Proechimys pattoni 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 : 238-239

Proechimys pattoni da Silva, 1998

TYPE LOCALITY: ‘‘ Brazil: Acre ; Igarapé Porongaba, right bank Rio Jurua´ , 72°47̍W, 8°40̍S.’’

DESCRIPTION: Proechimys pattoni is the third of the small spiny rats presently known from the Rio Jurua´. Individuals are slender, have relatively short ears and tail, and small hind feet (tables 60 and 64). The skull is small, relatively narrow with a pointed snout; the maxillary toothrow is very short (less than 7.5 mm) and the individual teeth appear especially tiny. The dark brown dorsal surface of the tail grades evenly, rather than contrasting sharply, with the paler brown to cream color of the ventral surface; the scales on the tail are relatively small (approximately 10 to 12 annuli per cm), but visibly conspicuous. Overall, the color of the body varies between Sanford’s Brown and Auburn (Ridgway, 1912), but is coarsely streaked with varying amounts of black both on the dorsum and sides; the interspersed thick dark brown aristiform hairs of the dorsum give it a somewhat darker aspect, but the contrast between the color of the dorsum and sides is not sharp. The color of the venter and chin is pure white as are the sides of the upper lips; a white spot is present at the base of vibrissae in most specimens. The white color of the inner surface of the hind limbs is interrupted by a dark ring around the tarsal joint. The juvenile pelage is uniformly grayish brown (age class 3); one specimen of age class 6 and one of age class 7 have adult, aristiform hair throughout the midback, and soft juvenile hair streaked with black and fulvous tips on the sides, shoulders, and rump. The dorsal surfaces of the hind feet are entirely white in most specimens, although in some the entire hind foot seems more golden than pure white or is slightly brownish on the sides, including the third and fourth digits, or across the entire middorsum. There are six tubercles on the plantar surface of the hind feet in most specimens (11 of 14), with a very long medial metatarsal tubercule (mmt) extending from about one to two thirds the distance between the calcaneus and the first predigital tubercule.

The baculum is massive in proportion to body size. It has a broad shaft, a thick and expanded base, and a long pair of divergent apical extensions separated by a wide and deep median depression (see fig. 137, and illustrations in Patton and Gardner, 1972, and da Silva, 1998).

The skull is small and rather delicate for a Proechimys (figs. 138 and 139), with overhanging supraorbital ledges but with only weakly developed beading extending onto the temporal region. A low but distinct postorbital process of the zygoma is present, usually formed solely by the squamosal. The floor of the infraorbital foramen is smooth lacking a groove or lateral flange. The incisive foramen is ovate to slightly squarish with flat posterolateral margins, an attenuate or dorsoventrally compressed maxillary portion of the septum, and a broad and short premaxillary portion, which usually is not in contact with the maxillary portion (fig. 140) The foramen of a specimen from eastern Perú was described and figured by Patton (1987: fig. 14d). The palate is smooth, without a median ridge. The mesopterygoid fossa is long and narrow (fig. 141), angle of indentation on posterior margin acute (50°– 60°), and may penetrate as far as the middle of M2. There are typically three folds on each of the upper cheekteeth, although occasional specimens have only two folds on PM4 and/or M3. Additional characters of the skull and phallus are described by da Silva (1998).

SELECTED MEASUREMENTS: Means and ranges of selected external and cranial measurements are given in table 64.

COMPARISONS: Table 60 compares characters of P. pattoni with those of other species from the Rio Jurua´. This species can be recognized by its small body size and relatively short tail, whitish feet, and very short toothrow; but these are all characters shared with both P. gardneri and P. kulinae . Indeed, as noted above under the accounts for these oth­ er two species and as emphasized in the multivariate analyses (figs. 135 and 136), all three are quite difficult to distinguish, except by the male phallus and baculum (fig. 137 see also da Silva, 1998: fig. 5), subtle differ­

ences in cranial characters, and karyotype ( P. kulinae from both P. pattoni and P. gardneri ). The individual cheekteeth appear distinctly smaller in pattoni than in the other two, although P. kulinae has the shortest toothrow (table 64) and significant differences in this character occur in comparisons between any pair of these three species (oneway ANOVA comparing all three, F 2,61 = 19.068, p <0.0001; Fisher’s PLSD p ­values in pairwise comparisons, <0.0001 [ gardneri vs kulinae ], 0.0192 [ gardneri vs pattoni ], and 0.0234 [ kulinae vs pattoni ]). The dark band around the ankle, the typical lack of contact between premaxillary and maxillary portions of the septum of the incisive foramen, the low and rounded but distinct postorbital process on the zygomatic arch, and the narrower and more deeply penetrating mesopterygoid fossa are a few external and cranial characters that help to distinguish P. pattoni from the other two species (table 60).

MOLECULAR PHYLOGEOGRAPHY: Cytochrome­b sequence data are available for only two localities in the Headwaters Region of the Rio Juruá (Igarapé Porongaba [locality 1] and Sobral [locality 4]; fig. 146, table 70). There is, however, as much variation among the five individuals from Porongaba as there is between any of these and the single specimen from Sobral. Given the close proximity of the two localities, even though they are on opposite sides of this river, the lack of demonstrable differentiation between localities should not be surprising. The average difference between all six individuals sequenced is only 1.3%.

DISTRIBUTION AND HABITAT: Proechimys pattoni is currently known from only five localities in western Amazonia: two in the Headwaters Region of the Rio Juruá (Porongaba, locality 1, and Sobral, locality 4) in Estado do Acre, Brazil, and three in eastern and southeastern Perú in the departments of Ucayali, Madre de Dios, and Puno (see da Silva, 1998, and figs. 132 and 145). All specimens from the Rio Juruá (n = 29) were collected in terra firme forest, five of these in disturbed areas dominated by bamboo. We captured P. pattoni using live traps placed on the ground close to fallen logs, at the base of trees, underneath dense ground cover, or in open forest understory; 59% of the spec­ imens were captured with Sherman and the remaining with Tomahawk traps. All but one young along with all subadult individuals (n = 6) were caught in Sherman traps, with most adults obtained (7 out of 11) in Tomahawk traps.

REPRODUCTION: We caught all specimens of P. pattoni during the rainy season in the months of February and early March. Since this species does not occur in downriver sites, no information is available regarding seasonality of reproduction on the Rio Jurua´ Of the 10 male specimens collected, we classified five as reproductively active, all of which were adults of age classes 8 to 10. One individual of age classes 9 was reproductively inactive. Four of the 19 females were pregnant; their ages ranged from toothwear class 6 to 10. No female was lactating. The modal litter size is 2, with the range 1–2.

KARYOTYPE: 2N = 40; FN = 56. This species has the same karyotype as P. gardneri (see above and da Silva, 1998). The karyotype of P. pattoni was also illustrated and described by Patton and Gardner (1972) un­ der the name P. guyannensis .

COMMENTS: Specimens of P. pattoni from Balta were referred by Patton and Gardner (1972) to P. guyannensis , whereas Patton (1987) placed them provisionally in his P. cuvieri species group. As discussed above this species and P. gardneri are extremely similar morphologically and are unusual among species in the genus in sharing identical karyotypes. However, da Silva (1998) recognized them as separate species because each forms a well supported and highly differentiated haplotype clade (figs. 134 and 146) and each is morphologically diagnosable, although their character differences are rather subtle. Each also has a relatively wide but nonoverlapping geographic range (fig 146). Future studies to determine areas of contact between the two species in western Amazonia will be the ultimate test to the degree of evolutionary independence of these two very similar yet distinct lineages.

SPECIMENS EXAMINED (n = 29): (1) 10 m, 18 f — INPA 2559–2562, 2564; MNFS 1087–1088, 1096, 1111, 1131, 1167, 1290– 1291, 1311, 1358; MPEG 25507–25511; MVZ 187194–187196, 187198–187202; (4) 1 f — MVZ 187197.