) in M. alstoni (see Voss et al., 2020: fig. 8 ). These species are craniodentally similar, but M. germana has a narrower postorbital constriction than M. alstoni (e.g., 7.6 ± 0.2 mm versus 8.2 ± 0.4 mm in males); the corresponding proportional difference (e.g., LPB/CBL × 100 = 16.9% on average in M. germana males versus 18.6% in M. alstoni males) is small but perceptible in side-by-side comparisons of representative series of skulls.

Marmosa germana occurs sympatrically with two other members of the subgenus Micoureus , from which it can be distinguished unambiguously by a combination of external and craniodental characters ( table 2); descriptive comparisons among these taxa were provided in the preceding account for M. jansae and in Voss et al.’s (2020) account for M. rutteri . Unfortunately, it is much more difficult to distinguish M. germana from M. constantiae , an allopatric species that occurs south of the Amazon ( Voss et al., 2019). Marmosa germana and M. constantiae are only distantly related ( fig. 1 ), but they are morphologically similar, at least in northern Peru, where the local form of M. constantiae lacks the white caudal markings of Bolivian and central Brazilian specimens ( Voss et al., 2019). Measurements of M. constantiae ( Voss et al., 2019: table 7) broadly overlap those of M. germana , and careful side-by-side comparisons of skins and cranial material have not revealed any consistent qualitative differences. For the moment, at least, the morphological discrimination of these phylogenetically distant and molecularly divergent taxa remains a research challenge.

REMARKS: Two putative species were associated with the name Marmosa germana by Voss et al. (2020), who assigned five cytochrome b sequences to the lineage they called “ germana A” and a single sequence to its sister lineage “ germana B.” The former sequences are from the Marañón-Napo interfluve in eastern Ecuador and northern Peru (Voss et al., 2020: fig. 1 ), whereas the latter is from the Napo-Putumayo interfluve. With so few sequences in hand, it is difficult to know whether the Napo really separates different mtDNA lineages of M. germana , but it seems harmless to adopt this as a working hypothesis. The sequence identified as germana B (obtained from LSUMZ 28018) differs from the sequences assigned to germana A (KU 157972, 157973; ROM 105521; TTU98988, 101236) by an average uncorrected distance of 4% (Voss et al., 2020: table 4), and the corresponding voucher material is also distinctive. Unlike any of the specimens from the Marañón-Napo interfluve, which have almost entirely gray-based ventral fur, LSUMZ 28018 has a narrow but apparently continuous midventral streak of self-buffy fur, and it has slightly narrower nasal bones. These are not very impressive differences, however, particularly given the handful specimens at hand, so we are reluctant to provide a new name for this unique specimen.

HABITATS: The dominant climax formation throughout the geographic range of Marmosa germana is evergreen lowland rainforest, but western-Amazonian rainforests include different kinds of vegetation associated with characteristic edaphic factors and successional stage. At the Reserva Nacional Allpahuayo-Mishana (locality 22; fig. 2 , appendix 1), Hice and Velazco (2012) trapped specimens of M. germana in vegetation locally known as varillal, which grows on patches of white sand on welldrained high ground ( fig. 8 ). By contrast with tall monte alto vegetation (in which M. jansae was trapped at the same locality; see above), varillal habitat has a lower (20 m) canopy of smaller trees, some of which have multiple stems; very few palms or lianas; and abundant terrestrial bromeliads that often carpet the forest floor. However, other specimens of M. germana were captured at this locality “in older secondary growth where it grades into less disturbed forest.” According to the authors, secondary vegetation occurs on abandoned agricultural plots that are usually sited on loamy soils, so the association between M. germana and white-sand forest is evidently not obligatory. Unfortunately, no other useful information is available about the habitats in which this species has been taken.

SPECIMENS EXAMINED (N = 18): Colombia — Caquetá, Tres Troncos (FMNH 70979). Ecuador — Orellana, 35 km S Pompeya Sur (ROM 105521); Pastaza, Río Pastaza (BMNH 54.265), Río Pindo Yacu (FMNH 43179), Sarayacu (AMNH 67292; BMNH 80.5.6.76, 80.5.6.77); Sucumbios, Marian (FMNH 124606). Peru — Loreto, Collpa Salvador (MUSM 17588), 25 km S Iquitos (TTU 98845, 98846, 98988, 101094, 101236), Quebrada Orán (LSUMZ 28018), San Jacinto (KU 157972), Santa Luisa (FMNH 87116), Teniente López (KU 157973).

Figures

FIG. 2. Collection localities of examined specimens of Marmosa germana, M. jansae, and M. perplexa. Numbers are keyed to entries in our gazetteer (appendix 1).

FIG. 3. Dorsal and ventral views of skins of Marmosa jansae (ROM 118880), M. perplexa (USNM 513425), and M. germana (FMNH 43179).

FIG. 4. Dorsal and ventral cranial views (×1.75) of Marmosa jansae (A, D; ROM 118880), M. perplexa (B, E; USNM 513425), and M. germana (C, F; TTU 101236).

FIG. 1. Maximum-likelihood (ML; left) and Bayesian (right) phylogenetic trees for putative species of Marmosa (Micoureus) inferred from concatenated sequences of six genes (after Voss et al., 2020: fig. 6). Numbers at nodes denote ML bootstrap support or Bayesian posterior probabilities. Outgroups are not shown. Dotted lines connect the same taxa across the two trees to highlight differences. Names of species treated in this report are in boldface.

FIG. 8. White-sand vegetation (varillal) at the Reserva Nacional Allpahuayo-Mishana, Loreto, Peru, habitat of Marmosa germana (photo by Christine L. Hice).