Proechimys echinothrix da Silva, 1998

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 : 228-232

Proechimys echinothrix da Silva, 1998

TYPE LOCALITY: ‘‘ Brazil, Amazonas: Colocação Vira­Volta, left bank Rio Juruá on Igarapé Arabidi, affluent of Paraná Breu 3°17̍S, 66°14̍W.’’

DESCRIPTION: This is one of the most easily distinguishable and among the largest spiny rats occurring in western Brazil. In general morphology, these animals are relatively robust, have long ears, a moderately long tail and large hind feet (tables 60 and 64). Overall, the color of the body is uniformly reddish­brown, coarsely streaked on the back and sides with varying amounts of black; the interspersed heavy, dark brown guard­hairs make the middorsum appear somewhat dark­ er, but this grades evenly into the brighter

and paler sides of the body. The aristiform hairs are long and much broader than those of any other species of spiny rats on the Rio Juruá (da Silva, 1998: table 4), with distinctly strong and blunt tips that are very conspicuous to the eye and touch, especially in the middorsal region. The color of the venter, chin, and inner surfaces of forelimbs and hind limbs is pure white. The tail is indistinctly bicolored, dark above and white ventrally. It is well haired, with the scales nearly completely obscured from view. The hind feet are long and narrow, nearly unicolored white in most specimens and lack a dark band on the ankle joint. The juvenile pelage varies from uniformly grayish brown (age class 1) to pale brown mixed with Sanford Brown (age class 6). The plantar surface of the hind feet has six tubercules; the lateral metatarsal tubercule is weakly to moderately developed, but is always visible and shorter than the medial metatarsal tubercule.

The baculum is massive and relatively short; its shaft is broad with a thick and expanded base and the distal end has a pair of divergent apical extensions that are separated by a shallow median depression (fig. 137; table 3 in da Silva, 1998). Da Silva (1998) figures and describes the soft anatomy of the male phallus.

The skull is moderately large, with a long and narrow rostrum (figs. 138 and 139) and a well­developed supraorbital ledge extending over the orbits but discontinuous across the parietals. The zygoma usually lack a postorbital process or, if present, it is low and rounded with equal contributions by the jugal and squamosal. A well­developed groove with a lateral flange is present on the floor of the infraorbital foramen. The incisive foramen is ovate to lyrate in general shape, with posterolateral margins mostly flat, or only weakly flanged with very shallow grooves extending onto the anterior palate, which lacks a median ridge (fig. 140). The premaxillary portion of the septum is long and narrow, extending between one half and two thirds of the length of the opening; the maxillary portion is typically attenuate and has weak to no contact with the premaxillary portion; and the vomer is visible in most specimens. The mesopterygoid fossa is moderate in depth but broad (fig. 141), with an angle of indentation into the posterior margin of the palate averaging about 70° and extending to the anterior half of M3. The median number of folds in all upper cheekteeth is three, although M3 occasionally only has two folds.

SELECTED MEASUREMENTS:: Selected external and cranial measurements are given in table 64.

COMPARISONS: This is one of the most readily identifiable species of Proechimys in western Amazonia. It can be distinguished from all other sympatric species by the following combination of characters: medium to large­sized body covered by very heavy aristiform hairs, especially along the dorsum long and very broad aristiforms with a distinctly sharp tip lacking any whiplike extension as seen in P. simonsi and P. steerei ; ears large; tail hairy and distinctly clear white below, almost brushy in comparison to other species; tail approximately two thirds of the length of the body, relatively and absolutely shorter than the tail of P. simonsi ; hind foot nearly unicolored white; cranial features include weakly developed posterior portion of the temporal ridges; uniformly three folds on all upper teeth; oval to lyre­shaped incisive foramen with an expanded and long contribution of the premaxillary portion (in contrast to P. simonsi , but similar to P. brevicauda ) but with both attenuate flanges and a maxillary portion of septum that lacks a keel (in contrast to P. brevicauda ); vomer visible although only barely so. The broad and short baculum of P. echinothrix distinguishes this species from all others, except P. cuvieri and P. pattoni , both of which are distinctive in other morphological features.

MOLECULAR PHYLOGEOGRAPHY: We have examined cytochrome­b sequences from individuals collected at each of the three localities from the Upper and Lower Central and Mouth regions of the Rio Jurua´, as well as from the upper Rio Urucu south of the Rio Solimões, and from the Rio Jaú and Rio Tiquie´, all to the north of the Solimões and west of the Rio Negro (fig. 145, above; table 69). Two clearly defined and deeply divergent clades are evident, one north of the Rio Solimões and the other to the south (fig. 145, below). These differ by an average of 10.8% All specimens from north of the Rio Soli­

mões, from the upper Rio Negro as well as the Rio Jau´, form a tight cluster of haplotypes that differ, on average, by only 1.4%. Specimens from the Rio Juruá and upper Rio Urucu, however, differentiate into two clades with the moderate divergence of 4.3% between them. One of these clades includes localities from the left bank of the Rio Juruá (Barro Vermelho [locality 12] and Colocação Vira­Volta [locality 14]); the other pairs the sole locality from the right bank of the Rio Juruá (Vai­Quem­Quer [locality 15]) with that from the upper Rio Urucu, which is further to the east. Thus, the Rio Juruá appears to separate this species into definable monophyletic haplotype clades.

MORPHOMETRIC VARIATION: Our samples are of limited size, but geographic divergence is evident in three morphometric variables in comparisons between individuals from opposite sides of the Rio Jurua´. Ear length (F 1,25 = 4.471, p = 0.0391), mastoid breadth (F 1,24 = 12.129, p = 0.0019), and occipital condyle breadth (F 1,23 = 6.552, p = 0.0178) are all significantly different by one­ way ANOVAs. Thus, there is minimal morphological differentiation to match that observed in mtDNA haplotypes.

DISTRIBUTION AND HABITAT: In her description of P. echinothrix, da Silva (1998) allocated only specimens from the Rio Juruá and upper Rio Urucu to this species, both sets of localities south of the Rio Solimões in westcentral Amazonian Brazil. However, she not­ ed that echinothrix ­like animals had been collected throughout the Parque Nacional Jaú northwest of the mouth of the Rio Negro in the central Amazon and that one of us (J. L Patton) had examined specimens from the Río Vaupés (= Uaupés) in Amazonian Colombia in the collection of the Instituto de Ciencias Naturales, Universidad Nacional de Colombia, in Bogotá that he believed to represent this same taxon. Finally, we have recently obtained specimens of this same taxon from Comunidade Colina, on the right bank of the Rio Tiquie´, a tributary of the Rio Uaupés. Thus, the range of this species, or a complex of closely related taxa, covers a large part of western Amazonian Brazil and

adjacent Colombia on both sides of the Rio Solimões, and is likely to be found in northeastern Perú as well (fig. 145, top). It’s distinctive external morphology should make this one of the more easily identifiable species of Proechimys in regional faunas of this area.

We collected specimens in the Rio Juruá only in the Lower Central (locality 12) and Mouth regions (localities 14 and 15), but on both banks of the river (figs. 133 and 145). Of the 50 specimens collected, 43 are from the terra firme forests of the Mouth Region, often at the edge of flooded várzea; the remaining seven specimens are from terra firme forest at Barro Vermelho (locality 12). We captured animals in Tomahawk (63%) and Sherman (29%) traps on the standardized lines; we also obtained specimens by hunting (2%) and in Victor snap traps (6%) set along other trails within terra firme forest. Young and subadult individuals were caught in both Sherman and Tomahawk traps in about equal numbers (46% and 53% respectively); all but one adult individual were caught in Tomahawk traps.

REPRODUCTION: We caught P. echinothrix only in the months of October and May. Our sample comprises an approximately equal number of males and females, as well as young and adult individuals. Eight of the 24 males were reproductively active, all of which were of age classes 9 and 10. In contrast, reproductively inactive individuals belonged to age classes 1 and 3 (1 specimen each), 5 (3 specimens), and 6 (4 specimens). We caught pregnant females in October (n = 2) and May (n = 9), in both dry and wet seasons. One pregnant individual was also lactating, suggesting postpartum estrous. Most pregnancies were in females of age class 9 or 10, although two individuals of age 6 were pregnant. Signs of estrous were first observed in individuals of age 5. The modal litter size was 2, the range 1–3.

KARYOTYPE: We prepared chromosomes from 31 individuals, including four from Barro Vermelho (locality 12) and all specimens from both Vira­Volta (locality 14) and Vai­Quem­Quer (locality 15). The karyotype is illustrated in da Silva (1998). It has a 2n = 32 and FN = 60, with the autosomes comprising two pairs of very large metacentrics, eight pairs of medium­sized to small metacentrics and submetacentrics, one pair of large and four pairs of small to medium­sized subtelocentrics. The X­chromosome is a small acrocentric and the Y­chromosome is a smaller acrocentric. The karyotype of P echinothrix is similar to that of P. simonsi which also has 2n = 32, but differs by having one extra pair of small subtelocentrics and lacking the large pair of acrocentric chromosomes (see Patton and Gardner 1972).

SPECIMENS EXAMINED (n = 50): (12) 3 m 4 f — INPA 2551–2552; JLP 15816; JUR 188; MPEG 25500; MVZ 187167–187168 (13) 1 f — MNFS 1792; (14) 8 m, 5 f — INPA 2550; MNFS 1694, 1698–1699, 1704 1714 –1716, 1719, 1723–1724; MVZ 187169, 187183; (15) 13 m, 16 f — JUR 273, 287, 290, 298, 301, 319, 324, 336, 342– 343, 356–358, 360, 377; MPEG 25501 MVZ 187170–187182.