FIELDIANA Zoology Published by Field Museum of Natural History New Series, No. 11 NEOTROPICAL DEER (CERVIDAE) PART I. PUDUS, GENUS PUDU GRAY PHILIP HERSHKOVITZ ^^^itli^tff^ neio MusEmA imm March 22, 1982 Publication 1330 NEOTROPICAL DEER (CERVIDAE) PART I. PUDUS, GENUS PUDU GRAY FIELDIANA Zoology Published by Field Museum of Natural History New Series, No. 11 NEOTROPICAL DEER (CERVIDAE) PART I. PUDUS, GENUS PUDU GRAY PHILIP HERSHKOVITZ Curator Emeritus Division of Mammals Department of Zoology Field Museum of Natural History Accepted for publication May 22, 1980 March 22, 1982 Publication 1330 Library of Congress Catalog Card No.: 81-67141 ISSN 0015-0754 PRINTED IN THE UNITED STATES OF AMERICA CONTENTS I. Introduction 1 Material 2 Abbreviations 2 Measurements 2 Review of Cervid Classification with Particular Reference to South American Genera 2 Odocoileine Origins and Radiation 8 Cheiridia 9 Limb Proportions 11 II. Genus Pudu Gray 13 Synonymic History 13 Conservation 14 Taxonomy 15 Generic Characters 15 Generic Comparisons 22 Fawnal Spotting Pattern "23 Ancestral Characters and Relationships 24 Origin, Dispersal, Spedation 25 III. The Species of Pudus, Genus Pudu Gray 43 Key to the Spedes 43 IV. Pudu mephistophiles de Winton 44 Synonymic History 44 Distribution 45 Conservation 48 Vernacular Names 50 Taxonomy 51 Characters 51 Comparisons 57 Spedmens Examined 58 V. Biology of Pudu mq^histophiles 58 Habitat 58 Habits 58 Maintenance in Captivity 59 Ungulate Assodations 59 Reproduction 59 VI. Pudu puda Molina 60 Synonymic History 60 Distribution 62 Conservation 62 Taxonomy 64 Characters 65 Karyotypes 70 Comparisons 70 Spedmens Examined 70 VII. Biology of Pudu puda 71 Habitat 71 Maintenance in Captivity 71 Ungulate Assodations 73 vi CONTENTS Numbers and Aggregations 73 Migrations 73 Locomotion, Play 74 Enemies 74 Diseases and Parasites 74 Food and Water 74 Earthquake Sensitivity 75 Temperature Regulation 75 Reproduction, Growth, Development 75 Longevity 79 Vm. Summary 79 IX. Acknowledgments 79 X. Gazetteer and Explanation of Maps 80 XL Literature Qted 83 LIST OF ILLUSTRATIONS 1. Dispersal and spedation of pudus 2 2. Left forefeet and hind feet of brocket, Mazama rufina, and muntjac, Muntiacus reevesi 6 3. Nares in Odocoileinae and Cervinae 7 4. Artiodactyl hind feet showing gradient of metatarsal bone reduction 10 5. Size relationship and body form of six South American deer 16 6. Body form of four South American deer 17 7. Fadal color pattern of Odocoileinae 18 8. Head characters of smallest South American deer 19 9. Fadal markings of five South American deer 20 10. Front and hind feet with outline of interdigital glands 21 11. Skulls, left side of adult female pudus and brockets 26 12. Skulls, dorsal surface of adult female pudus and brockets 27 13. Skulls, palatal surface of adult female pudus and brockets 28 14. Skulls, left side of adult male pudus and brockets 29 15. Skulls, dorsal surface of adult male pudus and brockets 30 16. Skulls, palatal surface of adult male pudus and brockets; inset, dedduous-like pm^ 31 17. Mandibles of pudus and brockets, lateral surface of left rami 32 18. Upper cheek teeth of adult pudus and brockets, and juvenal cheek teeth of Mazama chunyi 33 19. Lower cheek teeth of adult pudus and brockets, and juvenal cheek teeth of Mazama chunyi 34 20. Lower front teeth of pudus and brockets, labial surface 35 21. Lower front teeth of pudus and brockets, lingual surface 36 22. Permanent and dedduous indsor-canine series in pudus and brockets 37 23. Upper right and lower left second and third molars of pudus 38 24. Forefeet and hind feet of pudus with vestiges of metacarpal bones II and V (telemetacarpalia) 40 25. One-week-old southern pudu 41 26. One-week-old gray brocket 41 27. Red brocket doe with fawn 42 28. Northern pudus, adult male and female 43 29. Northern pudu, adult female 43 30. Geographic distribution of Pudu mqjhistophiles 46 31. Pelage and hair types of Pudu mephistophiles 53 32. Sectioned hair from Pudu mephistophiles 54 33. Saturation in mammalian tegumentary color 55 34. Color of hairs of Pudu mephistophiles and P. puda 56 35. Geographic distribution of Pudu puda 63 36. Southern pudus in Cologne zoo 67 37. Southern pudus in Berlin zoo 67 LIST OF TABLES 1. Homologies of artiodactyl carpal and tarsal bones 11 2. Artiodactyl limb proportions 12 3. External and cranial measurements of adult Pudu mephistophiles 52 4. External and cranial measurements of adult Pudu puda 68 5. Antler cycle in Pudu puda 69 6. Birth records of Pudu puda 72 7. Growth and development of a male Pudu puda in the Cologne zoo 76 8. Chronology of weight increment in Pudu puda 77 9. Three oldest longevity records for Pudu puda 78 10. Composition of milk produced by Pudu puda 79 I. INTRODUCTION The diminutive, spike-antlered deer of the genus Pudu are smallest of living true deer and smallest of New World ungulates. Of the two known species, the northern occurs in the high, equatorial Andes from southern Colombia to central Peru between latitudes 5°N and 12°S, and the southern is found in the southern half of Chile and in bordering parts of Argentina between latitudes 35°S and 53°S (fig. 1). This report is the first of a proposed series on the genera of neotropical Cer- vidae and part of a monograph begun more than three decades ago with the intent of completing a unified work. My preliminary publications in ensuing years include attempts to clarify the nomenclature and taxonomic status of some neotropical deer (1948, 1958a, 1958b), descriptions of a new subspecies and species of brockets of the genus Mazama (1959a, 1959b), and comments on South American cervid origin and distribution (1958c, 1966, 1972). Work on the mono- graph itself, which had already become a sizable manuscript, suffered delays and lengthy interruptions and seemed interminable as a single work. The present plan is to update and complete each generic review as a separate publication, with particular attention paid to taxonomy, zoogeography, and life history of the included species. In this report, the issues of origin, evolution, and supergeneric classification are touched upon mainly for orientation and refer- ence. Full discussion of these topics necessarily awaits completion of studies of both living and fossil forms. Detailed descriptions of more specific subjects, such as dental anatomy, hoofs, antlers, glands, genitalia, and other soft parts, will be presented in those reports where most appropriate or demanding. For example, because simple antlers of pudus and brockets are similar, the attention in this account focuses on other characters, such as limb bones, which distinguish pudus from all other deer. On the other hand, distinction of brockets {Mazama spp.) from their nearest relative, the white-tailed deer {Odocoileus virginianus) , is based largely on antler size and shape. It is proper, therefore, to reserve both description of the morphology and discussion of the evolution of branched antlers from the simple, spike-like form for reports on those particular deer. The final number of the proposed series will include a synthesis. The study of neotropical deer, particularly in the field, has been largely ne- glected. The 11 or so extant species exceed the combined number of spedes of the Odocoileinae and tribe Cervini native to other comparable zoogeographic regions. Man's persecution of deer is persistent and practically uncontrolled. Most of the species have already vanished from large areas of their former range. Under present conditions of mock conservation, some spedes may disappear com- pletely before the end of the century. Fig. 1. Dispersal and spedation of pudus. Dashed line at northeastern base of Andes indicates presumed temperate zone habitat of ancestral spedes during a gladal period. With onset of the postgladal period, pudus presumably moved with their natural temper- ate zone habitat to higher altitudes in the north (arrows) and to higher latitudes in the south (triangles) in the wake of retreating gladers. The widely separated populations then differentiated into Pudu mephistophiles in the north, and the more evolved P. puda in the south. Character for character, the southern spedes can be derived from the northern. Actual geographic range of P. mqjhistophiles, if not continuous, probably is — or was — less fragmented than indicated on the basis of available data. HERSHKOVITZ: NEOTROPICAL DEER 3 Material Specimens examined in the preparation of this report include skins, skulls, and postcranial skeletons representing 57 pudus, as well as an uncounted number of representatives of all other forms of neotropical deer preserved in the institutions listed under the "Abbreviations" section. In addition, reference was made, as occasion required, to other artiodactyls in the Field Museum collec- tions. Photographs of skulls and dentition are based on "normal" specimens in good condition selected from material at hand. In some cases, it was necessary to substitute mandibles or teeth of specimens other than those used for illustrating the main parts of the skull. Abbreviations AMNH — American Museum of Natural History ANSP — Academy of Natural Sciences, Philadelphia BM — British Museum (Natural History) FMNH — Field Museum of Natural History MCZ — Museum of Comparative Zoology, Harvard University USNM — National Museum of Natural History, Smithsonian Institution I take this opportunity to thank the authorities of the sister institutions men- tioned here for permission to study the material in their charge. Measurements Standard external and cranial measurements are used here. Limb bone mea- surements are of greatest length, irrespective of the points. Width of metapo- dials, however, is measured between widest points across the distal epiphyseal union. Combined length of the articulated thoracic and lumbar vertebrae is equated with trunk length (TL, table 2), and used as principal reference for proportional lengths of limb bones. In most cases, articulations were made man- ually and the vertebrae measured in sections. Carpal, tarsal, and phalangeal dimensions are omitted here. In comparisons of pudu postcranial skeletal bones with those of other ar- tiodactyls (table 2), ratios instead of actual measurements are given to eliminate or minimize actual size differences between species, sexes, age classes, and individuals. Review of Cervid Classification with Particular Reference to South American Genera The first comprehensive and soundly documented arrangement of deer, in- cluding the South American forms, is that by Brooke (1878, p. 889). His scheme, less a classification than a dichotomous key to the genera or generic groups, is abstracted as follows: I. Plesiometacarpi (proximal ends of lateral metacarpals persistent, fig. 2). A. Posterior portion of nasal cavity not divided into two chambers by vomer (fig. 3). Genera: Cervulus i=Muntiacus), Elaphodus, Cervus (sub- genera, Cervus, Rusa, Rucervus, Elaphurus, Axis, Pseudaxis, Dama). II. Telemetacarpi (distal ends of lateral metacarpals persistent, fig. 2). 4 FIELDIANA: ZOOLOGY A. Posterior portion of nasal cavity not divided into two chambers by vomer (fig. 3). Genera: Alces, Hydropotes, Capreolus. B. Posterior portion of nasal cavity divided into two chambers by vomer (fig. 3). Genera: Cariacus {=Odocoileus, with subgenera Odocoileus, Blas- tocerus [includes Ozotoceros=Blastoceros], Furcifer [=Hippocamelus], Coas- sus [=Mazama]), Pudua {=Pudu), Rangifer. The subfamilies Cervinae and Capreolinae, recognized by Pocock (1911, p. 971) in his arrangement of cervids, are equivalent to Brooke's Plesiometacarpi and Telemetacarpi. In 1923, Pocock (p. 206) elevated each well-marked genus or generic group of telemetacarpalid deer to subfanruly rank without indication of interrelationships, except through the following artificial key (abstracted): A. Telemetacarpalia (fig. 2). 1. Vomer dividing posterior nares (fig. 3). a. Cuboideonavicular and external cuneiform bones united (Pudinae, figs. 4, 24). b. Cuboideonavicular and external cuneiform bones separated (fig. 2). (1). Antlers present in males only (Odocoileinae). (2). Antlers present in both sexes (Rangiferinae). 2. Vomer not dividing posterior nares (fig. 3). a. Antlers absent; upper canine hypertrophied in males (Hy- dropotinae). b. Antlers present; upper canine minute or absent in either sex. (1). Muzzle short, not swollen (Capreolinae). (2). Muzzle very long, much swollen (Alceinae). B. Plesiometacarpalia (fig. 2). 1. Cuboideonavicular and external nuddle cuneiform bones fused (Mun- tiadnae = Cervulinae, fig. 2). 2. Cuboideonavicular and cuneiform bones discrete (Cervinae). Lydekker (1915, p. vii) grouped all deer in the family Cervidae (Section Pec- ora). The family, in turn, was divided into subfamilies Moschinae and Cervinae, the latter with genera Muntiacus, Elaphodus, Dama, Cervus, Elaphurus, Odocoileus, Blastocerus, Hippocamelus, Mazama, Pudu {Pudella, a subgenus), Capreolus, Ran- gifer, and Hydropotes. Living and fossil South American deer are central to the classification of the Cervidae presented by Carette (1922, p. 442). In the following, the two sub- families and attributed characters are abstracted, cervine generic names omitted: A. Cervinae (Eurasian). 1. Vomer incompletely ossified, the vertical plate not extending to pos- terior portion of nasal cavity. 2. Plesiometacarpalid or telemetacarpalid. 3. Metatarsal tuft, when present, on proximal half of metatarsus. 4. Upper canine tusk-like to absent. 5. Premaxillary and nasal bones usually articulating. 6. Antlers simple, dichotomous, branched, or absent. B. Neocervinae (New World). 1. Vomer completely ossified, vertical plate extending to posterior portion of nasal cavity. 2. Telemetacarpalid. 3. Metatarsal tuft, when present, on distal half of metatarsus. HERSHKOVITZ: NEOTROPICAL DEER 5 4. Upper canine small or absent. 5. Premaxillary and nasal bones usually not articulating. 6. Antlers. a. Absent {iBlastomeryx, North American Miocene). b. Simple {Mazama, Pudu). c. Bifurcate {Hippocamelus). d. Dichotomous (Blastocerus) . e. Branched {Odocoileus, Rangifer). Included within the "Neocervinae" are the extinct, branch-antlered Antifer, Paraceros, Morenelaphus, Pampaeocervus, and Epieuryceros. "Neocervinae," the group name erected by Carette, is not derived from a generic name and has no status. Kraglievich (1932, pp. 425-429) dismissed Carette's concept of "Neocervinae" as contrived and composite. His own arrangement, based primarily on antler form, is: a. Mazaminae {Mazama, Pudu). h. Odocoileinae {Odocoileus, Blastocerus, Hippocamelus). c. tAntiferinae {iAntifer). d. Aldnae {Alces). e. Cervinae {Cervus, Rangifer, fParaceros, fMorenelaphus [iPampaecervus, a subgenus]). f. Incertae sedis {iEpieuryceros). Living and fossil genera of North American deer (Cervidae) are classified in key form by Frick (1937, pp. 32, 34, 205) primarily on the basis of teeth and antlers. South American genera are included only "tentatively, for comparison — unstudied," in "Division C.-Cervini. Pleistocene — Recent," with the following subfamilies and their respective genera: Subfamily 7. Cervinae {Procoileus, Cervus). Subfamily 8. Odocoilenae {Odocoileus, Eucervus). Subfamilies 8 and 9a. Alcenae-Rangiferinae {Cervalces, Alces, Rangifer) Subfamilies 8 and 9a (continued). Odocoilenae and (?)Rangiferinae {Pudu, Hippocamelus, Blastocerus, Mazama, Ozotoceros, fPalaeodocoileus, fParaceros, fMorenelaphus, iAntifer). Simpson's (1945, p. 152) Cervoidea, with the single family Cervidae, contains the following subfamilies, with extant American tribes and genera included where pertinent: tPalaeomerydnae Moschinae t Dromomery d nae Muntiadnae Cervinae Odocoileinae Odocoileini {f Antifer, iEpieuryceros, fMorenelaphus, Odocoileus, Mazama, Hip- pocamelus, Blastocerus, Ozotoceros, Pudu) Aldni {iCervalces, Alee) Rangiferini {Rangifer) Hydropotini {Hydropotes) Capreolini {fProcapreolus, Capreolus) FRONT LIMB Mazama rufina 70563 cal caneam navicular — >^^^_^m— cuboid cuneiform — P33P^~-t5 DORSAL VENTRAL Muntiacus reevesi 15575 LATERAL DORSAL sesamoid HIND LIMB Fig. 2. Left forefoot and hind foot of brocket, Mazama rufina, with vestiges of metacarpal bones II and V distal in position (telemetacarpalia), and of muntjac, Muntiacus reei'esi, with vestiges of the same metacarpal bones proximal in position (plesiometacarpalia). Front digital bones oi Mazama rufina lacking. HERSHKOVITZ: NEOTROPICAL DEER Fig. 3. Nares are divided posteriorly by bony extension of vertical plate of vomer (arrow) in Odocoileinae, as in Mazama americana, but are undivided posteriorly in Cer- vinae, as in Capreolus capreolus. Simpson (1945, p. 267) discarded the prior subfamily group name Capreolinae for Odocoileinae, "because Odocoileus makes a better type." Flerov (1952) criticized Simpson's concept of cervoid relationships and rear- ranged the Cervoidea as follows: Moschidae Cervidae tPala eomery d na e tDromomery ci na e Cervulinae (=Muntiacinae) Cervinae {Capreolus, Cervus, Elaphurus, Alces, iPliocervus, iProcapreolus, ■UZervavitus, iEucladocerus, iPseudalces, iMegaloceros, iCervalces) Neocervini {Mazama, Pudu, Odocoileus [with subgenera Odocoileus, Blas- tocerus, Hippocamelus, Rangifer, iAntifer, fEpieuryceros, iPampaeocervus, iMorenelaphus ]) . According to Churcher (1966), iEpieuiyceros and iAntifer (in the preceding rearrangement) probably are congeneric with Blastocerus. 8 FIELDIANA: ZOOLOGY Cabrera (1%1, pp. x, 324), in his Catdlogo de los mamiferos de America del Sur, recognizes the extant South Anierican genera of Cervidae in the following order: Odocoileus, Blastocerus, Ozotoceros, Hippocamelus, Mazama, Pudu (with subgenus Pudella). For convenience of a single reference, a provisional classification of currently recognized genera and living spedes of South American cervids is provided herewith: Cervidae Gray 1821 Odocoileinae Pocock 1923 (New World deer) Pudini (new rank=Pudinae Pocock 1923) Pudu Gray 1850 (Pudus) P. mephistophiles de Winton 1896 P. puda Molina 1782 Odocoileini Simpson 1945 Mazama Rafinesque 1817 (Brockets) M. americana Erxleben 1777 M. chunyi Hershkovitz 1959 M. gouazoubira Fischer 1814 M. rufina Pucheran 1851 iCharitoceros Hoffstetter 1963 Hippocamelus Leuckart 1816 (Huemuls) H. antisensis D'Orbigny 1834 H. bisulcus Molina 1782 Blastoceros Fitzinger 1860 {=Ozotoceros Ameghino 1891 [Pampas deer]) B. bezoarticus Linnaeus 1758 Blastocerus Gray 1850 (?synonyms: iAntifer Ameghino 1889, iParaceros Ameghino 1889, iEpieuryceros Ameghino 1889 [Swamp deer]) B. dichotomus lUiger 1815 Odocoileus Rafinesque 1832 (synonyms: iPaleodocoileus Spillmann 1931, iProtomazama Spillmann 1931 [White-tailed deer]) O. virginianus Zimmermann 1780 fAgalmaceros Hoffstetter 1952 iRohnia Castellanos 1957 iMorenelaphus Carette 1922 (?synbnyms: iPampaeocervus Carette 1922, iHabromeryx Cabrera 1929) Representatives of all listed genera have been recorded from the Pleistocene of South America. In addition, deer fragments from the Uquian or the Pliocene- Pleistocene boundary in Argentina have been identified with Blastoceros {=Ozotoceros) , Antifer, and Habromeryx (Pascual et al, 1966, p. 21). Odocoileine Origins and Radiation It has been suggested by Matthew (1908; Flerov, 1952) that the Miocene Blas- tomeryx Cope 1877 from South Dakota, Wyoming, and Colorado is a primitive generic model from which New World deer can be derived. Blastomeryx, in turn, he declared, appeared to be derived from the Oligocene Leptomeryx. Judged by descriptions, skeletal parts of the known spedes of Leptomeryx are indeed com- paratively primitive, and antlers are lacking. However, the large body size, comparable to that of the larger spedes of Mazama, and, above all, the laniary or HERSHKOVITZ: NEOTROPICAL DEER 9 traguline-like upper canine remove Blastomeryx from the line leading to modern Odocoileinae, or, for that matter, to the Cervidae. Whatever the ancestral stock, it is no less likely that the Odocoileinae origi- nated in South America as in North America. The principal theater of radiation, however, almost certainly was South America. Pudus, tribe Pudini, with their mosaic of primitive, highly specialized and unique characters, appear to repre- sent a very early offshoot of the ancestral stock. Brockets and branch-antlered deer represent a second offshoot, if not the main stem itself. They have retained most of the basic ancestral characters and their evolution has been directed, in the main, toward increasing body mass and social complexity. Among the branch-antlered deer now living in South America, Odocoileus alone may have originated in Middle America, although its ancestral home may have been elsewhere. From Middle America, Odocoileus must have spread into South and boreal America with major radiation taking place in the latter. Cir- cumpolar caribou or reindeer, genus Rangifer, are the northernmost odocoileines. They could have diverged from the same stock that gave rise to modem Odocoileus. The extinct Argentine Morenelaphus with caribou-type antlers may represent a third group, or, perhaps, a subgroup of the second. Cheiridia The artiodactyl manus and pes evolved for support of the limbs on enlarged third and fourth ungual digits. The modification, doubtless beginning at a preancestral stage, involved loss of metapodial I and digit 1, disappearance of a discrete, carpal central bone, and, usually, the trapezium, reduction of digits 2 and 5 together with the corresponding metapodial bones, and elongation and coalescence of metapodials III and IV into a single, functional element — the cannon bone. Fusion of the middle metapodial bones progresses from midshaft toward both ends, with union of the distal portions lagging behind the proximal. A graded series of metapodials III and IV from separate to entirely fused, ex- cepting the distal articular facets, can be formed with species of suborder Suina, followed by the tragulid Hyemosc/iMS and selected forms of camelids, bovids, and cervids in the order named (fig. 4). Locomotorially, the front limb is specialized mainly for pulling, the hind limb mainly for pushing. As already noted, in nearly all living Old World cervids, only the proximal ends of the disappearing lateral metacarpals persist (plesiometacarpalid, fig. 2). In all living cervids of New World origin, the distal ends persist (telemetacar- palid, fig. 2). The latter condition may be more primitive because the metacarpal vestiges occur at the digital or functional end. However, functional difference between the two limb types is not apparent. Carpal bones, on the other hand, remained fairly constant throughout the order Artiodactyla. Individual forelimb and hind limb bones are serially homologous (table 1), but evolution of the homologues does not always proceed at the same rate or in the same direction. Metatarsi III and IV and corresponding digits do resemble their forelimb homologues. In contrast, metatarsi II and V have disappeared except sometimes for vestiges, particularly in pudus, without consistent orientation proximally or distally. The corresponding digits do not articulate with the rest of the skeleton, thus resembling the condition in plesiometacarpalid deer. The tarsal central or navicular bone, unlike the carpal central bone, is present and well developed. In Pecora and Tragulina, it is integrated with the cuboid to form c=^%r u b '- o ^' = 2 .| n "5 t« C 10 HERSHKOVITZ: NEOTROPICAL DEER 11 Table 1. Homologies of artiodactyl carpal and tarsal bones are shown horizontally. Names in columns marked "1" are used in comparative anatomy, those in columns marked "2" are the customary English names employed in this work. Names connected by a brace are of bones fused in all artiodactyls except as noted; those followed by a slash (/) are further fused with the next listed bone in suborders Tragulina and Pecora. Sesamoid bones, except the pisiform, are not named. Names and homologizations are adopted from Flower (1885, pp. 281-311; 339-360). Carpalia 1 Radiale Intermedium Ulnare Carpale 1 Carpale 2/ Carpale 3 Carpale 4 Carpale 5 Centrale =Scaphoid =Lunar' =Triquetrum or cuneiform =Trapezium^ =Trapezoici/ =Magnum =Undform = Central' 1 Tibiale Intermedius Fibulare Tarsale 1 Tarsale 2 Tarsale 3 Tarsale 4 Tarsale 5 Centrale Tarsalia 2 =Astragalus or Talus =Calcaneum =Internal cuneiform =Middle cuneiform =External cuneiform =Cunei- form^ =Cuboid/ =Cuboideonavicular3 = Navicular 'Lunar and central fused in Artiodactyla. ^Presumed lost in Artiodactyla except in Hippopotamus. ^Cuneiform and cuboideonavicular bones always fused in Tragulina, giraf- fid Okapia, and cervids Muntiacus and Pudu; middle and external cuneiforms (tarsales 2, 3) are distinct in Suina. the cuboideonavicular bone. External and middle cuneiform bones (cuneiforms 2 and 3) of the tarsus are fused, except in Suina, and the internal cuneiform is small but discrete in all artiodactyls. In Pudu and Cervulus {=Muntiacus), the cuboideonavicular bone is further united with outer and middle cuneiform bones into a single element, as in the chevrotains Tragulus and Hyemoschus (Tragulidae, Tragulina, Artiodactyla), and giraffids.' Fusion of cuboid, navicular, and outer and middle cuneiform bones into a single unit obviously evolved independently in the four disparate artiodactyl groups just dted. Functional difference between simplified and primitive ar- tiodactyl tarsi is questionable, particularly in Giraffa, where both types may occur in the same animal. Seemingly, the tragulid and pecorid tarsus operates as a unit irrespective of articular relationships between its components. Tarsal similarities among the four groups in question do not imply metatarsal similarities. Limb proportions differ widely among them. The tragulid metatar- sus is most primitive in retention of complete, albeit slender and elongate, metatarsal bones II and V; whereas the same bones are absent in giraffids, muntjacs, and deer, except pudus and possibly a few other spedes in which they may persist as vestiges. Furthermore, pudu metapodials, as represented by the cannon bone (III -I- IV), are relatively shorter and broader than those of other artiodactyls, and thus nearer the primitive condition still exhibited by living members of the suborder Suina. Limb Proportions Forelimb length, based on combined length of humerus, radius, and metacar- pus (H + R + MC, table 2) in pudus and the great majority of quadrupedal 'Consistently in Okapia (FMNH 26066, 104733, 108987) and variably in Giraffa (fused in both tarsi of FMNH 34930, fused in left, separate in right of FMNH 34424); aU zoo animals. u I CO ^ •§1 'I 5 — o II W — ' O f^ f^ f^ ^ op QO f*^ '— ^ QO ^ O r o in o^ -2 m t^ rj r- *; o ' a G O.' in 3 r} IN ^ HI 01 H O U H H 12 HERSHKOVITZ: NEOTROPICAL DEER 13 mammals, is less than hind limb length, based on combined length of femur, tibia, and metatarsus (F + T + MT). As shown in Table 2, length of forelimb (H + R + MC) to trunk length (TL), based on combined length of articulated thoradc and lumbar vertebrae, ranges from about 73% (Pudu) to about 104% (Mazama) in adult artiodactyls sampled. The average length of hind limb to trunk (TL) ranges from about 89% (Bison, followed by about 105% in Pudu and Oreo- tragus) to about 139% (Mazama gouazoubira) . Individual variation is considerable, and present ratios based on the sum of straight-line lengths of long limb bones are greater than — but probably directly proportional to — standing shoulder and rump heights in which limb bones articulate at angles. The cervid humerus is usually the longest thoracic limb bone. In pudus, it is relatively longer than in all other cervids and bovids measured. The humerus, however, contributes least to standing shoulder height. The cervid radius may be nearly as long as the humerus, sometimes slightly longer. The metacarpus is the shortest of the long thoradc limb bones, but nearly all its length enters into standing shoulder height. In most Bovidae, the radius is the longest forelimb bone, but in others, notably gazelles, the metacarpus is longest. In both cervid and bovid hind limbs, the tibia is usually the longest bone, the metatarsus being shortest of the three long bones. As in the forelimb, almost the entire length of the metatarsus enters into total standing rump height. Metapodials of pudus are not only relatively shortest among cervids, but also broadest. Ratio of metapodial width to length of the small Andean brockets, Mazama rufina and M. chunyi, appears to be nearly the same when greatest width is measured across the distal epiphysis. However, metapodial shafts of all brockets, particularly those of the metacarpus, are relatively narrower than those of pudus. II. GENUS Pudu GRAY Synonymic History Pudu Gray, 1850, Gleanings Knowlsely Menagerie, p. 69 — subgenus of Coassus Gray. Gray, 1852, Proc. Zool. Soc. London, 1850: 242— subgenus of Coassus Gray. Gray, 1852, Cat. Mamm. Brit. Mus., Ungulata, p. 240 — genus. Sclater, 1870, Proc. Zool. Soc. Lon- don, 1870: 116 — classification. Lydekker, 1915, Cat. Ungulate Mammals Brit. Mus., 4: 214— characters. Knottnerus-Meyer, 1907, Arch. Naturg., (73) 1: 15, 16, 97, 108— cranial characters. Carette, 1922, Rev. Mus. La Plata, 26: 412, 420, 422, 425, 428, 445— history; characters; classification. Pocock, 1923, Proc. Zool. Soc. London, 1923: 181, 207— characters; classification (genus of subfamily Pudinae). Kraglievich, 1932, Anal. Mus. Hist. Nat., Montevideo, (3) 2: 265 — characters; phylogeny. Pocock, 1932, Proc. Zool. Soc. London, 1932: 395^antlers; comparisons. Flerov, 1952, Fauna U.S.S.R., n.s. no. 56, Mammals, 1, (2): 218— description. Hershkovitz, 1958, Fieldiana, Zool., 36(6): 619 — distribution; Pudella a synonym. Cabrera, 1961, Rev. Mus. Argentino Cienc. Nat. "Bernardino Rivadavia," 4(2): 343 — subgenus of Pudu Gray; included species: P. pudu Molina. Haltenorth, 1963, Handb. Zool., 8(32): 4, 9, 42, 47, 48— subgenus of Mazama; characters; classification; Pudella Thomas a synonym. Hershkovitz, 1972, in Keast, Erk, and Glass (eds.). Evolution, mammals and southern continents, State U. of New York Press (Albany), pp. 362, 371, 392 — history, distribution, ecology, adaptation. Whitehead, 1972, Deer of the world. Viking Press, New York, pp. 64, 112, 169— spedes, P. pudu, P. mephistophiles; classification. Pudua Garrod, 1877, Proc. Zool. Soc. London, 1877: 18— emendation of Pudu Gray; spedes, Pudua humilis Bennett. Brooke, 1878, Proc. Zool. Soc. London, 1878: 926— characters. De Winton, 1896, Proc. Zool. Soc. London, 1896: 508— characters. Lydekker, 1898, Deer of all lands, p. 307— characters. 14 FIELDIANA: ZOOLOGY Pudella Thomas, 1913, Ann. and Mag. Nat. Hist., (8)11: 588 — type, Pudua mephistophiles de Winton. Lydekker, 1915, Cat. Ungulate Mammals Brit. Mus., 4: 217 — subgenus oiPudu Molina. Pocock, 1923, Proc. Zool. Soc. London, 1923: 207 — genus of subfamily Pudinae. Cabrera, 1%1, Rev. Mus. Argentino Cienc. Nat. "Bernardino Rivadavia," 4(2): 344 — subgenus of Pudu Molina. Brokx, 1972, J. Mammal., 53(2): 260 — may he Mazama. Mazama, Flerov (part, not Rafinesque), 1952 — Fauna U.S.S.R., n.s. no. 56, Mammals, 1, (2): 218 — part, M. mq)histopheles [sic] de Winton only. Haltenorth, 1963, Handb. Zool., 8(32): 47, 48 — part, Pudu Gray, a subgenus. Mezama [sic] Cabrera, 1%1, Rev. Mus. Argentino Cienc. Nat. "Bernardino Rivadavia," 4(2): 344 — misprint for Mazama in synonymy of subgenus Pudu. Coessus [sic], Cabrera, 1%1, Rev. Mus. Argentino Cienc. Nat. "Bernardino Rivadavia," 4(2): 344 — misprint for Coassus in synonymy of subgenus Pudu. Type species. — Coassus pudu [=Capra puda Molina], by virtual tautonymy. Included species. — Pudu mephistophiles de Winton, P. puda Molina. Distribution (fig. 1). — Temperate zone forests and fringing grasslands of southern Colombia, Ecuador, north central Peru, southern Chile, and adjoining parts of Argentina. Altitudinal range from sea level to about 1,000 m above in Chile and Argentina, and between 2,(X)0 and 4,000 m in Colombia, Ecuador, and Peru. Conservation A program for the study and management of deer threatened with extinction throughout the world was launched by the International Union for Conservation of Nature and Natural Resources (lUCN). The first of three reports presented by Cowan & Holloway (1973; 1974; 1975) lists 25 threatened forms, including the South American pampas deer {Blastoceros bezoarticus) , marsh deer {Blastocerus dichotomus), and the northern and southern huemuls {Hippocamelus antisensis and H. bisulcus). Certainly, the objectives and rationale of this conservation program apply no less to pudus than to any other deer spedes threatened with extinction. The geographic distribution of pudus given here (figs. 1, 30, 35), based on museum specimens and published records, does not reflect the present status of the animal in nature. In the years intervening since 1782, when the first pudu was made known to science, there has been enormous shrinkage in the peripheral and internal ranges of the genus. Hunting by man is a much greater menace to the survival of pudus than is habitat destruction. Pudus readily adapt as individuals to the marginal and even seemingly improbable habitats to which they are being driven. It is unlikely, however, that the spedes can remain viable for long under such drcumstances. Present conservation measures, based on establishment of national parks and nature reserves, are inadequate, because in virtually all cases, the legally protected animals are defended by nothing more potent than the mere letter of the law. As yet, Peru has no national park or nature reserve where pudus are definitely known to exist. Colombia has three parks that certainly harbor the northern pudu and three that probably do. Ecuador has one. Chile has four or five parks, and possibly more, where the southern pudu still may occur. Argentina can boast of at least three national parks where the southern pudu may still be found, if only as a vanishing spedes, unless a successful program for breeding the animal in semicaptivity is put into effect immediately. Information regarding past and recent status of each of the two kinds of pudus in the wild is included under various headings in the spedes accounts. HERSHKOVITZ: NEOTROPICAL DEER 15 Taxonomy The name Pudu was proposed to contain Capra puda Molina {Cervus humilis Bennett, a synonym) by Gray (1850; 1852a [1850], p. 242), as a subgenus of Coassus Gray 1843 [=Mazama Rafinesque 1817]. Soon after. Gray (1852b, pp. x, 240) raised Pudu to full generic rank. Characterization of the genus, however, was never adequate until Brooke (1874, p. 36; 1878, p. 926) pointed to the fused cuboideonavicular and outer and middle cuneiform bones of the tarsus. Pudua, introduced by Garrod (1877, p. 18), is merely an invalid emendation or Latinization of the original barbaric name. Pudella, erected by Thomas (1913, p. 588) for generic distinction of Pudu mephistophiles de Winton from P. puda Molina, was predicated on absence of preorbital glands, shallow lacrimal bone, contact between premaxillary and nasal bones, crown length of first indsor three to four times that of ia, narrower, more pointed hoofs, and rhinarium larger and projecting back in midline. Some of the foregoing characters are equivocal, and the others are usually evaluated as of specific grade in related genera. For example, in P. mephistophiles, the preorbi- tal glands with facial opening are present, not absent as claimed, but are little developed, the lacrimal fossa shallow. In P. puda, the glands are large, the external opening conspicuous, the lacrimal fossa correspondingly deep. These characters parallel those distinguishing most brockets from Mazama rufina, or Odocoileus virginianus from O. hemionus. The premaxilla does indeed contact the nasals in P. mephistophiles, but may or may not in P. puda. Middle lower indsors are not significantly different between these spedes; hoofs are nearly as long and narrow, but average deeper in P. puda, and therefore appear to be more blunt. Lydekker (1915, p. 217), dting only the fadal glands, lacrimal bones, and lower middle indsor crown length as diagnostic characters, reduced Pudella to subgeneric rank. Pocock (1923, p. 207), however, regarded Pudella and Pudu as distinct genera of his subfamily Pudinae. Flerov (1952, p. 200) referred only to the Mazama skull mismatched with the type skin of Pudu mephistophiles for sink- ing Pudella in the synonymy of Mazama. Evidently, he overlooked the fact that P. mephistophiles was typified by the skin only with assodated leg bones, and Pudella was based on the holotype skin plus two topotype skins and skulls of P. mephistophiles. In any event, Pudu and Mazama, with P. mephistophiles included, were recognized as distinct genera, but other South American genera are listed by Flerov as subgenera of Odocoileus. Haltenorth (1963, p. 48) compromised by relegating Pudella to the synonymy oiPudu and redudng the latter to a subgenus of Mazama. Otherwise, his arrangement of remaining groups of South American deer agrees with that of Flerov. Generic Characters External. — Smallest of Cervidae (figs. 5, 6), legs short, body low slung, rump little or not raised above withers, ears more or less rounded; combined head and body length less than 85 cm; height at shoulder less than 50 cm; tail obsolescent, less than 8% of combined head and body length and concealed in rump; weight less than 15 kg. General body color dominantly agouti with crown and ex- tremities more or less saturate and contrasting with trunk; ears without white edging; lateral narial area and chin rufous to blackish, never marked with white. Odocoileus virginianus Blastocerus dichotomus Mazama americana Pudu mephistopiles Pudu puda Hippocamelus antisensis Fig. 5. Size relationship and body form of six South American deer, from smallest (Pudu) to largest (Blastocerus); the numbers refer to approximate average shoulder height (cm). 16 HERSHKOVITZ: NEOTROPICAL DEER 17 H ippocamelus antlscusis Mazama americana Fig. 6. Body form of four South American deer, not to scale; compare dorsal contour, rump height, limb proportions, neck, muzzle, ears. From top, left to right, northern pudu (photograph by F. Kleinschmidt, courtesy Berlin zoo); southern pudu (photograph by H. Fradrich, courtesy Berlin zoo); northern huemul (photograph courtesy Berlin zoo); red brocket (photograph courtesy San Diego zoo). the pale, lateral rostral band absent (figs. 7, 8); buttocks without sharply con- trasting patch or band. Pelage of back harsh, basal portions of hairs pithy and brittle; hairs of crown whorled; paired frontal tufts present and unspotted in females; nuchal hairs always directed back. Frontal glands not evident in females; external opening of preorbital (lacrimal) gland small or large (figs. 8, 9); forefoot and hind foot pedal glands contained in shallow interdigital trough or small pouch (fig. 10); specialized tarsal and metatarsal tufts weakly developed or absent, corresponding glands not defined in dry skin. Fore and hind hoofs slender, length twice or more greatest width; fore and hind dew claws longer than wide; soles between outer and middle digits thickly clothed with hair (fig. 10). Cranial (figs. 11-17). — Rostrum short, the diastema shorter than alveolar length of upper cheek teeth (pm^-m^); length of nasals less than length of fron- tals at midline; superior border of orbit depressed below frontal suture at same plane; greatest supraorbital breadth of frontals less than width of brain case; upper portion of supraocdpital moderately inflated, its median protuberance 18 FIELDIANA: ZOOLOGY Fig. 7. Fadal color pattern of Odocoileinae, a composite of all spedes. posterior auricular border inner auricular surface outer auricular surface mental patch mandibular band buccal patch gular patch opening of preorbital pouch frontal spot or tuft (marks frontal gland in female) a. b. narial patch dark lateral rostral band J- k. c. pale lateral rostral band 1. d. rostral band m e. inferior orbital band n. f. g- h. i. superior orbital band or superdliary band anterobasal auricular patch posterobasal auricular patch anterior auricular border o. P- q- r. projecting beyond lambdoidal crest in lateral view; ascending ramus of pre- maxillary extending to, or widely separated from, lateral border of nasals; malar articulating with one-half or less inferior suture of fadal plate of lacrimal and hardly or not at all entering into composition of inferior surface of lacrimal fossa; transverse diameter of orbit greater than inferior margin of fadal plate of lacri- mal; auditory bullae little or not inflated. HERSHKOVITZ: NEOTROPICAL DEER 19 Pudu mephistophiles Pudu puda Mazama chunyi Fig. 8. Stylized heads of smallest South American deer showing individual color pat- tern, ear form, preorbital pouch opening, and dorsal outline of rhinarium. Dental (figs. 18-23). — Greatest mesiodistal crown length of permanent ii equal to or less than combined crown length of permanent i2-3 (fig. 22), all permanent indsors not in place before eruption of last molar; lower deciduous front teeth like replacements, but dii relatively broader, dia-s and dc narrower, more pointed, the entire series of front teeth approaching those of adult Mazama (fig. 22); deciduous upper canine usually present in Pudu mephistophiles, never persistent, but often succeeded by the permanent canine; dc absent in P. puda or, if present, shed after eruption of m^ and rarely, if ever, replaced (figs. 11, 13, 14, 16). Mazama gouazoubira Hippocamelus antisensis Fig. 9. Heads of five South American deer; compare fadal markings, muzzle length, preorbital gland, ear shape; antlers of P. mqjhistophiles partially rubbed, of M. rufina and H. antisensis in velvet. (Photograph of Pudu puda courtesy Cologne zoo; remaining photo- graphs courtesy Berlin zoo.) 20 FRONT HIND FRONT HIND Pudu mepbistophiles Pudu puda f i Mazama rufina Odocoileus virginianxis Fig. 10. Front and hind feet (left lateral and ventral surface) of pudus, brocket, and Virginia deer. Pouch of interdigital gland is indicated by dashed line. 21 22 FIELDIANA: ZOOLOGY On the basis of upper molars of a specimen of Mazama americana ("M. tema") and one of Odocoileus virginianus, Frechkop (1965) concluded that all Odocoileinae (his Capreolinae) were characterized by a short crista or crochet, springing from the inner distal enamel border of the hypocone, labeled "x" in his illustrations (Frechkop, 1%5, p. 4, figs. 1, 2, 3). In a spot check of Field Museum specimens, this crista (fig. 23, XI) appears to be absent or incipient in Pudu mephistophiles and Blastoceros bezoarticus, and rudimentary to moderately developed in unworn molars of P. puda, usually present in Mazama americana and M. rufina, present or absent in Mazama gouazoubira, Blastocerus dichotomus, Odocoileus hemionus, and O. virginianus. In two specimens at hand of Mazama chunyi, the character is present in m^'^ of the type, a juvenal with unerupted m^, and undeterminable in the considerably worn molars of an old female (fig. 18). Crest XI size is variable in all spedes. Where it appears to be absent in moder- ately worn teeth of a given spedes, it may be found in one or more unerupted molars, mainly m*, of some individuals of the same spedes. In some cases, a crest equivalent to crista XI springs from the medial border (V) of the hypocone. The comprehensive account of cervid dental morphology is reserved for a future artide of this series. Dental succession. — Sequence of eruption of permanent teeth determined from inadequate material in the Field Museum and extrapolations therefrom, is: ml-m2-m3-pm4-pm3-pm2-c (P. mephistophiles) ml-m2-(il or m3)-pm4-pm3-pm2-i2, (i3-c). Antlers. — Spike-like, usually less than 10 cm measured from burr and directed straight back in line with dorsal plane of frontal bone. Pedides develop during eruption of m^, burr and antlers form during eruption of m^, well-developed antlers present in all 17 males examined with complete dentition. Antler casting is not evident in present material. Published and secondary data indicate, however, that new antlers grow during spring to early summer, mature during summer to early fall, drop in winter. Antler cyde of southern hemisphere pudus acdimatized in northern hemisphere zoos adjusts to north- em seasons. Vertebral formula. — Seven cervicals, 12-13 thoradc, 6-7 lumbar, 4-5 sacral, 8 caudal; total, 38-39 vertebrae. Generally, form and relative size are the same as those of other cervids. Limb bones and proportions (fig. 24, table 2). — Metapodials extremely short and broad; cuboid, navicular and external and middle cuneiform bones fused into a single tarsal element; length of metacarpal splints II and V from approximately one-half to two-thirds length of metacarpals III and FV; central portions of metatarsal bones II and V frequently persistent as splints, each ankylosed to the adjacent metatarsal bone. Male genitalia. — In reference to Pudu puda, Brambell & Mathews (1977, p. 27) note, "The testes are subcutaneous, but not scrotal. The penis is not long but lies on the abdominal wall." Presumably, the same condition obtains in P. mephis- tophiles. Generic Comparisons Pudus, like other strictly New World deer, are characterized by retention of the distal portion of the obsolescent metacarpals II and V (telemetacarpalid) and HERSHKOVITZ: NEOTROPICAL DEER 23 by division of the posterior nares into two chambers by the vertical plate of the vomer (figs. 2, 3, 24). In exclusively Old World Cervinae, vestiges of the lateral metacarpals are proximal (plesiometacarpalid) and the posterior nares are undi- vided.^ Pudus are most readily distinguished from all other deer by their small size (fig. 5). Only the dwarf brocket, Mazama chunyi Hershkovitz, is similarly small. Pudus are further distinguished from other cervids by the low-slung body with- out highly raised rump; short, thick legs, comparatively small, more or less rounded ears, small eyes, thick rhinarium with rounded nostrils, muzzle with- out white narial and mental patches or spots (figs. 6, 8, 9), tail completely hidden by rump hairs; mandibular symphysis usually more angular in horizontal plane than in related deer, alignment of incisors and canines of each side more nearly oblique than transverse. Caudal vertebrae number only 8 in Pudu, 11 in Mazama, and 10 or more in other cervids. The shallow interdigital depression or pocket containing the pedal glands in Pudu is primitive compared with the well- developed interdigital pouch of the hind feet, at least, of other New World deer (fig. 10). The key osteological character separating Pudu from other true deer, i.e., Cervinae and Odocoileinae, is the consistent union of the cuboideonavicular and external and middle cuneiform tarsal bones into a single bone (figs. 2, 4, 24). Other living cervids with similarly fused tarsal bones are the Asiatic plesio- metacarpalid tufted deer (Elaphodus) and muntjac {Cenmlus Blainville, the valid name currently replaced by the erstwhile nomen nudum Muntiacus Rafinesque [Int. Comm. Zool. Nomencl. 1957, Opinion 460]). It is a curious coincidence, as shown by Pocock (1911, p. 952, figs. 134b-c), that the interdigital integumentary folds containing fore and hind pedal glands are about as simple in muntjacs and tufted deer as in pudus. The extremely short, broad metapodials of Pudu, with metacarpals proportionately wider than metatarsals, are a bovine-like character unmatched in other cervids (table 2). Dental peculiarities are absent in Pudu (figs. 18-23). The form of their lower incisors accords with the less specialized deer type characterized by the rectan- gular or moderately spatulate crown of ip The more specialized crown of ii in Mazama and a number of other New and Old World deer is broadly spatulate or cuneate. In the deciduous teeth of most pudus, however, a comparatively ex- panded crown of ii, and attenuate iz-s, Ci adumbrate the more specialized front teeth of brockets (figs. 20, 22). The upper canine persists in many adult P. mqjhistophiles, whereas only the deciduous upper canine appears occasionally in immature P. puda. Fawnal Spotting Pattern The Juvenal coat of Pudu puda is spotted, whereas that of P. mephistophiles is not. Inasmuch as adult P. mephistophiles is more primitive than P. puda in every respect, including coat color, it is logical to infer that the unspotted juvenal coat ^Among circumboreal cervids, the American moose or Palearctic elk (Alces) and caribou or reindeer (Rangifer) are telemetacarpalid, with the nares undivided in Alces and divided in Rangifer. The wapiti or deer (Cervus) is plesiometacarpalid, the nares undivided. Old World roebuck (Capreolus), Chinese water deer (Hydropotes) , and musk deer (Moschiis) also are telemetacarpalid. 24 FIELDIANA: ZOOLOGY color is also more primitive. Fawns of the Andean taruga, Hippocamelus antisen- sis, and the marsh deer, Blastocerus dichotomus, are also unspotted. Those of remaining neotropical deer are spotted (cf., figs. 25-27). Judged on its own merits, an unspotted, unmarked, or unpatterned coat characterized by agouti pelage is most primitive. Conversely, any coat color pattern that departs from the primitive agouti as a result of pigmentary dilution or loss is derived. In the case of white or whitish fawnal spotting, the pigmentary loss appears in the fetal pelage or lanugo. The juvenal character adumbrates the adult spotting that appears on some axis deer {Axis), fallow deer (Dama), and some Sika deer {Cer- vus). In these forms, the overall trend is toward paler coloration approaching albinism. The whitish spotting pattern similar to that of fawns, or a more advanced whitish spot and stripe, or entirely striped pattern appears in many mammalian lines. This characterizes the juvenal coat of swine and tapirs, and the adult coat of many rodents, e.g., ground squirrels {Spermophilus) , paca (Agouti), and the white stripe pattern of zebras. A blackish spotting coat pattern, noted in most felids and many other groups, is also common, whether in the juvenal or adult stages, or both. In these, however, the blackish spots may be reminders of a more primitive blackish coat. On the other hand, whitish spotting, which entails loss of melanin, is derived. Ancestral Characters and Relationships Pudus sometimes are referred to as dwarf or pygmy deer, and it has been suggested (von Lehmann, 1960, p. 50) that these and all other spike-antlered deer are literally dwarfed. The ancestral pudu is unknown and it cannot be ascertained whether it was larger or smaller than its descendants. Nevertheless, of the two living kinds, the smaller Pudu mephistophiles is evidently more primi- tive than the larger P. puda. Brockets, genus Mazama, are also small, ranging in size from Mazama chunyi, which is equal to the larger spedes of Pudu, and M. americana, which grades into the smaller forms of the otherwise much larger, white-tailed deer, Odocoileus virginianus. Thus, it can be asserted that the overall trend is toward greater body size and increasing antler size and complication. No doubt, dwarfed populations and races of deer develop from time to time, but probably most, if not all, are ecological rather than genetic demes or races. Small size and rudimentary antlers restricted to the male are primitive deer characters. Also, the smaller the deer, the smaller and more simple its antlers, irrespective of the age of the individual or grade of specialization of any or all other of its parts. Conversely, the larger the individual, the larger the antlers tend to grow, and the larger the race or spedes within a given phyletic line, the more complex the antlers tend to become. Thus, the extremely small body and simple antlers of Pudu, although primitive, do not in themselves imply an an- cestral condition or denote a close relationship to the small-bodied, spike- antlered Mazama. The affinities of Pudu, morphologically as well as historically, are with other neotropical cervids. However, the comparatively short muzzle, obsolescent tail, fused cuboideonavicular and external and middle cuneiform bones, and short, broad metapodials remove Pudu from all other deer and from consideration as an ancestral model. At the same time, primitive cervid characters of Pudu mephis- tophiles from which those of Mazama and other New World deer can be derived HERSHKOVrrZ: NEOTROPICAL DEER 25 include the relatively large and well-developed lateral metacarpals, vertically inclined antorbital vacuity, shape of lower indsors, the more than sporadic persistence of upper canines, the comparatively simple interdigital tegumentary fold for containment of pedal glands, absence of specialized tarsal and metatar- sal tufts and glands, comparatively short, rounded ears and small eyes, slightly raised rump, dominantly agouti pelage, and unspotted fawn. The remote, possibly Miocene-Pliocene, nearest common ancestor postulated here must have been sylvan, smaller than living pudus, body low slung, rump unraised, antlers simple or rudimentary and directed back parallel to the frontal plane. The tail must have been comparatively well developed, middle digits, including hoofs, long and mobile, facial or lacrimal glands small but not rudimentary, tarsal, metatarsal, and frontal glands and tufts rudimentary or absent, ears small, rounded, nuchal hairs directed back, coronal crests or whorls absent, adult pelage coarse but not pithy, general coloration agouti without special markings, fawn unspotted, metapodials broad, lateral metacarpals nearly or virtually complete, the corresponding digits or dew claws well developed, cuboideonavicular and cuneiform bones discrete, premaxillary and nasal bones in contact, antorbital vacuity small, its long axis vertical, crowns of lower indsors more nearly square than spatulate, upper canines small and persistent, vertebral formula 7C, 13T, 7L, 5S, IIC. More characters can be added, but with those mentioned, the others fall into place. Origin, Dispersal, Speciation The origin of Pudu in terms of time and place is unknown. The ancestral stock of the genus, and the Odocoileinae in general, is likewise unknown. As noted earlier (p. 8), the Palaeomerydnae, including Blastomeryx, the imputed ancestral root, is a spedalized traguline-like or moschine-like group that disappeared without known issue in the New World. In any event, it seems likely that Pudu and Mazama represent divergent lines of the tiny ancestral odocoileine, which must have lived during the early Pliocene or late Miocene in boreal America, Middle America, or South America. The genus Pudu itself must be regarded as South American, likely since Pliocene. The distribution and interrelationship of its two living spedes are clear, and their evolutionary history can be recon- structed from available morphologic and geographic data as well as from what is known of Pleistocene climates (Van der Hammen, 1974). It is believed that the immediate ancestor of modem pudus differed little from Pudu mephistophiles. During what possibly was a last gladal period, the former must have occupied the temperate zone forest fringe at the eastern base of the Cordillera Oriental, in what now is Colombia, Ecuador, Peru, and, perhaps, northern Bolivia. When the Andean gladers retreated, the temperate zone forest with its fauna followed in its wake. In the case of the pudu, some populations moved into higher altitudes at the same latitudes, clinging to the forest fringes forming on the shores of large rivers, such as the Putumayo, Pastaza, Huallaga, and Ucayali. Other populations moved southward along the base of the eastern Andes into higher latitudes as far as the Strait of Magellan and westward through forested passes into Chile. Evidently, no relic pudu populations per- sisted where the climate of the original temperate lowland habitat became tropi- cal or subtropical (fig. 1). Complete geographic isolation of northern and southern pudus is reflected in 3 < E uu 26 Fig. 12. Adult female pudu and brocket skulls, dorsal surface. 27 Fig. 13. Adult female pudu and brocket skulls, palatal surface. 28 •T3 C «5 D 3 29 Fig. 15. Adult male pudu and brocket skulls, dorsal surface. 30 Fig. 16. Adult male pudu and brocket skulls, palatal surface; inset, enlargement of premolars 2-4 with dedduous-like pm^. 31 Fig. 17. Mandibles of pudus and brockets, lateral surface of left rami. 32 Pudu mephistophiles 868A6 ^S^t&s^ Pudu puda 2A363 ;^#<^"^ Mazama rufma 70563 "^15^ ^%^^ Mazama gouazoubira 51091 Mazama americana 867A1 n&% Wazama chunyi (tvne) 73098 Fig. 18. Upper cheek teeth (pm^-m^) of adult pudus and brockets, and dpm^"^ m'"^ of Juvenal Mazama chunyi (type, AMNH). 33 Fig. 19. Lower cheek teeth (pirij-ms) of adult pudus and brockets, and dpm2.4, m,-2 of Juvenal Mazama chunyi (type). 34 3 3 35 36 Pudu mephistophiles juv, 88595 ^^ ' 88496 ^^- '-' Mazama chunyi juv. 73098 ad. 79912 Fig. 22. Permanent and deciduous incisor-canine series (i^a, c) in pudus and brockets; Mazama chunyi juv. 73098, type, AMNH. 37 Pudu mephistophiles Pudu mephistophiles Fig. 23. Upper right and lower left second and third molars of Pudu, the top m^ un- erupted of type, AMNH 73098. For explanation of symbols, see fadng page; system of symbols and homologies based on Hershkovitz (1977, pp. 298-301). 38 Explanation of Symbols Used in Figure 23. I eocene (paracone) 1 eoconid (protoconid) 2 protocone 2 metaconid 3 [paraconid, absent] 4 metacone 4 hypoconid 5 hypocone 5 entoconid a mesostyle (parastyle) a mesiostylid (parastylid) b distostyle (metastyle) b distostylid (hypoconulid) d plagioconule (metaconule) d plagioconulid e postentoconule e postentoconulid h entostylid-Zi i ectostyle-/ (stylocone) k ectostyIe-/f k ectostylid-/c* 1 ectostyle-/ (mesostyle) I ectostylid-/ (mesostylid) m ectostyle-m m ectostylid-m I eocrista I ecocristid II epi crista (crest labeled "+" by Frechkop, 1965, figs. 1,2) II epicristid III protoloph (protocrista) III protolophid IV plagiocrista (metaloph) IV' medial IV" median IV plagiocristid V entocrista VI distocrista (from postero- lingual cingulum C) VI distocristid VIII postentocristid XI crista-XI (individually variable in cervids) XII crista-XIl (individually variable in cervids; charac- teristic of notoungulates) ' ' e pretrigon basin or fossa e "trigonid" basin or fossa a trigon basin or fossa a talonid basin or fossa P talon basin or fossa /3 en entoflexus ex ectoflexus nex neoectoflexus talonid basin or fossid S posttalonid basin or fossid * Present in lower first molars only of individual figured. 39 FRONT LIMB ^ 1 VENTRAL iii-h ia-iv sesamoid- \l Pudu mephistophiles 86846 Pudu puda HIND LIMB Fig. 24. Left forefeet and hind feet of pudus with vestiges of metacarpal bones II and V distal in position (telemetacarpalia). Digits are missing in Pudu mqjhistophiles. 40 Fig. 25. One-week-old southern pudu fawn (photograph by H. Fradrich, courtesy Berlin zoo). Mazama gouazoubira Fig. 26. One-week-old gray brocket fawn (photograph by J. Fradrich, courtesy Berlin zoo). 41 42 FIELDIANA: ZOOLOGY Mazama amer icana Fig. 27. Red brocket doe with fawn (photograph courtesy Gladys Porter Zoo, Brownsville, Texas). the wide morphologic differences between them. The southern spedes, with its much greater distance from the hypothetical ancestral home, departs most from the postulated ancestral form. Hippocamelus, with its two spedes, the northern H. antisensis and the southern H. bisulcus, appears to have had a parallel history of dispersal and spedation. Most populations of the northern P. mephistophiles are still confined to the heads of the same Amazonian streams that defined their presumed habitats lower down. A few have spread across divides into watersheds drairung into the Caribbean or Padfic Oceans. Apparent absence of significant differences be- tween extant populations of northern Colombia, Ecuador, and Peru suggests that gene flow has been maintained between them throughout the major climatic event that shifted without notably altering the pudu habitat. Alternatively, the habitat shift was accomplished so swiftly or recently, with the actual linear distances covered so short, that no significant differentiation occurred in any population, geographically isolated as it may appear to be now or may have been then. HERSHKOVITZ: NEOTROPICAL DEER 43 III. THE SPECIES OF PUDUS, GENUS Pudu GRAY Key to the Species Size smaller; face, outer surface of ears, chin, forefeet and hind feet dark brown or black- ish; preorbital gland small, external opening inconspicuous; lacrimal fossa shallow or dish-shaped; long axis of antorbital vacuity nearly or quite vertical; fawns unspotted; (Colombia, Ecuador, Peru) Pudu mqyhistophiles (figs. 5, 6, 8, 9, 28, 29; p. 44). Size larger; face reddish buffy or brown; ears, narial patch, chin, forefeet and hind feet reddish; preorbital gland large, external opening large, conspicuous; lacrimal fossa deep or bowl-shaped; long axis of antorbital vacuity diagonal to nearly horizontal to basal plane of skull; fawns spotted; (Chile, Argentina) Pudu puda (figs. 5, 6, 8, 9, 36, 37; p. 60). Fig. 28. Northern pudus, adult male and female (photograph by H. Fradrich, courtesy Berlin zoo). Puau mephi stopni 1 es Fig. 29. Northern pudu, adult female, same individual shown in Figure 28. Note frontal tuft (photograph by H. Fradrich, courtesy Berlin zoo). 44 FIELDIANA: ZOOLOGY IV. Pudu mephistophiles de W7NTON NORTHERN PUDU (figs. 28, 29) Synonymic History Pudua mephistophiles de Winton, 18%, Proc. Zool. Soc. London, 1896(2); 508, pL 19 (animal) — part, not skulls shown in figs. 2, 4 [=Ma2ama]. Lydekker, 1898, Deer of all lands. Rowland Ward, London, p. 308, pi. 24, fig. 1 (animal) — characters. Lydekker, 1901, Great and small game of Europe, western Asia and America. Rowland Ward, London, p. 374 — characters. [Pudella] mephistophiles, Thomas, 1913, Ann. Mag. Nat. Hist., ser. 8, 11: 588 — classification. Pudu (Pudella) mephistophiles, Lydekker, 1915, Cat. Ungulate Mammals Brit. Mus., 4: 217; ECUADOR: Pichincha (Paramo de Papallacta); characters; type history. G. M. Allen, 1942, Extinct and vanishing mammals of the western hemisphere, p. 412 — ECUADOR; char- acters; type history; endangered species. Avila Pires, 1974, Publ. Biol. Inst. Invest. Gent., U.A.N.L., Mexico, 1:161 — distribution; conservation. P[udu]. mephistophiles, Hoffstetter, 1952, Mem. Soc. Geol. France, 31(66): 369— ECUADOR: Mazama fusca Spillmann, a synonym. Pudu mephistophiles, Hershkovitz, 1958, Mammalia, 22(4): 545 — metatarsal glands and tufts not defined. Hershkovitz, 1959, Proc. Biol. Soc. Washington, 72: 48-49, pis. 3C, 4D (skull), pi. 5B (front teeth) — comparisons. Cabrera, 1%1, Rev. Mus. Argentine Cienc. Nat. "Bernardino Rivadavia," 4(2): 344 — distribution; P. m. wetmorei Lehmann, a synonym. Crandall, 1964, Management of wild animals in captivity. U. of Chicago Press, pp. 557, 560, 594-5 — characters; no record of captivity. Grimwood, 1%9, Spec. Publ. no. 21, American Comm. Int. Wild Life Protection and New York Zool. Soc., p. 78 (map), 79 — PERU: Hudnuco (Pamacucho [sic]; Carpish summit on Huanuco-Tingo Maria road;- upper Rio Huallaga valley, above Huanuco; near ChagUa; Yanacocha, Rio Quero, 25 km NW Huanuco; between Huanuco and Ambo, 2700-3300 m; Quivilla road, 40 km from Huanuco); Junin (above San Ramon, 200-300 m; Rio Huasahuasi valley; upper Rio PaUcartambo, 10°50' S, 70°00' W); conservation; local names sac/ia cabra, antagllo. Dansie and Wince, 1971, Deer of the world. Hitchingsand Mason, Ltd., Plymouth, England, pp. 21, 23, fig. (map, distribution)— ECUADOR; PERU. Schauenberg, 1973, Mus. Geneve, 14(134): 12— figs. pp. 13, 14 (animal), fig. p. 15 (habitat), fig. p. 16 (head)— ECUADOR; COLOMBIA; history; conservation; distribution; occurrence in Peru questioned. P[udu]. mephistophiles mephistophiles, Mochi and Carter, 1953, Hoofed mammals of the loorld, Scribner's and Sons, Ltd., London, pi. VIII, text only. Pudu m[ephistophiles]. mephistophiles, Lehmann, 1960, Nov. Colombianas, Univ. Cauca, Popayan, 1(5): 282— ECUADOR; antler form. Pudu mephistophiles mephistophiles. Whitehead, 1972, Deer of the world. Viking Press, New York, pp. 29, 64, 165, map 9 (distribution) — distribution; classification. Pudua mephistopheles [sic], Thomas, 1908, Ann. Mag. Nat. Hist., ser. 8, 1: 350 — cranial characters: skull originally figured as corresponding to type skin =Mazama. Lonnberg, 1913, Ark. Zool., Stockholm, 8 (16): 33— ECUADOR: Pichincha (Papallacta, 12,000 ft). Pudu mephistopheles [sic], Soukup, 1%1, Biota, 3 (28): 331— PERU. Fradrich, 1975, Zool. Gart., 45(1): 67, 72, figs. 5, 6, 7 (animal), fig. 9 (skuU)— ECUADOR: Cotopaxi (W of Salcedo); maintenance in Berlin zoo. Brambell and Mathews, 1977. Proc. Symposium 1, Ass. Br. Wild Animal Keepers, p. 25 — characters. P[udu]. mephistopheles [sic], Brokx, 1972, J. Mammal., 52(2): 360 — upper canine present in specimen examined. Pudu (Pudella) mephistopheles [sic], Lehmann, 1945, Rev. Univ. Cauca, Popayan (Colombia), Zool., no. 6: 76 — comparisons. Pudella mephistophelis [sic], Cabrera and Yepes, 1940, Mamiferos Sud- Americanos. Hist. Nat. Ediar (Buenos Aires), p. 278— ECUADOR; PERU. M[azama]. mephistopheles [sic], Flerov, 1952, Fauna U.S.S.R., n.s. no. 56, 1(2): 218 (in text), 220 (in text) — probably a juvenal of M. rufina Pucheran. Haltenorth, 1963, Handb. Zool., 8(32): 48— ECUADOR; PERU; characters. M{azama\. m[ephistopheles\. mephistopheles [sic], Haltenorth, 1963, Handb. Zool., 8(32): 48 — classification. Mazama fusca Spillmann, 1931, Die Saugethiere Ecuadors im Wandel der Zeit. Univ. Cen- tral, Quito, 1: 37, 41 (measurements) — ECUADOR: Pichincha (type locality, Mt. Antisana near snow line); holotype, adult male, skull only, perhaps preserved in the Escuela HERSHKOVITZ: NEOTROPICAL DEER 45 Politecnica, Quito. Hoffstetter, 1952, Mem. Soc. Geol., France, 31(66): 369, regarded as a synonym of P. mqjhistophiles. Pudu (Pudella) mq)histophiles wetmorei Lehmann, 1945, Rev. Univ. Cauca, Popayan, (Co- lombia), Zool. no. 6: 76, figs. p. 78, 80 (animal; skull)— COLOMBIA: Cauca (type locality. Paramo de San Rafael, Purace, ESE Popayan, western slope of the Cordillera Central, 3400 m); holotype, adult male, skin and skull, Museo de Historia Natural, Universidad del Cauca, Popayan, no. 3801, collected 1 February 1945 by F. C. Lehmann. Cabrera, 1%1, Rev. Mus. Argentino Cienc. Nat. "Bernardino Rivadavia," 4(2): 344 — a synonym of P. mephistophiles de Winton. P{udu\. mq)histophiles ivetmoorei [sic], Hoffstetter, 1952, Mem Soc. Geol. France, 31(66): 369— COLOMBIA. P[udu]. m[ephistophiles]. wetmorei, Mochi and Carter, 1953, Hoofed mammals of the world, Scribner's and Sons, Ltd., London, pi. VIII, text only. Pudu mephistophiles wetmorei, Lehmann, 1959, Nov. Colombianas, 1(4): 202, fig. (animal), fig. 3 (antlered skull) — COLOMBIA: Cauca (Paramo de Guanacas, partial albino; Paramo de las Delicias, albinos); Valle (Paramos de Barragan, east of Tulua); description; skull; hunting; conservation. Whitehead, 1972, Deer of the world. Viking Press, New York, pp. 29, 64, 169, map 9 (distribution) — classification; distribution. P[udu]. mephistophelis [sic] wetmoorei [sic] Hoffstetter, 1952, Mem Soc. Geol. France, 31(66): 369— COLOMBIA. Pudu mephistophelis [sic] wetmori [sic], Keirans, 1973, J. Med. Entemol., 10(3): 254 — COLOMBIA: Cauca (Pusna, 2800 m); tick parasite {Ixodes montoyanus). M[azama\. m[ephistopheles]. [sic] wetmorei, Haltenorth, 1963, Handb. Zool., 8(32): 48 — COLOMBIA. Pudu sp. Keirans, 1973, J. Med. Entemol., 10(3): 254— COLOMBIA: Cauca (Malvasa, 3500 m, 3000 m); tick parasite (Ixodes montoyanus). Type. — Originally described from the skin and attached leg bones of an imma- ture female, and a mismatched skull of Mazama rufina, both presented in 1896 to the British Museum (Natural History), by Ludovic Soderstrom. The mismatch was discovered by Thomas (1908, p. 350) and the type was restricted to the skin BM 96.1.28.5, by Lydekker (1915, p. 217). Type locality. — Said to be the Paramo de Papallacta, but probably taken in temperate zone forests below paramo zone, southeast of Quito, eastern slope of the Cordillera Oriental, Pichincha, Ecuador. The altitude is said to be between 3,6(X)-3,700 m above sea level, but probably should be estimated at 200 or 300 m lower. Distribution (fig. 30) Restricted to the temperate zone forests and fringing grasslands or paramos above treeline in the Cordillera Central of southern Colombia, from the depart- ments of Valle and Tolima southward through the Cordillera Oriental of Ecuador and Peru into the department of Junin. The parts of the Andean chains actually occupied are poorly known, and distributional gaps between locality records in northern Ecuador and central Peru are unresolved. According to Lehmann (1945, p. 76), the Colombian pudu is known to occur in the Cordillera Central, from the Paramo and Valle de las Papas and probably the Paramo del Lebrero, northeast of the Valle de las Papas; northward through the paramos of Sotara, Purace, Las Delidas, Guanacas, and Moras, in the depart- ment of Cauca, to those of Las Hermosas in the department of Valle. The pudu has also been reported by Lehmann (1959, p. 203) to the north in the region of the Paramos de Barragan east of Tulua, department of Valle. In Lehmann's opinion, the range never extended farther north because of a deep depression in Fig. 30. Distribution of Pudu mepbistophiles de Winton, based on museum specimens and published 19th and 20th century locality records (dots). For locality names, see corre- sponding numbers in "Gazetteer" (p. 80). 46 HERSHKOVITZ: NEOTROPICAL DEER 47 the Andes that acts as a climatic barrier. Altitudinal range is given as between 3,000 and 4,000 m. Dr. Hernando Chirivi Gallego, head of the Laboratorio de Fauna of INDE- RENA (Institute de Desarrollo de los Recursos Naturales Renovables), Minis- terio de Agricultura, Bogota, informs me (/>i litt., 23 April 1979) that the range of Pudu mephistophiles may extend as far north as Mt. Quindio (4°42'N, 75°25'W). He adds that a skull of a pudu from the Paramo de Moras is in the INDERENA collection, and that pudus are well known by the name venado chonta in the paramos of Azufral, Cumbal, and Chiles in the department of Narifio. The parts of paramos presumably inhabited by pudus are actually sub- paramos, mostly — if not entirely — secondary in origin and characterized by low, thin, fire-resistant forest growths or matorrales, usually dominated by Polylepis. The steep, eastern slope of the Cordillera Central in the upper Rio Magdalena valley, below the Paramo de las Papas in the department of Huila, where 1 found signs and remains of pudus, ranges from 2,700 to 4,000 m above sea level and is thickly forested. The seeming absence of pudus from the Colombian Cordillera Oriental, which joins with the Cordillera Central at the Paramo de las Papas, is perplexing. The head of the Rio Magdalena, which marks the dichotomy between the eastern and western chains, is narrow and shallow enough at altitudes above 2,000 m for a pudu to jump, walk, or step across, and no doubt pudus do as occasions require. Farther down and eastward, the crest of the Cordillera Oriental loses altitude, and, passing the point (Picos de Fragua) where it turns sharply north- eastward, it forms a deep saddle at the divide between the Rio Suaza (Mag- dalena) and Rio Orteguaza (Caqueta). Here the climate is comparatively dry, upper tropical to subtropical, and acts as a barrier to northern spread of many temperate zone mammals, including pudus, but serves (or served) as a pass for tropical zone Amazonian mammals. The Colombian Cordillera Occidental, ex- cept for a few isolated peaks and subparamos, is too low to provide suitable habitat for pudus. The pudu has not been recorded from Carchi and Imbabura, the two Andean provinces of northern Ecuador, but is known there by the name sacha cabra. One was captured with hunting dogs in the virgin forests northeast of San Gabriel, Carchi, during my explorations there in 1934. Dr. Fernando Ortiz Crespo of the Pontifica Universidad Catolica del Ecuador in Quito, informs me (in litt., 6 March 1979) that he heard rumors of pudus being sighted around Volcan Chiles in the western Andes at the Ecuadorian- Colombian border. Dr. Ortiz mentions that the scarcity of the species may be due to competition with Mazama in forests between 2,500-3,200 m altitude, and to the absence of suitable habitat higher up. He observes, however, that the animal is rarely encountered because of its nocturnal or, at least, crepuscular habits and the difficulty of access to its extremely accidented habitat where it can be hunted only with dogs. He particularly notes that pudus are not confined to the isolated patches of Polylepis forest found in paramo and that these animals, including the holotype recorded from the Paramo de Papallacta, may have origi- nated in forests well below the paramo zone. The occurrence of pudus in the southern half of Ecuador, south of the Rio Pastaza, and northern Peru, north of Huanuco department, is not on record. However, Dr. Antonio Brack Egg, a distinguished naturalist, formerly of the Direccion General Forestal y de Fauna, Lima, believes {in litt.) that pudus may be present in the paramos of Huancabamba at the headwaters of the Rios Quiroz, 48 FIELDIANA: ZOOLOGY Chinchipe, and Huancabamba, in the departments of Piura and Amazonas on the Peruvian-Ecuadorian border. Grimwood (1%9, p. 79) provides the following details on the occurrence of pudus in central Peru: The occurrence of this genus within Peruvian limits does not appear to have beeri previously recorded. There is, however, a misidentified mounted specimen of Pudu mephistophiles (De Winton) in the museum of the Colegio "Leoncio Prado" at Huanuco, which was shot some years ago at Pamacucho, 4 kms. from that town; and in 1965 I obtained the skin and skull of a male of that species which had been killed near Carpish summit on the Huanuco-Tingo Maria Road. Subsequently, I have seen the animal alive in that region and have been shown skins or photographs of other animals killed there and in a number of localities in the upper Huallaga valley ab>ove the town of Huanuco. The distribution of this spedes appears to be very limited. At Carpish it is quite common in the forests on both sides of the divide, from the summit at approximately 3,000m. down to about 2,300m. on the eastern side and down to where forest gives way to cultivation at about 2,000m. on the western side. It is also known from similar country near Chaglla, 30 kms. further south. In the upper Huallaga basin the species occurs in a much drier habitat, and has been shot at Yanacocha, on the Rio Quere [sic=Quero], 25 kms. northwest [sic=SW] of Huanuco; at Pamacucho; and in a number of small patches of woodland between Huanuco and Ambo which lie at 2,700 to 3,300m., are no more than a few square kilometers in size, and are separated from each other by many miles of arid hillside. It is also reported to occur in an isolated patch of forest at 3,300m., 40 kn\s. from Huanuco on the Quivilla road. I believe Pudu mqjhistophiles also to occur in parts of the Department of Junin, although I have seen no specimen from that region. Local hunters, however, have unhesitatingly identified my Carpish specimen as the same animal as one known locally as "Antagllo" which is reported to be very plentiful at elevations of from 2,000 to 3,d00m. on the forested heights above San Ramon and also in the valley of the Rio Hausanhausi [sic], to the north. The antagllo is only doubtfully known in the Oxapampa area, but is reliably reported from the upper Rio Paucartambo at approxi- mately 76.00'W, 10.50'S. The Peruvian population of Pudu mephistophiles appears to be entirely isolated from the nearest known population in Ecuador, since no small deer of any kind is known in any area north of Huanuco where 1 have made enquiries. The specimen from Carpish mentioned above was shipped to the Field Museum but never arrived (Grimwood 1969, p. 80). To date, the only preser\'ed specimen of a Peruvian pudu known to me is the pelt from Bosque Zapatogocha, Huanuco, presented to the Field Museum by Peter Hocking. Early reports of the occurrence of pudus in southern Peru and Bolivia prove to have been based on pelts or bones of the dwarf brocket, M. chunyi (Hershkovitz, 1959a). Conservation The late Professor Lehmann warned in 1960 (p. 282) that the pudu was being hunted almost daily in the Colombian highlands by campesinos and often by villagers who sometimes killed as many as six or seven deer in a single organized hunt. At this rate of attrition, he argued, the pudu's extinction was foreseeable in the near future. Lehmann added that account must be taken of the limited range of the species (as known then) and of the increasingly large portions that man was converting to his own use. This alone, he declared, is enough to doom the pudu unless the government provides the vigilance and control needed to HERSHKOVITZ: NEOTROPICAL DEER 49 ensure survival of the natural fauna and flora by establishment of appropriate nature reserves. The Colombian government has since (1968) founded the Purace National Park and others. Pudus of the Purace region, the first known from Colombia, served Lehmann for the original description of his Pudu mephistophiles wetmorei. Oddly, Hunsaker (1972, p. 445) mentions the mountain tapir {Tapirus punchaque) but overlooks the pudu among the protected species of this park. Present status of P. mqthistophiles in Colombian National Parks has been summarized by Dr. Hernando Chirivi Gallego (see above, p. 47). The informa- tion is abstracted below with some comments. For location of parks listed, see "Gazetteer" (p. 80) and map (fig. 30, p. 46). Additional information on Colom- bian national parks is provided by Hunsaker (1972) and Harroy (1972). LOS NEVADOS, Cordillera Central, astride boundaries between departments of Caldas, Resaralda, Quindio, and Tolima; 4°45'N, 75°20'W; 50,000 ha (200 sq miles); established 1977. Not plotted on map. Figure 30. "Local informants distinguish a small reddish deer whose description agrees with that of Mazama rufina, and a smaller dark one like Pudu mephistophiles" (Chirivi"). This park lies within the range of Mazama rufina, but well north of the northernmost authenticated locality recorded for the northern pudu. LAS HERMOSAS, Cordillera Central, departments of Valle and Tolima, established 1977. "A mounted specimen of Pudu mephistophiles is in the Hunters Club of Ibague, Tolima. H was taken in the pdramo above Roncesvalles, Tolima" (ChirivQ. Lehmann (1945, p. 76) also recorded the pudu from the Valle side of the Paramo de las Hermosas. NEVADO DEL HUILA, departments of Huila, Tolima, and Cauca. "There is no information but in all probability the pudu is present" (ChirivO- PURACfi, Cordillera Central, Cauca; 80,000 ha (293 sq miles), 2,500-4,700 m; established 1968. "A specimen (skin and skull) from Laguna de Cusiyaco near the headwaters of Rio Magdalena, is preserved in the Instituto de Ciendas Naturales, Universidad Nadonal de Colombia, Bogota" (ChirivO. MUNCHIQUE, Cordillera Central, Cauca. "Dr. Heliodoro Sanchez, Chief, Division of National Parks, INDERENA, observed a pudu from a distance of a few meters on the western slope of Cerro de Munchique, between 2,500-2,600 m" (Chirivi). CUEVA DE LOS GUACHAROS, Cordillera Oriental, Huila, 15,000 ha (61 sq miles), 1,700-4,000 m. "There are vague rumors of the existence of pudus in the highest mountains surround- ing the Park" (ChirivQ. Pudu hunting in Colombia, Dr. Chirivi informs me, has been prohibited since 1969. However, without the necessary vigilance in the mostly unsupervised parks, survival of the species, where presently known to occur, is dubious. In Ecuador, the Parque Nacional de Cotopaxi harbors pudus. The pair cap- tured west of Salcedo and exhibited in the Berlin zoo (fig. 28) must have origi- nated in this park. There may be more havens. Dr. Fernando Ortiz Crespo informs me (see above, p. 47) that the Ecuadorian Ministerio de Agricultura and Ganaderia plans to establish nature reserves that include pudu habitats. Among areas under consideration are the slopes of Mt. Cotacachi, Imbabura Province; the eastern flanks of Mt. Tungurahua, Tungurahua Province; Mts. Altar and Sangay in Morona-Santiago; and the eastern slope of the Cordillera Oriental at 50 FIELDIANA: ZOOLOGY the latitude of Mt. Cayambe, Napo, which would incorporate the type locality of Pudu miyhistophiles. Dr. Ortiz adds that conservationists can speed up the estab- lishment of such reserves by exerting pressure on the highest governmental levels. I.R. Grimwood, Technical Adviser on Wild Life for the British Ministry of Overseas Development, on assignment to the Peruvian Servido Forestal y de Caza, is equally urgent in his assessment of the situation in Peru. He (1969, p. 80) gave this observation: It is improbable that this species will survive for much longer in the isolated forest patches in which it occurs in the upper Huallaga basin, where it is relentlessly hunted with dogs and guns. It has, however, outlasted the white-tailed deer in many of them because it does not break cover, as does that species. . . . The species is also hunted in the Carpish, Chaglla and San Ramon areas, but in these regions destruction of habitat for settlement is likely to present a greater threat. Control of hunting is, however, necessary. Pudu mephistophiles does not occur in any existing or projected national park or reserve [cf., Dourojeanni, 1968]. The creation of a special sanctuary, for which the high ground above the San Ramon gorge is suggested, is therefore desirable. In a personal communication dated 1 April 1979, Dr. Antonio Brack expresses doubt that the pudu ever occurred in the San Ramon region. In any event, he adds, the original environment has been destroyed and the region now is un- suited for a reserve. Brack further states that the only area where pudus are known to occur in abundance is in the Cordillera de Carpish, department of Huanuco, where the government plans to establish a reserve when funds and national conditions permit. In Brack's opinion, Pudu mephistophiles in Peru is not seriously threatened, as long as good habitat over an extensive zone of the eastern slope of the Andes remains closed to highways. A national park worth mention because it may shelter pudus in an area where never recorded is that of San Andres de Cutervo (6°13'S, 78°42'W), Cajamarca, altitude, 2,200-3,000 m. It is the smallest of Peru's national parks (2,500 ha). Harroy (1972, p. 77) describes it as ". . . totally protected: no occupation, no exploitation; lumbering is now forbidden almost everywhere. Staff headquarters at the town of San Andres de Cutervo, 3 km from the Park." Vernacular Names The Colombian pudu, according to Lehmann (1945), is called venado conejo (rabbit deer). Those with antlers in velvet may be distinguished as caciques (chiefs). Melanic individuals of the Valle de las Papas, Cauca, are recognized as a distinct variety called venado chonta (chonta is a species of palm with dark trunk). On the eastern slope of the Paramo de las Papas, Huila, where I worked, venado chonta is the name applied to the small, red mountain brocket, Mazama rufina; whereas the sympatric pudu is called venado churucu (or churuco, also used for the blackish, woolly monkey, Lagothrix lagothricha lugens). The most widely used name for the pudu in Ecuador and Peru is sacha cobra (wdld goat). The notation "local name ollique" appears on the collector's tag of a specimen from Cayambe, Ecuador, preserved in the British Museum (Natural History). Grimwood (1969) includes fl«teg//o as the name used in Junin, Peru. The Chilean name "pudu" now is universal for both recognized spedes of the genus Pudu. I follow Haltenorth (1963, p. 48) in distinguishing the two as "northern pudu" and "southern pudu." HERSHKOVITZ: NEOTROPICAL DEER 51 Taxonomy The original description of Pudu mephistophiles de Winton is based on a skin with foot bones intact, and the skull of a subadult brocket, apparently Mazama rufina Pucheran. The skin and mismatched skull were received from Mr. Ludovic Soderstrom, British consul in Quito, Ecuador. De Winton (1896, p. 511) exam- ined the foot bones still in the skin and observed that "Osteologically they agree with P. humulis [=Pudu puda], the ectocuneiform and navicular-cuboid bones being all in one." The erroneous association of the type skin and foot bones with a Mazama skull was discovered by Thomas (1908) when he received two additional specimens from Mr. Soderstrom, both with properly matched skins and skulls. The three known skins of P. mephistophiles from the type region in northern Ecuador average slightiy more saturate in general body color than southern Colombian representatives of wetmorei, and one of them agrees with the holotype of wetmorei as described. A skin from the Cordillera de los Llanganates, south of the type region of mephistophiles, with a buffy or olivaceous-appearing coat is absolutely indistinguishable from the majority of Colombian specimens I examined, including four topotypes of wetmorei. An adult male and female, also from Llanganates, reported by Fradrich (1975, p. 75) are said to "differ insignifi- cantly from the more northern wetmorei. " In the absence of any notable difference in color or any other character including antier shape, which varies even more widely, there appear to be no grounds for recognition of wetmorei as a race of P. mephistophiles. Characters Size. — Smallest of living deer; measurements are given in Table 3. Coloration and pelage (figs. 31-34). — General body color buffy agouti (oliva- ceous) to reddish brown agouti, the middorsum dominantiy dark brown, pelage stiff, adpressed, the hairs pithy; cover hairs of middorsum about 35 mm long, guard hairs like cover hairs, but about 5 mm longer, basal half of both types pale brown or grayish brown, terminal half dark brown (eumelanin) with buffy to reddish (pheomelanin) subterminal band 0.5 mm wide; fine brown curly wool hairs thinly spread and completely hidden to view by cover hairs. Head, in- cluding muzzle, chin, outer surface of ears, contrastingly colored dark brown or blackish (eumelanin), the hairs generally unhanded, specialized markings ab- sent; inner surface of ears sharply defined whitish, grayish, or pale buff; rostral hairs banded, usually with a single, narrow pheomelanic ring; orbital ring or stripe absent; neck, including nape, dominantiy buffy, the individual hairs with broad pheomelanic bands. Belly, inner sides of thighs and arms buffy to rufous, the hairs unhanded; forefeet and hind feet dark brown contrasting with buffy to reddish agouti of flanks; hoofs extremely long and narrow. Fawns unspotted, pelage coarse, less hispid, more relaxed, color pattern more uniform, without markedly contrasting dark head and legs. Color variation. — General body color of the holotype is reddish brown; back of head, nape, and shoulders more or less like back, rump more saturate than middle of back. Of two juvenals in the British Museum (Natural History), one, a topotype from Papallacta, is slightiy more saturate throughout than the holotype; the other, from Mt. Cayambe in type region, is slightiy paler. ft If ^ S « o c Q o c (B ^ 5 13 Z VI -Si c 2 1- (Q 3 01 2^ ■^ h TJ ;3 13 O « CL. £ 2 £ ■»-; O ax. 3 5 T3 w « "o T s 11 60 C c >.X> ^ N (A C ^■d 01 s J3 £5 3 m «* 01 J^ <0 fM 3 C/5 60 — 6 ^ i2 « US 0) > a3 .5 o fO w .J 09 < H Head and body , »-l rl (N in l-i J^ (Ni 00 IT) \o rj r5 ifi S o S \S I in CN 00 IT) n£i i-i I x> inoot^fOfO'^TfoOTtinvO'^o vovbtnro^^'^inrviTfKsdiriT*" I I rorOrorofOrOfOrrifrirnfOfrirriro Tj.r-irH rslOOfOCT^'^ oo i>iis!oovd(Ncoin'* tNinKinin rjint^ \0 |\D^0\O>.0t^vOsD^0 |\OsD>.OsO\0 sOvOsO :g iit^it^iiiiit>.iii < -■ O u o ^ O O T-i _ (U ^ ■^ ^ g 60 o j2 it ^X> (1t3 °° Q (ft *- )S _ -C (N in 00 f> <" ^ o in o Q o >>>>>> >>>>^^^5 U UJ PLI w -^' s _ p -J in • i; ^ CU 3 C-in « (0 O 01 (0 (Q O „ N ^ S;S £J O "C o vt: oj § CO "? S c ^ C .3 a; CiJ 2 ^ 0) t~N 01 O CT^ u? tfA.i.-!V.Uk J U. ...'....iT^ ■^^:.':.;-W'i.v,'"r"j;,4^»jjr n~] pheotnelanin slightly or not pigmented Fig. 34. Randomly selected hairs plucked from 11 chromogenetic fields of preserved skins of Pudu piida and P. mqjhistophiles. Skull (figs. 11-17). — Lacrimal fossa shallow or dish-shaped without fenes- trations, its ventral border hardly or not at all encroaching on malar and maxil- lary bones; superior border of ascending ramus of premaxillary expanded, con- tacting lateral margin of nasals; anterodorsal border of lacrimal flat; long axis of antorbital vacuity nearly or quite vertical; plane area of vacuity generally much less than one-half, rarely slightly more than plane area of facial plate of lacrimal; perforations on upper surface of zygomatic process of squamosal, if present, usually less than 1 mm in diameter each; angle of mandible produced well beyond posterior plane of condyle as an expanded subrectangular process; lunar notch between angle and condyle, deeply excised, concavity between base of angle and ramus well defined. HERSHKOVITZ: NEOTROPICAL DEER 57 Cranial measurements are given in Table 3. Dentition. — Upper permanent canine persistent in eight of 15 skulls with com- plete permanent dentition; deciduous canine present in two of three young with only first molars erupted, and in two of three with second molars erupted; metaloph and metaconule present, postentoconule absent in fully erupted mo- lars (fig. 23). The form of left pm^ in FMNH 24362 (fig. 16) is that of a deciduous tooth. In size and wear, however, it agrees with the upper third premolar of the right side and of other pudus. There is no sign of a replacement tooth. Ostensibly, the deciduous -like premolar is the permanent tooth. Antlers. — Adult male skulls examined were collected October through Feb- ruary and one in July in the department of Cauca, Colombia. All support polished antlers, the longest of which exceeds 90 mm (table 3). Antlers of the holotype of Pudu mephistophiles wetmorei, collected February 1945, are also polished (Lehmann, 1960, p. 283). The adult male of a pair of Ecuadorean pudus that arrived in the Berlin zoo during April rubbed the velvet from its antlers in June (Fradrich 1975, p. 76). The foregoing suggest that antlers of adult male pudus in southern Colombia and northern Ecuador may be shed from February into March, are in velvet from March into June and possibly July, and are polished from June or July until cast. I find no information on antlers of Peruvian representatives of the spedes. A skull of a three-horned pudu from Paramo de Guanacas, Cauca, Colombia, without date of collection, is figured by Lehmann (1959, p. 204). The antlers, still in velvet, are said to be fully developed and completely separated. Two antlers occupy normal positions and in the photograph appear to be more or less sym- metrical. The third antler, slightly shorter than the others, arises directly to the right of the normal right antler. Vertebral formula. — In a male (FMNH 86846) from Malvasa, Cauca, Colombia: 7 cervical, 13 thoracic, 6 lumbar, 4 sacral, 8 caudal. Comparisons Pudu mephistophiles is distinguis|ied from P. puda, its only congener, by char- acters already mentioned under generic headings (p. 15), key to the spedes (p. 43), and in the description of the spedes (above and pp. 65, 70). The only other deer with which the northern pudu need be compared is the smallest known brocket, Mazama chunyi. Using greatest skull length as most reliable indicator of overall body size in present material, that of one fully adult Mazama chunyi (all available) is 146 mm, whereas mean and extremes of 14 adult Pudu mephistophiles are 132.5 mm (127- 142). Externally, as represented by four skins, including holotype, M. chunyi more nearly resembles pale Colombian representatives of P. mephistophiles than the saturate Peruvian individual at hand. The brocket is separable by its white narial (trimmed away in holotype) and mental patches, tail distinct from rump, the underside sharply defined white, ears probably similarly short and rounded but edges white, their outer surface brownish like muzzle and crown, not black- ish, coronal hairs not whorled, body color dominantly pheomelanin, not eumelanin or agouti, lower portion of legs not markedly different in color from upper legs or flanks, and fawn spotted. 58 FIELDIANA: ZOOLCX^Y In Mazama chunyi, as in all brockets, external cuneiform bone is present as a discrete tarsal bone, and crown of first permanent incisor is broadly spatulate, its greatest length more than combined length of iz-a- For additional characters, see comparison between Pudu and Mazama (p. 22). Specimens Examined Total 29. COLOMBIA. Cauca: Paramo de Purace, 4 (USNM, 1; FMNH, 1; AMNH, 2); Malvasa, 15 (FMNH); Quintana, 1 (FMNH); La Vega, 1 (FMNH); Paletara, 2 (AMNH); Coconuco, 1 (AMNH). ECUADOR. Pichincha: Paramo de Papallacta, 2 including holotype (BM); Mt. Cayambe, 1 (BM); Tungurahua: Cor- dillera de los Llanganates, 1 (ANSP). PERU. Hudnuco: Bosque de Zapatogocha above Acomayo, 1 (FMNH). V. BIOLOGY OF Pudu mephistophiles Habitat In tracking the northern pudu, I found the animal confined to the central Andean forests in southern Colombia and the eastern Andean forests in north- ern Ecuador, at altitudes between 2,700 and 4,000 m. The severely acddented terrain where the animal still may be found probably is its last retreat rather than the predilect abode of this persistently persecuted little animal. The southern Colombian habitat is described by Lehmann (1945, p. 78) as the dense brush or matorrales of the paramos and the sedge and frailejdn (Espeletia) covered swamps of the Andean summits. I have never seen signs of pudus where grasses or frailejdn dominate. However, Lehmann points out that pudus sometimes attempt to escape pursuing dogs by climbing trunks of trees inclined over streams or bluffs, which occur only in the temperate zone forests below true paramo. In any event, pudus may invade borders of interdigitating grasslands or paramo to eat the succulent herbs and shoots that appear with the first rains after the dry season burnings. The foUov^ng is Grimwood's comparison (1969, pp. 77) of the habitat oiPudu mephistophiles in central Peru with that of the very similar-appearing pygmy brocket, Mazama chunyi of southern Peru: The two spades occupy the same type of habitat, avoiding the cold and damp of the highest woods and of the thickets above tree line, and [are] found on the eastern [Andean] slopes only from about 3,200 m downwards, in regions where [the alder] Alnus jorulensis is often the dominant tree. Both spedes are also known from a rather different, drier habitat, in isolated patches of forest or woodland in the high valleys immediately to the west of the main eastern watershed. Habits Little is known of the behavior of the northern pudu in the wild beyond what can be inferred from its habitat (see preceding section) and gross morphology. Grimwood (1%9, p. 77) compared the pudu with the very similar Mazama chunyi and noted that, "Both are very small nocturnal deer, found only singly or in pairs, which emerge in clearings or fields at dusk. Both are much given to whistling and both are said often to be dazed if disturbed by day, so that they can sometimes be caught by hand." Lehmann (1945, p. 78) observed that, when pursued by dogs, the pudu runs in a weaving pattern. Grimwood also noticed that, in flight, the animal does not HERSHKOVITZ: NEOTROPICAL DEER 59 break cover, but circles in the thickets and among the cliffs. Its long, fine, widespread hoofs facilitate climbing, and, as previously noted (Lehmann, 1945, p. 76), the pudu sometimes escapes dogs by running or climbing up inclined trees. Grimwood believes, on the basis of information received from hunters, that the little deer is as nimble on rocks as the African klipspringer (Oreotragus). Maintenance in Captivity Northern pudus are frequently captured alive, but, according to Lehmann (1945, p. 78), they do not survive long in captivity. One pudu in Lehmann's possession in Cali, Colombia, became very tame and particularly fond of milk, but died in two months. Lehmann adds that a friend, Guillermo Valencia, con- fessed his inability to keep captured pudus alive for more than a few weeks despite all efforts, including maintenance in a natural environment. The first northern pudus received in a European zoo were two females from Ecuador that arrived dead or dying 23 November 1968, at the Berlin zoological garden. This report, by Fradrich (1975, p. 73), includes a notice without particu- lars of the later accession in 1970 of a northern pudu by the same zoo. Sub- sequently, Fradrich continues, a pair captured in Ecuador, west of Salcedo be- tween Mt. Cotopaxi and the Cordillera de los Llanganates, arrived at th e Berlin zoo in good condition 12 April 1972. After the usual quarantine, the pair was placed on public view and fed a standard brocket diet (Fradrich, 1974). The male died 6 August 1972. The autopsy showed pasteurellosis, tracheobronchitis, and fibrinous pneumonia. The female died of food poisoning 11 October 1972. I find no other published records of Pudu mephistophiles in zoological gardens. Ungulate Associations Cervids occurring within parts of the range of Pudu mephistophiles are the brockets, Mazama rufina and M. americana, white-tailed deer, Odocoileus vir- ginianus, and taruga, Hippocamelus antisensis (fig. 5). The latter, known from the Andes of Ecuador, Peru, Bolivia, and northern Argentina is nearly, if not al- ready, extinct in Ecuador and Peru. The pinchaque or woolly or mountain tapir, Tapirus pinchaque, associates with the pudu in Colombia and Ecuador; its num- bers are dwindling. The dwarf deer, Mazama chunyi, is common in the southern Peruvian depart- ments of Cuzco, Puno, and Madre de Dios in habitats appropriate for Pudu mephistophiles, but the latter is unknown south of the department of Junin. Information on the distribution of small deer in southern Peru is meager, and identifications often have been equivocal or based on hearsay. The pudu, com- mon as it appears to be in central Peru, was not reliably reported from that country until 1969 (Grimwood, p. 79), and the first museum-preserved speci- men is recorded here. M. chunyi, long confused with P. mephistophiles, was not recognized as a distinct spedes until 1959. The two little deer may yet prove to be sympatric. Reproduction A female from Ecuador that arrived at the Berlin zoo 12 April 1972 died 11 October 1972 carrying a fully developed female fetus (Fradrich, 1975, p. 76). Grimwood (1969, p. 80) reports that, in central Peru, nearly full-term fetuses were taken from pudus killed in April and November. 60 FIELDIANA: ZOOLOGY The meager data indicate that, in Pudu mephistophiles, a gestation period of seven months is followed by a postpartum estrus as in Pudu puda. VI. Pudu puda MOLINA SOUTHERN PUDU (figs. 36, 37) Synonymic History Capra puda Molina, 1782, Saggia storia Nat. Chile, libr., 4: 308, 343. Capra puda, Anonymous ("An American Gentleman"), 1808, The geographical, natural and civil history of Chile, 1: 217, 240 — "Translation from the original Italian of Abbe don J. Ignatius Molina, Saggia . . . ." Printed for J. Riley, Middletown, Conn. [Ovis] Pudu [sic], Gmelin, 1788, Linnaeus Syst. Nat., ed. 13, 1: 201 — classification. Ovis Pudu, Shaw, 1800, Gen. Zool., 2(2): 392. [Aries]pudu, Fischer, 1814, Zoognosia Tab. Syn., 3: xvi, 406. A[ntelope]. Pudu, Blainville, 1816, Bull. Soc. Philomatique, pp. 76, 80 — classification. M[azama]. pudu, Rafinesque, 1817, American Monthly Mag., 1: 437 — regarded as a goat antelope. Mazama pudu, Vanoli, 1967, Saugetierk. Mitteil., 15(2): 155, figs 1-10 (adults, young) — CHILE: Osorno-Llanquihue; habits, habitat, reproduction; food; growth; development- antlers; enemies; disease; conservation. Mazama [{Oreamnos)] puda, Rafinesque, 1817, American Monthly Mag., 2: 44 — classification. M{azama (Pudu)]. pudu, Haltenorth, 1963, Handb. Zool., 8(32): 48— classification. Antilocapra pudu, Lesson, 1842, Nouv. Tabl. Reg. Animal, Mamm., p. 182 — classification. Cer^MS pudu. Gay and Gervais, 1846, Ann. Scd. Nat., (3)5: 88 — CHILE: classification; de- scription; measurements; Cervus humilis Bennett, a synonym. Gay, 1847, Hist. Chile, Zool., 1: 158, Atlas, pi. 9 (animal), pi. 1 (skull) — CHILE: from Cauquenes toChiloe; Indian name, pudu or puudu. Pucheran, 1852, Arch. Mus. Hist. Nat. Paris, 6: 482 — characters; measurements; history. Sclater, 1866, Proc. Zool. Soc. London, 1866: 104, fig. (animal) — menagerie specimen. Sdater, 1871, Proc, Zool. Soc. London, 1871: 238, pi. 17 (animal) — menagerie specimen; antler shedding. Garrod, 1877, Proc. Zool. Soc. Lon- don, 1877: 4, 13, 14 — alimentary tract; penis; cranial characters; classification. Huet, 1890, Bull. Soc. Nat. Acclim. France, 1888/1889, p. 86— coloration. C[ervus]. pudu, Wagner, 1855, Schreber's Saugth., Suppl., 5: 390, pi. 26 (animal) — characters; humilis Bennett a synonym. Coassus [(Pudu)] Pudu Gray, 1850, Gleanings Knowlsely Menag., p. 69 — characters. Gray, 1852, Proc. Zool. Soc. London, 1850: 242 — CHILE: Concepcion; Chiloe; classification; characters; part synonymy: Capra pudu Molina, Cen'ws pudu Gervais, Cervus humilis Bennett. Pudu pudu, Sdater, 1870, Proc. Zool. Soc. London, 1870: 116 — dassification. Wolffsohn and Porter, 1908, Rev. Chilena Hist. Nat., 12: 74 — CHILE: Valdii'ia; local names, venado, venadito, pudu (Araucano). Pocock, 1911, Proc. Zool. Soc. London, 1910: 967, 978 — interdigital gland. Lydekker, 1915, Cat. Ungulate Mammals Brit. Mus., 4: 215 — characters; synonymy; CHILE: Maule; Constitucion; Concepcidn (Concepdon); Cautin (Temuco). Synonyms, humilis Bennett, chilensis Gray. Pocock, 1923, Proc. Zool. Soc. London, 1923: 201, fig. 13 (foot gland), fig. 14fc (rhinarium), fig. 15 (ear) — external characters. Wolffsohn, 1923, Rev. Chilena Hist. Nat., 27: 160— CHILE: Concepcidn (Pedro Valdivia); Maule (Constitudon); measurements. Yepes, 1929, Rev. Chilena Hist. Nat., 33: 472 — CHILE: Maule (Constitudon) to Magallanes; ARGENTINA: western Neuquen to western Santa Cruz. Junge, 1932, Zool. Garten, 5: 249 — sensitivity to heat. Cabrera and Yepes, 1940, Mamiferos Sud- Americanos. Hist. Nat. Ediar (Buenos Aires), p. 278, pl. 88 (animal); characters; distribution. Yepes, 1943, Rev. Argentina Zoogeogr., 3: 75 — ARGENTINA: Neuquen (Cordillera del Trompul, San Martin de Los Andes; Brazo Co- rrentoso, south of Cordillera del Trompul; La Lolog, San Martin de Los Andes); local name, corzuelo. Osgood, 1943, Field Mus. Nat. Hist., Zool. Ser., 30: 228— CHILE: Magallanes (Isla Riesco; Rio Baker): Llanquihue (Cayetue [sic], Lago Todos Santos; Pe- trohue, Lago Todos Santos); Chiloe; Cautin (Pitrufquen); Concepcidn (San Pedro); Arauco (Santa Juana; Vega Blanca, Sierra Nahuelbuta); Valdivia (Valdivia); characters; variation; HERSHKOVITZ: NEOTROPICAL DEER * 61 habits. Schmidt, 1944, Argentinische Sdugetiere. (Buenos Aires); pp. 170, 237, 5 figs, (adults and young), 1 fig. (skull) — ARGENTINA: Rio Negro (Lago MascardQ; Neuqu^ (Rio Quillen). Flerov, 1952, Fauna U.S.S.R., n.s., no. 56, Mammals, 1(2): 220 (in text)— listed. Junge, 1954, Bol. Jardin Zool., Concepdon (Chile), 1: 3, fig. (animal) — two females in captivity. Hershkovitz, 1958, Mammalia, 22(4): 545 — metatarsal glands and tuft not defined. Mochi and Carter, 1953. Hoofed mammals of the world, Scribner's and Sons, Ltd., London, pi. VIII, fig. 16 (animal) — characters. Hershkovitz, 1959, Proc. Biol. Soc. Washington, 72: 45, pis. 3D, 4B (skull) — comparisons. Walker, 1964, Mammals of the world, Johns Hopkins Press, Baltimore, 2:1399, fig. (animal)— CHILE; "BOLIVIA." Crandall, 1964, The management of wild animals in captivity. U. of Chicago Press, pp. 557, 560, 594 — characters; exhibition in zoos. Greer, 1965, Publ. Mus. Midhigan St. U. Biol. Ser., 3(2): 146— CHILE: Malleco (Laguna Malleco; Galvarino, 10 km W, 61 m); Angol (Cordillera de Nahuelbuta); measurements. Junge, 1966, Int. Zoo Yearb. 6: 263 — CHILE: Angol; Malleco; Valdivia (Mariquina); habits in captivity. Windecker, 1967, Fr. Kolner Zoo, 10(2): 58, 4 figs. incl. 2 col. (animals) — Cologne zoo; characters of adult and young; gestation; antler shedding; diet in captivity; hunting in wild. Pefaur, Hermosilla, Di Castri, Gonzalez and Salinas, 1968, Rev. Soc. Med. Vet. (Chile), 18(1-4): 4, 5, 6, 7, 8, 11, 14 — CHILE: Curico (?); Nuble; Concepdon; Arauco; Bio-Bio; Malleco; Cautin; Valdivia; Osomo; Llanquihue; Chiloe; Aysen; Magallanes; distribution; disease (hoof-and-mouth); en- dangered spedes. Hick, 1968, Fr. Kolner Zoo, 4(10): 111, 16 photographs, 2 in color (female and young) — gestation, birth and chronology of growth and behavior in Co- logne zoo. Hick, 1969, Int. Zoo Yearb., 9: 110— CHILE: Llanquihue (Lago Llanquihue) — mating; antler growth and shedding; molt; birth, growth, behavior of young in Cologne zoo. Kohls, 1969, J. Med. Entomol., 6(3): 280— CHILE: Valdivia (Santo Domingo, 10 km SE Valdivia; Callumapu, 15 km W. Valdivia); ticks (Acarina): Ixodes taglei, I. stilesi. Hick, 1979, Fr. Kolner Zoo, 4(10): 113, fig. (week old young)— birth in Cologne zoo. Hick, 1971, Z. Kolner Zoo, 14(2); fig. p. 83 (mother v^ath F2 young S) — breeding in Cologne zoo. Dansie and Wince, 1971, Deer of the world. Hitchings and Mason, Ltd., Plymouth, England, pp. 21, 23, fig. (animal), fig. (map, distribution) — CHILE. Wetterberg, 1972, Oryx 11(5): 347, fig. (animal) fig. (map, Parque Nadonal Vicente Perez Rosales) — CHILE: Llanquihue (Parque Nadonal Vicente Perez Rosales); ARGENTINA: Neuquen (Parque Nahuel HuapO; habitat; conservation; food; habits; pre- dators. Koulischer, Tyskens and Mortelmans, 1972, Acta Zool. Path. Antwerp, 56: 25, 28, fig. 2 (karyotype) — karyotype (2n=70; FM=74). Whitehead, 1972, Deer of the world. Viking Press, New York, pp. 29, 64, 65, 169, map 9 (distribution), figs. 24a, b (animal) — characters; distribution. Avila Pires, 1974, Publ. Biol. Inst. Invest. Cient. U.A.N.L., Mexico, 1: 61 — distribution; conservation. Spotorno and Fernandez-Donoso, 1975, Mamm. Chrom. Newsl., 16(1): 17, fig. 10 (karyotypes) — CHILE: Llanquihue (Llan- quihue); karyotype (2n=70, FM=74); weights. Fradrich, 1975, Zool. Gart., 45(1): 67, figs. 1, 2 (animal), fig. 3 (mother with fawn), fig. 4 (fawn), fig. 8 (skull) — characters; mainte- nance in Berlin zoo; antiers; breeding. Czernay, 1977, Zool. Gart., 47(3/4): 226, figs. 2, 3 (growth graph), fig. 4 (male), figs. 5, 6 (mating), fig. 7 (fawn), figs. 8, 9, 10, 11 (mother and young), figs. 12, 13 (family group), fig. 4 (antlers in velvet). Brambell and Mathews, 1977, Proc. Symp. 1, Assoc. Brit. Wild Animal Keepers, p. 25 — characters, reproduction, management in captivity. P[udu]. pudu, Wolffsohn, 1927, Rev. Universitaria, Univ. Catolica Chile, p. 240— CHILE: hunting, domestication. Schneider, 1946, Bol. Soc. Biol. Concepdon, 21: 78 — CHILE: Concepdon (Nahuelbuta; Cayumanque; Manquimavida); habits. Hoffstetter, 1952, Mem. Soc. Geol. France, 31(66): 369— CHILE. Pudu puda. Flower, 1929, List of vertebrated animals Zool. Soc. London 1828-1927, p. 325— listed. P[udu]. puda, Zuckerman, 1952, Proc. Zool. Soc. London, 122: 901 — births in London zoo, 23 May 1932 to 29 May 1933. P[uda]. pudua [sic], Carette, 1922, Rev. Mus. la Plata, 26: 445 — characters. Pudu pudua, Castellanos, 1940, Rev. Chilena Hist. Nat., 44: 159— ARGENTINA: southern Andes near Chilean border. Pudua pudu, Lydekker, 1898, Deer of all lands, p. 307 — characters; distribution. Matschie, 1899, Sitz.-Bericht Gesellsch. Naturf. Berlin, p. 130— distribution. Lydekker, 1901, Great and small game of Europe, western and northern Asia and America. Rowland Ward, London, p. 373 — characters. 62 FIELDIANA: ZOOLOGY Nanelaphus Pudu, Fitzinger, 1873, Sitzb. Akad. Wiss. Wien (1873), 68(1): 361— classification. Ovis dubia Afzelius, 1815, N. Act. Reg. Soc. Sd., Upsalla, 7: 217 — new name for Copra pudu Molina; work not seen. Cewus humilis Bennett, 1831, Proc. Zool. Soc. London, 1831: 27 — CHILE: precise locality unknown; type female, skin originally in the collection of the London Zoological Soci- ety, now in the British Museum (Natural History) no. 55.12.24.248. Gay and Gervais, 1846, Ann. Sci. Nat., Paris, ser. 3, 5: 90 — a synonym of Cen'us pudu [sic] Molina. Fraser, 1849, Zoologica typica, pi. 14 and text — CHILE: Concepcion; Chiloe. C[enms]. humilis, Sundevall, 1846, K. Svenska Vet. Akad. Handl., 1844: 183— part, CHILE: specimen seen in Paris museum. Cervus [{Capreolus)] humilis, Lesson, 1842, Nouv. Tabl. Reg. Anim., p. 172 — classification. Pudu humilis, Gray, 1852, Cat. Mamm. Brit. Mus., Ungulata, p. 240, pi. 36, fig. 1 (skull captioned Pudu chilensis) — CHILE: characters; part synonymy. Gray, 1872, Cat. rumi- nant Mamm. Brit. Mus., p. 93. Gray, 1873, Handlist ruminant animals Brit. Mus., p. 163— type history; CHILE. Jentink, 1887, Cat. Osteol. Mus. Pays-Bas, 9: 153— CHILE. Jentink, 1892, Cat. Syst. Mamm. Mus. Pays-Bas, 11: 183— CHILE. Pudua humilis, Garrod, 1877, Proc. Zool. Soc. London, 1877: 18 — classification. Sclater, 1877, List Vertebrated Animals Zool. Soc. London, p. 139, fig. 18 (animal) — male ex- hibited live. Brooke, 1879, Proc. Zool. Soc. London, 1878: 927— CHILE. Sdater, 1883, List Vertebrated Animals Zool. Soc. London, p. 174, fig. 28 (animal) — 2 males, 1 female exhibited live. Sdater, 18%, List Vertebrated Animals Zool. Soc. London, p. 185 — 1 male, 2 females exhibited live. De Winton, 18%, Proc. Zool. Soc. London, 18%: 507, figs. 1, 3 (skull); characters; comparisons. Kidd, 1900, Proc. Zool. Soc. London, 1900: 686— hair pattern. Reiche, 1903, Anal. Mus. Nac. Chile, p. 46— CHILE: Arauco (Isla de la Mocha, extirpated). Coassus humilis, Riitimeyer, 1881, Abh. Schweiz. Palaont. Gesellsch., 8: 31, 93 — cranial characters; relationships; distribution. Pudu chilensis Gray, 1852, Cat. Mamm. Brit. Mus., Ungulata, pi. 36, fig. 1 (skull)— CHILE: (no predse locality); name in caption for figured skull or same spedmen denominated Pudu humilis in text (p. 240). Gray, 1873, Hand-list of edentate, thick-skinned and ruminant mammals in the British Museum, p. 163 — adult female, skull nos. 972 and 50.11.29.6 (same figured 1852), skin of type skull received from Zoological Sodety of London. "Chevreuil," Poeppig, 1829, Ferussac, Bull. Sd. Nat. Geol., Paris, 19: 96— CHILE: high altitudes; size. "pudu," Poeppig, 1835, Reise in Chile, Peru and Amazonstr., 1: 450 — CHILE: Bio Bio (Anhjco). Ulmer, 1959, America's First Zoo, 11(1): 7— exhibited 1877 in Philadelphia Zoological Garden. T}/pe. — None in existence; name based on native accounts of the animaL Type locality. — Southern provinces of Qiile. Distribution (fig. 35) Forests of southern Chile and some adjacent parts of south w^estem Argentina; in Chile, from the province of Maule, southward nearly or quite to the Strait of Magellan; in Argentina, from southwestern Neuquen, southward along the An- dean foothills and mountain passes into southwestern Santa Cruz province; altitudinal range from sea level to nearly 1,000 m above. Conservation Pudus are still hunted in Chile, among other areas, as a cheap source of protein or as an expensive form of sport and their habitat is rapidly being de- stroyed, except possibly in the following national parks: NAHUELBUTA, Arauco-MaUeco, 1,000-1,500 m altitude, 5,414 ha (20 sq miles). Pudus Fig. 35. Distribution of Pudu puda Molina, based on museum spedmens and published 19th and 20th century records (dots). For locality names, see corresponding numbers in "Gazetteer" (p. 80). 63 64 FIELDIANA: ZOOLOGY have been recorded from the Cordillera de Nahuelbuta, but no information is at hand regarding the present status of the native mammals of this park. TOLHUECA, Malleco, 1,500-2,000 m, 3,000 ha (12 sq miles). Although this park is within the geographic range of and provides a suitable habitat for pudus, it hardly is large enough to support more than a small population. VICENTE PtREZ ROSALES, Llanquihue, 135,175 ha (522 sq miles). According to Chileafi authorities surveyed by Harroy (1971, p. 168), hunting is permitted in the park. How- ever, Wetterberg (1972, p. 348) reports pudus abundant and generally unmolested in the area. VILLARICA, Cautln, 13,780 ha (53.1 sq miles); ". . . inaccessible and has few inhabitants," according to Harroy (1971, p. 168), and incorporates good pudu habitat. TORRES DEL PAINE, Magallanes, 10,000—20,000 ha. Three additional parks are located within the geographic range of Pudu puda, namely: LAGO SAN RAFAEL, Aisen (590,000 ha); PUYEHU£, Valdivia (75,000 ha); and PIRIHUEICO, Valdivia (13,950 ha). These are not regarded as conservation areas, due to the high number of human habitations and excessive exploitation, such as unrestricted or uncontrolled hunting, fishing, lumbering, stock grazing, and rapid destruction of forest. The future of the pudu in Argentina is clouded, Dennler de la Tour (1957, p. 159) observes, "Among the arumals most in need of protection in Argentina are the Andean deer 'huemul,' Hippocamelus bisulcus, and the dwarf deer 'pudu/ Pudua pudu. These are threatened not only by 'development' and poaching, but by the introduction of the European red and fallow deer in the park areas." A decade later, Nottebohm (1967, p. 117) warned that disappearance of the huemul and pudu from Argentine territory is imminent. The following is a list of Argentine national parks, from north to south, where pudus (and huemuls) live now or lived in the past: LOS COPAHUES, northwestern Neuquen, 80,000 ha (310 sq nules). The pudu is not known to occur in northern Neuquen at present, if it ever did before. LAGUNA BLANCA, central Neuquen, 1,000-2,000 m, 11,250 ha (40 sq miles); doubtful if the pudu occurred in the area surrounding the lake. LANlN, southwestern Neuquen, 600-3,776 m, 395,000 ha (1,500 sq miles). This area adjoins Nahuel Huapi National Park and has the same fauna, including pudu and huemul. NAHUEL HUAPt southwestern Neuquen, 720-3,554 m, 785,000 ha (3,000 sq miles) of which only 60,000 ha (260 sq miles) is nature reserve, the remainder being private lumbering, farming, and grazing property. According to Dennler de la Tour (1957, p. 160), "The huemul and the pudu have been almost eradicated by poaching, by the dogs of sheepbreeders and by the introduction of European deer." LOS ALERCES and ANEXO PUELO, western Chubut, 400-2,280 m, 280,000 ha (1,100 sq miles). Although the pudu has not been recorded here, the park lies within its probable geographic range and has a suitable habitat. PERITO FRANCISCO P. MORENO, west central Santa Cruz, 800-2,770 m, 115,000 ha (400 sq miles). Dennler de la Tour (1957, p. 163) includes the huemul and guanaco in the local fauna, but does not mention the pudu, which may occur nearby if not inside the park. LOS GLAQERES, southwestern Santa Cruz, 300-2,441 m, 670,000 ha (2,600 sq miles). Dennler de la Tour (1957, p. 162) reports that the "huemul and pudu are present, though reduced in numbers." Taxonomy Molina (1782, p. 308-309) described the "pudu {Capra puda) as a kind of wild goat about the size of a six-month-old kid, brown in color, the male with small horns. The animal is called venado or roebuck by the Spaniards but improperly as HERSHKOVITZ: NEOTROPICAL DEER 65 it has all the characters of a goat except for lack of a beard and in that its horns are smooth, rounded and divergent." Gmelin (1788, p. 201) referred the pudu to the genus Ovis, but rendered the original specific name "pudu," a misspelling used by nearly all subsequent au- thors. Rafinesque (1817a, p. 437) thought the pudu was akin to goat antelopes when he referred it to his newly described genus Mazama, together with Mazatna ovina {Oreamnos americanus americanus Blainville). Later he (1817b, p. 44) included Mazama puda, Mazama dorsata (=0. americanus americanus), and Mazama sericea (=0. americanus americanus) in "a particular subgenus (or perhaps genus) which I shall call Oreamnos, distinguished by the horns slightly curved backwards or outwards, often rough or annulated, and long hair, besides living in mountains: if the horns should prove hollow, it will have much affinity with the genus Rupricapa [sic]." This judgment was made earlier by Blainville (1816, p. 76). The true nature of Capra puda Molina remained equivocal until 1846 when Gay & Gervais (p. 88) demonstrated the cervid characters of two pudu skins and one skull that Gay brought to the Paris museum. They also showed that Cervus humilis Bennett, described in 1831 on the basis of a female that had lived in the gardens of the London Zoological Society in 1830, represented the same species. The name Pudu chilensis Gray appears first as the caption for the figure of a skull in Gray's Catalogue of the Specimens of Mammalia in the Collection of the British Museum (1852b, pi. 36, fig. 1). This name is valid, although it almost certainly is a lapsus for, and definitely a synonym of, P. humilis Bennett, the only spedes Gray (1852b, p. 240) actually described in the text. The preceding synonymy includes a complete bibliographical history of Pudu puda Molina. Characters Size. — Average larger than Pudu mephistophiles, smaller than Mazama chunyi. Coloration and pelage. — General body color buffy agouti with middorsum usu- ally contrastingly colored reddish brown agouti; pelage stiff, adpressed, the hairs pithy; dorsal cover hairs about 30 mm long, guard hairs like cover hairs about 5 mm longer, basal half of both types grayish, of terminal half brownish (eumelanin) with subterminal pheomelanin band on back, flank, rostrum, cheek, and neck; crown, outer surface of ears, supraorbital stripe or drcumorbi- tal band, narial and mandibular patches ochraceous to rufous, terminal portions of hairs uniformly pheomelanin, inner surface of ears, chin, throat like preced- ing but paler, the uniformly reddish parts contrasting with surrounding buffy or reddish agouti areas, including rostrum, cheeks, nape, neck, shoulders, flanks; chest, belly, inner sides of upper parts of front and hind limbs reddish, lower parts of limbs entirely reddish, hairs unhanded. Fawns spotted, pelage less hispid than that of adults. Color variation. — The dominant pheomelanin ranges in coloration from buffy to reddish or reddish brown. Of four adult males from Chiloe taken in January, one is in new buffy pelage, another is in new reddish pelage, and two are in old reddish pelage. The remaining adults and unspotted subadults at hand with known dates of capture are in unworn pelage, all from Chile, as follows: February — buffy 6 , Chiloe. April — reddish 9, Puerto Montt; rufous 9, Temuco. 66 FIELDIANA: ZOOLOGY April — buffy 9, Cautin. July — buffy 6 , Angol. September — buffy 9, Angol. October — buffy 9, Valdivia. November — buffy 6 , Petrohue. One female (without collection date) from San Pedro, Concepdon, is in worn, faded, buffy pelage. The condition of the fur in this example, however, may be the result of exposure to sunlight and wear after the specimen was prepared as a round study skin. Two undated males, one from Arauco, are in unworn, buffy pelage. Worn pelage appears to be associated with the reddish color, but not certainly with the buffy. New pelage, however, may be buffy or reddish. There is no indication in present material that an animal may molt from one dominant color to the other. Molt. — Winter molt begins in fall, summer molt in spring. Starting with birth- date, fawns begin to shed the juvenal spotted coat during the sixth week and lose it completely by end of the second month (Vanoli 1%7, for Chilean pudu; Hick, 1%7/1968, and Fradrich, 1975, for German zoo pudus). Rhinarium. — Not notably swollen or bulbous, posterodorsal border slightly convex or forming a straight line between nostrils. Tegumentaiy glands. — Preorbital gland large, the opening wide, conspicuously displayed; interdigital glands of forefoot contained in deep trough or pocket of hind foot in well-defined pouch; tarsal and metatarsal glands not evident in available material. External genitalia. — According to Garrod (1877, p. 9, fig. 16), the glans penis is similar to that of Odocoileus and Blastoceros {Ozotoceros) . Skull. — Lacrimal fossa large, bowl-shaped and often fenestrated, the maxillary bone contributing to ventral border, malar bone excluded; anterodorsal border of fadal plate of lacrimal bone slightly to moderately convex or convoluted; superior border of ascending ramus of premaxillary bone not expanded, and either contacting or widely separated from lateral margin of nasal bone; long axis of antorbital vacuity diagonal to nearly horizontal, plane area of vacuity from approximately one-half to nearly as large as plane area of fadal plate of lacrimal; perforations present on upper surface of root of zygomatic process of squamosal bone, the diameter of posteriormost usually more than 1 mm; angle of mandible subdrcular, produced little beyond posterior plane of condyle; lunar notch be- tween condyle and angle little exdsed, concavity between base of angle and ramus more or less defined. Cranial measurements are given in Table 4. Dentition. — Upper dedduous canine, present in some fawns, shed without replacement after eruption of first molar; metaloph absent in fully erupted mo- lars, postentoconulid and corresponding crista rudimentary in unerupted, ab- sent in fully erupted molars. Antlers (table 5). — Growth completed and spikes polished during summer, shed in winter same day new growth begins. In high, southern hemisphere latitudes, mature antlers are rubbed dean of velvet in December and dropped in July (Vanoli, 1%7). Antlers of pudus introduced into northern European zoos maintain southern periodidty until cast. During the next one or two years, antler development and other vital processes adjust to reversed seasons. Antlers attain ^ Pudu puda \ Fig. 36. Southern pudus, females, in Cologne zoo (photograph courtesy Cologne zoo). Fig. 37. Southern pudus in Berlin zoo. Left, breeding male; center, male in velvet; rear, five-month-old male; right, female. [Photograph by J. Fradrich, courtesy Berlin zoo; cap- tion from H. Fradrich (1975, p. 69).] 67 .C O) Q t^ QO in 00 On vO ^ v^ 5 so 5 in c^ "55 r^OTi»£ N CD So; 60— o o in vo rf o 1 00 O rri '^^ ON ON Tt tN. tX3 r<^ tv.'i-> fO'^ininLnLn'«a<'^'^^ I 6C6O 2 X S ■a "53 « '5 Tt in I I 1-1 I in in in o o 00 c S c5 *" « c C in in ^fjoinm I |T-H(N|r-irN|fMrH| I I 3 .C ■3 c J c V c o NO rNj in o t% in t^ ON in ON CNi vo cni Tj> I NO t^ t-^ 00 1^ oor^ I I I o♦'oo*o'o^D*o'o*o^^•oo ^ 1^ -«' ^ OP 5 — j: 'H 'i" ■»< '^ 'Oi "4* •1' 'i rs uuS° — — — — — — — — ^ ^c^ J tu a; n «5 > m > PC B o e o c (5 c m < < — 0" m > ^ £?33D3C/53333D3333 53333 "^333333333 0000--000000000 CCCCCi-CCCCCCCCC 3 3 c >^ n N x> v> 2^ UcC 00 00 s o ■DTD a S8:S )0 U, vo in in C3N OS in in ON a\ "i (JO J3. -5 ■5 •6 •6 •c •c •c •c TS TJ TJ T3 :(T3 :ns :ns ■n XX 1-1 li li v^ •c -c -c -c -c -c -c -c -c c c :(T3 -fZ 'ft -fZ "(^ :*^ 'fZ '-(Z •-i fc, ,N IN f^ C C §.■=00 nS §§ 3 c •5 5) 1 1 1 ift ^ 1 "<3 s a. 3 s 5-= a. mS c _1 ^-' i2 S § 1 in ^ D'2 TO I— I 3 . C > c o <2 Ob C C t>0 60 o o UU .5 .5 .£ .S .B .B .B .£ .S ti t; ojaiaiaiaiaiotaiaiiMt-. CQCQCQCQCQQQCQCQCQUJUJ I I I I I I I I 00 I I r4 t^ 0^ p r \0 CT^ rH fN rO I— 1 ON _ QQ ON ON ON ON On >,r "6 (3 oj a! H On O^ TO^'-'CCCCai ^^ IoSDD ^ • TO TO TO TO ^ rH < ^ \o 00 vO '-HjS.iS.in'^tNooin (N ' ' rH rH rH (sj CN o9P to ^ 06 ^ o^^C:|inino6fnro rHtN,(NrHrHr-irHfs (N t^ r'i rH rH . 1 5 C 01 01 C TO ON TO , 00 "^ rH ^^^^-^ . P 90 t;— — KCJ 1 OJ 1 rH rH tN in in ON rH cnJ 2 "O 1 1 33 ? tv rH rH S "C 00 oc oc * 2s ^^ rH U U S n >^ 00 Q.5 3 ON Tt UU c > ■5b c b ns u 3 ^ .2 = ON £ -^ rH ■" 01 >^ C J= TO S ; >^ O' C 1 5 ° £ ^ Sl 2^ 69 70 RELDIANA: ZOOl-OGY polished maturity from end of May to mid-July, are shed between mid- December and end of January (table 5). Yearling antlers are little more than buds. Second-year spikes grow to about 30 mm. Recorded lengths of third-to eighth year antlers range from 63 mm to 90 mm (Vanoli 1%7, p. 162). Vertebral formula. — In a male (FMNH 24363) from Quellon, Isla de Chiloe: 7 cervical, 12 thoracic, 7 lumbar, 5 sacral, 8 caudal. In two specimens (MCZ 5022; 6029) from Chile, precise locality unknown: 7, 7; 13, 13; 6, 7; 4, 4; 8, 5(-l-?). Karyotypes The chromosomes of a male pudu studied by Koulischer et al (1972) have a diploid number of 70, with fundamental number of 74. Sixty-six autosomes are acrocentric or telocentric and one pair is metacentric. The X chromosome, a metacentric, is the largest of the complement; the Y chromosome, also metacen- tric, is the smallest. A few satellites are seen attached to the largest autosome pair. Spotomo & Fernandez-Donoso (1975) found virtually identical karyotypes in a male and female pudu captured alive in Llanquihue. Satellites attached to the largest autosome pair are evident in the illustration of the karyotype, and the metacentric X chromosomes are by far the largest of all. The pudu karyotype is grossly similar to that of Odocoileus. The only other neotropical deer for which karyotypic data are available is the red brocket, Mazama americana, with formulae, as dted by Taylor et al (1969, p. 349), "2n=68 and NF=74 (Taylor et al [Taylor, Hungerford, and Snyder, unpublished] 1968)." According to the authors dted, an extra pair of metacentric autosomes in the brocket accounts for the lower diploid number of chromosornes. It is curious, however, that Benirschke (1977, p. 52, pi. 9) found 50 chromosomes in father and son and 49 in the mother of a family group of Mazama americana. Comparisons All distinctive characters of Pudu puda can be derived from an ancestral form, such as P. mephistophiles. Average size is larger, general body form less squat, more brocketlike, wdth slightly longer muzzle, less-rounded, more elongate ears, and larger eyes; tegumentary glands, most notably lacrimal and pedal, more developed; general body coloration of adult less agouti, the monochromal hairs pheomelanin; fawn with characteristic cervid spotted pattern; cranial parts more advanced, the lacrimal pit large, more deeply excavated, antorbital vacuity larger, its long axis more nearly horizontal; permanent upppr canines almost entirely lost. The same relationship holds for all other spedfic characters de- scribed elsewhere. In virtually every case the differences are quantitative. It remains to be tested whether or not complete reproductive isolation between the spedes has been achieved. Specimens Examined Total 31. CHILE. Concepcion: San Pedro, 1 (FMNH); Arauco: Santa Juana, 1 (FMNH); Sierra Nahuelbuta, 1 (FMNH); Isla Mocha, 1 (AMNH); Malleco: Angol, 3 (AMNH); Cautin: Maquehua, 1 (AMNH); Pitrufquen, 2 (FMNH); Tolten, 2 (FMNH); Valdivia: Valdivia, 1 (FMNH); Llanquihue: Petrohue, 1 (FMNH); Puerto Montt, 1 (AMNH); Chiloe: Isla Chiloe, 8(5 FMNH; 1 USNM; 2 MCZ); Lago Cucao, 1 (AMNH); Quellon, 6 (FMNH); Rio Inio, mouth, 1 (FMNH). HERSHKOVITZ: NEOTROPICAL DEER 71 VII. BIOLCX5Y OF Pudu puda Habitat The Chilean habitat of Pudu puda is the humid forest or Valdivian biotic dis- trict, which lies south of the Santiagan. As described by Osgood (1943, p. 30), the Valdivian . . . extends along the coast and through the mountains to the vicinity of the Gulf of Penas, about lat. 47°S. There is some interdigitation at the northern boundary where humidity increases with altitude, but the change is fairly abrupt from a region [the Santiagan] of moderate rainfall and relatively light vegetation to one of great humidity and dense forest. [Footnote: ". . .rainfall at Valparaiso 20-40 inches; at Valdivia, 80-200 inches."] In the heart of this district, vegetation is widely varied and highly exuberant. Heavy forest growth is combined with dense undergrowth and the numerous cryptograms that go wdth an atmosphere constantly charged with moisture. According to Wetterberg (1972, p. 348): . . . The most favorable pudu habitat in the [Vicente Perez Rosales National] Park is . . . classified as the Valdivian Forest Type, and the Coigue [Coihue] Forest Type. In the lower altitudes, pudu are often found in the understory thickets consisting of small trees of tiaca {Caldcluvia paniculata) , avellano {Guevina avellano), luma (Amonyrtus luma), arrayan (Myceugenella apiculata) and huedhued {Pernettya furiens). I never saw tracks at elevations above 1200 m. nor have sightings been reported by others. At these higher ranges, the understory is often quila {Chusquea quilla) and colihue (C. colue). Greer (1965, p. 146) notes that pudus have been seen regularly in the Laguna Malleco area, w^hich ". . . is heavily forested with coihue (Nothofagus dombeyi), rauli (N. alpina) and roble (N. obliqua) along with dense patches of cane [Chusquea sp.]." The habitat in adjacent parts of Argentina is similar to the Chilean. Maintenance in Captivity In sharp contrast to its northern congener, the southern pudu adjusts readily to captivity, whether as household pet in its native lands, or as resident in zoological gardens in Argentina, Chile, Belgium, England, Germany, Holland, and the United States. Doubtiess, it can survive in any temperate zone climate where deer are kept. The female sent to the London zoo in 1830, which served as type of Cervus humilis Bennett (=Pudu puda), may have been the first pudu shown in a European zoo. Since then, according to the International Zoo Yearbook, published since 1960 by the London Zoological Society, the animal has been kept and successfully bred in the Berlin, Cologne, Erfurt, Hellenthal, Wuppertal, Rotterdam, and London zoos. Birth records, for which dates are available, are listed in Table 6. A pudu breeding farm, that began with a pair captured in the wild, was established in 1938 in Neuquen, Argentina, by Otto Neumeyer (in Schmidt, 1944, p. 239). A total of eight individuals, including third-generation young, was reported in 1944. In 1976, according to the International Zoo Yearbook for 1978, only 12 animals were living in zoos, as follows: Antwerp 5, West Berlin 1, Cologne 1, Erfurt 1, Wuppertal 3, and Chilian (Chile) 1. The official studbook keeper, as of 1978, is J. P. Guillermo Staudt, Director, Wildfreigehege Hellenthal, 5374 Hellenthal-Eifel, West Germany, acting under supervision of Carlos Junge, Director, Jardin Zoologico de Chilian Viejo, Chilian, Chile. If s T3T» V •§•§ a •S-5 ^ S" 3 o ^ .£ CRRS S yer yer (196 (196 9671 <" « .- ._ 1-1 c c "o "o ^ ssg^:^ =2 ZZ>>I u. e o 1 fejqs i "O 1 1 r4 . bd «N - u j; 2 <-< ^ S 5 3 3 i2 < N N CO o o o On ON On >^ >. >, f^ fZ tz C C C 01 01 OJ N N N UUU CnI I I I I I I I S2 a rg I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I 'OCHCHO'00*OOt»OCHOO»0'Oo O 0*0*OCH >^ 00 I o a. a J c On « 2? T 1-H f^ ^ 2^^:^So^'^<^E;;<^Sl^OCTNt;;ONONLr f^^Ln ON fT\ tv So . r-i > . > > • ^, '"' >S >N >N ., >^'C 7^ >N >nV, >N >n?J >^^ * O L o c > c ri /^ \C 2 "5 c 2 2 c<- 2 = § =^SS gS^ s ■:, 3 < ►i^ I I I rH (N I I I I I I I I i I I I I I I I I I CC *. *_ oia;ooioioir-r-i- ■O -O ^-s-— ^^-i! .— -CCCCCCCCCCCCCCSKS a* 01 ^ ^iis 0*:= 0;= o_o o o o o_o o_o o_o o_o o^^^ 3 2r 2rx Icxp.'^n'^n'o o'o'o'o'o ©"o o"© ©"o 0*0 o o ot: 72 HERSHKOVITZ: NEOTROPICAL DEER 73 Ungulate Associations The huemul, Hippocamelus bisulats, is the only other native cervid in Chile, and the only one that might occur within the Argentine portion of the range of Pudu puda. Historically, the range of the grassland vicugna {Vicugna vicugna) and guanaco {Lama guanicoe) may have contacted that of the pudu in Chile and Argentina, but probably without sympatry. Numbers and Aggregations The Abbe Molina (1782, p. 308) notes that, with onset of winter in the moun- tains, pudus descended in flocks ("a truppe") to the warm coastal plains of the southern provinces. Gay (1847, p. 159) says they lived in small flocks in the central valley between the Cordilleras. However, accounts of modern authors indicate that the pudu, an inhabitant of forest thickets, is a sedentary and soli- tary animal except during the fall rut when it pairs and in spring when the single young accompanies its mother. Housse (1953, ex Wetterberg, 1972, p. 349) al- lows that pudus may have occurred in small herds before the last half of the 19th century, when the animals abounded, but "now that they have decreased, these groups of two or three are not a herd in formation but rather families not yet dispersed." Natives of the Nahuelbuta region informed Greer (1965, p. 146) that "pudus do not change places of habitation and are in the area in equal numbers in all seasons." Osgood (1943, p. 230) spoke of the animal living in very dense forest, and of only once catching a glimpse of one. Those collected on his two Chilean expedi- tions evidently were solitary animals driven from cover by dogs. Schneider (1946, p. 78) encountered the pudu singly or in pairs in the heavily forested parts of the Cordillera de Nahuelbuta, the Cayumanque, and Manquimavida. The semidomesticated pudus that Vanoli (1967) observed in gardens or small pre- serves were loners, pairs, or family groups consisting of parents with their young of the season. The largest known concentration of pudus lives on the privately owned Mar- garita Island, Lake Todos Los Santos, Llanquihue, within the Vicente Perez Rosales National Park. At this island, Wetterberg (1972, p. 348) reported "a herd of 10-15 with no natural predators or dogs to disturb them, and no interference apart from an occasional stick thrown at them when they wander too close to the vegetable garden. Some of these pudus were brought to the island by colonists on the mainland whose dogs had chased them into lakes or streams where they might have drowned." Wetterberg himself saw only two captive pudus in the town of Peulla, and a pair "casually walking down the road near the hotel at La Ensenada in the early evening." Migrations It has been said (Molina, 1782, p. 309) that herds of pudus come down from the Cordillera at the beginning of winter to feed on the plains of the southern provinces of Chile. Yepes (1943, p. 75) states that the deer entered Argentina from Chile through numerous passes in the humid, forested zones of the Andes. He records pudus from elevations as high as 900 m (Lago Lolog) above sea level, but most of the mountain passes to which Yepes alludes are much higher. The pudu does not descend the Argentine slope of the Andes to the Patagonian scrublands. 74 FIELDIANA: ZOOLOGY Occurrence of seasonal migrations has not been verified. According to Osgood (1943, p. 229), "In early accounts and many later ones based on them, [the pudu] is usually spoken of as an animal of the cordillera, but this appears not to be the case since it is actually recorded only from heavily forested regions either near sea level or at very moderate elevations." Present information indicates that the pudu is a sedentary, cryptic animal, its movements localized and likely limited to the narrowly circumscribed territory of the individual or family group, or to the quest of new territory by the matured young of the year. Locomotion, Play Information on pudu locomotion consists of a few observations of wild or captive animals in flight or play. Vanoli (1967, p. 155) compares the low, crouching, zigzagging flight of the short-legged pudu with movements of a young boar. Playful, developing young bound and gambol like kids (Hick, 1967/1968). Adults often engage in playful flight in early morning or evening, running and doubling until they pant with tongue hanging out (Vanoli, 1967). Junge (1966, p. 264) describes the female of the pair in the Chilian (Chile) zoo as "sometimes very playful toward evening and early in the morning. She runs about the enclosure like a puppy, zig-zagging, skipping and jumping on the spot till she is so exhausted that the foam shows on her muzzle. Sometimes the more sedate male joins in these games for a moment but he soon gives up or gives his high-spirited mate a rather firm dig in the ribs with his horns. The female then quickly ceases her playfulness and retires to the hut." Enemies Principal natural enemies of Pudu puda include the native foxes (Dusicyon spp.), puma {Felis concolor), spotted cats (F. guigna, F. colocolo), and possibly the eagle owl (Bubo). The introduced enemies — man and domestic dogs — have ex- terminated the pudu throughout most of its range and greatly reduced its num- bers in the remainder. Diseases and Parasites Hoof-and-mouth disease among free-living pudus was reported by Vanoli (1967, p. 162). He also mentioned infestation with ticks. Two kinds of parasites have since been identified as Ixodes taglei and /. stilesi by Kohls (1969, p. 280). Food and Water Free-living pudus in Chile, according to Vanoli (1%7, pp. 156-157) and others, feed on many kinds of native and introduced trees, shrubs, and vines. They browse mainly on foliage, but also eat bark, twigs, buds, blossoms, fruits, berries, acorns and other nuts, vegetables, and ferns. Branches of the Rosaceae, including roses, raspberries, and blackberries, are preferred above all (Junge, 1966, p. 264). Grass is seldom eaten, but plantain {Plantagd), clover, other weeds, and cereals are. Salt is taken eagerly. Pudus thrive on the regular vegetarian zoo fare, including commercial food pellets (Czernay, 1977; Hick, 1967/68). Caged young also eat insects (Hick, 1969, p. 115). Perhaps many kinds of invertebrates are consumed in the wild by pudus of all ages. Vanoli (1967) observes that. HERSHKOVITZ: NEOTROPICAL DEER 75 among the pudus in reserves and gardens of the Osorno-Llanquihue district, females are fastidious eaters, nibbling a leaf here and another there, seldom eating large quantities of any one kind, except roses, which they especially appreciate. In contrast, males, particularly old bucks, satisfy themselves with basic foods. The free-living, semi domesticated pudus that Vanoli (1967, p. 162) observed in Qiile drank water in summer and winter, even when available food was not dry. Earthquake Sensitivity The semidomesticated, free-living pudus of the Osorno-Llanquihue region observed by Vanoli (1967, p. 159) panicked with the onset pf the 1960 earthquake in southern Chile. Their lacrimal glands were everted as they fled from one hiding place to another. The subsequent eruption of volcanic ash caused them to cower in their retreats for long periods, with infrequent sallies for foraging. Unusual behavior of animals prior to earthquakes has attracted man's atten- tion through the ages. For detailed discussions of the subject, see Evernden (1976). Temperature Regulation Pudus normally live in cool, shaded forests. With regard to those in the Con- cepcion (Chile) zoo, Junge (1932, p. 241) noted, "Should the animal be exposed to the full glare of the sun for 1 or 2 hours it would die the evening of the same day." According to Junge (1966, pp. 263-264), pudus at the Chilian Viejo (Chile) zoo breathed with their mouths open when the temperature rose to 32° C (= 90° F). "Cooling the enclosure by spraying," he writes, "was not a success as the animals were more troubled by a damp heat than by ^ dry one. On very hot days they withdrew completely into the darkness of their huts to rest or ruminate." In winter, however, as shown by Vanoli (1967, p. 162, fig. 10), free-living pudus emerge into clearings to bask in the warmth of the high sun. Reproduction, Growth, Development (Tables 7-9) The female is sexually mature when about six months old (Vanoli, 1967, p. 160; Hick, 1967/68, p. 111). Males may become sexually mature at about the same age, according to Hick (1967/68, p. Ill; 1%9, p. 110), who has the word of a Chilean resident that a six-month-old female pudu "was covered by a male of the same age" and gave birth about seven months later. The successful mating of a nine-month-old Erfurt zoo male was reported by Czernay (1977, p. 230). Al- though Vanoli (1967, p. 157) mentions a fertile mating by a one-year-old male, he (1967, p. 163) sets the age of male sexual maturity at \y^ years. The rut occurs in the fall. Competition or battles between males for possession of the female is unknown. Vanoli (1967) summarizes the mating behavior of a pair of free-living, semidomesticated Chilean animals as follows. The buck is drawn by pheromones to the estrous female living within or near his home range. On meeting, they engage in bouts of butting, the blows usually directed to flanks and groin. This courtship intensifies to a point at which both animals hurl themselves against each other. The buffeting is followed by mutual groom- ing, with the female carefully licking the male from top to bottom and the buck reciprocating less diligently and thoroughly. The rut lasts several days, and. Table 7. Growth and development of a male Pudu puda in the Cologne zoo, Germany.* Day Remarks 1 Bom 13 November 1%7 before 8:00 A.M.; bleats; mother eats fetal membranes; fawn rises, walks to distant corner of cage, lies down; suckles standing up whilst being groomed by mother; weight "about 1000 g" shoulder height "about 25 cm" ( = 15 cm?). [Weight at birth 370 g, shoulder height 15 cm "after a few hours ran about and hid in a tree in the bush grass," (V). Born London Zoo, 1975, in usual sleeping area; father present; mother active but returns to feed young, licking fawn's anal region while suckling; first suckling attempt at about lOhr(B).] 2 Changes resting place repeatedly; bleats to call mother; utters alarm cry when frightened. [Definite suckling at 27 hr; suckles for about 5 min at 2-4 hr inter- vals; gains about 80 g daily (B).] 3 Rests hidden under straw; gambols; nibbles on lettuce and straw but does not swallow; suckles at intervals during day; preorbital glands everted when frightened; goes to mother for nursing. [Fed only in early morning, noon, and evening (V). Runs playfully (B).) 4 Suckles regularly at 10:00, 12:00, 16:00, and 18:00 hours; eats dirt; cowers when frightened. [Ingests dirt and gravel (V).] 5 Enjoys standing in water tank. 8 Tests food eaten by mother but spits it out. 9 [Tests solid food (B).] 12 Eats leaves. 14 Eats lettuce; stands in water trough for longer periods. [Height, 20 cm; eats tender leaves; runs along with adults (V).] 15 Straw covering cage floor replaced by sand and cinders; prefers to sit on cinders; rises on cage bars. 16 Cage movements follow regular pattern; changes resting place. 17 No longer hides; rests in center of cage. 19 Lowers head and butts mother. 22 Butts mother and observer (Hick); eats bananas and commercial food pellets for first time. 24 Following fright attempts to climb onto mother's back. 26 Takes to cover more readily when frightened; groomed less often by mother; alternates running with butting. 28 Eats hay; takes grasshopper from observer's hand; fawn and mother often rest near each other. 30 [Last seen suckling (B).] 33 Leaps over mother's back in play. 37 Licks salt stone. 38 Caught and ate cockroaches. 39 Mother goes to resting young for nursing. 40 Young nearly resembles mother in size, spotted coat disappearing; both use same place for defecation and urination. 52 Spot pattern virtually gone. 60 Weaned; rejected by mother when attempts to suckle. [Nearly as large as mother; shoulder height about 30 cm, spotted pelage gone (V). Faunal stripes disappear (B).] 62 Eats adult fare. 65 Accustomed to solid food; no longer attempts to suckle. 90 [Faunal spots gone (B).] 180 6 young sexually mature. [9 young sexually mature (V).] 270 [First antlers appear in S (V), 6 sexually mature (Czemay, 1977, p. 230).] * Source: Hick (l%7/68; 1%9), with interpolations in brackets from Vanoli, 1%7 (V), free living, Chile; Brambell & Mathews, 1977 (B), London zoo; and Czemay, 1977, Erfurt Zoo. 76 Table 8. Chronology of weight increment, actual and proportional, in a male and female Pudu pmda born in the Thiiringer zoo, Erfurt, Germany, from day 1 to day 98 for both, to sexual maturity (182 days) and beyond (205 days) for female only. 9* Daily average cJ** Daily average Day (kg) increment (kg) increment 1 0.78 11.5% - - 2 0.87 - .70 - 2-3 - 9.2% - 14.3% 4 1.03 - .90 - 4.5 - 3.4% - 7.8% 6 1.10 - 1.04 - 6-7 - 8.2% - 8.2% 8 1.28 - 1.21 - 8-11 - 2.7% - 5.4% 12 1.42 - 1.47 - 12-15 - 3.9% - 3.7% 16 1.64 - 1.69 - 16-17 - 3.4% - 2.1% 18 1.75 - 1.76 - 18-20 _ 3.8% - 3.3% 21 1.95 - 2.04 - 21-24 - 2.1% - 2.3% 25 2.11 - 2.23 - 25-29 - 2.7% - 2.1% 30 2.40 - 2.46 - 30-33 - 1.0% - 1.3% 34 2.50 - 2.69 - 34-37 - 1.6% - 0.4% 38 2.66 - 2.73 - 38-42 - 1.4% - 1.5% 43 2.85 - 2.94 - 43-47 - 1.8% - 1.6% 48 3.10 _ 3.17 - 48-55 - 0.8% - 1.1% 56 3.30 - 3.44 - 56-62 - 1.7% - 1.0% 63 3.70 - 3.67 - 67 3.85 - 3.80 - 63-70 - 0.7% - 0.8% 71 3.90 - 3.90 - 71-80 - 0.4% - 0.3% 81 4.05 - 4.0 ' - 81-98 - 0.6% - 0.6% 99+ 4.60 - 4.40 - 112 4.80 - ■- - 118 5.03 - - - 146 5.45 - - - 160 5.77 - - - 170 6.05 - - - 182 6.25 - - - 195 6.20 - - - 205 6.25 - - - Source: Czernay (1977, pp. 238, 239). * Born 2 May 1975 in Erfurt zoo. ** Bom 3 September 1974 in Erfurt zoo. + After 99 days, male resisted too vigorously for accurate weighing. 77 78 FIELDIANA: ZOOLCXiY Table 9. Three oldest longevity records for Pudu puda. irliest recorded Last recorded Zoo date Age date Interval locality Reference 10 July 1909 Adult 9 1 Jan. 1917 7 yr, 6 mo Berlin Fradrich (1975:71) 9 Nov. 1966 Adult 6 10 June 1974 (living) 8 yr, 7 mo Berlin Fradrich (1975:71) 14 Feb. 1866 Newborn 17 Feb. 1875 9 yr, 5 days London Jones, in Crandall (1964:595)» * The information was contained in an unpublished mimeographed manuscript dated 1958. 1 find no reference to this pudu in publications of the Zoological Society of London. during a three-day peak the buck mounts the doe repeatedly. He thrusts with- out panting or other signs of effort, but the female reacts to each thrust with a bleat. During the rut of the following year, Vanoli reports, the male served the same female, as well as a six-month-old, sexually mature virgin. No courtship pre- ceded his repeated mountings of the younger animal. On the approach of the buck, she arched her back, then stooped and pressed against the ground in response to the male's probings. She was silent throughout the act. However, the older doe responded wdth bleats to the thrusting of the male, as she had the year before. Newborn pudus are dropped in the spring of northern or southern hemi- spheres after a gestation of about seven months (202 to 223 days). Neumeyer (in Schmidt, 1944) reports six births in Neuquen, Argentina, from November through January, four occurring in November alone. Two births in the Osomo- Llanquihue region recorded by Vanoli (1967) occurred in November and De- cember. A late January birth in an unspecified Chilean locality was mentioned by Hick (1969, p. 110), and two pregnant females that arrived in the Berlin zoo produced in November of the same year according to Fradrich (1975). Eighteen other parturitions occurred in western European zoos during the following months; the details are given in Table 6. Jan. Feb. Mar. Apr. May June July Aug. Sept. Oct. Nov. Dec. 1 1 11 1 1 1 1 1 The 13 (72%) spring (April, May, June) and summer (July, August, September) births were produced by captive does that had borne young the preceding year in the same zoos. A wild, captured female that arrived pregnant in the Cologne zoo was responsible for the single November birth. The history of the lone January birth, also in the Cologne zoo, is not given. Fifteen births in the Berlin, Erfurt, Cologne, and Wuppertal zoos from 1967 to the end of 1973, compiled by Fradrich (1975, p. 70), are distributed by months as follows, with the January birth in the Cologne zoo excluded: April May June July August 14 6 3 1 Sex of 22 young born in the same four German zoos is 11 males and 11 females. Newborn listed in Table 6 include 10 males, eight females, and five with no sex recorded. Twins have not been recorded. Composition of milk is shown in Table 10. HERSHKOVITZ: NEOTROPICAL DEER 79 Table 10. Composition of milk produced on 32nd lactation day (4 June 1975) by wild- caught pudu {Pudu puda) maintained in Erfurt zoo since August 1971, compared with other ungulates and Homo. -1.5 Component Pudu Mule deer Roebuck Goat Cow Hui a. Fat 3-4 10.5 6.7 4.07 2.8-3.5 3.5 b. Albumin 2.03 9.56 8.8 3.76 3.5 1.2- c. Lactose 7-8 3.91 3.9 4.44 4.7 7 d. Potassium 81 - - - 150 53 e. Calcium 37 - - - 130 31 /. Phosphorous 46 - - - 97 15 g. Sodium 110 - - - 53 15 h. Chlorine 22 - - - 100 36 Values: a — c = g/100 ml; d — h = mg/100 ml after electrolysis. Source: Czernay (1977, p. 235). Longevity The three oldest longevity records for zoo-kept Pudu puda are listed in Table 9. Age at accession of each of the first two animals listed probably is more than one year. VIII. SUMMARY The South American deer of the genus Pudu, comprised of two of the smallest living species, is characterized and compared with other artiodactyls, particu- larly with respect to limb proportions and metapodial morphology. A hypotheti- cal reconstruction of pudu history is presented, and the genus is arranged sys- tematically as representative of tribe Pudini, subfamily Odocoileinae, family Cervidae. The northern Andean species, P. mephistophiles de Winton, is shown to be primitive, and the southern Chilean-Argentine species, P. puda Molina, derived. Both species are described, characterized, and compared: their geo- graphic ranges are plotted, and their life history notes are summarized. The status of wdldlife conservation of the pudu also is discussed. IX. ACKNOWLEDGMENTS Thanks are expressed to the authorities of the institutions listed in the "Ab- breviations" section for permission to study the material in their charge. Most of the pen drawings were prepared by Marion Pahl, former Field Museum staff artist; others were contributed by Zorica Dabich of the present staff. Most photographs of skulls were taken by Curtis Bean, a work-study student, later a volunteer, the remainder by staff photographer Ron Testa, who also processed the prints. Sarah Derr Bruner typed the manuscript, and technical assistant Barbara Brown helped with research and preparation of the manuscript for publication. I am indebted to the following persons and their institutions for providing photographs of living pudus and brockets used in the comparisons and descrip- tions of external characters: Dr. Don D. Farst, Director, and Violet K. Springman, Information Coor- dinator, Gladys Porter Zoo, Brownsville, Texas. Dr. Hans Fradrich, Assistant Director, Zoologischer Garten, Berlin. Uta Hick, Scientific Assistant, Zoologischer Garten, Koln (Cologne). Dr. Clyde A. Hill, San Diego Zoological Society. 80 FIELDIANA: ZOOLOGY X. GAZETTEER AND EXPLANATION OF MAPS (figs. 30, 35) Localities mentioned in the text are listed from north to south by country, province, or department, and by nearest town, subdivision, or topographic fea- ture where pudus were captured, observed, or reported as probably present. Documentation for specimens I examined is included with their respective localities. Documentation for other recorded localities is contained in the litera- ture dted in text, including the annotated synonymies for each of the two species. The numbers identify the localities plotted in Figures 30 and 35. COLOMBIA— PHd« mephistophiles Valle (Valle del Cauca) 1. Barragan,4'X)l'N, 75°52'W. 2. Las Hermosas (Paramo), 3°48'N, 75°56'W. 2. Las Hermosas. Cauca 3. Moras (Paramo), 2°50'N, 76°12'W. 4. Delicias (Paramo), 2°38'N, 76°14'W. 4. Malvasa, 2°34'N, 76°04'W; 3,000-3,500 m. K. von Sneidern, November-December 1956; January 1957, September 1957, Oc- tober 1957, December 1957; January-February 1958, 15 (FMNH). 5. Munchique (Cerro), 2°32'N, 76°57'W. 6. Pusna. See Pusha. 6. Pusha (Cerro), Totoro, 2°30'N, 76°24'W; 2,800 m. 6. Guanacas (Paramo), 2°27'N, 76°15'W. 6. Quintana, 2°24'N, 76°25'W. K. von Sneidern, November 1945, 1 (FMNH). 6. San Rafael (Paramo), 3,400 m. Adjacent to Paramo de Purace (q.v.). 6. Purace (Paramo), 2°21'N, 76°23'W; 2,800 m, 3,480 m, 3,500 m. F. Lehmann, July 1945, 1 (USNM), February, May 1953, 2 (AMNH); 1 (FMNH). 6. Coconuco, Paletara, 2°16'N, 76°22'W; 3,200 m. H. Van Deusen, 1 (AMNH). 7. Sotara (Paramo), 2°12'N,76°31'W. 7. Paletara, 2°10'N, 76°26'W. H. Van Deusen, 2 (AMNH). 8. La Vega, 2°05'N, 76°50'W; 3,500 m. K. von Sneidern, November 1956, 1 (FMNH). 9. Las Papas (Paramo), 1°55'N, 76°36'W; 4,000 m. 9. Cusiyaco (Laguna), 1°55'N, 76°36'W. 9. Lebrero (Paramo), NE Paramo de las Papas (q.v.). ' 9. Valle de las Papas, 1°52'N, 76°40'W. Tolima 10. Los Nevados. 11. Roncesvalles, 4°01'N, 75^2'W. Huila 12. Nevado del Huila. 13. San Antonio, upper Rio Magdalena, 1°57'N, 76°29'W; 2,700-4,000 m. P. Hershkovitz, August-September 1951 (observation). 14. Cueva de Los Guacharos, ca. 1°35'N, 76°00'W. Narino 15. Azufral (Paramo), 1°05'N, 77°43'W. 16. Cumbal (Paramo), 0°57'N, 77°52'W. 17. Chiles (Paramo), 0°48'N, 77°57'W. ECUADOR— Pudu mephistophiles Carchi 18. Atal, Hacienda El Vinculo, 0°35'N, 77°44'W. Cordillera Pimanpiro, W slope Cor- HERSHKOVITZ: NEOTROPICAL DEER 81 dillera Oriental. P. Hershkovitz, October 1934 (report). Pichincha 19. Cayambe (Mt.), 0°2'N, 77°59'W. L. Soderstrom, May 1898, 1 (BM). 20. Deleted. 21. Papallacta, 0°22'S, 78°08'W. L. Soderstrom 1896, 1 (BM), May 1897, 1 (BM). 21. Papallacta (Paramo), 3,600-3,700 m. See Papallacta. 22. Antisana (Mt.), 0°30'S, 78°08'W. Cotopaxi 23. Parque Nacional de Cotopaxi. See Salcedo. 23. Salcedo, W of, 1'^2'S, 78°34'W; 3,500 m. Tungurahua 24. Llanganates (Cordillera), 1°13'S, 78°15'W. Andes-Amazon Expedition 1935-1936, 1 (ANSP). PERU— Pudu mephistophiles Hudnuco 25. Quivilla road (Quivilla=9°32'S, 76°41'W); 3,300 m. 40 km NW Huanuco (q.v.). 26. Carpish (summit), 9°42'S, 76WW; 2,000-3,000 m. On Huanuco- Tingo Maria Road. 26. Zapatogocha (Bosque), Acomayo, 9°43'S, 76°04'W; 3,100 m. P. Hocking, November 1972, 1 (FMNH). 26. Chaglla, 9°50'S, 75°54'W. 27. Yanacocha, Rio Quero, 9°48'S, 76°25'W. 27. Huanuco, above, 9°55'S, 76°14'W. 27. Pamacucho, 4 km from Huanuco (Grimwood, 1969, p. 79). 27-28. Huanuco and Ambo, between, 9°55'S, 76°14'W— 10°07'S, 76°10'W; 2,700- 3,300 m. 28. Ambo, lOWS, 76°10'W. See Huanuco and Ambo, between. Junin 29. Paucartambo (Rio), 10°57'S, 75°16'W. 29. San Ramon, above, 11°08'S, 75°20'W; 2,000-3,000 m. 30. Huasahuasi (Rio), 11°15'S, 75°37'W. CHILE— Pwdw puda Maule 1. Constitucion, 35°20'S, 72°25'W; 1 m. 2. Cauquenes, 35°58'S, 72°21'W; 143 m. Concqjcion — Pedro Valdivia, not located 3. Concepcion, 36°50'S, 73°03'W; 13 m. 4. San Pedro, 36°51'S, 73°02'W; 17 m. CarlosS. Reed, 1 (FMNH). 5. Manquimavida, 36°57'S, 73°01'W. 5. Cayumanque, 36°40'S, 72°33'W. Bio Bio 6. Antuco, 37°20'S, 71°41'W; 602 m. Arauco 7. Santa Juana, 37°11'S, 72°52'W; 75 m. W. H. Osgood, 1 (FMNH). 8. Cordillera de Nahuelbuta. See Sierra Nahuelbuta. 8. Sierra Nahuelbuta, 37°48'S, 73°02'W; highest point, 1,440 m. W. H. Osgood, November 1939, 1 (FMNH). 8. Vega Blanca. See Sierra Nahuelbuta. 9. Isla de la Mocha. See Isla Mocha. 9. Isla Mocha, 38°22'S, 73°56'W. D. S. Bullock, December 1932, 1 (AMNH). 82 FIELDIANA: ZOOLOGY Malleco 10. Angol, 37°48'S, 72°43'W; 150 m. D. S. Bullock, September 1929, 2 (AMNH), July 1932, 1 (AMNH). 11. Laguna Malleco, 38°13'S, 71°49'W. 12. Galvarino, 38°24'S, 72°47'W; 61 m. Cautin 13. Temuco, 38°44'S, 72°36'W; 113 m. 13. Maquehue, Temuco. See Maquehua. 13. Maquehua, Temuco, 38°46'S, 72°39'W. D. S. Bullock, April 1910, 1 (AMNH). 14. Pitrufquen, 38°59'S, 72°39'W; 93 m. Carlos S. Reed, June 1926, 1 (FMNH), November 1955, 1 (FMNH). 15. Tolten, 39°13'S, 73°13'W. Carlos S. Reed, March, May 1926, 2 (FMNH). Valdivia 16. Callumapu, 39°36'S, 72°13'W. 17. Valdivia, 39°48'S, 73°14'W; 7 m. Carlos S. Reed, October 1936, 1 (FMNH). 17. Santo Domingo, 39°53'S, 73°12'W. 18. Mariquina, 40°08'S, 72°20'W. Llanquihue 19. Vicente Perez Resales, Parque Nadonal, 41°00'S, 72n5'W. 19. Petrohue, Lago Todos Santos, 41°08'S, 72°25'W. W. H. Osgood, November 1939, 1 (FMNH). 19. Cayetue. See Cayutue. 19. Cayutue, Lago Todos Santos, 41°14'S, 72°17'W; 221 m. 20. Lago Llanquihue, 41°08'S, 72°48'W; 51 m. 21. Llanquihue, 41°15'S, 73°01'W; 54 m. 22. Puerto Montt, 41°28'S, 72°57'W; 1 m. F. M. Chapinar, April 1920, 1 (AMNH). Chiloe — Isla Chiloe. Carlos S. Reed, December 1925, 2 (FMNH), February 1933, 1 (FMNH); B. Dean, 1 (USNM); Gerrard, 2 (MCZ); no data, 2 (FMNH). 23. Lago Cucao, Isla Chiloe, 42°38'S, 74°07'W. Junius Bird, 1936, 1 (AMNH). 24. Quellon, Isla Chiloe, 43°08'S, 73°40'W. W. H. Osgood, January 1923, 6 (FMNH). 25. Rio Inio, mouth, Isla Chiloe, 43°21'S, 74 WW. W. H. Osgood, January 1923, 1 (FMNH). Aisen 26. Rio Baker, 47°30'S, 73°37'W. Magallanes 27. Isla Riesco, 53°00'S, 73°37'W. v ARGENTINA— PMdH puda Neuquen 28. Quillen (Rio), 39°23'S, 70°56'W. 29. Lago Lolog, San Martin de los Andes, 40°05'S, 71°20'W. 30. Cordillera del Trompul, San Martin de los Andes, 40°08'S, 71°26'W. 30. Lago Lacar, 40°11'S, 71°30'W. 30. Brazo Correntoso, San Martin de los Andes, south of Cordillera del TrompuL 31. Lanin Parque National, 39°30'S, 71°30'W; 600-3,776 m. 32. Nahuel Huapi Parque National, 41°00'S, 71°30'W. Rio Negro 33. Lago Mascardi, 41°21'S, 71°52'W; 798 m. Chubut 34. Los Alerces Parque National, 42°50'S, 71°50'W. Santa Cruz 35. Perito Frantisco P. Moreno Parque National, 47°50'S, 72°15'W; 800-2,770 m. 36. Los Glatieres Parque National, 50°00'S, 73°30'W. HERSHKOVITZ: NEOTROPICAL DEER 83 XI. LITERATURE CITED Anonymous. 1%9. Tiergeburten im Kolner Zoo. Fr. Koln. Zoo, 12(2): 71-72. 1970. Tiergeburten im Kolner Zoo. Fr. Koln. Zoo, 13(2): 67. 1972. Saugetiere und Vogel die im Kolner Zoo geboren und grossgezogen wurden. Z. Koln. Zoo, 15(2): 57. .. 1973. Saugetiere und Vogel, die im Kolner Zoo geboren wurden. Z. Koln. Zoo, 16(2): 75. 1974. Saugetiere und Vogel, die im Kolner Zoo geboren wurden. Z. Koln. Zoo, 17(3): 103. .. 1977. Saugetiere und Vogel, die seit dem Erscheinen der Letzten Zoo-Zeitschrift im Kolner Zoo geboren wurden. Z. Koln. Zoo, 20(3): 101. .. 1978. Saugetiere und Vogel, die seit dem Erscheinen der Letzten Zoo-Zeitschrift im Kolner Zoo geboren wurden. Z. Koln. Zoo, 21(1): 27. Benirschke, K. 1977. Genetic management. Int. Zoo. Yearb., 17: 50-60, pi. 9. Blainville, H. H. de. 1816. Sur plusieurs especes d'animaux mammiferes, de I'ordre des ruminans. Bull. Sci. Soc. Philomat. Paris, 1816: 73-82. Brambell, M. R., and S. J. Mathews. 1977. Criteria for the management of pudu in captivity. Proc. Symp. 1, Assoc. Brit. Wild Animal Keepers, pp. 25-28. Brooke, V. 1874. On Sclater's muntjac and other species of the genus Cervulus. Proc. Zool. Soc. London, 1874: 33-42, pis. 8, 9. 1878. On the classification of the Cervidae, with a synopsis of the existing species. Proc. Zool. Soc. London, 1878(2): 883-998, pi. 55. Cabrera, A. 1961 (1958-1961). Catalogo de los mamiferos de America del sur. Rev. Mus. Argentino Cienc. Nat. "Bernardino Rivadavia," 4(2): 309-732. Carette, E. 1922. Cervidos actuates y fosiles de Sud America. Rev. Mus. de la Plata, 26: 393-472. Churcher, C. S. 1966. Observaciones sobre el status taxonomico de Epieuryceros Ameghino 1889 y sus espedes S. truncus y E. proximus. Ameghiniana, 4(10): 351-362, 13 figs. Cowan, I. M., and C. W. Holloway. 1973. Threatened deer of the world: Conservation status. Biol. Conserv., 5(4): 243-250, 1 fig. 1974. Threatened deer of the world: Research programmes for conservation. Biol. Conserv., 6(2): 112-117, 1 fig. 1975. Geographical location and conservation of the threatened deer of the world. Proc. Working Meeting lUCN Survival Service Comm. Deer Specialist Group. Longview, Washington, lUCN (Morges), pp. 11-22. CzERNAY, S. 1977. Zur Haltung und Zucht von Siidpudus (Pudu pudu Molina 1782) im Thiiringer Zoopark Erfurt. Zool Gart (n.f.), 47(3/4): 226-240, 14 figs. Dennler de la Tour, G. 1957. The national parks and reserves of Argentina. Oryx, 4(3): 159-167, 3 figs. DouROjEANNi, M. J. 1968. Estado actual de la conservacion de la flora y de la fauna en el Peru. Ciencia Interamericana, 9(1-6): 1-12, 8 figs. EvERNDEN, J. F. (ed.). 1976. Abnormal behavior prior to earthquakes, I. Office of Earth- quake Studies. U.S. Geol. Surv., Menlo Park. California, 429 pp. Flower, W. H. 1885. An Introduction to the Osteology of the Mammalia. Macmillan and Co., London, ed. 3, pp. i-ix, 1-383, 134 figs. Flerov, K. K. 1952. Musk deer and deer. Fauna of U.S.S.R. Mammals. (Moscow) 1(2): 1-257, 78 figs. (English translation published 1960 for the National Science Foundation, Washington, D.C., and the Smithsonian Institution by the Israel Program for Scientific Translations.) Fradrich, H. 1974. Notizen iiber seltener gehaltene Cerviden. Teil I. Zool. Gart. (n.f.), 44(4): 189-200, 10 figs. 1975. Notizen iiber seltener gehaltene Cerviden. Teil II. Zool. Gart., 45(1): 67-77, 9 figs. 84 FIELDIANA: ZOOLCX^Y Frechkop, S. 1%5. Notes sur les mammiferes. LII.-Au sujet d'un caractere dentaire des cervides. Bull. Inst. Roy. Sd. Nat. Belgique, 41(25): 1-10, 8 figs. Prick, C. 1937. Horned ruminants of North America. Bull. Amer. Mus. Nat. Hist., 49: i-xxviii, 1-669, 68 figs. Garrod, a. H. 1877. Notes on the visceral anatomy and osteology of the ruminants, with a suggestion regarding a method of expressing the relation of spedes by means of formulae. Proc. Zool. Soc. London, 1877: 2-18. Gay, C. 1847. Historia fisica y politica de Chile. Mus. Hist. Nat. Chile Santiago, 1: 1-496. Gay, C, and P. Gervais. 1846. Sur le Capra pudu et VEquus bisukus de Molina. Ann. Sd. Nat. Zool, Paris, ser. 3, 5: 87-94. Gmelin, J. F. 1788. Caroli Linne Systema Naturae per regna tria naturae . . . Leipzig, ed. 13, 1: [12], 1-500. Gray, J. E. 1850. Gleanings from the menagerie and aviary at Knowsley Hall. Hoofed quadrupeds. Knowsley [4], 1-76, 62 pis. 1852a. Synopsis of the spedes of deer (Cervina), with the description of a new spedes in the gardens of the sodety. Proc. Zool. Soc. London, 1850: 222-244, pis. 22-28. 1852b. Catalogue of the spedmens of Mammalia in the collection of the British Museum. Part III. Ungulata Furdpeda. London, pp. i-xvi, 1-286, 2 figs., 37 pis. Greer, J. K. 1965. Mammals of Malleco Province Chile. Publ. Mus. Michigan St. U., Biol. Sd., 3(2): 49-152, 8 pis, 5 figs. Grimwood, I. R. 1969. Notes on the distribution and status of some Peruvian mammals, 1968. Spedal Publ. no. 21. American Committee for International Wild Life Protection and New York Zoological Sodety. vi + 1-86 pp., illus. Haltenorth, T. 1963. Klassification des Saugetiere: Artiodactyla. I. Handb. Zool., Berlin, 8(32/1): 1-167, 16 pis. Harroy, J. P. 1971. United Nations list of national parks and equivalent reserves. Hayez, Brussels, ed. 2, 601 pp., illus. 1972. Addendum-Corrigendum to the second edition of the United Nations list of national parks and equivalent reserves. Hayez, Brussels, 107 pp. Hershkovitz, P. 1948. The technical name of the Virginia deer with a list of the South American forms. Proc. Biol. Soc. Washington, 61: 41-48. 1958a. The technical names of the South American marsh deer and pampas deer. Proc. Biol. Soc. Washington, 71: 13-16. -. 1958b. The metatarsal glands in white-tailed deer and related forms of the neo- tropical region. Mammalia, 22(4): 537-546. .. 1958c. A geographic classification of neotropical mammals. Fieldiana: Zool., 36(6): 583-620. 1959a. A new spedes of South American brocket, genus Mazoma (Cervidae). Proc. Biol. Soc. Washington, 72: 45-54, 4 pis. 1959b. A new race of red brocket deer {Mazama americana) from Colombia. Proc. Biol. Soc. Washington, 72: 93-96. -. 1966. Mice, land bridges and Latin American faunal interchange. In Wenzel, R. L. and V. V. Tipton, (eds.). Ectoparasites of Panama. Field Museum of Natural History, Chicago, pp. 725-751, figs. 151-154. 1972. The recent mammals of the neotropical region: A zoogeographic and ecologi- cal review. In Keast, A., Erk, F. G., and Glass, B. (eds.). Evolution, Mammals and Southern Continents. State U. of New York Press, Albany, N.Y., pp. 311-431, 16 figs. 1977. Living New World monkeys (Platyrrhini) with an introduction to Primates. Vol. 1. U. of Chicago Press, Chicago, 1,117 pp., 520 figs., 7 col. pis. Hick, U. 1967/68. Gegliickte Aufsucht eines Pudus {Pudu pudu Mol.) im Kolner Zoo. Fr. Koln. Zoo, 10(4): 111-117, 15 figs. 1969. Successful raising of a pudu, Pudu pudu, at Cologne Zoo. Int. Zoo. Yearb., 9:110-112. .. 1971. Saugetiere und Vogel, die in Kolner Zoo geboren und grossgezogen wur- den. Z. Koln. Zoo, 14(2): 83-84. HERSHKOVITZ: NEOTROPICAL DEER 85 1975. In Fradrich H., Notizen iiber seltener gehaltene Cerviden. Teil II. Zool. Gart. (n.f.), 45: 1-69. HoussE, p. R. 1953. Animales salvajes de Chile en su dasificacion moderna. Universidad de Chile (Santiago). (Work not seen, cited from Wetterberg, G. B. 1972. Pudu in a Chilean national park. Oryx, 11(5): 349.) HuNSAKER II, D. 1972. National parks in Colombia. Oryx. 11(6): 441-448, 4 figs. Jones, M. L. 1964. In Crandall, L.S., The Management of Wild Mammals in Captivity. U. of Chicago Press, Chicago, p. 595. JuNGE, C. 1932. Puduhirsche sind sehr hitzeempfindlich. Zool. Gart. (n.f.), 5: 241. 1966. Pudu (Pudu pudu) at Chilian Viejo Zoo. Int. Zoo. Yearb. 6: 263-264. Kohls, G. M. 1969. Ixodes taglei n. sp. (Acarina: Ixodidae) a parasite of the deer, Pudu pudu (Mol.), in Chile. J. Med. Entomol., 6(3): 280-283. KouLiscHER, L., J. Tyskens, and J. MoRTLEMANS. 1972. Mammalian cytogenetics. VII. The chromosomes oiCenms canadensis, Elaphurus davidianus, Cenms nippon (Temminck) and Pudu pudu. Acta Zool. Path. Antwerp., no. 56: 25-30, 3 figs. Kraglievich, L. 1932. Contribucion al conocimento de los dervos fosiles del Uruguay. An. Mus. Hist. Nat. Montevideo, ser. 2(3): 355-439, pis. 60-67. KuLLMANN, E. 1976. Jahresbericht 1975 der Aktiengesellschaft zoologischer Garten Koln. Z. Koln. Zoo, 19(2): 31-41. Lehmann, E. von. 1960. Das Problem der Grossenabnabnahme (Deminutions-Tendenz) beim Reh., Z. Jagdv^ssensch. 6(2): 41-51, 1 fig. Lehmann, F. C. 1945. Un venado del subgenero Pudella nuevo para la Cienda. Rev. Univ. Cauca (Popayan, Colombia) Zool. no. 6: 76-79, 2 figs. 1959. Pudu mephistophiles wetmorei Lehmann — Venado Conejo. Nov. Colombianas. Contrib. Qent. Mus. Hist. Nat. Univ. Cauca (Popayan, Colombia), 1(4): 202-204, 3 figs. 1960. Notas adidonales sobre Pudu mephistophiles wetmorei Lehmann. Nov. Colom- bianas. Contrib. Cient. Mus. Hist. Nat. Univ. Cauca (Popayan, Colombia), 1(4): 280- 283, 1 fig. Lydekker, R. 1915. Catalogue of the Ungulate Mammals in the British Museum (Natural History). Vol. IV. Artiodactyla, families Cervidae (deer), Tragulidae (chevrotains), Camelidae (camels and llamas), Suidae (pigs and peccaries), and Hippopotamidae (hip- popotamuses). British Museum, London, xxi -I- 1-438. Matthew, W. D. 1908. Osteology of Blastomeryx and phylogeny of the American Cervidae. Bull. Amer. Mus. Nat. Hist., 24: 535-562. Molina, G. I. 1782. Saggio sulla storia naturale del Chili. Tommaso d'Aquino Press, Bologna, Italy, 368 pp. NoTTEBOHN, F. 1967. Argentine fauna: Time of dedsion. Anim. Kingd. 70: 112-118, 9 figs. Osgood. W. H. 1943. The mammals of Chile. Field Museum Zool. Ser., 30(542): 1-268, 33 figs. Pascual, R., E. Ortega and E. Tonnl 1966. Las edades del Cenozoico mamalifero de la Provinda de Buenos Aires. In Borrello, A.V., Paleontografia bonaerense. Com. Inv. Cient. Prov. Buenos Aires. (La Plata), fasc. 4 (Vertebrata): 3-27. Pefaur, J. 1969. Consideradones sobre el problema de la conservadon de los mamiferos silvestres Chilenos. Bol. Informativo Santiago, no. 93/94: 4-10. PococK, R. I. 1911. On the spedalised cutaneous glands of ruminants. Proc. Zool. Soc. London, 1910: 840-986, 142 figs. 1923. On the external characters of Elaphurus, Hydropotes, Pudu, and other Cer- vidae. Proc. Zool. Soc. London, 1923: 181-207, figs. 2-17. Rafinesque, C. S. 1817a. Museum of natural sdences. Dissertation on water snakes, sea snakes and sea serpents. American Monthly Mag., 1: 431-437. 1817b. Descriptions of seven new genera of North American quadrupeds. Ameri- can Monthly Mag., 2: 44-46. Schmidt, H. 1944. Argentinische Saugetiere. Imprenta Merder, Buenos Aires, 258 pp., 117 figs., 2 col. pis. 86 FIELDIANA: ZOOLOGY Schneider, C. O. 1946. Catalogo de los mamiferos de la Provinda de Concepdon. Bol. Soc. Biol. Concepdon, 21: 67-83. ScLATER, P. L. 1871. Notes on rare or little known animals now or lately living in the Sodety's Gardens. Proc. Zool. Soc. London, 1871: 221-240, pis. 14-17. Simpson, G. G. 1945. The prindples of classification and a classification of mammals. Bull. Amer. Mus. Nat. Hist., 85: i-xvi, 1-350. Spotorno, O., and R. Fernandez-Donoso. 1975. The chromosomes of the Chilean dwarf-deer "pudu," Pudu pudu (Molina). Mamm. Chromosome Newsl., 16(1): 17-18, fig. 10. Taylor, K. M., D. A. Hungerford, and R. L. Snyder. 1969. Artiodactyl mammals: Their chromosome cytology in relation to patterns of evolution. In Benirschke, K., (ed.). Comparative Mammalian Cytogenetics. Springer Verlag, New York, pp. 346-356. Thomas, O. 1908. A new deer of the brocket group from Venezuela. Ann. and Mag. Nat. Hist., ser. 8, 1: 349-350. 1913. On certain of the smaller S. -American Cervidae. Ann. and Mag. Nat. Hist., ser. 8, 11: 585-589. Van der Hammen, T. 1974. The Pleistocene change of vegetation and climate in tropical South America. J. Biogeogr., 1: 3-26. Vanoli, T. 1967. Beobachtungen an pudus, Mazama pudu (Molina 1782). Saugetierk. Mitt., 15(2): 155-163, 10 figs. Wetterberg, G. B. 1972. Pudu in a Chilean national park. Oryx, 11(5): 347-351. 4 figs. Windecker, W. 1%9. Jahresbericht des Zoologischen Gartens Koln 1968. Fr. Koln. Zoo, 13(1): 3-12. 1971. Jahresbericht des Zoologischen Gartens Koln 1970. Zeitschr. Koln. Zoo, .14(1): 3-12. WiNTON, W. E. DE. 1896. On some mammals from Ecuador. Proc. Zool. Soc. London, 18%: 507-513, pis. 19, 20. Yepes, J. 1943. Localidades argentinas para el pudu chileno. Rev. Argentina Zoogeogr., 3(1-2): 75-76. ZucKERMAN, S. 1952. The breeding season of mammals in captivity. Proc. Zool. Soc. London, 122: 827-950. Field Museum of Natural History Roosevelt Road at Lake Shore Drive Chicago, Illinois 60605-24% Telephone: (312) 922-9410