Proechimys gardneri 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 : 232-236

Proechimys gardneri da Silva, 1998

TYPE LOCALITY: ‘‘ Brazil: Amazonas ; Altamira, right bank Rio Jurua´ ; 68°54̍W 6°35̍S.’’

DESCRIPTION: Proechimys gardneri is one of three small­bodied spiny rats within the Rio Juruá basin, with short ears, small hind feet, and proportionately long tail (tables 60 and 64). The tail is bicolored, dark brown above and cream to white below; the scales are relatively small, but not completely hidden by the hair. The overall color of the body is between Sanford’s Brown and Auburn (Ridgway, 1912), coarsely streaked with varying amounts of black both on the dorsum and sides; as with other species of spiny rats, the dorsum looks darker, especially on the rump, due to the presence of the heavy dark brown aristiform hairs. The venter and chin are pure white; in 12 out of 26 specimens the sides of the upper lips, sometimes confluent with a spot at base of vibrissae, are also white; the pure white of the inner surface of the hind limbs extends across the ankle along the hind foot in 7 of the 13 Rio Juruá specimens (the dark tarsal bank is incomplete) e color of the hind foot is yellowish­white rather than the pure white of the venter and inner thighs. In some specimens

the first and second digits of the hind foot, in combination or not with the distal portion of the digits, are brownish (for more information about the external morphology of this species, especially of young individuals from the Rio Juruá and elsewhere, see da Silva, 1998).

The baculum is massive and relatively long, especially in relation to the average body size, with short, broad, and distolaterally directed apical extensions separated by a shallow medium depression (fig. 137). The midshaft is relatively broad, the base is thick and expanded.

The skull is small and delicate, with a relatively long and narrow rostrum (figs. 138 and 139), and a beaded supraorbital ledge above the orbits that extends posteriorly as a weakly developed ridge on the anterior parietals. The postorbital process of the zygoma is obsolete (see da Silva, 1998: fig. 11). The floor of the infraorbital foramen is smooth, lacking a ventral canal. The incisive foramen is ovate to slightly lyrate in shape, with posterolateral margins flat or weakly flanged, at best outlining only a shallow groove in the anterior palate (fig. 140). The maxillary portion of the septum is dorsoventrally compressed posteriorly and narrow anteriorly, visible over almost half the length of the foraminal opening, and fully connected to the premaxillary portion, which is broad and usually about half the length of the foramen. The vomer is not visible on the ventral margin of the septum. The palate is smooth, without a median ridge. The mesopterygoid fossa is long and narrow (fig. 141), with an acute angle of indentation into the posterior margin of the palate averaging 61°, and penetrates deeply, often to the middle of M2. The cheekteeth are remarkably small, averaging only 7.5 mm in length. All upper teeth typically have three folds, although only two folds may be found on M 3 in some individuals.

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

COMPARISONS: This species can be confused only with the two other small­bodied taxa along the Rio Jurua´, P. kulinae and P. pattoni (figs. 135 and 136). Morphologically, P. pattoni and P. gardneri are very difficult to distinguish in that both have a short tooth­ row length, similar general pattern of the incisive foramen (although with much individual variation), ventral color of tail less brilliantly white than, for example, P. simonsi , but tail still bicolored in most specimens and more so in P. gardneri than in P. pattoni ; and relatively narrow, small white hind foot with a dark ring around the ankle. Subtle differences between them, however, are present For example, P. pattoni is slightly smaller than P. gardneri ; the aristiform hairs of P. pattoni are narrower but stiffer to the touch; the postorbital process of the zygomatic arch of P. pattoni is slightly but consistently more spinose; the maxillary and premaxillary portions of the incisive foramen are either not touching or are connected by a very attenuate keel in 6 of 11 specimens of P. pattoni (on the remaining specimens the maxillary portion was expanding in a spatulate shape in the area of contact between it and the premaxillary bone), whereas in P. gardneri the maxillary and premaxillary portions are in contact in all but one of the 32 specimens examined. In P. pattoni , the color of the dorsum and ventral surface of the tail does not contrast as sharply as in P. gardneri ; in fact the ventral side of the tail tends toward brown in several specimens and has relatively larger scales, and less hair than in specimens of P. gardneri . In addition to these external differences, the shape of the baculum (fig. 137) and the external morphology of the glans is clearly distinct between these two species (see da Silva, 1998) Comparisons to P. kulinae are given below in the account of that species.

MOLECULAR PHYLOGEOGRAPHY: The limit­ ed data on this taxon suggests some geographic substructure for the available cytochrome­b haplotypes. Specimens from the central Rio Juruá (Altamira [locality 9]) differ only by 2%, on average, from individuals obtained at two localities in northern Bolivia some 300 km to the south (fig. 146; table 70); those from the upper Rio Urucu, 500 km to the northeast, differ by only a slightly greater amount, at 2.2%.

DISTRIBUTION AND HABITAT: This species is known from only two localities in western Brazil and one in northern Bolivia (fig. 146, top). The distribution is perhaps delimited by the Rio Juruá on the west and the Rio Ma­ deira to the east. Specimens from the Rio Juruá are all from the terra firme forest at Altamira (locality 9) (table 63), with most (17 of 21) coming from the standardized trap plot. The majority (43%) of our sample was caught in Sherman traps, 38% came from Tomahawk traps, 14% were shot, and 5% was caught with Victor snap traps. Of the animals captured in both types of live traps, young and subadults (n = 8) were taken in Sherman traps while most adults (8 out of 9) were captured in Tomahawk traps. The high­ er proportion of adult animals in the larger Tomahawk traps may be associated with the behavior of these animals since the relatively small adult size (maximum weight about 270 g) should not show bias against use of the smaller Sherman traps.

REPRODUCTION: We obtained all specimens of P. gardneri near the beginning of the rainy season in the month of November. All females and all but one male were adults (age class 8 to 10); the single nonadult male was a young animal of age 2. All adult males (7) were reproductively active and 8 of the 13 females were pregnant, all of which were adults (age classes 8 to 10). The modal litter size was 2, range 1–3. Data are inadequate to judge seasonality of reproduction, but the one young individual and the large number of pregnancies clearly indicates that this species breeds at least during the dry season.

KARYOTYPE: We have karyotype data from 14 individuals from the type locality at Altamira, and from 6 individuals from the upper Rio Urucu. The diploid number is 40, the fundamental number 56, and the karyotype itself is figured in da Silva (1998). The autosomal complement includes seven pairs of medium­sized to small metacentrics and submetacentrics with one pair minute, two pairs of moderately small subtelocentrics, the smallest of which bear secondary constrictions on the long arms, and three pairs of medium­sized and seven pairs of small acrocentrics. The X­chromosome is a moderately small acrocentric and the Y is a small acrocentric. This complement is identical to that of P. pattoni (see Patton and Gardner 1972; da Silva, 1998), a feature that is concordant with the apparent sister relationship between these two species based on mtDNA sequence data (see above and fig. 134).

SPECIMENS EXAMINED (n = 21): (9) 8 m, 13 f ( INPA 2565–2569 ; JLP 16039; JUR

192; MNFS 853–854; MPEG 25512–25516; MVZ 187203–187209).