Mesomys Wagner, 1845

PATTON, JAMES L., DA SILVA, MARIA NAZARETH F. & MALCOLM, JAY R., 2000, Mammals Of The Rio Juruá And The Evolutionary And Ecological Diversification Of Amazonia, Bulletin of the American Museum of Natural History 2000 (244), pp. 1-306 : 190-191

Mesomys Wagner, 1845

Spiny tree rats

Spiny tree rats of the genus Mesomys are common components of the arboreal fauna of Amazonia, yet are only rarely collected unless effort is made to place traps in vine tangles or trees, or to hunt at night. Individuals will, on occasion, come to the ground and be taken in terrestrial traps baited with standard foods (mixtures of seeds, raisins peanut butter, etc.), and they are often com­

mon in tree falls or disturbed areas, as around villages where they can be found on rafters in houses. This is a relatively small­bodied animal, with the short and broad feet, large plantar pads, and sharply decurved claws indicative of arboreal habits. Among the echimyid climbing rats, Mesomys (and its close relative Lonchothrix ; see Lara et al., 1996) is unique in its combination of dorsal pelage comprised of broad and stiff spines, a sparsely haired tail that terminates in a distinct brush, or pencil, and small, rounded cheekteeth with flexi wearing to isolated oval fosettae or pits rather than tending to form transverse lamellae. Mesomys shares these tooth characters with the terrestrial spiny rats ( Proechimys , Hoplomys , and Trinomys ) and its subfamilial placement has been uncertain as a result (see Patton and Reig, 1989, for discussion, and fig. 119C and D for comparisons).

Mesomys occurs throughout Amazonia, from the eastern flank of the Andes in Colombia south to Bolivia and eastward to the Guianas and the mouth of the Rio Amazonas in eastern Brazil (Emmons, 1994; Emmons and Feer, 1997). Woods (1993) lists five species, two of which ( didelphoides Desmarest and obscurus Wagner) have now been allocated to the genus Makalata (Emmons 1993) . Of the other three, M. hispidus is considered to be widespread throughout Amazonia, M. leniceps is known only from the Andean foothills in northern Peru´, and M. stimulax ranges in eastern Brazil south of the Rio Amazonas (Emmons and Feer, 1997) Although the genus is in clear need of revisionary attention, the molecular data we have accumulated offer substantial help in apportioning the demonstrable geographic variation of this taxon into reasonable phylogenetic units.

Da Silva and Patton (1993) and Patton et al. (1994) provided an initial review of the degree of differentiation in the genus based on variation in the mitochondrial cytochrome­b gene. These papers documented well­defined regional geographic clades that shared deep divergences, suggestive of separate species status. In particular, they documented sympatry between spiny tree rats at two localities where cytochrome­b sequences differed by more than 13%, more so than the divergence between eastern Amazonian M. stimulax and any other sample from central Brazil to Perú and Bolivia. We expand on these accounts, and assign the broadly distributed haplotype clade we define below to the species M. hispidus . As a consequence we argue that the highly divergent and sympatric, but very localized clade is an unknown species, and we describe it here as a new species. Our rationale for these actions is given in the accounts below.

First, however, we describe the regional patterns of molecular divergence, based on all available geographic samples of spiny tree rats. We then describe the new taxon after which we examine patterns of differentiation within the dominant Amazonian taxon, M. hispidus , within the Rio Juruá basin.