< OF ILLINOIS LIBRARY AT URBANA-CHAMPA1GN eiaopv DEC 161996 •J£S>33*MCBjf|Biptans*rr n« > i» vt w ID D ECONOMY D THESIS □ ECONOMY AUTH. 1ST □ NO. VOLS. THIS TITLE COLOR 20706 ffftl M9W :s ISSN. BIX )ING WHEEL SYS. I.D. OUENCY TUBS CLOTH EXT STUB 157 HIS JOB NO. LEAF ATTACH. 2Z ! BIX "TBi 1l The Description of a New South American Hoticudo, ( Genus Oxymyctems (Sigmodont with a Critical Review of the Ge H A Philip Hershkoviti Ctober 31, 1994 ublication 1460 >se Mouse. lur^iii^a), ontent UBLISHED BY FIELD 71 The Description of a New Species of South American Hocicudo, or Long-Nose Mouse, Genus Oxymycterus (Sigmodontinae, Muroidea), with a Critical Review of the Generic Content Pictet 's hocicudo FIELDIANA Zoology NEW SERIES, NO. 79 The Description of a New Species of South American Hocicudo, or Long-Nose Mouse, Genus Oxymycterus (Sigmodontinae, Muroidea), with a Critical Review of the Generic Content Philip Hershkovitz Curator Emeritus Division of Mammals Department of Zoology Field Museum of Natural History Chicago, Illinois 60605-2496 Accepted April 19, 1994 Published October 31, 1994 Publication 1460 PUBLISHED BY FIELD MUSEUM OF NATURAL HISTORY © 1994 Field Museum of Natural History ISSN 0015-0754 PRINTED IN THE UNITED STATES OF AMERICA Table of Contents Abstract 1 Introduction 1 Abbreviations 1 Genus Oxymycterus 2 Geographic Distribution 2 Characters 2 External 2 Cranial 2 Os Rostri or Prenasal Ossification 5 Dental 5 Stomach 7 Glans Penis 9 Accessory Glands 13 Claws 13 Cytogenetics 13 Coloration and Habitat 14 Enemies 14 Relationships 14 Origin and Dispersal 15 The Species 15 Size Classes 21 Sympatry 22 Annotated Catalog of Described Forms of Oxymycterus 22 Habits and Habitat 36 Summary 39 Acknowledgments 39 Literature Cited 40 3. Diagram of first right upper molar 6 4. Diagram of first right lower molar 8 5 . Two sigmodontine stomach types: uniloc- ular-hemiglandular, bilocular-discoglan- dular 10 6. Pouched stomach lined with gastric epi- thelium in each of three diverse genera ..11 7. Glans penis of Oxymycterus paramensis 12 8. Glans penis of Oxymycterus roberti .... 13 9. Baculum of Oxymycterus roberti 14 10. Portion of male sigmodontine reproduc- tive tract 15 1 1 . Hands of non-burrowing and burrowing sigmodontines 17 12. Oxymycterus amazonicus (new species) skull and molars 18 13. Oxymycterus amazonicus, skulls com- pared 19 1 4. Oxymycterus angularis, upper and lower molars 20 15. Oxymycterus hiska, skull of paratype ... 21 16. Oxymycterus hispidus, skull and molars . . 22 1 7. Oxymycterus hispidus, unworn molars . . 23 18. Oxymycterus j uliacae, skull and molars .. 26 19. Oxymycterus nasutus, skull and molars . . 27 20. Oxymycterus paramensis, molars 28 21. Oxymycterus roberti, skull and molars . . 29 List of Tables List of Illustrations Pictet's hocicudo frontispiece 1 . Distribution of genus Oxymycterus .... 3 2. Map, type localities of nominal forms . . 4 1 . Geographic distribution and size classes ... 16 2. Measurements of type series of Oxymyc- terus amazonicus 24 The Description of a New Species of South American Hocicudo, or Long-Nose Mouse, Genus Oxymycterus (Sigmodontinae, Muroidea), with a Critical Review of the Generic Content Philip Hershkovitz Abstract The description of a new species of Oxymycterus entailed a review of the entire genus but without arriving at a definitive taxonomic revision. The many outstanding problems such as the complex dental morphology of the unworn molars, the function of the trumpet-shaped snout, the elongate front claws, and the origin, affinities, and biogeography of the genus, and of sigmodontines in general, could not be addressed within the limitations of current knowledge and available material. The size relationships of the 23 described forms are indicated in terms of size classes, from small to large. They are separated into two coordinate geographic groups, one Atlantic, the other Andean, both confined to between the south bank of the Rio Amazonas- Solimoes-Maranon system and the north bank of the Rio Parana system, with absence of hocicudos in the floodlands between. Each originally named taxon is listed with type data, including original measurements, putative geographic distribution, relevant taxonomic infor- mation, and some comparisons. What is known of habits and habitats is summarized. Notions of oxymycterine burrowing are dispelled, and generally accepted interpretations of supposed sigmodontine origins and dispersal are questioned. Introduction The following description of a new mouse of the genus Oxymycterus discovered in the lower basin of the Rio Amazonas in Brazil required a review of the genus and organization of the 23 forms de- scribed since the first by Felix Azara in 1801. Available material is inadequate for a definitive taxonomic revision, 18 of the 23 holotypes being in European museums, 1 1 of them known only from their types, and 2 without types. Neverthe- less, specimens at hand suffice for attainment of immediate objectives and organization of most of the accumulated data. Geographic boundaries of the genus (map, Fig. 1 ) are extrapolated from type localities (map, Fig. 2) and marginal localities. De- lineation of the range of each species, however, awaits taxonomic revisions based on direct com- parison of all holotypes or acceptable represen- tatives. For the present, published measurements of types and topotypes or near topotypes form the primary basis for arrangement of the material into size classes each with the appearance of a species or species-group. Recognized are three Atlantic Division size classes and three Andean Division size classes with no distinguishable intermediates. Abbreviations The following abbreviations are used for insti- tutions and terminology. fmnh = Field Museum of Natural History, Chicago mnr = Museu Nacional de Brasil, Rio de Ja- neiro mpeg = Museu Paraense Emilio Goeldi, Belem mvz = Museum of Vertebrate Zoology, Uni- versity of California, Berkeley F1ELDIANA: ZOOLOGY, N.S., NO. 79, OCTOBER 31, 1994, PP. 1-43 rnhms = Rijkmuseum Van Natuurlijke Histo- ric, Stockholm National Museum of Natural History, Smithsonian Institution, Washington, D.C. Universidade de Sao Paulo Museu de Zoologia, Sao Paulo USNM USPMZ CB = condylobasal length of skull CU = hind foot with claw E = ear from notch GSL = greatest skull length HB = length of head and body combined HF = hind foot length, with claw (cu) or without claw (su) IB = interorbital width ML = alveolar length of upper molar row SU = hind foot length without claw T = tail length ZB = zygomatic breadth Genus Oxymycterus Waterhouse Oxymycterus Waterhouse, 1 837:2 1 —subgenus of Mus. Tomes, 1861:285-genus. Tullberg, 1899-muroid and Oxymycterus anatomy; phylogeny. Thomas, 1909:236— characters; species. Tate, 1932b:616— taxonomic history; species. Gyldenstolpe, 1932: 128 -characters. Ellerman, 1941:8, 26, 317, 330, 367, 419, 420— characters; classification; species. Cabrera, 1961:465— catalog. Hooper and Musser, 1964:28— glans penis morphology; muroid classi- fication. Carleton, 1973:15, 24, 25— stomach mor- phology; relationships. Vorontsov, 1979— alimen- tary system (morphology; function; comparisons). Reig, 1987:360— tribe Akodontini (definition; ge- neric affinities; species). Musser and Carleton, 1993: 726— check list; taxonomic comments. Oxymicterus [sic], Tomes, 1861:285— generic rank. Type Species— Mus nasutus Waterhouse, by original designation. Geographic Distribution (maps, Figs. 1, 2) Hocicudos are confined to the middle latitudes of South America south of the Rio Amazonas- Solimoes-Maranon in Brazil and Peru. Eastward the range extends along the Atlantic coast of Brazil, Uruguay, the Province of Buenos Aires, Argenti- na, thence west along the northern side of the Rio Parana basin of Brazil, Argentina, Paraguay, B livia, and the northeastern or Amazonian versa of the Bolivian and Peruvian Andes to elevatio of 4000 m or more. Oxymycterus is unknown in the Amazonian ra forest or fioodlands between the Rio Tocantir or perhaps the more eastern Rio Guripi, to t eastern versant of the Andes. Excepted are inl trations of the lower reaches of the Rios Tocantii Xingu, Tapajoz and Madeira in Brazil, and hea waters of the Rio Madeira in Bolivia and Braz The genus is unknown north or west of the R Amazonas-Solimdes-Maranon. Characters External The 23 described forms of Oxymycterus ranj in size from about that of a large domestic mou (Mus musculus) to that of a large domestic r (Rattus rattus); tail from shorter (60%) to near as long as head and body combined, thinly hirsul the coarse annularly arranged scales not conceale hind foot long, stout, longer than ear measun from notch; heel hairy, sole with 6 pads, ck^ long, thin, the manual as long or longer than tl pedal; the middle manual claw longest and abo equal to combined length of corresponding ph langes; manual digit V with weak claw extendi] to slightly beyond base of IV, claw of I hard extending beyond carpals; ears and eyes not mar edly reduced; snout long, mobile, apparent adapted for rooting; general coloration agouti i dominantly brownish, blackish, reddish, orange i mottled, grayish hair bases often showing throuj on underparts, or may dominate; mammary fo mulae 2-1 = 6 or 2-2 = 8. Cranial (Figs. 12, 13, 15, 16, 18, 19, 21) Skull long, narrow; combined premaxillaries ar nasal bones often produced in front of incisors : an expanded tube or trumpet; nasals parallel-sid( to tapered behind, tips squared, length appro* mately 40% that of skull and increasing with ag cartilaginous septum of snout (os rostri) protru< ing; infraorbital foramina widely open; anterii zygomatic plate narrow, markedly reclined ar little or not visible viewed from above; zygomat FIELDIANA: ZOOLOG 40" 351 OXYMYCTERUS DISTRIBUTION 'Ft 95' 90* 85* 80" _\ S 1 I I L J. I /_ 70' G5* 60* 35" 50" 15' 40* 35* 30° 25" 20' 15" Fig. 1 . Geographic distribution of the genus Oxymycterus Waterhouse, an approximation based on type localities and marginal records. HERSHKOVITZ: SOUTH AMERICAN HOCICUDO Fig. 2. Type localities of all described species and subspecies of the genus Oxymycterus. FIELDIANA: ZOOLOGY arches slender, nearly parallel-sided or convergent anteriorly; braincase moderately inflated; inter- parietal bone greatly reduced or obsolete; supra- orbital borders may be rounded or square, more or less ridged or smooth; occipital bone often ridged, sometimes crested; palatal or incisive fo- ramina well open, produced back to about level of anterior third of first molar; palatal bridge broad, produced posteriorly from slightly in front to slightly behind posterior plane of last molars; mesopterygoid fossa wide, parapterygoid fossae narrower and shallow; bullae moderately inflated; braincase smooth, rounded, square, or with low temporal ridges that accentuate with age; molar rows parallel-sided to convergent posteriorly; mandible low, the angle lower than long. mys palustris, Peromyscus yucatanicus, and Neoto- ma cinerea among others cursorily examined. A functional relationship between presence of the ossification and shape, length, or any other mod- ification of the rostrum is not evident. My im- pression is that the ossification may be more prev- alent than where noted had it not been shorn during the skull cleaning process, perhaps because it seemed to be an adventitious growth. The function of the os rostri is unknown. It likely supports the soft tissue at the end of the snout. In Oxymycterus it may be related to rooting, but this may not be true of other taxa not known to root. In the Suidae with snout highly developed for rooting, the os rostri, an extension of the nasal septum, supports the fleshy outer tissue between the nostrils. Os Rostri or Prenasal Ossification (Figs. 15, 16A, 19A) A short bony or cartilaginous protrusion up to 2 mm in length rounds off the combined nasal bone tip in some sigmodontines. Hinojosa et al. ( 1 987, p. 5, Fig. 7 A) called attention to this feature in Oxymycterus and noted it in the photograph of a Blarinomys skull published by Matson and Abravaya (1977). A spot check of the sigmodon- tines in the Field Museum collection revealed a low number, from less than 1% to not more than 2%, in skulls of Abrothrix longipilis, Akodon ko- fordi, Oryzomys (Sigmodontomys) alfari, Oryzo- Dental (Figs. 3, 4, 12, 14, 16-21) Upper incisors opisthodont, moderately heavy; molars tetralophodont, hypsodont, inner and out- er cusps slightly oblique, the unworn outer cusps high, pointed, the inner terraced; procingulum (loph I) of m1 narrower than loph II (measured across protocone-paracone), anteromedian flexus pres- ent, the anterolabial and anterolingual conules subequal; m3 one-half or less bulk m2; paralophule and mesoloph present in unworn m1 but usually fused, sometimes present in m2; moderately worn Key to type localities of nominate species, Figure 2: 1. Fordlandia, Rio Tapajoz, Para, Brazil, 3°40'S, 55°30'W, near sea level (amazonicus). 2. Sao Lourenco da Mata, Pernambuco, Brazil, 8°00'S, 35°03'W, sea level (angularis). 3. Bahia (= Salvador), Bahia, Brazil, 1 2°59'S, 38°3 1 'W, sea level (hispidus; rostellatus). 4. Rio Mucuri, Bahia, Brazil, 18°05'S, 39°34'W, sea level (dasytrichus). 5. Paranahyba, Rio Jordao, Minas Gerais, Brazil, 18°26'S, 48°00'W, 700-900 m (roberti). 6. Roca Nova, Serra do Mar, Parana, Brazil, 25°30'S, 48°50'W, 1000 m {quaestor). 7. Joinville, Santa Catarina, Brazil, 26°18'S, 48°50'W, sea level (judex). 8. Taquara do Mundo Novo, Rio dos Linos (sic, = Sinos), Rio Grande do Sul, Brazil, 29°39'S, 50°47' W, 29 m (iheringi). 9. Maldonado, Uruguay, 34°54'S, 54°57'W, sea level (nasutus). 10. Ensenada, La Plata, Buenos Aires, Argentina, 34°51'S, 57°55'W (platensis). 11. Entre Rios, Argentina, "32°30'S" (rufus). 1 2. Rio Parana-i, near Caraguatay, Misiones, Argenti- na, 26°37'S, 54°46'W, 100 m (misionalis). 13. Sapucay, Paraguay, 25°40'S, 56°55'W, 220 m (de- lator). 14. Higuerilla, Valle Grande, Jujuy, Argentina, 23°35'S, 65°15'W, 2000 m (akodontius). 15. Carapari, Yacuiba, Tarija, Bolivia, 21°49'S, 63°46'W (jacentior). 1 6. Comarapa, 28 km W, Cochabamba, Bolivia, 1 7°5 1 'S, 64°40'W, 2800 m (hucucha). 17. Choquecamate, Rio Secure, Cochabamba, Bolivia, 16°55'S, 66°37'W, 4000 m (paramensis). 18. Charuplaya, upper Rio Secure, upper Rio Mamore, Cochabamba, Bolivia, 15°48'S, 66°30'W, 1350 m (doris). 19. Inca Mines, Santo Domingo, upper Rio Inambari, Puno, Peru, 15°30'S, 70°08'W, 1875 m (juliacae). 20. Yanahuaya, 1 4 km W, Puno, Peru, 1 4°1 9'S, 69°2 1 'W, 2210 m (hiska). 21. Limbani, Puno, Peru, 14°08'S, 69°42'W, 2810 m (nigrifrons). 22. Perene,Junin,Peru, 10o58'S,75°13'W,800m(/rtca). 23. San Ernesto, Mapiri, upper Rio Beni, La Paz, Bo- livia, 10°23'S, 65°24'W, 1000 m (iris). HERSHKOVITZ: SOUTH AMERICAN HOCICUDO NGUAL RIGHT M1 Fig. 3. Diagram of occlusal surface of a first right upper molar showing elements of the enamel pattern in all sigmodontine upper molars. Figure copied from Hershkovitz (1993, p. 12). FIELDIANA: ZOOLOGY crowns of m1-3 dished, 8-shaped, the lophules eroded; mesolophid, if present, often fused with mesostylid, if present. Remarks are present in the diagramatic representation of upper and lower first molars (Figs. 3, 4). Molars 2 and 3 show fewer, less well defined elements. A few typographical and accidental errors in the original diagrams (Hershkovitz, 1 993) reproduced here have been eliminated. The fully formed crowns of unworn molars of Oxymycterus display sharply each element of the akodontine molar. By the time the last molar has become fully functional, however, the particularly thin crown enamel of the first 2 teeth has mostly if not entirely been eroded, leaving the dentine fully exposed to the grinding, crushing forces of mastication. Without the enamel cover, the once high, pointed cusps are quickly flattened, lophs and lophids reduced to stubs, styles (ids) disap- pear, flexi (ids) become faintly outlined or reduced to mere indentations on the outer molar walls, and some may have an ephemeral existence as enamel islands of the crown. Adult molar crowns of all species become shal- low basins, the first trilaminar, the second 8-shaped, the third even further reduced. Except for size, adult molars seem to have lost taxonomic char- acter. All structural details of the unworn (ancestral?) Oxymycterus molars present at the eruptive stage Stomach New World cricetine stomachs are of the two morphological types defined by Carleton (1 973, p. 1 0). The primitive unilocular-hemiglandular type (Fig. 5A) is single-chambered with a shallow in- cisure angularis. The derived bilocular-discoglan- dular type (Fig. 5B) differs primarily by the deep incisura angularis that gives the bipartite appear- ance to the stomach. Of the 27 stomachs of representatives of South American sigmodontine genera investigated (17 figured) by Carleton, those of 20 conform to the basic unilocular-hemiglandular pattern. Notewor- thy is the stomach of Oxymycterus with the glan- dular epithelium confined to a diverticulum lo- cated on the greater curvature (Fig. 6). A minute aperture connects this glandular pouch with the lumen of the stomach. The diameter of the open- ing measured .35 mm in the specimen of O. pla- Explanation of symbols, Figure 3. Note: a-i inclusive t'. = procingulum or loph I; s, t, t', z = postcingulum or loph V. u. v. I-V. lophs of pentalophodont molars. w. a. anteromedian style (may be fused with b, c, or x. both). y. b. anterolabial lophule. z. c. anterolingual lophule. a'. d. anterolophule (may be fused with f ). e. anterior fossette. b'. f. plesiostyle (may be fused with d, h, or j). g. protostyle (may be fused with i). c'. h. anteroloph (may be fused with f, j, or both). d'. i. protoloph (may be fused with g, u, or both). e'. j. parastyle (may be fused with f, h, k, or a combi- nation). 1 . k. mesolophule (may be fused with h, j, or both). 2. 1. paracone. 3. m. paralophule (may be fused with n, o, or both); el- 4. ement may be multiplied. 5. n. mesoloph (when fused with o = mesolophostyle). o. mesostyle (may be fused with m, p, or both; when 6. fused with n = mesolophostyle of pentalophodont 7. molar). p. metalophule (may be fused with o, n, or both). 8. q. metacone. 9. r. posterolophule (may be fused with s). 1 0. s. posterostyle (may be fused with r, t, or both). t. posteroloph (may be fused with s). 11. posteroconule (may be fused with z, usually not differentiated from posteroloph, t). protolophostyle (may be fused with i). protocone. enteroloph (may be fused with x). enterostyle (may be fused with w). hypocone. distostyle. median fossette (may be coalesced with 3, or united with 4). posterior fossette (may be coalesced with 5, or unit- ed with 6). protolophule. hypolophule. mure (border between lingual and labial cusps and lophs). preflexus (anterior median fold), anteroflexus (anterior secondary fold), paraflexus (first primary fold), mesoflexus (first secondary fold), metaflexus (in absence of mesoloph [n] coalesced with first secondary fold [4]). posteroflexus (second secondary fold), supraflexus (anterior lingual fold; in absence of pro- toloph coalesced with first minor fold [8]). protoflexus (first minor fold), entoflexus (major fold). hypoflexus (in absence of enteroloph coalesced with major fold [9]). distoflexus (second minor fold). HERSHKOVITZ: SOUTH AMERICAN HOCICUDO LINGUAL I IV V © Ml (w) LABIAL RIGHT Mi Fig. 4. Diagram of occlusal surface of a first right lower molar showing elements of the enamel pattern in all sigmodontine lower molars. Figure copied from Hershkovitz (1993, p. 16). FIELDIANA: ZOOLOGY tensis and .35 mm and .50 mm in examples of O. rutilans(= O. rufus). Data are from Carleton (1973, p. 1 5). The bilocular-discoglandular type stomach of Peromyscus mexicanus is also characterized by a similar pouched stomach with the glandular ep- ithelium confined to the diverticulum. Its gastric secretion is discharged to the lumen through a tiny aperture (Fig. 6). The presence of a pouched diverticulum in a muroid stomach was first noted by Tullberg ( 1 899, p. 249) in an adult female Oxymycterus rufus. Her stomach measured 40 mm; small intestine, 190 mm; large intestine, 110 mm; caecum, 25 mm; inner surface of stomach entirely covered with stratum corneum; glands concentrated around the great curvature of the stomach wall formed the pocket, which opened into the lumen through the miniscule aperture. The diverticulum surrounded by the glandular epithelium is figured by Tullberg (1899, PI. XLI, Figs. 23, 24). The pouched stomach of a specimen of Oxy- mycterus nasutus was described and compared by Vorontsov ( 1 979, p. 1 84) with that of the pouched North American grasshopper mouse Onychomys with a bilocular-discoglandular type stomach. It is described as sacciform, fornix ventriculi . . . not marked and the main stomach-chamber . . . lined from esophagus to duodenum with corneous epithelium of esophageal origin. At the base, opposite to the opening between the esophagus and the stomach, there is a small ap- erture leading into an isolated chamber, the glandular diverticulum. The entire wall of the diverticulum is lined with considerably high fundic glands. The ap- erture of the diverticulum (as distinct from that of Onychomys) is on the left edge of its upper wall and is surrounded by a circular system of muscles forming a sphincter. As in Onychomys, the diverticulum of Oxymycterus presents a gigantic gland in which gastric juice is produced. This juice is periodically supplied to the corneous portion by opening the sphincter. Here the juice breaks down the proteins. Unlike Onychomys, the pyloric portion in Oxy- mycterus does not have any special corneous promi- nence for grinding the chitinous residue of insects and is not separated from the remaining part of the stom- ach. The pyloric sphincter muscles are not so prom- inent as in Onychomys. The microscopic anatomy of the stomach of Ox- ymycterus rutilans (= O. rufus) was studied by Echave Llanos and A. Vilchez (1964, p. 187). Glans Penis (Figs. 7, 8) The glandes penes of Oxymycterus paramensis and O. rufus have been described by Hooper and Musser (1964, p. 28). Both species, they found, Explanation of symbols. Figure 4. Note: a-h inclusive = procingulid or lophid I; a', r, r' = postcingulid or lophid V. I-V. lophids of pentalophodont molars. a. anteromedian stylid (may be fused with b, c, or both). b. anterolabial conulid (may be fused with c). c. anterolingual conulid (may be fused with b). d. labiolophulid (may be fused with g). e. anterior fossette. f. anterolophulid (may be fused with h, i, or both). g. prostylid (may be fused with d). h. anterolophid (may be fused with f, i, s, or com- bination), i. anterostylid (may be fused with f, h, s, or combi- nation), j. protoconulid (may be fused with g). k. protoconid. 1. paralophulid (may be fused with m, n, or both), m. ectostylid (may be fused with 1, n, o, or combi- nation), n. ectolophid (may be fused with 1, m, o, or combi- nation), o. hypoconulid (may be fused with m or n). p. hypoconid. q. posterolophulid (may be fused with r'). r'. posterolophid (may be fused with q). r. posteroconulid (usually not differentiated from posterolophid, r'). s. mesolophulid (may be fused with h). metaconid. metalophulid (may be fused with v, w, or both), mesolophid (when fused with w = mesolophostylid of pentalophodont molar). mesostylid (when fused with v = mesolophostylid). entolophulid (may be fused with v, w, or both), entoconid. distolophulid (may be fused with a'), posterostylid. median fossetid. posterior fossetid. protolophulid. hypolophulid. murid (zone between lingual and labial lophids and cuspids). preflexid (anterior median fold), anteroflexid (anterior lingual fold; in absence of anterolophid [h] coalesced with 3). metaflexid (first secondary fold), mesoflexid (first primary fold), entoflexid (second secondary fold; in absence of mesolophid [v], coalesced with 4). posteroflexid (second primary fold), supraflexid (anterior labial fold; in absence of la- biolophulid [d] coalesced with 8). protoflexid (first minor fold), ectoflexid (major fold). hypoflexid (in absence of ectolophid [n] coalesced with 9). distoflexid (second minor fold). HERSHKOVITZ: SOUTH AMERICAN HOCICUDO incisura angularis anterior end of duoderun ,■ glandular epithelium •fornix ventricularis UNILOCULAR - HEMIGLANDULAR A comified squamous epitheliun isura angularis esophagus anterior end of duoderun bordering fold glandular epithelium fornix ventricularis -//— comified squamous epithelium BILCCULAR - DISCCGLANDULAR B Fig. 5. Diagrams of two stomach types of New World sigmodontine rodents. A. Unilocular-hemiglandular of a South American complex penis type Oligoryzomys nigripes (Sigmodontinae). B. Bilocular-discoglandular of a North American simple penis type Neotomodon alstoni (Peromyscini). Illustrations from Carleton (1973, p. 1 1, Fig. 1) with labels added. 10 FIELDIANA: ZOOLOGY HERSHKOVITZ: SOUTH AMERICAN HOCICUDO 11 medial mound lateral mound lateral digit urethral process part prepuce folded back and cut medial digit crenate crater rim crater urethral process bacular shaft bacular base Fig. 7. Glans penis of the paramo hocicudo, Oxymycterus paramensis Thomas: Internal anatomy exposed by midventral excision; enlargement of spine-studded urethral process shown in inset; skin of entire glans similarly spiny. Figure copied from Hooper and Musser (1964, Figs. 5d, e) labels added. were phallically well differentiated from the other 1 5 genera of Neotropical sigmodontine rodents studied in detail. The Oxymycterus they described approach species of Thomasomys and Rhipidomys in some respects (e.g., shape of bone and external ap- pearance), and their relatively small lateral bacular mounds and segments are reminiscent ofAkodon, but they do not match any of those. Diagnostic features of the glans include: the crenate crater rim with essentially no bordering band of spine- less tissue; dissimilar bacular mounds, the short, erect, peaked, lateral pair contrasting with the laterally com- pressed medial one; and the robust spinous urethral flap. The distal segments of the baculum are short relative to the bone, the lateral pair particularly slight although slightly osseous. Each erect lateral digit is approximately one-half the length of the bent medial one. O. paramensis.— Gland oblong ... spinous, and without parotoid lobes but with six slight distal lobes, the pairs of three separated mid-dorsally and mid- ventrally by deep longitudinal troughs; the corre- spondingly hexalobate crater rim with essentially no ring of nonspinous tissue; bacular mounds dissimilar, the slight, peaked, lateral pair appressed tightly against the longer, deeper-than-wide medial one, its tip bent ventrad; dorsal papilla a low mound with at least two spines, one apically and one subapically; urethral flap robust (length and width about equal) and deeply cleft . . . the two erect attenuate processes spinous apically and ventrally (the spines concentrated in two rows). Bone of baculum [Fig. 9] slim in ventral view . . . with a narrow proximally projecting base (its ventral face almost flat and its dorsal face narrowly concave between lateral condyles) and a long terminally en- larged shaft; bone in lateral view more robust and gently curved dorsad, its configuration similar to that in Rhipidomys and Thomasomys; medial digit a slightly deeper-than-wide, slim, cartilaginous rod, its tip in- clined ventrad; each lateral digit also rod-like and erect, but smaller in diameter and height and with osseous tissue basally. O. rufus. —The specimens ofrufus are much like the example of paramensis except that they are larger . . . , the lateral mounds appear to be slightly smaller rel- ative to the medial one, the dorsal papilla more spi- nous (as many as three dorsal spines), the urethral flap also more densely spinous and longer (extending to distal limits of lateral mounds), the proximal face of bone slightly different, and the medial digit flexed more strongly ventrad. 12 FIELDIANA: ZOOLOGY •••-.. <#j g - fe • . '-; :- . a "■■■■< q-^^ ' A B Fig. 8. Glans penis of Robert's hocicudo, Oxymycterus roberti Thomas (fmnh 128320). A, ventral aspect; B, right lateral aspect; C, enlargement of spine-studded fragment of glans epidermis. Accessory Glands (Fig. 10) The glands of Oxymycterus delator (4 speci- mens) and O. rutilans (= O. rufus, 6 specimens) have been described by Voss and Linzey (1981, p. 14) as follows: Two pairs of preputial glands are present: the large lateral glands extend well beyond the ventral flexure of the penes while the more medial and ventral glands are considerably smaller and do not exceed the pre- puce in length; the medial glands are the larger of the two ventral prostatic pairs. Vesiculars [in both species] are lobed medially and along their greater curvatures . . . The subterminal flexures of the vesiculars of O. delator are rounded and smooth but they are irregu- larly lobed in some examples of O. rutilans [= O. rufus]. Claws (Fig. 11) Hand and foot claws of Oxymycterus range from 2.5 to 8 mm long measured in a straight line. The manual claws may be about 2 mm longer to shorter and usually thinner than the pedal claws. Because of the long manual claws, hocicudos have been thought to be burrowers. In the original description of Oxymycterus, Waterhouse ( 1 837, p. 21) described the claws of Mus nasutus as "long but slightly curved and formed for burrowing." The species, however, has never been witnessed burrowing or living in burrows. The genera of long-clawed truly fossorial sig- modontines are Geoxus Thomas, Chelemys Tho- mas, Notiomys Thomas, and Pearsonomys Pat- terson. These, all endemic to southern Chile and Argentina, have been described and compared by Patterson (1992b). The eastern Brazilian Kunsia Hershkovitz and Thaptomys Thomas complete the roster of true burrowing sigmodontines. Their claws are comparable to those of the non-burrowing Oxymycterus. Adaptation of certain akodonts to fossorial life as exemplified by Kunsia has been described by Hershkovitz (1966, pp. 87-95). Cytogenetics The karyotype of Oxymycterus paramensis, O. rufus, O. platensis (?= O. rufus), O. nasutus, and HERSHKOVITZ: SOUTH AMERICAN HOCICUDO 13 A B Fig. 9. Baculum (penis bone) of Robert's hocicudo. A, ventral aspect; B, right lateral aspect. scribed on the basis of either the eumelanic or phaeomelanic dominance of a single specimen, al- though both colorations may be present in the same population. Underparts may be dominantly grayish (eumelanic) or dominantly orange (phaeo- melanic), with the latter usually overlying the oth- er. Correlation between coloration of these mice and their habitat is not evident in present material. Activity of any species may be mostly nocturnal, diurnal, or both, midday excluded. Concealment, however, depends on the regular availability of suitable cover day or night, whether in caatinga, cerrado, or rain forest. Color differences between sympatric species reveal no pattern that to our limited knowledge can be correlated in either sym- patriot with habitat or behavior. Enemies Hocicudos, no less than others of their nature, are fair game for any predator of small mammals. Included among the most predaceous are the four- eyed opossums, Philander and Metachirus. Ho- cicudos themselves are also predaceous and can- nibalistic. O. angularis has 54 chromosomes (Vitullo et al., 1986) with no notable differences between each, the C- and G-banding patterns identical, according to Vitullo. The karyotype of Oxymycterus akodontius, ex- amined by Kajon et al. ( 1 984), has the same karyo- type, its fundamental number 64. The autosomes include 2 large subtelocentric pairs, 44 acrocen- trics of decreasing size, and 3 metacentric pairs. Examination of additional complements including that of O. rufus from Entre Rios, Argentina, and others believed to represent O. hispidus, all with identical karyotypes, adds to an impression that Oxymycterus may be cytogenetically monomor- phic. Coloration and Habitat The species vary from dominantly dark brown or blackish agouti to reddish or orange agouti on dorsal surface and sides. The molt is from blackish to reddish, often mottled. Species have been de- Relationships An oxymycterine generic group, presumed monophyletic, constructed by Hershkovitz (1966, p. 86), consisted of Oxymycterus, Podoxomys, Lenoxus, Microxus, and Abrothrix. Natural co- herence of the group has since been questioned. Reig (1987, p. 361) admitted the genus Lenoxus but only as "an exaggerated Oxymycterus in size and skull morphology." The short-clawed Abro- thrix, he believed, belonged with Akodon. Podox- omys and Microxus also appeared to be out of line, despite the opinion of Ellerman (1941). Hinojosa et al. (1987) also questioned the soundness of the generic assemblage as a tribe. The authors defi- nitely excluded Microxus but would add Blari- nomys and Geoxus. Neither of the last two is, in my opinion, related to Oxymycterus or to each other. Perez-Zapata et al. (1992, p. 220) pointed to the closer relationship between Microxus and Podoxomys than between either and Oxymycterus. Geographic isolation of Oxymycterus south of the Amazonian boundary also weighed against as- sumption of a close relationship to Podoxomys. 14 FIELDIANA: ZOOLOGY Accumulated knowledge of Oxymycterus indi- cates that what may have appeared to be the co- hesive features of a natural group were, for the most part, convergent characters among unrelated akodontines. Only Lenoxus has the semblance of a sister genus but is manually short-clawed. Claws figured by Hinojosa et al. (1987, p. 12) reveal the long claws of Oxymycterus inca con- trasted with short claws in non-oxymycterine Po- doxomys, Microxus, and Lenoxus. Correlated with the short claws are the rounded or pointed nasal tips {op cit., p. 13) compared with the squared nasal tips of Oxymycterus inca. Origin and Dispersal (Fig. 1) Oxymycterus has no known near living or ex- tinct relative north of the Rio Amazonas-Soli- moes-Maranon boundary. The geographic divi- sion indicates a far southern latitudinal origin with dispersal northward ultimately halted by the Am- azonian boundary and the upper Andean tree line. The dispersal must have been from a southern core with an eastern branch on the Atlantic coastal plains and highlands and a western branch along the ris- ing Andes with limited infiltration of marshes be- tween the divergent highland branches. Extensive floodlands are barriers to the nonaquatic Oxy- mycterus. Removal of the manually short-clawed Microx- us and Podoxomys from the illusory oxymycterine generic group eliminates all signs of an oxymyc- terine presence north of the grand river. No later than the Pleistocene, the Amazonian basin was a mediterranean sea separating the Guianan and Brazilian shields. Also known as Lake Amazonas, the basin was described by Campbell ( 1 990, p. 34) as "the largest freshwater lake the earth has ever known covering all the Amazonian Basin and con- tiguous areas to an elevation of approximately 300 m." The Species A total of 23 forms of long-nose mice have been named, 1 3 of them by Oldfield Thomas of the British Museum (Natural History). The first to be described was Azara's Rat cinquieme or rat roux, in 1801. It received the binomial Mus rufus by Fischer in 1814, and Mus rutilans, an objective HERSHKOVITZ: SOUTH AMERICAN HOCICUDO 15 Table 1 . Oxymycterus species: Arranged in size classes of Atlantic and Andean divisions; measurements are of holotypes, others shown in brackets [ ]. CB = condylobasal length; HF = hind foot with claw; ML = length of upper molar row. Atlantic Division sizes (mm) Andean Division sizes (mm) CB HF ML CB HF ML Large angularis hispidus judex rnisionalis platensis quaestor rostellatus* rufus dasytrichus4 Medium roberti Small amazonicus6 delator nasutus1 36.9' 39- 38.5 33.2 30+ 34' 34.5 36 33- 34 36 5.6 6.9' 5.8 5.7 5.4 5.8 Large doris inca iris juliacae 32.5 31 + 5.5 [32.3]8 33 5.7 379 33 5.7 36.4'° 32 8" [31.7]' [29.6] [27.6] Medium 33 [29]5 [4.9? paramensis nigrifrons jacentior akodontius*5 Small [32.5]12 [28.0]'3 32.614 27 27.5 30 28 26 + 4.9 5 5 5.2 [28.0] 28.5 [29.2] [5.0] 5.0 [4.8] hiska hucucha 25.1 24.1 25 21 5.01 4.1 1 Measured from original figures of holotype, which may be larger than in life. 2 Estimated. 3 Corpus 6" 3'" (= 163 mm), cauda 3" 10'" (= 80 mm). 4 Head and body 1" 10 1/3'" = 474 mm; only measurement. 5 Mean of 1 1, 18, and 17 individuals in same order. 6 Mean of 12 paratypes, HF dry. 7 Mean of five topotypes. 8 Mean of five from SE Peru. 9 Greatest skull length of holotype. 10 Greatest skull length of holotype. 1 ' Measurement out of line. 12 Greatest skull length; mean of nine from Cochabamba. 13 Mean of 13 paratopo types. 14 Dimension out of line; see p. 32. 15 Juvenal. synonym, by Olfers, in 1818. The last species, Ox- ymycterus amazonicus, is described below. Until the present, the numerous described forms of Oxymycterus had never been organized, their interrelationships never defined, their geographic distribution never mapped. Gyldenstolpe's (1932) arrangement of the species is haphazard, his di- agnoses and information borrowed uncritically from the original literature, his own contributions meager. Notwithstanding, his was the primary source of compiled information and remains use- ful for its morphometries and illustrations of type skulls and teeth. Cabrera's (1961) catalog of South American mammals lists all species alphabetical- ly, the higher categories in phylogenetic order. Ca- brera reduced the number of species by casting many as subspecies, his action based on original descriptions and reliance on his unequaled taxo- nomic acumen. The geographic distribution at- tributed to each taxon is, in most cases, an edu- cated guess. Musser and Carleton (1993) adopted Cabrera's systematic arrangement but recognized only species, the erstwhile subspecies reduced to synonyms of their presumed nominal senior rel- atives. The action may well have been dictated by editorial policy. 16 FIELDIANA: ZOOLOGY Fig. 1 1 . Right hands of two long-clawed sigmodontine mice. A. Non-burrowing Amazonian hocicudo, Oxymyc- terus amazonicus Hershkovitz (Brazil). B. Burrowing raton topo, Chelemys megalonyx Thomas (Chile). Left, palmar surface; right, dorsal surface. HERSHKOVITZ: SOUTH AMERICAN HOCICUDO 17 I 0 2 « 18 FIELDIANA: ZOOLOGY J" •3 [_ 3 E 1 § 5 g *"E O ,9 HERSHKOVITZ: SOUTH AMERICAN HOCICUDO 19 20 FIELDIANA: ZOOLOGY Fig. 15. Oxymycterus hiska. Skull: dorsal, ventral, and lateral views with mandible attached, of paratype (mvz 1715183); GSL = 30.6 mm; upper molar row, 4.8. Yanahuaya, Puno, Peru. Photographs courtesy of Hinojosa, Anderson, and Patton. Size Classes (Table 1) Geographic distribution of the species of Oxy- mycterus may be either Andean or Atlantic coastal and highland. Amazonian hocicudos are exten- sions of one or the other group. Bodily dimensions of adults of each geographic division sort into large, medium, or small. The criteria for estimates of size classes are con- dylobasal length of skull (CB), hind foot length with claw (HF), and upper molar row (ML). Each or any combination of measurements may be de- cisive. The dimensions given in Table 1 are those of the holotypes as originally published. Where lacking, the mean [in brackets] of topotypes or other representative adults is given. The data are incomplete but deemed adequate for rough esti- mates or perceptions of the values of each size class. All nominate forms are represented in the table. There appears to be little or no overlap in the measurements between each size class of each division. In effect, each size class of each geograph- ic division is the equivalent of a species group, in Fig. 14. Oxymycterus angularis. Molars: upper and lower little worn A (fmnh 2302<5) and little worn B (fmnh 21983); most to all lophs and lophules disappear or become indistinct after moderate wear; measurements of B, 5.4 mm (GSL = 36.8). HERSHKOVITZ: SOUTH AMERICAN HOCICUDO 21 Fig. 16. Oxymycterus hispidus. Skull (A, B, C, D); molars, E, upper; F, lower (mnr 320033); GSL = 38.8 mm; molars, 5.6. Caparao National Park, Minas Gerais, Brazil. most cases the species itself. In any case, phylo- genetic relationships between size classes are not implied. The hocicudos of both divisions might have diverged from a common ancestor, one fol- lowing the Atlantic dispersal route, the other the Andean, with the species of each division evolving independently. This, however, is speculative. Compared for size, class for class, those of the Atlantic Division are the larger and more diver- sified. throughout most of the geographic range of the genus. The larger species, generally identified as Oxymycterus rufus or O. hispidus, may coexist with medium-size roberti or small-size O. nasutus. Likewise, medium-size O. paramensis may cross trails with large O. inca, and small O. hiska or O. hucucha. Thomas (1896, p. 309) reported O. ihe- ringi and O. nasutus from Taquara, Rio Grande do Sul, but the current generic status of O. iheringi is under revision and not included in Table 1. Sympatry Sympatry, or the crossing of paths by congeneric species at any time within the life span of the taxa, has not been fully documented for Oxymycterus. There is evidence, nevertheless, that sympatry be- tween two or more congeneric species does occur Annotated Catalog of Described Forms of Oxymycterus All named forms are listed alphabetically except the first described as new. Each listing includes available type data and comments by this and oth- er reviewers. 22 FIELDIANA: ZOOLOGY Fig. 17. Oxymycterus hispidus. Molars, A, unworn upper; B, unworn lower (fmnh 945433). Primeiro Morro, Sao Paulo, Brazil. Oxymycterus amazonicus, new species (Figs. 12, 13) Oxymycterus sp. nov. A, Patterson, 1992a:27— BRA- ZIL: Para; Marai, Rio Tapajoz. Oxymycterus inca, Emmons and Feer (part, not Tho- mas) 1990:191 -BRAZIL: Para (south of the Rio Amazonas). Holotype— Adult male, skin and skull, Field Museum of Natural History no. 94524, collected 27 January 1 96 1 , by A. M. Olalla, original number 1066. Type Locality (Fig. 2)— Fordlandia, right bank, lower Rio Tapajoz, Para, Brazil, 3°40'S, 55°30'W. Distribution (Fig. 1)— Known from the south bank of the lower Rio Amazonas and lower parts of tributaries Tocantins, Xingu, Tapajoz, and mid- dle Rio Madeira. The range may extend eastward along the Atlantic coast to the Brazilian bulge, where Oxymycterus angularis is represented in the State of Ceara. In all cases, Oxymycterus seeks the higher ground or levees. Oxymycterus amazonicus is the only species of the genus known to occur in the lower Rio Ama- zonas basin. It may represent a relict population that once ranged from near sea level to the head- waters of the lower Amazonian tributaries, or it may have spread from the headwaters to the lower half of the Amazonian basin. More likely it has newly infiltrated from the east. Diagnosis— A chocolate-colored mouse with sharply defined orange to reddish orange under- parts; size small among Atlantic Division hoci- cudos species (Table 1); skull with rostrum com- paratively short, robust, nasals with little taper or HERSHKOVITZ: SOUTH AMERICAN HOCICUDO 23 Table 2. Measurements (in mm) of type series of Oxymycterus amazonicus. Means, extremes, and number of samples are included at bottom. Greatest Catalog Total Head and skull Museum number Date Sex length body Tail Hind foot Ear length Rio Tocantins, near Itupiranga, 26 km N, 30 km W Maraba, 05°09'S, 42°20'W MPEG 9134 1 1/6/76 6 217 121 96 27 13 34.2 Rio Xingii, 54 km S, 150 km W Altami ra, Para, 03°12'S, 52°12'W MPEG 8992 17/8/75 6 211 131 80 27 15 — MPEG 9000 1/7/75 $ 208 114 94 29 14 32.9 MPEG 9008 27/7/75 6 208 115 93 26 13 33.6 MPEG 9018 27/8/75 8 230 130 100 28 15 34.7 MPEG 9025 24/9/75 S 170 120 50 (bob) 28 14 32.3 MPEG 9029 19/11/75 6 220 125 95 29 12 31.2 MPEG 9050 27/7/75 <3 198 119 79 30 11 — MPEG 8999 25/6/75 9 213 132 81 27 13 31.9 MPEG 9002 2/7/75 2 200 120 80 26 14 31.7 MPEG 9031 28/11/75 $ 219 131 88 27 13 32.8 Rio Tapajoz, Fordlandia, Para, 03°40'S, 55°30'W FMNH 94523 27/1/61 6 215 130 85 28 — 32.1 Paratype 210 125 87 28 13 32.7 (140-230) (114-132) (75-96) (26-30) (11-15) (31.2-34.7) FMNH 94524 27/1/61 $ 221 136 75 27 — 32.0 Holotype 13 13 12 13 11 11 expansion to trumpet shape; tips nearly square and little produced beyond perpendicular plane of out- er incisive border; vestige of mesoloph (id) pres- ent, of upper usually fused with one or another lophule or adjacent cusp. External— (All dry skins, except of holotype and paratype, were prepared from liquid-pre- served specimens.) Pelage fairly dense, of dorsum about 1 cm long; upper parts of body dominantly brownish, sides more reddish, the subterminal pheomelanin bands wider; underparts orange but with slaty basal portion of hairs showing through on chest, belly, and limbs, ventral midline more reddish; crown and upper surface of muzzle red- dish brown like back; tip of nose with blackish patch or stripe; ears brown; hands (Fig. 1 1), feet, and tail blackish; frontal hairs tufted but condition may be an artifact of preparation; vibrissae thin, inconspicuous, manual digital vibrissae hardly ex- tending beyond distal phalanges, pedal digital vi- brissae extending halfway to claw tips; mystacial vibrissae barely extending to base of ears; skin between pedal digits palmate, outer pedal digit with claw extending to base of fourth digit, inner digit extending to base of second phalange of digit II, digit III longest, IV slightly shorter; length of each middle digit claw little more than half digital length; manual digits with claws slightly shorter or longer and weaker than pedal, the inner vestig- ial, the outer reduced; tail little more than half combined head and body length, hairs short, stiff, about 2-3 scales long, the slightly imbricated rhomboidal scales showing through, their arrange- ment annular with 3 hairs per scale. Nipples— 2 pair pectoral, 1 pair inguinal, total 6. Cranial— All skulls except those of holotype and paratopotype more or less damaged from ov- erboiling and careless cleaning; type skulls dam- aged by trap; rostrum thick, nasals broad, square at tip, obtusely pointed behind, narial aperture subtriangular, anterior border projected little be- yond perpendicular plane of anterior surface of incisors in types, more or less the same in type series; width of subovate incisive foramina 3 mm, posteriorly extending to second reentry fold of M1; temporal ridges poorly defined in types, more or less in others; interparietal bone 2.2 x 7.1 in ho- lotype, 3.3 x 9.4 in paratype; zygomatic arches slender, zygomatic plate moderately expanded, width about equal to length of m1, visible when viewed from above; width of mesopterygoid fossa 2.3 mm; additional measurements in Table 2. Dental— Molars of all specimens examined ex- cept types are cracked or broken from overboiling of skulls in cleaning process, inner or dentine sur- face of incisors stained brown except in types; up- per incisors short, comparatively stout, cutting edge plane, outer surface naturally pigmented orange; 24 FIELDIANA: ZOOLOGY Table 2. Extended. Condylo- Inter- basal Zygomatic orbital Braincase length breadth width width Nasal Incisive Palatal Zygomatic length foramina length Diastema plate Molar row 31.4 5.7 13.4 12.3 7.3 3.5 8.4 2.6 5.2 29.5 14.3 6.1 13.3 12.1 7.0 3.8 7.4 3.1 5.3 30.6 15.6 6.0 13.6 13.3 6.6 4.5 8.1 1.1 4.9 32.0 15.7 6.6 14.2 13.7 — — 8.6 2.9 5.2 28.6 14.4 — 13.6 11.7 6.6 3.8 7.1 2.8 4.8 29.2 14.0 6.2 13.4 — 7.0 3.9 7.6 2.7 5.1 28.0 14.2 5.8 13.3 — — — 7.7 0.7 5.0 28.9 14.6 6.4 13.5 11.6 7.0 4.2 7.9 2.6 4.9 28.8 15.1 6.3 13.7 12.0 7.0 — 7.4 2.8 4.5 29.8 15.1 6.3 13.8 11.7 6.7 4.5 7.9 2.8 5.3 28.8 6.1 13.5 11.5 _ _ 7.5 2.6 4.8 29.6 14.8 6.2 13.6 12.1 6.9 4.0 7.8 2.4 5.0 (28.0-31.4) (14.0-15.7) (5.7-6.6) (13.3-14.2) (11.4-13.7) (6.6-7.3) (3.5^1.5) (7.1-8.4) (0.7-2.9) (4.8-5.3) 29.4 — 6.4 13.5 11.4 6.8 3.9 7.8 2.4 5.0 12 9 11 12 10 9 8 12 12 12 molars hypsodont, 3-rooted, tetralophodont, crowns terraced, valleys deep, anterior median fold well denned, anterolophule, labiolophulid, ecto- lophid present, mesoloph (id) vestigial, absent or fused with 1 or 2 accessory lophs such as paralo- phule, metalophule (id) entolophulid, or an ad- jacent cusp (cf. Hershkovitz, 1993, Fig. 7). Holotype with occlusal surface of molars eroded to deeply dished 8-shaped configuration; procin- gulum of first molars with anterior median fold almost completely obliterated by wear; moderate- ly worn molars of paratopotype (fmnh 94523) ter- raced, the outer cusps sharply defined, anterolo- phule present; paracone of first and second molars with prominent paralophule, of another specimen (mpeg 0900) with 2 paralophules, the inner fused with stub-like mesoloph; mandibular teeth of ho- lotype excessively worn, of paratopotype with an- terior median fold of m, sharply defined, labiolo- phulid and ectolophulid of m, 2 well developed, anterior labial fold well defined in all molars; me- solophid vestigial in all lower molars. Variation— Of the 14 prepared specimens, all but the holotype and paratotype were skinned out of formalin. They agree well with the type descrip- tion except pelage is in better condition despite effects of liquid preservative; pelage of underside of limbs shows more gray at base; cheeks grade from dominantly brown to entirely orange like neck; basicranium, interorbital region, and tur- binates considerably damaged or destroyed by the cleaning method; temporal ridges of skull more marked than in holotype; palate extends to or slightly beyond line drawn across posterior border of last molars. Exposed dentine of upper and lower incisors with a purplish stain presumably acquired during immersion in formalin. Comparisons— The small size of Oxymycterus amazonicus obviates the need for comparisons with species of the large- and medium-size classes of the Atlantic Division (Table 1). Of the two other small-size-class species of the Atlantic Division, the Uruguayan Oxymycterus nasutus is readily distinguished from the dark brown, nearly black- ish O. amazonicus by its bright ochraceous col- oration throughout, softer pelage, pale hands and feet, bicolor tail, narrower, more tapered nasals, smooth braincase, slender mandible, and longer, finer lower incisors. Specimens of O. delator with some small-size- class measurements were not available for com- parisons. The original description is of a dark- colored animal with undersurface notably paler than that of O. amazonicus, head with a light patch behind and above each eye absent in amazonicus and all other hocicudos known to me, and tail thickly and uniformly haired as contrasted with the virtually bare tail of other hocicudos. Descrip- tion of the nasals suggests the narrower characters of O. nasutus. Outstanding in Thomas's original HERSHKOVITZ: SOUTH AMERICAN HOCICUDO 25 Fig. 18. Oxymycterus juliacae. Skull (A, B, C, D); molars, E, upper; F, lower (fmnh 52622); GSL = 34.9 mm; molars, 5.3. Segrario, Puno, Peru. description are a combined head and body length of 155 mm coupled with a tail length of 106 mm. Both measurements belong with such large-size hocicudos as the Paraguayan O. rufus or O. misio- nalis. Cranial and hindfoot measurements, how- ever, are those of small hocicudos (Table 1). Oxymycterus hiska and hucucha of the Andean Division are extremely small and need no com- parison with O. amazonicus (Table 1). O. ako- dontius, assigned to the medium-size class (Table 1 ) is a juvenal. Had it been given a chance, it would have grown into its proper medium-size class, or perhaps the larger class. Remarks— The series of Oxymycterus amazo- nicus was loaned by the Museu Paraense Emilio Goeldi for identification and possible description. Field Museum Curator of Mammals Bruce Pat- terson, who was at the time studying some Am- azonian material received from the Royal Natural History Museum, Stockholm, for identification, recognized a specimen of the new species among the rodents. He generously added it to those being described in this paper. Other specimens of the then undescribed O. amazonicus were recorded by Emmons and Feer (1990, p. 191) under the name of the very similar appearing but larger Andean Oxymycterus inca. Specimens Examined— Total 15. BRAZIL: Para (Rio Tapajoz, Fordlandia [0340/5530], holotype and paratopotype, fmnh; Marai = Mararu, Rio Tapajoz [0226/5442], 1 [rnhms]; Maraba [0521/ 4900], Rio Tocantins, 2 [mpeg]; Altamira [0312/ 5212], Rio Xingu, 9 [mpeg]; Mato Grosso (Rio Aripuana, Humboldt, 1 [usnm]). Oxymycterus akodontius Thomas Oxymycterus akodontius Thomas, 1921:615. Cabrera, 1961:466 — distribution; "possibly only a color 26 FIELDIANA: ZOOLOGY Fig. 19. Oxymycterus nasutus. Skull (A, B, C, D); molars, E, upper; F, lower (fmnh 276523); GSL = 33.3 mm; molars, 4.7. Maldonado, Uruguay. morph of O. paramensis jacentior." Reig, 1987: 361 —may be a variant of paramensis. Holotype— Male, skin and skull, British Mu- seum (Natural History) no. 21.11.1.72, collected 8 May 1921, by E. Budin, no. 1465. Type Locality (Fig. 2) — Higuerilla, Valle Grande, Jujuy, Argentina; 23°35'S, 65°15W, 2000 m. Distribution— Known from type locality only. If conspecific with Oxymycterus paramensis, the range extends through the Andes from southeast- ern Peru (Puno) through Bolivia into northwestern Argentina (Jujuy). Original Measurements of Holotype— HB, 116; T, 79; HF, 26 (su); E, 18.5; GSL, 31; CB, 27; ZB, 13.5; IB, 6; ML, 5. Remarks— The taxonomic status of medium- size O. akodontius is uncertain. It was originally described as dark-colored juvenal, size about as in O. paramensis. Oxymycterus angularis Thomas (Fig. 14, molars) Oxymycterus angularis Thomas, 1909:237-238. Vieira, 1 953:2 16-BRAZIL: Alagoas (Manga- beira); measurements. Gyldenstolpe, 1932:131, 149— type data; measurements. Cabrera, 1 96 1 :466— listed; distribution. Reig, 1987:361— angularis and roberti "tentative" synonyms. Musser and Carleton, 1993:726-listed. Holotype— Male, skin and skull, British Mu- seum (Natural History) no. 3. 10. 1.56, collected 14 August 1903 by Alphonse Robert, no. 1706. Type Locality (Fig. 2). Sao Lourenco da Mata, Pernambuco, Brazil, 8°00'S, 35°03'W, 30 m. HERSHKOVITZ: SOUTH AMERICAN HOCICUDO 27 Fig. 20. Oxymycterus paramensis nigrifrons. Mo- lars, A, upper; B, lower (fmnh 52623); GSL =31.2 mm; molars, 4.4. Limbani, Puno, Peru. Distribution— Coastal Brazil, in the states of Pernambuco and Alagoas. Original Measurements of Holotype— HB, 160; T, 100; HF, 30 (cu); E, 21; ML, 5.6; "of another specimen," GSL, 36.3; CB, 33; ZB, 17; IB, 7.1; ML, 5.6. Remarks— Oxymycterus angularis, a large spe- cies of the Brazilian bulge, needs comparison with Bahian O. hispidus, its nearest geographic ally. A series of seven specimens from Caruari, Pernam- buco, inland from Sao Lourenco, the type locality, agrees with the description of angularis as well as with that of hispidus. The form of the zygomatic root or plate of hispidus as described by Thomas is difficult to visualize, but those of specimens at hand are not distinctive. A second series of five specimens from Vicosa, Alagoas, which agrees with the Pernambuco angularis, is less distant geo- graphically from the hispidus type locality and fits the original description of hispidus as far as it goes. Measurements of three specimens from Ceara at the top of the Brazilian bulge are smaller throughout than those of angularis or hispidus. They agree best with our referred medium-size O. roberti from Brasilia. Oxymycterus dasytrichus Schinz Mus dosytrichos [sic] Schinz, 1821:288— prepublica- tion of the original Wied-Neuwied (1826) descrip- tion. H[ypudaeus]. dasytrichos [sic], Wied-Neuwied, 1826: 425 — BRAZIL: Bahia (Rio Mucuri, type locality). Hypudeus dasytrichus, Tate, 1932b: 17 — name emended. [Oxymycterus] dasytrichus, MusserandCarleton, 1993: 727 — synonym of O. rufus. Oxymycterus rufus dasytrichus, Cabrera, 196 1:468 — BRAZIL: Bahia (Rio Mucuri, type locality); rostel- latus Wagner, a synonym; distribution. Lectotype— Skin mounted with skull in, Amer- ican Museum of Natural History no. 559; lecto- type selected by Avila Pires (1965, p. 15). Type Locality (Fig. 2)-Rio Mucuri (18°05'S, 39°34'W, sea level), Espirito Santo; a second spec- imen from nearby Camamu, Espirito Santo, re- ceived from Herr Freyreiss. Distribution— Known from type region only. Original Measurements of Holotype— HB, 10 Vi" (= 474 mm). Oxymycterus delator Thomas Oxymycterus delator Thomas, 1903a:489. Musserand Carleton, 1993:726, 727— compared with the "larg- er, rufescent rufus from Entre Rios." Reig, 1987: 361— distinction from O. rufus questioned. Voss and Linzey, 1981:14— morphology of male acces- sory glands. Holotype— Male, skin and skull, British Mu- seum (Natural History) no. 3.4.7.18, collected 24 October 1902 by William T. Foster, no. 880. Type Locality (Fig. 2)— Sapucay, Paraguay, 25°40'S, 56°55'W, 220 m. Distribution— Known only from the type lo- cality. Original Measurements of Holotype— HB, 155; T, 106; HF, 26/28.5; GSL, 34.5; CB, -; ZB, 14; IB, 5.1; ML, 5. Remarks— Described as a very large uniformly blackish species. Musser and Carleton (1993, p. 726) compared it with "larger rufescent O. rufus 28 FIELDIANA: ZOOLOGY Fig. 21. Oxymycterus roberti. Skull (A, B, C, D); molars, E, upper; F, lower (fmnh 1283213); GSL = 37.2 mm; molars, 4.1. Parque Nacional de Brasilia, D.F., Brazil. from Entre Rios" without resolving its status as a species. The type measurements combine the body and tail measurements of a large species and the cranial and hindfoot measurements of a small one. Correct appreciation of the coloration of delator requires special viewing conditions. The color, ac- cording to Thomas (1903a, p. 489), approximates Ridgeway's "clove-brown not rufous at all. This is when viewed from behind and above, but if the specimen is held between the light and the ob- server, with its nose toward him, its upper surface appears blackish with a purplish sheen . . . ." Ob- viously, delator is chromatically very variable. Oxymycterus don's Thomas Oxymycterus doris Thomas, 1916:478^79. Musser and Carleton, 1993:727— synonym of O. inca. Oxymycterus inca doris, Cabrera, 1961:467— classi- fication; distribution. Holotype— Male, skin and skull, British Mu- seum (Natural History) no. 2.1.1.95, collected 21 May 1901 by Perry O. Simons, no. 1441. Type Locality (Fig. 2)— Charuplaya, Rio Se- cure, upper Rio Mamore, Cochabamba, Bolivia, 15°48'S, 66°30'W, 1350 m, 1400 m. Distribution— Known only from type locality. If conspecific with Oxymycterus inca, the range extends from central Peru (Junin) south into Co- chabamba, Bolivia. Original Measurements of Holotype— HB, 150; T, 126; HF, 31 (su); E, 21; GSL, 37.5; CB, 32.5; ZB, 16; IB, 6.6; ML, 5.5. Remarks— Cabrera (1961, p. 467) holds that the large-size doris may be no more than a color phase of "Oxymycterus inca iris Thomas." HERSHKOVITZ: SOUTH AMERICAN HOCICUDO 29 Oxymycterus hiska Hinojosa, Anderson, and Pattern (Fig. 15, skull) Oxymvcterus hiska, Hinojosa, Anderson, and Patton, 1987:4, 10, 14, Figs. 2, 3 (skull), Fig. 4 (molars). Musser and Carleton, 1993:726— listed. Holotype— Adult female, skin, skull, body in alcohol, Museum of Vertebrate Zoology, Univer- sity of California, no. 171519; five topotypes; col- lected 5 August 1 985 by J. L. Patton, original num- ber 12257. Type Locality (Fig. 2)— Yanahuaya, 14 km W, Puno, Peru, 14°19'S, 69°21'W, 2210 m. Distribution— Known from type locality only. Original Measurements of Holotype— HB, 100; T, 77; HF, 25 (cu); E, 16 (dry); CB, 25.10; ZB, 17.57; IB, 6.77; ML, 5.01. Remarks— The original description including measurements and figures of the skull indicates valid distinction of small hiska from small hucu- cha. A photograph of the paratype skull (mvz 171578) is shown (Fig. 15). The geographic variation, however, has not been studied. Original Measurements of Holotype— Ft B, 162; T, 115; HF, 34 (cu); E, 16 (dry). Remarks— Judged only by the original mea- surements and figures, O. hispidus may be the larg- est of the nine large Atlantic Division species of the genus. The figured skull measures about 36.9, crown length of upper molar row, 6.9. As depicted, the reddish brown upper parts of hispidus, the red- dish head and cheek, comparatively dark under- parts, reddish tail and upper surface of hands and feet, but more particularly the description in text, fit our Pernambuco series completely, especially mnr 32 1 . Measurements of a blackish female from Bahia (Juquila Tres Bracos, usnm 545060) are HB, 157; T, 118; HF, 34; E, 24; GSL, 37.6; ML, 5.7. The name hispidus has been widely used for the large Oxymycterus of eastern Brazil. It has also been treated as a junior synonym of O. rufus Fi- scher and as a senior synonym of judex, misionalis, rostellatus, and quaestor. Their holotypes or rep- resentatives await comparison with each other. Oxymycterus hispidus Pictet (Fig. 17, molars) Oxymycte rus hispidus Pictet, 1843:37, PI. 10 (animal), PI. 11, Figs. 9-14 (skull, teeth, soles, palms). Pictet, 1 843:2 12 -characters. Leuderwaldt, 1929:27- BRAZIL: Sao Paulo (Una de Sao Sebastiao). Vieira, 1953:143- BRAZIL: Sao Paulo (Sao Sebastiao); Rio de Janeiro (l\ha Grande). Musser and Carleton, 1993: 727— synonyms: judex, misionalis, quaestor. Oxymycterus hispidus hispidus, Cabrera, 196 1 :466 — classification; distribution. 0[xymycierus]. hispidus, Reig, 1987:361— "tenta- tive" synonyms: judex, misionalis, quaestor. H[esperomys\. (Oxymycterus) hispidus, Thomas, 1 884: 450 — classification. M[us]. hispidulus Schinz, 1845, 2:179— renaming of Oxymycterus hispidus Pictet mistakenly believed preoccupied by Mus hispidus Lichtenstein, 1828. Holotype— Male, skin and skull (mounted?), in Musee d'Histoire Naturelle de Geneve, no. 275/ 47, received from M. Blanchet. Type Locality (Fig. 2)— Bahia, Brazil, 12°59'S, 38°3 l'W, sea level. Distribution— If conspecific with most large- size Atlantic Division species including Oxymyc- terus rufus, a senior synonym, the range of the hispidus group would extend from Alagoas and Pernambuco, Brazil (angularis), in the north, to Buenos Aires and Los Rios, Argentina, in the south. Oxymycterus hucucha Hinojosa, Anderson, and Patton Oxymycterus hucucha Hinojosa, Anderson, and Pat- ton, 1 987: 1 5, Figs. 2, 3 (skull, Fig. 4 [molars]). Mus- ser and Carleton, 1993:727 — "Morphologically similar to O. hiska.'" Holotype— Adult male, skin, skeleton, tissues in liquid nitrogen, American Museum of Natural History no. 260583 (nk 1 2028, University of New Mexico), prepared by S. Anderson, 4 September 1 984, field number 8 1 76; two additional Bolivian specimens (1 from Totora, 20 mi E; 1 from Epi- zana, 101 km W, 2989 m). Type Locality (Fig. 2)— Comarapa, 28 km W (Santa Cruz) but in Cochabamba, Bolivia, 1 7°5 1 'S, 64°40'W, 2800 m. Distribution— Known from type locality only. Original Measurements of Holotype— HB, 1 16 + ; T, 60 + ; HF, 23 (cu); E, 14; CB, 25.41; ZB, 13.01; IB, 5.39; ML, 4.26. Remarks— O. hiska was said to be "distinguish- able from all previously described species by smaller size . . . ." O. hucucha was described in the same paper (p. 1 5) as slightly smaller than O. hiska. Compared with iheringi, questionably an Oxy- mycterus, hucucha is about the same or slightly smaller in all dimensions. 30 FIELDIANA: ZOOLOGY \J\Oxymycterus iheringi Thomas Oxymycterus nasutus, Hensel (not Waterhouse), 1 873: 43, Figs. 19, 29 (molars)- BRAZIL: Rio Grande do Sul. Musser and Carleton, 1993:727— listed. H[esperomys]. nasutus, Leche (not Waterhouse), 1886: 700, Figs. 29, 30 (molars)- BRAZIL: Rio Grande do Sul. Oxymycterus iheringi Thomas, 1896, p. 308. Thomas, 1902a:62- BRAZIL: Parana (Roca Nova, Serra do Mar, 100 m). Oxymycterus iheringi, Massoia, 1963:129, Figs. 1-4 (skull)— ARGENTINA: Misiones (Tobunas, Ruta 14, km 352; Puerto Gisella); characters; compari- sons; taxonomy; URUGUAY: Canelones (Arroyo Sarandi). Massoia and Fornes, 1 969:3 1 5, Fig. 1 (an- imal), Fig. 2 (palate), Fig. 3 (molars); taxonomic history; characters; comparisons with O. nasutus. Musser and Carleton, 1993:727— listed. M\icroxus]. iheringi, Thomas, 1909:237— classifica- tion. Microxus (?) iheringi, Gyldenstolpe, 1932:134— list- ed. Microxus iheringii [sic] Vieira, 1953:145— BRAZIL: Rio Grande do Sul (Sao Lourenco). Massoia, 1 963: 135— Vieira's identifications questioned. Akodon {Microxus) iheringi, Cabrera, 1961:458 — classification; distribution. cation; distribution; doris, irisjuliacaelistcd as sub- species. Oxymycterus incae [sic], Reig, 1987:361 —inca, doris, iris, juliacae recognized as subspecies. Holotype— Male, skin and skull, British Mu- seum (Natural History) no. 0.7.7.45, collected 10 April 1900 by Perry O. Simons, no. 925. Type Locality (Fig. 2)— Perene, Junin, Peru, 10°58'S, 75°13'W, 800 m. Distribution— Known from type locality only in central Peru. If combined with its ascribed ju- nior synonyms (doris, iris, juliacae), the range would extend from central Peru south through the Andes into Cochabamba, Bolivia; altitudinal range from about 800 m (Perene, Peru) to about 1900 m (juliacae, Peru). Original Measurements of Holotype— HB, 135; T, 105; HF, 30/33; E, 21; GSL, 35; ZB, 18; IB, 6.7; ML, 5.7. Remarks— This is the large-size-class species of the Andean Division. Holotype— Female, skin and skull, British Mu- seum (Natural History) no. 86.9. 16.8, collected by Hermann von Ihering. Type Locality (Fig. 2)— Taquara do Mundo Novo, Rio dos Sinos (Linos), Rio Grande do Sul, Brazil, 29°39'S, 50°47'W, 29 m. Distribution — Known with certainty only from type locality; recorded from Parana, Brazil, Uru- guay, and Misiones, Argentina. Original Measurements of 2 Holotype— HB, 100; T, 94; HF, 23.5 (su); E, 16.7; CB, 24.7; IB, 6; ML, 4.2. Remarks— The claws, both manual and pedal, of referred specimens of O. iheringi are small, al- most wispy, and partially hidden by long hairs. This may account, in large part, for exclusion by some authors of iheringi from the genus Oxymyc- terus. The status of the species and related newly collected material is presently under study. Re- tention of the species in the genus Oxymycterus is to be considered provisional. Oxymycterus inca Thomas Oxymycterus inca Thomas, 1900:298. Musser and Carleton, 1993:727— Emmons and Feer (1990) identification of "Amazon Basin" long-nose form questioned; synonyms: doris, iris, juliacae. Oxymycterus inca inca Cabrera, 1961:467— classifi- Oxymycterus iris Thomas Oxymycterus iris Thomas, 1 90 1 : 1 83. Oxymycterus inca iris, Cabrera, 1961:467— distribu- tion; "may be no more than a color phase of inca." Holotype— Male, skin and skull, British Mu- seum (Natural History) no. 1.1.1.76, collected 5 September 1900 by Perry O. Simons, no. 1218. Type Locality (Fig. 2)— San Ernesto, Mapiri, upper Rio Beni, La Paz, Bolivia, 10°23'S, 65°24'W, 1000 m. Distribution— Known from type locality only but see Oxymycterus inca, above. Original Measurements of Holotype. HB, 160; T, 102; HF, 30/33; E, 22; GSL, 37; IB, 6.5; HL, 5.7. Oxymycterus paramensis jacentior Thomas Oxymycterus paramensis jacentior Thomas, 1 925:580 Thomas, 1926:323-BOLIVIA: Tarija (Pino) 1800 m. [Oxymycterus] paramensis jacentior, Yepes, 1935 232-ARGENTINA: Salta. Oxymycterus paramensis jacentior, Cabrera, 1961 467— classification; distribution. HERSHKOVITZ: SOUTH AMERICAN HOCICUDO 31 Holotype— Male, skin and skull, British Mu- seum (Natural History) no. 25.2.1.49, collected 2 August 1924 by Emilio Budin, no. 1772. Type Locality (Fig. 2)— Carapari, Yacuiba, Tarija, Bolivia, 21°49'S, 63°46'W, 1000 m. Distribution— Known only from Tarija De- partment, southern Bolivia, between 1000 and 1800 m. Original Measurements and Holotype— HB, 146; T, 1 19; HF, 28 (su); E, 198; GSL, 36.2; CB, 32.6; ZB, 16.8; IB, 6; ML, 5. Remarks— The condylobasal length 32.6 places the holotype with the large-size group of Andean hocicudos (Table 1). All other dimensions, how- ever, and its classification as a subspecies of O. paramensis point to a medium-size species. The holotype, an "old male," was described as slightly larger "than typical paramensis" its "skull rather larger and heavier." Oxymycterus judex Thomas Oxymvcterus judex Thomas, 1909:238. Gyldenstolpe, 1932:131, PI. 16, Figs. 1-1 b (skull), Fig. 18 (upper molars) — BRAZIL: Santa Catarina, Stockholm Museum. Musser and Carleton, 1993:727— syn- onym of O. hispidus. Oxymycterus hispidus judex, Cabrera, 1961 :466 — classification; distribution. Holotype— Male, skin and skull, British Mu- seum (Natural History) no. 9.11.19.23, collected by W. Ehrhardt, no. 2 1 . Type Locality (Fig. 2)— Joinville, Santa Ca- tarina, Brazil, 26°18'S, 48°50'W, sea level. Distribution— Known from type locality only, but see Oxymycterus hispidus, above. Original Measurements of Holotype— HB, 1 52; T, 1 29; HF, 34.5; E, 22.5; GSL, 42; ZB, 1 7.6; IB, 6.8; ML, 5.8. Remarks— Oxy myct erus judex was described as "most nearly allied to O. quaestor but is distin- guished by the larger size and much larger brain case." Separation of the forms by Thomas was based on measurements of a representative each of the two taxa from the same region, one being insignificantly larger than the other. Cabrera re- classified them as subspecies without direct com- parison. Oxymycterus juliacae J. A. Allen (Fig. 18, skull, molars) Oxymycterus juliacae J . A. Allen, 1900:223. Thomas, 1 90 1 a: 1 89 - PERU : Puno (Rio Inambari). Thomas, 1902b: 140-BOLIVIA: Cochabamba (Charuplaya, 1350 m). Sanborn, 195L23-PERU: Cuzco (Ha- cienda Cadena). Sanborn, 1953:6 — PERU: Puno (Sandia). Goodwin, 1953:322— type history. [Oxymycterus] juliacae, Musser and Carleton, 1993: 727— a synonym of O. inca. Oxymycterus inca juliacae, Cabrera, 1961:467— clas- sification; distribution. Holotype— Male, skin and skull, American Museum of Natural History no. 15804, collected 29 November 1899 by H. H. Keays, no. 12. Type Locality (Fig. 2)— Inca Mines, Santo Do- mingo, upper Rio Inambari, Puno, Peru, 15°30'S, 70°08'W, 1875 m. Distribution— Andes of southern Peru, in Puno and Cuzco, south into Cochabamba, Bolivia. In effect, the same range as that of Oxymycterus inca Thomas, 1900. Original Measurements of Holotype— HB, 138; T, 112; HF, 29/32; E, 16; E, dry, 13; GSL, 36.4; ZB, 16; IB, 6.7; ML, 8. Remarks— Published measurements of O. ju- liacae agree with those of O. inca except the molar row is given as 8 mm. This length exceeds any other known for Oxymycterus. The largest for any other holotype is 5.8 mm, in this case O. judex and O. quaestor. The largest molar row measured in this study, 6.4 mm, is that of a specimen from Carucari, Pernambuco, a near topotype of O. an- gularis. Oxymycterus misionalis Sanborn Oxymycterus misionalis Sanborn, 1931:1. Sanborn, 1947:250-type history. Carleton, 1973: 15 -stom- ach morphology. Musser and Carleton, 1993:727 — synonym of O. hispidus. Oxymycterus hispidus misionalis, Cabrera, 1961:466 — classification; distribution. Holotype— Female, skin and skull. Field Mu- seum of Natural History no. 26756, collected 15 September 1926 by Colin C. Sanborn, no. 1242. Type Locality (Fig. 2)— Rio Parana-i, affluent of Rio Parana, about 100 mi S Rio Iguassu, Mi- siones, Argentina, 26°37'S, 54°46'W, 100 m. Distribution— Known from type locality only but see Oxymycterus hispidus, above. Original Measurements of Holotype— HB, 174; T, 143; HF, 36; E, -; GSL, 42.8; CB, 38.5; ZB, 17; IB, 6.2; ML, 5.7. 32 FIELDIANA: ZOOLOGY O. misionalis may indeed be conspecific with hispidus, or more likely rufus, the older name. Oxymycterus nasutus (Fig. 19, skull, molars) Mus nasutus Waterhouse, 1 837: 16. Waterhouse, 1 839: 56, PI. 17, Fig. 2 (animal), PI. 33, Fig. 7 (skull, dentition). Mus (Oxymycterus) nasutus, Waterhouse, 1837:31 — classification. Holochilus nasutus, Gray, 1843:1 14— BRAZIL. Hesperomys nasutus, Burmeister, 1879:214 (part)— BRAZIL: Rio de Janeiro (Novo Friburgo). Figueira, 1894:18-URUGUAY: common everywhere. Hesperomys (Oxymycterus) nasutus, Thomas, 1884: 450— classification. Ihering, 1892:109— BRAZIL: Rio Grande do Sul; characters. Ihering, 1894:20— BRAZIL: Sao Paulo. H[esperomys (Oxymycterus)] nasutus, Ihering, 1885: 424 -BRAZIL; characters. Oxymycterus nasutus, Pelzeln, 1883:74— BRAZIL: Sao Paulo (Ypanema). Vorontsov, 1 979:20, 41,95, 184, 234, 277, 281, Fig. 100 (stomach), Fig. 125 (intestine). Musser and Carleton, 1993:727— listed; distribution. 0[xymycterus] nasutus, Thomas, 190 lb: 5 31— range from Uruguay into Rio Grande do Sul, Brazil. Tho- mas, 1 9 1 4:244-URUGUAY: La Plata estuary (type locality); BRAZIL: Rio Grande do Sul. Sanborn, 1929:157-URUGUAY: Maldonado (Maldonado; 1 5 km N of San Carlos); San Jose (Arazati). Vieira, 1953: 143 -part, BRAZIL: Parana (Castro); Sao Paulo (Campos do Itatiaia). ARGENTINA: Entre Rios; measurements questionable. Oxymycterus nasutus, Musser and Carleton, 1993: 717-listed. [Hesperomys (Oxymycterus) rufus], Trouessart (part), 1 88 1 : 1 42 —classification. Oxymycterus rufus nasutus, Cabrera, 1 96 1 :469— clas- sification; distribution. Holotype— Skin and skull, British Museum (Natural History) no. 55.12.24.176, collected by Charles R. Darwin on the voyage of the Beagle. Type Locality (Fig. 2)— Maldonado, Uruguay, 34°54'S, 54°57'W, sea level. Distribution— From La Plata Estuary north through Uruguay, Paraguay, and southeastern Brazil in the States of Rio Grande do Sul, Santa Catarina, and Parana. Original Measurements of Holotype— HB, 131; T, 68; HF, 26.5. Remarks— Oxymycterus nasutus is the small- >ize species sympatric with the large rufus (or his- ndus) throughout Uruguay, Paraguay, southeast- rn Brazil, and the La Plata estuary region in Ar- entina. The species had been treated mistakenly is a synonym of O. rufus by Hershkovitz (1959, ). 339). Oxymycterus paramensis nigrifrons Osgood (Fig. 20, molars) Oxymycterus paramensis nigrifrons Osgood, 1 944: 1 97. Sanborn, 1947:250— type history. Sanborn, 1950: 15 -PERU: Puno (Limbani); BOLIVIA: La Paz (Nequejahuira; Rio Aceramarca; Cocopunco; Okara; Tacacama). Cabrera, 1961:468— distribution. Holotype— Female, skin and skull, Field Mu- seum of Natural History no. 52629; collected 29 September 1941 by Colin C. Sanborn, no. 2743. Type Locality (Fig. 2)— Limbani, Puno, Peru, 14°08'S, 69°42'W, 2810 m. Distribution— Andes of southern Peru and northern Bolivia pending comparison with O. par- amensis (s.s.) and O. p. jacentior. Original Measurements of Holotype— "Av- erage of eight adult topotypes": TL, 235; T, 90; HF, 30; GSL, 32.7; ZB, 14.3; IB, 6.3; ML, 5. Remarks— The name of the taxon is derived from a supranarial blackish patch "uniformly present and has not been found in any other form examined except in very incipient form." The tip of the snout is indeed blackish in all 24 specimens of the type series and also in nearly every specimen of Oxymycterus I have examined from throughout the range of the genus. Holotype and topotype of O. amazonicus at the northeastern extreme of the generic range are as well marked with a rostral stripe as any nigrifrons from Limbani, Peru. Sep- aration from juliacae or paramensis (s.s.) may be questioned. Oxymycterus paramensis Thomas Oxymycterus paramensis Thomas, 1902b: 139— BO- LIVIA: Cochabamba (Choquecamate; Alicuni, 2600 m; Choro, 3500 m); local name huacucha. Thomas, 1 9 1 8: 1 88- ARGENTINA: Leon; Jujuy; local name, hocicudo; agrees with type series in every detail. Yepes, 1933:48 -ARGENTINA: Salta (Aguarau, 700 m). Hooper and Musser, 1964:28, Fig. 5 (glans penis)— description of glans, baculum. Musser and Carleton, 1993:727— listed; synonyms: nigrifrons, jacentior. 0[xymycterus] paramensis, Thomas, 1925:580— PERU: Cuzco (Ollantaytambo); ARGENTINA: Leon; Jujuy. Oxymycterus paramensis paramensis, Gyldenstolpe, 1932:130— "most closely allied to O. roberti." Ca- brera, 1961:468— classification; distribution. Holotype— Male, skin and skull, British Mu- seum (Natural History) no. 2.1.1.90, collected 15 July 1901 by Perry O. Simons, no. 1504. Type Locality (Fig. 2)— Choquecamate, upper -1ERSHKOVITZ: SOUTH AMERICAN HOCICUDO 33 Rio Secure, Cochabamba, Bolivia, 16°55'S, 66°37'W, 4000 m. Distribution — Andes, from Puno, Peru, throughout Cochabamba, Bolivia, into north- western Argentina in the departments of Tarija, Jujuy, and Salta. Original Measurements of Holotype— HB, 130; T, 102; HF, 25/27.5; E, 18; GSL, 32.5; ZB, 14.2; IB, 5.6; ML, 4.9. Remarks— Oxymycterus paramensis may be the medium-size species sympatric with the larger O. inca throughout most of its range. Oxymycterus platensis Oxymycterus platensis Thomas, 1914:244. Thomas, 1917:100-ARGENTINA: Buenos Aires (Isla Ella, delta of Rio Parana). Massoia, 1961:124— AR- GENTINA: Buenos Aires (Punta Lara, Rio de La Plata); habits; measurements. 0[xymycterus]. platensis, Reig, 1987:861 — may be subspecies of O. rufus. Carleton, 1973:15, stomach morphology. [Oxymycterus] platensis, Musser and Carleton, 1993: 727— synonym of O. rufus. Oxymycterus rufus platensis, Cabrera, 1961:469 — classification; distribution. Crespo, 1962:1 16— AR- GENTINA: Buenos Aires (Azul). Oxymycterus rutilans platensis Massoia and Fornes, 1969:3 16- ARGENTINA: habits; reproduction. Crespo, 1964b:62- ARGENTINA: Cordoba (Ya- cambo). Crespo, 1964a: 101 -ARGENTINA: Bue- nos Aires (Abra de Ventana; Tornquist, 500 m). Massoia and Fornes, 1964:30- ARGENTINA: Buenos Aires (Delta del Parana); in owl pellets; local name hocicudo rojuzochico. Holotype— Male, skin and skull, British Mu- seum (Natural History) no. 99. 1 0.4. 1 , collected 24 June 1896 by C. Spegazzini. Type Locality (Fig. 2)— Ensenada, Rio Santi- ago, La Plata, Buenos Aires, Argentina, 34°5 1 'S, 57°55'W, sea level. Distribution— States of Buenos Aires and Cor- doba, Argentina. Original Measurements of Holotype— HB, 140; T, 111; HF, 28 (su); E, 16.5; GSL, 36; CB, 33.2; ZB, 17.2; IB, 6.2; ML, 5.4. Remarks— Treatment of platensis as a synonym or subspecies of Oxymycterus rufus is arbitrary. Oxymycterus quaestor Thomas Oxymycterus quaestor Thomas, 1903b:226 — BRA- ZIL: type; Santa Catarina; Rio de Janeiro (Tere- zopolis). Davis, 1945:280- BRAZIL, reproduc- tion. Vieira, 1 953: 1 44 - BRAZIL: Sao Paulo (Icatu, Estrada Noroeste; Paranapema); measurements. Oxymycterus hispidus quaestor Cabrera, 196 1:467 — classification; distribution. Carvalho, 1 965:25 1 — BRAZIL: Sao Paulo (Boraceia). 0[xymycterus]. quaestor, Carleton, 1973:15 — stom- ach morphology. [Oxymycterus] quaestor, Musser and Carleton, 1993: 727— synonym of O. hispidus. Oxymycterus questor [sic], Gyldenstolpe, 1932: 131 — characters; measurements of holotype. Holotype— Female, skin and skull, British Mu- seum (Natural History) no. 3.7.1.80, collected 2 November 1901 by Alphonse Robert, no. 892. Type Locality (Fig. 2)— Roca Nova, Serra do Mar, Parana, Brazil, 25°30'S, 48°50'W, 1000 m. Distribution— Southeastern Brazil in the States of Rio de Janeiro, Sao Paulo, Parana, and Santa Catarina. Original Measurements of Holotype— HB, 140; T, 100; HF, 30/34; E, 22; GSL, 39.5; ZB, 16.2; IB, 6.1; ML, 5.8. Remarks— Oxymycterus quaestor has been classified by Cabrera ( 1 96 1 , p. 467) as a subspecies of O. hispidus, but evidence for this decision was not forthcoming. Oxymycterus roberti Thomas (Fig. 21, skull, molars) Oxymycterus roberti Thomas, 1 90 1 b:530. Avila Pires, 1960:39 (part), fig. p. 45 (skull)- BRAZIL: Minos Gerais (Conceicao do Mato Dentro, NNE Lagoa Santa); characters, taxonomy; Oxymycterus rufus Winge (not Fischer), a synonym. Cabrera, 1961: 468— distribution. Musser and Carleton, 1993:727— listed; distribution. Holotype— Male, skin and skull, British Mu- seum (Natural History), no. 1.1 1.3.51, collected 5 July 1901 by Alphonse Robert, no. 741. Type Locality (Fig. 2)— [Paranahyba], Rio Jor- dao, southwestern Minas Gerais, Brazil, 18°26'S, 48°00'W, 700-900 m. Distribution— In Brazil, western Minas Ge- rais, Goias, and Distrito Federal. Original Measurements of Holotype— HB, 127; T, 1 10; HF, 30/33; ear, 22; bregma to nasal tip, 26; ZB, 15.3; IB, 6.7; ML, 5.2. Remarks— Oxymycterus roberti appears to be a well-defined species intermediate between the large- and small-size hocicudos. 34 FIELDIANA: ZOOLOGY Oxymycterus rostellatus Wagner Hesperomys (Oxymycterus) rostellatus Wagner, 1 842: 361. Wagner, 1843:44— English translation of the original description. H [espe romys]. rostellatus, Wagner, 1843:514, PI. 202A (animal) -BRAZIL. Oxymycterus rostellatus, Giebel, 1857:191, PI. 5, Fig. 4 (skull)— osteology. Gyldenstolpe, 1932:128 — characters; type said to be in "Naturhis. Museum, Vienna, no. 420 9." Vieira, 1953:144-BRAZIL: Sao Paulo (Butantan). [Oxymycterus] rostellatus, Musser and Carleton, 1993: 727 — synonym of rufus. Oxymycterus rosettellatus [sic], Vorontsov, 1979:95, Fig. 47 (upper molars)— alimentary tract. [Hesperomys (Oxymycterus) rufus] rostellatus, Trouessart, 1881:142— classification. Holotype— According to Wagner (1842, p. 361) who described the animal, the holotype is in the Munich Zoological Museum, purchased from Brandt. Gyldenstolpe (1932, p. 128) says the type, i female, is in the Vienna Natural History Mu- seum, no. 420. Type Locality (Fig. 2)— "Brasilia"; restricted 3y Gyldenstolpe (1932, p. 128) to Bahia, this can De Sao Salvador, or Salvador, Bahia, Brazil, 12°59'S,38°31'W, sea level. Distribution— Eastern Brazil. Original Measurements of Holotype— HB, )" 3'" (163); T, 3" 10'" (80 mm). Remarks— The name rostellatus Wagner has >een used reservedly not only because of questions egarding the status of the older name dasytrichus )ut also because of the questionable type locality iscribed to rostellatus. Authors have found such tames as hispidus, rufus, and quaestor more con- venient if not more appropriate geographically. A uture reviser of the genus may locate the holotype )f rostellatus. In a comparison of Oxymycterus quaestor with i speciman said to be "0. rostellatus Wagner, from irazil," Thomas (1903b, p. 227) described its 'posterior nares [as] level with the back of the econd molar," an extreme condition. Oxymycterus rufus Fischer Rat cinquieme ou rat mux, Azara, 1 80 1 , 2:94— French translation of original Spanish description. Hocicudo Azara, 1802:80— Original Spanish descrip- tion. Azara, 1809:328— travels. Le rat roux du Paraguay, Desmarest, 1804:24— ver- sion of original French description. Mus rufus Fischer, 1814:71. Desmarest, 1819:62 — name based solely on the rat roux of Azara. Des- marest, 1822:305 -PARAGUAY. Rengger, 1830: 230— PARAGUAY (Asuncion); characters. Tate, 1932a:l— rat roux of Azara, an Oxymycterus. M[us]. rufus, Brants, 1827:142— characters; synony- my. Mus rufus, Langguth, 1966:285— M[us]. rufus Fischer, 1814 (p. 21) available for the rat roux of Azara. Hesperomys rufus, Burmeister, 1954:183— characters; taxonomy; Mus fossorius Lund a synonym. H[esperomys]. rufus, Bertoni, 1914:73 — PARA- GUAY: Puerto Bertoni. H[esperomys]. rufus, Ihering, 1885:424— baculum. [Hesperomys (Oxymycterus)] rufus, Trouessart, 1881: 142. H[esperomys]. (Oxymycterus) rufus, Thomas, 1884: 450 — classification. Hesperomys (Oxymycterus) rufus, Ihering, 1894:21 — BRAZIL: Sao Paulo. Tullberg, 1900:249, PI. 56 (foot), PI. 1 5 (skull), PI. 28 (molars), PI. 30 (scapula), PI. 36 (palate), PI. 41 (stomach), PI. 45 (intestine)— characters. Yepes, 1938:53 — part, distribution. Vieira, 1953:145-part, BRAZIL: Sao Paulo (Pe- nis; Sao Sebastiao). Oxymycterus rufus, Pelzeln, 1883:75 — BRAZIL: Sao Paulo (Harare). Winge, 1888:36, PI. 1, Figs. 10, 11 (head, hind foot), PI. 2, Fig. 14 (skull)- BRAZIL: Minas Gerais (Lagoa Santa, Recent and Pleisto- cene); characters; comparisons; measurements. J. A. Allen, 1 9 1 6:572-BRAZIL: Matto Grosso (Cam- pos Novos). Shufeld, 1926:563, PI. 1, Fig. 2, PI. 3, Fig. 8— BRAZIL: Santa Catarina, characters; os- teology. Yepes, 1938:53— part, distribution. Vieira, 1953: 1 45 -part, BRAZIL: Santa Catarina (Colonia Hansa). Hooper and Musser, 1964:291— glans pe- nis. Vorontsov, 1979:20, 41, 231, Fig. 100 (stom- ach)—alimentary tract, morphology, function. Musser and Carleton, 1993:727— synonyms: das- ytrichus, platensis, rostellatus, rutilans. 0[xymycterus]. rufus, Thomas, 1914:244— ARGEN- TINA: Corrientes (Goya specimens regarded as typ- ical). Yepes, 1935:232- ARGENTINA: EntreRios; Corrientes. Oxymycterus rufus rufus, Cabrera, 1961:469— classi- fication. Oxymycterus rutilans, Hershkovitz, 1959:339— na- sutus Waterhouse regarded a synonym. Vaz-Fer- reira, 1960:66 -URUGUAY: Laval lega; Maldona- do; near streams and banados. Reig, 1964:213— ARGENTINA: Buenos Aires (General Pueyrredon; Mar Chiquita; Baliza San Andres; Arroyo Brusqui- tas; Arroya Corrientes), population dynamics; hab- itats; diet. Reig, 1965:208- ARGENTINA: Buenos Aires (SE coast); associations; embryos (avg. 4). Reig, 1965:2 13- ARGENTINA: Buenos Aires (Arroyo Brusquitas; Arroyo Loberia; Baliza Cantu; Arroyo Corrientes; Baliza San Andres; Petit Hotel; Arroyo Chapadmalal; Santa Clara del Mar; Serra La Peri- grina; Laguna de los Padres); population fluctua- tions; habits; habitat. Voss and Linzey, 198 1: 14 — male accessory glands, morphology. [Mus?] rutilans Illiger, 1815— nomen nudum. M[us]. rutilans Olfers, 1818:209— characters. 0[xymycterus]. rutilans, Carleton, 1973:15, Fig. 5 (stomach)— stomach morphology. Oxymycterus nasutus Bertoni (not Waterhouse), 1914: 1ERSHKOVITZ: SOUTH AMERICAN HOCICUDO 35 73-PARAGUAY. Vieira, 1953:143-ARGEN- TINA; Entre Rios; measurements. [Oxymycterus] nasutus Yepes (not Waterhouse), 1935: 232— ARGENTINA: Entre Rios (Yaguarasapa); lo- cal name anguya pittha. uals. Notwithstanding, nine names (Table 1) have been erected for the large-size Atlantic Division hocicudo. Most of them, however, are now re- garded as synonyms of O. hispidus for lack of bet- ter knowledge of O. rufus or O. hispidus. Holotype— Not known to be in existence, name based on the rat cinquieme ou rat roux of Azara. Type Locality (Fig. 2)— "Paraguay," in the original description, but the hocicudo (i.e., the rat roux of the French edition) was said to have been found at 32°30'S in the Rio Parana drainage sys- tem; this restricts the type locality to northern Entre Rios, Argentina. Distribution— Recorded as rufus and rutilans from Paraguay and Uruguay and in Brazil from Santa Catarina, Mato Grosso, occasionally from Sao Paulo, and Minas Gerais; if conspecific with Oxymycterus hispidus, its geographic distribution becomes the same (p. 30). Original Measurements of First Type— TL, 8'/2" (230); T, Vh" (95); HF, 13'" cu (35.2); second type, TL, 9Va" (250). Remarks— The original description of rufus, the rat cinquieme ou rat roux, was based on a large individual preserved in brandy and presented to Azara (1801, p. 94) by his friend Nosedo. Place of capture was not given. The hocicudo or long- nose rat of the Apuntamientos, Azara's ( 1 802, p. 80) original Spanish account of the translated French version, was shot by Azara about 9:00 a.m. while shotgun fishing along a stream at 32°30'S in the Rio Parana basin. The mouse was spied eating a piece of meat Azara had casually thrown away and which landed at the entrance to the rodent's nest. Azara added that another specimen pre- served in spirits was brought to him by his friend Nosedo. It was said to be indistinguishable from the first in size, proportions, and coloration. Although the name Oxymycterus rufus is based on Azara's rat roux, the locality 32°30'S belongs to its conspecific of the Spanish account. To avoid strife and confusion, the point 32°30'S is herewith transferred to the rat roux and becomes, as it has been, the type locality. A dominantly reddish, large-size hocicudo from Entre Rios, Argentina, has not been seen by me in the course of this study. This leaves moot the question of relationship between Oxymycterus ru- fus and the next older named O. hispidus. Either or both names have been regarded as available for the large-size coastal hocicudo. The reddish color ascribed to the original specimen of O. rufus is an erythricism of common occurrence in many pop- ulations that may also include melanistic individ- Habits and Habitat What is known of habitat and behavior of the 23 forms of hocicudos, or long-nose mice (also hog-nose mice), is derived from observations of the large-size Oxymycterus rufus (or hispidus), the medium-size O. roberti, and the smaller O. na- sutus, all of the Atlantic Division, and the medi- um-size O. paramensis of the Andean Division. Borchert and Hanson (1983) observed the ef- fects of fire and flooding on the small mammals of the cerrado or scrub savanna of Brasilia, D.F., in central Brazil. Oxymycterus roberti was one of seven sigmodontine species live-trapped in the valley side wet campo, also known as campo limpol or brejo, a formation of grasses, sedges, and shrubs. These wet campos may flood seasonally on a small j scale, or are permanently flooded. Captures werei 42% Bolomys lasiurus ("Zygodontomys lasi-\ urus"), 31% Oxymycterus roberti, 20% "Plectomysl paludiocola," a nomen nudum later described as] Akodon lindberghi. Also captured were OligoryA zomys eliurus, Akodon arviculoides [= A. cursor], and Calomys callosus. The preferred habitat of Oxymycterus roberti was the wet campo. It was also taken in grassy margins along the wet campo transition and wetj campo-gallery forest border. Borchert and Hanson) never captured the mouse in gallery forest. Theyj found O. roberti equally abundant on burned and untouched areas. The examined stomachs con-j tained 74% insects, earthworms, and termites. The! population density of O. roberti fluctuated least of j the three most abundant species mentioned above, j Activity was about equally nocturnal and diurnalj where fire had not destroyed the cover. Lacher et al. (1989) studied the structure of aj small mammal community in two contiguous butj floristically different savannas in the same cerradoi worked by Borchert and Hanson. The drier grass-' land, they found, had fewer species and 38% ofj the overall population density of the more humid grassland. Microhabitat generalist Bolomys lasi-> urus was abundant in both formations. The mi-, crohabitat specialist Oxymycterus roberti pre-, ferred the more humid habitats with soft soil and; thicker grass cover. An investigation of Oxymycterus roberti pre- 36 FIELDIANA: ZOOLOGY dation on eight species of cerrado termites was made by Redford (1984). It was shown that most hocicudos preferred termite species with soldiers whose defense was mandibular biting but with some chemical backup. Least preferred were ter- mites depending entirely on chemical defense. Species classified as intermediate in preference used both defenses. Oxymycterus roberti was called a burrowing mouse by Redford, following a long tradition of vernacular usage by mammalogists. I saw no signs of burrowing by Oxymycterus when trapping this species and other sigmodontines in the same area. A 7-month study of recapture of small mam- pals in the same habitat by Nitikman and Mares (1987:85) took into account two species of small [marsupials and eight sigmodontine rodents, in- cluding Oxymycterus roberti. The five individuals of this species captured were described as char- acteristic residents of brejos, the permanently in- undated savanna that borders gallery forest. None of the marked animals seemed to be resident in the study area. Three were caught only once and none was caught in more than one trapping period. All but one capture was made in part of the grid dominated by bracken fern; one individual was caught inside gal- lery forest. Captures were made in four of the seven months of the study. The species however was scarce during both wet seasons. Other workers, including myself, found Oxymyc- terus abundant at all times. The big reddish hocicudo, Oxymycterus rufus, of Buenos Aires Province, observed by Reig ( 1 964) was found in every habitat type whether meadow, tall grass, woodland, streamside, or rocky out- croppings. The population density fluctuated con- siderably throughout Reig's investigations. Cen- jsused together with Akodon azarae, it peaked at 84% of total captures of small mammals of the Istudy area. In January, however, the O. rufus pop- ulation dropped to 64.7% of captures, that of A. azarae to 24.4%. Trapping was in open and wood- ed areas. In more humid sites, the catch was 58.7% for O. rufus and 33.3% for A. azarae. Retrapping the first locality in April yielded 4 1 .8% for O. rufus ind 41.1% for A. azarae. In another area at about :he same time, the population density of O. rufus vas 13%, whereas A. azarae abounded at 69.5%. The cause or significance of the contrasting den- iities was not explained, but habitat and resource itilization of the two species are different. Stom- ich contents of O. rufus consisted mostly of mol- usks, earthworms, myriopods, and bits of rodents. The diet of A. azarae, in contrast, is mostly her- bivorous. The behavior of Oxymycterus rufus at Punta Lara, Buenos Aires, Argentina, was closely fol- lowed by Kravetz (1973). It was most abundant where cover was densest, a logical correlation. Maximum population density occurred in winter, minimum in summer (February). Breeding began late in September and continued into May. Most of the population was renewed annually. Activity was mostly nocturnal with a peak between 1 7 and 21 hr, another between 3 and 9 hr. The sigmo- dontines captured in a bamboo patch with an area of 3200 square meters were 67 Oxymycterus rufus, 16 Scapteromys tumidus, 13 Akodon azarae, 1 Rattus rattus, and 1 Oryzomys flavescens. Hoci- cudo stomachs contained 20% oligochetes, 60% arthropods, 5% other animals, 15% vegetation (mostly seeds and pollen), and traces of mice. An 1 8-month ( 1 968-1 970) survey of the biology of the pampa rodents of the Balcarce area in Bue- nos Aires Province, Argentina, was made by Dalby (1975). His comparisons of life histories of Oligo- ryzomys nigripes, Akodon azarae, and Oxymyc- terus rufus (as "O. rutilans^) with those of North American ecological equivalents were most infor- mative. The hocicudo, Dalby noted, actually had no biological equivalent, but its habits coincided in many respects with those of the North American grasshopper mouse Onychomys and the shrew Blarina. Its biomass in the study area was highest, about 40% of all myomorphs captured. Mean and extreme weights in grams for scrotal males were 92.4 ± 1.56 (62-125) 82 samples; for perforate females, 76.2 ± 2.56 (46-110) 39. Sex ratio for trapping between 1 December 1968 and 30 April 1970, compared at 36 intervals, averaged 69.2% for females. Breeding occurred at all seasons, lit- ters averaged 3.1 young, weaning 14 days, and sexual maturity 3 months. Survival rate for mice was high, 70-76% in summer, 92% in winter, con- trasting with that of the Punta Lara O. rufus. Dalby noted tail autotomy in Oxymycterus, a common event among rodents with loose-fitting tail skin or, as in spiny rats (Echimyidae) and other vertebrates, loose caudal vertebral joints. Many of the Oxymycterus I collected were indeed "bobbed." As a rule, the skinned, injured portion of the tail dries and becomes detached, or it may be bitten off by its owner. Dalby found no burrows or runways attributable to the hocicudo, "contrary to its common name of burrowing mouse." In his opinion the "long foreclaws and shrew-like pointed nose function in digging and rooting for subsurface invertebrates." In comparing an Argentine Oxymycterus rufus with the semi-aquatic Scapteromys tumidus, Mas- IERSHKOVITZ: SOUTH AMERICAN HOCICUDO 37 soia and Fornes (1964a, p. 294) noted that the hocicudo swims only to survive. Submerged, its fur becomes completely soaked. A female they captured had six embryos at term. The type specimens of the Argentine Oxymyc- terus akodontius were two of five snap-trapped by E. Budin. The others, he said (in Thomas, 1921, p. 6 1 5), were eaten by rats, their skulls totally de- stroyed. When rats attack a dead specimen they always com- mence by eating the brain. One of these specimens was caught in a Tuco-tuco [Ctenomys] hole. I have observed both these animals dig their holes like the Yeoxus [Geoxus] of the south, making small hillocks of earth over them. The burrows are round and clearly visible. Hocicudos live in humid places among the hills, in the thickest parts of the woods. These Budin notes were either garbled or mis- attributed. Oxymycterus has never been seen bur- rowing as alleged by Budin. Furthermore, sig- modontines that live, as reported by Budin, "in humid places ... in the thickest parts of the woods" do not burrow. Wied-Neuwied (1826) found the holotype of Oxymycterus dasytrichus along the Rio Mucuri in Bahia, Brazil. He also observed this mouse in the deep forest along the shore of Lago d'Arara. It seemed to live underground, he reported, but he could not determine whether or not it burrowed. Nowhere in eastern Brazil, he declared, "have I seen signs of burrows." Burmeister (1854, p. 183) also thought Oxy- mycterus rufus was a burrower because its ap- pearance suggested similarities with microtines. Life history studies of the rodents of Uruguay by Barlow (1969) included Oxymycterus nasutus (his O. rufus nasutus), misnamed "burrowing mouse." Barlow found the hocicudo in such im- probable burrowing resorts as wet meadows with stands of bunch grass (Paspalum sp.); tall grass adjacent to streams and rivers; and drier parts of marshy areas among clumps of pampas grass (Cortaderia selloand) outside the frequently inundated zone. Although taken in the same trapline with the hydrophilic Scapteromys tumidus and Bolomys ob- scurus, O. rufus [= nasutus] was most often taken in the slightly drier places frequented by Oligoryzomys flavescens, with which it was commonly associated. O. rufus [= nasutus] constructs no runways of its own but uses those of Cavia pamparum, the broad trails of Hydrochoerus hydrochaeris in stands of tall bunch grass, or natural pathways through such vege- tation. Odour. An example caught in a live trap smelled strongly of an odour distinctly reminiscent of acrolein (acrylic aldehyde), a chemical used commercially to warn of leakage of toxic odourless gases from me- chanical systems .... This acrid, penetrating scent may serve a similar warning function in these rodents, possibly serving to discourage potential predators. Food Habits. These mice are primarily entomoph- agous .... Glandular and structural modification of the stomach include a thickening of the fundus. This and other characters presumably associated with the entomophagous habit are shared with North Ameri- can grasshopper mice (Onychomys) .... The contents of 12 stomachs analyzed in detail comprised inver- tebrated material, on the basis of the per cent of stom- achs containing the item, as follows: Coleoptera (Tenebrionidae, Scarabaeidae, Lampyridae, Chryso- melidae, Carabidae, Pselaphidae, Dytiscidae) 100.0%; Hymenoptera (Formicidae) 83.3%; Diptera (Sciari- dae, Bibionidae, Chironomidae, Tipalidae) 83.3%; Hemiptera (Timgidae, Pentatomidae) 4 1 .6%; Orthop- tera (Gryllidae, Acrididae, Gryllacrididae) 25.0%; Lepidoptera (Noctuidae, unidentified families) 25.5%; Homoptera (Cicadellidae, Fulgoridae) 16.6%; Oligo- chaeta (earthworms) 33.3%; Chilopoda (centipedes) 25.0%; Gatropoda (slugs) 16.6%; Aranea 16.6%; plants (unidentified fragments) 16.6%. In many marshy places Scapteromys tumidus and Oxymycterus rufus [= nasutus] are the principal in- sectivorous rodents. The two differ in microhabitat preferences and time of activity; Oxymycterus rufus [= nasutus] is primarily diurnal, although specimens have been trapped at dusk and in the early evening. Consequently competition with the nocturnal Scap- teromys tumidus would probably be minimal. Breeding. Available data indicate that the breeding season of this mouse is protracted in Uruguay. Fe- males were taken only in February, March, May, Sep- tember and December but samples for each of these months included pregnant or lactating individuals. Likewise, males with scrotal testes of breeding size (9 x 6 mm and larger, as demonstrated by histologic analysis of testicular biopsies) were found in the above months. Seven gravid females averaged 2. 1 (1-4) em- bryos ranging in length from 4 to 29 mm in length. One embryo was in the left uterine horn and three were in the right. Massoia and Fornes (1964, p. 294) reported a female with six embryos. Oxymycterus ru- fus [= nasutus] possesses eight nipples, and litters larg- er than those indicated by the embryo counts might be expected. The only behavioral report of an Andean Di- vision hocicudo (Oxymycterus paramensis) dis- counting that of Budin (above) is by Mares (1981, p. 181), as follows: This is one of the most shrew-like of the akodont rodents and inhabits the forest floor in mesic areas of north-central Salta. Some specimens were taken under logs in dense verdant second growth. One was taken from the sand-rock embankment of the Rio Pescado. This is an uncommon species in northeast Argentina and appears to be nocturnal. It probably is limited to the northern wet forests, although it possibly will be found in mesic forest enclosures in central, or even south-central Salta. Two individuals captured in September were not breeding— one had small abdominal testes, whereas the other had large (10.4 m length) inguinal testes. 38 FIELDIANA: ZOOLOGY \ Nothing in the above account would lead one to believe that Oxymycterus paramensis is a bur- rower or that its habitat is suitable for burrowing, yet Mare's account of the species is headed "ho- cicudo parameno-burrowing mouse." Remarks— In my short experience live-trap- ping the Brazilian Oxymycterus rufus, O. roberti, and one or two undescribed short-clawed species related to O. (?) iheringi, I saw no burrows or signs of any that may have been made or used by these sigmodontines. The long manual claws, seemingly burrowing tools, are used in foraging or for grap- pling subsurface organisms and tearing apart cov- ered termite runways. Their mobile snouts aid in the search and probably for rooting. Summary The South American long-nose sigmodontines of the genus Oxymycterus, called hocicudos in Spanish and Portuguese, are small to medium-size terrestrial mice, with long front claws, tail shorter than head and body combined, a habitat prefer- ence for damp ground and tall grass, and a diet of mostly arthropods and molluscs. They are not nat- ural climbers, swimmers, or burrowers. Hocicudos are confined to the wide part of the continent between the south bank of the Rio Ama- zonas-Solimdes-Maranon and north bank of the Rio Parana system, with the Atlantic Ocean on the east and the Andes on the west to about 4000 m above sea level. Demographically, Oxymycterus is one of the most speciose and populous of ako- dontines. It has no phylogenetic ties with sigmo- dontines to the north or south of its known range. The suggestion that Podoxomys, Microxus, Le- noxus, Abrothrix, and Oxymycterus formed a nat- ural tribal group has been roundly rejected. The 23 described hocicudos, the first in 1801 by Felix Azara, had never been organized into phylogenetic or geographic order. Material avail- able in the Field Museum and some borrowed from other institutions, however, possessed the information necessary for the arrangement of Ox- ymycterus into two branching divisions, one At- lantic, the other Andean, the intervening separa- tion filled by the Amazonian floodplain. Each division, in turn, consists of a number of species more or less comparable in size from small to large without overlap if all critical size dimensions are weighed together. The northern and southern fluvial boundaries of the genus are not mere geographic barriers. They suggest historic markers that point to the possi- bility that Oxymycterus, if not all other complex penis type sigmodontines, are Patagonian (or Weddellian) in origin. In their northward spread (Fig. 2) hocicudos only now are infiltrating the Amazonian basin, which until Pleistocene time had been a huge freshwater lake. The foregoing scenario is conjectural but per- haps less so than the presumption of a sigmodon- tine entry into South America from North Amer- ica with no evidence of prior existence on that continent by the progenitor(s). Evidence of exis- tence of South American descendants and their pathway to where differentiation of Oxymycterus took place is also lacking. The phylogenetic and biogeographic data of South American sigmodontines, particularly ako- dontines, adduced by Smith and Patton (1993) from mitochondrial DNA with hypothetical pa- leontological assistance may have no support for my offhand proposition of a southern origin and northward spread. On the other hand, they do not justify their faith in a North American origin of complex penis type sigmodontines (as distin- guished from neotomyini-peromiscini cricetids) and invasion of South America via the Panaman- ian route. A third option for consideration in the possible origin of South American sigmodontines is the rafting of some of the basic stock from Africa, maybe in the Miocene (Hershkovitz, 1972, pp. 324, 354). The African origin of Neotropical ca- viomorph rodents and platyrrhine monkeys is rel- evant. This paper is not a taxonomic revision of the genus Oxymycterus. It is a compilation and or- ganization of mostly available information essen- tial for the description of the new species of Ox- ymycterus. The organized data at the same time throw light on certain perhaps unsuspected aspects of sigmodontine origins, dispersal, and behavior. Acknowledgments The scientific illustrations in this paper were executed by project artist and technical assistant Kathleen Kozol Telfer; the manuscript was com- puterized by volunteer Eunice Hoshizaki. My thanks to both for their contributions. My thanks are also extended to Diane White and Linda S. Dorman of the Field Museum's Department of Photography, headed by John Weinstein. Re- HERSHKOVITZ: SOUTH AMERICAN HOCICUDO 39 search Associate Barbara E. Brown helped in many ways to advance completion of this paper. The critical review of this manuscript by my colleague Bruce D. Patterson, Philip Myers of the University of Michigan, and two anonymous re- viewers is deeply appreciated. Financial support for the southeastern Brazilian project is received from the Barbara E. Brown Mammal Research Fund. Literature Cited Allen, J. A. 1990. On the mammals collected in south- eastern Peru by Mr. H. H. Keays with descriptions of new species. Bulletin of the American Museum of Nat- ural History, 13: 219-227. . 1916. Mammals collected on the Roosevelt Brazilian Expedition with field notes by Leo E. Miller. Bulletin of the American Museum of Natural History, 35: 559-610. Avila Pires, F. E. 1960. Roedores coleccionados na region de Lagoa Santa, Minas Gerais, Brasil. Arquivos do Museo Nacional, Rio de Janeiro, 50: 25^5. . 1965. The type specimens of Brazilian mam- mals collected by Prince Maximilian zu Wied. Amer- ican Museum Novitates, 2209: 1-21. Azara, F. 1801. Essais sur l'histoire naturelle des quadrupedes de la province du Paraguay. Paris, 2: 499 pp. . 1 802. Apuntamientos para la historia natural de los quadrupedos del Paraguay y Rio de la Plata. Madrid, 2: i-iii + 328 pp. 1809. Voyages dans lAmerique Meridionale . . . Walkenaer, ed., Dentu, paris, 4 vols, with atlas. Barlow, J. C. 1969. Observations on the biology of rodents in Uruguay. Life Sciences Contributions, Roy- al Ontario Museum, 75: 1-59. Bertoni, A. de W. 1914. Fauna Paraguay. Catalogos sistematicos de los vertebrados del Paraguay. Des- cription fisica y economica del Paraguay (Asuncion), 59(1): 1-86. Borchert, M., and R. L. Hansen. 1983. Effects of flooding and wildfire on valley side wet campo rodents in central Brazil. Revista Brasileira de Biologia, 43(3): 229-240. Brants, A. 1827. Het geschlacht der Muizen, Berlin, i-xii + 1 90 pp. Budin, E. in O. Thomas. 1921. On a further collection of mammals from Jujuy obtained by Sr. Budin. Annals and Magazine of Natural History (9), 8: 615. Burmeister, H. 1854. Systematische Uebersicht der Thiere Brasiliens. Erster Theil Saugethiere (Mam- malia). Georg Reimer, Berlin. . 1879. Description physique de la Republique Argentine d'apres des observations personnelles et etrangeres. (Traduit de lAllemand avec le concours de E. Daireaux). Buenos Aires, 3(1): vi + 555 pp. Cabrera, A. 1961. Catalogo de los mamiferos de America del Sur. Revista Museo Argentino de Cien- cias Naturales, "Bernardino Rivadavia," 4(2): 309- 732. Campbell, K. E., Jr. 1990. The geologic basis of bio- geographic patterns in Amazonia, pp. 34-35. In Pe- ters, G., and R. Hutterer, eds., Vertebrates in the Trop- ics. Proceedings of the International Symposium on Vertebrate Biogeography and Systematics in the Trop- ics. Koenig Zoological Research Institute and Zoolog- ical Museum, Bonn. Carleton, M. D. 1973. A survey of gross stomach morphology in New World Cricetinae (Rodentia, Mu- roidea), with comments on functional interpretations. Miscellaneous Publications, Museum of Zoology, University of Michigan, no. 146. Crespo, J. A. 1 962. Una mutation de pelage en el zorro Colorado Dusicyon culpaeus (Molina). Neotropica, 8(27): 115-116. Dalby, P. L. 1 975. Biology of pampa rodents, Balcarce Area, Argentina. Publications, Michigan State Uni- versity Biological Series, 5(3): 155-271. Davis, D. E. 1945. Annual cycle of plants, mosquitos, birds and mammals in two Brazilian forests. Ecolog- ical Monographs, 15: 243-295. Desmarest, A. G. 1804. Nouveau Dictionaire d'His- toire Naturelle. Paris, 24(6): 5-58. . 1819. Article Rat. Nouveau Dictionaire d'His- toire Naturelle. Paris, 29: 40-8 1 . 1 822. Mammalogie ou description des especes de mammiferes. Second partie, i-viii + 277-555. Echave Llanos, J. M., and C. A. Vilchez. 1964. An- atomica microscopica del estomogo del raton hoci- cudo Oxymycterus rutilans. Revista de la Sociedad Argentina de Biologia, 40: 187-192. Ellerman, J. R. 1 94 1 . The families and genera of liv- ing rodents with a list of named forms by R. W. Hay- man and G. W. C. Holt. Vol. II, Family Muridae. British Museum (Natural History), London. Emmons, L. H., and F. Feer. 1990. Netropical Rain- forest Mammals. A Field Guide. University of Chi- cago Press, Chicago, 281 pp., 7 pis. Figueira, 1894. Contribution al conocimiento de la fauna Uruguaya. Anales Museo Nacional, Montevi- deo, 2: 187-217. Fischer von Waldheim, G. 1813-1814. Zoognosia tabulis synopsis illustrata, in usum prae lectionum Academiae imperialis medico— chirurgicae mosquen- sis edita, vol. 2, 1813, vol. 3, 1814, Moscow. Giebel, C. G. 1857. Odontographie, Vergleichende Darstelung des Zahnsystems der lebenden und fossilen Wirbelthiere. Abhandlungen Naturwissenschaften Verein Sachsen and Thiiringen, Halle. Goodwin, G. 1953. Catalogue of the type specimens of Recent mammals in the American Museum of Nat- ural History. Bulletin of the American Museum of Natural History, 102: 221-41 1. Gray, J. E. 1843. List of the Specimens of Mammals in the Collection of the British Museum. British Mu- seum (Natural History), Lodon, xxviii + 216 pp. Gyldenstolpe, N. C. G. 1932. A manual of neotrop- ical sigmodont rodents. Kiingliche Svenska Veten- skapsakad, Handlinger Series 3. B, no. 3: 1-164. Hensel, R. 1873. Beitrage zur Kentniss der Saugethiere 40 FIELDIANA: ZOOLOGY Siid-Brasiliens. Abhandlungen K. Akademie Wissen- schaften, Berlin, 1872: 1-130. Hershkovttz, P. 1959. Nomenclature and taxonomy of the neotropical mammals described by Olfers, 1818. Journal of Mammalogy, 40: 337-353. . 1966. South American swamp and fossorial rats of the Scapteromyine group (Cricetinae, Muridae) with comments on the glans penis in muroid taxon- omy. Zeitschrift fur Saugetierkunde, 31(2): 81-149. . 1972. The Recent mammals of the Neotropical region: A zoogeographic and ecological review, pp. 31 1^31. In Keast, A., F. L. Err, and B. Glass, Evo- lution, Mammals and Southern Continents. State Uni- versity of New York, Albany. . 1993. A new central Brazilian genus and spe- cies of sigmodontine rodent transitional between ako- donts and oryzomyines with a discussion of muroid molar morphology and evolution. Fieldiana: Zoology, n.s., no. 75: 1-18. Hinojosa, F., S. Anderson, and J. L. Patton. 1987. Two new species of Oxymycterus (Rodentia) from Peru and Bolivia. American Museum Novitates, no. 2898: 1-17. Hooper, C. T., and G. G. Musser. 1964. The glans penis in neotropical cricetines (Family Muridae) with comments on classification of muroid rodents. Mis- cellaneous Publications, Museum of Zoology, Uni- versity of Michigan, no. 123. Ihering, H. 1885. Zur Kentniss der brasilianischen Mause und mause plagen. Kosmos, 13(2): 423-437. . 1892. Os mamiferos do Rio Grande do Sul. Annuario Estado do Rio Grande do Sul, 1893, de Graciano A. de Azambuya, pp. 3(96)-30(123) (re- print). . 1 894. Os mammiferos de S. Paulo. Catalogo. "Diario Official," Sao Paulo, pp. 3-30. Illiger, C. 1815. Ueberblick der Saugethiere nach ih- rer Verteilung iiber die Weltheile. Abhandlungen K. Akademie Wissenchaften, Berlin (1804-181 1). . 1815 (1804-1811). Prodromus Systematis Mammalium et Avium. Berlin, viii + 301 pp. Kajon, A. E., O. A. Scaglia, C. Horgan, C. Velazquez, M. S. Merani, and O. A. Reig. 1984. Tres nuevos cariotipos de la tribu Akodontini Rodentia, Criceti- dae. Revista Museo Argentino de Ciencias Naturales "Bernardino Rivadavia," 13(1-60): 461-469. Kravetz, F. O. 1 973. Estudio del regimen alimentario, periodos de actividad y otros rasgos ecologicos en una poblacion de "Raton Hocicudo" (Oxymycterus rufus platensis Thomas) de Punta Lara. Acta Zoologica Lil- loana, 29: 201-212. Lacher, T. E., M. A. Mares, and C. J. R. Alho. 1989. The structure of a small mammal community in a central Brazilian savanna. Advances in Neotropical Mammalogy, 1989: 137-162. Langguth, A. 1966. Application to place on the ap- propriate official list the names given by G. Fischer, 1 8 1 4 to the cricetid rodents described by F. de Azara in the French translation of Essais sur FHistoire Na- turelle des quadrupedes der Paraguay, 1801. Bulletin Zoological Nomenclature, 23: 285-288. Leche, W. 1886. Ueber einige sudbrasilianische Hes- peromys-Arten. Zoologische Jahrbuch, 1: 687-703. Leuderwaldt, H. 1929. Resultados de uma excursao scientifica a Ilha de Sao Sebastiao em 1925. Relacao. Revista do Museu Paulista, 16: 22-65. Mares, M. A., R. A. Ojeda, and M. P. Kosco. 1981. Observations on the distribution and ecology of the mammals of Salta Province, Argentina. Annals of the Carnegie Museum, 50(6): 151-206. Mares, M. A., M. Braun, K. Janet, and D. Gettinger. 1989. Observations on the distribution and ecology of the mammals of the cerrado grasslands of central Brazil. Annals of the Carnegie Museum, 58(1): 1-60. Massola, E. 1 96 1 . Notas sobre los cricetidos de la selva marginal de Punta Lara (Mammalia, Rodentia). Pub- licaciones del Museu Municipal de Ciencias Naturales y Tradicional de Mar del Plata, 1(4): 1 15-134. . 1963. Oxymycterus iheringi : (Rodentia-Crice- tide) nueva especie para La Argentina. Physis, 24(27): 130-136. . 1964a. Notas sobre el genero Scapteromys (Ro- dentia-Cricetidae). I. Sistematica, distribution geo- grafica y rasgos etoecologicos de Scapteromys tumidus (Waterhouse). Physis, 24(68): 279-297. Massoia, E., and A. Fornes. 1964b. Pequenos mami- feros (Marsupialia, Chiroptera y Rodentia) y aves ob- tenidos en regurgitaciones de lechuzas (Stringiformes) del Delta Bonaerense. Delta del Parana, (4) 6: 27-34. . 1969. Caracteres comunes y distintivos de Ox- ymycterus nasutus (Waterhouse) y O. iheringi Thomas (Rodentia, Cricetidae). Physis, 28: 315-321. Matson, J. D., and J. P. Abravaya. 1977. Blarinomys breviceps. Mammalian Species, 74: 1-3. Musser, G. G., and M. D. Carleton. 1993. Family Muridae, pp. 501-755. In Wilson, D. E., and D. H. Reeder, eds., Mammal Species of the World: A Tax- onomic and Geographic Reference. Smithosonian In- stitution Press, Washington, London. Nitikman, L. A., and M. A. Mares. 1987. Ecology of small mammals in a gallery forest of Central Brazil. Annals of the Carnegie Museum, 56(2): 75-95. Olfers, I. 1818. Bemerkungen zu Illiger's Ueberblick der Saugethiere nach ihrer Betheilung iiber die Weltt- heile riichsichtlich der Siidamerikanischen Arten (Spe- cies). In Eschwege, W. L., ed., Journal von Brasilien, Weimar, 15(2): 192-237. Osgood, W. H. 1944. Nine new South American ro- dents. Field Museum of Natural History Zoological Series, 29: 191-204. Patterson, B. D. 1 992a. Mammals in the Royal Nat- ural History Museum, Stockholm, collected in Brazil and Bolivia by A. M. Olalla during 1934-1938. Field- iana: Zoology, n.s., no. 66: 42 pp. . 1992b. A new genus and species of long-clawed mouse (Rodentia: Muridae) from temperate rain for- ests of Chile. Zoological Journal of the Linnean So- ciety, 106: 127-145. Pelzeln, A. 1 883. Brasilische Saugethiere. Resulte von Natterer's Reisen in dem Jahren 1817 bis 1835. K. K. Zoologisch-botanischen Gesellschaft, Wien (Beiheft), 33: 1-140. Perez-Zapata, A., D. Lew, M. Aquilera, and O. A. Reig. 1992. New data on the systematics and kar- yology of Podoxomys roraimae (Rodentia, Cricetidae). Zeitschrift fur Saugethierkunde, 57(4): 216-224. HERSHKOVITZ: SOUTH AMERICAN HOCICUDO 41 Pictet, F.-J. 1843. Second notices sur les animaux nouveu ou peu connus du Musee de Geneve. Memoire 44, Livraison, 2: 27-40; Memoires de la Societe de Physique et Histoire Naturelle de Geneve, mammi- feres, 9, 10: 82 pp. (1841-1844). Redford, K. H. 1984. Mammalian predation on ter- mites: Tests with the burrowing mouse (Oxymycterus roberti) and its prey. Oecologia (Berlin), 65: 145-152. Redford, K. H., and G. A. B. Fonseca. 1986. The role of gallery forests in the zoogeography of the cer- rado's non-volant mammalian fauna. Biotropica 18(2): 126-135. Reig, O.A. 1 964. Roedores y marsupiales del Partido de General Puerredon y regiones adjacentes (Provincia de Buenos Aires, Argentina). Publicaciones del Museo Municipal de Ciencias Naturales de Mar del Plata, 1(6): 203-224. . 1965. Datos sobre la comunidad de pequenos mamiferos de la region costera del Partido de General Puerredon y de los partidos Iimitrofes (Provincia de Buenos Aires, Argentina). Physis, 25(69): 205-21 1. 1987. An assessment of the systematics and evolution of the Akodontini, with the description of new fossil species of Akodon (Cricetidae: Sigmodon- tinae), pp. 347-399. In Patterson, B. D., and R. M. Timm, eds., Studies in Neotropical Mammalogy, Es- says in Honor of Philip Hershkovitz. Fieldiana: Zo- ology, n.s., 39. Rengger, J. R. 1830. Naturgeschichte der Saeuge- thiere von Paraguay. Schweighauserschen Buchhand- lung, Basel, 16 + 394 pp. Sanborn, C. C. 1929. The land mammals of Uruguay. Field Museum of Natural History, Zoological Series, 17(4): 147-165. . 1931. A new Oxymycterus from Misiones, Ar- gentina. Proceedings of the Biological Society, Wash- ington, 44: 1-2. 1947. Catalogue of type specimens of mam- mals in the Chicago Natural History Museum. Field- iana, Zoological Series, 32(4): 209-293. . 1950. Small rodents from Peru and Bolivia. Publicaciones Museo de Historia Natural "Javier Pra- do," Universidad Nacional Mayor de San Marcos (A), Zoologia, (5): 1-16. 1951. Mammals from Marcapata, southeast- ern Peru. Publicaciones Museo de Historia Natural "Javier Prado," Universidad Nacional Mayor de San Marcos (A), Zoologia, (6): 1-26. -. 1953. Mammals from the Departments of Cuz- co and Puno, Peru. Publicaciones Museo de Historia Natural Javier Prado (A) (Zoologia), 12: 1-8. Schinz, H. R. 1821. Das Thierreich eingetheilt nach dem Bau der Thiere als Grundlage ihrer Naturges- chichte und der engleichenden Anatomie von dem Herr Ritter von Cuvier. Vol. 1. J. G. Cott'aschen Buchhandlung. Stuttgart und Tubingen. Cuvier's Das Thierreich, 1: 288. . 1845. Sy stematische Verzeichniss aller bis jetz bekantenn Saugethiere oder Synopsis Mammalium nach dem Cuvier' schen System. Soluthurn, 2: i-iv, 1- 574. Shufeld, R. W. 1926. Obervacoes sobre certos peixes e mammiferos do Brasil e mais particularmente sobre sua osteologia. Revista Museu Sao Paulo, 14: 503- 561, 23 pis. Smith, M. F., and J. L. Patton. 1993. The diversi- fication of South American murid rodents: Evidence from mitochondrial DNA sequence data for the ako- dontine tribe. Biological Journal of the Linnaean So- ciety, 50: 149-177. Tate, G. H. H. 1 932a. The South American Cricetidae described by Felix Azara. American Museum Natural History Novitates, 557: 1-5. . 1 932b. The taxonomic history of the South and Central American akodont rodent genera: Thal- pomys, Deltamys, Thaptomys, Hypsimys, Bolomys, Chroeomys, Abrothrix, Scotinomys, Akodon (Chalco- mys and Akodon), Microxus, Podoxomys, Lenoxus, Oxymycterus, Notiomys, and Blarinomys. American Museum Novitates, 582: 1-32. Thomas, O. 1884. On a collection of Muridae from central Peru. 1 884 Proceedings of the Zoological So- ciety of London, 1884: 447^58. . 1896. On new small mammals from the Neo- tropical region. Annals and Magazine of Natural His- tory, (6) 18: 301-314. . 1900. Descriptions of new rodents from west- ern South America. Annals and Magazine of Natural History, (7) 6: 294-302. . 1901a. New mammals from Peru and Bolivia with a list of those recorded from the Inambari River, upper Madre de Dios. Annals and Magazine of Natural History, (7) 7: 178-190. — . 1901b. On mammals obtained by Mr. Al- phonse Robert on the Rio Jordao, S.W. Minas Gerais. Annals and Magazine of Natural History, (7) 8: 526- 536. — . 1 902a. On mammals from the Serra do Mar, Parana, collected by M. Alphonse Robert. Annals and Magazine of Natural History, (7) 9: 59-64. . 1 902b. On mammals from Cochabamba, Bo- livia, and the region north of that place. Annals and Magazine of Natural History, (7) 9: 125-143. — . 1 903a. New forms of Sciurus, Oxymycterus, Kannabateomys, Proechimys, Dasyprocta and Calu- romys from South America. Annals and Magazine of Natural History, (7) 11: 487^*93. 1903b. New species of Oxymycterus, Thricho- mys and Ctenomys from South America. Annals and Magazine of Natural History, (7) 11: 226-229. . 1909. Notes on some South- American mam- mals, with descriptions of new species. Annals and Magazine of Natural History, (8) 4: 230-242. . 1914. New South-American rodents. Annals and Magazine of Natural History, (8) 14: 240-244. . 1916. Two new Muridae from South America. Annals and Magazine of Natural History, (8) 18: 473- 480. . 1917. On small mammals from the delta of the Parana. Annals and Magazine of Natural History, (8)20:95-100. — . 1918. On small mammals from Salta and Jujuy collected by Mr. E. Budin. Annals and Magazine of Natural History, (9) 1: 186-193. . 1920. Descriptions of new rodents from west- 42 FIELDIANA: ZOOLOGY ern South America. Annals and Magazine of Natural History, (7) 6: 294-302. . 1921. On a further collection of mammals from Jujuy obtained by Sr. Budin. Annals and Magazine of Natural History, (7) 8: 608-617. — . 1925. The Spedan Lewis South American Ex- ploration—I. On mammals from southern Boliva. An- nals and Magazine of Natural History, (9) 15: 575- 582. . 1926. The Spedan Lewis Exploration— II. On mammals collected in the Tarija Department, south- ern Bolivia. Annals and Magazine of Natural History, (9)17:318-328. Tomes, R. F. 1861. Report of a collection of mammals made by Orbert Salvin Esq. F.Z.S. at Duefias Gua- temala; with notes on some of the species, by Mr. Frazer. Proceedings of the Zoological Society, London, pp. 278-288. Trouessart, E.-L. 1881. Catalogue des mammiferes vivants et fossiles. Bulletin Societe Etudes Scienti- fiques, Angers 1880, 10(1): 177-209. Tullberg, T. 1 899. Ueber das System der Nagethiere. Eine phylogenetische Studie. Nova Acta Regiae So- cietatis Scientiorum Upsaliensis (3rd Series), 18: 514 pp., 57 pis. Vaz-Ferreira, R. 1960. Nota sobre Cricetinae del Uruguay. Archivos Sociedad Biologica, Montevideo, 24: 66-75. Vieira, C. O. da C. 1953. Roedores e lagomorfos do estado de Sao Paulo. Arquivos Zoologica Estado Sao Paulo, 8(5): 129-168. VrruLLO, A. D., M. S. Mirani, O. A. Reig, A. E. Kajon, O. SCAGLIA, M. B. ESPINOSA, AND A. PEREZ-ZapATA. 1986. Cytogenetics of South American akodont ro- dents (Cricetidae): New karyotypes and chromosomal banding patterns of Argentinian and Uruguayan forms. Journal of Mammalogy, 67(1): 69-80. ybRONTSOv, N. H. 1962. The ways of food specializa- tion and evolution of the alimentary system in Mu- roidea. Symposium Theriologicum. Proceedings of the International Symposium on Methods of Mammalog- ical Investigation held in Brno, Czechoslovakia from August 26 to September 5, 1960. Czechoslovak Acad- emy of Sciences, Praha. . 1979. Evolution of the alimentary system in myomorph rodents. Nauk, Siberian Branch, 235 pp., 148 figs., 14 pis. Translated from original Russian [published 1967] for the Smithsonian Institution and the National Science Foundation, Washington, D.C., by the Indian National Scientific Documentation Cen- ter, New Delhi, 346 pp., 148 figs. Voss, R. S., and A. V. Linzey. 1981. Comparative gross morphology of male accessory glands among neotropical Muridae (Mammalia: Rodentia) with comments on systematic implications. Miscellaneous Publications,Museum of Zoology, University of Mich- igan, 159: 1-41. Wagner, A. 1 842. Diagnosen neuer Alien Brasilischer Saugethiere. Archiv fur Naturgeschichte, Berlin, (8) 1: 356-362. . 1843. Saugethiere in Abbildungen nach der Natur mit Beschreibungen. In von Schreber, J. S. D., Saugethiere Supplement, Erlangen, 3: xiv + 614. Waterhouse, G. R. 1837. Characters of a new species of the genus Mus from the collection of Mr. Darwin. Proceedings of the Zoological Society, London, 1837: 15-22, 27-29. 1839. The zoology of the voyage of H.M.S. Beagle under the command of Captain Fitzroy, R. N., during the years 1832 to 1836. Part 2, Mammalia, i- ix + i-v + 97 pp. Elder and Company, London. Wied-Neuwied, M. 1826. Beitrage zur Naturges- chichte von Brasilien, vol. 2. Weimar Gr.H.S. priv Landes-Industrie-Comptoirs, 6 1 6 pp. WrNGE, O. 1888. Jordfunde og nulevende Gnavere (Rodentia) fra Lagoa Santa, Minas Geraes, Brasilien. E Museo Lundii, 1(3): 200 pp. Yepes, J. 1933. Nuevos roedores para la fauna argen- tina. Revista Chilena Historia Natural, 37: 46-49. . 1935. Epitome de la sistematica de los roedores argentinos. Revista Instituto de Bacteriologico, Bue- nos Aires, 7(2): 213-269. — . 1938. Disquisiciones zoogeograficas referidas a mamiferos comunes a las faunas de Brasil y Argen- tina. HERSHKOVITZ: SOUTH AMERICAN HOCICUDO 43 >rtd Monk >hilin Ho: Jats of the Philippine Islai no. 69, 1992. 44 pages, 6i \ Taxonon s., no. 70, ;e R. Heaney, Paul D. H a pages, IS illUs., 20 tat )lication numbe is residents add current 1440,$: • ' a bles. H EC KM AN BINDERY INC. NOV 96 u.«i To iW N.MANCHESTER, iound-To-Plu>^ |ND,ANA 46962