ABSTRACT

On the basis of morphological criteria, five main groups of Philander were recognized, though some of them seemed to be composites of more than one species and subspecies. Within the "frenatus group" the greatest resolution was achieved. The subgroup endemic to the Chiquitano dry forest biome of east-central Bolivia, currently not recognized taxonomically, showed enough morphological distinctness to be considered a valid form of Philander, which we argue should be upgraded to species level, Philander crucialis (Thomas, 1923).

INTRODUCTION

Grey four-eyed pouched opossums are didelphid marsupials of the genus Philander widely distributed throughout the Neotropics, from east-central Mexico to northeastern Argentina. Several authors have reviewed the contents of the genus, but their taxonomic conclusions have differed considerably. On the basis of morphological and molecular criteria, 6 species of Philander are currently recognized: P. andersoni, P. deltae, P. frenatus, P. mcilhennyi, P. mondolfii and P. opossum. For the last species, no less than 5 subspecies have been recognized. Neither a review of geographic variation in Philander nor a revision of currently recognized forms has been undertaken in a broad and integrative sense.

The main aim of this research was to clarify the taxonomic status of some Philander taxa, by examination of skull morphometry and external morphology, to assess if patterns of geographic and taxonomic variation observed were consistent with the current classification of the genus.

MATERIALS AND METHODS

460 skulls (and skins when available) representing most nominal taxa of Philander from Central and South America were examined (deposited in the collection of mammals of the NMNH and AMNH). Only skulls of adult specimens were considered in the morphometric analysis. Because didelphids exhibits sexual dimorphism in size, all specimens were segregated by sex and studied separately (266 males and 194 females). To assess size differences between taxa comparisons were made only between individuals of the same sex and age.

Twenty characters were measured from each skull: condylobasal length (CBL), condyloincisive length (CIL), palatal length (PL), palatal width (PW), zygomatic breadth (ZB), least interorbital breadth (LIB), least postorbital breadth (LPB), breadth palatal shelf (BPS), braincase breadth (BB), maxillary toothrow length (CM4), molar toothrow length (M1M4), rostral length (RL), rostral width (RW), nasal length (NL), nasal breadth (NB), rostrum breadth across frontals (RBF), rostrum breadth across jugals (RBJ), mastoid breadth (MB), occipital condyle breadth (OCB), basioccipital length (BL). Pelage patterns and colors were also examined.

Specimens were grouped based on consideration of morphometrical, external morphological and geographic criteria. Discriminant-function Analysis (DFA) was performed to test our hypothesized groupings . To assess differences in skull size and shape, among and within groups, Principal Component Analysis (PCA) was performed.

The map of Figure 1 shows all specimens localities. The four hypothesized groups of Philander (frenatus, andersoni, nigratus and opossum) are shown in different colors.

Figure 1

Figure 4

DISCUSSION

The combined analysis of both skull morphometrics and external characters allowed the recognition of five main morphological groups of Philander (Figures 1 and 4). The distinctness of each group was well supported by the data, though some of them may still be composites of more than one species or subspecies.

For the "frenatus group" the greatest resolution was achieved. As shown in figures 2a and 2b, that group is clearly separated from other Philander in the first canonical root. The size component is explaining the 83% of the variation in the sample.

The separation of nigratus and andersoni from the "opossum group" can be seen in the second and third canonical roots (mainly shape differences) respectively. P. nigratus and P. andersoni are currently recognized as valid monotypic species of Philander. Although nigratus and andersoni are hard to take apart on the basis of external morphology, they can be discriminated morphometrically. Within the "opossum" group, considerable variation was observed. Further analyses suggest that, the Andes may separate two recognizable subgroups within the "opossum group", which may remain a composite even of distinct species and subspecies.

The "frenatus group" comprises small-sized animals, with pale gray dorsal pelage. Members occur in the Atlantic Forest and Cerrado southern Brazil, southern Paraguay, north-central Bolivia, and up north to Venezuela, inhabiting drier forests between the Andes and the Amazon Basin (Fig.1).

Some members of the "frenatus group" are found in sympatry with P. nigratus in central Peru (see Figure 1). Of the groups compared, these two were the most different from each other in the analysis, both in skull morphometrics and in external traits. Perhaps because of that differentiation, those two forms can coexist, avoiding direct competition for resources. No other examples of sympatry are known in Philander.

Within the "frenatus group", a morphologically distinctive subgroup is restricted to Santa Cruz Province, east central Bolivia and in western Mato Grosso State, Brazil (Figs. 4 and 5). Though currently not recognized taxonomically, this subgroup showed consistent morphometric (Figures 3a and 3b) and pelages attributes relative to other populations referred to the "frenatus group". It must be considered a unique form of Philander, which we suggest should be recognized as a valid species. The name crucialis Thomas, 1923, currently in the synonymy of opossum, is available for this unique taxon. The topotypical material studied perfectly matches Thomas's description of crucialis (type locality: Santa Cruz de la Sierra, Bolivia).

CONCLUSIONS

Although the classificatory schemes and various aspects of the Natural History of Philander are far from complete, the results of the present study contribute to a better understanding of the Philander species complex and set the background for further and complementary studies aiming to resolve long standing problems regarding the number of species and subspecies and their geographical ranges.