Oryzomys yunganus Thomas 1902

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 : 156-159

Oryzomys yunganus Thomas 1902

TYPE LOCALITY: ‘‘Charuplaya, 1350 m,’’ Río Secure´, Departamento de Cochabamba Bolivia, 16°36̍S, 66°37̍W (see Musser et al. 1998: 52).

DESCRIPTION: This is the second species of Oryzomys found within the Rio Juruá basin with the derived cephalic arterial pattern. As noted above and as detailed by Musser et al (1998), O. yunganus is extremely similar in both external and cranial features to O. perenensis , but can be distinguished from that species by a combination of pelage and cranial qualitative features (table 38). However

as also emphasized by these authors, careful examination of voucher specimens will likely be required to distinguish O. yunganus and O. perenensis . Certainly, identification of these in the field during live­trapping studies cannot be readily accomplished, and will require considerable experience with both taxa. See tables 38 and 40 and figures 99 and 100.

SELECTED MEASUREMENTS: Means, standard errors, and ranges of selected external and cranial dimensions are given in table 39.

GEOGRAPHIC VARIATION: Samples of O. yunganus are limited in size, but available for most localities and, thus, all geographic regions along the length of the Rio Jurua´. The pattern of nongeographic variation due to age and sex is of the same nature as described above for O. perenensis , with virtually no contribution of sex to the overall variation (average of 1.2% across the four external and 20 cranial variables; table 44). Not surprisingly, age exhibits a much greater proportion to the total pool of variation, averaging 35.7% across all variables for adult individuals. Also as was characteristic for O. perenensis , there is virtually no among­locality differentiation detectable, with only 3.1% of the variation referable to this component. As a consequence, it is not possible to distinguish among any of the samples of O. yunganus , whether comparisons are made along or across the river. Again, this species is morphologically uniform throughout the Rio Juruá basin, an observation fully concordant with our limited molecular haplotype data.

MOLECULAR PHYLOGEOGRAPHY: While our sampling for molecular variation is not nearly as extensive as it was for O. perenensis (above, and Patton et al., 1996a), we have examined 25 individuals from 12 localities. Eleven separate haplotypes were recovered from sequences of the initial 414 bp of the cytochrome­b gene. An unrooted network of these is shown in figure 84. A few haplotypes are reasonably divergent from all others, differing by six or seven steps from adjacent ones in the network, but most differ only by one or two steps. The average number of steps between adjacent haplotypes is 3.2, which is, however, nearly three times that observed for haplotype variation in O. perenensis (Patton et al., 1996a, as O. capito ). Nev­ ertheless, this still represents an average divergence of less than 1%. Viewed geographically, there is some apparent structure along the river, but none related to samples from opposite banks. With the exception of a single haplotype from locality 2 and another from locality 12, there is an exact correspondence between position of haplotypes in the network and the geographic placement of the localities from which they were sampled (fig 105). One haplotype is broadly shared between the Headwaters localities of Nova Vida (locality 3) and Sobral (locality 4) as well as all of those from the Upper Central Region (Condor, Penedo, and Nova Empresa [localities 6, 7, and 8, respectively]). The remainder of those haplotypes from Headwaters sites are closely related to this broadly

distributed one, with the single exception from the left bank opposite Igarapé Porongaba (locality 2), noted above. This group shares a much greater degree of similarity among themselves than do the cluster of haplotypes from either Lower Central or Mouth sites. It is also noteworthy that only two of the four haplotypes found in upriver sites are apparently limited to single localities, while each of the seven downriver haplotypes are confined to a given site. Samples equivalent to those of O. perenensis would be required for proper hierarchical comparisons between these two species, but the limited data suggest that O. yunganus contains more divergent haplotypes that exhibit greater geographic structure along the river. The greater reliance on terra firme habitats, the lower reproductive potential, and the smaller population sizes estimated from trapping data for O. yunganus (see below) could contribute to the types of differences in molecular population characteristics we have observed between this species and O. perenensis .

DISTRIBUTION AND HABITAT: This species is not as abundant as O. perenensis at any given locality, nor within the entire Rio Juruá basin. It was captured, however, at one or more sites within each of the four regional areas sampled, from the headwaters to the mouth of the river. It was also present in all habitats except the river­edge grasses and Cecropia On balance, it was more common in terra firme forest than in várzea, the opposite of the pattern of habitat usage exhibited by O perenensis (table 43). The difference in the relative abundances of these two species in different habitats is highly significant (X 2 = 39.533, p <0.001), although there is also heterogeneity in habitat occurrence across the sample sites for each species.

REPRODUCTION: We trapped reproductively active males (those with scrotal testes and enlarged vesicular glands) at all sites where the species was present and at all seasons over the year of our sampling. Pregnant, lactating, or postlactating females were also obtained at all sites; therefore, at least some females breed in all seasons over the year. The modal litter size was 2 (range 1–4, n = 14) however, with the mean litter size significantly lower than that for O. perenensis (t = 3.191, df = 13, p = 0.0071). Moreover while more than 50% of all reproductively active females of O. perenensis were pregnant at any single sampling period, pregnancy rates were much lower for O. yunganus The proportion of reproductively active (pregnant, lactating, or postlactating) females in each toothwear age class was also significantly different between the two species (X 2 = 31.769, df = 4, p <0.001), with O yunganus apparently delaying breeding to an older age, on average. This is not true for males, however, where both species exhibit the same proportions of breeding and nonbreeding individuals in each age class (X 2 = 2.536, df = 4, p = 0.469). Consequently, the differences in modal litter size, apparent pregnancy rates, and longer time to reproductive maturity in females may explain, at least in part, the differences in relative abundances of these two species of Oryzomys within the Rio Juruá basin. This may also be a general pattern, as O. perenensis has been uniformly much more commonly taken at localities where both species are sympatric (Musser et al., 1998).

KARYOTYPE: Chromosomal preparations were made from 20 individuals, four from localities in the Headwaters Region, nine from the Upper Central Region, one from the Lower Central Region, and six from the Mouth Region. The diploid number is uniformly 58, the fundamental number 62. The autosomal complement comprises 25 pairs of large to small acrocentric and three pairs of small metacentric elements. The X­chromosome is a medium large acrocentric and the Y­chromosome is a small acrocentric chromosome. This karyotype is the same as that reported for specimens from eastern Perú by Gardner and Patton (1976).

SPECIMENS EXAMINED (n = 138): (1) 9m, 4f — MNFS 1093, 1101, 1171–1172, 1194, 1228, 1265–1267, 1295, 1323, 1380, 1401; (2) 4m, 4f — MNFS 1181, 1240, 1245, 1247, 1275, 1303, 1347, 1389; (b) 1m — MNFS 1004; (3) 6m, 7f — JUR 212, 214, 227, MNFS 1555–1556, 1588, 1608, 1610– 1611, 1630, 1650, 1652, 1657; (4) 1m, 1f — MNFS 1444, 1455; (6) 7m, 3f — JLP 15523, 15535, 15571, 15605, 15650, 15720–15721, MNFS 527–528, 553; (7) 32m, 16f, 1 unknown — JLP 15242, 15250, 15257, 15262– 15263, 15265–15266, 15275, 15281–15283, 15288–15290, 15301, 15303, 15312, 15319, 15321, 15323–15329, 15361, 15443, 15476, 15495, 15500, 15519, MNFS 335, 341, 349, 367–368, 373, 384, 386–388, 390–392, 412, 419, 511, 519; (8) 8m, 9f — JUR 2, 5, 7–8, 12–13, 35, 37, 41, 45–46, 73–75, 110; (9) 1m, 2f — JLP 16013, 16021, 16068; (9a) 2m, 1f — MNFS 921, 923, 929; (10) 2m — MNFS 868, 918; (11) 2f — JLP 15759, MNFS 697; (12) 3f — JLP 15784, 15829– 15830; (14) 5m, 7f — JUR 440, 445, 463– 464, 474, 483, 489, 494, 509, 524, 543, 557.