ECHIMYIDAE GRAY, 1825

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 : 169-170

ECHIMYIDAE GRAY, 1825

The terrestrial spiny rats and tree rats of the family Echimyidae are second only to the Muridae as the most diverse family of rodents within lowland Amazonian forests Some taxa, especially the terrestrial spiny rat genus Proechimys , are also among the most abundant in any local mammalian commu­

nity. On the other hand, virtually all arboreal genera are poorly represented in museum collections, due either to inadequate sampling of the canopy or to true rarity.

We encountered five genera of echimyids in the Rio Juruá basin: eight species of Proechimys (see da Silva, 1995, 1998); two species of the spiny tree rat, Mesomys ; two species of bamboo rats, Dactylomys ; and one species each of the red­nosed tree rat, Makalata , and brush­tailed rat, Isothrix . Proechimys were exceedingly numerous at all sites, generally the most common of all mammals trapped. Nearly 40% of our specimens were of this genus. We heard the distinctive nocturnal calls of Dactylomys at most sites, and all along the river during travel periods, but few specimens were acquired, and none in our traps. Mesomys were among the more common members of the canopy community, as ascertained by our platform trapping program. Isothrix was considerably less common, yet encountered routinely at most sites, both captured in our canopy traps or observed at the entrances of tree­hole nests. Finally, Makalata was uncommon throughout the river, as individuals were only rarely captured by any means and it was never seen at night while hunting, unlike the other arboreal taxa.

Generic limits and relationships among echimyids are poorly resolved for the most part. Lara et al. (1996) argue that the extant generic lineages diverged nearly simultaneously from a common ancestor in the Miocene. Thus, the rapidity of this early radiation, coupled with combinations of uniquely derived traits and broadly shared primitive ones, has resulted in considerable confusion in the diagnosis of genera and in the allocation of species to them. The traditional separation of extant genera into three subfamilies (summarized in Woods, 1993)—the Dactylomyinae ( Dactylomys , Kannabateomys , and Ollalamys), Echimyinae ( Echimys , Isothrix , Makalata , and Nelomys ), and Eumysopinae ( Hoplomys , Mesomys , Proechimys , and Trinomys )—is not supported by molecular analyses of either proteins or mtDNA sequences (Patton and Reig, 1989; Lara et al., 1996).

The difficulties in defining and diagnosing genera are exacerbated severalfold when one considers the problems of circumscribing species in nearly all of these groups. For example, the terrestrial spiny rats of the genus Proechimys are among the most notorious of Neotropical taxa for posing difficulties in the recognition of species boundaries. The confusion surrounding species recognition in this genus was aptly stated by Oldfield Thomas (1928: 262) some 70 years ago in his often quoted statement: ‘‘The bewildering instability of characters... makes it at present impossible to sort them according to locality into separate species, subspecies, or local races.’’ Nevertheless, great strides have been made in recent years in understanding the patterns of morphological character variation in this difficult genus (Patton and Gardner 1972; Gardner and Emmons, 1984; Patton 1987; Aguilera and Corti, 1994; da Silva 1995, 1998). The number of actual species of Proechimys , including those sympatric at single sites, remains underappreciated for the most part, and we know little or nothing about even the most simple ecological relationships among them (but see Emmons 1982; Malcolm, 1992). As documented by da Silva (1995, 1998), at least eight species of Proechimys are present along the Rio Jurua´ This is double the number we expected based on the prior work by one of us in adjacent Perú (Patton and Gardner, 1972; Patton 1987). As we also document below, and as already alluded to in earlier publications on this fauna (da Silva and Patton, 1993, 1998 Patton et al., 1994, 1996a), additional species of Proechimys and other echimyid genera can be recognized elsewhere in Amazonia based on the Rio Juruá materials. One of these, a Mesomys , appears to be unknown to science, and we describe it here. For the others, we assign existing names in the appropriate generic accounts, below.

The genera of echimyids present in the Rio Jurua´, indeed throughout much of Amazonia are each recognizable by a combination of external (spinose or nonspinose fur, brushy or naked tail) and craniodental (shape of the interorbital region; development of supraorbital ledges; laminate or nonlaminate cheekteeth; parallel or diverging maxillary toothrows) features. In the accounts below, we describe and figure each taxon that we found within the Rio Juruá basin.