o OJ Q: Q. a I B R.AR.Y OF THE UN I VERSITY Of ILLI NOI5 590-5 FI Return this book on or before the Latest Date stamped below. Theft, mutilation, and underlining of books arc reasons for disciplinary action and may result in dismissal from the University. University of Illinois Library MOV 1 1 19f7 L161— O-1096 EVOLUTION OF NEOTROPICAL CRICETINE RODENTS (MURIDAE) WITH SPECIAL REFERENCE TO THE PHYLLOTINE GROUP PHILIP HERSHKOVITZ FIELDIANA: ZOOLOGY VOLUME 46 Published by CHICAGO NATURAL HISTORY M! S DECEMBER 20, W&2 CORRIGENDA FIBLDIANA: ZOOLOGY VOLUME 46 Page 84, footnote 1: Add to subgenera of Oryzomys, the following, Macruroryzo- mys, Microneclomys^ Page 97, caption for figure 18B, third line: Read confluence for "coalescence." Caption for figure 18C, third line: Read confluent for "coalesced." Page 117, line 16 from bottom: Read ectolophid for "ectoloph." Page 153, line 13 from bottom: Read provenance for "precedence." Page 346, line 12 from bottom: Read 14 for "12." Line 3 from bottom: Read 13 for "12." Page 347, line 7 from top: Read 11 for "10." Page 357, in column under "tail pilosity:" Insert X for Urco, and read 14 for "13." Page 403, on map: Read BRAZIL for "BOLIVIA." a«i 21, 22, InUruhaag* vor4 lttt«rgr*d« (Ut>« with «ord i«t«rty«f»fl (liM 22). 139, fi*« 36, pq^ttilar length is from honwtUun as 1 1 s»lat >I Ion th nho-il-t ho fr-t ' or front of iuci ,or ,. FIELDIANA: ZOOLOGY A Continuation of the ZOOLOGICAL SERIES of FIELD MUSEUM OF NATURAL HISTORY VOLUME 46 CHICAGO NATURAL HISTORY MUSEUM CHICAGO, U.S.A. 1962 EVOLUTION OF NEOTROPICAL CRICETINE RODENTS (MURIDAE) WITH SPECIAL REFERENCE TO THE PHYLLOTINE CROUP 3 • £: C8 S ca Ml «3 Be **p2 ° 2 *^ ^a C ^ ^5j 3 o ^e OH WCL, QP..-2 ft 3 o. 6 - ••3 01 03 »* EVOLUTION OF NEOTROPICAL CRICETINE RODENTS (MURIDAE) WITH SPECIAL REFERENCE TO THE PHYLLOTINE GROUP PHILIP HERSHKOVITZ Curator, Division of Mammals FIELDIANA: ZOOLOGY VOLUME 46 Published by CHICAGO NATURAL HISTORY MUSEUM DECEMBER 20, 1962 Preparation of the manuscript was aided by a grant (G-10753) from the National Science Foundation. Library of Congress Catalog Card Number: 62-22372 PRINTED IN THE UNITED STATES OF AMERICA BY CHICAGO NATURAL HISTORY MUSEUM PRESS CONTENTS PAGE LIST OF ILLUSTRATIONS . 9 PRELIMINARY REMARKS Introduction 13 Material 15 Acknowledgments and Abbreviations 15 Measurements 16 GENERAL DISCUSSION Origin, Evolution and Dispersal 16 Classification and Interrelationship 19 Adaptive Radiation '25 Juvenal Characters and Development 27 Reproduction 28 Pocket Populations 28 Sympatry and Allopatry 31 Crop Variation 36 Niche Variation 38 Sociability 41 Ratadas and Plant Fruiting Cycles 42 Hibernation 46 Enemies 46 Ectoparasites 47 SOME SPECIAL EXTERNAL AND CRANIAL CHARACTERS Feet 51 Tail 53 Bony Palate 54 Supraorbital Region 57 Baculum 58 DENTAL CHARACTERS AND EVOLUTION Enamel Folds in Rodent Molar Evolution 69 Procingulum and Postcingulum 74 The Mesoloph in the Pentalophodont and Tetralophodont Patterns ... 76 The "Pseudomesoloph" and the Dental Pattern in Cricetine Systematics 80 Dynamics of Rodent Molar Evolution 82 Plication 83 Planation 86 Hypsodonty Lamination 92 Involution 93 Triangulation 95 Fusion . 97 4 CONTENTS PAGE Cylindrification 98 Differential Evolutionary Rates in Upper and Lower Molars 99 Third Molar 100 Molar Roots 101 Upper Incisor 101 Lower Incisor 107 Incisor Indices 107 SYSTEMATIC REVISION OF PHYLLOTINES Characters of the Phyllotine Group of Cricetine Rodents 116 Key to Genera and Species 118 Calomys Section Genus Calomys Waterhouse 123 Diagnostic Characters and Key to the Species of Calomys 137 Calomys sorellus Thomas 137 Calomys laucha Olfers 142 Calomys laucha laueha Olfers 149 Calomys laucha lener Winge 157 Calomys lepidus Thomas 160 Calomys lepidus lepidus Thomas 162 Calomys lepidus ducillus Thomas 163 Calomys lepidus argiirus Thomas 164 Calomys lepidus carillus Thomas 164 Calomys callosus Rengger 165 Calomys callosus callosus Rengger 171 Calomys callosus expulsus Lund 174 Genus Eligmodontia F. Cuvier 175 Eligmodontia typus F. Cuvier 183 Eligmodontia typus typus F. Cuvier 184 Eligmodontia typus puerulus Philippi 186 Genus Zygodontomys J. A. Allen 196 Zygodontomys brevicauda cherriei J. A. Allen 203 Zygodontomys brevicauda ventriosus Goldman 204 Zygodontomys brevicauda seorsus Bangs 204 Zygodontomys brevicauda sanctaemartae J. A. Allen 204 Zygodontomys brevicauda brunneus Thomas 204 Zygodontomys brevicauda punctulatus Thomas 204 Zygodontomys brevicauda stellae Thomas 205 Zygodontomys bretncauda thomasi J. A. Allen 205 Zygodontomys brevicauda brevicauda J. A. Allen and Chapman .... 205 Zygodontomys brevicauda microtinus Thomas 205 Zygodontomys lasiurus fuscinus Thomas 205 Zygodontomys lasiurus pixuna Moojen 206 Zygodontomys lasiurus lasiurus Lund 206 Zygodontomys lasiurus brachyurus Wagner 206 Zygodontomys [tlasiurus] lenguarum Thomas 206 Zygodontomys [tlasiurus] tapirapoanus J. A. Allen 207 Phyllotis Section Genus Pseudoryzomys Hershkovitz 208 Pseudoryzomys wavrini Thomas 215 CONTENTS 5 PACK Genus Phyllotis Waterhouse 1217 Phyllotis danrini Complex 234 Phyllotis haggardi Thomas 256 Phyllotis andittm Thomas 260 Phyllotis danrini Waterhouse 269 Phyllotis danrini posticalis Thomas 282 Phyllotis danrini magister Thomas 288 Phyllotis danrini definitus Osgood 296 Phyllotis danrini limatus Thomas 299 Phyllotis danrini ntpestris Gervais 302 Phyllotis danrini danrini Waterhouse 320 Phyllotis danrini fulrescens Osgood 325 Phyllotis danrini xanthopygus Waterhouse 327 Phyllotis danrini caprinus Pearson 330 Phyllotis danrini wolffsohni Thomas 339 Phyllofis osilae J. A. Allen 344 Phyllotis osilae osilae J. A. Allen 380 Phyllotis osilae phaeus Osgood 384 Phyllotis osilae tucumanus Thomas 388 Phyllotis osilae nogalaris Thomas 390 Phyllotis micropus Waterhouse 391 Phyllotis pictus Thomas 404 Phyllotis boliviensis Waterhouse 410 Phyllotis boliviensis boliriensis Waterhouse 416 Phyllotis boliviensis flavidior Thomas 419 Phyllotis sublimis Thomas 419 Phyllotis sublimis sublimis Thomas 427 Phyllotis sublimis leucurus Thomas 428 Phyllotis gerbillus Thomas 430 Phyllotis amicus Thomas 438 Phyllotis griseoflarus Waterhouse 441 Phyllotis griseoflavus griseoflavus Waterhouse 451 Phyllotis griseoflarus domorum Thomas 458 Phyllotis edithae Thomas 461 Phyllotis hypogaeus Cabrera 462 Genus Galenomys Thomas 464 Galenomys garleppi Thomas 468 Genus Andinomys Thomas 473 Andinomys edax edax Thomas 481 Andinomys edax lineicaudatus Yepes 482 Genus Chinchillula Thomas 485 Chinchillula sahamae Thomas 490 Genus Euneomys Coues 493 Euneomys chinchilloides chinchilloides Waterhouse 498 Euneomys chinchilloides petersoni J. A. Allen 499 Euneomys noei Mann 499 Euneomys mordax Thomas 500 Euneomys catenatus Ameghino 500 ?Euneomys fossor Thomas 500 6 CONTENTS PAGE LITERATURE CITED 503 INDEX . .511 TABLES OF MEASUREMENTS AND COMPARISONS 1. Comparative size of bacula in the Phyllotis ssp 65-68 2. Calomys sorellus Thomas 189 3. Calomys laucha laucha Olfers 190 4. Calomys lepidus Thomas 191 5. Calomys callosus callosus Rengger 192 6. Calomys callosus expulsus Lund 193 7. Eligmodontia typus typus F. Cuvier 194 8. Eligmodontia typus puerulus Philippi 195 9. Zygodontomys lasi ur us Lund and Z. bretricaudaJ. A. Allen and Chapman . 207 10. Pseudoryzomys wavrini Thomas 215 11. Phyllotis haggardi Thomas and P. fuscus Anthony 260 12. Phyllotis haggardi Thomas 261 13. Phyllotis andium Thomas and P. darwini posticalis Thomas from Casa- palca, Lima, Peru 264 14. Phyllotis andium Thomas 267, 268 15. Phyllotis darwini Waterhouse, breeding data 278-281 16. Phyllotis darwini posticalis Thomas 286, 287 17. Phyllotis darwini magister Thomas and sympatric series of P. d. rupestris Gervais 291 18. Phyllotis darwini magister Thomas and P. d. rupestris Gervais, altitudinal distribution 294 19. Phyllotis darwini magister Thomas 295 20. Phyllotis darwini definitus Osgood 297 21. Phyllotis darwini limatus Thomas 301 22. Phyllotis darwini rupestris Gervais from successively higher altitudes . . 307 23. Phyllotis darwini rupestris Gervais 312-317 24. Phyllotis darwini darwini Waterhouse 324 25. Phyllotis darwini fulvescens Osgood 326 26. Phyllotis darwini xanthopygus Waterhouse 328 27. Phyllotis darurini caprinus Pearson 338 28. Phyllotis darwini wolffsohni Thomas 342 29. Ph yllotis osilae J. A. Allen, breeding data 356 30. Comparisons of sympatric series of Phyllotis osilae J. A. Allen and P. dar- wini Waterhouse 357 31. Comparison of Phyllotis osilae osilae J. A. Allen from Arapa, Puno, Peru, with sympatric P. darwini rupestris Gervais 359 TABLES OF MEASUREMENTS AND COMPARISONS 7 PAGE 32. Comparison of Phyllotis osilae osilae J. A. Allen from Asillo, Puno, Peru, with sympatric P. d. rupestris Gervais 360 33. Comparison of Phyllotis osilae osilae J. A. Allen from Chucuito, Puno, Peru, with sympatric P. darwini rupestris Gervais 361 34. Comparison of Phyllotis osilae osilae J. A. Allen from Huacull-mi, Pun:>, Peru, with sympatric P. darwini rupestris Gervais 362 35. Comparison of Phyllotis osilae osilae J. A. Allen from Juli, Puno, Peru, with sympatric P. darwini rupestris Gervais 363 36. Comparison of Phyllotis osilae osilae J. A. Allen from Occomani, Puno, Peru, with sympatric P. darwini rupestris Gervais 364 37. Comparison of Phyllotis osilae osilae J. A. Allen from Pairumani, Puno, Peru, 13,500 ft., with sympatric P. darwini rupestris Gervais .... 365 38. Comparison of Phyllotis osilae osilae J. A. Allen from Pairumani, Puno, Peru, 13,000 ft., with sympatric P. darwini rupestris Gervais . . 366, 367 39. Comparison of Phijllotis osilae osilae J. A. Allen from Pomata, Puno, Peru, with sympatric P. darwini rupestris Gervais 368 40. Comparison of Phyllolis osilae osilae J. A. Allen from Puno, Puno, Peru, 5 km. W., with sympatric P. darwini rupestris Gervais 369 41. Comparison of Phyllotis osilae osilae J. A. Allen from Puno, Puno, Peru, 15 km. W., with sympatric P. darwini rupestris Gervais 370 42. Comparison of Phyllotis osilae osilae J. A. Allen from Umayo, Puno, Peru, with sympatric P. darwini rupestris Gervais 371 43. Comparison of Phyllotis osilae osilae J. A. Allen from Yunguyo, Puno, Peru, with sympatric P. darwini rupestris Gervais 372 44. Comparison of Phijllotis osilae osilae J. A. Allen from Huaracondo, Cusco, Peru, with sympatric P. darwini poslicalis Thomas 373 45. Comparison of Phyllotis osilae osilae J. A. Allen from Ollantaytambo, Cusco, Peru, with sympatric P. darwini posticalis Thomas 374 46. Comparison of Phyllotis osilae osilae J. A. Allen from Hacienda Urco, Cusco, Peru, with sympatric P. darwini posticalis Thomas .... 375 47. Comparison of Phyllotis osilae phaeus Osgood from Limbani, Puno, Peru, 13,000 ft., with sympatric P. darwini posticalis Thomas 376 48. Comparison of Phyllotis osilae phaeus Osgood from Limbani, Puno, Peru, 11,000 11,500 ft., with P. darwini posticalis Thomas from Limbani, 15,000 ft 377,378 49. Comparison of Phyllotis osilae osilae J. A. Allen from Choro, Cocha- bamba, Bolivia, with P. darwini posticalis Thomas from Palmira, Apurimac, Peru 379 50. Phyllotis osilae osilae J. A. Allen 382, 383 51. Phyllotis osilae phaeus Osgood 387 52. Phyllotis osilae tucumanus Thomas 389 53. Phyllotis micropus Waterhouse 400 54. Phyliotis pictus Thomas .409 55. Phyllotis bolirienxis bolitriensis Waterhouse 418 56. Phyllotis boliviensis flavidior Thomas 418 57. Phyllotis sublimis sublimix Thomas and P. s. leucurux Thomas .... 429 58. Phyllotis gerbillus Thomas . . 433 8 TABLES OF MEASUREMENTS AND COMPARISONS PAGE 59. Phyllotis amicus Thomas 440 60. Phyllotis griseoflavus Waterhouse, sex ratios and reproduction 450 61. Phyllotis griseoflavus griseoflavus Waterhouse 456, 457 62. Phyllotis griseoflavus domorum Thomas 460 63. Andinomys edax edax Thomas 483 64. Andinomys edax lineicaudatus Yepes 483 65. Chinchillula sahamae Thomas 491 66. Nominal forms of Euneomys 502 LIST OF ILLUSTRATIONS Habitats of phyllotines in Peru Frontispiece PACK 1. Map showing distribution of phyllotine rodents in South and Central America 14 2. Chart showing interrelationship of phyllotines and other American crice- tines 22 3. Hind feet of phyllotine rodents 52 4. Hard palates in Muridae 55 5. Glandes pene and bacula of cricetines; simple and complex types ... 59 6. Bacula of phyllotines 60 7. Bacula of Phyllotis darwini complex 63 8. Bacula of Phyllotis andium 64 9. Enamel pattern of the molars of Muridae 71 10. Evolution of the occlusal surface of a generalized cricetine molar . 77 11. Molars of Aporodon tenuirostris and Reithrodontomys fulvescens .... 79 12. Lower first molar of Peromyscus truei gilberti 81 13. Upper second molar of the Eocene Theridomys 85 14. Molar planation 87 15. Hypsodonty in rodents 91 16. 17. Lamination and involution in rodent molars 94, 96 18. Fusion and cylindrification in rodent molars 97 19. Forms of incisors in phyllotines 103 20. Types of incisors in rodents 105 21. Incisive and molar planes 108 22. Skull, showing cranial measurements used in text 109 23. Murid skull, dorsal aspect Ill 24. Murid skull, ventral aspect 112 25. Murid skull, lateral aspect 113 26. Murid mandible . . 115 27. Map showing distribution of the genera of the Calomys section in South America and Panama 122 28. Map showing distribution of the species of Calomys 125 29. Skulls of Calomys lepidus lepidus, C. laucha laucha, C. sorellus, and C. cal- losus callosus 131 30. Skulls and mandibles of Calomys lepidus lepidus, C. laucha, C. sorellus and C. callosus callosus 132 31. Skulls of Baiomys taylori and Calomys laucha 133 10 LIST OF ILLUSTRATIONS PAGE 32. Palate of Baiomys taylori, Calomys laucha and C. sorellus 134 33. Molars of Calomys sorellus, C. laucha, C. lepidus and Baiomys taylori . . 135 34. Skulls of Zygodontomys brericauda, Calomys callosus and Phyllotis griseo- flavus 136 35. Map showing distribution of Calomys sorellus 139 36. Skull of Calomys sorellus 141 37. Map showing distribution of the subspecies of Calomys laucha .... 145 38. Head and feet of Calomys laucha 146 39. Map showing distribution of the subspecies of Calomys lepidus . . . .159 40. Skulls of Calomys lepidus 161 41. Map showing distribution of the subspecies of Calomys callosus .... 167 42. Skulls of Calomys callosus callosus 168 43. Palate and molars of Calomys callosus 169 44. Molars of Calomys callosus, Zygodontomys bretricauda and Phyllotis griseo- flarus 170 45. Map showing distribution of the subspecies of Eligmodontia typus . . . 177 46. Skulls of Eligmodontia typus 178 47. Palate, skull and mandible of Eligmodontia typus 179 48. Molars of Eligmodontia typus 180 49. Stages of molar planation shown in Zygodontomys, Eligmodontia and Phyllotis amicus 181 50. Map showing type localities of the nominal subspecies of Zygodontomys brericauda and Z. lasiurus 197 51. Skull and mandibles of Zygodontomys brevicauda 200 52. Map showing distribution of the genera of the Phyllotis section .... 209 53. Plantar surface of right hind foot of Pseudoryzomys wavrini and Oryzomys palustris 210 54. Skulls of Pseudoryzomys waerini, Calomys callosus and Oryzomys palustris 212 55. Skulls and mandibles of Pseudoryzomys wavrini and Oryzomys palustris . 213 56. Molars of Pseudoryzomys warrini, Calomys callosus and Oryzomys palus- tris 214 57. Map showing distribution of the species of Phyllotis 218 58. Skulls of Phyllotis griseoflarus, P. pictus, P. darmni, P. micropus, P. sub- limis and P. amicus 224 59. Skulls of Phyllotis griseoflarus, P. darmni, P. micropus, P. sublimis, P. pictus and P. boliriensis 225 60. Mandibles of Phyllotis griseoflavus, P. darmni, P. micropus, P. sublimis, P. pictus, P. boliriensis and Mystromys albicaudatus 226 61. Skulls of Phyllotis darmni, Galenomys garleppi, Andinomys edax, Chin- chillula sahamae and Euneomys chinchilloides 227 62. Skulls of Phyllotis darmni, Galenomys garleppi, Chinchillula sahamae and Euneomys chinchilloides 228 63. Mandibles of Phyllotis darmni, Galenomys garleppi, Andinomys edax, Chinchillula sahamae and Euneomys chinchilloides 229 64. Skulls of Mystromys albicaudatus and Phyllotis darwini 230 LIST OF ILLUSTRATIONS 11 PAGE 65. Skulls of Mystromys albicaudatus and Phyllotis darwini 231 66. Palate and molars of Zygodontomys brevicattda, Mystromys albicaudatus and Phyllotis darwini 231 67. Molars of Mystromys albicaudatits, Phyllotis griseoftavus, I', darwini, P. micropits, P. pictus, P. sublimis and P. amicus 232 68. Molars of Phiillotis darwini and P. micropus 233 69. Molars of Phyllotis darunni, Galenomys garleppi, Andinomys edajc, Chin- chillula sahanuie and Euneomys chinchilloides 233 70. Map showing distribution of the subspecies of Phyllotis darwini . . . 235 71. Map showing distribution of Phyllotis haggardi and P. andium in Ecua- dor 239 72. Map showing distribution of Phyllotis andium, P. haggardi and the subspecies of Phyllotis darunni in Peru 240 73. Map showing distribution of the subspecies of Phyllotis darwini in south- ern Peru 243 74 Map showing distribution of the subspecies of Phyllotis darwini and P. osilae in southern Peru, Bolivia, Argentina and Chile 245 75. Map showing distribution of the subspecies of Phyllotis danvini in Ar- gentina and Chile 249 76. Map showing contact between Phyllotis osilae and P. darwini posticalis and probable origin of P. osilae from ancestral form of P. darwini posti- calis 252 77. Map showing distribution of the subspecies of Phyllotis osilae and P. dar- wini in southern Peru, Bolivia, northern Argentina and Chile. . . . 253 78. 79. Skulls of Phyllotis haggardi and P. darwini rupestris 272, 273 80. Abnormal skulls of Phyllotis darwini 274 81. Left molars of Phyllotis darwini 275 82. Age-size variation in Phyllotis darwini limatiis 298 83. 84. Skulls of Phyllotis darwini caprinus and P. d. wolffsohni . . . 332, 333 85. Upper molars of Phyllotis darwini caprinus and P. d. wolffsohni . . . . 334 86. Comparisons of Phyllotis osilae and P. darwini 349 87. 88. Skulls of Phyllotis darwini and P. osilae 354, 355 89. Map showing distribution of Phyllotis micropus 393 90. Skulls of Phyllotis micropus 396 91. Molars of Phyllotis micropus 397 92. Map showing distribution of the subspecies of Phyllotis pictus . . 403 93. Skulls of Phyllotis pictus 406 94. Map showing distribution of the subspecies of Phyllotis bolipienxi* . 411 95. Skulls of Phyllotis boliviensis 96. Molars of Phyllotis boliviensis .414 97. Map showing distribution of subspecies of Phyllotis sublimis . . 420 98. Skull of Phyllotis sublimis 423 99. Molars of Phyllotis sublimis 424 100. Map showing distribution of Phyllotis gerbillus and P. amicus 101. Skulls and molars of Phyllotis gerbillus and P. amicus 436 12 LIST OF ILLUSTRATIONS PAGE 102. Skulls and molars of Phyllotis amicus and P. gerbillus 437 103. Map showing distribution of the subspecies of Phyllotis griseoflavus . . 443 104. 105. Skulls of Phyllotis griseoflavus 444, 445 106. Molars of Phyllotis griseoflavus 447 107. Molars of Phyllotis hypogaeus 463 108. Map showing type localities of Galenomys garleppi, Phyllotis hypogaeus, and P. edithae 465 109. 110. Skulls of Galenomys garleppi and Phyllotis sublimis 467, 469 111. Molars of Galenomys garleppi and Phyllotis sublimis 470 112. Map showing distribution of the subspecies of Andinomys edax . . . 472 113. Skulls of A ndinomys edax 475 114. Molars of Phyllotis griseoflavus and Andinomys edax 476 115. 116. Molars of Neotoma (Hodomys) alleni, Neotoma (Teonoma) cinerea, Neotoma (Neotoma) mexicana, Andinomys edax and Phyllotis griseo- flavus 478, 479 117. Map showing distribution of Chinchillula sahamae 484 118, 119. Skulls of Chinchillula sahamae 487 120. Molars of Chinchillula sahamae 488 121. Map showing type localities of the subspecies of Euneomys chinchilloides 492 122. Skulls of Euneomys chinchilloides 495 123. Molars of Euneomys chinchilloides 497 Evolution of Cricetine Rodents INTRODUCTION ' Phyllotines are grazing, or pastoral, cricetines confined to the Neotropical region. They may be as small as house mice or as large as Norway rats. Their ancestors, along with other grazing and aquatic mice, were among the first cricetines to invade South Amer- ica over the Panamanian land bridge. Today, nearly every impor- tant grazing and grazing-browsing range of the Neotropical region from Costa Rica south to the Straits of Magellan, from sea level to more than 5000 meters above, or to the highest points capable of sustaining mammalian life, is occupied by one or more species of 1 This paper was submitted for publication in 1956 and was in galley proof when a published revision of the genus Phyllolis by Oliver P. Pearson came to hand in December, 1958. Pearson's work includes many advances over our erstwhile knowledge of the systematics and ecology of the species and subspecies included by me in the Phyllotis darwini complex. Many of Pearson's contributions are based on a considerable amount of material studied and collected in the field by the author himself. A part of this collection, totaling approximately 400 specimens including the bacula of 113, has since been made available to me by Dr. Seth B. Benson and Dr. Oliver P. Pearson of the Museum of Vertebrate Zoology, Uni- versity of California. I am particularly grateful to them for their kindness. My study of this material has resulted in an interpretation of the inter-relationship of the members of the Phyllotis darwini complex which differs in some ways from Pearson's own. A number of other important contributions to our knowledge of cricetine morphology and phylogeny also have appeared since 1956. Those pertinent to the present study have been taken into account either in the text or in footnotes. A series of short papers by Vorontsov (1957, 1959a, b, 1960a, b) on the anatomy of soft parts, zoogeography, adaptive radiation and systematics is of particular interest. Unfortunately, Vorontsov's documentation in support of his generaliza- tions and broad conclusions is sparse. The number of specimens dissected is not given and the variety of New World cricetines examined is small. Strangely, Vorontsov gives no indication that he is familiar with the bulk of the recent litera- ture on New World cricetines published in North and South America. Finally, on March 1, 1962, I received the second part of Dr. Angel Cabrera's Catalog dc los mamiferos de America del Sur. His classification of the species of I'hyllotix is that of Pearson. His treatment of the genera Calomyx and Eligmodontia may re- flect the content of some of the correspondence between him and me and, possibly, my determinations of a number of cricetines in the collection of the Buenos Aires museum. Other revisions made by Cabrera were not anticipated and correspond- ing changes in the text could not always be made at this date except by way of footnotes. 13 14 FIELDIANA: ZOOLOGY, VOLUME 46 90 70 50 20 20 40 i i PHYLLOTIS SECTION 20 20 40 90 70 50 30 FIG. 1.- — Map of South and Central America showing distribution of the phyllotine rodents. phyllotines (fig. 1). Rapid and pervasive dispersal of the group was promoted by the availability and abundance of natural pastures, by the absence of predatory insectivores, and by the paucity of small, grazing competitors. Man-made pastures and settlements in erst- while forest country are also being added to the range of phyllotines. Phyllotis darwini and P. micropus have followed man's clearings along river borders and through forests of Chile and western Argentina. Zygodontomys is rapidly becoming the common plague rodent in trop- HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 15 ical regions where previously forested areas have been transformed into pastures. Individuals of Phyllotis darwini, P. griseoflarm and the Mws-like Calomys callosus and C. laucha have invaded man's habitations. These species compete with introduced murines in farms, villages and even in large metropolises. They are among the most successful of indigenous mice and are abundant, often ex- cessively so, throughout their extensive range. One of them, Calo- mys laucha, has been accidentally imported into the tropical coastal areas of northern South America where it is now well established. This is an extension of range far removed from the natural habitat of the species in the temperate zones of southern South America. MATERIAL Approximately 2500 specimens of phyllotines were examined. Of these, 2308 are formally listed. The remainder represent Zygodontoniys, a genus not revised to the subspecies level, though a provisional clas- sification of its nominal subdivisions is given. Characterization of the non-phyllotine Euneomys, included in this monograph, is based on 20 specimens. Present material represents all but 5 of the 132 described forms of phyllotines. These 5, and 16 uncritically listed named forms of Zygodontomys, are included in the total of 60 species and subspecies recognized as valid. New names have not been proposed here. \CKNOWLEDGMENTS AND ABBREVIATIONS Thanks are expressed to the authorities of the institutions listed below for permission to study and report upon the specimens of phyllotines in their charge. The following abbreviations for the names of the institutions are used in the text and lists of specimens examined. A MNH = American Museum of Natural History BM = British Museum (Natural History) CNHM = Chicago Natural History Museum LAC M = Los Angeles County Museum RNHL=Rijksmuseum van Natuurlijke Historie, Leiden MACN=Museo Argentine de Ciencias Naturales, Buenos Aires MCZ= Museum of Comparative Zoology, Harvard University MVZ= Museum of Vertebrate Zoology, University of California A critical reading of a draft of this manuscript by Professor Bryan Patterson is gratefully acknowledged. I am specially thank- 16 FIELDIANA: ZOOLOGY, VOLUME 46 ful to Staff Artists E. John Pfiffner and Marion Pahl and Photog- raphers John Bayalis and Homer V. Holdren for the illustrations. MEASUREMENTS All measurements in the tables are given in millimeters. Figures in parentheses are the extremes, the mean precedes the parentheses, and the number of samples measured follows. Cranial terms and measurements are explained in figures 22-26. Except as noted, external measurements are those of the collectors and are not uniformly comparable. Unless otherwise indicated, measurements of "hind foot, dry," include the longest claw and were taken by the author. Ordinarily, altitudes and distances are given in the universal metric system. However, measurements recorded in feet and miles on specimen labels and in published descriptions have not been changed in the following lists of localities. Some collectors and authors have perversely converted into the English system altitudes and distances originally or officially recorded in the metric system. To reconvert such measurements here would compound errors of arithmetic and errors of judgment. ORIGIN, EVOLUTION AND DISPERSAL The earliest New World cricetines may have appeared during early Oligocene or even late Eocene. They may have originated in Eurasia and ranged into North America, or they may have arisen in North America and extended into the Old World. Whatever the place of origin, modern Old World cricetines are relatively few in kind, specialized in form, and pastoral in habitat. American species, on the other hand, are highly varied and occupy a diversity of pas- toral and sylvan niches. The ancestral form was essentially terrestrial and, judged by pres- ent habitats of cricetines with unmodified molars, a forest dweller. Its diet consisted mainly of succulent herbs, fruits, fungi, soft seeds, shoots, exposed roots, insects, and other small invertebrates. The molars were characterized by low crowns, a double row of moder- ately raised and rounded cusps, and three connecting crests, or lophs. These are the pentalophodont molars described in detail elsewhere (p. 76). They are designed for efficient mastication of food chiefly by crushing and pounding motions of the lower jaws. Other char- HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 17 acters of the ancestral species include a Mus-tike form and size, tail approximately equal to combined head and body length, toes well developed, sole and tail coarsely scutellated, skull delicate and gen- erally smooth, rostrum moderately elongated, hard palate wide, comparatively short and unfurrowed, upper incisors simple, slender and recurved. A Juvenal pelage distinguishable from that of the adult by its darker color and finer texture is another character of archaic sylvan cricetines. Although the first cricetines (or nearly related contemporaries) are not known, they may have been very much like modern representatives of Thomasomys, or typical Pero- myscus. The evolution of cricetines resembles that of horses in several important respects. Both categories began as forest dwellers in early Tertiary and both succeeded in spreading over most of the Holarctic region by late Tertiary. Where rainfall failed and forests were re- placed by savannas, progressive forms of early equids and cricetines gradually changed into grazers. The transition from sylvan to pas- toral life was made possible by the transformation of crested, low- crowned molars into plane, high-crowned molars capable of with- standing the attrition of hard grasses and associated grit. In crice- tines, the higher the molars, the earlier their eruption and function. This correlation corresponds to the presence and use of milk pre- molars by young ungulates. Shortening of the nestling stage and suppression of the distinctive Juvenal pelage in pastoral cricetines are other adaptive characters. These compare with adaptations of colts and other new-born grazing mammals which permit them to fend almost entirely for themselves in competition with adults.1 1 Vorontsov (1960b, p. 155) believes that the transition from sylvan to pas- toral habits is also "expressed ecologically in the reduction of the animals' mobil- ity, in the increase of the total volume of food consumed, in the transition from nocturnal to 24-hour activity, in the disappearance of the instinct of food pro- vision. . . . Morphologically, this transition is revealed in the reduction of the organs of sense (smell, sight, taste), in the simplification of the organs of move- ment, in the reduction of olfactory lobes of the cerebrum and cerebellum . . . and essential transformation in the alimentary system." There is no evidence that the locomotory organs of New World pastoral crice- tines are less versatile than those of their comparably specialized sylvan counter- parts. The contrary may even be true in specific cases. All cricetines are nocturnal and some species are also diurnal. Stresses and environmental pressures frequently cause diurnality in otherwise nocturnal species. Undoubtedly, pastoral species, because they are more frequently and easily observed by man, appear to be more commonly diurnal than sylvan species. Vorontsov's statement that pastoral spe- cies are losing their instinct for food hoarding must be a mistake of the translator (I have not seen the original paper in Russian, if published). Pastoral species are, of course, more regular and consistent storers than sylvan species. Vorontsov him- self (1960a, p. 983), speaking of Palearctic cricetines, emphasizes that the "features of their winter biology (food storage and hibernation) permit the present hamster 18 FIELDIANA: ZOOLOGY, VOLUME 46 New World grazing, or pastoral cricetines emerged in three cen- ters. The first was in the western United States during the Oligo- cene. Eumys Leidy and Leidymys Wood are two noteworthy products of the center. The same or similar genera may have migrated into the second center in Middle America when grasslands were estab- lished there, not later than early Pliocene. It is more likely, however, that the Middle American forms arose independently from the same generally distributed sylvan stock. No fossil records exist to throw light on this subject. The third center of origin of pastoral cricetines is in South America, beginning in late Pliocene or early Pleistocene. The order of invasion of South America by cricetines was determined by the normal sequence of ecological succession on the Panamanian land bridge. The first cricetine habitats established were pastoral, specifically aquatic and palustrine. Savannas, then forests, each with its fauna, followed. It is significant, then, that the progressive and specialized pastoral, rather than the archaic and sylvan cricetines were the first to enter South America. The highly specialized, aquatic ichthyomyines may have been in the vanguard of the inva- sion. As forests superseded the grasslands of the Andean and coastal regions of northwestern South America, pastoral cricetines (except- ing aquatic forms like Rheomys and Ichthyomys, whose habitats and modes of life were least affected) were pushed into the savannas of the Guianas and into the punas and pampas of the southern half of the continent. Pastoral cricetines of the Guianan region (including northeastern Brazil) are the phyllotine Zygodontomys, the sigmodonts, Sigmodon and Holochilus, and the relict akodonts, Chalcomys, Microxus and the oxymycterine Podoxomys. All Patagonian cricetines, except the ubiquitous Oryzomys (Oligoryzomys) nigripes (=longicaudatus), are pastoral. Each modern phyllotine species probably originated somewhere within its present range. The ancestral phyllotine was almost cer- tainly Middle American in origin. Its identity with Pleistocene (and late Pliocene?) fossils of the United States, recorded under such phyl- lotine names as Hesperomys (=Calomys) and Eligmodontia, cannot to survive. . . ." Differences in the sense organs and central nervous system cer- tainly exist between many species of cricetines, but a cleavage in these respects between New World sylvan and pastoral forms has not been shown. Evolution of a two-chambered stomach in Old World pastoral species from a primitive sac-like structure such as occurs in New World sylvan Nectomys squamipes was demon- strated by Vorontsov (1957, p. 526). On the other hand, the stomach of Old World pastoral Calomyscus and nearly all New World pastoral forms examined by Voront- sov remains single-chambered. HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 19 be demonstrated. What appears, nevertheless, is that molars of pas- toral cricetines are remarkably similar in all geographic centers of origin, whether New World or Old World. CLASSIFICATION AND INTERRELATIONSHIP Cricetines, including phyllotines, are here regarded as a subfamily of Muridae. Differences between the Murinae and Cricetinae are weak and depend almost entirely on whether or not stylar processes on the inner side of the upper molars are hypertrophied into func- tional cuspules. The addition of one or more such cuspules results in the complicated triserial tubercular pattern characteristic of Murinae as distinguished from the primitive biserial pattern retained in the Cricetinae. Actually, the difference is no more significant than the dental ones separating such cricetines as Oryzomys, Pero- myscus (s.s.), Thomasomys and Aporodon, all with a functional meso- loph fused with the mesostyle, from other cricetines such as Neotouia, Sigmodon, Hypsimys, Ichthyomys, etc., without mesoloph and meso- style. Indeed, the Murinae and all other supergeneric groups of liv- ing Muridae, save some Malagasy species, diverged from a cricetine stock in which the mesoloph was being eliminated. Notwithstand- ing gross differences between the extremes of the groups concerned, the record of change from species to species, genus to genus and sub- family to subfamily, is still preserved in Recent forms. Perhaps more basic than the structure of the molars as a criterion in the classification of myomorph rodents is the design of the baculum and glans penis. Two basic types of penes in the Cricetinae, and possibly in the Muridae as a whole, have been demonstrated by Hooper (1959, 1960) and Burt (1960). The simple type of penis con- tains a baculum made up of a bony shaft and a single cartilaginous segment at the tip. The baculum of the complex type is composed of a bony shaft with three osseous, cartilaginous or soft tissue finger- like processes attached to the tip (fig. 5). According to Hooper (1960, p. 17) the complex or quadripartite baculum with its correspondingly complex glans penis "is the common and geographically widespread one. It is predominant in cricetids (sensu Simpson, 1945) in both the Old and New Worlds, and likely it is typical of murids. It occurs in Mus, Rattus, Apodemus, Micromys, Acomys, Arvicanthis, NcKokia, Meriones, Rhambomys, Cricetus, Cricetulus, Mesocricetus, most if not all microtines, and most if not all South American cricetines [except Aporodon and Tylomys]. 20 FIELDIANA: ZOOLOGY, VOLUME 46 "While the complex plan is widespread, both taxonomically and geographically, the simple type is restricted to a few North American kinds. Of the 50 or more genera studied to date, the following twelve are characterized by simple glandes and bacula; Peromyscus, Reithro- dontomys [including Aporodon], Neotomodon, Ochrotomys, Onycho- mys, Baiomys, Scotinomys, Nelsonia, Neotoma, Xenomys, Ototylomys and Tylomys." This assemblage of genera, concludes Hooper (1960, p. 19), "has the aspect of a distinct natural unit, of subfamily or fam- ily rank, which like the Heteromyidae and Geomyidae, is endemic to the New World. The possibility that it is a natural group now requires intensive investigation, utilizing all pertinent anatomical data." That these twelve or thirteen genera comprise a natural group distinct from other cricetines seems to be confirmed by their cranial, dental and external characters. The differences between the genera inter se are typical of radiating forms and each genus could have diverged from a common Peromyscus-like ancestor. As a result of convergence, sylvan and pastoral peromyscines strongly resemble their ecological equivalents among South American cricetines. The similarities between sylvan Peromyscus of the simple penis group and sylvan Thomasomys of the complex penis group, however, may be more than superficial. These genera occupy a position very near the point of dichotomy between their respective phyletic lines and their kinship is correspondingly close. On the other hand, resem- blances between pastoral Baiomys with simple penis and pastoral Calomys with complex penis are superficial and clearly the result of convergence. The trenchant differences between these two genera are discussed elsewhere (p. 126). The systematic value of the structure of the cricetine penis can- not yet be fully appreciated or properly interpreted. The penes of the vast majority of the species and genera remain to be studied and critically compared and there is no fossil record. What is known of the differences between the two types of bacula and their glandes is not imposing. Conceivably, if not predictably, the two types may prove to be fully intergrading. No other basic anatomical character absolutely distinguishes any one supergeneric group of murids from another. Nevertheless, such characters are important and those of the penis or its parts may be transcendental. The taxonomic significance of the structure of the cricetine ali- mentary tract, particularly of the stomach, has been examined by Vorontsov (1957, p. 526). This investigator demonstrated the pres- ence of a two-chambered stomach in representatives of all Palearctic HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 21 Cricetinae except Calomyscus. Among New World cricetines he found a simple sac-like stomach in Nectomys squamipes, Sigmodon hispidus, Reithrodontomys megalotis and, from published data, in Peromyscus leucopus, Neotoma floridana and Oxymycterus rujus. In addition, Vorontsov noted the tendency for the formation of a three- chambered stomach in Haplomylomys californicus and that the sin- gle-chambered stomach of Nectomys squamipes most nearly resembled the postulated primitive cricetine condition (see also footnote, p. 18). From the foregoing, it is presumed that microtines and South American cricetines, both with complex glans penis, and peromys- cines, with simple glans penis, arose independently from a common cricetine ancestor. The ancestral type may have been very much like modern Thomasomys with the complex glans penis and a single- chambered stomach. Presumably, most New World cricetines, whether sylvan or pastoral, retain the simple stomach. Old World cricetines and murines may have diverged from a common stock which, in turn, was itself an independent offshoot from the original thomasomyine stock. Only Calomyscus among Old World murids has kept the structurally simple stomach. Widest cleavage among South American cricetines exists between forest, or sylvan, forms and their progressive pastoral relatives (fig. 2) . Morphological differences between the two divisions are of the same order, though weighed on finer scales, as those distinguishing prim- itive browsing horses from their grazing descendants. In the case of murids, however, it is common for certain primitive ancestral stocks to exist contemporaneously with their modern specialized descend- ants and connecting relatives. The older sylvan division includes thomasomyines (Nyctomys, Oto- nyctomys, Phaenomys, Rhipidomys, Thomasomys [Wilfredomys Avila Pires, a synonym]) and oryzomyines (Oryzomys [sensu lato], Neso- ryzomys, Neacomys, Scolomys, Megalomys). In these, each tooth of the molar series is pentalophodont, i.e., with mesoloph (-id) present and fused with mesostyl(id) (figs. 4, A; 10, A; 11, A, B; 13). Sylvan cricetines are primarily inhabitants of broad-leaf forests (selva). The terrestrial species habitually nest on the ground or in dry burrows. They may forage in trees, streams and lakes as well as on the forest floor. Arboreal cricetines nest, feed and breed in trees but some- times descend for food and water. Semi-aquatic forms nest on the banks of streams or lakes. There are no fossorial sylvan cricetines. The pastoral division includes akodonts, phyllotines, sigmodonts, ichthyomyines, oxymycterines and every other South American crice- PASTORAL CHINCHILLULA ' ANDINOMYS GALENOMYS \ PHYLLOTIS \ ^'T V- ENTALOPHODONT SYLVAN CRICETiNES FIG. 2.- — Interrelationship of the phyllotine genera and their morphological and ecological relationship to certain other South American cricetines. The progressive pastoral forms with tetralophodont molars evolved from sylvan pentalophodont stock. Cricetines with a simple type of penis probably evolved from a sylvan pentalophodont stock with the complex type of penis. 22 HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 23 tine not named above as a member of the sylvan division. Their molars are simplified, i.e., with mesoloph absent, or, if present, re- duced and never fused with the mesostyle. The mesolophid is uni- versally absent or obsolete. The molar pattern is tetralophodont, or may be reduced to trilophodont, bilophodont or cylindrodont (figs. 4, B-D; 10, B-F; 11, C, D). Pastoral cricetines inhabit pampas, punas, llanos, savannas, tundras, scrublands, rock lands, deserts and other types of grazing and grazing-browsing habitats. They also occur in thick second-growth woodlots during early serai stages, and in narrow leaf, or coniferous forests. Most pastoral species are terrestrial and nest on the ground or in burrows. Semiaquatic forms shelter along the banks of watercourses. A few species are fossorial and certain scansorial species have acquired arboreal nesting habits in scrub country subjected to periodic inundations and destruction of ground cover usually by burning. A number of sylvan cricetines may be excurrent elements in pas- toral regions, especially along streams and gallery forests. Con- versely, pastoral species may be incurrent in forests where water- ways, grassy banks or deforestation provides avenues of ingress. Sylvan and pastoral members of the peromyscine group differ in the same way as the corresponding divisions of the South American cricetine group. In some cases, however, species or even subspecies notably of the genus Peromyscus (cf. Hooper, 1957) appear to be intermediate between pastoral and sylvan divisions in habitat and molar design. They are best included in the pastoral division toward which they are evolving rather than with the sylvan group whence they departed. Certain other species, such as the woodrat (Neotonia floridana), with dentition highly specialized for pastoral life, have acquired sylvan habits in parts of their range. All microtines and Old World cricetines are pastoral. Their habi- tats and habits are similar to those of South American pastoral crice- tines and their dental evolution has followed parallel lines. Phyllotines belong to the pastoral division of South American cricetines. They may be recognized chiefly by their generally Mus- or vole-like appearance, light or moderately heavy skull without salient modifications, long Oryzomys-Yike palate with conspicuous posterolateral pits, long slender incisive foramina, and tetralophodont molars, i.e., molars with mesoloph (-id) obsolete or absent (fig. 4, B). Phyllotines probably evolved from the same line that gave rise to akodont rodents. The sigmodonts (Sigmodon, Reithrodon, N colo- ny s, Holochilus) are more specialized but appear to be progres- 24 FIELDIANA: ZOOLOGY, VOLUME 46 sive offshoots from the main phyllotine line. The Neotropical fish eaters typified by Ichthyomys and Rheomys, also with basic tetra- lophodont molars, are highly specialized derivatives of pastoral forms. The insectivore-like oxymycterines represent still another radiation of basic pastoral stock (fig. 2). Phyllotines are separable into two sections. The more primitive Calomys section includes the buno-brachyodont Calomys, Eligmo- dontia and Zygodontomys. The Phyllotis section, with high-crowned, terraced or planed molars (fig. 14), consists of Phyllotis, Pseudory- zomys, Galenomys, Andinomys and Chinchillula. The Mus-\ike Calomys sorellus and C. laucha are the most generalized and smallest members of the phyllotine group. The microtine and hamster- like Andinomys and Chinchillula, respectively, are the most special- ized and largest. External and cranial differences between the ex- tremes are hardly more trenchant than those separating species and subgenera. On the other hand, dental differences between Calomys and either Andinomys or Chinchillula are of the same order as those distinguishing a browsing buno-brachyodont ungulate of the Oligo- cene from its modern grazing platy-hypsodont descendant. In phyl- lotines, however, evolution in dental topography from low crown to high crown, from crested to plane, from tetralophodont to bilo- phodont (fig. 10), can be traced as a gradual transition through living species. Besides the generalized tetralophodont dental pattern, the species of Calomys exhibit in their crania the two basic supraorbital types (fig. 23) found in cricetines. One is with parallel- or concave-sided supraorbital region, the other with divergent-sided supraorbital region. All generalized external characters of phyllotines are found in Calomys. It may be assumed, therefore, that the pre-phyllotine ancestor differed little in external and cranial characters from such forms as Calomys sorellus and C. laucha. The ancestral molars, how- ever, must have been marked by the presence of a well-developed mesoloph. This is requisite for derivation of phyllotines from the basic pentalophodont cricetine stock. The remarkable resemblance between Calomys and the distantly related tetralophodont Baiomys of North America is principally superficial and appears to be the result of convergence (cf. p. 126). Certain gross resemblances between phyllotines and South African Mystromys (cf. p. 221) are obviously the result of parallel develop- ment of widely separated lines of murids. HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 25 The most specialized phyllotines grade into the sigmodont group (for characterization see Hershkovitz, 1955, p. 640).' Molars of sigmodonts are, in general, more advanced, with the enamel folds more compressed and tending toward lamination. The palate and the pterygoid region are more complicated, the zygomatic plate ex- tremely developed. Externally, sigmodonts, which have developed such well-defined vole-like, aquatic, and leporine types, may be nearing the end of their adaptive radiation. External, cranial and dental criteria for classification of the cricetines mentioned above are discussed under separate headings below. ADAPTIVE RADIATION The repeated establishment of meadows, prairies, tundras and coniferous forests in the wake of the several glacial retreats in the northern hemisphere provided the stages for explosive developments of pastoral murids from sylvan stock. With the close of the last glacial period, pastoral forms diversified into many types, each adapted to a special niche. Notwithstanding differences in general environment, diet and masticatory systems, adaptive radiation in pastoral species was remarkably similar to that of sylvan species where newly invaded habitats resembled sylvan habitats. In the case of Neotropical cricetines, the semiaquatic, pastoral marsh rat, Holochilus, is almost indistinguishable in external and cranial characters from its sylvan counterpart, the oryzomyine rat, Nectomys squamipes. Pastoral but semiarboreal Phyllotis griseoflavus mimics its sylvan opposite numbers among scansorial or more or less arboreal species of Thoma- somys, Peromyscus (Megadontomys) and Oryzomys (Oecomys) concolor. Not only do these species agree in body size, proportionate tail length and limb modifications; they have also developed a broad, divergent-sided supraorbital region. Truly arboreal pastoral forms comparable to sylvan Rhipidomys, Nyctomys and Tylomys do not exist. Among hystricomorph rodents, however, the grazing-brows- ing species of Dactylomyinae nest and feed in the evergreen pastures of bamboo grove canopies; presumably, the ancestral dactylomyines left the ground when their terrestrial habitat was subjected to pro- tracted floodings or was choked by superseding forests. The most 1 In the description of the cranial characters of the sigmodont rodents, the anterior border of the zygomatic plate is said to be "convex" (loc. cit., last line). The term should read "concave." See plate 29 of the cited work. 26 FIELD IANA: ZOOLOGY, VOLUME 46 common pastoral cricetine types are vole-like phyllotines, akodonts, etc. The sylvan counterpart is Oryzomys caliginosus. Perhaps be- cause of the abundance of moist, dark cover on forest floors or because of periodic flooding in many parts of their range, sylvan cricetines have not become specialized for subterranean life. At best they are weak burrowers. In contrast, the pastoral akodont Notiomys, with stout limbs, long claws, minute ears, and thick coat, is well adapted for a burrowing life. Pastoral Oxymycterus, Thap- tomys, and Bolomys are subfossorial, and Blarinomys is the extreme example of talpine form and habits among cricetines. The majority of phyllotines are generalized pastoral types. None is fossorial, although most phyllotines do occupy, whether for refuge or nesting, burrows made by other animals. Only the monotypic phyllotine Pseudoryzomys is sufficiently modified for aquatic life for comparison with such semiaquatic cricetines as Oryzomys palustris Zygodontomys, notably in its fan-shaped hind foot, also bears some resemblance to large palustrine types of oryzomyines. Zygodonto- mys, however, is basically vole-like and apt to be confused with the similarly vole-like Sigmodon, with which it shares habitats where the ranges of the two overlap. Indeed, the microtine or hamster-like form has been the principal evolutionary line of adaptation followed by phyllotines. Phyllotis sublimis, P. boliviensis, Galenomys garleppi and tiny Calomys lepidus, all of the Andean highlands, are short- tailed vole-like inhabitants of grass and rock communities. The last species superficially resembles species of Old World Cricetulus. The most remarkable hamster-like phyllotine is the large, heavy-bodied and variegated Chinchillula sahamae, of the same habitat. Eligmodontia typus has radiated in a direction unique among phyllotines and New World cricetines in general. It is remarkably gerbil-like in color, size, shape of body, specialization of hind legs and feet, and choice of habitat. The Piura desert mouse, Phyllotis gerbillus, is similar in size, color and habitat but its limbs are not gerbil-like. A third type of adaptive radiation among phyllotines is the semi- arboreal. Phyllotis griseoflavus has already been mentioned as the example. This species nests on and above the ground but is pri- marily a ground feeder. Ability to climb is in itself a generalized phyllotine character. The vaulted skull and proodont upper and procumbent lower incisors found in the vole-like phyllotine Galenomys garleppi of the high Andes are characters pointing to leporine habits. Similar char- HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 27 acters exist in the Arctic hare, Lepus arcticus. The habitats of both animals are also similar. JUVENAL CHARACTERS AND DEVELOPMENT In grassland cricetines the first post-natal pelage is essentially as in adults, though it is usually thinner, paler, and more grizzled on the upper parts and sides of the head and body. Molt from Juvenal pelage to typically adult pelage is generally unmarked by lines of transition and is completed during an early stage of development, possibly at weaning. Growth of tail, hind foot, and ear in young individuals is faster than growth of head and trunk. After sexual maturity, the extrem- ities have attained almost maximum size and their rate of growth declines sharply. Growth of head and body, on the other hand, con- tinues apace throughout life. Measurements reveal that in any one population extremities of juvenals and subadults are proportionately larger than in adults of the same population, and that differences in skull lengths between young and extremely old adults are greater than between young adults and weaned juvenals. Breeding begins at a growth stage when cranial characters would be subadult or even Juvenal. Sexual maturity precedes anatomical maturity in all mammals but pastoral mice are particularly pre- cocious breeders. The braincase is proportionately larger, the supraorbital region broader in juvenals. In some phyllotines, notably Phyllotis griseo- flavus, Calomys callosus and Zygodontomys, the broad Juvenal supra- orbitals have been retained in the adult and have developed beaded projecting ledges. The young of most phyllotines acquire a fully erupted and func- tional set of molars while still suckling. This phenomenon provides maximum molar surface in response to an early need for grinding siliceous grasses and other harsh, highly cellulose plants. The newly erupted hypsodont or subhypsodont molar is pyramidal in form. As a consequence the grinding surface of the tooth increases with wear concurrently with the growth of the mouse and its need for more food. Some dental elements, such as the mesoloph, normally absent in the adult molars of phyllotines, may be present, though weakly developed, in the newly erupted third molar. Conversely, dental characters of a caenogenetic nature may also appear in the unworn or unerupted third upper molar. Some such characters are ephe- 28 FIELDIANA: ZOOLOGY, VOLUME 46 meral and of no significance beyond the brief span of their existence. Others may have some selective value and may develop into impor- tant factors in the economy of the adult. The fully adult pelage and dentition acquired at an early age by juvenals of most phyllotines contrast sharply with the distinctively dark pelage and retarded last molar in juvenals of forest-dwelling cricetines with the primitive, or pentalophodont, type of molars. REPRODUCTION In Neotropical regions with well-defined seasonal fluctuations in temperature, breeding activity among cricetines is greatest in sum- mer (December-March) and least in winter (June-September). Breeding becomes more nearly uniform throughout the year in latitudes nearer the equator. Here, too, some periodicity may be maintained locally, but mating does not take place at the same time in all localities. In effect, reproduction, like molt, is uninterrupted in species occupying geographic ranges which embrace highly diversi- fied but locally constant climates. The condition of the mammae in skins of specimens examined indicates that from two to at least four litters are produced annually by the adult female of all species of Phyllotis. According to col- lectors' data, a litter contains from three to ten young. Litter sizes and reproductive rates in phyllotines are comparable to those of cricetines of the Temperate Zone of North America. Data on sexual behavior, compiled from the literature, are given in the accounts of individual species. POCKET POPULATIONS Small mammals of a sedentary nature, such as most phyllotines, are often distributed in pockets, when their population is at mean or low levels. A pocket is a small isolated ecological community in- habited by an inbreeding population, or colony, of animals. The pocket ordinarily contains all elements needed for maintaining its resident population and is usually sufficiently diversified ecologically to include a number of niches. A desert oasis and an alpine meadow are examples of pockets in so far as concerns the small mammals restricted to each. A bamboo grove or a wood-lot surrounded by grassland or swamp or, conversely, a pasture or swamp surrounded by forest, each with its special complement of resident animals, are also examples of pockets. HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 29 The actual size of a pocket is related to the size and mobility of its residents. A pocket in terms of a cricetine rodent is not a pocket in terms of a large ungulate or carnivore. On the other hand, micro- scopic organisms infesting a pocket of mice are also residents of pre- cisely the same pocket. Here the focus of infection and the pocket are identical and the parasites are subject to the same dynamics of the pocket as their hosts. The size of the population depends on the carrying capacity of the pocket. Usually, fluctuations in population are directly related to fluctuations in food supply. Rate of reproduction, predation, dis- ease, seasonal changes in climate and terrain are controlling factors. At times of extraordinary abundance of food the population of a pocket may grow to exceed by several times the ordinary carrying capacity of the range. With swift termination of the special food supply most of the population erupts into the inhospitable country- side in search of nourishment. Nearly all migrants die within days (see ratadas, p. 42). A few return to their original pocket. Others may breach barriers and find other pockets where they may either be absorbed by established colonies, if any, or start new colonies. The stability of a pocket population is relative. Its isolation may break down during population explosions or other crises. Floods not only have the power to destroy but may also be the agents for trans- porting individuals from one pocket to another or for seeding new pockets. Predators or disease may wipe out the population of a pocket and fires may destroy the pocket as well as its population. Pockets themselves may be early serai stages and each major change in suc- cession often involves the displacement or replacement of the earlier pocket population. Cultivation of land by man regularly results in the establishment of new pockets and the destruction or modification of old ones. As pockets expand their populations expand. Two or more pockets which grow and join to form one large subclimax or climax merge their populations into one race. Secondary fragmen- tation of the climax redivides the racial population into relict pocket populations. The pocket population is the basic stock for experiments in evo- lution and the pocket is the primary testing laboratory. The pop- ulation of a relict pocket is, perhaps, the most stable for experimental purposes and the relict pocket itself provides the most uniform selec- tive pressures for testing new characters or new combinations of char- acters. When the population is low and the reproductive rate high, the establishment of a trait throughout the pocket population can 30 FIELDIANA: ZOOLOGY, VOLUME 46 be rapid. When the population is at an eruptive peak, diffusion of the trait throughout the range of the race or species becomes possible. The rate of evolution is measured in generations, not calendar years. Pocket populations of small rodents such as phyllotines breed early, usually within one month of birth, and several times each year. The average life span of the individual, however, is less than 6 months. This combination of fecundity and longevity results in rapid popu- lation turnovers with a minimum of ancestral genetic drag. The diagnostic characters of old established neighboring pocket populations may be concordant where ecological conditions are sim- ilar, discordant where they are notably different, or clinal where they are varied or intermediate. Differences between allochronic but sympatric populations may be of the same order as differences be- tween synchronic but allopatric populations. Montane mammalian populations that occupy pockets at the highest altitudinal levels may be called pinnacle populations. They are derived from and may retain their connection with the mass of the species population at lower levels. Pinnacle populations of the same mountain system generally develop along parallel or convergent lines but the relationship between them is through the parental stock at lower levels. Where the parental stock has disappeared each pin- nacle population becomes a relict evolving along independent lines. Oasis populations are usually of waif or accidental origin and their superficial characters are generally concordant, their under- lying ones usually discordant. Peripheral pocket populations inhabit the outer borders of the range of the species or subspecies. The diagnostic characters of these populations are apt to be clinal, with each population, in the case of subspecies, interbreeding almost or quite continuously with the neigh- boring populations of the same or related subspecies. A species or subspecies generally includes a number of discordant as well as concordant and clinal pocket populations. There are, nevertheless, valid subspecies and even species (cf . Phyllotis gerbillus) with their respective geographic ranges restricted to a single pocket. Basic data for these generalizations on pocket populations are contained in the following accounts of variation between the sub- species of Phyllotis darwini, P. osilae and others. The habitats of local populations of Phyllotis described by Pearson (particularly in his section on sympatry (1958, pp. 397-405) are, in many instances, examples of pockets. HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 31 SYMPATHY AND ALLOPATHY The practical concept of sympatric is taxonomic, not biogeo- graphic. The word, however, can refer to any and all organisms occupying a common geographic area. In order to conserve and em- phasize the taxonomic sense of sympatric, a restricted form of the term must be used for representatives of sibling species or different subspecies of the same species whose trails cross during the course of their ordinary movements. A more precise term for this form of sympatric relationship is synecetic, proposed by Harper, et al. (1961, p. 211). Thus, all synecetic representatives of species or subspecies are sympatric but all sympatric representatives of species or subspe- cies are not synecetic. Two mammalian subspecies with ranges which interdigitate but do not overlap may be regarded as sympatric but cannot be classified as synecetic. The geographic ranges of fossorial, terrestrial, aquatic and arboreal species of mammals may all over- lap)— actually they would form a mosaic — but if the animals them- selves do not ordinarily cross trails they are not synecetic. Sympatry, when it occurs, follows divergence of the species or subspecies in geographic isolation. Sympatry may endure for the life of all but one of the coexisting taxons, or it may be cyclical or ephemeral. The spatial ebb and flow of otherwise geographically contiguous or narrowly separated taxons may result in periodic or rare overlappings in range. Migratory species coexisting transitorily with closely related forms are not true sympatriants. A misleading simulation of sympatry may result from the mingling, during an eruption, of two or more well-defined and erstwhile geographically isolated pocket populations. Sympatric taxons may or may not be reproductively isolated but synecetic taxons are necessarily so. Sympatry, or syneceticity between two or more taxons, is generally regarded as prima facie evidence for their specific separation. There are, however, notable ex- ceptions among mammals, including the rodents under consideration. Allopatric, the antonym of sympatric, could be correspondingly restricted to geographically contiguous sibling (cognate) species or subspecies. Use of the term for widely separated taxons of any grade dilutes or nullifies its special taxonomic implications. The intimately related and similar-appearing members of the Phyllotis darwini complex offer prime examples of the significance of sympatry and allopatry in taxonomy. All other phyllotines can be classified on the basis of morphology alone. 32 FIELDIANA: ZOOLOGY, VOLUME 46 The northernmost member of the Phyllotis darwini group is Phyl- lotis haggardi of the Ecuadorian Andes. It is treated as a full species on the basis of seemingly trivial yet consistent characters which dis- tinguish it from its nearest geographic ally, Phyllotis andium. Pres- ent data indicate that haggardi and andium are completely allo- patric (fig. 72), but the nearer they are geographically the more similar they appear to be. Whether the cline is a product of con- vergence or of intergradation is a question that cannot be answered conclusively without the test of sympatry. In contrast with the above, specific separation of Phyllotis andium from P. darwini depends entirely on evidence of sympatry. The former is recorded as coexisting in central Peru with three subspecies of the latter. The differences between P. andium and sympatric populations of P. darwini limatus, and sympatric and at least one synecetic population of P. d. posticalis are not impressive. They may be greater than between some allopatric populations of the same taxons including P. andium and less than others including P. andium. On the other hand, P. andium differs grossly from P. darwini defi- nitus with which it coexists in the only locality where the latter is known to occur. Whatever the nature of the differences, where an- dium and darwini are sympatric they do not intergrade and where they are allopatric they cannot interbreed. Two subspecies of the same species, P. darwini rupestris and P. d. magister, coexist in southwestern Peru without intergradation (fig. 73). Unlike the preceding example, where these races are allopatric they can and do intergrade through an intermediate sub- species. The larger race, Phyllotis darwini magister, is a poorly de- fined, questionably separable western Andean extension of P. darwini posticalis. P. d. rupestris also seems to stem from posticalis stock but it has spread southward along the eastern slope of the moun- tains and has become significantly smaller and paler. Where rupestris meets magister, differentiation of the former already has reached the point of reproductive isolation from the latter. Inasmuch as both rupestris and magister grade into posticalis, all are regarded as con- specific. On the opposite side of the Andes in southern Bolivia and north- ern Argentina, Phyllotis darwini rupestris and P. d. caprinus are fully intergrading, though their ranges may overlap at some points. The available evidence, meager and unverified as it is, indicates that the two races are not synecetic (figs. 74, 77). HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 33 The most interesting and perplexing case of sympatry among phyllotines is that between Phyllotis darwini and P. osilae. The range of osilae extends as a narrow band along the eastern Andes from central Cusco, Peru, in the north to southern Catamarca, Argentina, in the south (fig. 74). Representatives of osilae and darwini are recorded as sympatric in 25 localities distributed over the entire length of this range. Two subspecies of P. osilae and four of P. darwini are involved. Whether or not the two species are actually synecetic in any one locality is moot. Pearson (1958, p. 399) believes that "sympatry of these morphologically similar forms is based upon different habitat preferences."1 If this is true, the case for regarding osilae and darwini as distinct species rather than sympatric subspecies is considerably weakened. The greatest number of sympatric series are concentrated in the Lake Titicaca drainage basin in Puno, Peru. They represent sub- species Phyllotis osilae osilae and P. darwini rupestris. The number of specimens in the sympatric pairs ranges from one osilae and two rupestris from Puno, Puno, to eighteen osilae and three rupestris from Pairumani. The best numerical representation of the two species is the series from Chucuito, which consists of five osilae and eight rupestris. In no case are the sympatric series large enough to use for calculating the full range of their respective variation. The dif- ferences between the species are subtle, and the larger the sympatric series of osilae and rupestris the fewer and narrower the differences become. In any case, sympatric populations of osilae and rupestris are more nearly alike than allopatric members of the same races. Differences between P. osilae osilae and P. darwini posticalis re- corded from three sympatric localities in Cusco, Peru, are fine enough to be overlooked or dismissed as insignificant by the unwary taxono- mist. Differences between allopatric populations of these subspecies are of the same order or greater, never less. P. osilae pkaeus is easily distinguished from a peripheral series of P. darwini posticalis and the differences between allopatric popula- tions may be greater or less. 1 This statement appears to be based on impressions of random trapping results rather than on a scientifically conducted census. Pearson does not relate his catch to the number of trap nights or supply information regarding his trapping methods. The reader is left to believe that other conclusions might be reached by other col- lecting methods or by other collectors. Judged by data on the labels of his speci- mens, Pearson's trapping consisted of one or two night stands in each of the vast majority of localities worked. The trapping sites were generally along highways or railways in areas with ecologies repeatedly disturbed and constantly controlled by man during one or two millenniums. 34 FIELDIANA: ZOOLOGY, VOLUME 46 The third concentration of sympatric localities is in central Bo- livia, where the range of the localized Phyttotis darwini wolffsohni sprawls over a section of the range of P. o. osilae. The sympatric populations of the two forms are similar superficially but differ in size and in cranial and dental characters. The differences between the two species here are greater than elsewhere irrespective of the races compared or their geographic relationship. Nevertheless, the tendency toward convergence in sympatry is so strong among phyllo- tines that P. darwini wolffsohni more nearly resembles sympatric populations of the species P. osilae than it does the nearest geo- graphic representatives of the subspecies Phyllotis darwini rupestris with which it intergrades through P. d. caprinus. Not one of the characters that serve to separate sympatric repre- sentatives of osilae from darwini is valid for distinguishing all osilae from all darwini. The differences between sympatric populations are of the same order as differences between pocket populations and sub- species of either species. There is no evidence of intergradation be- tween osilae and darwini where they are sympatric, and there is no evidence that a population of one species of a sympatric pair inter- grades with an allopatric (i.e., contiguous) population of the other species (cf. Tables 30-49). Where sympatric osilae and darwini are most nearly alike, their resemblance suggests either a recent and comparatively slight diver- gence from a common ancestor or a convergence in sympatry after initial differentiation in geographic isolation. Absence of inter- grades or hybrids points to the latter explanation as more plausible if not most probable. Representatives of the two species that differ most in sympatry occur at the far extremes of their respective geographic ranges. Their greater divergence involved movements over greater distances through more diversified environments during longer periods of time. Nevertheless, where they meet, the species tend to converge in super- ficial characters at least. All the foregoing runs counter to character displacement defined by Brown and Wilson (1956, p. 63) as "the situation in which, when two species of animals overlap geographically the differences between them are accentuated in the zone of sympatry and weakened or lost entirely in the parts of their ranges outside this zone. The charac- ters involved in this dual divergence-convergence pattern may be morphological, ecological, behavioral, or physiological. Character HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 35 displacement probably results most commonly from the first post- isolation contact of two newly evolved cognate species. Upon meet- ing, the two populations interact through genetic reinforcement of species barriers and /or ecological displacement in such a way as to diverge farther from one another where they occur together. Ex- amples of the phenomenon, both verified and probable, are cited for diverse animal groups [none mammalian] illustrating the various aspects that may be assumed by the pattern." Divergence in allopatry, i.e., in different but contiguous ecologies, and convergence in sympatry, i.e., in the same broad ecology, are an evolutionary pattern quite the reverse of "character displacement." This pattern is not peculiar to members of the Phyllotis darwini com- plex. It is common among closely related mammals of all orders. As a rule, convergent characters in sympatry are superficial, usually adaptive, and superposed on the valid diagnostic characters of the species. They are, in effect, similar, if not identical, responses of closely related species to the same environment. On the other hand, truly distinctive specific characters are those which evolve in allo- patry as unique responses of the organism to its peculiar environment. The degree of divergence depends on the period and circumstances of geographic isolation and not at all on a mechanism which is trig- gered by the first post-isolation contact as suggested by Brown and Wilson. Divergent populations which make contact on common ground after reproductive isolation has been attained, may continue to diverge, and they probably do in some respects. The more im- portant and stronger tendency, however, is toward convergence.1 The basic specific characters, which evolved in allopatry, become stabilized in sympatry. The great amount of variation and the instability of the charac- ters or combination of characters distinguishing sympatric popula- tions of osilae and darwini suggest that the species themselves are not stable. It is not improbable that future collecting will reveal the existence of one or more intergrading populations. Meanwhile, Phyllotis osilae is regarded as something less than a full species but more than the conventional subspecies; hence the binomial. Its posi- tion as an infraspecific category, however, coordinate with the com- 1 This statement is not to be interpreted as a denial of the fact that many sib- ling species do differ most in sympatry. I myself (1960, p. 529 and pi. 5) pointed out that sympatric representatives of the cricetines Oryzomys (Oecomys) concolor and O. (O.) bicolor are easily recognized, but that some allopatric individuals of the same species may be indistinguishable. Additional examples among mammals can be cited, but nothing known of the evolution of these paired species supports the concept of "character displacement." 36 FIELDIANA: ZOOLOGY, VOLUME 46 parably ill-defined Phyllotis andium and P. haggardi, is indicated by inclusion of osilae in the Phyllotis darwini complex. Sympatry also occurs between Phyllotis darwini and its most nearly related congeners outside the complex. The range of the western Peruvian Phyllotis amicus overlaps that of P. darwini pos- ticalis, but nothing is known of the habits of the latter. The mor- phology of P. darwini and P. griseoflavus suggests some ecological segregation where they are sympatric. The two species, however, are evidently synecetic, and individuals of both have even been found in the same burrow. P. darwini has also been found in common col- lecting localities with the Patagonian P. micropus but it is assumed this occurs where man has altered the habitat of the second to per- mit invasion by the first. Calomys, the only other undoubted polytypic phyllotine genus, comprises four species. Where their ranges overlap, C. laucha is the pampa mouse, while the larger C. callosus is restricted to gallery and scrub forests. Calomys sorellus and the smaller, shorter-tailed C. lepidus have been taken in the same localities in Peru. It is virtually certain that they, too, are ecologically segregated. CROP VARIATION No two generations of a population develop in the same way and at the same rate. Most common differences between generations, or crops, are in size, weight, pelage, color, reproductive rate, and population density. Such differences between generations, or sea- sonal crops, of the same population are expressions of crop variation. The young of grazing cricetine rodents, including phyllotines, acquire adult characters and habits precociously as compared with the young of forms that undergo a prolonged development in the comparatively stable environment of maternal nest, den or burrow. Phyllotines born during the dry season are promptly exposed to rigorous living conditions. Available food may be barely sufficient to supply the young animal with energy needed for foraging and for maintaining body weight commensurate with increasing size. On the other hand, phyllotines born in the rainy season may be literally bedded in an abundance of tender shoots of succulent plants. The quantity and quality of such food, combined with the ease with which it is taken and eaten by the mouse, must contribute to development of potential maximum in size of skeletal and dental parts. Thus, an individual born and developed during the fruitful rainy season may HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 37 be larger, and, because of less molar abrasion, erroneously adjudged younger, than a comparably aged product of the barren dry season. When phyllotines of one local crop survive several seasons no doubt they tend to appear like old adults of other crops of the same locality. Few small rodents, however, live through the combined life spans of two or more successive crops. The average life span of cricetines in nature is less than six months. Blair (1948) calculated the longevity of wild living individuals of comparable Michigan Peromyscus maniculatus, P. leucopus, and Microtus pennsylvanicus, as varying from four to five months. Snyder (1956, p. 27), using finer methods, determined the average life span of Peromyscus leucopus as five months. Hence, a phyllotine may be born, reach sexual maturity, reproduce and die during the three to six months duration of a normal dry or rainy season anywhere within its range. No field studies of crop variation in phyllotines have been made. However, results of an 11-year investigation of cyclical fluctuations in the cotton rat (Sigmodon hispidus) population of a fallow field in Georgia, are pertinent. The investigator, Odum (1955) found that "cotton rats appear on the runways when a week or less old . . . and begin to breed when two months or less in age. . . . Since few indi- viduals may be expected to live more than six months, the population turnover may often be complete every six months. With a species such as the cotton rat which begins to breed at an early age and rarely lives long enough to reach maximum size it is clear that average measurements of 'adults' will vary according to the stage in the population oscillation. . . . Therefore, size should not be used as a criterion for the establishment of geographic races or subspecies unless the sample is very large and the stage in the population cycle from which it was taken is known." A study by Martin (1956, p. 388) of a Kansas population of the prairie vole (Microtus ochrogaster Wagner) revealed that during an unusually dry summer reproduction was inhibited, adult voles lost weight, and young born in the spring ceased to grow. Fitch and Rainey (1956, p. 528) conducted a similar study of the eastern woodrat (Neotoma floridana Ord) and found that "growth rate and adult weight are influenced to a large extent by season." They did not overlook the importance of individual differences as controlling factors. In the works cited above, organic variations are correlated with population cycles, and with seasonal changes in the quantity and quality of the food supply, temperature, and rainfall. The impor- 38 FIELDIANA: ZOOLOGY, VOLUME 46 tance of food alone among wild brown rats (Rattus norvegicus Berkenhout) was determined experimentally by Calhoun (1949, pp. 1113-1122). The rats were kept in a specially designed pen. Those born near the abundant and continuously maintained food supply grew more rapidly, attained larger size, and were higher in the social order than those born farther away. Other field studies of less nearly related mammals support the concept of crop variation. Scheffer (1955, pp. 493-513) was pri- marily concerned with the relationship between body size and nu- merical abundance in seals. He also reviewed some of the literature regarding the same relationship in the brown rat, house mouse, vole, muskrat, gray squirrel, raccoon, deer, and man. As a corollary, Scheffer noted (op. cit., p. 513) that "body size is, of course, not only a genetic product but a quantitative measure of environmental response. When certain body characters vary significantly in size from those of the generation closely removed, it may be argued that here are members of two 'races' separated only by a thin partition of time. The argument is unrealistic but it does make a point: where slight differences are critical, the systematist should choose samples for comparative study taken from similar population density environments. This precaution extends the meaning of Cazier and Bacon (1949 [Bull. Amer. Mus. Nat. Hist., 93: 343-388]): The specimens should be from similar climatic, edaphic, and biotic situations . . . taken within a limited range of time so as to preclude possible confusion resulting from contamination by seasonal vari- ance.' ' Here I must interpose that the systematic mammalogist can deal only with the material at his disposal and has little or no knowledge of particular habitats or population densities. He need only learn to discount crop and other levels of intraspecific variation in evaluating the taxonomic status of the sample. Environmental factors that combine to modify each crop are separated in time from those modifying the next crop. Modifications within a crop produced by environmental factors separated in space are discussed under the next heading. NICHE VARIATION Differences between individuals living in one section of the range of a population and individuals of the same species inhabiting another section of the same range are expressions of what is here termed niche variation. Calhoun's experiment with brown rats, cited above as an illustration of crop variation, is also a valid demon- HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 39 stration of one kind of niche variation. Likewise, individuals or small colonies of phyllotines living nearest the best food supply can reasonably be expected to be better developed than their neighbors not as favorably stationed. Certain phyllotines, notably Phyllotis darwini and Calomys laucha, reside in human dwellings, storehouses, well-watered gardens, refuse dumps and similar habitats with a suitable and permanent food supply. Inhabitants of such special niches can acquire somatic characters that distinguish them from members of the same population, or crop, living in natural and less favored surroundings of the immediate vicinity. Some of the strik- ing morphological differences distinguishing neighboring colonies of tuco tucos (Ctenomys) and similar burrowing rodents (Thomomys, Tachyorycles, etc.) may be no more than manifestations of niche variation. An investigation of house mice (Mus musculus} inhabiting four different environments revealed manifestations of niche variation. It was shown by Laurie (1946) that the mice in cold stores where the temperature never exceeded 15° F. lived normally, bred and produced regularly the year round, had more embryos, and were larger and heavier than mice of the other three habitats. The cold store mice ate meat exclusively. Mice of corn ricks had the highest rate of productivity, while those of urban habitats had the lowest. Mice living in flour depots were average producers. Their food consisted of white flour only. This diet may have been the cause of a skin condition characterized by thin pelage, bald patches, and torn and crumpled ears. A minority of the mice was affected and most of them lived in the same depot. The mice of the four environ- ments were taken from a number of places in London, Oxford and other localities. For all taxonomic purposes, however, the environ- mentally different habitats could have been adjacent rooms in a building in Oxford where most of the urban rats were trapped. Striking differences in color may also be manifestations of niche variation. The unusually pale color in two species of bats, Tadarida brasiliensis and Myotis velifer, living in certain Texas caves is the result, according to Constantine (1957, p. 465), of the bleaching effect of environmental ammonia gas and water vapor. Later Constantine (1958, p. 513) confirmed his conclusions by reproducing experimen- tally the bleaching effect of ammonia and high humidity on live Tadarida. The distinction between crop and niche variation is blurred or erased in nearly all experiments in which attempts are made to deter- 40 FIELDIANA: ZOOLOGY, VOLUME 46 mine the effects of environment on organic growth and development. Nevertheless, such experiments often produce significant differences between the test animals and sometimes suggest the kinds of differ- ences which might exist in nature between groups of individuals of the same population separated in time or space. In one experiment, Ogle (1934a, p. 628) found that female white laboratory mice sub- jected to a warm (31°-33° C.), humid (75%) environment had fewer fertile matings, smaller litters and less viable offspring than mice of the same strain kept in a uniformly cool (17.8° C.) environment. The males (Ogle, 1934b, p. 635) in the warm moist room developed very short and slender bodies but the tails of each were 1 cm. longer than combined head and body length. Males acclimated in the cool room increased in body weight and length at practically the same rate as the control mice kept at 21°-27° C. Their tails, however, were 1 cm. shorter than combined head and body length. In a sim- ilar experiment by Biggers, Ashoub, McLaren and Michie (1958, p. 144), who used different hot and cold environmental temperatures, the results were generally the reverse. Females in hot (28.09° C., 87% humidity) and cold (4.78° C., 77.48% humidity) environments had a higher mortality of prenatal and postnatal young than females in the temperate or control (21° C., 69.03% humidity) environment. Furthermore, "no significant differences in rate of growth or develop- ment were found between the mice in the hot and temperate environ- ments. Both growth and development were markedly retarded in the cold." Changes in the gross appearance and chemical composition of the teeth of rats may also be induced by chronic exposure to adverse en- vironments. Harris, Mefferd and Restivo (1960, p. 476) showed that the incisors in rats acclimated for 3 to 6 months at a temperature of 36° C. usually acquired a heavy reddish-brown deposit. The incisors in rats maintained at a high altitude (380 mm. Hg.) remained normal. Food intake, hence masticatory activity, was approximately the same in the heat and altitude groups. Rats exposed to a combina- tion of heat and high altitude revealed significantly reduced concen- trations of calcium, phosphate and magnesium in their incisors. As a final example, Harrison (1958, p. 892) investigated such characters as longer tails, larger feet, thinner pelage, smaller body size and general acclimatization of mice (Mus musculus) subjected to a hot environment by a number of experimenters including Ogle (1934b, p. 635), mentioned above. Harrison concluded (p. 900) "that at least some of the changes, both physiological and morphological, HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 41 which occur when mice are reared at high temperatures, are in their over-all effect adaptive." Trapping of mice for taxonomic studies is almost always limited in time and restricted in place. The sampling of any species thus secured is often representative of only one crop and /or niche. Some- times a few survivors of an older crop, or strays from a different niche, may be included in the sampling. Such individuals help form a more nearly exact concept of variation within the species. Occa- sionally crop and niche variants are mistakenly adjudged distinct species or perhaps sympatric(!) subspecies. SOCIABILITY Most phyllotines are social or gregarious and many are commen- sal as well. Darwin (in Waterhouse, 1839, p. 44) reported that six specimens of the type series of Hesperomys bimaculatus (= Calomys laucha) were discovered in the same burrow. The social and gre- garious habits of Phyllotis sublimis were observed by Pearson (1951, p. 147), and Thomas (1900, p. 467) noted that the type and eight topotypes were dug out of one hole. Because of their preeminently grazing habits, populations of species of Phyllotis, Andinomys edax, Chinchillula sahamae and, possibly, Galenomys garleppi, tend to con- centrate in the midst of localized stands of the more succulent grasses. In some cases, sociability develops into colonialism through commensal ism. Small populations of Phyllotis darwini become inte- grated with colonies of tuco tucos (Ctenomys) or cavies (Cavia, Galea, and others) by sharing their living quarters and feeding grounds. In- dividuals of Phyllotis darwini have also been found in the same bur- rows with Octodon degus, 0. bridgesi, Abrocoma bennetti, Akodon oliva- ceus, Abrothrix longipilis, Oryzomys longicaudatus, Rattus rattus, and Marmosa elegans (fide Wolffsohn in Thomas, 1927, p. 556). Pearson (loc. cit.) noted Phyllotis boliviensis living in a viscacha colony and P. sublimis sharing homes with nearly every other cricetine of the same region in the Peruvian altiplano. De la Barrera (1940, p. 572) records Phyllotis griseoflavus and P. darwini each living together with Microcavia australis in burrows of the latter in Mendoza, Argentina. Sociability in phyllotines is not restricted to natural habitats. Phyllotis darwini, P. griseoflavus, Calomys laucha and C. callosus are common tenants of man's habitations, where they compete success- fully with the introduced commensals Rattus and Mus. 42 FIELDIANA: ZOOLOGY, VOLUME 46 Nothing is known of the habits of the highly localized Piura desert mouse Phyllotis gerbillus. The ease with which the collector Celestino Kalinowski secured a series of 17 topotypes suggests, how- ever, that this species is gregarious, like most of its relatives. On the other hand, there is nothing in the records of Eligmodontia typus to indicate that this gerbil-like mouse is anything but solitary out- side the breeding periods. RAT AD AS AND PLANT FRUITING CYCLES Devastations of cultivated fields and store rooms and invasions of human habitations and villages by inordinately large numbers of hungry rodents are termed ratadas in most parts of rural South America. The expression rat plague refers to the same phenomenon but has been used interchangeably with the differently meaning bubonic plague. Ratadas are commonplace in South America and a number of them have been recorded. Only exceptionally were the rodents involved identified. It is certain, however, that phyllotines, notably the ubiquitous species of Phyllotis and Calomys, are among the anonymous "mice" of many ratadas mentioned in the literature. Ratadas originate in restricted localities when an extraordinary surplus of natural food, usually the fruit of one or two kinds of plants, occurs together with ideal cover and nesting sites for mice. In this environment multiplication of the population of one or a few species of mice assumes explosive proportions, and the number of individuals soon exceeds by many times the ordinary carrying capacity of the range. With termination of the fruiting season and deterioration of protective cover, the hungry, exposed rodents break out into the surrounding countryside. By the time a new balance is reached between the individual and its range, there has been a precipitous drop in the rodent population and considerable destruction of the cultivated crops in the overrun area. The many dead mice often observed after a ratada peak are the victims of exposure, hunger, thirst, abnormally vicious inter- and intra-specific competition, cannibalism, and wholesale killing by man and predator. Some- times disease strikes the rodents with catastrophic effects. Another cause of high mortality may be the radical change in diet for the generations of mice that developed chiefly, if not entirely, after weaning, on one plant food. In many parts of South America ratadas are correlated with the fruiting of diverse species of bamboo. One of the earliest recorded HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 43 took place in Valdivia, Chile, in 1780. This ratada is cited by Philippi and Landbeck (1858, pp. 81-82) from a manuscript written by the historian and naturalist Claudio Gay. The manuscript adds that periodic plagues of pericotes (native name for small rodents in general, and often for Phyllotis darwini and Rattus in particular) scourged all parts of Chile. The ratadas were known to occur in cycles of 17 to 20 years that coincided with the cyclic fruiting and dying of the "coligue," a bamboo used by the Indians for lance shafts. Philippi and Landbeck (op. cit.) also mention a sudden out- break of rats in the Araucanian districts of southern Chile, in 1681. No explanation is given for the ratada and no source for the reference. However, in 1869 or 1870, Philippi (1879) observed a ratada in Valdivia, where he was in residence. On that occasion all the "coligue" bamboos of the province produced seed at the same time and died immediately thereafter. Philippi noted also that ratadas occur "sometimes in the south of Chile, Araucania, Valdivia, and Llanquihue, when the coligue and other species of Bambuseae have flourished and fructified, an occurrence which happens every 15-20 years." The common native mice that might have participated in these ratadas are species of Phyllotis, Akodon and Oryzomys nigripes (= longicaudatus) . According to Schneider (1946, p. 77), Oryzomys nigripes was probably the species which devastated the fields of Con- cepcion Province, Chile, during the notorious ratada of July, 1877. The best account on the subject of the correlation between ratadas and the fruiting season of certain species of bamboos was provided by Orville A. Derby (1879, p. 65): "From time to time in all parts of Brazil the plantations are subject to the depredations of armies of rats that issue from the forests and consume everything edible that comes in their way. During a recent excursion in the province of Parana Mr. Derby found an almost universal lack of corn throughout the province, due to such invasion of rats, by which almost the entire crop of last year had been destroyed. This in- vasion, or plague as it is called, is said to occur at intervals of about thirty years, and to be simultaneous with the drying of the taquara, or bamboo, which everywhere abounds in the Brazilian forests. The popular explanation is that every cane of bamboo sprouts with a grub, the germ of a rat, within it, and that when the bamboo ripens and dies the germ becomes a fully-developed rat and comes out to prey on the plantations. An educated and observant Englishman, Mr. Herbert H. Mercer, who has resided a number of years in the province and had an opportunity of studying the phenomenon, 44 FIELDIANA: ZOOLOGY, VOLUME 46 furnished Mr. Derby the following rational and curious explanation : The bamboo arrives at maturity, flowers and seeds at intervals of several years, which doubtless vary with the different species. The period for the species most abundant in Parana is thirty years. The process, instead of being simultaneous, occupies about five years, a few of the canes going to seed the first year, an increased number the second, and so on progressively, till finally the remaining and larger portion of the canes seed at the same time. Each cane bears about a peck of edible seed, resembling rice, which is very fat and nourishing, and is often eaten by the Indians. The quantity produced is enormous, and large areas are often covered to a depth of five or six inches. After seeding the cane dies, breaks off at the root and falls to the ground, the process of decay being hastened by the borings of larva which live upon the bamboo and appear to be particularly abundant at seeding time. These larva have doubtless given rise to the story of the grub developing into a rat. New canes spring up from the seed, but require seven or eight years to become fit for use, and thirty to reach maturity. With this sudden and constantly increasing supply of nourishing food for a period of five years, the rats and mice, both of native and imported species, in- crease extraordinarily in numbers. The fecundity of these animals is well known, and the result after four or five years of an unusual and constantly increasing supply of excellent food and in the absence of enemies of equal fecundity, can readily be imagined. The last of the crop of seed being mature and fallen to the ground, the first rain causes it to decay in the space of a very few days. The rats, suddenly deprived of food, commence to migrate, invading the plantations and houses and consuming everything that does not hap- pen to be repugnant to the not very fastidious palate of a famishing rodent. If this happens at the time of corn planting, the seed is consumed as fast as it can be put into the ground. Mr. Mercer, who plants annually about fifty acres of corn, replanted six times last year, and finally gave up in despair. The mandioca is dug up; the rice crop, if it happens to be newly sown or in seed, is consumed, as is also everything in the houses in the way of provisions and leather, if not carefully guarded in tin trunks." The 30-year blooming cycle of the taquara and its correlation with ratadas was corroborated by Pereira (1941). The plant was identified as the taquara lixo (rough bamboo, Merostachys fistulosa Doell). Pereira also reviewed part of the history of ratadas in southern Brazil, beginning with a citation of observations made by HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 45 the French botanist, August St. Hilaire, in 1832 (Voyage dans 1'interieur du Bre"sil). Giovannani, Vellozo and Kubiak (1946) continued the investigation of ratadas in Parand. This time the outbreak was correlated with the fruiting of the taquara lisa (smooth bamboo, Merostachys sp.). Here, again, exhaustion of their food supply in the bamboo groves obliged the mice to erupt into the neighboring cultivated fields. The species involved in this ratada included the phyllotine Calomys laucha tener Wagner, and Oryzomys nigripes Olfers (Oryzomys eliurus Wagner), Holochilus brasiliensis leucogaster Brandt and Akodon nigrita Lichtenstein (the rato pitoco). Crespo (1944) studied a ratada of Calomys laucha laucha Olfers and Akodon benefactus Thomas that developed in fallow fields on the outskirts of Buenos Aires, Argentina. In this case, prolonged warm weather and heavy rainfall stimulated superproduction of natural food and heavy growth of cover favored by the mice. An abrupt change to abnormally dry, cold weather destroyed forage and cover, and obliged the rodents to invade adjoining cultivated fields. A ratada in south coastal Peru was investigated by Gilmore (1947). Though the affected area lies within the range of Phyllotis darwini, Gilmore found that only the rata-muca (Oryzomys xanthae- olus ica Osgood) was involved. The cause for the outbreak was not determined. I have found the phyllotine Zygodontomys brevicauda in near plague numbers in some northern Colombian localities. I also en- countered other South American rodents, notably the forest dwelling Oryzomys laticeps and Oryzomys caliginosus, some species of Thoma- somys, the spiny rat Proechimys guyannensis, and the grassland Heteromys anomalus and Sigmodon hispidus, in excessive numbers. The history of ratadas of marsh rats (Holochilus) was recorded else- where (Hershkovitz, 1955, p. 655). The meager data in literature, and personal observations in the field, indicate that fluctuations in populations may be as drastic in tropical and south temperate latitudes as in Arctic and north tem- perate ones. Ratadas in the Neotropical region are usually composed of one, sometimes two, rarely more, species of the local rodent fauna. Environmental conditions that promote the production of a ratada of one species may not alter significantly the population of another species living in the same area. The one or two species least affected by adverse conditions in a given locality are most likely to become plagues during optimum conditions. Species able to thrive in newly invaded territory are also notoriously ratada-prone. 46 FIELDIANA: ZOOLOGY, VOLUME 46 HIBERNATION Hibernation, or aestivation, as a regular seasonal function, has not been positively demonstrated in any South American mammal. There are suggestions, however, that sloths and certain small Patago- nian marsupials, bats, a marmoset1 and rodents may become torpid at irregular intervals. The mouse opossums, Marmosa pusilla (=ele- gans) and Dromiciops australis, store fat in their bodies, notably at the base of the tail, in advance of winter and become lethargic during particularly cold spells. Some phyllotines are also prone to torpidity when the outside temperature is low. One of the lauchas secured by Thomas (1916b, p. 184) in Buenos Aires province during the winter month of June, 1896, is said to have been "dug up, semi-torpid, in very cold weather, from about 6 inches below the surface of the ground." The species (described as Hesperomys murillus, a synonym of Calomys laucha laucha) is not a burrower and Thomas does not explain its presence underground or state that he himself found it there. Henry Durnford (in Thomas, 1898a, p. 210) observed that whereas Phyllotis griseoflavus was excessively numerous and overran the Colony in Chubut, Argentina, during the summer, it disappeared in the wintertime. Durnford believed that the animal became dor- mant. Pearson (1951, p. 147) also puzzled over the sudden disap- pearance of Phyllotis sublimis Thomas from the Peruvian altiplano during the rainy season months of October, November and Decem- ber. However, in regions where the most drastic environmental change is in rainfall, not temperature, absence of trap or sight records need not be the result of hibernation. A lush growth of food cover plants during the rainy season makes mice sedentary, less exposed to predators and less, if at all, trap prone (see also discussion of Ratadas, p. 42). ENEMIES Phyllotines, together with Akodon (sens, lat.), are the dominant mammals of the grasslands, scrublands and deserts of South Amer- ica. As such they are basic food for the carnivores living in the same range. The most important predators include the weasel (Mustela frenata), the huron (Grisonella) , species of foxes and wild cats, owls and hawks. In tropical savannas inhabited by Zygodontomys, snakes feed largely on small rodents. 1 Callithrix jacchus, in captivity (Kraft, 1957, Saugeth. Mitt., 5: 175). HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 47 Precocious development and the short suckling period of young phyllotines help reduce loss of population to nest-raiding predators. ECTOPARASITES The distributional patterns of ectoparasites often provide im- portant clues to the origin and phylogeny of their mammalian hosts. Modern parasitologists explore this relationship, and notable contri- butions on the subject have already been made by many investi- gators. In addition to those cited below, the works of Hopkins and Rothschild (1953-56) and Johnson (1957) on fleas (Siphonaptera) may be consulted to advantage. Ferris and Stojanovich (1951) mono- graphed sucking lice (Anoplura) and Vanzolini and Guimaraes (1955) have discussed the geographic distribution and host associations from a historical point of view. Few studies have been made of the ecological relationship be- tween disease-transmitting ectoparasites and Neotropical rodents. The most outstanding are those of de la Barrera. One of his works on sylvatic plague and life histories of rodent vectors is cited below; others are mentioned in the present account of Phyllotis griseoflavus (pp. 441-461). A list of some of the fleas, lice, mites, ticks and staphylinid beetles infesting phyllotines is given herewith. The authorities for published records are shown in parentheses following the name of the parasite. Heretofore unpublished records of specimens in Chicago Natural History Museum are distinguished by the symbol CNHM. The identifications were made by Dr. Rupert Wenzel, Curator of Insects of this institution, who also checked the nomenclature of all parasites listed. An interrogation (?) sign before the name of a para- site means that the host association is doubtful. Host names given below are those recognized in the revision. For concordance between host names originally used by the para- sitologists cited and those adopted here, see the synonymies and geographical distributions in the text that follows. GALOMYS Section Calomys sorellus Thomas Tick Ixodes (Eropalpiger) andinus Kohls (CNHM) Calomys laucha Olfers Fleas 48 FIELDIANA: ZOOLOGY, VOLUME 46 Polygenis rimatus Jordan (Costa Lima and Hathaway, 1946) Craneopsylla minerva wolffhuegeli Rothschild (Hopkins and Rothschild, 1956) Calomys lepidus lepidus Thomas Fleas (l)Plocopsylla pallas Rothschild (Hopkins and Rothschild, 1956) (l)Tetrapsyllus bleptus Jordan and Rothschild (Traub, 1952) (l)Neotyphloceras crassispina crassispina (CNHM) (l)Agastopsylla nylota (CNHM) C!)Sphinctopsylla inca Rothschild (Hopkins and Rothschild, 1956) (l)Dysmicus danger Rothschild (Johnson, 1957) Calomys callosus Rengger Lice Hoplopleura hesperomydis Osborn (Hopkins, 1949; Ferris, 1919-35) Hoplopleura nesoryzomydis Ferris (Hopkins, 1949) Hoplopleura affinis Burmeister (Ferris, 1919-35) Flea Craneopsylla minerva wolffhuegeli Rothschild (Hopkins and Rothschild, 1956) Calomys callosus expulsus Lund Flea Polygenis bohlsi bohlsi Wagner (Costa Lima and Hathaway, 1946) Zygodontomys lasiurus Lund Louse Hoplopleura affinis Burmeister (Hopkins, 1949) Zygodontomys brevicauda J. A. Allen and Chapman Louse Hoplopleura nesoryzomydis Ferris (Hopkins, 1949) PHYLLOTIS Section Phyllotis haggardi Thomas Flea Nosopsyllus londiniensis Jordan (Johnson, 1957) Phyllotis andium Thomas Flea Craneopsylla minerva minerva Rothschild (CNHM) Tick Ixodes (Exopalpiger) andinus Kohls (CNHM) Beetle Amblyospinus ancashi Seevers (Seevers, 1955) Phyllotis darwini posticalis Thomas Fleas (l)Plocopsylla pallas Rothschild (Hopkins and Rothschild, 1956) (t)Dysmicus claviger Rothschild (Johnson, 1957) Neotyphloceras crassispina crassispina Rothschild (CNHM) (l)Agaslopsylla nylota Traub (Traub, 1952) HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 49 (l)Tetrapsyllus bleptus Jordan and Rothschild (Traub, 1952) Cleopsylla townsendi Rothschild (CNHM) Tick Ornithodoros sp., larvae (CNHM) Phyllotis darwini rupestris Gervais Fleas Delostichus phyllotis Johnson (Johnson, 1957) Neotyphloceras crassispina crassispina Rothschild (CNHM; Traub, 1952; Costa Lima and Hathaway, 1946) Tetrapsyllus bleptus Jordan and Rothschild (CNHM) (l)Polygents bytunis Jordan and Rothschild (Costa Lima and Hathaway, 1946) Plocopsylla inti Johnson (Johnson, 1957) Sphinctopsylla inca Rothschild (Johnson, 1957) Louse Hoplopleitra affinis Burmeister (Hopkins, 1949) Mite Euschongastia phylloti Wharton (Wharton, 1948) Tick Ornithodoros sp., larvae (CNHM) Phyllotis darwini darwini Waterhouse Fleas Neotyphloceras crassispina chilensis Jordan (Costa Lima and Hathaway, 1946) Plocopsylla wolffsohni Rothschild (Costa Lima and Hathaway, 1946; Hopkins and Rothschild, 1956) Cleopsylla townsendi Rothschild (Johnson, 1957) Phyllotis darwini xanthopygus Waterhouse Fleas Plocopsylla chiris Jordan (Costa Lima and Hathaway, 1946; Hopkins and Rothschild, 1956) Plocopsylla achilles Rothschild (Costa Lima and Hathaway, 1946) Tiamastus subtilis Jordan and Rothschild (Costa Lima and Hathaway, 1946) Craneopsylla minerva wolffhuegeli Rothschild (Hopkins and Rothschild, 1956) (l)Polygenis by turns Jordan and Rothschild (Johnson, 1957) Tiamastus subtilis Jordan and Rothschild (Johnson, 1957) Phyllolis osilae J. A. Allen Flea (l)Polygenis thurmani Johnson (Johnson, 1957) Phyllotis micropus Waterhouse Lice Hoplopleura affinis Burmeister (Hopkins, 1949) Hoplopleura reducla Ferris (Hopkins, 1949) Phyllotis boliviensis Waterhouse Flea Cleopsylla townsendi Rothschild (Johnson, 1957) Neotyphloceras crassispina crassispina Rothschild (CNHM) 50 FIELDIANA: ZOOLOGY, VOLUME 46 Louse Hoplopleura affinis Burmeister (Hopkins, 1949) Phyllotis sublimis Thomas Flea Craneopsylla minerva minerva Rothschild (Johnson, 1957) Beetle Amblyopinus sp. (Seevers, 1955) Phyllotis pictus Thomas Fleas (l)Agastopsylla pearsoni Traub (Traub, 1952) (l)Plocopsylla enderleini Wagner (CNHM) Cleopsylla townsendi Rothschild (CNHM) Sphinctopsylla inca Rothschild (CNHM) Neotyphloceras crassispina crassispina Rothschild (CNHM) Ectinorus ineptus Johnson (Johnson, 1957) Louse Hoplopleura affinis Burmeister (Hopkins, 1949) Tick Ixodes sp. nymphs (CNHM) Phyllotis griseoflavus griseoflacus Waterhouse Fleas Craneopsylla minerva wolffhuegeli Rothschild (Costa Lima and Hathaway, 1946; Johnson, 1957) Dysmicus barrerai Jordan (Costa Lima and Hathaway, 1946) Dysmicus hapalus Jordan (Costa Lima and Hathaway, 1946) Hectopsylla stomis Jordan (Costa Lima and Hathaway, 1946) "Polygenis litargus puelche Del Ponte and Reisel" nomen nudum (Costa Lima and Hathaway, 1946) Polygenis occidentalis Cunha (Costa Lima and Hathaway, 1946) Polygenis platensis cisandinus Jordan (Costa Lima and Hathaway, 1946) Pulex irritans Linnaeus (Costa Lima and Hathaway, 1946) Neotyphloceras crassispina hemisus (Jordan, in de la Barrera, 1940) Louse Hoplopleura affinis Burmeister (Hopkins, 1949) Hoplopleura reducta Ferris (Ferris, 1919-35) Phyllotis amicus Thomas Fleas Neotyphloceras crassispina Rothschild (Costa Lima and Hathaway, 1946) Rhopaplopsyllus cacicus Jordan and Rothschild (Traub, 1952) Andinomys edax Thomas Fleas Dysmicus budini Jordan and Rothschild (Costa Lima and Hathaway, 1946) Neotyphloceras crassispina hemisus Jordan (Costa Lima and Hathaway, 1946) (l)Ectinorus disjugis Jordan (Johnson, 1957) HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 51 Chinchillula sahamae Thomas Fleas Cleopsylla townsendi Rothschild (Johnson, 1957) Neotyphloceras crassispina crassispina Rothschild (Traub, 1952) Tetrapsyllus bleptus Jordan and Rothschild (Traub, 1952) Plocopsylla enderleini Wagner (CNHM) C!)Agastopsylla pearsoni Traub (Traub, 1952) Mite Euschonygastia phylloti Wharton (Wharton, 1948) Beetle Amblyopinus sp. (Seevers, 1955) Euneomys chinchilloides Waterhouse Fleas Barreropsylla excelsa Jordan (Johnson, 1957) Plocopsylla wolffsohni Rothschild (Johnson, 1957) Ectinorus onychius onychius Jordan and Rothschild (Johnson, 1957) Tetrapsyllus rhombus Smit (Johnson, 1957) SOME SPECIAL EXTERNAL AND CRANIAL CHARACTERS Feet Figure 3 The hind feet of all phyllotines except Pseudoryzomys are small, generally weak in structure, and adapted primarily for locomotion on the ground. Most species are good climbers but in none is the foot obviously specialized for an arboreal habitat. The hind foot of Pseudoryzomys has evolved into the natatorial type, characterized by stout proportions, reduction of outer toes, weak claws, development of interdigital webbing — most markedly between the middle digits — the absence of superfluous hair, particu- larly from the plantar surface of the heel, and obsolescence of the fifth (outer posterior) postdigital plantar tubercle (fig. 53). The hind foot of Zygodontomys also departs somewhat from the usual vole-like phyllotine structure and approaches that of the palustrine type charactersitic of terrestrial species of Oryzomys. The foot is stout, fan-shaped, often dark, with relatively thick, long claws and short, thin, digital bristles. The most remarkable adaptation is the spade-shaped, gerbil-like hind foot of Eligmodontia. The form of the extremity and the hairy cushion of its sole are unique among cricetines. That Eligmodontia B D E to FIG. 3. — Plantar surface of left hind foot in dry skin: A, Phyllotis darwini; B, P. pictus; C, P. gerbillus; D, P. amicus; E, Calomys spp.; F, Eligmodontia typus; G, Galenomys garleppi. 52 HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 53 is a desert jumping mouse is inferred from the structure of its limbs and knowledge of its habitat. The hairy heel in phyllotines and a scattering of hairs between the plantar tubercles in some species are secondary developments, presumably adaptive, that partially replace the primitive scaly armor. The extremely hairy sole of Galenomys may be correlated with the cold climate of its Andean habitat, but sympatric phyllo- tines are not so benefited. In a like manner the sole of Phyllotis gerbillus, of the Piura desert, possesses an appreciable covering of bristles while that of other coastal desert phyllotines is bare. The fifth hind digit of all species of the genus Phyllotis is suffi- ciently elongated to be of assistance in climbing. The species of Calornys are similarly endowed, but their hind foot is extremely delicate (fig. 38). The hind foot of Andinomys, Chinchillula and Galenomys is of the ambulatory type common to savanna species, but nothing definite is known regarding locomotion in these genera. The hind feet of most phyllotines are remarkably uniform in structure, but the same types are common to unrelated cricetines inhabiting savanna or scrub country. The feet of Eligmodontia, Pseudoryzomys, Zygodontomys, and Galenomys are divergent from the generalized form but here, too, similar structures have evolved independently in other murids. The forefoot of cricetines is smaller and proportionately shorter than the hind foot, and its first digit is greatly reduced. The pollex bears a nail in phyllotines and all other cricetines except the special- ized fossorial forms (Notiomys, Blarinomys, Oxymycterus, Scaptero- mys] where it is provided with a claw. After locomotion the princi- pal functions of the forefoot are the seizure and conveyance of food to the mouth. The forefoot may develop exceptionally long claws for digging in some non-phyllotines, but it shows no specializations for a particular mode of locomotion comparable in degree to those of the hind foot. Tail Tail length approximately equal to combined head and body length is characteristic of the more generalized, terrestrial cricetines. In a plastic, wide-ranging form like Phyllotis darwini, tail length varies from about one-third less to two-fifths more than head and body length. A tail with this variability adapts itself easily to ter- restrial, fossorial, scansorial, saltatorial and semi-aquatic modes of life. The specialized tail, whether consistently longer or shorter than 54 FIELDIANA: ZOOLOGY, VOLUME 46 combined head and body length, can easily be derived from the flexible, medium length, all-purpose type. Tails of vole-like, bur- rowing Phyllotis sublimis, Calomys lepidus, Galenomys garleppi and Chinchillula sahamae, are one-third to three-fourths of the head and body length. Tails of the less specialized Andinomys edax, Phyllotis micropus, P. boliviensis, P. pictus, Zygodontomys, Calomys laucha, C. sorellus and C. callosus average between 75 and 100 per cent of head and body length. Highly specialized saltatorial rodents may be of the bipodal or leporid type, with tail very short or vestigial, or they may be of the tripodal or ricochetal type (cf. Hatt, 1932, pp. 602 et seq.), with tail stoutly developed and, with rare exceptions, longer than combined head and body length. No cricetine belongs definitely to either saltatorial category, but Eligmodontia shows a slight tendency to- ward the second. The elongated, tufted, balancing type of tail of scansorial or arboreal rodents is found in Phyllotis griseoflavus and P. amicus. In specialized, short-tailed forms, the tail of juvenals is always proportionately longer to head and body length than in adults. On the other hand, the tail of young individuals of long-tailed species is usually proportionately shorter. Bony Palate Figure 4 The postdiastemal portion of the hard palate of nearly all crice- tine species may be classified as short or long and as wide or narrow. These terms are defined as follows: Short palate: With median posterior borders of palatines not extending behind posterior plane of third molars. Long palate: With median posterior borders of palatines extend- ing behind posterior plane of third molars. Wide palate: With distance between inner borders of first molars greater than length of either molar. Narrow palate: With distance between inner borders of first molars less than length of first molar. In many cases the narrow palate is marked by a pair of deep lateral troughs separated by a prominent median longitudinal ridge. These features indicate that the narrow palate derives from a longitudinal corrugation of the wide palate. SHORT WIDE A If LONG WIDE B Y" SHORT NARROW C LONG NARROW D FIG. 4. — Hard palates in Muridae: A, short, wide (Thomasomys); B, long, wide (Phyllotis); C, short, narrow (Neotoma); D, long, narrow (Antinomy*} . Molars pentalophodont in A, and derivatives of the tetralophodont pattern in B I). 56 FIELDIANA: ZOOLOGY, VOLUME 46 The wide, short, uncomplicated palate such as occurs in Pero- myscus (s.s.), most species of Thomasomys (fig. 4, A), Rhipidomys and other related genera, is here regarded as primitive. Brachyo- dont, bunodont molars, whether pentalophodont or tetralophodont, are commonly associated with the wide, short, simple palate. On the other hand, the narrow palate, either long or short, is rarely associated with the generalized or low crown, crested molars of cricetines. In a few forms (e.g., Irenomys) the palate has remained primitively short but is clearly evolving from wide to narrow, while the molars have already reached an advanced stage of hypsodonty and lamination. A clearly defined narrow and short palate, however, does not appear in any South American cricetine. Among North American peromyscines, the palate of woodrats (Neotoma, Xenomys) is narrow and short (fig. 4, C) and may become even shorter as it constricts pari passu with increasing hypsodonty. The long, postdiastemal portion of the palate evolved from the short, wide palate by a closing of most of the gap between the median posterior portions of the paired palatine bones. It represents com- pletion of the roofing-over process of the secondary or mammalian palate and is always correlated with complications of the posterior palatal and postpalatal regions. The evolutionary stages are evident in living forms. Thus in juvenals with unerupted third molars this type of palate is still simple and comparatively short. In akodonts and such peromyscines as Haplomylomys, Podomys, Baiomys, Reithro- dontomys the palate is wide, with a narrow range of variation from short to long. Usually, the posterior palatal area in this group is in- flated, and the pits are roofed (as seen from below) . The wide palate of the dentally primitive Aporodon also varies from short to barely long. In phyllotines (fig. 4, B), the postdiastemal portion of the pal- ate is consistently long and is distinguished by the presence of more or less conspicuous posterolateral palatal pits or fossae. Here the palate is wide in all genera except Andinomys and Chinchillula. It is widest in the brachyodont, bunodont Calomys, Eligmodontia, and Zygodontomys and becomes progressively narrower with increasing hypsodonty. The narrow palate in Andinomys (fig. 4, D) and Chin- chillula is associated with extremely high, flat-crowned molars. The palate of sigmodonts (Holochilus, Sigmodon, Neotomys, Reithrodori) is generally wide; the molars, however, are more ad- vanced than those of phyllotines and the postpalatal pterygoid re- gion is extremely complicated. The palate of oryzomyine rodents is definitely long and like the wide type of most phyllotines. Identical palatal structures in buno- HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 57 brachy-pentalophodont oryzomyines and buno-brachy-tetralopho- dont phyllotines may be independent developments or may point to a common ancestry. A primitive palate may be associated with any advanced stage of evolution in other cranial, dental, and external characters. Con- versely, a specialized palate may be associated with generalized characters in other parts of the organism. The structure of the palate alone, however, tends to be constant or to vary in but one direction in any given species or genus. The combination of postdiastemal palatal and molar characters is an important indicator of generic and supergeneric relationships. The accompanying diagram (fig. 2) shows the palatal-molar relation- ship between phyllotines and those cricetine groups with which comparisons are made. Proportional length of the diastemal portion of the palate, whether long or short, is also an important character in the system- atics of the Muridae, as well as in other rodent groups. However, among phyllotines and most pastoral cricetines in general, the char- acter is of little significance. A discussion of its implications, there- fore, is omitted here. In another section (p. 102) the elongated type of the diastemal portion of the palate is mentioned in con- nection with dental specializations and feeding habits. Supraorbital Region Figure 23 In most phyllotines, as in Muridae in general, the borders of the supraorbital region are either divergent or concave mid-frontally. In some, the sides of the supraorbital region vary from parallel to concave or divergent. In the concave type, the distance across the sides of the supraorbital region, measured at the mid-frontal plane, is always less than the greatest width of the rostrum. In the diver- gent type, the width across the sides is always more. The primitive murid supraorbital region was almost certainly of the concave type but with sides nearly parallel mid-frontally. This condition persists in many cricetines and in the phyllotine Calomys lepidus, Phyllotix darwini and others. In Calomys sorellus the sides vary from slightly concave to parallel mid-frontally, while in Calomys laucha the sides vary from parallel to slightly divergent. The supraorbital regions of juvenals of either type is generally broad without notable constriction or divergence. In adults spe- 58 FIELDIANA: ZOOLOGY, VOLUME 46 cialization of the supraorbital region has led either to further con- striction of the concave type with formation of longitudinal frontal ridges or to greater expansion of the divergent type with development of beading and overhanging frontal ledges. Exaggeration of either the concave or the divergent type of supraorbital region is also char- acteristic of old age. Tendency toward constriction of the supraorbital region is pres- ent but hardly noteworthy in Phyllotis pictus and P. boliviensis. In closely related Euneomys chinchilloides, and in sigmodonts like Reithrodon physodes, Neotomys ebriosus, and Holochilus magnus, there is marked constriction. Extreme constriction of the supra- orbital region is attained in microtines. Increasing expansion of the divergent type of supraorbital region may be traced in phyllotines from the nearly parallel, square-sided condition in Calomys laucha to the widely divergent, beaded, ledge- like supraorbital borders in Calomys callosus, Zygodontomys and Phyllotis griseoflavus. As in the concave type, development of diver- gent-sided supraorbitals has not gone far in phyllotines. Most ad- vanced conditions among other cricetines appear in arboreal forms such as Tylomys, Ototylomys, Nyctomys and the Oecomys section of Oryzomys. The correlation between an arboreal habitat and a diver- gent-sided supraorbital region appears in Old World Muridae as well. There is no apparent reason for the predominance of the ex- panded, divergent-sided type of supraorbital region among scan- sorial and arboreal species of Muridae (including Cricetinae). Also, no explanation is offered for the greater prevalence of the constricted concave type of supraorbital region among voles and vole-like mice. In either case, modification of the supraorbital regions appears to be of specific rather than generic grade. Baculum Figures 5-8 The two basic types of bacula, simple and complex, are defined and their systematic significance is discussed in a preceding section (p. 19). The baculum of all phyllotines of which the element has been studied is complex (figs. 5, B; 6). This type is characterized by a bony shaft with an expansion at the base and three soft, cartilaginous or more or less ossified, digitate processes at the tip. The base (fig. 6) is marked by a pair of lateral condyles that are thickened or knobby HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 59 in some specimens, undercut or channeled and curled upward in others. The condyles may be connected along the median line by an expanse of bone or they may be separated by a median incision. B COMPLEX FIG. 5. — Simple and complex types of glandes pene and bacula. A, glans penis and baculum (dotted) of Baiomys musculus (after Hooper [1959, pi. 4a]); B, glans penis and baculum (dotted) of Phyllotis dar- urini. The basal contour of the shaft varies from either concave or convex to nearly plane. Shafts with the convex base predominate among cricetines with the complex type of bacula. The same shape of base appears to be universal among microtines and in cricetines with the simple type of baculum and glans. The concave base (figs. 5, 6) is found only in Phyllotis dar- wini among strictly South American cricetines, and Burt (1960, pi. 24a) figures the baculum of only one North American species, Oryzo- mys alfaroi, with the same kind of base. The base of the nearly re- lated Oryzomys melanotus, figured by the same authority (pi. 24e) is convex. Bacula of juvenals of phyllotine species with either convex or con- cave bases, are practically indistinguishable. Evidently, both types of bases can be derived from the Juvenal of either. All bacula of adult Phyllotis darwini wolffsohni examined resemble those of the juvenals of either P. osilae or P. darwini (fig. 7). Of eight bacula of adult P. darwini caprinus, five are of the Juvenal type as in wolffsohni, the remaining three are of the usual concave type characteristic of the baculum of P. darwini. The difference in size between Juvenal and adult type bacula in adult caprinus is not related to the overall size of the animals themselves (Table 1). 81270 81274 75438 81282 75543 75380 69155 71257 FIG. 6. — Bacula of phyllotines; distal digitate processes not shown. EXPLANATION OF FIGURE 6 (5 digit numbers are of specimens in Chicago Natural History Museum, 6 digit numbers are of specimens in the Museum of Vertebrate Zoology, University of California.) 75453 Phyllotis darwini (Huancavelica, Huancavelica, Peru) 115856 osilae (Pairumani, Puno, Peru) 120025 andium (Matucana, Lima, Peru) 23290 " micropus (Rfo Inio, Chiloe Is., Chile) 53166 amicus (Chosica, Lima, Peru) 46123 griseoflavus (Belen, Catamarca, Argentina) 81270 " gerbillus (Piura, Piura, Peru) 81274 " " (Piura, Piura, Peru) 75438 pictus (Tambo, San Miguel, Ayacucho, Peru) 81282 (Quilcayhuanca, Ancash, Peru) 75543 Calomys sorellus (Ocros, Ayacucho, Peru) 75380 71257 Zygodontomys (Muzo, Boyaca, Colombia) 69155 " (Socorre, Cordoba, Colombia) 60 HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 61 A parallel relationship in size and shape of bacula exists between those of Phyllotis osilae osilae and P. osilae phaeus. The baculum of the first is comparable in size but not in shape to that of adult P. dar- wini. The single available baculum of an adult P. o. phaeus is Juvenal in size and, though distinguishable from the baculum of sympatric specimens of P. darwini, can just as readily be assigned to P. osilae as to allopatric P. darwini on the basis of its shape (fig. 7). The size and shape of the baculum in adult phyllotines varies locally and geographically (fig. 8, Table 1). The smallest baculum occurs in the large race Phyllotis darwini wolffsohni. The baculum of the largest race, P. d. magister, is no larger than that of the smallest race, P. d. rupestris. The difference in size of bacula of the Camataqui and Tilcara series of P. d. caprinus (fig. 7) is startling. That of the former is small and agrees with wolffsohni, while that of the latter is nearly as large as the baculum of rupestris. The Camargo and Tarija specimens show the same relationship, respec- tively. This accords with variation noted in the external, cranial and dental characters of P. d, caprinus. Presence of thickenings or slightly developed processes on the mid-shaft region appears to be a local variable in P. darwini. The processes are present in all seven bacula of P. darwini magister which were examined and in the two preserved bones of the intergrading series of P. d. posticalis from Puquio, Peru (fig. 7). Thickenings of the mid-shaft region occur sporadically in P. d. rupestris but not in representatives of the populations with which magister is sym- patric. Three of the four preserved bacula of P. d. posticalis from Limbani, Peru, are also thickened at mid-shaft but the single bacu- lum of sympatric P. osilae is not. There is a very slight mid-shaft thickening in some bacula of caprinus but the shaft appears to be smooth and evenly tapered in wolffsohni. There is no relationship between cyclical changes in size of testes and size of baculum in adult phyllotines. A correlation between the size and form of the baculum or the phallus as a whole and the adaptation of the animal to its external environment is not evident. The penis is designed for the deposition of sperm in that part of the female genital tract where fertilization may proceed most effectively. Variations in the form and function of the female reproductive tract exert selective pressures on the male reproductive system. The penis that cannot insure delivery of sperm to egg cannot insure the survival of its kind. EXPLANATION OF FIGURE 7 (Numbers are of specimens in the Museum of Vertebrate Zoology, University of California.) Phyllotis osilae osilae 114702 Pairumani, Puno, Peru 114696 120163 Leon, Jujuy, Argentina Phyllotis osilae phaeus 116177 Limbani, Puno, Peru Phyllotis darwini rupestris 114684 Pairumani, Puno, Peru (compare with sympatric P. osilae osilae above) 115887 (compare with P. d. posticalis at right and P. d. magister below). Phyllotis darwini posticalis 115809 Puquio, Ayacucho, Peru (compare with P. d. rupestris at left and P. d. magister below). 120003 Ayacucho, Ayacucho, Peru (compare with P. osilae phaeus above). 116136 Limbani, Puno, Peru (compare with sympatric P. osilae phaeus above). Phyllotis darwini magister 115874 Tarata, Tacna, Peru 115878 (compare both samples with P. d. rupestris and P. d. posticalis above). Phyllotis darwini wolffsohni 120196 Comarapa, Santa Cruz, Bolivia 120194 (compare both samples with P. d. caprinus and P. andium below and P. o. phaeus above). Phyllotis darwini caprinus 120204 Tilcara, Jujuy, Argentina 119975 Camargo, Chuquisaca, Bolivia 120208 Tilcara, Jujuy, Bolivia 120201 Tarija, Tarija, Bolivia (compare with P. d. wolffsohni and P. osilae osilae above). 120132 Camataquf, Tarija, Bolivia. Phyllotis andium 120034 Surco, Lima, Peru 120037 120025 Matucana, Lima, Peru (compare all samples with P. osilae above). 62 116177 115887 115809 120003 116136 115874 115878 120196 120194 120204 119975 120208 120201 120132 120034 120037 120025 FIG. 7. — Bacula of species and subspecies of the Phyllotis danrini complex; distal digitate processes not shown. See opposite page for explanation. 63 Chongoyape Yungay Ticopampa FIG. 8.- — Bacula of Phyllotis andium showing individual and local variation; distal digitate processes not shown. (All 81257 81255 81252 81253 81251 81256 81238 81247 81240 81243 81242 81246 81233 81236 81232 81235 EXPLANATION OF FIGURE 8 bacula in Chicago Natural History Museum. Specimens shown from left to right are listed in order by their catalog numbers.) Chongoyape, Lambayeque, Peru Yungay, Ancash, Peru Ticopampa (Hacienda Catac), Ancash, Peru 64 HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 65 TABLE 1. —Individual and Geographic Variation in Size of Bacula in Phyllotis ssp. Head Baculum width and Cat. No. Locality length and Testes body length length Skull length Weight of animal Phyllotis danvini limatus 114704 Atico, Arequipa, Peru 30x36 14 121 31 .7 49 114705 Atico, Arequipa, Peru 28x34 6 112 29 .4 38 116677 Atico, Arequipa, Peru 30x39 6 114 29 .3 47 120062 Chosica, Lima, Peru 26x37 11 107 29 .6 Phyllotis darwini posticalis 120086 Casapalca, Lima, Peru 30x39 31 .2 115809 Puquio, 15km. NE., 29x38 12 134 33 .2 72 Ayacucho, Peru 116024 Puquio, 21 miles ENE., 30x42 12 128 33 .2 64 Ayacucho, Peru 120003 Ayacucho, Ayacucho, Peru 14x33 4 113 30 .8 114679 Limbani, SmilesSSW., 30x40 12 127 32 .9 70 Puno, Peru 114680 Limbani, 8 miles SSW., 31x29 8 131 31 .9 55 Puno, Peru 114683 Limbani, 8 miles SSW., 32x39 13 125 32, ,8 75 Puno, Peru 116136 Limbani, 4 miles SSW., 33x35 9 123 30, 2 50 Puno, Peru Phyllotis darwini magister 115792 RioTorata, 15km. NE., 19x — 8 122 30 .9 Moquegua, Peru 115874 Tarata, 1 km. N., Tacna, Peru 28x41 12 133 33 .4 71 115877 Tarata, 1 km. N., Tacna, Peru 27x42 13 135 34 .5 93 115878 Tarata, 2 km. N., Tacna, Peru 25x35 11 133 31 .6 70 115879 Tarata, 2 km. N., Tacna, Peru 26x38 11 127 32 .8 75 115881 Tarata, 4 km. N., Tacna, Peru 26x39 11 131 32 .9 72 115883 Tarata, 4 km. N., Tacna, Peru 25x43 11 144 34 .5 89 Phyllotis darwini rupestris 115780 Arequipa, 7 km. E.f 26x32 9.5 105 28 .8 40 Arequipa, Peru 115784 Arequipa, 7 km. E., 21x32 7 88 27 .3 25 Arequipa, Peru 115785 Arequipa, 7 km. E., 22x31 8 96 27 .2 26 Arequipa, Peru 115786 Arequipa, 12 km. SSW., 27x36 9 104 29 .2 Arequipa, Peru 115787 Arequipa, 12 km. SSW., 23x36 10 102 29 .2 Arequipa, Peru 116124 Arequipa, 52 miles ENE., 28x37 110 29 .5 48 Arequipa, Peru 66 FIELDIANA: ZOOLOGY, VOLUME 46 TABLE 1. — Individual and Geographic Variation in Size of Bacula in Phyllotis ssp. (continued) Cat. No. Locality 116126 Huaylarco, Arequipa, Peru 116127 Imata, Arequipa, Peru 116128 Salinas, Arequipa, Peru 114684 Pairumani, Puno, Peru 115887 Pairumani, Puno, Peru 1 14685 Santa Rosa, Puno, Peru 1 14686 Santa Rosa, Puno, Peru 115816 Have, Puno, Peru 115817 Have, Puno, Peru 115819 Juli, Puno, Peru 115822 Pampa de Ancomarca, Puno, Peru 115823 Pampa de Ancomarca, Puno, Peru 115825 Puno, 5 km. W., Puno, Peru 1 1 5888 Puno, 5 km. W., Puno, Peru 1 1 5827 Puno, 82 km. W., Puno, Peru 115891 Puno, 15 km. W., Puno, Peru 1 1 5830 Rio Santa Rosa, Puno, Peru 115831 Rio Santa Rosa, Puno, Peru 1 1 5870 Pomata, Puno, Peru 116130 Asillo, Puno, Peru 116134 Tincopalca, Puno, Peru 1 1 6690 Juliaca, Puno, Peru 115799 Torata, 10 km. NEM Moquegua, Peru 115812 Lago Suche, Moquegua, Peru 115832 Tarata, 4 km. N., Tacna, Peru 115837 Tarata, 4.5 km. N., Tacna, Peru 115838 Tarata, 4.5km. N., Tacna, Peru 1 1 5840 Tarata, 2 km. N., Tacna, Peru 115843 Tarata, 2 km. N., Tacna, Peru 115845 Tarata, 3 km. N., Tacna, Peru 115846 Tarata, 4 km. N., Tacna, Peru 115847 Tarata, 5 km. N., Tacna, Peru 1 1 5849 Tarata, 20 km. N., Tacna, Peru 115853 Tarata, 20 km. N., Tacna, Peru 1 1 6788 Tarata, 8 miles NE., Tacna, Peru 1 1 6792 Caritaya, Tarapaca, Chile 1 1 6782 Caritaya, Tarapaca, Chile 1 1 6783 Caritaya, Tarapaca, Chile 1 1 6784 Caritaya, Tarapaca, Chile 1 1 6778 Toconce, Antof agasta, Chile 1 1 6780 Toconce, Antof agasta, Chile Baculum width and length Head and Testes body length length Skull length Weight of animal 29x39 11 129 32.0 67 26x36 12 120 30.3 50 29x41 119 49 34x42 14 118 32.0 62 31x40 13 121 31.1 63 30x38 10 120 31.0 56 25x34 12 105 29.3 44 28x40 12 114 30.7 54 34x39 12 107 31.4 53 16x30 5 92 27.1 31 26x32 11 110 28.5 46 28x36 11 127 31.0 69 27x38 12 118 31.2 60 25x29 10 109 28.8 42 21x35 11.5 103 28.0 38 30x42 126 32.8 68 30x39 11 133 31.7 19x35 10 108 30.0 27x40 11 119 31.2 64 27x38 11 118 30.2 46 — x38 11 124 31.7 34x41 11 123 31.0 55 25x36 9 97 28.5 40 26x34 11 111 31.7 61 28x36 11 123 31.6 n 25x34 9.5 94 26.8 32 i 28x41 11 116 30.1 58 26x37 10 103 28.6 39 26x — 8 99 27.7 35 28x38 9 101 30.0 39 27x37 10 113 29.5 52 — x32 94 28.2 32 i 25x36 12 112 30.4 56 i 26x35 10 111 31.5 52 i 29x37 9x6 116 29.9 51 29x38 110 29.9 43 26x37 10 114 29.1 50 28x32 10 104 42 — x36 10 117 52 30x37 8 113 29.5 50 31x37 118 28.3 43 HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 67 TABLE 1. — Individual and Geographic Variation in Size of Bacula in Phyllotis ssp. (continued) Head Cat. No. Baculum and Weight width and Testes body Skull of Locality length length length length animal Phyllotis dancini wolffsohni Tapacari, Cochabamba, Bolivia 13 x 33 6 135 31.8 57 Tapacari, Cochabamba, Bolivia 10 x 28 4 103 28.1 Tapacari, Cochabamba, Bolivia 10x27 4 117 30.8 52 Punata, 8 miles NE., 10x29 4 114 29.0 49 Cochabamba, Bolivia Punata, 10 miles NE., 8x27 4 112 29.4 42 Cochabamba, Bolivia Comarapa, Santa Cruz, Bolivia 14x33 5 129 31.0 55 Comarapa, Santa Cruz, Bolivia 7x25 6 116 30.5 52 Phyllotis danrini capritms Tilcara, Jujuy, Argentina 24 x 37 10 116 30.4 Tilcara, Jujuy, Argentina 28x40 11 116 31.1 Tilcara, Jujuy, Argentina 26x38 11 123 32.2 Tilcara, Jujuy, Argentina 15x31 7 114 29.7 Camargo, Chuquisaca, Bolivia 26 x 38 7 122 30.7 47 Camataqui, Tarija, Bolivia 17x32 7 118 31.4 56 Camataqui, Tarija, Bolivia 1 1 x 23 4 116 28.1 35 Camataqui, Tarija, Bolivia 12x23 4 101 29.0 39 Tarija, Tarija, Bolivia 13x32 5 126 31.7 Phyllotis osilae osilae Calacala, Puno, Peru 29x42 10 118 51 Calacala, Puno, Peru 33x44 10 125 28.2 55 Calacala, Puno, Peru 29 x 39 12 124 30.8 53 Calacala, Puno, Peru 29 x 43 7 124 30.8 55 Pairumani, Puno, Peru 28x43 12 131 32.0 69 Pairumani, Puno, Peru 24 x 38 9 105 27.6 32 Pairumani, Puno, Peru 32x42 10 129 31.5 57 Pairumani, Puno, Peru 29 x 43 11 124 31.9 66 Pairumani, Puno, Peru 30x43 11 119 31.0 58 Juli, Puno, Peru 27x38 10 122 31.4 70 Puno, 15km. W., Puno, Peru 26x37 12 120 30 . 7 51 Puno, 5km. W., Puno, Peru 30x37 10 116 31.3 52 Pomata, Puno, Peru 25 x — 10 126 32.0 69 Pomata, Puno, Peru 25 x 42 9.5 121 31.0 56 Pomata, Puno, Peru 23x36 8 122 29.8 60 Pomata, Puno, Peru 22x41 7 115 30.9 52 Pomata, Puno, Peru 28 x — 9 127 31.1 62 Pomata, Puno, Peru 23x38 8 118 30.6 47 Asilla, Puno, Peru 24 x 40 7 115 30 . 4 68 FIELDIANA: ZOOLOGY, VOLUME 46 TABLE 1. — Individual and Geographic Variation in Size of Bacula in PhyUotis ssp. (continued) Cat. No. Locality 1 1 6686 Arapa, Puno, Peru 120163 Leon, Jujuy, Argentina 116177 Limbani, Puno, Peru Head Baculum and Weight width and Testes body Skull of length length length length animal 23x36 4 112 30.3 46 11x31 4 109 29.0 19x32 6 126 31.5 54 PhyUotis 120008 Canta, Lima, Peru 120017 Huaros, Lima, Peru 120022 Matucana, Lima, Peru 1 20023 Matucana, Lima, Peru 120024 Matucana, Lima, Peru 120025 Matucana, Lima, Peru 120034 Surco, Lima, Peru 120035 Surco, Lima, Peru 120037 Villavista, Lima, Peru 81232 Hacienda Catac, Ticopampa, Ancash, Peru 81233 Hacienda Catac, Ticopampa, Ancash, Peru 81235 Hacienda Catac, Ticopampa, Ancash, Peru 81236 Hacienda Catac, Ticopampa, Ancash, Peru 81238 Yungay, Ancash, Peru 81240 Yungay, Ancash, Peru 81247 Yungay, Ancash, Peru 81243 Yungay, Ancash, Peru 81246 Yungay, Ancash, Peru 81242 Yungay, Ancash, Peru 81251 Chongoyape, Lambayeque, Peru 81252 Chongoyape, Lambayeque, Peru 81253 Chongoyape, Lambayeque, Peru 81255 Chongoyape, Lambayeque, Peru 81256 Chongoyape, Lambayeque, Peru 81257 Chongoyape, Lambayeque, Peru andium 14x29 100 27.5 14x29 103 28.8 14x31 106 28.0 11x23 95 26.6 16x31 103 27.8 16x30 111 29.3 9x25 99 28.4 14x30 105 28.4 17x29 116 30.5 14 x- 123 29.3 14x — 111 27.7 14 x — 109 27.8 14x — 110 27.3 17x29 119 30.2 15x25 97 27.5 13x — 113 28.2 12x24 112 26.6 15x27 117 28.3 14x — 117 28.5 13x — 99 26.4 17x — 99 26.7 14x26 95 25.0 15x27 103 25.7 20 x — 114 27.6 17x29 108 27.2 HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 69 DENTAL CHARACTERS AND EVOLUTION Enamel Folds in Rodent Molar Evolution The crown pattern of the molars of cricetines and many other rodents can be most accurately described in terms of the enamel folds. A name for each definable fold is as necessary a tool in the study of rodent phylogeny as a name for a primary cusp or loph. The fold terminology used here was proposed by me in 1944 (p. 14). Another very useful terminology based on beaver molars was devised by Stirton (1935, p. 392). However, the definitions of the folds as originally presented by these authors lack the precision and scope needed for present purposes. Enamel folds are now defined as fundamental dental features. It follows that all folds developed in the same way in the same place are homologous. The letter symbols used in the following definitions are also shown in figure 9. Major fold (MF) : Based on posterior enamel wall, or ridge, of protocone(-id) and, except where enteroloph or ectolophid is well developed, continuous with anterior enamel ridge of the hypocone (-id). The major fold is the most convenient base reference for orientation and homologization of all quadritubercular molar ele- ments. First minor fold (NF 1): Based on posterior enamel ridge of anteroconule(-id) and continuous with anterior border of proto- cone(-id); non-existent in absence of anteroconule(-id). Second minor fold (NF 2) : Based on anterior enamel ridge of posteroconule(-id) and continuous with posterior border of hypo- cone(-id); present in a few cricetines (e.g., Punomys; third upper molar of Neotomys), in some murines and in many caviomorphs. First primary fold (PF 1) : Of upper molar, based on anterior enamel ridge of paracone and continuous with posterior border of anteroloph; may be obsolete or absent in molar with reduced or excessively worn procingulum. Of lower molar, based on posterior enamel ridge of metaconid and continuous with anterior ridge of either entoconid, or, if present, mesolophid. Second primary fold (PF 2) : First fold of tribosphenic, or primitive therian upper molar (Simpson, 1936), and most persistent through attrition and specialization; based on anterior ridge of metacone and continuous with posterior ridge of either paracone or, if present, mesoloph. Of lower molar, based on posterior ridge of entoconid and continuous with anterior ridge of posterolophid. EXPLANATION OF FIGURE 9 MF major fold NF 1 first minor fold NF 2 second minor fold PF 1 first primary fold PF 2 second primary fold SF 1 first secondary fold SF 2 second secondary fold IF 1 first internal fold IF 2 second internal fold NAMES OF FOLDS AIF AMF ASF ALF ABF APF anterior internal fold anterior median fold anterior secondary fold (upper molars only) anterior lingual fold (upper molars only) anterior labial fold (lower molars only) anterior primary fold (lower molars only) NAMES OF CUSPS Upper Molars (A) 1. Protocone 2. Hypocone 3. Paracone 4. Metacone 5. Mesoloph 1 6. Mesostyle / 7. Anteroloph 8. Anterolabial style 9. Anteroconule 10. Anterolingual style 11. Anterolabial conule 12. Anterolingual conule 13. Anteromedian style 14. Posteroloph 15. Posterostyle 16. Posteroconule 17. Enteroloph 18. Enterostyle 19. Paralophule 20. Metalophule a. Antero-median protolophule b. Postero-median c. Antero-median hypolophule d. Postero-median " mesolophostyle (-id) Procingulum Postcingulum Lower Molars (B) 1. Protoconid 2. Hypoconid 3. Metaconid 4. Entoconid 5. Mesolophid 6. Mesostylid 7. Anterolophid 8. Anterolingual stylid 9. Anteroconulid 10. Anterolabial stylid 11. Anterolingual conulid 12. Anterolabial conulid 13. Anteromedian stylid 14. Posterolophid 15. Posterostylid 16. Posteroconulid 17. Ectolophid 18. Ectostylid 19. Metalophulid 20. Entolophulid a. Antero-median protolophulid b. Postero-median c. Antero-median hypolophulid d. Postero-median " 70 LABIAL LINGUAL (3) M F UPPER RIGHT MOLARS LOWER LEFT MOLARS FIG. 9. — Master plan of the occlusal surface of rodent molars showing all elements present in the enamel pattern of Muridae. Supernumerary lophules (-ids) present in the third molars of some forms (cf. Otomyx, fig. 16, A) are not shown. For key to numbers and abbreviations see opposite page. 71 72 FIELDIANA: ZOOLOGY, VOLUME 46 First secondary fold (SF 1) : Of upper molar, based on anterior enamel ridge of mesoloph ; sometimes divided into lateral and median portions by a paralophule; non-existent in absence of mesoloph. Of lower molar, based on posterior enamel ridge of anterolophid ; non- existent in absence of anterolophid. Second secondary fold (SF 2) : Of upper molar, based on anterior border of posteroloph and continuous with posterior enamel ridge of metacone; non-existent in absence of posteroloph. Of lower molar, based on posterior enamel ridge of mesolophid; sometimes divided into lateral and median portions by an entolophulid ; non-existent in absence of mesolophid. First internal fold (IF 1): Enamel island defined by internal enamel ridges or lophules(-ids) connecting protocone(-id) with paracone (or metaconid); may be coalesced with first primary or first secondary fold. Second internal fold (IF 2) : Enamel island defined by internal enamel ridges or lophules(-ids) connecting hypocone(-id) with meta- cone (or entoconid); may be coalesced with second primary or second secondary fold. In rodents with premolars absent, the procingulum of the first molar is hypertrophied and plicated. The resultant lophs and con- ules duplicate in appearance the structures immediately behind and partially replace the functions of the premolar lost in front. The folds defining the components of the procingulum are defined as follows: Anterior median fold (AMF) : Loop defining internal borders of anterolingual and anterolabial conules(-ids) ; may be weakly defined in the specialized molar or isolated and coalesced with anterior in- ternal fold defined below. Anterior secondary fold (ASF) : Defines opposing enamel walls of anterolabial conule and anteroloph. Anterior lingual fold (ALF) : Defines opposing enamel walls of anterolingual conule and anteroconule. Anterior primary fold (APF) : Defines opposing enamel walls of anterolingual conulid and anterolophid. Anterior labial fold (ABF): Defines opposing enamel walls of anteroconulid and anterolabial conulid. Anterior internal fold (AIF) : Enamel island defined by internal enamel ridges or lophules(-ids) connecting anterolingual and antero- labial conules(-ids); not to be confused with secondarily isolated anterior folds with which it may be coalesced. HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 73 Supernumerary folds: In some rodents, for example, Otomys (fig. 16, A), there is a tendency for the development of supernumerary crests posteriad to the posteroloph(-id) or posteroconule(-id) of the third molar. The folds defining these conules are treated as super- numerary secondary folds (SF 3, SF 4, etc.) or supernumerary minor folds (NFS, NF4, etc.). Summary. — Folds based on the main cusps of the primitive therian molars are the major and the two primary folds. The minor and secondary folds are based on lophs and other secondarily de- veloped dental elements. Folds based on homologous cusps and lophs are correspondingly homologous. Folds based on derivatives of the anterior cingulum of the first molar and of the posterior cingu- lum of the third molar are analogous or, rather, homoplastic struc- tures. Simple accessory styles are not defined by folds, even when originating from or secondarily fused with, elements identified with folds. The first fold to characterize the tribosphenic upper molar is the second primary (PF 2). The major fold (MF) is the second to appear in phylogeny. This fold delimits the hypocone and marks the quadritubercular upper molar. The first fold of the tribosphenic lower molar was one between paraconid and metaconid. In therians, such as rodents that have lost the paraconid, the first primary (PF1) and the major fold (MF) are the oldest. In the upper molar, the first primary and the two secondary folds delimit the lophs already present in the earliest cricetine molars. Internal and median folds appear last and are secondary modifications developed within the cricetine line. Minor folds inaugurate the cycle of specialization. The second minor fold, especially, is a development of the grazing type of molar in several rodent lines. In lower molars major and minor folds may be regarded as equiva- lents of corresponding folds of the upper molars. Primary and secondary folds, on the other hand, are not equivalents of the same named folds of the upper molars. The same terminology is used, nevertheless, for convenience. Once the maximum number of folds have become established in a species, their order of isolation and disappearance, whether through attrition in the individual or specialization in the line of descent, is the inverse of their order of appearance in phylogeny. Thus, the second primary fold, the first fold to define the tribosphenic upper molar, is the last to disappear. 74 FIELDIANA: ZOOLOGY, VOLUME 46 Procingulum and Postcingulum Homologies of the primary cusps and lophs of cricetine cheek teeth defined by Wood and Wilson (1936) have been generally ac- cepted. However, attempts to apply extensively the Wood and Wilson definitions for derivatives of the cingulum meet with diffi- culties. The cingulum of the quadritubercular molar in rodents is dis- continuous. The anterior and posterior portions are most de- veloped, and, in the more generalized tooth, they give rise to tubercles similar in form and structure to primary cusps and lophs. Lateral portions of the cingulum may appear in the form of well-defined ledge-like outgrowths supporting one or more stylar processes. These processes, however, may be present with only the merest trace of their original lateral cingular base. The terms anterior cingulum and posterior cingulum have been applied not only to these basic portions of the cingulum but have been used at times for elements secondarily derived from the cingula. To avoid ambiguity, the terms procingulum and postcingulum, pro- posed by me in 1955 (p. 651), are reintroduced for the complex derived from the anterior cingulum and the complex derived from the posterior cingulum, respectively. These terms are particularly useful in the analysis of murid molars. They are defined as follows: Procingulum, upper molars: Anterior cingulum with the follow- ing complex of functional elements derived from it: (a) Anteroloph (fig. 9, A 7): Anteroloph of Wood and Wilson ("anterior cingulum" of some authors). (b) Anterolabial style (fig. 9, A 8) : When present usually fused with anteroloph; if not discrete, its presence indicated when combined elements extend to lateral margin of tooth. (c) Anteroconule (fig. 9, A 9): Small anterior cone defined by first minor fold. (d) Anterolingual style (fig. 9, A 10) : When present never well developed and generally juxtaposed against anteroconule. (e) Anterolabial conule (fig. 9, A 11) : Always present in first molar with well-developed procingulum; absent as discrete element in second and third molars. (f) Anterolingual conule (fig. 9, A 12) : Like preceding, always present in first molar with well-developed procingulum; absent in second and third molars. HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 75 Note: Anterolabial and anterolingual conules may have been originally independent and discrete derivatives of the anterior cingu- lum, and the two may subsequently have fused into a single cone, the anterocone of Wood and Wilson. The possibility, however, that a simple anterocone is the primitive structure in rodents cannot be discounted. Anteromedian style (fig. 9, A 13) : When present arising from anteromedian border of anterior cingulum; rarely well de- *T*"il rf"Vt"Vrf-Wl (g) veloped. Postcingulum, upper molars: Posterior portion of cingulum with the following complex of functional elements derived from it: (a) Posteroloph (fig. 9, A 14) : Posteroloph of Wood and Wilson ("posterior cingulum" of some authors). (b) Posterostyle (fig. 9, A 15) : When present usually fused with posteroloph; if not discrete, presence indicated when com- bined elements extend to lateral margin of tooth. (c) Posteroconule (fig. 9, A 16): Lateral posterior cone defined by second minor fold; analogous to anteroconule in origin. (d) Supernumerary crests: See definition below (p. 76). Procingulum, lower molars: (a) Anterolophid (fig. 9, B 7): Less developed than anteroloph. (b) Anterolingual stylid (fig. 9, B 8): Usually absent; when present less developed than corresponding element (antero- labial style) of upper molar. (c) Anteroconulid (fig. 9, B 9): Frequently present when cor- responding element (anteroconule) of upper molar is absent. (d) Anterolabial stylid (fig. 9, B 10): Generally present in juxta- position with anteroconulid and more developed than cor- responding element (anterolingual style) of upper molar. (e) Anterolingual conulid (fig. 9, B 11). (f) Anterolabial conulid (fig. 9, B 12). Note: Anterolingual and anterolabial conulids together compose the anteroconid of Wood and Wilson. (g) Anteromedian stylid (fig. 9, B 13): Arising from anterior border of cingulum ; rarely present. 76 FIELDIANA: ZOOLOGY, VOLUME 46 Postcingulum, lower molars: (a) Poster olophid (fig. 9, B 14), and poster ostylid (fig. 9, B 15): When present, generally better developed than correspond- ing elements of upper molar. The posteroconulid (fig. 9, B 16) may be present or absent irrespective of presence of posteroconule. (b) Supernumerary crests: See definition below. Supernumerary crests, upper and lower molars: A crest, or series of crests, in the form of cuspules, lophs or laminas, may develop posteriad to the posteroloph(-id) or posteroconule (-id) of the third molar. Such supernumerary structures are absent in cricetines and murines but are present in otomyines (fig. 16, A), microtines and gymnuro- myines (fig. 16, D). They are here classified as postero- supernumerary crests, lophs, or laminas. The folds defining them are a continuation of the numerical sequence of secondary folds and minor folds (see p. 73). The Mesoloph in the Pentalophodont and Tetralophodont Patterns Figure 10 The complex, or pentalophodont, cricetine molar is characterized by five primary transverse crests. Those of the upper molar are defined by I, procingulum; II, paracone; III, mesoloph fused with mesostyle; IV, metacone; V, postcingulum. The primary transverse crests of the lower molar are defined by I, procingulum; II, meta- conid: III, mesolophid fused with mesostylid; IV, entoconid; V, postcingulum. The procingulum and postcingulum are described (see p. 74). All molars of oryzomyine, thomasomyine and some peromyscine rodents (see footnotes 1-3, p. 84) are pentalophodont. The secondarily simplified, or tetralophodont, cricetine molar is marked by the same primary transverse crests that characterize the complex molar except that crest III, the fused mesoloph-mesostyle (-id) complex is disjunct, incomplete or absent. Derivatives of the tetralophodont molars are the trilophodont molar, with crest III and either crest I (procingulum) or crest V (postcingulum) obsolete or indefinable; the bilophodont molar, with crests I, III, and V obsolete or indefinable; and the cylindriform molar, with only crest II cer- 77 78 FIELDIANA: ZOOLOGY, VOLUME 46 tainly definable. No sharp line can be drawn between the tetralo- phodont and derivative patterns. As a rule, even the cylindriform molar in the uneroded state retains signs of the erstwhile discreteness of one or more of crests I, IV, and V. Cricetines with basic tetralo- phodont patterns are all the New and Old World species except those of the groups mentioned in the preceding paragraph. There is considerable variation in the degree of disjunction, re- duction, or obsolescence, of crest III, in certain cricetines of the tetralophodont groups. Here, mesoloph(-id) and mesostyle(-id) are independent of each other. Either element may be present in the absence of the other, or both may be present in some molars and absent in others of the same specimen. In no case are mesoloph (-id) and mestostyle(-id) simultaneously present in all upper and lower molars. Hooper (1952, pp. 177-183) attempted to show that the dental pattern of the tetralophodont group, represented by Reithrodontomys (s.s.), graded into that of the pentalophodont group, represented by Aporodon. His observations were based on 2486 specimens of all species of the genera concerned. Hooper analyzed only the first and second upper molars and made no distinction be- tween independent mesoloph and mesostyle and the same elements fused as a single functional element, the mesolophostyle. Examina- tion of the same species in the collections of Chicago Natural History Museum and the University of Michigan Museum of Zoology reveals that in Aporodon a fused mesoloph-mesostyle is present in all upper molars and a fused mesolophid-mesostylid is present in all lower molars. On the other hand, in no species of Reithrodontomys (s.s.) are there either fused or independent mesoloph and mesostyle in the third upper molar, and in no species are there fused or independent mesolophid and mesostylid in any of the lower molars (fig. 11). In worn first and second upper molars of some specimens of Reithro- dontomys, the crown surfaces of mesoloph and mesostyle may be abraded to the same level and appear to be a single continuous mesoloph-mesostyle. The unworn molars in the same species, how- ever, reveal the mesoloph and mesostyle as truly discrete elements. The evidence presented here, and, inadvertently, by Hooper (see fig. 11) shows no intergradation in dental characters between Aporodon and Reithrodontomys. The discrepancy seems to be slight but is highly significant in a group of intimately related forms where differences are more often relative than absolute, and where parallel development of similar but independently acquired characters is commonplace. mis ml ms Lmsd msd FIG. 11.— A, B, right upper and left lower molars of Aporodon tenniroslris, showing well-developed mesoloph-mesostyle(-id) (mis); C, D, right upper and left lower molars of Reithrodontomys fulvescens with mesoloph (ml) obsolescent, mesolophid absent, and independent mesostyle (ms) and mesostylid (msd) present. Note the advanced stages of lamination in ma and sigmation in ms. (After Hooper, 1952, figures 4C, D, 5C, D.) 79 ,1-msd 80 FIELD IANA: ZOOLOGY, VOLUME 46 The mesoloph(-id) arose in some groups of Rodentia before crice- tines were defined. The mesostyle antedates the Rodentia. It is moot whether the ancestral cricetine molar was characterized by an independent mesoloph(-id) and mesostyle (-id) or by the fusion of the two elements. Either arrangement could have arisen indepen- dently and either could develop from the other. It is significant, however, that in living New World cricetines the fused mesoloph- mesostyle(-id) occurs only in the more generalized molars. With spe- cialization, mesoloph(-id) and mesostyle (-id) degenerate, lose their triturative function and become obsolete or disappear altogether in hypsodont, laminate, triangulate and all other advanced types of cricetine molars. On the other hand, some Old World murids, such as the Malagasy Gymnuromys (fig. 16, D), have retained the fused mesoloph-mesostyle(-id) through the processes described below (pp. 86-93) as planation, hypsodonty and advancing lamination. In another phyletic line, represented by Neocometes Schaub and Zapke (fig. 16, C) of the European Miocene, the fused mesoloph- mesostyle(-id) persisted through stages of planation and lamination. The "Pseudomesoloph" and the Dental Pattern in Cricetine Systematics In some cricetines a "pseud omesoloph (-id)" (fig. 12) appears in one or more molars, most commonly the first or second upper and the first lower. This false "mesoloph" is the paralophule (fig. 9), an enamel extrusion of the posterior border of the paracone. The false "mesolophid" is the entolophulid (fig. 9), which originates from the anterior border of the entoconid. In contrast, the true mesoloph (-id) is derived from the longitudinal enamel ridge (the mure, or Langs- grat) connecting protocone with hypocone (protoconid with hypo- conid). In many species lophule(-id) and mesoloph (id) are present in the same tooth as discrete elements. In cricetines such as the phyllotine Zygodontomys, without mesoloph (-id), the "pseudomeso- loph," i.e., the paralophule, is sometimes well developed. Its link- age with the mesostyle adds to its simulation of the true mesoloph. Hoffmeister (1951, p. 14, fig. 7) described and figured a "pseudo- mesolophid" (i.e., entolophulid) in mT of Peromyscus truei Shufeldt. Although he labeled and treated the element as a "mesolophid," Hoffmeister clearly distinguished (in text) this structure from the true mesoloph present in the upper molars of the same animal. Some time after the manuscript of this paper was completed and submitted for publication, Hooper's (1957) remarkably detailed HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 81 study of the accessory elements of the molars of the cricetine genus Peromyscus (sensu lato) came to hand. Unfortunately this work fails to distinguish paralophule and entolophulid from mesoloph and mesolophid, respectively. Thus, some of the 8 diagrams presented by Hooper (op. cit., fig. 1, p. 9) to show the relationship between mesostylid entolophulid ("pseudomesolophid") FIG. 12. — Left lower first molar of Peromyscus truei gilberti (after Hoffmeister, 1951, p. 14, fig. 7). The mure (or Langsgrat) is the median longitudinal enamel ridge connecting metaconid with entoconid. The true mesolophid (absent) is a diverticulum of the mure. The "pseudomesolophid" is the ento- lophulid, a ridge protruding from the entoconid. (For names of all parts of the tooth see fig. 9, p. 71.) what are termed mesoloph (-id) and mesostyle(-id), require other in- terpretations. The first diagram (with caption " + ") shows meso- style(-id) free, as described. On the other hand, the succeeding diagrams, labeled 1 to 5 inclusive, show mesostyle and mesostylid fused with paralophule and entolophulid, respectively, not mesoloph or mesolophid, as implied in the caption and text. The ridge with which the mesostyle (-id) is coalesced in diagram 6 may be the true mesoloph (-id). The correct identification of that element can be determined only in the real tooth. The transverse ridge in diagram 7 is rightly identified as the mesoloph (-id) fused with the mesostyle(-id). Bader (1959, p. 600) used Hooper's system for a statistical analysis of the dental traits in three species of Florida Peromyscus. Although he recognized that the true mesoloph (-id) and pseudo- mesoloph(-id) are not homologous, Bader treated them as one in computing their percentage of penetrance in each of the first two upper and lower molars. Misidentification or confusion of paralophule and entolophulid with mesoloph and mesolophid, respectively, may lead to a miscon- struction of the roles played by each of these elements in the evolu- tionary process. Paralophule and entolophulid are specializations that evolve after the mesoloph and mesolophid have become estab- 82 FIELDIANA: ZOOLOGY, VOLUME 46 lished. They may also originate while the mesoloph and mesolophid are becoming obsolete or have disappeared altogether. In the first instance, paralophule and entolophulid behave as accessories of the mesoloph and mesolophid, respectively. In the second instance, they act as functional replacements, i.e., as "pseudomesoloph" and "pseu- domesolophid," respectively. The metalophule and metalophulid (fig. 9) are analogous elements that may also be confused with the mesoloph and mesolophid, respec- tively. They are of infrequent occurrence and of no concern in the present discussion. Regardless of the identifications made or the terminologies used, Hooper's statistical analyses, figures and graphs clearly reveal the precise styles (-ids), lophs (-ids), and lophules(-ids) involved in the enamel pattern of each of the 17 species of Peromyscus studied. The accumulated data prove that in all populations of some species of Peromyscus molar patterns are essentially stable. In other species, the patterns are geographically variable. Where such variability exists, Hooper (op. cit., p. 48) points out that "populations that inhabit arid situations . . . have simpler teeth, with fewer and smaller styles and lophs, than populations of the same species that live in denser cover in more humid areas. This is in harmony with a cur- rent hypothesis, namely that the mesoloph (-id) is vestigial or absent in molars of cricetines inhabiting open country and scrubland. (Hershkovitz, 1955: 644.)" Actually, the phenomenon exemplifies more than a hypothesis. It is a demonstration of a basic principle in cricetine evolution. Variability in the dental pattern of some species led Hooper (op. cit., p. 53) to the conclusion that extraordinary reliance cannot be placed on accessory lophs and styles in the systematic analysis of the genus Peromyscus. No one would quarrel with this opinion. It is necessary to emphasize, however, that Peromyscus, as currently con- stituted, is unique among cricetines in containing some species which are ecologically and morphologically transitional between sylvan and pastoral groups. No question arises regarding the validity of dental characters in the systematics of species and genera whose evolution from sylvan to pastoral forms is no longer marked by contemporary annectants. Dynamics of Rodent Molar Evolution Dental modifications in the phylogenetic line leading to rodents were associated with hypertrophy of the incisor pair accompanied by HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 83 a progressive decrease in number of teeth and a compensatory in- crease in the functional area and efficiency of the cutting, crushing, or grinding surfaces of the remaining cheek teeth. Loss of the primitive first and third incisors was adequately compensated by development of ever-growing median incisors that maintain virtually permanent cutting surfaces on an even and efficient operating plane. Loss of the first two premolars, and suppression, or reduction in number or size, of the last two, may be attributable in part to the increasing demands for calcium and alveolar space by the middle incisors and in part to the usurpation of their functions by those teeth. In many rodents, additional specializations of the incisors as seizers, diggers or triturators are correlated with simplification and reduction in num- ber of the true molars. In extreme examples, the molars may degen- erate into simple tools that serve merely for gripping food, or for mixing and blending with saliva the food already adequately com- minuted by the incisors. With establishment of the definitive rodent dental formulae, earli- est molar modifications leading to the generalized pattern in cricetines and other rodent groups was accomplished by the process of plication, described below. Subsequent changes in molar structure have been effected by forces acting on the vertical and horizontal planes of the teeth. The processes described in the following as planation and hypsodonty operate in opposing directions of the vertical axis of the molar. Lamination, involution, triangulation and fusion model the molar in a horizontal sense. The first three horizontal forces are progressive and result in secondary complications. The fourth hori- zontal force, fusion, is retrogressive and results in secondary simpli- fications of the generalized plicident pattern. All processes involved are determined by inherent factors controlling growth and develop- ment. They operate in a definite direction throughout the evolu- tion of the organism. Plication Plication, or infolding, of the enamel covering of the primitive upper brachyodont cheek tooth defines the hypocone, procingulum, postcingulum, connecting ridges between the main tubercles, certain styles, and a mesoloph (cf. pp. 76-80; fig. 10, A). The mesoloph, when present, develops from the longitudinal ridge (mure of Wood and Wilson, 1936; Langsgrat of Schaub, 1925) connecting the pro- tocone and the hypocone and extends transversely nearly or quite to the outer border of the molar. Plication of the lower cheek 84 FIELDIANA: ZOOLOGY, VOLUME 46 tooth originates similar topographic features. The additaments of plication enumerated above result in a complex crown pattern of 5 transverse ridges (anticlines of Stehlin and Schaub, 1951, p. 30) separated by 4 valleys (synclines of Stehlin and Schaub, loc. cit.) opening on the outer side of the upper tooth and the inner side of the lower. The oldest known rodent with the complex plicident or pentalophodont molar pattern is the late Eocene Theridomys Jourdain (fig. 13) of Europe. Here the middle transverse ridge of the molar is the mesoloph fused with the mesostyle. Early crice- tines such as Eumys, Schaubeumys, Cricetodon, Heterocricetodon, etc., show the mesoloph and mesostyle as either fused or independent structures. The same basic pattern, with slight modifications, has been acquired, probably independently, in the upper molars of the more generalized representatives of Castoridae, Eutypomyidae, Eomyidae, etc. (cf. Wood, 1947; 1955, pp. 165-185). Recent American cricetines with complex plicident molars typi- fied by the fused mesoloph and mesostyle include the species of oryzomyine1, thomasomyine2 and some peromyscine3 rodents (figs. 4, A; 10, A; 11, A, B). These mice are primarily forest-dwellers. In deforested country they may survive in wooded ravines, shrub- covered cliffs and along brush-lined streams. They feed on her- baceous plants, pulpy fruits, soft seeds, fungi, and invertebrates. Grains do not constitute a significant part of the normal diet of woodland-inhabiting mice. Except for minor exceptions, it is un- likely that their diet and feeding habits have changed appreciably through geologic time. Adaptive radiation among these mice has affected external characters most, dental structure least. Known fossils of early Tertiary rodents of Europe and North America currently classified as cricetines are not ancestral to present forms. Most of them are more specialized dentally, but they may be offshoots of the same basic forest-inhabiting stock that has per- sisted without important modifications to this day. A second division of cricetines with plicident molars consists of nearly all remaining forms, including the phyllotines of this mono- graph, sigmodonts, akodonts and pastoral peromyscines (Podomys, 1 Composed of Oryzomys (subgenera:Microryzomys, Oecomys, Melanomys,Neso- ryzomys), Neacomys, Scolomys, Nectomys (s.g., Sigmodontomys) , Megalomys. For characters see Hershkovitz (1944, pp. 12-13). 2 Thomasomys (synonyms: Aepeomys, Inomys, Delomys, Erioryzomys, Wilfred- omys), Phaenomys, Nyctomys, Otonyctomys, Rhipidomys. The "Acodon" dorsalis Hensel, of Schaub (1925, p. 89, pi. 5, figs. 8, 10), the molars of which were care- fully described by him as typical of the present group, is a species of Thomasomys. 3 Peromyscus (Peromyscus, s.s. [part]; Ochrotomys, s.g.; Megadontomys, s.g.). HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 85 Haplomylomys, Onychomys, Baiomys, Reithrodontomys [s. s.], Neo- tomodon, Scotinomys, Nelsonia, Neotoma, Xenomys, Ototylomys and Tylomys). Their molars are characterized by all topographic fea- Pf, SFI SF2 PF2 FIG. 13. — Primitive pentalophodont molar plan in Theridomys (European Eocene; after Stehlin and Schaub, 1951, fig. 23): A, upper right second molar; B, buccal aspect of same. For explanation of symbols see figures 9 and 10; mls= fused mesoloph and mesostyle. tures of the primitive plicident tooth, except that the mesoloph (and mesolophid) is never locked with the mesostyle (and meso- stylid) in the unworn tooth; the mesoloph (id) is either reduced or absent in m1^-1, vestigial or absent in my--, and absent in m3. Thus, the unmodified crown pattern of the upper molar normally consists of 4 complete transverse ridges and 3 valleys opening on the labial border. The mesoloph sometimes present in m1 ? may be sufficiently developed to appear as an incomplete middle or "5th" crest (fig. 10, B). An occasionally present "pseudomesoloph"(-id) in the form of either a thickened mesostyle or an enamel evagination from the posterior border of the paracone (or from the anterior border of the entoconid) does not affect the basic crown pattern (fig. 12). According to Schaub (1925, p. 91) and Stehlin and Schaub (1951, pp. 169, 324, and elsewhere in the text), the simplified or tetralopho- dont cricetine molar differentiated from the complex, or pentaloph- odont, molar by regressive loss of the true mesoloph. This view is certainly correct. It is evidenced by the progressive simplification 86 FIELDIANA: ZOOLOGY, VOLUME 46 of the molar to the trilophodont, bilophodont and cylindriform pat- terns (fig. 10), by analogy with other regressive elements in the molars of living cricetines and all other mammals, and, most con- vincingly, by ontogeny. The American fossil record is poor. There is a predominance of specialized pastoral forms, which are uncritically classified. Nevertheless, it does not contradict the present concept of phylogenetic sequence. The possibility that cricetines with simpli- fied molars are collateral offshoots from a common ancestor with plicident molars in the premesoloph stage of development, can be discounted. The rudiment of a mesoloph, often present in tetra- lophodont molars of cricetines, is a vestige derived from, and not an emergent structure leading to, the pentalophodont molar pattern. The fused mesoloph-mesostyle(-id) is present only in bunodont, brachyodont molars. It may also be absent in such molars but it is never present in the specialized, hypsodont molars of cricetines. Living cricetines with the simplified molar pattern are primarily pastoral, i.e., inhabitants of deserts, tundras, punas, savannas, scrub- lands and coniferous forests. Sometimes they occur as intrusive ele- ments in bogs, meadows, or rocky situations within deciduous forests. A few aquatic species have survived in riverine habitats subsequently surrounded by climax forests; others may have invaded streams with banks deforested by man. The vast majority of pastoral species are ground-nesting, but many are good climbers and some have become semi-arboreal. A few species are aquatic. All feed mainly on grasses, lichens, mosses and grains. Some include pine needles and bark; others add insects, worms and small fish to the diet. The impact of this food, especially of harsh or siliceous grasses, on the molars, is considerably greater than that of the diet of forest-dwelling species with pentalophodont molars. The natural selective effects of the different feeding habits may account for the initial differentiation and subsequent radiation of the simplified, or tetralophodont, molar pattern. The six processes of molar evolution described below have acted on the tetralophodont cheek tooth. Only the first process, planation, has had a moderate effect on the complex pentalophodont molar of some American cricetines. Planation Figures 14, 49 Once a complicated system of dental peaks, ridges and valleys has been developed, the process of planation may set in. As a rule, crested terraced plane FIG. 14. — Molar planation: evolutionary stages from primitive crested to specialized plane shown in front views of molars (upper row) and side views of molar rows. l = protocone(-id); 3 = paracone in upper, metaconid in lower (after Hershkovitz, 1955:649). 87 88 FIELDIANA: ZOOLOGY, VOLUME 46 it advances simultaneously with hypsodonty. In either primitive (i.e., complex) or simplified molar types, early stages of planation are bilaterally unequal. Here the inner side of the upper molars and the outer side of the lower rise less rapidly. The terraced molar (fig. 14) that results is a recognizable and phylogenetically important stage in the succession of dental topography. In many species with plane molars, a terraced condition may persist in the newly erupted but unworn molars of juvenals. As a rule, the upper molars are in a more advanced stage of planation than the lower, and the third molar is more advanced than the other molars of the same jaw. The pentalophodont pattern in living New World cricetines tends to retain its primitive bunodonty. Nevertheless, planation has pro- gressed as far as the terraced stage in a number of oryzomyine and thomasomyine species.1 On the other hand, the tetralophodont molars of New World cricetines show all stages of planation from the fully bunodont, or the crested, through the terraced to the completely plane. Examples of complete planation in molars of non-cricetine rodents with or without mesolophs, and in other orders of mammals, are too well known to require citation. Hypsodonty Figure 15 Hypsodonty is the evolutionary process that provides a longer wearing surface by an increase in the depth of the tooth. It is a rejuvenating process that counterbalances the effects of planation on teeth exposed to excessive attrition. Hypsodonty may result in either an elongate or a prismatic crown or in hypertrophy of the main tubercle or tubercles of the crown. These effects of hypsodonty may be described in two categories, as follows: 1. Coronal hypsodonty: Vertical elongation of the entire crown of the tooth at the expense of the root. This kind of hypsodonty is found only in molariform teeth specialized for crushing and grinding. It is typically present in grazing animals. New World cricetines with the primitive pentalophodont type of molars are forest-dwellers and, in the main, non-grazers. Coronal hypsodonty among them is absent or, at most, incipient. The quasi-pentalophodont molars of the Chi- nese hamster Cansumys, however, are not only hypsodont but seleno- dont as well, an exceptional tooth form in cricetines. The degree of 1 In the Malagasy murid Gymnuromys (fig. 16, D) specialization of the penta- lophodont molar has reached the wholly plane condition. HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 89 hypsodonty in cricetines with the simplified or basic tetralophodont molar pattern reflects the extent to which each species has become adapted to the diet of savanna and scrubland vegetation. In most phyllotine rodents, the molar crowns are slightly to moderately hyp- sodont. The condition is more pronounced in sigmodonts and neoto- mines. Among murids in general, maximum development of coronal hypsodonty at the expense of the roots has been attained by micro- tines. Caviomorphs offer additional examples of molar crown hypso- donty. The classic example of coronal hypsodonty is the prismatic molar of the modern equine. 2. Tubercular hypsodonty: Elongation of the coronal tubercle, or tubercles, at the expense of the remainder of the tooth, including the root. This type of hypsodonty is an adaptation for seizing, grappling, stabbing, cutting, chopping or cracking. It may appear in any kind of tooth. The middle incisor of all rodents is an example of tubercu- lar hypsodonty. The chisel-shaped tooth has developed in depth at the expense of its own root and the erstwhile root-encasing alveoli of the lost incisors, the canines and one or more of the premolars. The elephant tusk is another notable example of tubercular hypso- donty of the incisor. Canine teeth of many mammals are also hypsodont tubercles, but the root has persisted in most. Tubercular hypsodonty of the molariform teeth is well developed in many marsupials, insectivores and insectivorous bats. It is also present in some cricetines, particularly the insect-eating Onychomys and in the fish-eaters Rheomys, Ichthyomys, etc. Here, the molar cusps have developed into prong-like projections for seizing and cracking elusive, hard-coated, mobile prey. The spiked cusps of the insect- and fish-eating rodents are quickly eroded but not before the peculiar seizing and chewing patterns have become established. In the central African murid Deomys (Deomyinae) the molar tubercles are notably elongated and those of both sides of the upper molars are hooked backward. Evidently, the primary function of the mo- lars of Deomys (fig. 15, B) is the grappling and chopping of such wiggling, soft-bodied prey as earthworms and caterpillars. The molars appear to be of little value as crushers and grinders even when worn. The history of hypsodonty and its correlation with the type and mineral content of plants eaten by herbivorous mammals has been reviewed by White (1959, p. 211). This author defines three kinds of molar hypsodonty. Type 1, called cusp hypsodonty, corre- sponds roughly to my coronal hypsodonty. I do not, however, C '•v 03 .s •i g C s> ^ 03 ^ ^ C *""{ ^ O 3 *^^» s--^ *«* =c .^ en ^ 0) ^ £ o o ** Q 73 O • "8 en C a oj ^ 'oT »•*! *"~1 J> 03 ^H ^H •* _C ' ^ O> 73 £ 13 >«i c S o . 'en CD o 0 Q O O> T? ^ -4-5 73 B C3 ^ o> 'E 3 CO S w Jl ^ .2 ~ 73 £2 £ ,S > ^ ^H 03 . 03 S s K J * .S ^ •£« V •^ CO j£ C «5 <3 w ,| t*> — ^j t~, " O « § 1- O g 73 C F^ ^3 0) g OJ , 5 a r^ . •^ 03 ^ a 3 *^ U s -1 3 f-t f\ § -g o> 'en u 3 C "o ^ 1? ^ J^ '^ "03 *o 2 3 ^S QJ 0 _C «*-" «1 C 0 en O en X! Ol £•> S fe O _W +9 -£ « -C >» 5 5 psodonty ropean m O 1 a O O en a x . C8 §g- o i 1 £ en *S >> 3 ^ j_ a ^ >, •^ r^i 03 -2 >> c H 5 H i -« I- | O> u ^2 3 -C •" C -M -1 £ O v E-< H tyj -C j . 3 CO . C8 S H << ^3 m £ d Q ^ C o o> 03 j C^ 73 £ EH 90 CO 91 92 FIELDIANA: ZOOLOGY, VOLUME 46 regard the Suidae, included by White in type 1, as a family char- acterized by hypsodont molars irrespective of some individual ex- ceptions. White's type 2, tooth-base hypsodonty, should, by defini- tion, exclude anything assigned to type 1. It seems to me, however, that molars of most of the taxons included in type 2 can just as well be referred to type 1, while others, as suggested by White, can be treated as intermediate between types 1 and 2 or made subjects of a separate category. Type 3, named root hypsodonty, includes animals with molars which, in my opinion, can be in- differently referred to either type 1 or 2, and molars of other animals which are rootless according to all authorities, including White. Elsewhere, White (cf. p. 260) equates hypsodonty with "high crowned" only. Whatever species or genera he may have had in mind in erecting the three types of hypsodonty, White cites only supergeneric taxons, mostly families and orders, as products of each type. He feels (p. 261) that not enough is known of the larger groupings of hypso- dont mammals to permit a satisfactory categorization of the lower taxonomic grades. In the present discussion, however, I prefer to regard hypsodont teeth as parts of specific animals. Such teeth can be classified by any convenient standard which serves to illus- trate the effects of hypsodonty on their form and function. Lamination Figures 16, 17 Lamination is the process of transection of a molar crown by con- fluence of a fold of one side of the tooth with another of the opposite side. The lamina formed is composed of one or more elements (cusp, conule, loph or "lophostyle") of one side of the tooth combined with at least one analogous element of the opposite side of the tooth. Lamina are classified according to their orientation relative to the antero-posterior axis of the molar. 1. Transverse: Formed by confluence of opposing, transversely directed folds. Transverse folds are common among rodents in gen- eral. They are present in the simplified, or tetralophodont, pattern of some living cricetines, including phyllotines with unworn or slightly worn molars (fig. 44, c [m4], fig. 69, e [mT]), but are absent in the pentalophodont molars of New World cricetines. However, they occur in a highly specialized group of the European Miocene repre- sented by Neocometes Schaub and Zapfe (fig. 16 C). HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 93 2. Oblique: Formed (a) by confluence of a fold of one side with either the anterior or posterior alternating fold of the opposite side, or (b) as a transverse lamina but with the entire molar rotated approximately 20° to 90°. Oblique lamina are present in the molars of many caviomorphs and other rodents, including some living species of tetralophodont cricetines (fig. 68, c [m3-]). 3. Chevron: Formed by confluence of obliquely directed opposing folds. These folds are present in some caviomorphs and murines but are absent in cricetines. The figure 8 pattern in simplified tetralophodont molars of many cricetines is a condition of incomplete transverse lamination between opposite folds. The same design may also occur as a residue in lami- nated molars worn to the base of the folds. Subdivision of a lamina into its original components may be ef- fected by the processes of involution and triangulation, defined below. Involution Figures 16, 17 The whole or part of a planed molar crown may be modified into a C-, or E-, or 5-shaped pattern by varying degrees of penetration of each of the enamel folds. For this modeling process, the embryo- logical term involution is used. Thus, involution, like lamination, is one of the manifestations of plication in the specialized molar. The C (or < ) pattern is derived from involution of a single lamina (partial or complete; see Lamination above). A double C (or ^) pattern can be derived from a bi- or trilaminate molar. A C-shaped pattern may form, also, in third molars in which all folds except either the major or the second primary fold are obsolete or absent as a result of simplification or excessive wear. This C-shaped pat- tern of involution is not to be confused with a similar pattern formed by secondary fusion (see below) of two lamina on only one side of the tooth. The C-shaped pattern is not present in phyllotines, except sometimes secondarily in the third molar, as a result of excessive wear, rather than involution (cf. fig. 67, g [m*]). The sigmoid or 5-shaped pattern is derived from involution of a simplified molar with the procingulum and postcingulum greatly re- duced or absent (fig. 69, e, fig. 67, b-g [m7]). A modified sigmoid pattern in upper molars retains indications of the first primary and, sometimes, of the second secondary folds. In the simple sigmoid pattern, first primary and second secondary folds are suppressed. pstc. P»2 FIG. 16. — Lamination and involution (for names of all cusps, lophs and folds see fig. 9). See opposite page for explanation. 94 HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 95 The sigmoid pattern in lower molars is an inverted mirror image of that of the corresponding upper molars of the same side. Sigmodont rodents are so called because of the S-shaped pattern of their third lower molars. In Euneomys, the sigmoid pattern characterizes all molars. Both sigmoid and modified sigmoid patterns occur among phyllotines. The E, or epsilon, pattern is the sigmoid pattern with a well- developed procingulum in the upper molars and a well-developed postcingulum in the lower molars (fig. 67, d-f [m^]). Involution appears most commonly in cheek teeth in the planed, hypsodont stage of phylogeny. The process may operate before, after, or simultaneously with, the processes of lamination and tri- angulation. Triangulation Figures 16, F, 69, c, 115, 116 The process of triangulation is the modeling of circular or ovate cusps, cingula, and lophs of a planed hypsodont molar into more or less triangular-shaped elements with apices at crown borders. Tri- EXPLANATION OF FIGURE 16 A. Transverse lamina in the African murid (Otomyinae) Otomys ores/ex Thomas (upper right molar row). Note the supernumerary laminules and folds of the postcingulum (pstc.) of ma. B. Partial transverse lamination in the South American murid (Cricetinae) Irenomys longicaudalus Philippi (upper right molars). Note the 8-shaped pattern of m2. C. Transverse, oblique and involuted lamina in the European Miocene murid (Cricetinae?) Neocometex brunonis Schaub and Zapfe (lower left molars, after Schaub and Zapfe, 1953, pi. 1, fig. 13). D. Transverse, oblique and involuted lamina in the Malagasy murid (Neso- myinae) Gymnnromys roberti Forsyth Major (lower right molars, after Stehlin and Schaub, 1951, fig. 577, reversed). The resemblance in enamel pattern between Gymnnromys and Neocometex is striking. However, the order of size of the molars in the former is the reverse of that in the latter and indicates that each genus may belong to a different phyletic line within the Muridae. E. Oblique partial lamination in the West Indian caviomorph (Capromyidae) Plagiodontia hylaeum Miller (? = P. aedium F. Cuvier] (upper right premolar and molars). F. Chevron-shaped partial lamination in the South American caviomorph (Abrocomidae) Abrocoma bennetti Waterhouse (lower left premolar and molars); the general pattern of the upper cheek teeth is as in B, an unusual combination. G. Chevron-shaped lamina in the Philippine murid (Murinae) (Crateromys Kchadenbergi Meyer (lower left molars). 96 FIELD IANA: ZOOLOGY, VOLUME 46 FIG. 17. — Lamination and involution (for names of all cusps, lophs and folds, see fig. 9). A. Lamination and involution in second right upper molar of the South Amer- ican caviomorph Dinomys brannicki Peters. B. Lamination and involution in second right upper molar of the South American caviomorph Dactylomys dactylinus Desmarest. C. Involution in second left lower molar of the South American cricetine Phyllotis micropus, with resultant E=, or epsilon pattern. D. Involution in third left lower molar of Phyllotis micropus, with resultant modified S= pattern. E. Involution in second left upper molar of the South American cricetine Euneomys chinchilloides, with resultant S= , or sigmoid pattern. angulation may precede, follow, or operate simultaneously with, lamination or involution, or both, in the same tooth row. The process of triangulation has reached its highest degree of develop- ment in microtines. Good examples of triangulation also occur in the grass- and scrubland species of Neotoma. There are some in- stances of triangulation among phyllotine rodents, but Andinomys is the only notable example. HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 97 Fusion Figures 10, 18 Reduction or elimination of a fold between two dental elements leads to the partial or complete fusion of these elements. Reduction or elimination of the elements themselves is not fusion. Individual folds are reduced by compression and eliminated by isolation and /or coalescence. nil) Pt2 D FIG. 18.- -Fusion and oylindrification (for names of all cusps, lophs and folds, see fig. 9 and also compare with fig. 10). A. Isolation of the major fold (m/) in the second right upper molar of the heteromyid Dipodomys merriami Mearns, with resultant fusion of protocone (/) and hypocone (2). B. Lamination and fusion in second right upper molar of Heleromys gaiimfri J. A. Allen and Chapman. Fusion of hypocone (2) and posteroconule (16) is the result of isolation of second minor fold (nf 2} and its coalescence with second secondary fold («/ 2). C. Fusion of all cusps in first and second right upper molars of the geomyid Orthogeomys cuniculux Elliot as a result of isolation, compression and elimination of coalesced major fold (mf) and second primary fold (pf 2). The two molars have become cylindriform. D. Fusion of all topographical elements in each of the upper right cheek teeth of the late Oligocene cylindrodont Tsagnnomys altaicus Matthew and Granger, as a result of the elimination of all folds. All cheek teeth have become cylindriform. 98 FIELD IANA: ZOOLOGY, VOLUME 46 Compression. — With advancing hypsodonty and lamination (in- cluding involution and sigmation) enamel folds tend to become narrower antero-posteriorly (cf. Euneomys, fig. 69, e). In the Lago- morpha the folds are compressed to transverse serrated lines. Isolation. — True internal folds (fig. 9, IF) are isolated remnants of larger folds that opened on the margin of the tooth. They are present in the molars of New World pentalophodont cricetines but, with a few exceptions, have disappeared from the molars of New World tetralophodont cricetines. In pentalophodont cricetines, the first secondary fold, between paracone and mesoloph, is commonly isolated but the dental elements themselves are normally discrete, except in an extremely worn tooth. In some pentalophodont and many tetralophodont cricetines, the second secondary fold of the upper and the second primary fold of the lower molars, most fre- quently the third, may be mere enamel islands before they are effaced by slight attrition. Complete isolation, or loss, of the second sec- ondary fold in tetralophodont cricetines and other rodents, results in a trilophodont molar pattern. Elimination of the second second- ary and first primary of the upper molars produces the bilophodont or 8-shaped pattern. Isolation of the major and second primary folds that define the figure 8 leads to complete fusion (not loss) of all com- ponents of the primitive tri- or quadritubercular molar. Coalescence. — Two or more folds of the same side of the molar may coalesce as a consequence of compression, isolation, or obsoles- cence, of a cuspule or loph. The result is the ultimate loss of the identity of one of the folds. The internal folds of pentalophodont cricetines are often coalesced with either a primary or a secondary fold. In tetralophodont cricetines, where the mesoloph is absent or obsolete, the first secondary fold is coalesced with the second primary fold and has no separate identity. Confluence of a fold of one side of the molar with another of the opposite side, such as the union of major and primary folds in some species, results in lamination and is not coalescence. However, such confluent folds are also subject to the process of fusion. They may become compressed, isolated, and eventually eliminated. Cylindrification Figures 10, 18 The end product of molar evolution in most pastoral rodents and many other mammals is the cylindriform molar. This is a completely HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 99 fused molar. The process of cylindrification is, therefore, an exten- sion of fusion. However, the structural and time gaps between fusion of a secondary fold in the primitive pentalophodont molar and elimi- nation of the major fold in a simplified prismatic molar are so great that cylindrification must be recognized as a distinct stage, the last stage, in a line of molar evolution. Fusion and cylindrification are retrogressive processes as they apply to the enamel pattern of the molars. Evolution of the organ- ism itself, however, may be progressive pari passu with retrogression and loss of some or all of its dentition. Cylindrification is most common among the Edentata. The cy- lindrical cheek teeth of the Oligocene cylindrodontid rodent Tsagan- omys Matthew and Granger are representative. The processes of fusion and cylindrification are well exemplified in modern forms of Heteromyidae and Geomyidae. Differential Evolutionary Rates in Upper and Lower Molars of New World Cricetines Evolution of each upper molar in cricetines proceeds at a different rate and does not always involve the same stages of evolution as its equivalent in the lower jaw. Evolution of the last molar, upper or lower, is more advanced than in the tooth before it. Differential rates of evolution between molars individually and between those of the upper and lower jaws collectively, result from the position of the teeth in their mortar-pestle relationship. Structural changes in the upper molars, the passive mortar, reflect the shifting movements and changing shape of the pounding, pulverizing lower molars, the pestle. Because of its terminal position, the first molar has the most varied functions, hence the most complex design. The second molar, situated at the center of the masticatory surface, has the simplest functions and is the most uniform in structure. The third molar, crowded toward the fulcrum, has a minimum use and tends to dis- appear altogether. This is true of cricetines but not of rodents in general (cf. fig. 16, A, D, where the third molar has evolved into the dominant tooth). The process of planation, which normally proceeds apace with the compensatory process of hypsodonty, is usually more advanced in the upper molars. Among phyllotines, especially in Phyllotis, slightly terraced lower molars are commonly associated with plane upper molars. In any case, the third molar, upper or lower, is more 100 FIELDIANA: ZOOLOGY, VOLUME 46 planed than the second and the second is more planed than the first. At the same time, the upper molars may be laminating while the lower molars are undergoing triangulation or involution. The order in which each molar element degenerates and disap- pears is generally the reverse of the order in which it first appeared. The older the element in phylogeny the longer it persists in the indi- vidual and in the species. The last important structural complex to formulate the cricetine pentalophodont molar, namely the fused mesoloph-mesostyle(-id) is the first to disintegrate into its compo- nents and disappear altogether. As a rule, the mesolophid disap- pears first, then the mesoloph. Procingulum and postcingulum break down next. However, the evolutionary relationship between these complexes in upper and lower molars is not on the same order as that existing between mesolophostyle and mesolophostylid. Although the upper procingulum is less specialized than the lower, it usually becomes obsolete first. The original crests of the quadritubercular rodent molar are the last to be affected. The hypocone(-id) may be lost entirely but the protocone(-id), paracone, metacone(-id) and entoconid degenerate by fusion (cf. p. 99) into the cylindriform molar. The first cylin- driform cricetine molar is m-. This tooth is also the first to disap- pear in more advanced murids (cf. Desmodilliscus, Rhynchomys, etc.). The order of disappearance of enamel folds corresponds to that of the elements they define. Thus, in the upper molar, the first second- ary fold is lost first, the second primary fold last. Degeneration of molars through wear, i.e., ontogeny, follows essentially the same sequence as degeneration through phylogeny. Third^Molar The cricetine third molar is the last to appear in ontogeny and the first to disappear in phylogeny. All evolutionary stages of sim- plification from pentalophodont to cylindriform molar are displayed first in the third molar. This tooth is usually terraced and frequently plane, while first and second molars are still crested. The specialized character of lamination is most frequently established first in the third upper molar. This is balanced by the appearance of sigmation in the third lower molar. The second minor fold, a novelty of the tetralophodont molar, first appears in the third molar, usually the upper. Fusion and loss of cingula and hypocone occur in the third molar, usually first in the upper, well in advance of first signs of the same degeneration in other molars. HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 101 Phylogenetically significant dental characters in the process of disappearance become progressively less evident on the crown sur- face of unworn teeth. Conversely, new characters in the process of establishment become progressively more evident and persist longer through wear. A mesoloph in the process of disappearance (palin- genesis) and a second minor fold in the process of establishment (caenogenesis) may occur simultaneously in the unerupted or unworn third upper molar. The third molar of most pentalophodont, or sylvan, cricetines is not functional before the animal is subadult and sexually mature. In tetralophodont, or pastoral, cricetines, including phyllotines, the third molar erupts before or shortly after birth and is functional in the juvenal mouse. Time of eruption of the third molar advances with increasing complexity of the high-crowned molar. Conversely, eruption is retarded with degeneration of the low-crowned molar. Molar Roots The primitive number of roots of the cricetine first molar may have been four. The roots become fewer as the tooth becomes more specialized, and the reduction in number is greater in the lower than in the upper first molar. In the following list the number of upper and lower roots of the first molar of each phyllotine species is shown : } = Phyllotis darwini; P. griseoflavus; P. gerbillus; P. pictus; Pseudoryzomys wavrini; Galenomys garleppi; Eligmodontia typus; Calomys spp. ~ = Phyllotis amictis. 4 = Phyllotis sublimis; P. micropus; Calomys spp. -1.,4 = Phyllotis boliviensis. II = Zygodontomys spp. % = Andinomys edax; Chinchillula sahamcu. Upper Incisor Figures 19, 20 The generalized upper incisor of rodents is slender or moderately thickened, with the terminal half slightly recurved and the cutting edge lying on a horizontal plane or at a slight angle. The anterior 102 FIELDIANA: ZOOLOGY, VOLUME 46 face is smooth and normally pigmented yellow, orange or reddish. As a rule, the vertical length, measured in front, is about one-half the length of the diastema. The generalized upper incisor, common to most rodents, has been drastically modified in many species. The specialized upper incisor may be markedly thick or slender, extremely recurved (opisthodont), straightened (orthodont), thrust forward (proodont), or flexed inwardly. The anterior face may be unpigmented, grooved, or striated. The cutting edge may be acutely angular, and the whole tooth may be disproportionately lengthened or shortened. All modifications of the upper incisor are related to molar specialization. In many cases, a correlation with modifica- tions of the palate is obvious. A few of the more striking divergents from the generalized form of the upper incisor are described in the following. Degenerative. — The upper incisor may degenerate progressively with deterioration of the cheek teeth. This process is generally asso- ciated with a compensatory elongation and attenuation of the muzzle, especially notable in the diastemal portion of the palate. Incisor degeneration has not occurred among phyllotines. Other murids, however, afford examples. The upper incisor of the Celebesian Echi- othrix leucura Gray is short, weak, faintly grooved and orthodont. The diastemal portion of the palate in this species is elongate. The diastema of the lower jaw, however, is not proportionately produced but the very long slender lower incisor bridges the gap to make occlu- sion. The inner row of cuspules of the upper molars are reduced and the corresponding ones of the outer side of the lower molars are obso- lete or fused.1 The process of incisor degeneration associated with loss of pigmentation, diastemal elongation, and molar degeneration is extreme in the Philippine shrew rat Rhynchomys soricoides Thomas (fig. 20, A-C). Here, the first and second molars of upper and lower jaws are practically functionless, the third molars have disappeared and the lower incisors are short and extremely fine. In all, the max- illo-mandibular region approaches the condition existing in the xen- arthrous and monotreme anteaters. Rodents of the type described are small-eyed burrowers. They probably feed mainly on earth- worms and grubs in addition to soft, ripe fruit. Seizer-digger. — The upper incisor may increase its utility while the cheek teeth become simplified in form, specialized in function, and reduced in number. The highly specialized seizer-digger incisor 1 This condition simulates a reversal from the typical triserial cuspidate pattern of murines to the biserial arrangement characteristic of cricetines. OPISTHODONT D ORTHODONT B PROODONT ai- Fio. 19. — Forms of incisors in phyllotines: AC, curvatures with reference to the basal-incisive plane (a-6) and vertical-incisive plane (c d). D, incisors of Andinomys edax: 1, front view of upper and lower incisors; 2, cross section of upper incisors above cutting edge. E, incisors of Phyllotis danrini: 1 3, variation in outline of cutting edge of upper incisors; 4, lower incisor; 5, cross section of upper incisor above cutting edge. 103 104 FIELDIANA: ZOOLOGY, VOLUME 46 is, as a rule, slender, markedly elongate, proodont, slightly or not at all scored longitudinally on its anterior face, and lightly or not at all pigmented (fig. 20, D). It is adapted for use as a pick in digging and for cropping short shoots or seizing small insects, larvae, and earthworms in tunnels or in rocky or cavernous situations where the hands are of little or no assistance. The gnawing power of the seizer- digger is probably reduced. The diastemal portion of the palate in rodents with this type may be long or short but is always attenuate. Phyllotines with specialized seizer-digger upper incisors are Gale- nomys, Phyllotis boliviensis and individuals of P. sublimis (figs. 59, 62). Some species of murines combine moderately elongated proo- dont upper incisors of the seizer-digger type with an elongation of the diastemal portion of the palate. In these, the mandibular dia- stema remains short while the lower incisor lengthens extraordinarily to make occlusion with the upper. The Burmese and Assamese Rattus manipulus Thomas is a moderately specialized example of this type with molar degeneration incipient. According to Roonwal (1949, p. 100) this is a burrowing rat that feeds largely on insects and earthworms. The seizer-digger characters are more advanced in the incisor of the Philippine striped rat, Chrotomys whiteheadi Thomas. Its molars are more simple, with the third nearly vestigial. This animal is also an earthworm eater (Rabor, 1955, p. 207). The closely related Philippine rat, Celaenomys silaceus Thomas, is even more specialized in the same direction. Both Chrotomys and Celae- nomys are divergents from the same murine stock that gave rise to Rhynchomys. They are much less nearly related to Hydromys with which they are currently associated in the superfluous subfamily Hydromyinae. Extreme examples of the seizer-digger incisor occur in such unrelated rodents as the Andean cavy (Monticavia), the Chilean coruro (Spalacopus) , the viscachas (Lagostomus), the African mole rat (Cryptomys), the Asiatic burrowing vole (Ellobius) (fig. 20, D), and the Oregon gopher, Thomomys bulbivorus. In all, the incisors are wholly or partially unpigmented. Triturator. — The specialized triturator incisor is short or long, orthodont or opisthodont, thick, often deeply grooved longitudinally and usually well pigmented (fig. 20, E). Triturators are efficient gnawers, choppers, and whittlers, and they function as hoes if used in digging. Triturators supplement, and may tend to usurp, the grinding function of cheek teeth. The diastemal portion of the pal- ate of rodents with the triturator type of incisor tends to become thick and shorter relative to the basicranial length. The upper in- HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 105 r FIG. 20. — Rodent incisors: A, degenerative type in themurid (Rhynchomyinae) Rhynchomyx soricoidex Thomas, palatal aspect showing long palate and right upper molars; B, side view of same skull; C, mandible and left lower molars of same; D, xeizer-digger type in the murid (Microtinae) Ellobius talpinus Pallas; K, tritura- tor type in the muroid (Rhyzomyidae) Rhizomys pruinosm Blyth. cisor of some fossorial forms such as the African Tachyoryctes and the Indo-Chinese Nyctoleptes, combine the special digging qualities of both triturator and seizer-digger types. Their incisors are elongate, forward-projecting, thick and heavily pigmented. The triturating type of upper incisor is not highly developed in phyllotines. The orthodont upper incisor of Chinchillula sahamae 106 FIELDIANA: ZOOLOGY, VOLUME 46 is comparatively heavy but not definitely classifiable as a specialized triturator. The upper incisor of Phyllotis micropus is the heaviest in the genus but is still very near the generalized type. Incisors of Calomys, Eligmodontia, Zygodontomys and the species of Phyllotis, other than the one mentioned above, are generalized. Many exam- ples of highly specialized triturators occur in geomyids, bathyergids, caviomorphs and castorids. The example figured (fig. 20, E) is of the Asiatic bamboo rat, Rhizomys pruinosus. Flexed. — The upper incisor of Andinomys (fig. 19, D) is inwardly flexed, its cutting edge oblique with the sharp point on the outer side. The paired incisors form a two-pronged instrument. The tips of the lower incisors taper to a point that fits the angle between the upper incisors. Incisors of certain fish-eaters (Ichthyomys, Anotomys), the African dormouse (Claviglis), and the tortoise-headed rat (Lophi- omys) are of the same type. In Rhagomys and the wood rat (Neo- toma), the cutting edge of the upper incisors also forms prongs but the teeth are not flexed inwardly. Even more remarkable than the shape of the incisors is the great diversity in the form and feeding habits of the many species of rodents to which they are common. Grooved. — Grooves or striations on the anterior face of the upper incisors occur sporadically throughout the Rodentia. Their function or structural significance is unknown. With the exception of Apo- rodon, grooves or striations are normally found only among tetra- lophodont rodents. They are often present and usually well defined in the specialized triturating type of incisor but are absent or poorly defined in the seizer-digger type. The systematic significance of grooved incisors in murids is slight or nil. Nevertheless, their appeal as obvious "key characters" is great. At one time or another the name Reithrodon, used for a mono- typic genus of groove-incisored cricetines, was applied to such di- verse, groove-incisored forms as Phyllotis pictus, Sigmodon alstoni, Irenomys longicaudatus, Euneomys chinchilloides, Chelemyscus fossor, and some species of Reithrodontomys. Among phyllotines, incisors of some individuals of Phyllotis boliviensis are scored by longitudinal striae hardly or not at all visible to the naked eye. In P. sublimis distinct striae are usually present and in P. pictus striae are constantly present and usually well defined. Striae are sometimes found in the upper incisors of Chin- chillula sahamae. In Euneomys, a genus nearly related to phyllotines, the incisor is deeply grooved and of the triturator type. HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 107 Lower Incisor Variation in length, shape, form of cutting edge, and angle of projection of the lower incisor is correlated with mandibular mobil- ity, the size and shape of the mandibular processes, and the structure of the muzzle as a whole. Upper and lower incisors, therefore, do not show as close a morphological correspondence as exists between upper and lower molars.1 Analysis of variation in the cricetine lower incisor and the interpretation of its significance are subjects that have not been studied. However, the description of the man- dible of each species in the following pages includes a reference to the degree of projection of the lower incisor root. Also, the length and angle of the exposed anterior portion of the lower incisor are noted in the description of Galenomys (p. 464, fig. 63), in comparisons be- tween Phyllotis and the African Mystromys (p. 221, fig. 60), and in the general discussion of the upper incisor in the preceding section. Grooving of the lower incisors is rare in murids, absent in crice- tines. Where it does occur, as in the African Otomys, it is a variable character and is normally associated with the grooved triturator type of upper incisors. Incisor Indices Relationship of the incisors, upper or lower, to the molar plane and to the basal-incisive plane (see definitions below) varies from species to species. Characters revealed by measurements of these planes are not described in the text (see figs. 59, 60, 62, 63). Planes of reference used in the illustrations are defined as follows (figs. 19, 21): Basal-incisive plane. — Defined by a line from the tips of the upper incisors to the most inferior point of the basicranium (basioccip- ital, tympanic bulla, paroccipital process, or pterygoid, as the case may be). Vertical-incisive plane. — Defined by a line projected through, or tangential to, the upper or lower incisor and vertical to the basal- incisive or -mandibular plane. Basal-mandibular plane. — Defined as the line projected by those parts of the inferior border of the mandible that are in contact with a plane surface. 1 There is no foundation for surmising, as does Landry (1957, p. 223), that rodents seem to have "evolved a stable lower incisor form and variously adapted the upper incisor and rostrum to fit it." 108 FIELDIANA: ZOOLOGY, VOLUME 46 Molar plane. — Defined by a line connecting the peak of the first cheek tooth with the peak of the last cheek tooth of either upper or lower jaw. FIG. 21.— Side view of skull, showing incisive and molar planes: a, vertical- incisive plane; b, molar plane; c, basal-incisive plane; d, mandibular plane. \ greatest length condylobasal palatal bullar length less tube FIG. 22. — Dorsal and palatal aspects of skull, showing cranial measurements used in text, o, width of rostrum; b, least interorbital breadth; c, zyjjomatic breadth; d, mid-frontal width; e, width of braincase. 109 EXPLANATION OF FIGURES 23-25 1. Nasal 2. Frontal 3. Parietal 4. Interparietal 5. Occipital 6. Premaxillary 7. Maxillary 8. Jugal 9. Squamosal or temporal 10. Palatine 1 1 . Presphenoid 12. Basisphenoid 13. Parapterygoid plate (often fused to basisphenoid) 14. Pterygoid process (often fused to basisphenoid) 15. Alisphenoid (more or less fenestrated) 16. Auditory, or tympanic, bulla 17. Mastoidal, or periotic, capsule, or petrosal 18. Orbitosphenoid 19. Lachrymal a. Capsular projection for upper incisor b. Preorbital foramen c. Antorbital, or infraorbital, foramen d. Antorbital bridge (of maxillary) e. Incisive, or anterior palatal, foramen /. Palatal process of premaxillary g. Palatal process of maxillary h. Anterior palatal pit (and foramen) i. Posterolateral palatal pit (and fora- men) j. Parapterygoid fossa k. Sphenopalatine vacuity I. Mesopterygoid fossa (between pterygoid processes) m. Hamular process of pterygoid o. Petrotympanic fissure p. Foramen ovale q. Auditory bullar tube (eustachian canal) r. Auditory meatus s. Occipital condyle t. Foramen magnum u. Paroccipital process v. Glenoid fossa w. Carotid canal and fissures x. Mastoidal process y. Lambdoidal crest z. Temporal ridge aa. Optic foramen 66. Sphenopalatine foramen cc. Anterior lacerated foramen dd. Zygomatic plate (of maxillary) ee. Hamular process of squamosal //. Temporal vacuity (dorsal and ventral) 110 FIG. 23.— Parts of the dorsal aspect of a murid skull. Small figure shows divergent-sided frontals and angular fronto-parietal sutures. See p. 110 for key to numbers and letters. Ill 17 U FIG. 24. — Parts of the ventral aspect of a murid skull. See p. 110 for key to numbers and letters. 112 ~ c et £ E c Q ,0 o •— I 0. $ OT 113 EXPLANATION OF FIGURE 26 a. Ramus g. Mental foramen 6. Angular process h Inferior masseteric ridge c. Condyloid process . „ , n ., i. Supenor masseteric ridge d. Coronoid process e. Capsular projection *- Mandibular foramen /. Sigmoid notch k- Symphysis 114 FIG. 26.— Parts of the mandible of a murid. A, lateral aspect; B, medial aspect. See p. 114 for key to letters. 115 SYSTEMATIC REVISION OF PHYLLOTINES Characters of the Phyllotine Group of Cricetine Rodents External characters. — Body and limbs adapted for terrestrial life with normal ambulatory, slightly saltatorial, scansorial or natatorial movements; notable specializations for burrowing, jumping, climb- ing or swimming absent; body size varying from extremely small to moderately large for cricetines; pelage normally thick, the individual hairs long, fine and soft to moderately coarse, never hispid or spiny; juvenal pelage usually similar in color and texture to adult pelage; color of nose, ears and rump not reddish or in marked contrast with remainder of upper parts; tail from less than one-half of combined head and body length to one and one-half times longer, moderately to well haired, often pencilled and faintly or sharply bicolor for at least two-thirds of its length; eyes normal; ears moderate to large in size, always more than one-half of length of hind foot; hind foot (fig. 3) not markedly enlarged, well haired above, heel beneath hairy, plantar surface bare or hairy; pollex with nail; claws of fore and hind feet not specially modified, the middle claw, measured on ven- tral surface, one-half or less of length of corresponding digit. Cranial characters (figs. 22-26). — Palatal bridge produced pos- teriorly to or behind posterior plane of last molars, posterior border rounded or square, with or without a short median spine; postero- lateral portion of each palatine bone often excavated or depressed and always marked by a distinct perforation or by a reticulation of perforations or pits; mesopterygoid fossa more or less U-shaped or M-shaped, the pterygoid processes parallel-sided or slightly divergent posteriorly, but never defining a sharp A-shaped fossa along their entire length; parapterygoid fossa, seen from ventral surface, usually shallow and subtriangular or ovate in outline, lateral wall not parallel to inner wall (hamular process) but gradually merged into dorsal surface and anterior wall; anterior wall of parapteryoid fossa usually flattened, rarely undercut to form a deep fossa; sphenopalatine vacui- ties usually large, the lateral wings of the basisphenoid and pre- sphenoid thinly ossified and largely fenestrated; incisive foramina long and narrow, pointed behind and, with few individual excep- 116 HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 117 tions, extending posteriorly to or behind anterior plane of first mo- lars; rostrum not markedly tapered dorso-ventrally from base to tip; anterior process of premaxillary projected slightly or not at all be- yond an tero- vertical plane of incisors and never united with antero- lateral border of nasals to form a tubular projection or trumpet; portion of premaxillo-maxillary suture below antorbital foramen situated about midway between first molar and incisor and form- ing a 90°-135° angle to horizontal plane of palate; nasals behind roughly square, rounded or obtusely triangular, terminating from slightly in front of to slightly behind fronto-maxillary suture, never with a long pointed process extending deeply between frontals; supra- orbital edges square, ridged, beaded or produced as ledges, never evenly curved; zygomatic arches complete; interparietal well devel- oped, at least transversely, and with a distinct suture separating each of its borders from at least two- thirds of the width of parietal and occipital, respectively; opening of antorbital foramen on dorsal surface deeply excised, semicircular or ovate in outline, with more than one-half to nearly full width of zygomatic plate clearly exposed to view when seen from above; inner side of medio-lateral axis of antorbital bridge of zygoma never directed posteriorly; zygomatic plate comparatively high and broad, width more than one-half of least interorbital breadth, anterior border plane, slightly convex, con- cave, or, if deeply excised and provided with prominent spine on upper corner, the incisors not grooved. Dental characters. — Molars well developed, mesoloph(id) absent or vestigial, at least in m} §; mesostyle(id) when present never fused with mesoloph(id); ectoloph, enteroloph and second minor folds ab- sent; first upper molar with three or four roots, lower with two to four roots; occlusal surface of m^ square or broadly rectangular in outline, its width at posterior half approximately two-thirds or more of great- est length of tooth ; third molar usually fully erupted and functional in active juvenals; ma from one-half to three-fourths of length of m- ; procingulum of mr not isolated from cusps by confluence of first minor and first secondary folds; major and primary folds well opened, not compressed antero-posteriorly; first primary fold of m-3 reduced or absent, its length, when present, one-half or less of combined length of second primary and internal folds (usually fused except, frequently, in Calomys), long axis of second primary fold inflected forward, second secondary fold reduced to a small enamel island, or absent; second primary fold of ms obsolete or absent; outer surface of each incisor smooth or marked by one or two weakly defined grooves. 118 FIELDIANA: ZOOLOGY, VOLUME 46 KEY TO SECTIONS, GENERA AND SPECIES OF PHYLLOTINE RODENTS 1. Upper incisors orthodont, proodont, or opisthodont, with anterior surface smooth, grooved or striated; upper molars high or low crowned, with flat or terraced grinding surfaces, the anterior enamel walls of at least the inner upper and outer lower cusps not thrust upward and back to form rounded crests; procingulum of mx simple or bilobate, its occlusal surface flat, laminate, sub- triangular, or subcordate in outline; mesostyle absent or extremely rudimen- tary; ear, measured from notch, shorter to longer than hind foot; mammae never more than 8 (PHYLLOTIS section) 4 1. Upper incisors opisthodont, their anterior surface never marked by vertical grooves or striae; upper first and second molars low or moderately high crowned with crested surfaces, the anterior enamel wall of each of the four principal cusps and of procingulum of m1 thrust upward and inclined posteriorward to form well-defined rounded crests in unworn tooth; procingulum of miuniconu- late or biconulate, its outline in worn tooth broadly ovate with or without an anterior notch; mesostyle present or absent, when present usually fused with paracone. Ear, measured from notch, never longer than hind foot; toes of hind foot variable in length; mammae 6 to 14 (CALOMYS section) 2 2. Mesopterygoid fossa as wide as or wider than parapterygoid fossa measured at same plane; sides of supraorbital region markedly divergent, the edges strongly beaded, m- with or without anterior median fold, the procingulum never clearly divided into two conules in adult; mesostyle often well developed and fused with paracone; m^ nearly as long as to slightly longer than m^. Fifth hind toe short, its tip, less claw, not reaching distal end of first phalanx of fourth toe; mammae 8 Zygodontomys (p. 196) 2. Mesopterygoid fossa narrow, its width less than width of parapterygoid fossa measured at same plane; sides of supraorbital region divergent or nearly par- allel-sided, the edges beaded or square. mx with or without anterior fold, pro- cingulum uniconulate or biconulate; rrig distinctly shorter than m^; mesostyle absent or extremely rudimentary. Fifth hind toe long or short; mammae 6 to 14 3 3. Sole with three middle postdigital tubercles more or less united to form a thickly haired cushion; first postdigital tubercle small, naked; fifth postdigital tubercle reduced or absent; tarsal tubercle small, naked; fifth hind toe long, its tip, less claw, reaching distal end of first phalanx of fourth toe or beyond. Eligmodontia (p. 175) 3. Sole with six well-defined, naked plantar tubercles; fifth hind toe, less claw, barely or not extending as far as distal end of first phalanx of fourth toe. Calomys (p. 123) 4. Sole with three middle postdigital tubercles coalesced to form a thick, well- haired pad; size small, greatest length of skull less than 28 mm.; incisors opisthodont, never grooved Eligmodontia (p. 175) (Individuals with extremely worn molars may be confused with the Phyllotis section, hence Eligmodontia is keyed in this as well as in the Calomys section where it belongs.) 1*. Sole with three middle postdigital tubercles discrete or partially fused at their base, the tubercles (not surrounding area) entirely naked or set with a few short inconspicuous bristles; length of skull variable; angle of incisors variable, their anterior face smooth or grooved 5 5. Upper incisors opisthodont, ungrooved; zygomatic breadth not more than dis- tance from posterior tips of nasals to anterior border of interparietal; inter- orbital breadth at mid-frontal plane less than greatest width of rostrum; sides of supraorbital region divergent and slightly ridged or beaded; mesopterygoid fossa at base of hamular processes at least as wide as parapterygoid fossa meas- ured at same plane; sole, including plantar surface of heel, entirely bare; fifth hind toe, less claw, not reaching distal end of first phalanx of fourth; first hind toe, less claw, not reaching base of first phalanx of second toe; ear, measured HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 1151 from notch, less than 70 per cent of hind foot length; tail sparsely haired, untufted, and not markedly elongated or shortened. Pseudoryzomys icavrini (p. 208) 5. Upper incisors orthodont, proodont or opisthodont, grooved or smooth, zygo- matic breadth more than distance from posterior tips of nasals to anterior border of interparietal; interorbital breadth at mid-frontal plane more or less than greatest width of rostrum; sides of supraorbital region variable, the edges square, beaded or ridged; width of mesopterygoid fossa variable; plantar sur- face of heel hirsute, remainder of sole with or without hair; fifth hind toe long, its tip, less claw, extending to distal end of first phalanx of fourth toe; first hind toe, less claw, reaching at least to base of first phalanx of second toe; ear measured from notch, more or less than 70 per cent of hind foot length; tail variable in length, more or less hirsute, with or without a terminal tuft 6 6. Supraorbital region comparatively broad, the sides diverging from slightly behind angle of fronto-maxillary suture; interorbital breadth at mid-frontal plane more than greatest width of rostrum; supraorbital edges square or beaded, with or without projecting ledges 7 6. Supraorbital region narrow, the sides either parallel or convergent from slightly behind angle of fronto-maxillary suture to form constriction at approximately mid-frontal plane; interorbital breadth at mid-frontal plane equal to or, usually, less than greatest width of rostrum; supraorbital edges square, pinched or, sometimes, weakly beaded or with projecting ledges 11 7. Size large, greatest length of skull in fully adult more than 30 mm., width of braincase more than 13 mm., alveolar length of molar row 5 mm. or more, antero-posterior length of bulla (less tube) more than 5 mm. Phyllotis griseoflavus (p. 441) (Also "Graomys edithae" Thomas, p. 461) 7. Size small, greatest length of skull less than 30 mm., width of braincase less than 13.2 mm., alveolar length of molar row less than 5 mm., antero-posterior length of bulla (less tube) 5 mm. or less 8 8. Bullae little inflated, their antero-posterior length (less tube) approximately equal to or less than length of molar row Catomys (p. 123) (Individuals with extremely worn molars may be confused with species of the Phyllotis section, hence Calomys is keyed here as well as in the section with crested molars.) 8. Bullae well inflated, their antero-posterior length (less tube) always more than length of molar row 9 9. Habitat in western Peru 10 9. Habitat in northern Andes of Argentina Phyllotis hyppgaens (p. 462) (See also "Graomys edithae" Thomas, p. 461) 10. Ears, from notch, more than 20 mm., hairs of under parts and thighs not wholly white Phyllotis amicus (p. 438) 10. Ears, from notch, less than 20 mm., hairs of under parts and thighs wholly white Phyllotis gerbillm (p. 430) 11. Anterior face of each incisor marked by one or two vertical grooves, visible to the unaided eye 12 1 1 . Anterior face of each upper incisor not grooved though often with fine vertical striae hardly or not at all perceptible to the unaided eye. ... 14 12. Second upper molar with major and second primary folds extending obliquely beyond midline of tooth; procingulum reduced (obsolete or absent in worn tooth), the occlusal surface forming an S-shapea pattern; upper molar row markedly divergent posteriorly; incisors opisthodont; postero-lateral palatine excavations deep and greatly enlarged antero-posteriorly . .Euneomya (p. 493) 12. Second upper molar with major and second primary folds not extending be- yond midline of tooth; major fold in worn tooth more or less at right angles to midline; outline of occlusal surface of moderately worn tooth more or less 8-shaped, with or without a notch in upper outer corner to mark position of 120 FIELDIANA: ZOOLOGY, VOLUME 46 procingulum ; upper molar row more or less parallel-sided; incisors orthodont to proodont; postero-lateral palatine excavations not markedly developed . . 13 13. Tail less than 70 mm. and less than 65 per cent of combined head and body length; hind foot (dry, with claw) less than 25 mm. . .Phyllotis sublimis (p. 419) 13. Tail more than 70 mm. and more than 55 per cent of combined head and body length; hind foot (dry, with claw) usually more than 25 mm. Phyllotis pictus (p. 404) 14. Alveolar length of molar row more than 6.6 mm 15 H. Alveolar length of molar row less than 6.6 mm 17 15. Incisors orthodont or slightly opisthodont; second upper molar with procingu- lum obsolete or absent; primary and major folds of m3- penetrating as far as but not beyond midline of tooth; white of hips and lower sides of rump sepa- rated by black of outer sides of thighs Chinchillula sahamae (p. 415) 15. Incisors opisthodont; second upper molar with procingulum well developed; second primary fold of m2- penetrating beyond midline of tooth and apex of major fold; hips, rump and outer sides of thighs more or less uniformly colored buff to brown 16 16. Medial anterior borders of upper incisors meeting to form a broadly angular trough when seen in cross section; palatine bones deeply excavated postero- laterally; cusps of first two upper molars triangular in outline and arranged in zigzag pattern. Habitat, Andes of Peru, Bolivia and northern Argentina. Andinomys edax (p. 472) 1 6. Medial borders of upper incisors meeting in midline to form a convex outline when seen in cross section; palatine bones slightly or not at all excavated postero-laterally; cusps of first two upper molars elliptical or ovate in outline with those of outer side nearly or quite opposite their analogues of inner side. Habitat, Andes of southern Chile and southern Argentina. Phyllotis micropus (p. 391) 17. Plantar surface of hind foot well haired; skull markedly vaulted mid-dorsally. Galenomys (p. 464) 1 7. Plantar surface of hind foot bare; dorsal contour of skull flat or slightly convex, not vaulted 18 18. Prpcingulum of ma reduced or obsolete, of ma obsolete or absent; incisors opisthodont, their anterior aspects smooth; greatest zygomatic breadth nearly always less than distance from tips of nasals to posterior borders of last molars; tail from about 40 per cent shorter to over 25 per cent longer than combined head and body length 19 1 8. Procingulum of m2 well developed and persistent except in greatly worn tooth, of ma present but not always well developed; incisors proodont, orthodont or opisthodont, with or without vertical striae or weakly defined grooves on ante- rior aspects; greatest zygomatic breadth more or less than distance from tips of nasals to posterior border of last molars; tail always shorter than combined head and body length 20 19. Upper incisors slender, combined width of upper cutting edges less than alveo- lar length of m1; molar rows parallel-sided or slightly convergent or divergent posteriorly; lower first molar 4-rooted; zygomata not greatly expanded, tend- ing to be parallel-sided, their greatest breadth less than distance between posterior tips of nasals and anterior border of supraoccipital; tail longer or shorter than combined head and body length. Phyllotis darwini complex (p. 234) 19. Upper incisors thick, combined width of cutting edges equal to or more than alveolar length of m1; molar rows usually divergent posteriorly; lower first molar 3-rooted; zygomata well expanded posteriorly, the sides usually markedly divergent antero-posteriorly, giving a triangular outline to mid- cranial region; greatest breadth across zygomatic arches subequal to dis- tance between posterior tips of nasals and anterior border of supraoccipital; tail always shorter than combined head and body length. Phyllotis micropus (p. 391) HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 121 20. Upper incisors ungrooved, opisthodont; tail longer or shorter than combined head and body length . . 19 20. Upper incisors ungrooved or grooved, proodont or orthodont; tail always shorter than combined head and body length . .21 21. Tail less than 70 mm. and less than 65 per cent of combined head and body length; hind foot (dry, with claw) less than 25 mm.; lower first molar 3-rooted. Phyllotis sublimis (p. 419) 21. Tail more than 70 mm. and more than 55 per cent of combined head and body length; hind foot (dry, with claw) more or less than 25 mm.; lower molar 3- or 4-rooted 22 22. Incisors without striae or grooves; width across bullar meati one-half or more of greatest length of skull; color of upper parts of body pale; prominent ochra- ceous preauricular tufts present; guard hairs of rump not unusually long or conspicuously displayed; lower first molar 3-rooted. Phyllotis boliviensis (p. 410) 22. Incisors with weakly defined striae or grooves; width across bullar meati in fully adult individual less than one-half of greatest length of skull; upper parts of body pale; no preauricular tufts; guard hairs of rump unusually long and prominently displayed; lower first molar 4-rooted. . .Phyllotis pictus (p. 404) CALOMYS ZYGODONTOMYS ELIGMODONTIA FIG. 27.- — Distribution of the genera of the Calomys section of phyllotine rodents in South America and Panama. For Costa Rican portion of range see map, fig. 1. 122 CALOMYS SECTION The characters of the section are given in the key (p. 118). For its geographic range, see maps (figs. 1 and 27). Genus CALOMYS Waterhouse Calomys Waterhouse, 1837, Proc. Zool. Soc. London, 1837: 21 — subgenus of Mus (included species: bimaculatus Waterhouse, type, elegans Waterhouse, gracilipes Waterhouse). Burmeister, 1854, Thiere Brasiliens, 1: ix, 168— subgenus of Hesperomys, part (species: expulsus Lund, lasiurus Lund). Thomas, 1884, Proc. Zool. Soc. London, 1884: 449 — subgenus of Hespero- mys, part (species: bimaculatus Waterhouse, type, gracilipes Waterhouse). Hesperomys Waterhouse, 1839, Zoology Voy. "Beagle," pt. 2, no. 4, p. 75 — part (Mus bimaculatus only). Wagner, 1843, Schreber's Saugthiere, Suppl. 3: xii, 510 — part (species: bimaculatus Waterhouse, gracilipes Waterhouse, callosus Rengger, laucha Desmarest). Baird, 1859, Mammals of North America, p. 453 — part (characters, type of tribe Sigmodontes Baird). Fitzinger, 1867, Sitz. Akad. Wiss. Wien, 56, (1), p. 26— part (species: callosus Rengger, expulsus Lund). Coues, 1874, Proc. Acad. Nat. Sci. Philadelphia, p. 177, footnote — part (Mus bimaculatus Waterhouse desig- nated type). Winge, 1888, E Mus. Lundii, 3: 11 — (species: simplex Winge [Pleistocene), molitor Winge [Pleistocene], tener Winge, expulsus Lund). Thomas, 1888, Proc. Zool. Soc. London, 1888: 133— merged with Cricetus. J. A. Allen, 1891, Bull. Amer. Mus. Nat. Hist., 3: 291— "no substantial basis as a generic name." Osgood, 1909, North American Fauna, no. 28, p. 11 — taxonomic history. Thomas, 1916, Ann. Mag. Nat. Hist., (8), 17: 141 — (genus ; synonym, Calomys Waterhouse, 1837, preoccupied [!] byCaWo- mys d'Orbigny and Geoffrey 1830; species: callosus Rengger, carilla Thomas, ducilla Thomas, expulsus Lund, gracilipes Waterhouse, laucha Desmarest, lepidus Thomas, sorella Thomas, tener Winge, venustus Thomas). Tate, 1932, Amer. Mus. Nov., no. 541, p. 11 — taxonomic history. Ellerman, 1941, Families and genera of living rodents, 2: 445— part (all forms listed except Paralomys Thomas with Phyllotis gerbillus Thomas). Osgood, 1947, Jour. Mamm., 28: 168 — characters and relationships. Eligmodontia Thomas, 1896, Proc. Zool. Soc. London, 1896: 1020— part (Calo- mys Waterhouse and Hesperomys Waterhouse in synonymy). Necromys Ameghino, 1889, Act. Acad. Nac. Cie'nc., Cordoba, 6: 110, 120 type by monotypy, Necromys conifer Ameghino. Type species. — Mus (Calomys) bimaculatus Waterhouse, by origi- nal designation (=Calomys laucha Olfers). 123 124 FIELDIANA: ZOOLOGY, VOLUME 46 Included species. — Calomys laucha Olfers, sorellus Thomas, lepidus Thomas, callosus Rengger. Distribution Figures 27, 28 Temperate zone grasslands, scrublands, and forest fringes of South America from the pampas of Argentina north into Paraguay, Uruguay and southeastern Brazil, west into the Andes of Peru, Bolivia and the Sierras Pampeanas of Argentina; the genus may be represented in the Bolivian Chaco and has been accidentally im- ported into northeastern Venezuela, Curacao and Aruba where it is now established; altitudinal range from sea level to approximately 5000 meters above. Characters External. — Small, generally Mus-like in appearance; pelage as a rule long, thick and smoothly adpressed or moderately lax, upper parts buff or tawny to drab, under parts white to dark gray; tail usually shorter than combined head and body length, sometimes slightly longer, uniformly colored to sharply bicolor, without ter- minal brush; hind foot (fig. 3) small, first toe less claw not extending to base of second; fifth toe less claw not extending beyond articula- tion between first and second phalanges of fourth toe; plantar sur- face naked, with 6 small tubercles; ears small, their length from notch never more than hind foot length; postauricular tufts present or absent; mammae, 6 to 14. Cranial (figs. 29-32). — Skull lightly built; upper surface of fron- tals usually convex, sometimes flat, with or without a slight median longitudinal depression; distance across mid-frontal region more or less than greatest width of rostrum; sides of supraorbital region par- allel, concave or divergent, with or without beading or overhanging ledges; outline of fronto-parietal suture usually crescentic, some- times angular; zygomatic arches little expanded, usually convergent anteriorly, rarely parallel-sided; anterior border of zygomatic plate plane or slightly concave; pitted posterolateral palatal depressions shallow or deep; paired anterior palatal pits situated on maxillaries or across maxillo-palatal sutures on anterior half of palatal bridge; width of mesopterygoid fossa at anterior base of pterygoids less than width of parapterygoid fossa measured at same plane; parapterygoid fossa shallow or moderately excavated; antero-posterior length of FIG. 28. — Distribution of the species of Colomyv. 126 FIELDIANA: ZOOLOGY, VOLUME 46 bulla (less tube) from slightly more to slightly less than length of molar row; posterior border of mandibular ramus deeply excised. Dental (fig. 33). — Upper incisors opisthodont to nearly proodont, ungrooved; molar rows generally parallel-sided; upper first molar four-rooted, lower three- to four-rooted; m7 more than one-half to nearly as long as m^; crown crested, the cusps tuberculate, with an- terior enamel walls of upper molars slightly more projecting than posterior enamel walls, and both inclined posteriorly with a slight inward rotation; principal cusps of outer side arranged in echelon or in opposition with respect to those of inner; cusps of upper mo- lars ovate in outline, of lower ovate or tending to become triangular; anterior fold of m1 well developed, the procingulum distinctly bi- lobate in Juvenal with outer conule larger than inner; mesoloph absent; an inconspicuous mesostyle sometimes present; first primary fold of m-~- well developed, at least in unworn teeth; second second- ary fold obsolete in upper molars; first minor fold reduced or absent in m^, obsolete or absent in m^; major fold of ma sometimes coalesced with second primary fold; first internal fold sometimes, second in- ternal fold rarely, discrete in m^; second internal fold of m-3- discrete or coalesced with second primary fold; anterior fold of mT present at least in unworn tooth; minor fold present in m^ and in unworn m-s; second primary fold of m^ present, of m^ reduced or obsolete. Comparisons Small size, variable number of mammae, unspecialized hind foot, crested molars and complete absence of mesoloph are salient char- acters of Calomys. Superficial resemblance between the species of Calomys, Mus musculus, Phyllotis amicus and Pseudoryzomys wavrini is dealt with under appropriate species headings. Similarity between North American Baiomys and the three smaller species of Calomys (figs. 31-33) appears to be the result of convergence. Baiomys is practically inseparable from Calomys laucha in size, external and dental characters, and general shape of skull. It is readily distin- guished, however, from all South American cricetines except Aporo- don and Tylomys by its comparatively simple glans penis and bipartite baculum (cf. p. 58). It is separated from all phyllotines, including Calomys, by slight dorsal excision of antorbital foramen; nearly transverse, not markedly anteriorward, orientation of medio- lateral axis of antorbital bridge; by postero-lateral palatal pits often wholly or partially concealed from ventral view by paired swellings on posterior border of bony palate. Characters distinguishing Bai- HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 127 omys from Calomys include the hypertrophy into a process of the burr which marks the origin of the superficialis part of the masse ter muscle at root of anterior border of zygomatic process; mesopterygoid fossa always as wide as or wider than parapterygoid fossa measured at the same plane at anterior base of pterygoid processes; paraptery- goid fossa shallower; fifth toe of front and hind feet longer, always extending to base of second phalanx of fourth, or slightly beyond, and pale postauricular patches absent. Characters distinguishing Baiomys from Calomys are as trenchant as any which separate genera and supergeneric groups. Neverthe- less, their small size and M Ms-like appearance misled Husson (1960) and Packard (1960) into assuming the existence of a relationship between Baiomys and Calomys that is much closer than permitted by the facts. Husson's comparisons of Baiomys with Calomys are based on individually variable characters discussed below under the heading Calomys laucha (p. 143). Packard's (1960, p. 664) conclu- sion "on the basis of internal morphological characters studied (audi- tory ossicles, hyoid apparatus, and baculum) [that] Baiomys seems to be more closely related to a South American hesperomine [sic], perhaps Calomys, than to any North American cricetine" appears to be baseless. Bacula of Baiomys as figured by Packard (1960, p. 603) and other authors are fundamentally different from the baculum of Calomys and all other South American cricetines examined by me and others (cf. Hooper, 1959, p. 10). Regarding the auditory ossi- cles, if, as averred by Packard (1960, p. 606), those of Baiomys "closely resemble" those of Calomys and Thaptomys and "differ greatly" from those of Zygodontomys, then they must be dismissed as being devoid of taxonomic value on the generic level. Thaptomys is not particularly related to Calomys and has nothing to do with Baiomys while Zygodontomys is a phyllotine very nearly related to Calomys. As for the hyoid, Packard (1960, p. 603) finds that on the basis of shape "Baiomys seems to be most closely related to Ochro- tomys." No comparisons were made with the hyoid of Calomys and no evidence is presented that the hyoid of this or any other South American rodent was examined. Packard's corollary (1960, p. 664) that "pygmy mice were more widely distributed in the past than they are at present [and that] part of the ancestral stock . . . may have emigrated from North America into South America in a brief period in the Pliocene" is entirely without foundation. There is a striking resemblance between Calomys callosus, Zygo- dontomys and Phyllotis griseoflavus with respect to their broadly expanded supraorbital regions (fig. 34). 128 FIELDIANA: ZOOLOGY, VOLUME 46 Variation In all species color of head, upper parts and sides of body are Mus-like, under parts white to gray with or without a light buffy wash. Pale and dark color phases with intermediate stages are found in all populations. Juvenals, however, are often darker than adults. The maximum number of mammae noted in Calomys callosus and C. laucha is 14, but most females of these species show only 8. In C. sorellus the normal number is 8, but usually only 4 or 6 show. In C. lepidus there are 8 to 10 mammae, but only 6 may be detect- able in the dry skin. Pale postauricular tufts are present in all spe- cies, but not in all individuals of the species. The tail is always shorter than combined head and body length in lepidus and callosus and is usually shorter or sometimes as long or slightly longer in laucha and sorellus. Hind foot is extremely variable in size. Often an "abnormally" short- or large-footed individual is found in a large series with which there is agreement in all other characters. The same is true of other size characters. An extremely large individual of an older generation than that of other individuals of the same pop- ulation may appear deceptively like the sole representative of a dis- tinct species. Conversely, a young, precociously developed individual with worn molars may be outstanding in a series of older individuals with normally developed features. Remarks The crested, comparatively low-crowned molars of Calomys are primitive phyllotine characters, but absence of the mesoloph is an advanced cricetine character. Small size, proportioned ears, vari- able number of mammae, and terrestrial habits are indications of the generalized cricetine pattern, but short outer toes and tendency for reduction of tail are specializations. In general, however, Calomys appears to be the phyllotine that has departed least from the hypo- thetical forest-dwelling ancestral stock. The most generalized member of the genus is Calomys sorellus. C. laucha is practically indistinguishable but shows more than an incipient lateral expansion of the supraorbital region. This charac- ter attains maximum development in Calomys callosus. A parallel development of the expanded type of supraorbital region has oc- curred in the nearly related Zygodontomys and the more distant Phyllotis griseoflavus. HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 129 Taxonomic History Calomys Waterhouse, 1837, is the earliest valid name for rodents hitherto referred to Hesperomys Waterhouse, 1839, as restricted by Thomas in 1916. The name Calomys fell into disuse because it was erroneously believed to be a homonym of Callomys d'Orbigny and Geoff roy, 1830, a genus of Chinchillidae. However, Article 56 (a) of the International Code of Zoological Nomenclature states that "even if the difference between the genus-group names is due to only one letter, these two names are not to be considered homonyms." Use of Calomys by Jordan (1888, p. 321) with Hesperomys a synonym, is as a composite of 5 North American species of cricetines, none of which was included or is now included in the Calomys of Waterhouse. Calomys Jordan, therefore, falls as a junior homonym. Use of the name Hesperomys Waterhouse is one of the most em- broiled in the taxonomic history of cricetines. It was proposed, in 1839, to contain all American cricetines with the same biserial molar cusp arrangement seen by Waterhouse in the teeth of the mouse he described two years earlier as Mus (Calomys) bimaculatus. In addition to bimaculatus, Waterhouse included in Hesperomys the species [Peromyscus] leucopus, Symidon [sic= Sigmodon] hispidus, the "spe- cies of Neotoma," [Oryzomys] galapagoensis, [Phyllotis] xanthopygus and [Phyllotis] darwini. This broad, essentially supergeneric con- cept of Hesperomys was revised by Wagner (1843, p. 510) to include the South American Scapteromys, Oxymycterus, Abrothrix and Calo- mys, all described by Waterhouse in 1837 as subgenera of Mus. Wagner, and after him Burmeister (1854, p. ix), continued to em- ploy Hesperomys as a "supergenus" of South American cricetines. The literature of purely North American cricetines also became in- volved with Hesperomys. Baird (1859, p. 453) made it type of the tribe Sigmodontes (=Cricetinae of modern authors) and referred 10 species, now included in Peromyscus, to the genus Hesperomys. Still no genotype was designated, but one of the 10 species mentioned, leucopus, was also included by Waterhouse in the original definition of Hesperomys. Finally, Coues (1874, p. 177), pointed out that "Waterhouse, in drawing his comparison between Mus and the New World mice, took M. rattus and M. bimaculatus for such purpose; we may properly therefore elect the latter as technically the type." This statement is the first clear and unequivocal determination of the status of Hesperomys. The name, therefore, by virtue of its designated genotype, is an absolute synonym of Calomys Waterhouse. 130 FIELDIANA: ZOOLOGY, VOLUME 46 Hesperomys of Winge (1887, p. 11) is equivalent to Calomys as defined here, while his Calomys is the Oryzomys of modern literature. Winge distinguished Hesperomys from his Calomys (= Oryzomys) by the presence of a well-developed mesoloph in the molars of the latter and its absence in those of the former. He also observed that the molars of Hesperomys and Cricetus were similar in this respect. Thomas (1888, p. 133) went farther and proposed that the name Hesperomys "be abolished altogether and [its] species united with the Old-World Hamsters under the name Cricetus." J. A. Allen (1891, p. 291) also voted for the elimination of Hesperomys as an invalid name, but he objected to the transfer of its species to Crice- tus. Thomas (1896, p. 1020) compromised by sinking both Hespero- mys and Calomys in the synonymy of Eligmodontia Cuvier. This solution endured until 1916, when Thomas (1916a, p. 141) revived Hesperomys (instead of Calomys, supposedly invalidated by Cal- lomys d'Orbigny and I. Geoffrey), as the generic name for lauchas. FIG. 29. —Dorsal and palatal aspects of skulls of a, Calomyx hpidux lepidiis; b, C. lattcha laucha; c, C. sorellux; d, C. callosus callows. (Enlarged.) 131 FIG. 30. —Side view of skull and outer surface of mandible of a, Calomys lepidus lepidus; b, C. laucha; c, C. sorellus; d, C. callosus callosus. (Enlarged.) 132 1 2 : O s o 5 II _ « 11 133 FIG. 32. — Palate of a, Baiomys tay- lori; b, Calomys laucha; c, Calomys sorellus. (Enlarged.) 134 FIG. 33. — Right upper molars and lower left molars of a, Calomys sorellus; b, C. laucha; c, C. lepiditx; d, Baiomys taylori. (Enlarged.) 135 FIG. 34.— Dorsal and palatal aspects of skulls of a, Zygodontomys brevicauda; b, Calomys callosus; c, Phyllotis griseoflavus. (Enlarged.) 136 HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 137 DIAGNOSTIC CHARACTERS AND KEY TO THE SPECIES OF CALOMYS 1. Borders of supraorbital region of skull divergent from front to back, strongly beaded and forming broad ledges in fully mature individuals; mid-frontal width always more than greatest width of rostrum; interparietal usually well developed antero-posteriorly; greatest length of adult skull at least 24 mm.; alveolar length of molar row 4.0 mm. or more; head and body length more than 80 mm., tail more than 60 mm callostus 1. Borders of supraorbital region parallel, divergent or slightly concave mid- frontally, the edges square, never beaded and never forming ledges except, sometimes, in very old individuals; mid-frontal width more or less than greatest width of rostrum; interparietal usually narrow antero-posteriorly; greatest length of skull less than 26.5 mm.; alveolar length of molar row less than 4.3 mm. ; head and body more or less than 80 mm., tail more or less than 60 mm. . . 2 2. Tail less than 50 mm. long, whitish or pale gray; heel and proximal portion of soles covered with hair, the tarsal tubercle partially hidden from view; sides of supraorbital region of skull parallel or very slightly concave mid-frontally and never forming ledges; mid-frontal width less than greatest width of rostrum; posterior tip of lower incisor indicated as a ridge without capsular projection on outer side of mandible lepidus 2. Tail more than 30 mm. long, brown above, paler beneath; heel, only, covered with hair; sides of supraorbital region of skull divergent or parallel, the edges square or tending to form ledges in old individuals; mid-frontal width more or less than greatest width of rostrum; posterior tip of lower incisor encased in projecting capsule 3 3. Supraorbital borders usually parallel or slightly divergent mid-frontally, the edges never forming ledges; mid-frontal width equal to or less than greatest width of rostrum; habitat Peru sorelliw S. Supraorbital borders usually divergent throughout their length except at ex- treme anterior portions, the edges tending to form ledges in old individuals; mid-frontal width equal to or more than greatest width of rostrum . . . .laucha Calomys sorellus Thomas Eligmodontia sorella Thomas, 1900, Ann. Mag. Nat. Hist., (7), 6: 297. Osgood, 1914, Field Mus. Nat. Hist., Zool. Ser., 10: 166— PERU: La Libertad (mountains northeast of Otusco). [Hesperomys] sorella, Thomas, 1916, Ann. Mag. Nat. Hist., (8), 17: Hi- classification. Hesperomys sorella, Pearson, 1957, Breviora, Mus. Comp. Zool., no. 73, p. 1 — PERU: Puno (3 miles northeast of Arapa; 5 miles south of Asillo; Hacienda Calacala, 7 miles southwest of Putina). Calomys lepidus sorellus, Cabrera, 1961, Rev. Mus. argentino Cienc. Nat., "Bernardino Rivadavia," 4: 481 — classification. Hesperomys frida Thomas, 1917, Smithsonian Misc. Coll., 68, no. 4, p. 1— PERU: Cusco (type locality, Chospioc). Thomas, 1920, Proc. U. S. Nat. Mus., 58: 230, pi. 14, fig. 3 (skull) PERU: Cusco (Huaracondo; Chospioc; Torontoy; Querefrata; Anta). Calomys frida frida, Cabrera, 1961, Rev. Mus. argentino Cienc. Nat., "Ber- nardino Rivadavia," 4: 478 — classification. Hesperomys frida miurus Thomas, 1926, Ann. Mag. Nat. Hist., (9), 17: 314 —PERU: Junin (type locality, Yana Mayo, Rfo Tarma). Thomas, 1927, op. cit., (9), 20: 602— PERU: Paaco (Huariaca; Alcas). 138 FIELDIANA: ZOOLOGY, VOLUME 46 EXPLANATION OF FIGURE 35 Calomys sorellus: collecting localities and collectors. Type localities in boldface. Calomys sorellus (1-35) PERU 1. Huamachuco, 8 miles south, La Libertad. — P. O. Simons. 2. Otuzco, mountains northeast of, La Libertad. — W. H. Osgood and M. P. Anderson at 10,000 feet. 3. Macate, Ancash. — W. H. Osgood and M. P. Anderson. 4. Tullparaju, Huaraz, Ancash. — C. Kalinowski at 4300 meters. 4. Quilcayhuanca, Huaraz, Ancash. — C. Kalinowski at 4000 meters. 5. Catac (Hacienda), Ticapampa, Ancash. — C. Kalinowski at 3500 meters. 6. Cullcui, Rfo Maranon, Huanuco.— J. T. Zimmer at 10,400 feet. 7. Huanuco Viejo, Huanuco.— J. T. Zimmer at 12,700 feet. 8. Panao Mountains, Huanuco. — J. T. Zimmer at 10,300 feet. 9. Huanuco, Huanuco.— J. T. Zimmer at 12,200 feet. 10. Alcas, Huariaca, Pasco. — R. W. Hendee at 11,500 feet. 10. Huariaca, Pasco.— R. W. Hendee at 9000 feet. 11. La Quinua, Pasco.— E. Heller at 11,600 feet. 12. Carhuamayo, Junfn. — C. C. Sanborn at 14,500 feet. 13. La Oroya, Junm.— M. P. Anderson at 12,000 feet. 14. Yana Mayo, Rio Tarma, Junin. — Type locality of miurus; R. W. Hendee at 8500 feet. 15. Piso (Hacienda), Locroja, Huancavelica. — C. Kalinowski. 16. Mayoc, Locrojo, Huancavelica. — C. Kalinowski. 17. Huancavelica, Huancavelica.- — C. Kalinowski at 3680 meters; A. R. G. Morrison at 12,000 feet. 18. Lircay, Huancavelica. — C. Kalinowski at 3310 meters. 19. Tambo, San Miguel, Ayacucho. — C. Kalinowski. 20. Huanta, Ayacucho. — C. Kalinowski. 21. Puente Pajonal, Ocros, Ayacucho. — C. Kalinowski at 1900 meters; includes Hacienda Pajonal, 1900 meters. 22. Ocros, Ayacucho. — C. Kalinowski at 3150 meters. 23. Andahuaylas, Apurimac. — C. Kalinowski; locality includes Hacienda Mo- zabamba, 2300 meters; Hacienda Palmar, 2200 meters; Uripa, 3100 meters; Hacienda La Laguna, 3040 meters. 24. Quebrada Matara, Apurimac. — C. C. Sanborn; includes Hacienda La Vic- toria, 7000 feet; Hacienda Matara, 6200 feet. 25. Chirapata, Cosireni Pass, Cuzco. — E. Heller. 26. Torontoy, Cuzco.— E. Heller at 12,000 feet. 27. Chospioc, Cuzco.— Type locality of frida; E. Heller at 10,000 feet. 28. Huaracondo, Cuzco.— E. Heller at 10,850 feet. 29. Anta, Cuzco. — O. Garlepp. 30. Urco (Hacienda), near Calca.— J. M. Schmidt at 9500-10,000 feet. 31. Cuzco, Cuzco. — C. C. Sanborn. 32. Querefrata, Cuzco. — Not located; E. Heller. 33. Arapa, 3 miles northeast, Puno. — O. P. Pearson at 12,600 feet. 34. Asillo, 5 miles south, Puno.— O. P. Pearson at 13,000 feet. 35. Calacala (Hacienda), 7 miles southwest of Putina. — O. P. Pearson at 13,000 feet. FIG. 35. -Collecting localities of Calomyx sorcllm. See opposite page for explanation. 139 140 FIELDIANA: ZOOLOGY, VOLUME 46 Type.— Female, British Museum (Natural History) no. 0.6.6.29; collected November 28, 1899, by Perry 0. Simons. Type locality. — Eight miles south of Huamachuco, La Libertad, Peru; altitude, 3500 meters above sea level. Distribution (fig. 35). — Peruvian Andes from the department of La Libertad southward into Cusco and Puno. Altitudinal range from approximately 2000 to 4600 meters above sea level. Characters. — Larger of the two known Peruvian species of Calo- mys; pelage long, soft, moderately lax, underfur thick; tail from approximately 60 per cent to fully as long as head and body com- bined, exceptionally slightly longer; tail bicolor except, often, at extreme tip; fifth hind toe long but not extending beyond articula- tion between first and second phalanges of fourth toe; white or buffy postauricular tufts present but not always conspicuous; upper parts of body ochraceous to tawny mixed with black; a more or less de- fined, sometimes broken, ochraceous lateral line present; tip of muzzle usually ochraceous; under parts gray, rarely with a weak buffy wash; mammae normally 8, but usually only 4 or 6 detectable in dry skin ; borders of supraorbital region of skull (figs. 29, 30, 36) slightly di- vergent, the edges square, never beaded; mid-frontal width about equal to or, usually, less than greatest width of rostrum; inter- parietal extremely reduced antero-posteriorly; other external and cranial characters as given in the key (p. 137); dental characters (figs. 32, 33) as for the genus. Comparisons. — Calomys sorellus is distinguished from the allo- patric C. laucha chiefly by larger average size, fifth hind toe slightly longer, sides of supraorbital region less divergent, their edges less sharply squared. Comparison with the partially sympatric C. lepidus is made elsewhere (p. 160). The superficial resemblance between C. sorellus and Phyllotis amicus is restricted to character of pelage, color, and small body size. Variation. — A tawny and a dark, or olivaceous, phase are pres- ent in most series. Individuals may molt from one phase to the other. The molt cycle is individual rather than seasonal in present material. A gradient or other pattern of geographic variation in color or size is not evident. Taxonomy. — Specimens from northeast of Otuzco, La Libertad, here regarded as most nearly representing the type of sorellus Thomas, were identified as such by Osgood (supra cit.). Evidently this species was overlooked by Thomas when he described one of HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 141 FIG. 36. —Calomys sorellus. Dorsal and ventral aspects of skulls of two adults from Lircay, Peru. (X 2.) its representatives from southern Peru under the new name Hes- peromys frida. It was compared with Phyllotis, and some differences between Calotnys frida and callosus ("//. venustus") were noted. Calomys sorellus was again ignored when Thomas erected Hespe- roniys frida miurus. This was characterized as shorter-tailed than typical frida. Tail measurements of miurus and frida are given below, following their respective combined head and body meas- urements: 142 FIELDIANA. ZOOLOGY, VOLUME 46 miurus: head and body, 86; tail, 76; tails of additional specimens, 70-80 mm. frida: head and body, 102; tail, 91; tails of additional specimens, 82, 85, 85, 87 mm. The measurements reveal that the tail of the type of miurus is approximately 47 per cent of total length and that of frida is also about 47 per cent of total length. Furthermore, the tail length of miurus is approximately average for the series to which it belongs, while that of the type of frida is extremely long. Habits and habitat. — The two specimens recorded by Osgood from the northern Peruvian Andes northeast of Otuzco, at 11,000 feet above sea level, "were caught in tall grass and weeds growing about the base of rough limestone exposures on the very top of the mountains." This is all that is known of the life history of Calomys sorellus. Measurements. — See Table 2 (p. 189) . Specimens examined. — 130, all in collection of Chicago Natural History Museum. PERU. — La Libertad: Otuzco, 2. Ancash: Macate, 1; Hacienda Catac, Ticapampa, Tullparaju, Huaraz, 2; Quilcay- huanca, Huaraz, 2. Huanuco: "Panao Mountains," 3; Huanuco, 8; Huanuco Viejo, 1; Cullcui, Rio Maranon, 1. Pasco: Quinua, 9. Junin: Yana Mayo, 1; Oroya, 2; Carhuamayo, 1. Huancavelica: Hacienda Piso, Locroja, 3; Mayoc, Locroja, 1; Huancavelica, 15: Lircay, 11. Ayacucho: Tambo, San Miguel, 10; Huanta, 4; Ocros, 4; Puente Pajonal, Ocros, 3. Apurimac: Andahuaylas, 16; Quebrada Matara, 9. Cusco: Chirapata, Cosireni Pass, 1; Hacienda Urco, near Calca, 15; Cusco, 5. Calomys laucha Olfers. (Synonymy under subspecies.) Distribution (maps, figures 28 and 37). — Grasslands and scrub- lands in southern Minas Geraes, south through the Brazilian coastal states into Paraguay, Uruguay and the Sierras Pampeanas and plains of Argentina as far as the Rio Negro; west to the Andean foothills of extreme southern Bolivia; established as import colonies in Monagas, Venezuela, and the Netherlands West Indian islands of Curacao and Aruba; altitudinal range from sea level to approxi- mately 2500 meters above. Characters. — Smallest phyllotine within its geographic range; pel- age long, thick, smoothly adpressed; tail from approximately 35 per cent to 108 per cent of combined head and body length, brownish HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 14:1 above, beneath paler but never sharply defined white; fifth hind toe, less claw, not extending to tip of first phalanx of fourth toe (fig. 38); pale buff to white postauricular patches usually present; upper parts of body buff to tawny finely mixed with black; an ochraceous lateral line often present; under parts of body from sharply defined white to moderately well-defined grayish with or without an ochraceous wash, base of hairs of belly and chest slaty; mammae from 8 to 14; borders of supraorbital region of skull (figs. 29-31) divergent, the edges square, never beaded; mid-frontal width equal to or more than greatest width of rostrum; interparietal usually narrow antero- posteriorly. Other external and cranial characters as given in the key (p. 137) ; dental characters (figs. 32, 33) as for the genus. Comparisons. — Remarkable resemblances between the non- phyllotine Baiomys and Calomys laucha have been discussed under the generic heading (p. 127 and figs. 31, 32). Characters used by Husson (1960, p. 38) for distinguishing the two genera are variable and confused. This author compared his six specimens of "Baiomys" hummelincki ( = Calomys laucha) with four of the same species from Argentina, one specimen of C. lepidus from Peru, two of Baiomys taylori from Texas and four of B. musculus from Mexico. He con- cluded that Baiomys differed from Calomys laucha by its smaller, more rounded m-, with major fold obsolete or absent, and by the shape of its interparietal, which was said to be broader antero- posteriorly and shorter transversely, extending only about one-half the distance across the braincase. In the approximately 100 Chicago Natural History Museum specimens of Baiomys represented by B. taylori and B. musculus, m^ is often, if not usually, smaller and more rounded in outline than in most comparable specimens of Calomys laucha. The third molar of the latter, however, is so variable in size and shape that all conditions found in Baiomys are frequently duplicated in Calomys. Unworn m^ of Calomys laucha is usually indistinguishable from worn or unworn ma of Baiomys. The major fold is usually open to the margin in Calomys but sometimes appears only as an enamel island, quite as in "Baiomys" hummelincki. In Baiomys musculus the major fold is usually an enamel island, the marginal flexure being obsolete or absent. In contrast, the major fold of m:' in B. taylori is open at the margin in nearly all but the most worn teeth. Packard's (1960, p. 586) figures of the enamel pattern of the molars of Baiomys are not clear on this matter. His figure of m:i in B. taylori lacks the marginal flexure or any other indi- cation of the presence of the major fold, while that of B. tnusculux shows a weak marginal flexure. Variation in the structure of the O 1 co o CO O {3 0 WD 1 s O5 5> c J « ts 3 w 8 w « C i-I J • (-1 .£ a "S £ "So to *S *^ d) 1 § J M- 3 ^S = p i s i ^ 3 £U pq ^ £ H ^ s c S • ^ p*i 0 ' ^ a; ^ c2 | S 3 w pi 10 .- "O , O i 13 1 ' ^ i 4J O T3 . C J • 1 \^ ^ ^ nJ • JURE 37 rs. Type localities in boldface. Chumbicha, Catamarca (415 meters).- meters. Goya, Corrientes.^ — R. Perrens. Yacanto, near Villa Dolores, Cordoba.- 1000 meters. Canada Honda, San Juan.^ — E. Budin a Buenos Aires, Buenos Aires. La Plata, Buenos Aires.^ — O. Thomas. Ensenada, Buenos Aires.- — C. Spegazzin San Miguel, Buenos Aires. — J. Crespo. Henderson, Buenos Aires.^ — W. H. Osgo Torrecito, Buenos Aires.^ — W. H. Osgooi Los Inglesas, Buenos Aires. — H. E. Box Dorrego (=Coronel Dorrego), 10 km. Aires (113 meters). — C. C. Sanborn. RoUfo m. ,„,.., i},,,,,,,... AI',.QO n TV,™ Algarrobo, Buenos Aires (47 meters). — J Pichi Mahuida, La Pampa (119 meters) meters. Chimpay, Rfo Negro (170 meters). — C. Choele-Choel, Rio Negro.— E. Budin. :ERLANDS WEST INDIES (35-36) /-l A T) -0:4.4.. .. T> 1ir«~ lection. Aruba.^ — P. Wagenaar Hummelinck; A. ZUELA (37) Maturin, 42 km. south, Monagas. — B. B v~/ o K W Qti « O5 O i— 1 coco^f-'tfiocot— 0005 c > i— i eg CO T(< EH u 0 ZD X^ C~ ^ 1-1 eg (M eg eg eg eg eg eg eg eg eg c 5 co co co co y c o co fq co O 1 ^ > Z ^ O as d 0 0 1 o OT- £ S d o OS ^ ^ ? ^ o £ ? us a> <3 '-£ Z | £ i C 02 3 $ — 1 I * s| 2 J s PH M « O c 0 ^ ,-; c < 1 « > C 0 ^ 3 1 ?m. x .s s o ^ CO pq * W | 0 f- VI • 4J CO -U CO W ^ 2 W "* 1. p~* 05 i-^* 0) t-> O cS 2 *• 4 TJ pj ° ^-x * CO C ^ C C fe s « 0 c7 fi « i S * " "S TJ S § s § s. - — - -JS ' to a) f fe £ £ s s s 3 1 i 1 c » fe « •§ § i 1 3 i 3 1 0 V2 H P CO C 2 3 1-2 : « ^> as « i « § «3i ^ ^** i -e 05 e R «§ -U. ^ ° ^ 1 .S rt = JO •§ { 3 .5 ^ S3 ^ " ° SI fig ^ 1. • N 0 S*" 5 C tM s 3 o) .(Me 3 o C ^ = J oo -^ a E- ? C - 3 C a 5 -g ^ PQ ^ ^ 5 S -| H « £ 5 H 1° | C F^H v. * 9 7 » r 1 1 S i 1 I'l , § £ EH ^ PH 99 _ >s « ,2 -2 »^ _f* *^ tt O S W (j U 8 * <5 S « :§• ( £ • fe 6 > = S .« ^ u 8 1 . - <3 .J . °. _^ ,,-wJ o O ^ «T >> i H. rf ^^ "^ CO ^ 1 i 2 co c .2, 1. 1 3 . 1 W 3 >> v- * a 3 S3 ' i^-1 ^ e^ EH'C | "^e-rt JlJ03^ J O £~ _2 t~5 .^ s (J2; ^ ^C5 ja eg ci o T -3 H. ^ 1 °. :§ § i S { ?! J^I^J ; Jl "S 'S '+3 5 S £ ^ C _j "O "C '"^ .2 2 2. — «" ^utec'3 s^Xs J-si gllll §1-1 s o £ c« 2 £ £ 3 2 G -g .g .5 3 _n ^) w ^^ w n . i u< ^ /—^ C O Og o ce ~ — — -• E-" CQ rh O O ique Mend., 1: 333. 150 FIELDIANA: ZOOLOGY, VOLUME 46 [Mus?] laucha Illiger, 1815, Abh. Akad. Wiss. Berlin, 1804-1811: 108— nomen nudum. Af[z- mys bimacitlatHS bimaculatus Yepes, Rev. Inst. Bacteriol., 7: 1935: 227" only. Type, — None preserved; technical name based on Azara's pub- lished description of three specimens. Type locality. — One of Azara's specimens was taken in a pile of maize stalks on a farm in the province of "Buenos Aires," and a second in the "pampas" about 25° S. Lat. Buenos Aires in Azara's time included present-day Uruguay and the Argentine provinces of Misiones, Corrientes and Entre Rios. Azara did no work in the modern province of Buenos Aires. The second locality, at the twenty-fifth parallel, may be construed as being in the neighbor- hood of Asuncion, Paraguay, to which area the type locality is now restricted. Paraguay was selected by first reviser Olfers as type locality. Distribution (fig. 37). — Paraguay, Uruguay and Argentina from Rio Negro north into the Andean foothills of southern Bolivia. Alti- tudinal range from sea level to not over 2500 meters above. An established colony of imports 42 km. southeast of Maturin, Monagas, northeastern Venezuela, may have been transported from Bolivia, according to Butterworth (1960, p. 517). The precedence of the colonies in Curasao and Aruba, Netherlands West Indies, has not been determined. Characters. — Those of the species. Variation. — Generally as given for the species. Lauchax from Uruguay and eastern Argentina are slightly darker throughout than those of the Argentine Andes. A series of 10 specimens from Gua- challa and one specimen from Orloff, both localities in eastern Chaco, Paraguay, are nearest the Andean lauchas but average smaller and paler throughout with white under parts, lower third of sides of body, muzzle, arms and upper surface of hind feet. The Guachalla speci- mens, taken September 3-5, 1945, are in prime pelage; the Orloff in- dividual, dated November 15, 1945, is in old worn pleage. There 154 FIELDIANA: ZOOLOGY, VOLUME 46 is complete agreement in external characters between the Guachalla specimens and the colored figure of the type of Mus bimaculatus Waterhouse (1839, pi. 12), taken in June. Of 13 specimens from 42 kilometers southeast of Maturin, Mo- nagas, Venezuela, which were collected and recorded by Butter- worth (1960, p. 517), 4 are described as juvenals; the remaining 9, represented by skins and 6 skulls, are at hand. They include one juvenal and two subadults. Pelage of all is comparatively short and thin as in old summer pelage of their southern hemisphere relatives. Upper parts agree in color with the series from Guachalla, Chaco, Paraguay; under parts vary from white to grayish. The tail aver- ages shorter, actually and relative to head and body length, than in any other series examined but in this respect they are also most like the Paraguayan lauchas. Taxonomy. — First accurate characterization of mice of the re- stricted genus Calomys was presented by Felix Azara in the French edition (1801, p. 102) of his account of the mammals of Paraguay. The vernacular name "laucha" was applied to the three speci- mens described. In the later Spanish edition (1802, p. 96) only one laucha was described. There also appears in the Spanish edi- tion a small mouse called "bianco debaxo" or white-bellied mouse. Tate (1932b, footnote, p. 11) thought the bianco debaxo was a laucha for which the technical name Mus dubius Fischer was available. Cabrera (1961, pp. 477-478) agreed and treated the white-bellied bimaculatus Waterhouse and the gray-bellied bonariensis Osgood, as subspecies of Calomys dubius. Measurements given by Azara for the three lauchas of the French edition and those for the laucha and the bianco debaxo of the Spanish edition are as follows, in millimeters converted from Spanish inches. Hind foot Ear Total with from Edition length Tail claw crown laucha cf French 110 46 15 7 laucha 9 French 100 — — • — - laucha 9 French 81 — — — laucha - Spanish 127 54 16 9 bianco debaxo - Spanish 135 42 18 13.5 External measurements of the bianco debaxo are not those of any known Calomys. The animal might be a Thalpomys but this possi- bility need not be discussed here. Specimens from the type locality in the vicinity of Asuncion are not available. Topotypes of H. laucha laucha may be nearer the HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 155 small white-bellied specimens from Guachalla, Chaco, or they may be like the larger, darker Uruguayan and eastern Argentine lauchas. Whatever the affinities, too little is known of seasonal and geographic variation to prove subspecific distinction between any two popula- tions of the species. The figured skull accompanying the original description of the Uruguayan Mus bimaculatus Waterhouse (cited in the above syn- onymy) is inaccurate in several respects. It is shown with the in- terorbital constriction exaggerated and the nasals tapered behind to a point, instead of being square or broadly rounded. The mouse described and figured by Waterhouse (supra cit.) as Mus gracilipes is smaller than the type of bimaculatus but the difference in size lies well within the limits of individual variation. Measurements given by Gyldenstolpe (1932, p. 143) for "Hesperomys gracilipes" do not correspond to the type, as implied, or to any other member of the genus Calomys. The body size of the type of Hesperomys laucha musculinus Thomas, from Maimara, Jujuy, equals that of small adults of the larger species C. callosus Rengger. However, the molar row length in musculinus, given as 3.6 mm., is diagnostic of Calomys laucha. Presumably because of the disparity in gross size between the types of musculinus and laucha, Thomas subsequently raised musculinus to specific rank and described cortensis as a subspecies of it. In this connection Thomas observed that the type series of cortensis in- cluded "a single old male skull — one of those overgrown examples which often render distinction of size so difficult -[that] measures no less than 26 mm. in greatest length; but this is obviously abnor- mal, the type being of about the usual adult size [24.5 mm.]." This statement adds weight to my opinion that the type of musculinus, also with skull length 26 mm., is an "overgrown" individual of C. laucha. Specimens at hand from the type region of musculinus and its erstwhile subspecies cortensis cannot be separated from other Argentine and Uruguayan lauchas. Hesperomys murillus, described by Thomas in 1916, was recorded by the same authority in 1897 as Oryzomys laucha and again under this same name in 1898. Subsequent specific separation was based on the insignificant character of tail length, "which nearly or quite equals the head and body in length." Hesperomys bimaculatus bo- nariensis Osgood, like murillus, is nothing more than another name for the same small pampas Calomys already described from the same region as laucha, bimaculatus and gracilipes. No representative of 156 FIELDIANA: ZOOLOGY, VOLUME 46 Hesperomys murillus cordovensis is at hand, but there is nothing in the original description or in the distributional pattern of related forms that suggests racial distinction of the Cordoba laucha. A specimen from Chumbicha, Catamarca, collected by Emilio Budin and identified by Thomas as cordovensis, is not separable from other Argentine lauchas. The confusion attending the systematics of Calomys is com- pounded by use of such similar sounding trivial names as murillus, muriculus, musculinus, and cortensis, cordovensis. Baiomys hummelincki Husson from the island of Curacao, Nether- lands West Indies, most nearly resembles new summer pelage repre- sentatives of typical Calomys laucha laucha but is more brightly colored ochraceous-buff and smaller throughout. The original de- scription is based on a skin and skull (type) and a skull only from Curacao, and four specimens from Aruba represented by two skins with skulls and two specimens in alcohol. External measurements given are minimal for Calomys laucha but were taken from the shrunken, preserved specimens. Cranial measurements are also small. The figure of the type skull and its greatest length, 18.6, are those of a Juvenal. The largest skull, evidently of an adult, measures 19.3. The figures of the head and feet of one of the speci- mens in alcohol from Aruba are of a Juvenal. Copies of these excel- lent drawings are used here (fig. 38) for illustrating the external characters of Calomys laucha. Husson's description of the Curacao and Aruba specimens of Calomys laucha (= Baiomys hummelincki) and two skins and three skulls at hand of the same material agree best with the Venezuelan import colony. It has been suggested that the latter may have been transported from Bolivia but it is equally possible that it proceeded from Curacao or Aruba. This still does not determine the native habitat or habitats of the original imports. In any event, the differences between the West Indian and Venezu- elan populations are of the same order as differences between pocket populations in the southern hemisphere. Suppression of the winter pelage in the northern hemisphere lauchas is a niche variable. More collections are needed at regular intervals for measuring the rate of evolution of Calomys laucha in Venezuela and the Netherlands West Indies. Measurements. — See Table 3 (p. 190). Specimens examined. — 67. PARAGUAY. — Chaco: Guachalla, 10 (CNHM); Orloff, 1 (CNHM). URUGUAY.— Minos: Paso de Averias, 2 (CNHM) ; Polanco, 1 (CNHM). ARGENTINA.— Jujuy: HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 157 Maimara, 2 (MACN). Tucuman: Conception, 3 (CNHM). Cata- marca: Chumbicha, 1 (CNHM). Buenos Aires: Torrecita, 13 in- cluding the type of bonariensis Osgood (CNHM); near Henderson, 4 (CNHM); Dorrego, 10 km. northwest, 7 (CNHM); Algarrobo, 2 (MACN). Rio Negro: Chimpay, 7 (CNHM); Choele-Choel, 2 (MACN). VENEZUELA.— Monagas: Maturin, 9 (LACM). CURACAO.— Plan tage Jongbloed, 1 (RNHL). ARUBA. Shiri- bana, 1 (RNHL); Airport Dakota, Boca Morto, 1 (RNHL). Calomys laucha tener Winge Hesperomys tener Winge, 1888, E Mus. Lundii, 1, no. 3: 15, pi. '2, fig. 3 (skull). Gyldenstolpe, 1932, Kungl. Sv. Vet. Akad. Handl., (3), 11: 75— selection of lectotype. [Hesperomys] tener, Thomas, 1916, Ann. Mag. Nat. Hist., (8), 17: 141 — classi- fication. Eligmodontia tener, Shufeldt, 1926, Rev. Mus. Paulista, 14: 503, 508, 563, 568, pi. 2, fig. 6 (skin)— BRAZIL: Sao Paulo (Piracicaba). Type. — Female, Zoological Museum, Copenhagen, no. 252. Type locality. — Rio das Velhas, Lagoa Santa, Minas Geraes, eastern Brazil. Distribution (fig. 37). — Eastern Brazil; known from the states of Minas Geraes and Sao Paulo but probably occurs, also, in Goias and the southeastern coastal states of Brazil. Characters. — The original description of tener is based on a com- parison with the sympatric congener Hesperomys callosus expulsus. It remains to be shown that tener differs subspecifically from typical laucha. Remarks. — Vieira (1953, p. 148) has recorded "Hesperomys tener" from Sao Paulo. Measurements given for two specimens are not those of any known species of Calomys. Cabrera's (1961, p. 481) inclusion of "Ksperomys [sic] lepidus lepidus Sanborn, Public. Mus. Javier Prado, Zool. num. 5, 1950: 3," in the synonymy of Calomys tener Winge is an obvious typograph- ical error. Cabrera (op. cit., p. 480) cites the same reference cor- rectly under Calomys lepidus lepidus Thomas. Measurements. — Those of the lectotype and two cotypes are from the original description. External measurements taken from the dry skin: head and body, 95 ("stretched"), 82, 74; tail, 67 ("stretched"), 57, 61; hind foot, 16.5, 15, 16; ear, 13, 13.5, 12.5; greatest length of skull (of lectotype only), 19.7; zygomatic breadth, 12; upper molar row, 3.3. Specimens examined. — None. 158 FIELDIANA: ZOOLOGY, VOLUME 46 EXPLANATION OF FIGURE 39 Calomys lepidus: collecting localities and collectors. Type localities in boldface. Calomys lepidus lepidus (1-7) PERU 1. Carhuamayo, Junfn. — C. C. Sanborn at 14,500 feet. 2. Junin, Junin. — C. Jelski. 3. Tambo Polanco, San Miguel, Ayacucho. — C. Kalinowski. 4. San Jenaro, Santa Inez, Huancavelica. — C. Kalinowski. 5. Lauramarca, Cusco (3996 meters) . — O. Garlepp. 6. Lucre, Cusco. — O. Garlepp at 3500 meters. 6. Sucre, Cusco = Lucre (?.».). 7. La Raya Pass, Cusco.— E. Heller at 14,010 feet. Calomys lepidus ducillus (8-19) PERU 8. San Ant6n, Puno. — P. O. Simons at 3800 meters. 9. Picotani, Puno.— C. C. Sanborn at 14,000 feet. 10. Posoconi, Puno.— C. C. Sanborn at 13,500 feet. 11. Santa Lucia, Puno. — M. R. Portugal. 12. Chucuito, Puno. — C. C. Sanborn. 13. Collacachi, Puno. — C. C. Sanborn. 14. Yunguyo, 6 miles south, Puno. — C. C. Sanborn at 13,000 feet. 15. Pairumani, Puno.— O. P. Pearson at 13,000 feet. 16. Huacullani, Puno. — C. C. Sanborn. 17. Sumbay, Arequipa. — C. C. Sanborn. 18. Salinas, Arequipa. — C. C. Sanborn and J. M. Schmidt at 13,500 feet. CHILE 19. Ojos de San Pedro, Antofagasta. — C. Koford. Calomys lepidus carillus (20-21) BOLIVIA 20. La Cumbre, La Paz.— G. H. H. Tate at 15,200 feet. 21. Choro, Cochabamba. — P. O. Simons at 3500 meters; F. Steinbach at 3500 meters. Calomys lepidus argurus (22) ARGENTINA 22. Abrapampa, Jujuy. — E. Budin. Calomys lepidus lepidus " " _/ •// ducillus O " " •// w canllus FIG. 39. —Collecting localities of the subspecies of Calomyx Icpidns. See opposite page for explanation. 159 160 FIELDIANA: ZOOLOGY, VOLUME 46 Calomys lepidus Thomas. (Synonymy under subspecies.) Distribution (figs. 28, 39). — Temperate zone grasslands and scrub- lands of the southern half of Peru, western Bolivia, extreme north- western Argentina and northeastern Chile; altitudinal range between 3000 and 5000 meters above sea level. Characters. — Smallest of phyllotines in average size; pelage long, thick, smooth or crinkly; tail between one- third and two- thirds of length of head and body combined, whitish, buffy or gray above, slightly paler beneath, terminal portion sometimes darker than basal portion; fifth hind toe, less claw, not extending to tip of first phalanx of fourth toe; heel and proximal portion of sole hairy, the tarsal tubercle partially hidden; whitish postauricular patches present; upper parts of body buffy to tawny more or less suffused with gray anteriorly, lined or peppered with black posteriorly, the black tips of guard hairs prominent only in juvenals; under parts sharply de- fined gray or white with plumbeous basal portions of hairs showing through; lateral lines, when present, not well defined, often meeting posteriorly at base of tail; mammae 8 to 10 but usually not more than 6 detectable; supraorbital region of skull (figs. 29, 30, 40) par- allel-sided or slightly concave mid-frontally, the edges square, never beaded; mid-frontal width less than greatest width of rostrum; pos- terior tip of lower incisor forming a ridge without capsular projection; dental characters (fig. 33) as for the genus. Comparisons. — The small size and vole-like appearance are alone sufficient to distinguish Calomys lepidus from its three Mus-\ike con- geners. Cranially, C. lepidus is nearest sorellus. The externally sim- ilar though much larger vole-like Phyllotis sublimis and Galenomys garleppi inhabit the same parts of the altiplano of southern Peru and western Bolivia. Juvenals of Phyllotis sublimis have been con- fused with adult Calomys lepidus, but differences between the two are obvious when size and shape of ears and hind feet, cranial char- acters, and size and structure of incisors and molars are taken into account. Variation. — A bright buff, or tawny, color phase characterizes some series, and a dusky gray phase characterizes others. In the bright phase, head and shoulders are contrastingly grizzled in speci- mens from the more arid localities. In the dusky phase, upper parts and sides of body are fairly uniform in color except for the darker mid-dorsal line or band. Color of tail varies independently, but there are more dark tails among individuals of the dusky phase than in HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 161 FIG. 40. —Calomys lepidus. Dorsal aspect of skulls showing variation in size (-X 2). specimens of the bright phase. The color phases may represent two stages of seasonal pelage succession or each may be a constant in a given locality. The data are tabulated as follows: No. of Date of Calomys lepidus lepidus (Peru) specimens capture Phase Carhuamayo, Junfn 2 Feb. 21-23 bright Tambo Polanco, Ayacucho 1 Nov. 27 dusky San Jenaro, Huancavelica 1 Dec. 24 dusky Calomys lepidus ducillus (Peru) Yunguyp, Puno 3 Posoconi, Puno 5 Santa Lucfa, Puno 1 Sumbay, Arequipa 1 Huacullani, Puno 3 Chucuito, Puno 1 Collacachi, Puno 8 Picptani, Puno 1 Salinas, Arequipa 5 Calomys lepidus carillus (Bolivia) Choro, Cochabamba 1 June 27 dusky La Cumbre, La Paz 7 Feb. 12-15 dusky Variation in tone of bright and dusky phases parallels the con- dition in Phyllotis darwini from the same localities. The bright Carhuamayo and dusky Ayacucho and Huancavelica specimens of Calomys I. lepidus are much more saturate than corresponding rep- resentatives of C. I. ducillus. The Bolivian carillus is represented by individuals of one color phase which are slightly darker than dusky specimens of ducillus and practically indistinguishable from dusky lepidus. May 1-2 May 7-9 July 23 Aug. 18 Aug. 25-26 Sept. 8 Sept. 11-14 Sept. 15 Oct. 6-7 dusky dusky bright bright bright bright bright bright bright 162 FIELD IANA: ZOOLOGY, VOLUME 46 Most juvenals resemble small adults and growth continues through- out life. Habits and habitat. — The following observations on Calomys lepi- dus ducillus are quoted from Pearson (1951, p. 141) : "Only seven of these little mice were caught, most of them at night in grassy places or in stone walls such as those surrounding corrals in open country. Hesperomys was apparently more abundant than this catch indicates, for the owls at Pairumani (Puno, Lake Titicaca region) caught more Hesperomys than they did any other species. The nocturnal habits and grassland preferences of these mice make them natural owl prey, and it is possible that they are not attracted to trap baits (nuts were used for bait in most cases). Phyllotis sublimis was frequently trapped nearby. "A female on August 29 had an open vagina and thick uterine horns, but no female earlier than this contained embryos and none was caught later. Two males in July had testes 5.5 and 7 mm. long. "The only ticks encountered on any animal in the region were found on the ear of a Hesperomys from Pairumani." The only other note on the habits of Calomys lepidus is a mis- leading one made by Thomas. In connection with the description of C. lepidus argurus, Thomas (1919d, p. 130) observed that it was the "dry-area representative of the forest H. carillus." The type and only specimen of Calomys lepidus carillus known to Thomas is from the grasslands of Choro, Cochabamba, in the Bolivian Andes. Calomys lepidus lepidus Thomas Hesperomys (Calomys) bimaculatus var. lepidus Thomas, 1884, Proc. Zool. Soc. London, 1884: 454, pi. 42, fig. 2 (color), pi. 44, figs. 10, 11 (skull), fig. 12 (ear), fig. 13 (hind foot). [Hesperomys} lepidus, Thomas, 1916, Ann. Mag. Nat. Hist., (8), 17: 141 — classification. E[ligmodontia] lepida, Thomas, 1900, Ann. Mag. Nat. Hist., (7), 6: 298— comparison in text. Hesperomys lepidus lepidus, Sanborn, 1950, Mus. Hist. Nat. "Javier Prado," (A), Zool., no. 5, p. 3 — PERU: Junin (Carhuamayo). Hesperomys carillus marcarum Thomas, 1917, Smithsonian Misc. Coll., 68, no. 4, p. 1 — PERU: Cusco (type locality, Lauramarca). Thomas, 1920, Proc. U. S. Nat. Mus., 58: 231— PERU: Cusco (La Raya Pass; Sucre [= Lucre]. Calomys lepidus marcarum, Cabrera, 1961, Rev. Mus. argentine Cienc. Nat. "Bernardino Rivadavia," 4: 480 — classification. HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 163 Type.— Male, British Museum (Natural History) no. 85.4 1.43; collected by Constantin Jelski. Type locality. — Junin, Junin, Peru. Distribution (fig. 39). — Andes of central Peru in the Lago Junin and Rios Apurimac-Urubamba drainage systems, departments of Junin, Huancavelica, Ayacucho, Apurimac and Cusco; altitudinal range, from approximately 3500 to 4500 meters above sea level. Characters. — Upper parts tawny with or without a dark lining on posterior half. Variation. — As described for the species (p. 160). Remarks. — Specimens from Carhuamayo, Junin, are practically topotypes. One specimen from Huancavelica and another from Aya- cucho agree with the published descriptions of Calomys carillus mar- carum Thomas from Cusco. In view of the small difference of a clinal nature between lepidus of central Peru and ducillus of southern Peru the occurrence of a geographically intermediate race is extremely doubtful. Measurements. — See Table 4 (p. 191). Specimens examined. — 4, all in collection of Chicago Natural History Museum. PERU. — Junin: Carhuamayo, 2. Huancavelica: San Jenaro, Santa Ine"z, 1. Ayacucho: Tambo Polanco, San Miguel, 1. Calomys lepidus ducillus Thomas Eligmodontia ducilla Thomas, 1901, Ann. Mag. Nat. Hist., (7), 7: 182. (Hesperomys) ducilla, Thomas, 1916, Ann. Mag. Nat. Hist., (8), 17: 141 — classification. Hesperomys ducillus, Pearson, 1951, Bull. Mus. Comp. Zool., 106: 141 PERU: Puno (Pairumani). Hesperomys lepidus ducillus, Sanborn, 1950, Mus. Hist. Nat. "Javier Prado," (A), Zool., no. 5, p. 4— -PERU: Puno (Posoconi; Hacienda Picotani; Chucuito; Hacienda Collacachi; Santa Lucfa; Yunguyo); Arequipa (Sa- linas; Sumbay). Hesperomys lepidus ducillus, Koford, 1954, Invest. Zool. Chilenas, 2: 6, p. 95 — CHILE: Anlofagasta (Ojos de San Pedro). Calomys lepidus ducillus, Cabrera, 1961, Rev. Mus. argent ino Cienc. Nat. "Bernardino Rivadavia," 4: 480 — classification. Type.— Female, British Museum (Natural History) no. 1.1.1.1; collected June 28, 1900, by Perry O. Simons. Type locality. —San Anton, Lake Titicaca region, Puno, Peru; altitude, 3800 meters above sea level. 164 FIELDIANA: ZOOLOGY, VOLUME 46 Distribution (fig. 39). — Altiplano of southern Peru, northeastern Chile, western Bolivia and probably northern Argentina; altitudinal range, 3500 to 4500 meters above sea level. Characters. — Paler than typical lepidus. Variation. — As described for the species (p. 160). The series from Salinas, another from Collacachi, and the only specimen from Huacullani exhibit a strongly contrasting grizzling on head and shoulders. Remarks. — Specimens from Picotani and Posoconi are practically topotypical. The individual from the first locality is in the bright color phase, the series of five from the second locality is in the dusky pelage phase. Measurements. — See Table 4 (p. 191). Specimens examined. — 28, all in collection of Chicago Natural History Museum. PERU. — Arequipa: Salinas, 5; Sumbay, 1. Puno: Huacullani, 3; Yunguyo, 6 miles south, 3; Santa Lucia, 1; Picotani, 1; Chucuito, 1; Posoconi, 3 miles west of Asillo, 5; Hacienda Colla- cachi, 8. Calomys lepidus argurus Thomas Hesperomys carillus argurus Thomas, 1919. Ann. Mag. Nat. Hist., (9), 4: 130. Type. — Female, British Museum (Natural History) no. 19.8.1.21; collected February 17, 1919, by Emilio Budin. Type locality. — Abrapampa, Jujuy, Argentina; altitude, 3500 meters above sea level. Distribution (fig. 39). — Known from type locality only. Characters. — According to the original description, "essential characters apparently quite as in the true H. carillus, but colour throughout much paler." Remarks. — The pale argurus may prove to be the northern Argen- tine representative of Calomys lepidus ducillus. Thomas ignored both lepidus and ducillus in his description of argurus. Measurements. — See Table 4. Specimens examined. — None. Calomys lepidus carillus Thomas Eligmodontia carilla Thomas, 1902, Ann. Mag. Nat. Hist., (7), 9: 133. [Hesperomys] carilla, Thomas, 1916, Ann. Mag. Nat. Hist., (8), 17: 141 — classification. Thomas, 1920, Proc. U. S. Nat. Mus., 58: 231 — comparison. HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 165 Hesperomys lepidus montanus Sanborn, (part), 1950, Mus. Hist. Nat. "Javier Prado," (A), Zool., no. 5, p. 4— BOLIVIA: La Paz (type locality, La Cum- bre, 15,200 feet altitude); the type and one subadult only. Type. — Male, British Museum (Natural History) no. 2.1.1.55; collected July 1, 1900, by Perry 0. Simons. Type locality. — Choro, Cochabamba, Bolivia; altitude 3500 me- ters above sea level. Distribution (fig. 39).— Upper Rio Mad re de Dios drainage sys- tem in La Paz and Cochabamba Departments, Bolivia; altitudinal range, 3500 to 4800 meters above sea level. Characters. — Slightly darker than dusky color phase representa- tives of ducillus. Variation. — As for the species (p. 160). Remarks. — The difference between dusky phase near topotypes of Calomys lepidus ducillus and Bolivian carillus is slight. However, the suspicion that the dark Bolivian specimens represent the bright color phase of carillus makes at least provisional recognition of the name advisable. Hesperomys lepidus montanus Sanborn is based on a series of 4 subadults including the type, and 4 juvenals referable to Phyllotis sublimis. Two of the subadult Calomys are indistinguishable from a topotype of carillus, a specimen not available at the time of the description of montanus. The third subadult and, presumably, the type, are darker. Measurements. — See Table 4 (p. 191). Specimens examined. — 4. BOLIVIA. — Cochabamba: Choro, 1 (CNHM). La Paz: La Cumbre, 3 (AMNH). Calomys callosus Rengger. (Synonymy under subspecies.) Distribution (figs. 28, 41). — Forest fringes and scrublands of southern Brazil, Paraguay, Bolivian Chaco; plains of northern Ar- gentina, and eastern slopes of Andes from La Paz, Bolivia, into the Sierras Pampeanas in Argentina as far as Cordoba; altitudinal range from near sea level to approximately 200 meters above. Characters. — Largest species of the genus; pelage long and mod- erately lax, or short and adpressed; tail from approximately 60 per cent to 90 per cent of combined head and body length, faintly to sharply bicolor; fifth hind toe, less claw, extending from base to middle of first phalanx of fourth toe; pale postauricular patches pres- 0 O 00 OU 1 - o= 5 o' I f m « !! « 1 E c> 7 CQ .. c -H O TT « •c S i g D X « 5, » «H O ra *H O z o — h w >» A 5? ^^ --T —•" * U£§ < S ~ "^ S £ ~ -j jg N3 »dcdt-^oc oi <3o'~ B E > c, • £-§ c- T & 1 « c^ §5 O _o c o .c§ K o L~ t> 10 '3_ • m * 1— E § s j .2 1 1 1 1 * § 2" « W. - 2 - «ug2§5 S g § § 2 .2 x CQ 166 O o 167 168 FIELDIANA: ZOOLOGY, VOLUME 46 ent or absent; upper parts of body ochraceous to tawny with a fine or coarse mixture of black or dark brown; weakly defined ochraceous lateral band sometimes present; under parts gray with or without a FIG. 42. — Calomys callosus callosus. Dorsal aspect of skulls showing variation in size (-X2). thin buffy wash; mammae 8 to 14; borders of supraorbital region of skull (figs. 29, 30, 42, 54,6) divergent from anterior angle backward, strongly beaded and forming broad ledges in fully mature individ- uals; mid-frontal width always more than greatest width of rostrum; interparietal well developed; dental characters (figs. 43, 44, 56,6) as for the genus. Comparisons. — Calomys callosus is readily distinguished from all other members of the genus by larger size and specialized form of the supraorbital region. Large individuals of C. callosus resemble small ones of Zygodontomys (for comparison see p. 199 and figs. 34 and 44). C. callosus and C. laucha are sympatric throughout the greater part of their respective ranges. Both are Afws-like in ap- pearance and, to a certain extent, in habits. Diagnostic characters given in the keys and descriptions are adequate for separating them. Variation. — Individual and geographic variations in size lie with- in the same narrow limits. Color of adults in a given season is uni- form throughout the range of the species. However, one specimen from Goias, Brazil, has dark under parts. Young individuals aver- age darker than fully mature ones. Winter pelage is soft, long, and HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 169 thick; summer pelage is short, thin and, in warmer localities, slightly hispid. Pattern of seasonal variation is local and, in Andean local- FIG. 43. — Calomys callosus. a, palate; 6, lower molars of juvenal; c, d, worn upper and lower molars respectively, of adult. ities, may depend on altitude. Most marked condition of winter molt is found in specimens taken in June from Calilegua, Jujuy, and from localities in Tucuman. In a large series from Concepcion, Tucuman, 3 specimens taken in May are in good pelage, 9 speci- mens taken in June are molting and 4 are in good pelage. In the same series, 2 specimens taken in December are in well-marked summer molt. Four specimens taken in February from Paraguay FIG. 44. — Upper and lower right molars of a, Calomys callosus; b, Zygodontomys breeicauda; c, Phyllotis griseoflaviis. 170 HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 171 have a short new summer pelage that averages darker than the longer winter pelage of other Paraguayan members of the species. Habits and habitat. — Calomys callosiLs is the common mouse of gallery and scrub forests. It has invaded cultivated fields and houses, where it lives side by side with Mus musculus. In his de- scription of C. callosus, Rengger stated that the mice dig tunnels several feet long by about two inches wide and nest there in pairs. Rengger also observed the species foraging in daylight for seeds and roots. Rodents recorded by Kuhlhorn (1952, p. 116; 1954, p. 72) as Hesperomys musculinus or H. laucha musculinus are certainly refer- able to the large species Calomys callosus. Their weight, given as 20-30 grams for subadults and 35-38 grams for fully developed in- dividuals, corresponds to the weight of mice the size of C. callosus and averages more than twice the weight of mice the size of C. laucha. Kiihlhorn's specimens were taken in gallery forest where H. laucha is unknown, and no specimens were found, after intensive trapping for several weeks, in open country, or in campos, where H. laucha might occur. Kuhlhorn noted that the mice were most active at night but could still be observed in their runways between 6 and 7 o'clock in the morning. Although traps were always set, only one Calomys was taken at noon. The runways measured from 3 to 5 centimeters in width. Evidently, a mouse used only its own runway system. Once it was trapped out, no other mouse moved in. According to Kuhlhorn, the species nests in holes in underbrush or rotting tree stumps. The above observations were made during the dry season in April and May, 1938. No nestlings, embryos or lactating females were taken at that time. Stomach contents revealed mainly vege- table matter with a small quantity of insects. Unfortunately, mice collected by Kuhlhorn on Krieg's South American expedition were lost during the war. Calomys callosus callosus Rengger Mus callosus Rengger, 1830, Natiirg. Saug. von Paraguay, Basle, p. 231. Thomas, 1916, Ann. Mag. Nat. Hist., (8), 17: 182— species unidentifiable. Oryzomys callosus, Thomas, 1898, Boll. Mus. Zool. Anat. Comp., Torino, 13, no. 315, p. 3— PARAGUAY: Conception (Rfo Apa, Concepci6n); Chaco (Rio Pilcomayo); ARGENTINA: Jujuy (San Lorenzo) ; Corn entes (Goya). Eligmodontia callosa, Thomas, 1902, Ann. Mag. Nat. Hist., (7), 9: 241— ARGENTINA: Cordoba (Cruz del Eje). Hesperomys callosus, Thomas, 1916, Ann. Mag. Nat. Hist., (8), 17: 141- classification. 172 FIELDIANA: ZOOLOGY, VOLUME 46 H[esperomys] callosus callosus, Yepes, 1935, Rev. Inst. Bacteriol., Buenos Aires, 7: 227— ARGENTINA: all northern and central provinces. Oryzomys (?) venustus Thomas, 1894, Ann. Mag. Nat. Hist., (6), 14: 359— ARGENTINA: Cordoba (type locality, Cosqufn). [Hesperomys] venustus, Thomas, 1916, Ann. Mag. Nat. Hist., (8), 17: 141 — classification. Hesperomys venustus, Thomas, 1919, Ann. Mag. Nat. Hist., (9), 4: 154 — ARGENTINA: Salta (Tartagal). Thomas, 1920, op. cit., (9), 5: 190 — ARGENTINA: Jujuy (Villa Carolina). Thomas, 1925, op. cit., (9), 15: 578— BOLIVIA: Tarija (Yacuiba; Carapari). Thomas, 1926, op. cit., (9), 17: 604— ARGENTINA: Tucumdn (Cerro del Campo, Burruyacu). Calomys venustus venustus, Cabrera, 1961, Rev. Mus. argentine Cienc. Nat. "Bernardino Rivadavia," 4: 481 — fecundus Thomas, a synonym. Eligmodontia callosa boliviae Thomas, 1901, Ann. Mag. Nat Hist., (7), 8: 253 — BOLIVIA: La Paz (type locality, Rfo Solocame, 1200 meters above sea level; Yungas; Astilleros). [Hesperomys callosus} boliviae, Thomas, 1916, Ann. Mag. Nat. Hist., (8), 17: 141 — classification. Hesperomys callosus boliviae, Osgood, 1916, Field Mus. Nat. Hist., Zool. Ser., 10: 207— BOLIVIA: Beni (Trinidad). Sanborn, 1950, Mus. Hist. Nat. "Javier Prado," (A), Zool., no. 5, p. 3— BOLIVIA: La Paz (Pitiguaya). Hesperomys venustus callidus Thomas, 1916, Ann. Mag. Nat. Hist., (8), 17: 182— ARGENTINA: Corrientes (type locality, Goya, altitude, 600 feet). Hesperomys muriculus Thomas, 1921, Ann. Mag. Nat. Hist., (9), 8: 623 — BOLIVIA: Santa Cruz (type locality, San Antonio, Parapeti, about 250 km. south of Santa Cruz de la Sierra). Hesperomys fecundus Thomas, 1926, Ann. Mag. Nat. Hist., (9), 17: 321 — BOLIVIA: Tarija (type locality, Tablada, altitude, 2000 meters). Hesperomys musculinus, Kiihlhorn (not Thomas), 1952, Zeitschr. Saugetierk., 18: 116— habits. Hesperomys laucha musculinus, Kiihlhorn (not Thomas), 1954, Saugetierk. Mitt., 2: 72— BRAZIL: Mato Grosso (Alto Parana; Rio Ivinheima). [?] Necromys conifer Ameghino, 1889, Act. Acad. Nac. Cienc. Cordoba, 6: 120, pi. 4, figs. 17, 18 (dentition)— ARGENTINA: Buenos Aires (type locality, Buenos Aires [Pleistocene]). Type. — Not known to be in existence. The bulk of Rengger's Paraguayan collection was lost before it could be sent to Europe. It is possible, however, that a few specimens reached the natural history museum of Aarau, Switzerland. Type locality. — "Taken on the banks of the Rio Paraguay at about 27 degrees latitude," i.e., opposite mouth of Rio Bermejo, Department of Villa del Pilar, Paraguay. Distribution (fig. 41). — Mato Grosso, Brazil, Paraguay, Bolivian Chaco, plains of northern Argentina and eastern slopes of Andes HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 173 from La Paz, Bolivia, into the Sierras Pampeanas of Argentina as far south as Cordoba; altitudinal range to 2000 meters. Characters. — Those of the species except that the under parts are not dark gray. Taxonomy. — The original description of Mus callosus Rengger, published in 1830, is unmistakably that of the larger of the two known species of Paraguayan Calomys. There is absolutely no basis for the conjecture advanced by Thomas (1916b, p. 182) that callosus is the same as Mus musculus. Rengger compared callosus with, not to, as implied by Thomas, the Old World house mouse because the two animals are superficially similar and frequently occupy the same habitat. On pages 228-229 of his work Rengger makes clear the dis- tinction between the "vier einheimische [cricetine] and zwei aus- landische [murine] Nager" he found in Paraguay. Regarding meas- urements, those of callosus are practically the same as those of a number of specimens recorded or described by Thomas as members of the large species of Hesperomys (= Calomys). The name callidus proposed by Thomas for mice identified by authors as callosus Rengger is a gratuitous synonym. The characterization of Eligmodontia callosa boliviae Thomas does not indicate substantial difference from Paraguayan callostis. Meas- urements of the types of the two forms are not significantly different, and specimens at hand from various localities in the Amazonian drainage of Bolivia are not separable from the Paraguayan form. The name fecundus is based on southern Bolivian individuals with nothing more distinctive than possession of 14 mammae instead of the more frequent 10. In addition to the type, Thomas recorded under the name Hesperomys venustus many specimens from various localities in northern Argentina and southern Bolivia. Abundant material at hand from the same regions shows nothing by which venustus can be separated from so-called fecundus or either of them from representatives of true callosus. Another named form of south- ern Bolivian callosus is muriculus Thomas. This is said to be "readily recognizable by its comparatively dark colour." Measurements and coloration of the type indicate a young individual. An adult from Tacuara, Chuquisaca, just northwest of San Antonio de Parapeti, type locality of muriculus, is indistinguishable from comparable specimens of callosus. Measurements. — See Table 5 (p. 192). Specimens examined. — 113. BRAZIL.— Mato Grosso: Descal- vados, 7 (CNHM); Urucum de Corumba, 9 (CNHM). BOLIVIA. 174 FIELDIANA: ZOOLOGY, VOLUME 46 — Beni: Trinidad, 1 (CNHM). Santa Cruz: Buena Vista, 4 (MACN, 1; CNHM, 3); San Carlos, Ichilo, 2 (MACN, 1; CNHM, 1); Tacu- ara, 1 (CNHM). Tarija: Villa Montes, 2 (CNHM). PARAGUAY.— Chaco: Guachalla, Rio Pilcomayo, 6 (CNHM); Rio Pilcomayo, 15 miles above mouth at Rio Paraguay, 5 (CNHM). ARGENTINA. —Salta: Aguaray, 3 (MACN) ; Rio Santa Maria, 2 (MACN). Jujuy: Calilegua, 11 (CNHM). Tucuman: Burruyacu, 3 (CNHM); Tafi Viejo, 2 (MACN); Conception, 55 (MACN, 3; CNHM, 52). Calomys callosus expulsus Lund Mus expulsus Lund, 1839, Ann. Sci. Nat., Paris, 11: 233— BRAZIL: Minas Gerais (Rio das Velhas) ; name only. Lund, 1841, Afh. K. Danske Vidensk. Selsk. Nat. Math., 8: 134, 266, 280, 294— description and records. Schinz, 1845, Syn. Mamm., 2: 194 — diagnosis. Hesperomys expulsus, Burmeister, 1854, Syst. Ueber. Thiere Brasil, 1: 175 — BRAZIL: Minas Gerais (Lagoa Santa). Giebel, 1855, Odontographie, p. 50, pi. 21, fig. 76 (dentition). Giebel, 1857, Abh. naturw. Ver. Sachsen u. Thiiringen, Halle, 1: 191 seq., p. 5, fig. 8 (skull) — osteology. Winge, 1888, E Mus. Lundii, 1, no. 3: 16 (description), pi. 1, fig. 1 (head), fig. 2 (hind foot), pi. 2, fig. 4 (skull), 4a (molar) — BRAZIL: Minas Gerais (Lagoa Santa). Gyldenstolpe, 1932, K. Sv. Vet. Akad. Hand!., (3), 11: 75. [Hesperomys] expulsus, Thomas, 1916, Ann. Mag. Nat. Hist., (8), 17: 141 — classification. [Oryzomys] expulsus, Trouessart, 1898, Cat. Mamm., p. 528 — classification. [Zygodontomys] expulsus, Trouessart, 1905, Cat. Mamm., Suppl., p. 423. Type. — In Zoological Museum, Copenhagen. Type locality. — Lagoa Santa, Minas Gerais, Brazil. Distribution (fig. 41). — Eastern Brazil; known only from Minas Gerais and Goias but the range probably extends from the Rios Sao Francisco and Parana to the coast. Characters. — Possibly darker throughout than typical callosus. Remarks. — Cotypes of Mus expulsus Lund have been redescribed and figured by Giebel (supra cit.) and Winge (supra cit.). The ex- tremely detailed and accurate evidence leaves no doubt of the spe- cific identity of expulsus with callosus. Subspecific status of expulsus may be postulated, however, on biogeographic grounds. A specimen from Sao Domingo, Goias, the only one examined that can be re- ferred to expulsus, differs from typical callosus by its extremely dark under parts. Measurements. — See Table 6 (p. 193). Specimens examined. — 1. BRAZIL. — Goias: Barro do Rio Sao Domingo, 1 (CNHM). HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 175 Genus Eligmodontia F. Cuvier Eligmodontia F. Cuvier, 1837, Ann. Sci. Nat., Paris, Zool., (2), 7: 168. Thomas, 1916, Ann. Mag. Nat. Hist., (8), 17: 140- included species: typus Cuvier, type [elegants Waterhouse a synonym), hirtipes Thomas, moreni Thomas, morgani J. A. Allen. Elygmodortia Wiegmann, 1838, Arch. Naturg., 4: 388 misprint. Heligmodontia Agassiz, 1846, Nomencl. Zool. Mamm., Addenda, 5, Index Univ., pp. 136, 175 — emendation. Elimodon Wagner, 1841, Arch. Naturg., 1: 125 — misspelling. Palmer, 1904, N. Amer. Fauna, no. 23, p. 256 — "Elimodon," cited from Fitzinger, 1867. Eligmodon Thomas, 1896, Ann. Mag. Nat, Hist., (6), 18: 307— emendation. Type species. — Eligmodontia typus Cuvier. Included species. — Eligmodontia typus Cuvier. Distribution Figures 27, 45 From the Straits of Magellan northward through Magallanes Chile into Argentina as far as the Rio Parana on the east, and along the Andes on the west into northern Chile, and the Lakes Poopo and Titicaca drainage basins of western Bolivia and extreme southern Peru; altitudinal range, sea level in southern latitudes, to more than 4500 meters above in higher latitudes. EXPLANATION OF FIGURE 45 Eligmodontia typus: collecting localities and collectors. Type localities in boldface. Eligmodontia typus puerulus (1-12; 36-38) CHILE (1-3) 1. Churiguaya(?), Puna Zone, Tarapaca. — Presumably in vicinity of Parina- cota; G. Mann. 2. San Pedro, 20 miles east, Antofagasta. - C. C. Sanborn at 12,600 feet. 3. San Pedro de Atacama, Antofagasta (3223 meters). F. Philippi. BOLIVIA (4-6) 4. Oruro, Oruro. —P. O. Simons at 3700 meters. 5. Challapata, Oruro. —P. O. Simons at 3750 meters. 6. Pampa Aullagas. P. O. Simons at 3800 meters. ARGENTINA (7-12) 7. Abrapampa, Jujuy. — E. Budin at 3500 meters. 8. Chorillos, Los Andes (=Salta). -E. Budin at 4500 meters. 9. Corral Quemado, Catamarca. — E. S. Riggs. 10. Chumbicha, Catamarca. — E. Budin at 600 meters. 176 FIELDIANA: ZOOLOGY, VOLUME 46 11. Chilecito, La Rioja.— F. P. Moreno at 1200 meters. 12. Media Agua, San Juan.— L. E. Miller and H. S. Boyle at 2200 feet. PERU (36-38) 36. Santa Rosa, Puno.— 0. P. Pearson at 14,000 feet. 37. Moquegua. — O. P. Pearson. 38. Tacna.— 0. P. Pearson. Eligmodontia typus typus (13-35) ARGENTINA (13-34) 0. Corrientes, Department. — Supposed type region, but genus may not occur here. 13. Delta del Parana, Buenos Aires. 14. Bahia Blanca, Buenos Aires. — C. Darwin. 15. Peru Station, F.C.O., La Pampa.— F. H. F. Parkes. 16. Pichi Mahuida, La Pampa. — E. Budin at 650 meters. 17. Chos Malal, Neuquen.— E. Budin at 800 meters. 18. Agrio (Rfo), Neuquen. — J. Yepes. 19. Zapala, Neuquen. — E. Budin at 1062 meters. 19. Las Lajas, Rfo Agrio, Neuquen. — E. Budin. 20. Choele-Choel, Rio Negro.— E. Budin. 21. Pilcaniyeu, Rio Negro.— H. E. Box; E. Budin. 22. Huanu Luan, Rio Negro. — J. C. Peters. 23. Rfo Chubut, Chubut.— C. Burmeister. 23. Rawson, Chubut.— E. Budin. 24. Pico Salamanca, Chubut (=Comodoro Rivadavia). — E. Budin at 100 meters. 25. Piedra Clavada, Rfo Deseado, Santa Cruz (=Comodoro Rivadavia). — E. Budin at 200 meters. 26. Swan Lake, Santa Cruz. — A. E. Colburn. 27. Upper Rfo Chico, Santa Cruz.— A. E. Colburn. 28. Arroyo Else, Santa Cruz. — Used incorrectly for "Basaltic Canyons" ( ~Ztt*f'*fe^~Z-*f'*f'*r »r eoco-^rTCOcocococococococo I COCOCOCOCOTCOCOCOTCOCOCO °. cooo^^i- o ~ co' co' co' co' i •'ti 7 ,w ^ o 10 oo •— <^ ^ ^ ® t~" *o~ o ^ ^ ® •"* °~» " c~ o> 10 o co co cooTt-{i-ioo'cM ccTcot>-coco'io'co'»O'-o CO , — v Cj ^T oo ^-i 05 ^ oo t- ^ a £ hTs8?^ T i S" s ^* t> oo "5 ^^t~ oo c- t> t- • S cs «- -1 >-|Ci-lo.d'T^'^'-|Ct "^"^QC!^ I I tf oooo -2 -g ^ o. *• •!? oo co o 3 •— oo o • «— i i i— i - - c-i — "j^'^'^iOiOi ""'C O) « "c^-o^ N^N "I ISJ I U 00 CO T)O|— it-lO ^HU ja ~ toco -•Tfoo^-^O'TOi o>*~ljj"rt-a H co »OO i iooocoo^'t-r^j-r^'O — -— t~ oo t— 60 t^ O^ t~ t— "O v. 'C J< " — • • - •* — ' — '' — rt ea o ' oo co co ^^ - — •,— 100^* • — •• — ^ ^H „?.. -O —i m « *— ^ f~* on » » ^•i ^ O (f^ rt ^ rt ^ C-l •-cajosooa>ooooo>'-iooQOooo>o»oiooooooooajOioi-Ht^ »• •> ? G3 TS 5 s P II ijlr -* ' 189 "* g c3 ±3 O ^ I-6 ^ C^J i-HCO 10 10 o ;* ^o i CO CO* coco oq co P CO 2 «: -^ £ CO COU3 1 ^ S OO ^f CO °° 'S "3 CO co co CO co, N~ M* ^ cS 2 CO CO Tf CO t- t- 10 t- co •«* Tf c> ncO^o'cOT}' US co' ob oq T3 .2 co co co co co co co co co co co c* D CO CO CO CO co' co IM' CO g "3 a CO 'c 'Sfi c" ^4 i-HCO IO 1O o 'C *" 0 , — s ,, — ^ J o.2 *£^ CO C~* CO S" t- 0 £ ,_* r-5 co (^J I « SB •c 1— 1 1— 1 1— I 1—1 o «- u i 1 1 1 o °i • «H w . CO C & ( (~^ J_^ ] * TJ* "? "o C/TT3 "C ^ i— i 1-1 1— 1 1— I I™H o^ oi o «^ CO 5^ Tf •<* t>CO OO CO i-H Tj«eo const- - 00 lOO IM o co' 06" CO N C COrHrHrH COrH COCOrH 1-HrH,- HO -HC^CO c — s a.3 ~ ^ s w >c o S ^ 03 I .§ ^ 13 ^_, & I I I I I co^usiovo^col O H I I I I I TH rH TH rH TH rH TH [ .2 a> -u o £ 2 '~l "^ .fi t« Jis''-|i-1 ^-Tf«~~"''t^h*''-'T3 .S"-^ » •— iO Q5 i i O CD **i *^ ^ •— ' *™ 3T37 ^,0*0^7,2 Mg >•&.& «>> S .S co „ oo ^.^ oo co I f c« rtCO'C £ a>^ •gS ^t^oo opS°°rQ<1i«)'30|| t^ FHCO 05U3rH OrH||"fecO'3gS.J-3 TJ< Tj- TJ. >. t- 2B,LL"3-So2 Xffi o -^ OOCOTf"O5U3COCOO5"500lOt-CO"2 rHOOOOO5CDOOt-OOOOC-C-t-OOC~O5 a .- o> o rr 4-> O. *-• r Sau S C 3 « 00 t£> tD «5 t— _ co co co co co wee co w w co w co co co O j= H t-; oo eg' eg co eg -H eg CM eg eg eg ^H S 3 DC eg eg eg eg eg eg s £ »-i o w «o •§ Oi t^ oj oo oo os «o*co -HCOIO TO* to -c.5 II a£ I' Sc •o o s.s-.s-^ •sbs.^ 'C S'C c °~0 % 0 c— 8 ° |c-1- - .— ^ 3 »: «. .— -a rt O_bC S ll!s H^|« 2 2«t c 0.0 rt- ^ E Sa *- Crs •* o £-"2 c O B »- O TI is s 3| w s. ^ ^ gwK^OOft, .= 5 OOJJ jj c; 'C OH ? CO e, < 191 « o s^; a S - Ooira CD kO -_ OS O -C *" "*" ^ 1*"? T* ^ *? ti c3 o eo" T}< CO C5 g\j ^ 4 1 H -f ,^ ^ •^ ^ Tj< X I ^_- 5)< " £ co co" co T-H 1-4*1— 1 10 i- -1 CO CO -^ CO M5 Co"i-l T • t- co Tf co CO co Tj< T)< C & c •o _w o' o co" ^ ? CO C r! . "£j r* 50" Oj 4J 1— I "^ Tf Tj< . U3 1 i-H i— 1 i— i 1 5 358 g-0 1 ^ 0 g °> *-< rH CO r-T •* 1 oo_ *C •& _a, bfl I* CO r-T co co -^i a j C ^ M S fi 4 t -H OJ W ^ oo'i-i oo co 10" c j us co c- i-Tco a -^ > — *' ~j Q^ W -* •<* •<*' -* o 310 CO C0«0 -.tflO^-f ^* ^* 03 ^"2 1 o •c.Sf'S) -jj ^* CO ^* 05 C"O O • "J^ d) T-H O t- ,-. « 5 1 « S - w g x gj co t: t: ^ o? 2? P S S CM co CO ^ j <; •SCO bo |f I-H oo* •s-_ — * ._ t^ O CO O » ( D co"**^- *" _r^; cd co •«# ^J* CO b™ T ' < ^ rt a3 co SS°i oo* CO_^ CO CO O i S 3 C M 2 ^rf COCO t~ 00 CO CO CO CO IO O OS CO OS CO O O OO i -H ° ^ S CO t- CO t> OO° IO CO U 3OSCO CO COO5 COOOC- - t> IO ! oi £ § 2 CM (M CO (N (M (M C •3 CO CO CO CO CO CO CO C 1 CO CO C "* ' OH ^ 0 ^* S2 co "§ w ^ o •S .2 2 r% 1 0* * 1 * s oo | oo 06" oo os t~ jv S 8 s CO jji SJ ^ W 1 1 i-H I H i— 1 | i— 1 1— 1 1— 1 1— 1 1— 1 — • '~~ •) 1 ~~ ««-. O t4_| o m o 4J ^-^ ft^ ^. i UJ CD a . a O W 1O CO ... U 9 "-1 CVI J] co co £} c- 1 r . *• K ^ f—t \ *^ 1 ^ L "" "* ^ c '"^ c O ^3 fl) 'S3 °° oo O5 O5 2^1- H ^O — *^3 WJ t cC j ^^ 7 00 U5 ^H C 1 rt fi ctf t^,oo_ «£,!>. t^ 05^ t; .S 'B .S O IO~ OO CO lO^CO 1 C ^ o o ^o T^ t^ oo oo o 3^0 10 CO bO'C bo oo t- t- oo co t~ 1 c< sos oooooot- t-ooa 3 00 CO » 'C o 'C o _, o g ^ r« 51 2 H § S CO . ^~*. co •V o J~ o -0 ^^ C 1 1-1 CO I-H JO w **1 10" **1 ^i CO CO lo co" ^ S S rt ^O "o O o ctf O i— * i— < CO O rH CO OO i-H T-I OS S o S W111 co^oT > > || ^E^Q _) Q « « PL| ' o fucumonus >• • phoeus •' - si » nogo/oris M I S" /"**-; / ~ P/)y//ofis darwini a posh'calis limatus ropes fri's FIG. 77.— Collecting localities of the subspecies of Phyllotix oxilae. Shaded areas show approximate ranges of the subspecies of I'hyllotis darwini in southern Peru, Bolivia, and northern Argentina and Chile. See opposite page for explanation. 253 254 FIELDIANA: ZOOLOGY, VOLUME 46 Explanation of Figure 77 (continued) 16. Occomani, Puno. — C. C. Sanborn at 12,700 feet. 17. Puno, Puno (3822 meters). —A. K. and O. P. Pearson at 13,000 feet; M. C. Portugal. 18. Chucuito, Puno.— C. C. Sanborn at 12,800 feet. 19. Pairumani, Puno.-C. B. and M. Koford at 13,000 feet; A. K. and O. P. Pearson at 12,800 and 13,500 feet. 20. Juli, Puno. —A. K. Pearson at 12,700 feet. 21. Pomata, Puno.— A. K. and O. P. Pearson at 12,500 feet. 22. Yunguyo, Puno. — C. C. Sanborn at 13,000 feet. 23. Huacullani, Puno. — C. C. Sanborn at 12,700 feet. BOLIVIA 24. Achacachi, La Paz.— R. M. Gilmore at 3800 meters. 25. La Paz, La Paz (3720 meters). 26. Choro, Cochabamba. — Type locality of lutescens; P. O. Simons at 3200- 3500 meters; F. Steinbach at 3500 meters. 27. Choquecamate, Cochabamba. —P. O. Simons at 4000-4300 meters. 28. Colomi, Chapare, Cochabamba. — F. Steinbach. 29. Tiraque, Cochabamba. —Collected at 15 miles ESE.; O. P. Pearson at 10,500 feet. 30. Punata, Cochabamba. —Collected 10 miles NE.; 0. P. Pearson at 10,500 feet. 31. Vacas, Cochabamba. — J. Steinbach at 3800 meters. 32. Lagunillas, Potosf (3500 meters).— P. O. Simons. 33. El Cabrado, Potosf. —P. O. Simons at 3500 meters. 34. Tarija, Tarija.— Collected 10 miles NW.; 0. P. Pearson at 8200 feet. 35. Pinos, Tarija.— E. Budfn at 1700 meters. ARGENTINA 36. Leon, Jujuy.— E. Budin at 1500 meters; O. P. Pearson at 5800 feet. 37. Yala, Jujuy, mountains west of. — W. H. Osgood at 10,000 feet. Phyllotis osilae tucumanus (38-46) ARGENTINA 38. Cienaga, Tucuman. 39. Tan del Valle, Tucuman.— L. E. Miller and H. S. Boyle at 7000 and 9500 feet. 40. Cumbre de Mala Mala, Sierra de Aconquija, Tucuman. — Type local- ity of tucumanus; E. Dinelli at 3300 meters. 41. Cerro San Javier, Tucuman.— E. Budfn at 2000 meters. 42. Norco, Vipos, Tucuman.— E. Budfn at 1500-2000 meters. 43. Aconquija, Catamarca.— J. Mogensen at 3000 meters. 44. Rfo Vallecito, Andalgala, Catamarca.— J. Crespo at 2900 meters. 45. Otro Cerro, Catamarca. — E. Budfn at 3000 meters. 46. Chumbicha, Catamarca.— E. Budfn at 500-600 meters. Phyllotis osilae phaeus (47-53) PERU 47. Limbani, Puno.— C. C. Sanborn at 11,000 feet; C. B. Koford at 11,500 and 13,000 feet. HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 255 Explanation of Figure 77 (continued) BOLIVIA 48. Sorata, La Paz (2657 meters). - H. E. Anthony and G. Ottley. 49. Tacacoma, La Paz.— H. E. Anthony and G. Ottley. 50. Alaska Mine, La Paz.— G. H. H. Tate at 13,700 feet. 51. Pongo, La Paz.— G. H. H. Tate at 12,000 feet. 52. Aceramarca, La Paz.— G. H. H. Tate at 10,800 feet. 53. Yanacachi, La Paz. Phyllotis osilae nogalaris (54) ARGENTINA 54. 1 1 mu, -i ill. i. Valle Grande, Jujuy.— E. Budfn at 2000 meters. 256 FIELDIANA: ZOOLOGY, VOLUME 46 The Species of the Phyllotis darwini Complex Four species are recognized. They are so closely related inter se that if large series from certain crucial areas had not been avail- able for study, the four might well have been regarded as one. There is a probability, nevertheless, that even more material from as yet unexplored regions may reverse this conclusion and prove that all members of the P. darwini group are still intergrading segments of a single species. A key to the species based on morphology alone would not be feasible, if at all possible. The key devised by Pearson (1958, pp. 407-408) to all species of the genus Phyllotis depends largely and, in critical areas, entirely, on geographic data. This is extreme and prejudices taxonomic judgments. Only the species of the dar- wini complex lack trenchant diagnostic characters. The members of the group, however, are mostly or wholly allopatric to each other. Several occupy parts of a geographic gradient in size and some differ in proportion of parts and the shape of the baculum. These considerations are combined in the following "key" to the species. 1. Phyllotis haggardi. — Andes of Ecuador. Size small; tail short-haired, less than 95 mm. and always shorter than head and body combined. 2. Phyllotis andium. — Andes of Ecuador and Peru as far south as Lima. Average size greater than haggardi; tail short-haired and more than 95 mm. or longer than head and body combined; base of baculum con- vex or plane (osilae type, fig. 7). 3. Phyllotis darwini. — Peru from Junin and Lima departments south into Bolivia, Chile and Argentina. Average size greater than andium; tail short- or long-haired and longer or shorter than head and body com- bined; base of baculum concave or plane (darwini type, fig. 7). 4. Phyllotis osilae. — Andes of southern Peru (Cusco-Puno), Bolivia, northern Argentina; size and proportions as in P. darwini; tail short-haired; ears usually less than 23 per cent of head and body; m2-3 with first primary fold obsolete or absent in moderately worn tooth; second primary fold of m3 nearly always an enamel island; base of baculum convex or plane (osilae type, figs. 7, 86). Phyllotis haggardi Thomas Phyllotis haggardi Thomas, 1898, Ann. Mag. Nat. Hist., (7), 2: 270. Thomas, 1912, op. cit., (8), 10: 409-410— comparisons. Lonnberg, 1913, Ark. Zool., Stockholm, 8, (16): 27— Pichincha (Mt. Pichincha; Quito). Stone, 1914, Proc. Acad. Nat. Sci. Phila., p. 11 — Pichincha (Hacienda Garzon, Mt. Pichincha, 12,000 ft.). Allen, 1916, Bull. Amer. Mus. Nat. Hist., 35: 120 —Pichincha (Mt. Antisana, 12,000 ft.). Pearson, 1958, Univ. California Publ. Zool., 56: 441 — characters; comparisons. HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 257 Phyllotis haggardi haggardi, Pearson, 1958, Univ. California Publ. Zool., 56: 442 — Imbabura (Mt. Mojanda); Pichincha (Guaillabamba; Imbuyo; Je- rusaten; La Providencia; Perucho; Puellaro; Puembo; Tanda; Ventanilla; Quito; Mt. Pichincha); characters; comparisons; distribution. Hesperomys elegans, Tomes (not Waterhouse), 1860, Proc. Zool. Soc. London, I860: 213— Chimborazo (Pallatanga). Phyllotis elegantulus Thomas, 1913, Ann. Mag. Nat. Hist., (8), 11: 139— type locality Pallatanga, Chimborazo Province, Ecuador. Phyllotis haggardi elegantulus, Pearson, 1958, Univ. California Publ. Zool., 56: 443 — status of type. Phyllotis fuscus Anthony, 1924, Amer. Mus. Nat. Hist. Novit., no. 114: 1 — type locality Contrayerbas, Azuay, Ecuador, 11,000 ft. Goodwin, 1953, Bull. Amer. Mus. Nat. Hist., 102: 324 — history of type; measurements. Phyllotis haggardi fuscus, Pearson, 1958, Univ. California Publ. Zool., 56: 442 — Azuay (Contrayerbas); Chimborazo (Mt. Chimborazo; Urbina); Pichin- cha ("Galaya" [ = Saloya]; Guamanf; Mt. Antisana; Mt. Corazon, Cota- pafi (Mt. Illiniza); Napo-Pastaza (Papallacta); characters; comparisons; distribution. Type.- -Skin and skull, sex unknown, British Museum (Natural History) no. 98.5.1.11; collected October 1897, by Ludovic Soder- strb'm. Type locality. — Mt. Pichincha, above Quito, Pichincha Province, Ecuador; altitude, 340(MOOO meters. Distribution (figs. 70-72).— Andes of Ecuador, from just north of the Equator to about 3° 30' south in the Province of Azuay; alti- tudinal range from approximately 2000 to over 4000 meters above sea level. External characters. — Smallest species of the P. darwini complex; tail less than 95 millimeters and shorter than combined head and body length; ears short, apparently less than 23 per cent of com- bined head and body length; upper parts of body buffy to dark brown with head and rump not markedly different from back; under parts gray, well defined from sides, a minute pectoral streak rarely present; tail bicolor, short-haired, pencil little developed. Cranial characters (figs. 78, 79). — Interorbital region narrow, par- allel-sided or, usually, concave mesially, the edges square; proximal ends of nasals usually pointed and terminating behind plane of fronto- premaxillary sutures; posterolateral palatal fossa well excavated, the pits large and usually situated anteriad to plane of posterior border of palate; antero-posterior length of bulla, less tube, sub- equal to alveolar length of molar row. Dental characters. — Generally as in P. darwini; alveolar length of molar row, 3.9-5.0 (49 specimens). 258 FIELDIANA: ZOOLOGY, VOLUME 46 Comparisons. — Phyllotis andium is the only closely related spe- cies found in or near the range of P. haggardi. It is distinguished by longer tail and cranial characters described under the species heading. Comparisons with other species are made elsewhere (p. 263) . Variation. — Pale and dark populations are scattered throughout the range of the species. As a rule, dark populations occur at the altitudinal peaks of the range, i.e., on the high humid grasslands, or paramos, of Mounts Pichincha, Mojanda, Corazon, Antisana, etc., from 3500 meters to more than 4000 meters above sea level. Most of these dark populations are isolated from each other. Pale popu- lations are found at the opposite altitudinal extremes, in the iso- lated arid and semi-arid intermontane valleys, between 200 and 2500 meters above sea level. Dark and pale populations grade in- sensibly into each other, or into the main population mass occu- pying the broad and continuous intermediate altitudinal zone. There is an increase in size and relative length of tail from north to south. Because of the uneven topography and the great diversity of local environments, the gradient is correspondingly un- even, obscured at some points and disrupted at others. Although recognizing the cline in size, bodily proportions and color, Pearson (1958, pp. 441-443) admitted three subspecies of Phyllotis haggardi. The nominate form was restricted to the type locality in the Pichincha-Quito area, thence northward about 30 miles to Mt. Mojanda. P. k. elegantulus was recorded from its type locality only in Pallatanga, Chimborazo, while the range of fuscus engulfed those of the other two. A specimen in Chicago Natural History Museum from Saloya, misspelled "Galaya" on the museum label, was referred by Pearson to fuscus on the assumption it originated in the province of Chimborazo. Actually, Saloya lies west of Mt. Pichincha in Pichincha province. The Saloya speci- men was taken by F. Mena. Two specimens from Chimborazo, in Chicago Natural History Museum, were also collected by A. Mena. Taxonomy. — Phyllotis haggardi was originally characterized as "The Ecuadorean [sic] representative of the Chilian Ph. darwini . . . distinguished by its smaller size and smaller ears." Present material confirms the smaller size of haggardi. Collectors' ear meas- urements are not available for the vast majority of the specimens examined. The ears, however, do appear to average smaller in proportion to head and body length than those of most members of the group. The type, collected by Soderstrom, is an imperfect specimen with incomplete skull and without the usual collector's HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 259 data. Its published external measurements were taken from the dry skin. Specimens at hand include 79 individuals from the type region, also collected by Soderstrb'm. Most of them are juvenals and subadults. The color and texture of many skins indicate undue exposure to drying and fading. No external measurements are given and accurate ones cannot be made on the distorted specimens. There is considerable mismatching between skins and skulls of the same species while some skins are coupled with skulls of mice of the genera Akodon, Thomasomys and Reithrodontomys. It is assumed that data given on the original labels of some of these Sb'derstrom specimens are likewise untrustworthy. Phyllotis elegantulus Thomas, from Pallatanga, Chimborazo Prov- ince, was described as "a small species allied to Ph. amicus, but with shorter tail." The type was skinned out of spirits and the original description is based on the dried, made-up skin. Color of the type, said to be "somewhat modified . . . but apparently as in Ph. amicus," agrees as well with that of Phyllotis haggardi. External measure- ments and cranial characters given by Thomas for distinguishing elegantulus from amicus are precisely those that distinguish haggardi from amicus. Present material from the type region of elegantulus is unquestionably representative of haggardi. Pearson (1958, p. 443), who examined the type and what appeared to be a topotype, con- firmed the identification of elegantulus with the nominate form of haggardi rather than with fuscus. He, nevertheless, retained ele- gantulus as a valid subspecies, chiefly to avoid placing the "clearly defined P. h. fuscus into synonymy under the almost unknown elegantulus." Phyllotis fuscus Anthony, from Contrayerbas, Azuay Province, Ecuador, was distinguished from topotypes of haggardi by darker color and longer rostrum. The Contrayerbas series at hand does average darker than Mt. Pichincha and Quito haggardi, but the difference is local, if not seasonal. A large series of Phyllotis from Mt. Antisana just southeast of Quito also averages darker than topotypes of haggardi and is quite as dark as the Contrayerbas Phyllotis. The longer rostrum attributed to fuscus as compared with topotypes of haggardi merely reflects the increase in size of the species from north to south. The proportions of the parts, however, remain practically the same, as shown in Table 11. Speci- mens listed are those of fully mature adults with skulls in good condition. 260 FIELDIANA: ZOOLOGY, VOLUME 46 TABLE 11. — Measurements of topotypes (Mt. Pichincha) of Phyllotis haggardi Thomas and topotypes (Contrayerbas) of Phyllotis fuscus Anthony Greatest length Nasal Ratio Locality Specimen no. of skull length % Mt. Pichincha1 CNHM 53305 26.5 11.7 44 MACN 31150 26.7 11.3 42 31151 25.7 11.3 44 AMNH 46825 26.6 10.8 40 46826 26.2 10.5 40 46829 26.5 11.3 43 46836 27.0 11.6 43 Contrayerbas2 AMNH 620503 29.5 12.6 43 61957 29.0 12.3 42 61959 28.7 12.4 43 61960 28.3 11.5 40 61961 28.7 11.0 38 1 Means and extremes of nasal length of 11 specimens measured by Pearson (1958, p. 465), 10.93±0.14 (10.4-11.7). - Means and extremes of nasal length of 5 specimens measured by Pearson (1958, p. 465) 12.34±0.12 (12.0-12.6). The difference of 1 millimeter in the mini- mum length of the nasals in the two sets of measurements does not affect the con- clusions. If nasals of no. 61961 were 12.0, the ratio would be 42%. 3 Type of Phyllotis fuscus Anthony; measurements from original description. Measurements. — See Table 12. Specimens examined. — 124, all from Ecuador. Pichincha: Mt. Pichincha and vicinity including Quito, 49 (AMNH, 40; CNHM, 4; MACN, 2; MCZ, 3); Mt. Mojanda, 12 (AMNH); south slope of Mt. Mojanda and vicinity of Guaillabamba, 7 (AMNH); Rio Guaillabamba, 1 (AMNH); Jerusalem, Guaillabamba, 4 (AMNH); Puellaro, 6 (AMNH); Puembo, 1 (AMNH); Jiron, 1 (AMNH); Saloya, 1 (CNHM); Mt. Antisana, 16 (AMNH, 13; MCZ, 3); Mt. Corazon, 12 (AMNH). Cotapaxi (formerly Leon): Mt. Illiniza, 1 (MCZ). Napo-Pastaza: Papallacta, 1 (AMNH). Chimborazo: Urbina, 1 (AMNH); Mt. Chimborazo, 3 (CNHM). Azuay: Contra- yerbas, 8 (AMNH). Phyllotis andium Thomas Phyllotis andium Thomas, 1912, Ann. Mag. Nat. Hist., (8), 10: 409. Osgood, 1914, Field Mus. Nat. Hist., Zool. Ser., 10: 165— Cajamarca (Hacienda Limon near Balsas; Hacienda Llagueda); Libertad (mountains near Otuz- co). Thomas, 1926, Ann. Mag. Nat. Hist., (9), 18: 162— Amazonas (Condechaca; Chachapoyas; Molinopamba). Thomas and St. Leger, 1926, op. cit., (9), 18: 348— Amazonas (Goncha). Pearson, 1958, Univ. California Publ. Zool., 56: 408, 440 — characters; comparisons; distribu- tion [for locality records see explanation of figs. 71, 72]; synonyms, me- lanius, stenops, tamborum, fruticicolus. A «^ 0) I_ - * O o 01 CM t- o Q. "3 j= *" •«t lO _>> i _> *J u- bcja co cg c— I Tj< ?O T f o co «o" ;| *c 0 c 1— 1 E ri C - £ C 2 £ lO O t- O -H OS r-l -H c> •. to co co O ^H O CM r* ~* >-• OS 1— I OS f-H '-I •2 s. E - i'rtrt — -° - •3 1 1 1 •§ S- 8 5 E £ V s _ ~ 261 262 FIELDIANA: ZOOLOGY, VOLUME 46 P[hyllotis] andium, Thomas, 1926, Ann. Mag. Nat. Hist., (9), 17: 615— Ama- zonas (Chachapoyas); stenops a synonym. [Phyllotis lutescens] andium, Osgood, 1944, Field Mus. Nat. Hist., Zool. Ser., 29: 194 — classification. Phyllotis melanius Thomas, 1913, Ann. Mag. Nat. Hist., (8), 11: 407— part, skull only; ECUADOR: (type locality Porvenir, Bolivar, 1800 meters). Pearson, 1958, Univ. California Publ. Zool., 56: 440— type a composite of skull of Phyllotis andium from Canar and skin of Akodon aerosus from Porvenir; type restricted to skull only. Phyllotis andium stenops Osgood, 1914, Field Mus. Nat. Hist., Zool. Ser., 10: 165— PERU: (type locality Rio Utcubamba, 15 miles above Chacha- poyas, Amazonas, altitude, 7500 feet). Phyllotis tamborum Osgood, 1914, Field Mus. Nat. Hist., Zool. Ser., 10: 165— PERU: (type locality Tambo Carrizal, east of Balsas, Amazonas). Phyllotis andium tamborum, Thomas, 1926, Ann. Mag. Nat. Hist., (9), 17: 614— PERU: Cajamarca (Celendin); Amazonas (Tambo Carrizal; Leima- bamba). Phyllotis fruticicolus Anthony, 1922, Amer. Mus. Nat. Hist., Novit., no. 32: 1— ECUADOR: (type locality Guachanama, Loja, Ecuador, 9050 ft.). Goodwin, 1953, Bull. Amer. Mus. Nat. Hist., 102: 323— history of type; measurements. Phyllotis daricini posticalis, Thomas (not Thomas), 1927, Ann. Mag. Nat. Hist., (9), 20: 602 -PERU: Hudnuco (Llata); Pasco (Huariaca); Ancash (Pira). Type. — Adult male, skin and skull, British Museum (Natural History) no. 99.9.9.68; collected 18 April, 1899, by Perry O. Simons. Type locality. — Canar, Canar Province, Ecuador; altitude, 2600 meters. Distribution (figs. 70-72). — From Tungurahua Province in the Andes of central Ecuador south through the Andean departments of Peru between the coastal and Rio Huallaga drainage basins as far as northern Junin and western Lima; altitudinal range from approxi- mately 200 to 4800 meters above sea level. External characters. — Average size larger than P. haggardi, smaller than P. darwini; tail 95 millimeters or longer and usually (always in Ecuador) longer than combined head and body length; ear, from notch, 25 millimeters or less; upper parts nearly uniformly warm brown; sometimes buffy or slightly grizzled but with little or no con- trast between anterior third and posterior parts of body; underparts gray, often with a thin wash of pale buff; a poorly defined pectoral patch or streak rarely present; tail short-haired with poorly developed pencil, sharply bicolor but tip often uniformly brown. Cranial characters. — Interorbital region parallel-sided, or, usually, concave mesially, the edges square, never beaded or forming ledges, HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 1>63 the median frontal sulcus weakly defined or absent; proximal ends of nasals usually truncate and terminating approximately in line with fronto-premaxillary sutures; posterolateral palatal fossa poorly defined, the pit (or pits) large or small and situated on a line with, or, usually, anteriad to, plane of posterior border of palate; posterior border of palate round or square and with or without median cleft or median process; width of mesopterygoid fossa, measured at base of hamular processes subequal to or less than width of parapterygoid fossa measured at same plane, sides parallel or slightly concave medially, sometimes slightly divergent; bullae moderately inflated, antero-posterior length, less tubes, 3.8-4.8 and always less than alveolar length of molar row; auditory bullae comparatively little inflated, Eustachian tubes short to moderately elongate. Dental characters. — Generally as in the P. darwini group but m1 more often S-shaped than 8-shaped, pf 1 of m^ more often ab- sent or obsolete in moderately worn tooth, pf 2 of nv1 nearly always an enamel island; molar rows parallel or slightly convergent pos- teriorly; alveolar length 4.1-5.3 (190 specimens). Baculum. — See discussion on p. 58 and figures 7 and 8. Comparisons. — Two species and three subspecies of the Phyllotis darwini complex occur in or near the range of P. andium. The Ecuadorian P. haggardi is slightly smaller with tail actually and proportionately shorter. P. darwini posticalis and P. d. limatus average larger throughout with bullae usually more inflated. P. d. definitus is absolutely larger with molars larger, body more warmly colored throughout, the under parts washed with ochraceous, pectoral patch well defined. Separation of Phyllotis andium from P. haggardi and P. darwini posticalis is not always clear. Evidently, speciation was recently accomplished and only in critical geographic areas is it sufficiently defined for recognition. Specific distinction between P. andium and P. darwini posticalis is predicated principally if not solely, on the basis of a small series of each species collected by Pearson near Casapalca, Lima, Peru. Here the representatives of the two species differ in virtually all dimensions, notably in inflation of bullae and proportional width of interorbital region (Table 13). Some of these differences may be age factors, the andium series being younger. The posticalis series, however, is also darker, more gray, with a coarser ticking, has more prominent guard hairs, hairier tail and its median palatal spine is poorly defined or absent. % t: auids IB^BIBJ 02 = = = u -O O. rt q^3uai ||ti3js t^ t^» oo oo t^~ oo oo oo :MOJ JBjoaAiv 2 o> (X t^ oo o ci ^o t>« ^ »~^ rt A\OJ JBJO3AIV ^' Tt W3 Tf 10 10 10* IO* J qi3U't| oooo -H^HOOO 3 Itn>js :SIBSBM 1 oo «o co Oi TJ< o A SIBSBN 2 ^ S 2 2 S ^ S o £ o i aBjjnq XJQ^. 0303^0-^ oqi-noiOi •2 -ipnB jo q^Sua^j ^ "*' "* ^ 1 "w q^Suaj {(nj^s ^«o«n^o rj<^ o S "o £ (Aip) ;ooj puiH SSScg e^So3e5 1 s $ eg-^oso t-co-^to c ^IB^ i— i i— it— i T-> Mcgo—" t, o> 03 £ •« Jjj OOiOOt^ »H tO tO »H S Apoq PUB psajj oosoicft egcgr-io ^ | ^H ^H ^H ^H ^H QQ *o CO 1—t fe 3 •5J "i n ^^ ^ j 5 *5 t"* 03 03 o: -^^ 03 .^ ^ ^4 ^.J" o >— i eg co 10 «o t- oo o-§ *^ ^H^-,^H^I oooooooo -3 a O oooo oooo =§S j oooo oooo "S't; c 0) ^gcgcgegeg-gcgcgegcg % % .E, (^ S *^ 1 1 ss^ Is s * o 264 HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 265 Specimens of darwini and andium from other localities in the area of sympatry are imperfect for comparative purposes. One specimen of Phyllotis darwini limatus (MVZ 120033) from Surco, Lima, is a hydrocephalous Juvenal with an abnormally inflated braincase (fig. 80,a). Its skull cannot be fairly compared with those of three adult andium from the same locality. However, external differences be- tween the Surco darwini and sympatric andium are the same as ob- tain between the two Casapalca species except that the tail of dar- wini is not appreciably longer-haired. One specimen of P. darwini limatus from 1 mile east of San Bartolom£ is also pale, with a more coarsely ticked coat and slightly hairier tail as compared with a series of 10 specimens of andium from the same locality. The andium series also shows well the shorter and broader muzzle, less inflated bullae and other distinguishing cranial characters of the species. Some allopatric series of andium, notably one from Chongoyape, Lambayeque, differ even more widely from the sympatric series of posticalis while others overlap in all characters except relative degree of inflation of the bullae. The slight but real difference in the auditory bullae is fairly constant throughout the geographic range of both species. Unfortunately bullae cannot be accurately compared in terms of the linear measurements used here. Evidence of sympatry between P. andium and P. haggardi is lack- ing. Both species become larger, their tails longer, from north to south. The size cline in andium begins along the same coordinate south of the point where the known range of haggardi ends. On the other hand, P. andium remains nearly the same size with the same long tail where its range parallels that of P. haggardi. This phenomenon is the dubious basis for specific separation of the two forms. Cranial characters used by Pearson (1958, p. 437) for distinguishing haggardi from andium include "longer palate, posteropalatal pits usually sunk in depressions, tooth rows tending to converge posteriorly, zygomatic arches more widely flaring." These characters though valid in present material are of the kind which in larger samplings usually prove to be local or individual variables. The less nearly related Phyllotis amicus is smaller, with longer tail, larger ears, skull with broad divergent-sided supraorbital region, incisors more delicate and m^ with procingulum more developed. Variation. — The various series of andium from the Department of Lima are fairly uniform in cranial characters, notably in the presence of a well-developed median spine on the posterior border of the palate. This feature is insignificant or absent in the vast 266 FIELDIANA: ZOOLOGY, VOLUME 46 majority of andium from elsewhere. Specimens from localities in Lima coincident with or near the range of dark-colored Phyllotis darwini posticalis are dark, others from paler Phyllotis darwini limatus territory (Surco, Santa Eulalia) are correspondingly paler. A series of 36 specimens from Macate, northern Ancash, taken from 1 February through 6 March, 1914, by M. P. Anderson, is more brown, on an average, than andium from the Department of Lima. A series from Yungay, Ancash, collected March 4-6, 1954, southeast of Macate, is more somberly colored because all 11 specimens are in old, worn pelage. Nevertheless, patches of new pelage on the dorsum of one specimen (CNHM 81246) are colored as in the Macate series. Thirty-three specimens from Huaras, Quilcayhuanca and Hacienda Catac, in the Department of Ancash, are paler than those from Yungay. They were taken in February 1954. All are in old pelage, but some show patches of new pelage. None compare with the Macate specimens which are in prime pelage although taken in the same general region during the same month 40 years earlier. The series from Huaras includes individuals nearly as pale as coastal Phyllotis darwini limatus. Eight small, pale, long-tailed mice from Chongoyape, Lamba- yeque, a locality 240 meters above sea level on the coast of northern Peru, closely resemble large individuals of the nearly related Phyllotis amicus. This population may represent a desert race characterized by small size, extremely pale color and long tail. In most of the Chongoyape specimens the molar crowns are considerably worn, the dentine deeply excavated. As noted elsewhere, unusually small size appears to be correlated with excessive molar wear. Evidently, in the Chongoyape ambience the mice were obliged to feed on tough- fibered plants with low nutritional value. Taxonomy. — The type of Phyllotis andium stenops Osgood, from Rio Utcubamba, Chachapoyas, is the only fully mature individual of a series of ten from the Chachapoyas area in southern Amazonas. The series was distinguished from Cajamarca representatives of andium "principally by the narrowness of [the] nasals." It has already been shown by Thomas and Pearson that subspecific dis- tinction of stenops by this or any other character is untenable. The type of Phyllotis tamborum Osgood, from the mountains east of Balsas, southeast of Chachapoyas, is a juvenal absolutely indistinguishable from Chachapoyas stenops. Phyllotis fruticicolus Anthony, from Guachanama, southern Ecua- dor, is, as shown by Pearson (1958, p. 441), indistinguishable from X CO X 5? oo t- oo s oo c~ 1 - W oo c7 OJ eg^ eg_ eg .1, 10 TO us eg^ OO OS rH 1C rH ,_, OO S" P P oo" oo" P~ oo~ eg eg eg eg eg eg M i 1 l ' i 1 » Tf kOrlj eg e>i t~ c- to to to" o M O O OS rH rH OS TJ« 00 TO 00 CO TO rH eg eg eg eg eg eg Cg rH rH rH rH rH rH 1-1 III I I 00 it tO O t~ OS co°° oornoooeg OO"~lrH OOrHt-tOOSt— COT*""! rHegrHOOCgTOrHrHrHOrHrHOO rH OS rH OS rH rH rH rH rH rH _ rH OS ECUADOR Punfn 88, Canar1 Hi Yunguilla Valley 90, Guachanama2 91 Guachanamd 84 OO ^ 1 o eg l ?i T! — 1 rH eg rH rH 1 1 t— eg 1 00 •=. to to ~ o - 1 S t^ eg rH O 2 S N 0 t- el i - oT O Os — ' jg w t TO if /• - 00 "3 frt "*1* S rH OS * rt ~ Tn 01 Chongoyape Cabache § ^ 9 Chachapoya Chachapoya Carrizal6 - \ ~ c a CO Cajamarca Limon near Otuzco Llagueda Macated" cf - - 1 od - - ^ Yungay Huaras Quilcayhuar Catac Cullcui 267 1.2 £ os co co eg d o •o c as aS o> (-< 73 C 73 aS 0) c 73 IV C T3 OS 0) o o JS "be O) ^: M (V E-1 .c "Si O 1— 1 II c ~a of eg" oo CD oc ^_ !. ^3 __ o 11 TJ- T}- CO T).' CO Tf t- G ) "O 3 "3 "as "c3 oc* -^ c > ^f CD -5j< CO 4 C OS cr o> 7 cr 0) cr c S3 TT T}« if 5 -n- -5t rf. ^j. d CC a > , H-, «« as (continued) CO 0) CO CO d os° 06" oc" t- os co eg t~ TJ" OO -^f o eg i o > ^ c JS 1 — l "^ ' r'i i O eg £ "be o 73 c JS bo o eg -a c bo o c bo 1 '3 02 a £ o ~ c os oc > •— ' i 00 OC OO OC i eg 0s] eg eg t- os T3 O .0 73 .2 "2 _c 5 J2 T3 C C T3 C o aj -s £ as ~ 73 £ i eg "l "l -*-* QJ 1 rt? M Co OS as X ca K L !"H o •; _aj O) f S JS 0) (D X as *3s eg eg eg eg eg eg eg _> "2 c -c o> c do c 1 OJ JC .•> — o _c IS IS IS IS o> *c K •i2 ~' CJ ^i eg X if5 M *M eg •*?* t"~ ^* os - ~ £ — *** ^— ' — CC Q • i , f 2 H eg — * o -H o eg 3 O) 1 5 "S . c c C a cc t~T CD" o -* o ~ os "5 o CD" ~ ^ eg eg •M 0) "Cj y C Q) c "1 ) ^s -^ _3 *^ ~" tT bo C bo c tT bo C fe bo O) be c *c o> bo H 73 t^ t^ os £ || u -C u •s" c 2 -2 ^ 2 3C 2 c^ c -0 •o 00 eg ^ "1 0 CD "~ t^^ ^5 ^5* 1^5 1^11 El?| X O "o S u Ij 5 =1 DC 'S 1 0 « l§ o> a *-T "1 eg —* OC -H T O TJ< ~"1 o ^ o os i -I § §, •s -^ ~ § S CJ •ij s -g =c C "r-r £ — DO «o go a: is oc 01 o go oc Jo •" oc DC t !<•= j 5 ^. « -2 "S •• '? *" QJ s—^. § oc gQ o "^ <— • T ^ T __, "T1 32 -^ "as |{j 1 1 5 1 a a r= Q. >> >» rt >> f_ EH H H a := >j as E-1 E-1 ^ ^ c^ cE cE cS- -HCD C 8? CJ II s S E 0 = C 2 u o «= r- . c. II 73 2 ^ -a '_' II T3 "C o o O O JO .0 ^2 268 HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 269 andium. The description was based on a comparison of the type, an immature male, and two immature topotypes with Phyllotis haggardi. According to Pearson (1958, p. 440) the name Phyllotis melanius Thomas, 1913, is based on a skin of an Akodon aerosus and a skull of a Phyllotis andium. Skin and skull were collected by P. O. Simons, the former in Porvenir, Bolivar, Ecuador. Both bear Simons' field number 258 and the British Museum (Natural History) number 99.9.9.107. Pearson restricted the name Phyllotis melanius to the skull alone. He further demonstrated that a skin only of Phyllotis andium, also collected by P. 0. Simons and with the same field number, was almost certainly the correct skin of the type skull. Inasmuch as Phyllotis does not occur in Porvenir, Ecuador, and the skin with the same field number as the type skull originated in Cafiar, Ecuador, the type locality of melanius was redetermined as Canar by Pearson. Phyllotis melanius thus becomes a topotype and absolute synonym of P. andium. Measurements. — See Table 14. Specimens examined. — 212. ECUADOR. — Chimborazo: Punin, Quebrada Chalan, 4 (AMNH). Azuay: Pauijchi, 1 (AMNH); Yun- guilla Valley, 7 (CNHM). Loja: Guachanama, 2 including type of fruticicolus (AMNH). PERU.— Piura: Ayavaca, 1 (CNHM); Huan- cabamba, 5 (AMNH, 3; CNHM, 2). Lambayeque: Chongoyape, 8 (CNHM); Cabache, 1 (CNHM). Amazonas: Chachapoyas, 13 including type of stenops (CNHM, 10; AMNH, 3); San Pedro, south of Chachapoyas, 6 (CNHM); Tambo Carrizal, the type of tamborum (CNHM). Cajamarca, 10 (AMNH, 1; CNHM, 9); Haci- enda Limon, 2 (CNHM). La Libertad: Hacienda Llagueda, 2 (CNHM); mountains near Otuzco, 2 (CNHM). Ancash: Macate, 41 (AMNH, 2; CNHM, 36; USNM, 3); Yungay, 11 (CNHM); Huaras, 14 (CNHM); Quebrada Quilcayhuanca, Huaras, 11 (CNHM); Hacienda Catac, Ticapampa, 8 (CNHM). Huanuco: Cullcui, 3 (CNHM); Ambo, 10 (AMNH, 1; CNHM, 8; USNM, 1). Lima: Matucana, 7 (CNHM, 3; MVZ, 4); Surco, 3 (CNHM, 1; MVZ, 2); Casapalca, 4 (MVZ); Canta, 1 mile west, 2 (MVZ); Huaros, 8 (MVZ); San Mateo, 6 (MVZ); Villavista, 8 (MVZ); Zarate, 10 (MVZ); San Bartolome" Station, 1 (MVZ). Phyllotis darwini Waterhouse. (Synonymy under the species and subspecies.) Distribution (figs. 57, 70). — From the departments of Junin, cen- tral Peru, south along the Pacific coast and highlands through Peru, 270 FIELD IANA: ZOOLOGY, VOLUME 46 Chile, Bolivia and Argentina to the Straits of Magellan, and east to the Atlantic coast in the Argentine departments of Santa Cruz and Comodoro Rivadavia; altitudinal range from sea level to 5000 meters above or to snow line. Characters. — External, cranial and dental characters are those of the P. darwini group; base of baculum convex or nearly plane as described elsewhere (p. 58, figs. 5-8). Comparisons. — Phyllotis darwini figures in comparisons made be- tween members of the darwini complex, with other species of the genus Phyllotis (p. 236) and with phyllotines in general. The geo- graphic relationship between P. darwini and other members of the complex is discussed under the heading Sympatry and Allopatry (p. 31). Differences between P. darwini and sympatric representa- tives of P. osilae are shown in Tables 30-49, and pp. 346-357. The differences are idealized in figure 86. Comparisons between P. dar- wini and sympatric series of P. andium are made in the account of the latter. Variation. — GEOGRAPHIC: The ten subspecies of P. darwini rec- ognized here are, from north to south, definitus, posticalis, limatus, rupestris, magister, wolffsohni, caprinus, darwini, fulvescens and xan- thopygus. The palest race, Phyllotis darwini rupestris and P. d. limatus live in the driest parts of the range of the species; the most saturate race, fulvescens, inhabits areas of high rainfall. The finely graduated increase in humidity along the Pacific Coast from southern Peru to southern Chile is correlated with a correspondingly fine transition from limatus and rupestris, through more intensely col- ored darwini, to saturate fulvescens. The boundaries of the ranges of most subspecies conform closely to the hyetal lines which de- marcate the principal zones of rainfall. There is no such correla- tion between altitude per se and color or any other subspecific character. A size gradient between subspecies is not apparent. P. d. rupestris averages smallest but intergrades directly or indirectly with all other races. P. d. rupestris becomes larger from western Peru and Chile eastward into the high Andes but without significant differentiation in proportions or other characters. This even gradient is no basis for divisions into the two or more subspecies (chilensis, vaccarum, ricardulus*) recognized by other authors. On the other hand where the change in size is abrupt as in the case of rupestris and magister no cline is evident but two sharply defined taxons are. HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 271 Tail length averages slightly less than combined head and body length in present samples of rupestris, darwini and xanthopygus, slightly more than head and body length in posticalis, wolffsohni and magister, considerably more in limatus (118:100), and most in caprinus (122:100). It is significant, however, that practically the full specific range of variation of tail length relative to head and body length obtains in such races as P. d. posticalis (72 137:100) and rupestris (81-132:100). More specimens of these than of other races were examined. The principal obliquity in cranial characters is the strongly divergent-sided supraorbital region in most specimens of wolffsohni. The same race is also noteworthy for the consistent S-shaped pat- tern of its second molars and, to a slightly lesser extent, of its third upper molars. Phyllotis darwini caprinus bridges the gap between wolffsohni and other races of darwini (figs. 82-84). LOCAL: Size differences between two neighboring micropopu- lations, or pocket populations, of a subspecies may be greater than the average size difference between two neighboring subspecies. The tail may average considerably longer than combined head and body length in some local concentrations of individuals and considerably shorter in others. For example, in the 5 specimens of P. darwini posticalis from San Jenaro, Santa Inez, Huancavelica, the ratio of the tail to head and body is 83, 85, 86, 86, 86, respectively; in the 4 specimens of posticalis from Urco, Cusco, the tail ratio is 117, 121, 126, 136, respectively. A group of individuals inhabiting a com- paratively humid pocket of an otherwise generally dry region may be slightly darker than average in coloration. Pelage of populations inhabiting warm coastal areas may be shorter, thinner, and coarser in texture than that of populations of the same race living near the border of perpetual snow. Local differences in the chemical composition of solubles in drinking water may cause striking local differences in bulk and density of cheek teeth. These and similar gross deviations from the overall racial pattern may be impressive when subjected to statistical analysis. For the most part, however, they are reflections of age, crop and random genetic variations. INDIVIDUAL AND AGE: The classical concept of the existence of a direct relationship between the age of an individual and its size, pelage, and dental conditions can be applied with some degree of confidence only to individuals of the same crop. It cannot be applied generally to phyllotine rodents or, for that matter, to any mammal without qualifications which are frequently nullifying. Pel- X T) C a cd I Q 272 _ - (V 273 8 BE _ "H | «-i 5 ° - »— i '. 3 0) - Q c. C ^ DO *o w ^v 3 • CO 3 § "o >-5 S 73 "1 o<- o" ^- o" c C5 co Of O '-s O O . • IM o" c^T o" o cT >J ^ u . C?>-s'l-s'^r>-S>-3>-90 Ol-50^< O O O l-f O « O 1 d ,0 E c" "-C BJ 3 o C K 2 ^ oj "S. ^; OJ X § PH 0) 4-5 4->' N . _J yf **-* t*-^ 5 . rH 0 0 0 0 s 3 3 « r ; C 0 0) bo Q. C a ~a ~5 ^03 § § Ills o££ o |M|» ^ ^ 0 bO C ^-^T 0 C g.'3 3 3 .C * *• ^H O) ^1 l> > C TO 3 CO ^ — , i—* CT O b 11 !*-! o 1 £-d1^ f &. flfllll •ti rt«rt2°2 .3 rtSS-0'^ cr. co ajvr ...^V1^ | i«-3-!|.2<*>--B'-*-,2 w x 3 » 2 ? 2 » -5 2 P° OT R. c >^ "D 15 *- •b ^. M x ^^ s "*> TjT cc 0 K rt 3 w "° T3 *e c a « •c . S -E 3 0 C «- rt >5 C ^* o ° S b-o C8 C o O O O >-s S « E « o "I t-"eo TjT . . °i '-L o cT o t-T cT m >-» o o cT cT o" cT cd ~ _, e eertj::5 i ^ C H H c* O cS CC _C M .£.5=33 m^ •— C8 OO iJ'^CiT1 2.C8 C> o^oC l~'e«hL|o»22JS §OH TO c±ic303rt 2 ^^- >> 0- c >>2 S •s . j^agSSsfiflSsiesS «^ rt MM ^ ^^ A " t- fc -- F ^ iaPllIlfla-illsllllli £ 62.1 g.l |=3Sg5^.gm.^a£.S ^5& O3J S^-sE-'OlZODico^iJCiiiJ E-i^E-1 3 a < H 9 « £ a s => | t 33 OJ •-5 "-5 < OJ 279 'C Ml 00 *fcj 2j:^^::'^ ^^ w, cc g. R, "^ a 3 s S §, s e o ^ a« s a CQ V i 1 2 1 •3 5 y 11 a, • o ^1 •° *? rH M — < ^5 cT ^* 3 S w "^ /J S t. "*•* Pjj oT TjT bo -G TS S -> 0 "b s « _ ^ ^- '"I ^ o" °°I "^ oT T— 1 i jj * o oT l-H "^ "^. '""I ^- cT *"?, » 3 ^2 ^ ~? ; Locality1 Punata, Cochabamba' Comarapa, Santa Cru Caritaya, Tarapaca5 Tilcara, Jujuy6 La Paz, La Paz8 Oruro, Oruro8 Uyuni, Potosi8 Villazon, Potosi8 Limbani, Puno & £ >= °3 « 2 % » § ^ o3 t; £ 03 a; :C55, S-S S :c'| OSOToo*" -«<3— oo & £ O EH 0-3| ^^ §.5,^^ g •.OOfSaj^o-p -r-)3 3^^^ ^o^ i 3> i'^go'S -^lll-a -^^OV_TZ^^ o) o3 ^r-i^ ^ S TON&ogoPH-t°.2,rt"!_o S 1 3 8 -1 « 2 :c S S g o3rSoOrtSc3o3o3o3.S O > PH EH EH M EH H O U J Huaylarco, Arequipa Tincopalca, Puno Puno, Puno Asillo, Puno ^ H 3 a I, c S O X5 O 280 ft. « > "y - manuscript data on skin tags of specimens collected Koford, and E. Heller (concluded). — - £ 1 w S •b 00 » * " o °i2 o i~i 00 °°I 'VJ **? "•! '^-i t- rt (M i-H -HO* oo E "3 £ u 0) co co d •-• M „ •22 _c . 1 >. - C 0 o 0 0 cT^ O~O CO _c o 03 tc 0> 0 05 rt E «' C 0) fi d be —Reproduction in Phylloii UJ > XI Locality1 Torata, Moquegua Arequipa, Arequipa Arequipa, 12 km. SSW, A Salinas, Arequipa Imata, Arequipa Torata, Moquegua Lago Suche, Moquegua Tala, Moquegua Caccachara, Puno Puno, Puno Rfo Santa Rosa, Puno Peru unless otherwise indi 0-7) represents the numbe Jng. present. 4-14) represents length of es. s. • and Tarata rupestris are f f/1 •^ _e a j O) • £fe.l "3. a* -S S S> cr . a. 03 C r — 3 ^C ^J .C 3 | Is V. S «= *- ^ S c 5o -2 '& ^~ > E .S rt 'Z 03 03 e i- •-• JS 0) _C_ •S'? J ? ^: 3 ."= ? > 0 — .- - L* ™ £ 03 > be ° — <-• ^ ' | 03 || || ll ^ II II II 03 -C Er~ o C c o. 1 <; w4 J f^ W "-5 tC -U •< H CO 281 282 FIELDIANA: ZOOLOGY, VOLUME 46 The data may be summarized as follows: 1. Highest proportions of gravid and suckling females occur from November to February and in August. 2. Most males are in breeding condition from November to May. 3. Lowest proportions of gravid and suckling females occur dur- ing September and October. 4. Most non-breeding males occur from July to October. 5. Means and extremes of the number of embryos are 4.6 (2-7) 12 females. 6. Sexual maturity is attained at a stage of external, cranial and dental development which would be classified as subadult or juvenal. 7. The annual breeding cycles and reproduction in Phyllotis darwini correspond roughly to those of Phyllotis osilae for which comparable data are available (p. 351, Table 29) and agree with the generalized observations on altiplano cricetines made by Dorst (1958, p. 563). 8. The existence of a local and/or regional breeding cycle or cycles may be suggested but is not established by available data. Present information is derived from kill-trapped individuals taken during one or two successive trapping nights per locality. Such samples do not necessarily reflect the true breeding condition of the population at any moment or month or even a fair approximation of the ratio of individuals belonging to each sex and age class. Phyllotis darwini posticalis Thomas Phyllotis darwini posticalis Thomas, 1912, Ann. Mag. Nat. Hist., (8), 10: 406. Thomas, 1920, Proc. U. S. Nat. Mus., 58: 230 — part, Cusco (Huaracondo [part]; Chospioc; Ollantaytambo [part]; Puquiura[?]). Pearson, 1958, Univ. California Publ. Zool., 56: 414— characters; distribution (see p. 242 for locality records); abrocodon Thomas a synonym. P[hyllotis] d[armni] posticalis, Osgood, 1915, Field Mus. Nat. Hist,, Zool. Ser., 10: 190, footnote — Junin (Oroya). [Phyllotis darmni] posticalis, Osgood, 1944, Field Mus. Nat. Hist., Zool. Ser., 29: 194 — possibly recognizable as a subspecies; abrocodon Thomas a synonym. Phyllotis abrocodon Thomas, 1926, Ann. Mag. Nat. Hist., (9), 17: 316— PERU: Junin (type locality, La Oroya, Lake Juntn, 3750 meters). Type. — Adult female, skin and skull, British Museum (Natural History) no. 0.7.7.38; collected by Perry O. Simons, 26 February, 1900. HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 283 Type locality. — Galera, a railway station on the crest of the Cordillera Occidental, Yauli Province, department of Junin, near boundary of Lima Department, Peru; altitude, 4800 meters above sea level. Distribution (figs. 70, 72, 73, 76). — Andes of Peru from northern Junin and bordering parts of Lima southward into Huancavelica, Ayacucho, Apurimac, Cusco and the upper Rio Inambari basin in Puno; altitudinal range approximately 1900 to 5000 meters above sea level. Characters. — Average size slightly less than definitus and mag- ister, larger than limatus and rupestris; paler than definitus and sympatric populations of Phyllotis osilae but darker, more gray, than average magister, rupestris and limatus; dorsal surface pale brown finely peppered with black, giving a dark grizzled effect; side of head, rostrum, and shoulders usually more gray than trunk; well-defined gray underparts with a light, sometimes a moderate, wash of buff or ochraceous; pectoral streak or patch present or absent; tail usually longer than combined head and body length, thinly haired or well-haired, bicolor except sometimes at tip, rarely nearly entirely dark brown; ratio of tail to combined head and body length =70-137: 100; auditory bullae moderately inflated; aver- age antero-posterior length, less tubes, about 4.8 mm. (4.4 — 5.4, 95 specimens) and always less than alveolar length of molar row; mesopterygoid fossa comparatively broad, the anterior border usu- ally square or slightly rounded, rarely pointed as in definitus; postero- median process of palate little developed or absent; posterolateral palatal pits usually anteriad to posterior border of palate but may be on a line with or posteriad; supraorbital edges sharp or rounded, the edges often raised to form a well-defined though shallow frontal sulcus, proximal ends of nasals pointed to truncate and terminating from in front of to behind fronto-premaxillary sutures; molars mod- erate to large, combined alveolar length, in adults, 4.9-6.1 mm. (99 specimens) or 17-19 per cent of greatest skull length. Variation. — Smallest members of the race occur at the southern extreme of the range, in Cusco, where contact with equally small Phyllotis osilae occurs, and at the northern extreme, in Junin and Lima, where similarly sized P. andium lives. Populations of large posticalis including some quite like Phyllotis darwini magister, occur in the central and southwestern parts of the range in the depart- ments of Huancavelica, Ayacucho and Apurimac. 284 FIELDIANA: ZOOLOGY, VOLUME 46 Size of molars as measured by their alveolar lengths varies in adults from 17 to 19 per cent of greatest skull length. The narrow range in proportional length indicates that longer molar, or alveolar, rows, are associated with larger skulls, smaller rows with smaller skulls and that the length of the row increases with cranial growth until old age when molar attrition is greater. In a series of 5 speci- mens from Hacienda Palmira, Apurimac, the largest skull, with length 33.0 mm., has the largest molars, with an alveolar row length of 6.0 mm. The second smallest fully adult skull, with length 30.6, has the smallest molars, with an alveolar row length of 5.5. The smallest adult skull (30.1 mm.) has an alveolar row of 5.6. The ratio of alveolar row to skull length is approximately 18 per cent in each of the five specimens of the series. The largest set of molars in the Palmira series is the least worn, the smallest the most worn. Palest populations of posticalis are in Cusco (Cusco; Sacsa- huaman). They grade into pale rupestris of Arequipa and Puno. Others from Cusco (Ccolini) and Puno (Limbani) are virtually in- distinguishable from extremely dark Bolivian representatives of Phyllotis osilae. More typically colored posticalis from Ayacucho, Apurimac and Huancavelica are like the darker series of magister from Chihuata, Arequipa. Northward, in Junin, posticalis is more brown, and distinction from neighboring Phyllotis andium is based on characters other than color. Nowhere, however, does posticalis become as warmly brown as P. darwini definitus. In general, pale populations occur in drier areas, dark populations in more humid regions. The tail is generally bicolor except sometimes for a brownish tip. A single specimen from Ccolini, Marcapata, Cusco, collected 27 August, 1953, is unusually dark with terminal two thirds of tail uniformly brown. Hairiness of tail varies locally to a marked degree. In the de- partments of Lima, Junin and Puno, tails are generally hairy with well-developed pencils. In Huancavelica, Apurimac and Ayacucho, the tails of some populations are quite hairy while those of others are thinly haired with virtually no pencil. In still others, tails range from thin to thickly haired. A buff y or brownish pectoral streak or patch is frequently present and often correlated with dark underparts and a thinly haired tail as in Phyllotis osilae. Cranial characters common to P. osilae may also be associated with the osilae-\ike external characters. Speci- mens of posticalis from several localities in Huancavelica (Piso; HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 285 Lircay; Palmira) are like P. osilae and except for their darwini type penis bones could be so classified. In some Cusco localities (Huaracondo; Ollantaytambo; Urco) and Puno (Limbani) where posticalis and P. osilae occur, the former is more typically darwini in external and cranial characters. The front of the upper incisors is white with an orange mottling in two of five specimens from Palmira, Apurimac. This character served as the basis for the description of Phyllotis abrocodon Thomas from La Oroya, Junin. The incisors are normally pigmented orange in two specimens at hand which formed part of the original abro- codon series of twelve examined by Thomas. Pink-stained skulls in posticalis have been noted by Pearson (1958, p. 415). He suggested that the stain may have been derived from ingested cochineal. I tested the theory by feeding cochineal to laboratory mice (Mus musculus). Their bones stained the same as in the Phyllotis. The pink skulls of P. darwini posticalis in Chicago Natural History Museum and the University of California Museum of Vertebrate Zoology (MVZ) are listed as follows: Catalog Date of Number Locality Collection 20917 La Oroya, Junm 28 June, 1925 75421 Puente Pajonal, Ayacucho 22 October, 1953 75422 Puente Pajonal, Ayacucho 22 October, 1953 75423 Puente Pajonal, Ayacucho 23 October, 1953 75424 Puente Pajonal, Ayacucho 23 October, 1953 120004 (MVZ) Ayacucho, Ayacucho 30 August, 1955 75425 Piso, Huancavelica 8 December, 1953 75426 Piso, Huancavelica 8 December, 1953 75436 Mayoc, Huancavelica 5 December, 1953 75439 Mayoc, Huancavelica 4 December, 1953 75440 Mayoc, Huancavelica 5 December, 1953 120006 (MVZ) Rfo Mantaro, Huancavelica 29 August, 1955 120007 (MVZ) Rfo Mantaro, Huancavelica 29 August, 1955 There are two more skulls from La Oroya, and two from Aya- cucho, all unstained. Two skulls of Calomys sorellus from Puente Pajonal, collected at the same time as the Phyllotis, are similarly stained pink. Taxonomy. — Phyllotis darwini posticalis combines in one or an- other of its populations all external, cranial and dental characters of the forms originally described as Phyllotis andium, P. osilae, P. magister and P. definitus. Absolute distinction of posticalis and P. osilae depends solely on slight and dubiously constant differences in the structure of the penis (fig. 7). The same may be true of P. a oo —~ _C5 ^* CO ? i « « ^— • W3 ^T • t^J i IO IO i IO > bO ca TlJ oo" t~ 10 co" r-T w i— T — C*o U^ iri id co id id co* »d id id 10 id id ^ id — S t^ eg co •0 10 to' 0 0» 10* C^ rn" rf os" eg t-" oo' ^" rf c^ eg" t- t- t~ os cd id t> 10 10 co "ia'S *** CO t>- • • S ,-( co 10 £ • TH t— • OS s,« t~- 1 1 ft io" T1 *~i '"i V '-i b? *" *° co |d co co « id id co to «o co Tt "* 10 NN X •M _*h 4Mh co co co co" t^ •*" «o oT eg' co" eg" os" co to" ^ oo" TH" co eg" (N co to 10 id id to *d id co id id co to 10 to 10 10 10 IO j •^ a . 1 . ( . 1 . 1 1 "^ j Tf ' ^m ^. ~ "w ^L ^* i/^ 10 eg ' — ^ ' ^ co co CO , — ^ ' — ^ co 2.2 °* H ca j^ eg °. ^ 10" TT co" o eg oq ®. co" 2 few W co co £J rj | | CO CO co w co eg °]3 co | . c ^ eg t^" co o ^™ oo i— i co" oo to eg" oo" w id" - ~ — ~- i— 1 O5 CO CO* °* i— 1 TH eg o o eo oo t-' o 2 3 SS co Ojj^ 03 co 23- ^ e*' co co co co eg £J, ^ § W °». "? «? t-id"coco"o»d'ioco"to' OS" l-H oo" 10 co" t>^ oo T-T os ^ * O TH TH •» co co co rocococotococococo co co co co co eg eg co eg CO co i 10^ U5 i °^ c5 c^ e^ eg eg S L 0 A l*i] "^ eg eg cf 1 CD IO 1 1 •"# t— t— ^^ oo" to" co" — . 2J. fj" 513" £J. |* U 10 eg eg eg co eg eg eg eg co eg eg to 1 * * to i •* -^^^ ^*> .» * • ^^ * * " " ,T* " * ifc ; H N 1 >] cgcg> ca ca CJ g ^2 £ ft"1 1 s £ 2 c8 " w S 3 *« ca o ^ >, >, lillLbl caSca^^.as.tiS ca u c c •2, ^2 liliiii t = _£ 000 ^'-sOSSPnWHjM — — <3 pL| PH H^ O2 O S 3 286 Sol > » J3 co os 10 oo 10 •— i id id id id id 10 o $ ML. >»^J N: — "J5J ,-* O Ol ^ -S *°. °°. r* -*t co os co co eo eg Phy i — E co ri o t^ o o eo e S-. ;! 5 O r — i 1 r? cr. II -r' o -r ri •c i - c W c 1 Tl cr. i -r' 0 T3 c 5 |3 -' C <— i 1 II C eo - — ^^ o - ^ 71 I cr C5 O IO OS oo II •a o J2 T3 e OS ^ ~ 5 I C 03 S .g 2. ~ iT c. £ >. -r M a> — bx ,£ - r r^ ~ C "c •/. — TZ: ~ ^ v -r '3 H -r' g EH ed M E"1 o ri a — Ja • j= £ ed jg •5 "S "S H >. 00 Qj M ^ ^ C r JT _!_• C au i ~ ~ .— '3 >» •c o £ T3 M 1 description. description. combined hea ead and body >, -r r jQ -r r; ~ omparisons w EH ~ tf _T >. — T3 o T3 C cd T3 rt V O a z r: ^c c ~ I £ B J= W V . _ tx "^ c c ^, c .- L-, •c ». IE IE * :5 IS 1 | I n o ~ 8 2 £ o c •a ~ ":? ~ ed a >. £ 1 . c = ri to m .£ _2 C £ I « ^i1 ed r emen *S "3 t ^ ~ "H c 1 O. — *» 8 us i- C t» ft £ ~ en cd - easur = — V ~ ] ~ j— O o O O § o oo o . " ' to t- £ £ ^ — " _r ^ "* •« *tr • •~ "o o 5 CO co" to" O o & • L." t_ a £ £ c 1 1 ,8 - •u ~ _: ^ M :: _£ ^ tx c _£ ex IO* v c _2 03 CTJ 0) be & c c c ,2 j: -? o e IO c "cd "1 1 oo »H . . .^^ IH . CO CO .._ z o o ~ - M - ^ r^ O r 4> a> m CO EH •f M b* v: — 5. cx • — • ~ ~ fli " 'Sj *S ^-* * _ . 5T — . 3 n r ed ^r ^. ^ :— EH H EH EH EH -7. H ^ > E- O CM 287 288 FIELDIANA: ZOOLOGY, VOLUME 46 andium. P. definitus grades into posticalis while magister is virtu- ally indistinguishable. The variability, plasticity and geographic position of posticalis are such that all species of the darwini group can be derived from it. Measurements.— See Table 16 (p. 286). Specimens examined. — 118. PERU. — Apurimac: Hacienda Pal- mira, 5 (CNHM). Ayacucho: Ayacucho, 2 miles south, 3 (MVZ); Puente Pajonal, Ocros, 4 (CNHM) ; Puquio, 35 km. ENE, 2 (MVZ) ; Puquio, 15 km. NE, 1 (MVZ); Tambo, Polanco, 1 (CNHM). Cusco: Ampay, 3 (CNHM); Ccolini, Marcapata, 1 (CNHM); Chospioc, 5 (USNM); Cusco, 4 (CNHM); Fundo Perayoc, 1 (CNHM); Haci- enda Ccapana, Ocongate, 1 (CNHM); Hacienda Paullo Grande, Calca, 2 (CNHM); Hacienda Urco, 4 (CNHM); Huaracondo, 1 (USNM); Ollantaytambo, 1 (USNM); Sacsahuaman, 5 (CNHM). Huancavelica: Hacienda Piso, Locroja, 2 (CNHM); Huancavelica, 16 (CNHM) ; Lircay, 4 (CNHM) ; Mayoc, 3 (CNHM) ; Rio Mantaro, 2 (MVZ); San Jenaro, Santa Ine"s, 5 (CNHM). Junin: Carhua- mayo, 8 (CNHM); Junin, 2 (CNHM); La Oroya, 3 (CNHM). Lima: Casapalca, 11 (MVZ). Puno: Limbani, 8 miles SSW, 16 (MVZ); Limbani, 4 miles SSW, 2 (MVZ). Phyllotis darwini magister Thomas Phyllotis magister Thomas, 1912, Ann. Mag. Nat. Hist., (8), 10: 406. Pearson, 1958, Univ. California Publ. Zool., 56, (4): 429— characters. Phyllotis magister magister, Pearson, 1958, Univ. California Publ. Zool., 56, (4): 430 — PERU: Arequipa (Arequipa; Chihuata, 2 miles east; Mt. Misti; Tingo; Yura); Moquegua (Rio Torata); Tacna (Tarata); taxonomy. [?] [Phyllotis darwini] limatus, Thomas (part, not Thomas), 1920, Proc. U. S. Nat. Mus., 58: 230— PERU: Arequipa (Arequipa). Type. — Male, skin and skull, British Museum (Natural History) no. 0.10.1.31; collected 29 March, 1900, by Perry 0. Simons. Type locality. — Arequipa, Arequipa, Peru; altitude, 2300 meters. Distribution (figs. 70, 73) . — Western slope of the Andes of south- ern Peru in the departments of Arequipa, Moquegua and Tacna; altitudinal range from approximately 2300 to 4000 meters above sea level. Characters. — Largest race of Phyllotis darwini; color intermediate between rupestris and posticalis; dorsal surface buffy or ochraceous finely pointed with black; head and shoulders sometimes contrast- ingly paler than back; underparts grayish, usually washed with buffy and with the plumbeous basal color of the hairs showing through; pectoral spot or streak present and well defined; tail bicolor, thinly HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 28!) to moderately haired and shorter to longer than head and body combined, average 103 per cent (80-119, 22 specimens). Auditory bullae moderately inflated, average anteroposterior length, less tubes, about 5.1 mm. (4.8-5.3, 18 specimens), and always less than alve- olar length of molar row; mesopterygoid fossa comparatively broad with anterior border square or slightly rounded, rarely pointed as in definitus, posteromedian palatal spine hardly or not at all devel- oped ; posterolateral palatal pits anteriad to posterior border of pal- ate; supraorbital region usually with a median sulcus, sometimes flattened, edges sharp or rounded, pinched upward or divergent and sometimes forming narrow ledges posteriorly; proximal ends of nasals pointed to truncate and terminating from slightly behind to slightly in front of fronto-premaxillary sutures; combined alveolar length of molars in adults, 6.1 mm. (5.7-6.8, 22 specimens) or 19 per cent (17-20) of greatest skull length. Variation. Topotypes and other specimens from localities near Arequipa (Yura; Tingo; Mt. Misti) are palest with pectoral streaks well defined. A series of three grayish brown individuals from Chihuata, 3125 meters altitude and about 3 kilometers east of Arequipa, is darkest, the pectoral streaks extremely weak, the tails more hirsute than in the preceding. Eleven specimens of magister taken near Tarata, Tacna, at altitudes ranging from 11,000 to 12,800 feet, are as pale as the Arequipa magister and slightly darker throughout than eleven specimens of rupestris variously collected between 11,500 feet and 13,500 feet in the same area. A distinct pectoral streak is present in all the magister but in only two rupestris from 13,200. Three magister and six rupestris from Rio Torata, Moquegua, altitude 12,200 feet, are slightly paler than their respec- tive relatives of the above-mentioned localities, the rupestris more markedly so than the magister. Palest individual of magister is an old male from Yura, near Arequipa, with grayish mottled head and shoulders. It is slightly darker than the average of a sympatric series of 29 rupestris. None of the Yura rupestris has a pectoral streak, tails of all are hairier, and one, with gray shoulders and head grayish as in the Yura magister, has completely white under- parts. P. (I. magister averages slightly larger than posticalis but propor- tions of external cranial characters are approximately equal in both races. The largest magister at hand is slightly smaller than the largest posticalis but the smallest adult magister is appreciably larger than the smallest adult posticalis (cf. Tables 16, 19, pp. 286, 295). 290 FIELDIANA: ZOOLOGY, VOLUME 46 Nasals are generally expanded anteriorly in magister but in three topotypes they are tapered as in most rupestris; supraorbital region pinched, or broad and flat, the edges sometimes slightly divergent and forming narrow ledges (cf. MVZ 115792, Rio Torata; MVZ 115883, Tarata; compare with caprinus and wolffsohni); postero- lateral palatal pits described as large by Pearson (1958, p. 429) appear to me to be as small as in comparable series of any taxon of the Phyllotis darwini group. In no specimen of magister, as here defined, are the pits as large as in any of the four known specimens of Phyllotis darwini definitus. Mesopterygoid fossa in one specimen of magister from Tarata, Tacna, is asymmetrically pointed in front; in the remaining skulls the anterior border of the fossa is square or rounded, the pterygoid processes parallel-sided or slightly diver- gent posteriorly; bullar tubes are relatively short with ventral sur- face generally convex. Molars are actually (but not relatively) larger in magister than in other races; first primary fold of m^ usually persistent in moder- ately worn tooth, occasionally in moderately worn m^; second molar with a modified S- to 8-shaped pattern; m^ usually with pf 2 re- duced to an enamel island. The mid-shaft area of all bacula ex- amined is thickened (fig. 7). The same feature appears occasionally in other groups of the species. Taxonomy. — There is no evidence of intergradation between mag- ister and rupestris in two known areas of contact in southwestern Arequipa (Arequipa; Yura), one in western Moquegua (Torata) and one in eastern Tacna (Tarata, 3 sites at 11,500, 12,200, 12,800 feet, respectively). Each series of magister from each trap line is readily distinguished from its sympatriot (or perhaps compatriot) of presumably the same trap line, by overall larger size, darker color, well-defined pectoral streak and slightly less hairy tail (Table 17, p. 291). There are too few samples, however, to give assurance that complete reproductive isolation prevails at these critical points. All characters of magister, whether singly or in combinations, are of common occurrence throughout the range of Phyllotis darwini. In- dividuals practically indistinguishable from magister appear in Peru (posticalis) , Bolivia (wolffsohni), northern Argentina (caprinus), southern Argentina (xanthopygus) and Chile (darwini). Most of these large, magister-like individuals or populations have no direct contact with the populations so named. Arequipa magister, however, does merge both morphologically and geographically with Ayacucho and Apurimac posticalis. It is also evident from material at hand >.!, C . •§! .-Q; O O Q 3^T3~T?yT3_T3"T3 DC-cC-aQTjQTsQ-c P ,. = : in sc "r1 - ^•"jr~js ^"^js s — -*- «- — — -~ — - — ^C -C *-* £"3 •2 u 1 Si ft. u s s w o S e- C/2 /• -rW^C/ia- »: CO w /W tfiC/2 ^C/3 V. Ul v. m v. _Q; s CM ^ ^ § _ -* H 1 3 L> — - C_£ <•» sO oo o> .= •s. ._ * - — .. 01 X -S. ~H ' o» 2 o oc os oo" £ r; — - ^ o" t^ oo" t~ oo" oo" o t_ S) o - ••N, c i en f -oooo»t~o>t~t-ooocn T3 y . J >-i <-i C*J 2 •M CO ^ CD i^ °u . »^ ^^ 1C ^ rt 0. E •3 ^ G ^^ « 0 ^ 00 r> Tf U5 *? CO • CD *O CD to sc >> .., be — .-T ? ? oi of o oT o" oT "^ » — g— yj u 5 •*1>CDl"t/itO IC^O *\ •* E „ o c 5 co *^i t^» oj » 01 t^~ ^f os *c 2 '% co zo a 5 CD to 10 lO CD "*T O t/D 1/5 10 - ^^ co ~ £ *J ^ •M CO —< be £ ri »\r I-^T ^ S V. " • OS • OO CD to 0-J — •Si S _c co" £3 o S ri 06 ^f o ~ S» 3 ti M SJ - c | CO f CO OJj CO CO °. 01" o ^ o" oi os" . 3 C ^ O V- i *^ 3; oo o oo' o-i oo' co os' 3 ^^ 1 O W gj, CO^ Ol CO C^i ^o, •S 5 ^ oo co" « 3 O t~- OS »— i CO t~- OS OO »— * to 'C o 'f M CD ~ 01 D M' oo' Os' Os' — i t-^ M' O O Os' w 'S C co co 7 icocjcMoicoeMcocococM oj rt •S CD o f CD •-! — T3 v; •~i 5i ^ "™i C SS 2 o oo" 10 /yi co M CD oT Ol 1* i-i '^f O to i— i •S .***> ^j ^^ ^"1. — 1~^ [ »— < f— ^ 1— 1 f— « I— g oo H 2 : t-1 "5 *£ —^ -riT ocT "*• r> o co — Si E 2 ^ oo oo" - i«'ooiot--T-ro'cocot^!-roo" £ ** y lO — . "> ico'-'oj'— ico-^--rOco'-< **" S C •2 §? o • C r" — *u >, "M r^ T3 "1. •1* C^J H. S o O •" 5 2 O_| CO ^ 10 ^ co ^ -r oo" •M ^' O-l 'M 'I' ^' 3 "I 0 c >- ^ S 00 tf •o ^2 5 ^ oo" ^ -i 2 "M" ^'" 1 1 • » os ^— os * *[ — t , *. o cB - ^" •- — O> 00 - 'i' »O — C c -r t~ oo" co co" °l »-- "- o" co" 5-rooocooscoOt-i'— i 1 «: < I-C?-T?-T l§. E"1 o o o o o o "*" !S • >> ii^ p o p o o o ° J' X* ^'S'^ r _ 'i i: -" -' ^f £[ ?J ^r S £ . o '3 '3 '= !-^-i'5532f3225 Hfii o —• cr c J OJ Ol S c S-s do " •55-0-5 — Op 291 292 FIELDIANA: ZOOLOGY, VOLUME 46 that rupestris in departing from the area of sympatry becomes larger and darker in northeastern Arequipa (chilensis of Pearson, cf. Cailloma series), and grades into the same peripheral populations of posticalis (fig. 73). P. darwini limatus of the southwest coast of Peru is another annectant form. It more nearly resembles magister than rupestris of southern Peru by its larger size, tail proportionately longer, underparts darker, sometimes with a faint pectoral streak. P. d. limatus grades into posticalis but its precise geographic relationship with magister is unknown. Intergradation between magister and posticalis is best illustrated with a series of three specimens collected by Pearson at Chihuata, about 143^ kilometers east of Arequipa. The specimens are in old pelage and sensibly darker, their pectoral streaks much weaker than in other examples of magister from the Arequipa region. These Chihuata magister are virtually indistinguishable from three speci- mens collected by Pearson in the vicinity of Puquio, Ayacucho, and identified by him as Phyllotis darwini posticalis. The only notable difference of "significance" between the two series is the absence of a pectoral streak in two of the Puquio specimens and its mere indication in the third (MVZ 116024) by a few pale buff hairs. Resemblance between the Chihuata magister and a series of five specimens from Puente Pajonal, Ayacucho, and another of the same number from Palmira, Apurimac, with pectoral streaks better developed, is complete enough to make distinctions between magister and southern populations of posticalis an arbitrary matter. Phyllotis darwini posticalis, as now understood, is a composite of discrete magister-like and rupestris-\ike populations which inter- breed at points of contact throughout most of the range. In south- eastern Peru, where the magister and rupestris types diverge more from each other, interbreeding seemingly does not occur and the two forms behave and appear as distinct species at known points of contact. Magister represents the end of the speciation ring in posticalis. Farther south, in Chile and northern Argentina, fully in- terbreeding pockets of large and small populations reappear. Some have received distinctive names (vaccarum, ricardulus, etc.) but all are treated here as members of a single race, rupestris. Phyllotis definitus Osgood was classified by Pearson as a subspe- cies of Phyllotis magister on the basis of large size and correspond- ingly large molars. The same characters, however, are shared with all other large members of the darwini group. The real differences HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 293 between magister and definitus (cf. following section and p. 297) are accentuated by a geographic gap of 1100 kilometers which is largely occupied by Phyllotis darwini posticalis. Whatever the relationship between magister and definitus it can be derived only from their com- mon relationship with posticalis. That definitus grades into posti- calis is demonstrated elsewhere. It is shown now that magister also combines so intimately with populations of posticalis less than 300 kilometers away (Puquio; Yura and Chihuata) that no line of sep- aration can be drawn between the two forms. A parallel morphological and geographical relationship exists be- tween magister and wolffsohni of the eastern Andes of Bolivia. Pear- son (1958, p. 431) pointed to the situation but dismissed the possibility of subspecific relationship between the transandean forms. It is shown in the following section and in the account of wolffsohni that no real differences separate the two forms and that intergradation takes place through geographically intermediate rupestris and caprinus (fig. 77). Remarks. — According to Thomas, the type of magister was "caught at the same place [i.e., Arequipa] as a number of examples represent- ing the northern form of Ph. darwini." Corroborating evidence of sympatry was gathered by Pearson (1958, p. 400). He observed that "Phyllotis magister magister is sympatric on the Pacific slope of southern Peru between 7,000 and 13,000 feet with P. darwini rupestris, one of the smallest of the subspecies of darwini. In the three areas where P. magister magister has been taken (Departments of Arequipa, Moquegua, and Tacna), darwini shares its habitat but unlike magister ranges besides into the more scantily vegetated des- ert at lower elevations and onto the altiplano at higher elevations. Actually, darwini tends to avoid the densest brush. I have taken magister only in the zone of maximum vegetation on the Pacific slopes of southern Peru, a zone where bushes and cacti form a rela- tively thick cover. In a collection made between January 23 and 27, 1952, at various elevations above Tarata, Department of Tacna, the two species were represented in the ratios shown in Table 2. If traps had been set at even lower altitudes, the catch of magister would presumably have decreased and that of darwini increased." The cited Table 2 is reproduced in my Table 18. The interpola- tions in brackets are explained in the text. I cannot understand the logic behind Pearson's conclusion that at levels lower than 11,000 feet "the catch of magister would pre- sumably have decreased and that of darwini increased." Elsewhere 294 FIELDIANA: ZOOLOGY, VOLUME 46 TABLE 18. — Altitudinal distribution of Phyllotis darwini magister Thomas and P. darwini rupestris Gervais captured above Tarata, Depart- ment of Tacna, Peru (Table 2, ex Pearson, 1958, p. 400). Elevation Number of Number of (in feet) magister darwini [15,200] [0] [1] 14,600 0 [0] 12 [5] 13,600-13,200 0 [0] 6 [3] [13,000] [0] [2] 12,800 7 [3] 5 [5] 12,200 3 [1] 6 [2] 11,500 14 [3] 10 [4] 11,000 7 [4] 0 [0] [Totals 31 (11) 39 (22)] in the same region Pearson had taken magister together with rupestris at altitudes as low as 7000 feet. Furthermore, figures in Pearson's Table 2 (Table 18, above) showing the number of specimens taken at each station above Tarata may be questioned. The interpolated figures in brackets alongside those quoted represent the actual number of specimens listed by Pearson (1958, pp. 417, 418, 431) as "Specimens examined" from the altitudinal stations shown. The total of magister and darwini in Table 2 (Table 18) is 31 and 39, respectively. In his "Specimens examined" section, Pearson lists only 11 and 22 specimens, respectively, from the same stations. The last pair of figures agrees with the number of specimens made avail- able to me for study by the Museum of Vertebrate Zoology. The Tarata material was collected from 23-28 January, 1952. The field numbers for the Tarata series of magister and darwini combined run from a low of 738 to a high of 755 (A. K. Pearson) and a low of 3546 to a high of 3566 (0. P. Pearson). If all numbers in the two series were of specimens of magister and darwini, assuming no other kinds of animals were taken, the totals would be 39 or still 31 less than the totals in Pearson's Table 2. Unless the figures of his Table 2 include specimens first identified in the field and then discarded, no credence can be placed in them. The statistical analysis of the measurements of magister com- pared with sympatric populations of darwini given by Pearson in his figure 9 (p. 430) also raises questions. In his "Specimens exam- ined," Pearson lists a total of 8 magister and 11 darwini (i.e., rupestris) from Arequipa proper. Presumably all are of the original Simons 1899 collection. The measurements given in Pearson's figure 2, how- HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 2H5 TABLE 19. —Measurements of Phyllotis danrini magister Thomas Locality Head and body Tail Hind foot Ear Arequipa' 152 158 32 29 Arequipa 118,138 128,109 28,28 27, Mt. Misti 118 118 27 26 Chihuata 135,134,143 129,134,122 27,28,29 27,27,25 Tingo 124 106 27 Yura 144 138 28 27 Rio Torata 108,108,122 127,127,145 27,28,29 24,26,25 Tarata, 1 km. N 118,121,133 ,135 124,125,131 ,129 27,28,28,29 26,25,26,24 Tarata, 2 km. N 133,131,127 130,134,143 28,28,29 25,25,25 Tarata, 3 km. N 131 143 30 25 Tarata, 4 km. N 110,142,144 131,154,156 29,28,32 24,26,29 Alveolar Skull, greatest Zygomatic length of Locality Weight length breadth molar row Arequipa1 36.8 18.5 6.0 Arequipa — -.32.3 -.16.0 6.1,6.0 Mt. Misti — 31.3 15.7 6.0 Chihuata 67,66,88 32.6,33.3,34.2 16.5,17.0,16.7 6.0,6.1,6.0 Tingo - 32.6 15.5 6.2 Yura — 32.0 16.8 6.3 Rio Torata 29.9,30.9,30.9 15.7,15.8,15.9 5.7,6.2,6.0 Tarata, 1 km N 58,55,93,71 31.3,31.8,33.4,34.5 16.2,16.4,17.4,17.9 6.1,6.1,6.3,5.9 Tarata, 2 km N 70,58,75 31.6,32.1,32.8 16.6,16.2,17.1 6.0,5.9,6.8 Tarata, 3 km N 72 32.9 16.7 5.7 Tarata, 4 km N 44,65,89 30.1,33.2,34.5 16.2,17.1,17.0 5.9,5.9,6.4 1 Type of magisier; measurements from original description. 296 FIELDIANA: ZOOLOGY, VOLUME 46 ever, are for 15 magister and 37 darwini. Obviously, Pearson in- cluded other "Arequipa region" specimens which may or may not be sympatric and synchronic. Similarly, 13 Tacna and Moquegua magister are collectively compared with 9 darwini from the same Departments. Measurements. — See Table 19 (p. 295). Specimens examined. — 22. PERU — Arequipa: Chihuata, 2 miles east, 3 (MVZ); Arequipa, 2 (AMNH; CNHM); Mt. Misti, 1 (CNHM); 'tfngo, 1 (AMNH); Yura, 1 (CNHM). Moquegua: Rio Torata, 15 km. NE of Torata, 3 (MVZ). Tacna: Tarata, 11 (MVZ). Phyllotis darwini definitus Osgood Phyllotis definitus Osgood, 1915, Field Mus. Nat. Hist., Zool. Ser., 10: 189. Phyllotis magister definitus, Pearson, 1958, Univ. Cal. Publ. Zool., 56: 431 — Ancash (Macate); taxonomy. Type. — Adult male, skin and skull, Chicago Natural History Museum no. 21125; collected 14 February, 1914, by M. P. Anderson. Type locality. — Macate, north-central Ancash, Peru; altitude, ap- proximately 2800 meters. Distribution (figs. 70, 73). — Known from type locality only. Characters. — More brown than posticalis or rupestris, molars larger; dorsal surface warm brown, the anterior half slightly griz- zled; underparts washed with ochraceous and not sharply defined from sides; a well-defined pectoral patch present; bicolor tail sub- equal to or shorter than combined head and body length; auditory bullae moderately inflated, antero-posterior length, less tubes, about 5.0 mm. (4.7-5.1, 4 specimens) and less than alveolar length of molar row; bullar tubes short and with a pronounced keel on ventral sur- face; anterior border of mesopterygoid fossa pointed; posterior me- dian border of palate without spinous process; posterolateral palatal pits large and on a line with or anteriad to anterior border of meso- pterygoid fossa; supraorbital edges rounded, the ridges hardly raised; proximal ends of nasals truncate and terminating on a line with or slightly behind premaxillary sutures; molars large, combined alveo- lar length, 6.0-6.2 mm. (4 specimens) or 19-20 per cent of greatest skull length. Variation. — Most of the slight amount of individual variation noted is detailed in the characterization of the subspecies. It must be emphasized that definitus is based on four specimens representing a single age group of a single crop. HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 297 TABLE 20.— Measurements of Phyllotis darwini definitus Osgood Head and Hind Skull, greatest Zygo- matic Alveolar length of Macate body Tail foot length breadth molar row C 21125 d" 131 132 28 31.8 16.8 6.3 C 21126 9 124 120 27 30.2 16.2 6.1 C 21127 9 135 124 28 32.2 16.2 6.2 C 21128 9 121 111 29 30.7 16.1 6.0 C = Specimens in Chicago Natural History Museum. 1 Type. Taxonomy. — Present classification of Phyllotis definitus as a sub- species of P. darwini nearest related to posticalis depends partly on negative evidence. Morphological intergradation between definitus and posticalis can be demonstrated. There is no reason to believe that geographical intergradation does not occur also. Pearson (1958, p. 414) postulated that "collecting in north-central Peru will probably reveal posticalis or some other race of darwini in the high mountains of Ancash." This "other race of darwini" can be no other than definitus. Pearson, however, treated Phyllotis definitus together with Phyllotis magister as a species distinct from darwini. Except for parallel development in size, there is nothing to support this thesis. Definitus and magister are separated by a geographical gap of 1100 kilometers and intergradation between the two can be effected only through the geographically and morphologically intermediate Phyl- lotis darwini posticalis. Phyllotis darwini definitus is much larger and heavier than sym- patric Phyllotis andium, more richly colored brown, underpants more ochraceous with a well-defined pectoral patch always present, tail bicolor to tip; anterior border of mesopterygoid fossa not rounded or square, the palate relatively narrower with well-marked longi- tudinal ridges. Allopatric individuals of andium of equal body size are more lightly built with smaller crania and molars. Specimens of P. darwini posticalis, on the other hand, commonly equal and even exceed the size and bulk of definitus and in many the molars are as large. However, the short, thick rostrum and pointed ante- rior border of the mesopterygoid fossa characteristic of definitus are uncommon in posticalis. Huancavelica representatives of posticalis, notably a series of four specimens from Lircay, most nearly rasem- ble the four known specimens of definitus. Measurements. — See Table 20. Specimens examined. — 4 including the type, all from Macate, Ancash, Peru (CNHM). FIG. 82. — Age-size variation in Phyllotis darurini limatus: a, skull and molars of a juvenal from 4 miles east of Santa Eulalia, Lima, Peru; b, extremely old indi- vidual from 3 miles west of Atico, Arequipa; c, juvenal from Hacienda San Jacinto, lea, lea. (Skulls about X 2; molars about X 10.) 298 HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 299 Phyllotis darwini limatus Thomas H[esperomys] Darwinii, Tschudi, 1844, Fauna Peruana, 1: 16, 181, 184 — PERU: Lima (Huacho). Phyllotis darwini limatus Thomas, 1912, Ann. Mag. Nat. Hist., (8), 10: 407 — part, type only. Pearson, 1958, Univ. California Publ. Zool., 56: 413 —PERU: Arequipa (Atico; Chucarapf; Tambo); lea (lea); Lima (Chos- ica; Huinco; La Palma; Nafta; Oscolla; Rimac Valley; San Bartolom6 Station; Santa Eulalia; Surco; Verruga Canyon; Yaca). Type.—Ma\e, skin and skull, British Museum (Natural History) no. 0.5.7.43; collected 29 January, 1900, by Perry O. Simons. Type locality.— Chosica, near Lima, Peru; altitude, 850 meters. Distribution (figs. 70, 72, 73).— Coast and foothills of the western slope of the Andes of Peru in the departments of Lima, lea and Arequipa. The range may continue along the coast into northern Chile where intergradation with P. d. darwini may occur (cf. Pearson, 1958, p. 421). Characters. — General color and body size as in rupestris, tail pro- portionately longer; head grayer than trunk; lateral line not always well defined and rarely continuing around base of rump; underparts sharply defined whitish to dirty gray lightly washed with buff; a weakly marked pectoral streak sometimes present; toil bicolor, well- haired with well-developed pencil and nearly always longer than head and body combined (tail: head and body=118 [98-126]: 100, 25 specimens); auditory bullae well inflated, antero-posterior length, less tubes, subequal to or, usually, longer than alveolar length of molar row; other cranial characters as in rupestris; enamel pattern of m1 usually more S- than 8-shaped (fig. 82); first primary fold of moderately worn m-8- well defined and, usually, at least indicated in moderately worn ma; second primary fold of ma much more fre- quently open to the margin than forming an enamel island; alveo- lar length of molar row 4.6-5.2 or 17-18 per cent of greatest skull length. Variation. — The palest samples are from lea and western Are- quipa (Chucarapf; Atico), the darkest are seven specimens from Chosica collected in August, 1955, by 0. P. Pearson. In contrast, a series of three specimens from the same general locality taken in January 1942, by C. C. Sanborn, are more buff, less gray on upper parts and distinctly less gray on underparts. In all, they resemble more nearly the palest representatives of the race from lea and Are- quipa than they do their topotypes. Pearson alludes to the differ- ence in color by remarking that his "specimens captured in 1955 in 300 FIELDIANA: ZOOLOGY, VOLUME 46 relatively wild parts of the Rimac Valley [Chosica; San Bartolome"; Rimac Valley; Surco; Nafia; Santa Eulalia] have grayish bellies reminiscent of those mice living in houses. This discoloration may be due to smoke from the railroad and from the annual burning of cotton plants in the valley." Whatever the cause, a test with detergents proves that even with grime removed the darker color of the underparts persists as a character of Pearson's specimens from Chosica, "Rimac Valley," San Bartolom£ and Surco. Under- parts of the three specimens from Nafia in the Rimac Valley are whitish and those of the five specimens from Santa Eulalia are inter- mediate. All series in question with underparts washed or unwashed can be precisely matched in color with series from temperate zone localities in various parts of Peru and northern Chile where neither railroads nor cotton plant burning occurs. It is interesting to com- pare Pearson's explanation for the difference in color between the allochronic Chosica series and the somewhat analogous Salinas series of rupestris (cf. p. 308). Two specimens collected 19 March, 1951, in Atico, Arequipa, by 0. P. Pearson are in new buffy pelage while two specimens taken 20 May, 1952, are in old worn and distinctly redder pelage. The terminal portion of the tail is marked by a white band in the type of limatus (Chosica, Lima, Peru) and two of ten topotypes examined. It also occurs in one of five specimens from "Rimac Valley," in one of two specimens collected at Nana, one in Surco and in the single specimen from Oscolla, all localities in the Rimac Valley above Lima. The latter (CNHM no. 53056) was captured alive 28 December, 1941, and kept in a laboratory in Lima until its death, 10 February, 1942. It is remarkable for white patches on the upper surface of the distal fifth of its tail, and the unusual color of its body. The anterior half of the upper parts is pale as in average Arequipa rupestris or limatus, the posterior half more warmly colored than in any rupestris or limatus. The general coloration may represent that of the wild Oscolla population but this is unlikely. No specimen from other localities in the same general region resem- bles it. Possibly the color pattern was evoked by the unnatural laboratory environment. The Surco individual (MVZ 120033) with braincase monstrously enlarged by hydrocephaly has already been noted by Pearson (loc. cit.) and is figured here (fig. 80, a). The well-defined first and second primary folds in the molars of P. darwini limatus is a character common to most representatives of P. darwini of western Peru and northern Chile. > £.2 us eg eg us eg eg o lO lO US oo •**" co o o" ~ 10 eg* o -«1« lO lO 10 lO lO lO •*? lO JO 10 I o eg S3 oq q oq •«* 10 T' - E o JS ~ = — 3 -^ w J to eg i CO t- ""• eg^ cg^ e^ t-T t-T to eg eg eg 10 ri w 71 ^ 55^ i - 7! US to eg_ to* e-i_ us ^T -- t-^ to" -^ to" us 71 71 eg eg eg e-j eg — ••t /: vr ^" — — **i* co C N oo Oi e-i eo 1 eg* CO o us US -T t" ^" ^r o »o CO T7 CO •* DO ei "~ £ us 7 1 r - i o i^ — t - S oo to W r H 10 ri ia o" -r r o* ,r 1 _o A "$ ^ A "o •7; 1 — * _w G E •*-* y "* 02 K PL, '7. o *$ -^ £ U s -C •_ ' / : -=. U cs « cffi '§ CS 1-7 C/3 Z I S a w 3 CO nipa ico 301 302 FIELDIANA: ZOOLOGY, VOLUME 46 There is no consistent correlation between skull size and molar wear in individuals of different series. A Juvenal from Santa Eulalia, Lima, with greatest skull length of 25.4 and alveolar length of molar row 4.9, has slightly more worn first and second molars than another juvenal from lea with skull 30.5 and alveolar length 5.4. This last, in turn, is appreciably larger than an apparently old and fully grown individual from Chucarapi, Arequipa, with skull 29.5, molar row 4.9 and crowns more worn than those of all but one of the 36 specimens of limatus at hand (fig. 82). The difference in size and tooth wear between examples cited above is ostensibly related to the quantity, quality and availability of nourishment in the habitat of each. Taxonomy. — Phyllotis darwini limatus Thomas was originally de- scribed as the "northern representative of Ph. darwini [but] paler and [with] longer tail than the typical Chilean form. Specimens referable to it are in the [British] Museum from various parts of the highlands of Peru and Bolivia, including Caylloma, Arequipa, La Paz, etc." In effect, limatus of Thomas, as originally conceived, is equivalent to rupestris Gervais, a name which, at the time, was linked with the genus Akodon and therefore not taken into account. Pear- son (1958, p. 413) redefined limatus and restricted it to its present range along the coast and Andean foothills of Peru. He suggests that limatus "may intergrade with P. d. posticalis on the west slope of the Andes somewhere south of the Rimac Valley (central Lima) . . . but in the valley it is separated from posticalis by a zone of brush occupied by Phyllotis andium." He also believes that limatus "prob- ably intergrades with the more inland P. d. rupestris in some valleys in the department of Arequipa and possibly in northern Chile." Measurements. — See Table 21. Specimens examined. — 37. PERU: Arequipa: Atico, 3 miles west, 4 (MVZ) ; Chucarapi, Tambo Valley, 2 (MVZ). lea: Hacienda San Jacinto, 2 (CNHM). Lima: Chosica, 10 (CNHM, 3; MVZ, 7); Nana, 2 (MVZ); Oscolla, 1 (CNHM); Rimac Valley, 5 (MVZ); San Bartolom£ Station, 4 (MVZ); Santa Eulalia, 5 (MVZ); Surco, 1 (MVZ); Veruga Canyon, 1 (MCZ). Phyllotis darwini rupestris Gervais Mus rupestris Gervais, 1841, Voy. "Bonite," Zool., 1: 51. Gervais, 1841, Bull. Soc. Philom., Paris, p. 115— "BOLIVIA" [ = Chile]: Antofagasta (Cobija, skeleton only). Gay, 1847, Hist. Chile, Zool., 1: 115, atlas, pi. 6, fig. 1 (skull and dentition), pi. 7 (animal figured not the type, may be Phyllotis darwini darwini). HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 303 M[us (Abrothrix)\ rupestris, Gervais, 1849, Diet. Univ. Hist. Nat., 10: 731 characters; classification. Abrothrix rupestris, Lesson, 1842, Nouv. Tabl. Reg. Anim., p. 136 classi- fication. I'hyllotis darwini rupestris, Osgood, 1943, Field Mus. Nat. Hist., Zool. Ser., 30: 205-6— taxonomy; CHILE: Antofagasta (San Pedro de Atacama; 20 miles east of San Pedro, Rfo San Pedro); BOLIVIA: Potosl (Lipez); ARGENTINA: Jiijuy (Casabindo). Mann, 1945, Biologica, Trab. Inst. Biol. Univ. Chile, fasc. 2, p. 39— CHILE: Tarapacd (Valley of Oases; Pampa of Tamarugal). Mann, 1950, Investigaciones Zool. Chilenas, fasc. 2: 5— CHILE: Tarapacd. Pearson, 1958, Univ. California Publ. Zool., 56: 417 — revision; distribution [for locality records see p. 244). Mus capita Philippi, 1860, Reise durch die Wiiste Atacama, p. 159, Zool., pi. 2, fig. 2 (animal), pi. 5, fig. c (skull) CHILE: Antofagasta (type locality, Hueso Parado, near Taltal); name preoccupied by Mutt capita Offers, 1818, an Oryzomys. Philippi, 1900, Anal. Mus. Nac. Chile, Zool., Ent., 14a: 10, pi. 1, fig. 2 (animal) — characters. Hcsperomys glirinus Philippi, 1896, Anal. Mus. Nac. Chile, Zool., Ent., 13a: 19, pi. 7, fig. 3 (animal, feet, molars) CHILE: Antofagasta (type locality, San Pedro de Atacama). Mus glirinus, Philippi, 1900, Anal. Mus. Nac. Chile, Zool., Ent., 14a: 59 characters. Hesperomys lanatus Philippi, 1896, Anal. Mus. Nac. Chile, Zool., Ent., I3a: 19, pi. 7, fig. 2 (animal, skull, molars)— CHILE: Antofagasla (type local- ity, San Pedro de Atacama). Mas lanatus, Philippi, 1900, Anal. Mus. Nac. Chile, Zool., Ent., 14a: 59 characters. rhyllotis arenariu* Thomas, 1902, Ann. Mag. Nat. Hist., (7), 9: 224 BO- LIVIA: type locality, Uyuni, Potosf, altitude, 3670 meters; Orura; Sucre; Polosi; PERU: Arequipa. Thomas, 1913, op. cit., (8), 11: 139 ARGEN- TINA: Jujuy (Cerro de Lagunita, 4500 meters). Yepes, 1933, Rev. Chilena Hist. Nat., 37: 47— ARGENTINA: Catamarca (Laguna Blanca); Los Andes (San Antonio); Jujuy (Sierra de Zenta). I'hlyllotis] arenarius, Yepes, 1935, Rev. Inst. Bacteriol., 7: 223 characters; distribution. [rhyllotis lutescens] arenarius, Osgood, 1944, Field Mus. Nat. Hist., Zool. Ser., 29: 194— classification. Phyllolis darwini raccamm Thomas, 1912, Ann. Mag. Nat. Hist., (8), 10: 408 —ARGENTINA: type locality, Las Vacas, Mendoza, 2500 meters; Mrn- doza (Puente del Inca). Phyllotis darwini taccarum, Osgood, 1943, Field Mus. Nat. Hist., Zool. Ser., 30: 205- characters; distribution; CHILE: Coqitimbo (Bafios del Torn; Paiguano); Atacama (Domeyko); ARGENTINA: Ncuqufn (Chos Malal); Mendoza (Punta Vaca). Yepes, 1935, Rev. Inst. Bacteriol., 7: 223, pi. 4, fig. 2 (animal)- characters. Yepes, 1936, Nov. Reunion Sor. Argentina Patol. Reg. en Mendoza, Publ. 31: 710 -ARGKNTINA: Mrndma (Ma- largiie [Canada Colorada]). Pearson, 1958, Univ. California Publ. Zool.. 304 FIELDIANA: ZOOLOGY, VOLUME 46 56: 419 — revision; synonyms, oreigenus Cabrera, wolffhuegeli Mann; dis- tribution [for locality records see p. 244). Phyllotis ricardulus Thomas, 1919, Ann. Mag. Nat. Hist., (9), 3: 493— ARGENTINA: type locality, Otro Cerro, 45 km. west of Chumbicha, "Rioja" (=Catamarca), 3000 meters. Thomas, 1919, op. cit., (9), 4: 130 —part; ARGENTINA: Jujuy (Abrapampa; Casabindo). Thomas, 1920, op. cit., (9), 6: 417— ARGENTINA: La Rioja (La Invernada; Potre- rillos). Thomas, 1921, op. cit., (9), 8: 611— ARGENTINA: Jujuy (Sierra de Zenta). Thomas, 1926, op. cit., (9), 18: 194 —part; BO- LIVIA: Potosi (Lipez). Yepes, 1935, An. Mus. argentine Cienc. Nat., 38: 335— ARGENTINA: Jujuy (La Laguna). Yepes, 1936, Physis, 12: 37— ARGENTINA: La Rioja (San Antonio; Puesto Viejo; Chilicito; Las Piriquitas). [Phyllotis darwinii] ricardulus, Osgood, 1944, Field Mus. Nat. Hist., Zool. Ser., 29: 194 — "most of the specimens referred by Thomas to his ricardulus probably should be called rupestris." Phyllotis darwini ricardulus, Pearson, 1958, Univ. California Publ. Zool., 56: 418— revision; ARGENTINA: Catamarca (Chumbicha; Otro Cerro); Los Andes (San Antonio de los Cobres); Salta (G. M. Sola; Incamayo). Phyllotis oreigenus Cabrera, 1926, Rev. Chilena Hist. Nat., 30: 319, fig. 42 (molars) — ARGENTINA: type locality, Laguna Blanca, Catamarca, 3100 meters; Los Andes (Salar de Antafalla). Yepes, 1935, An. Mus. argentine Cienc. Nat., 38: 335 — listed. Phyllotis wolffhuegeli Mann, 1944, Biologica, Trab. Inst. Biol. Univ. Chile, fasc. 1 : 102 (name), p. 108 (description), fig. 7 (alimentary canal), pi. 8 (alimen- tary canal), pi. 9 (skin), pis. 10, 11 (skull)— CHILE: type locality, Boca- toma, Lo Valdes, Cajon del Rio Volcan, Santiago, 1800 meters. Phyllotis osgoodi Mann, 1945, Biologica, Trab. Inst. Biol. Univ. Chile, fasc. 2: 81, figs. 32-36 (anat.), pi. 30 (skin), pi. 31 (skull)— CHILE: type locality, Parinacota, puna of Tarapaca. Mann, 1950, Investigaciones Zool. Chile- nas, fasc. 2: 5 — CHILE: Tarapaca. Phyllotis arenarius chilensis Mann, 1945, Biologica, Trab. Inst. Biol. Univ. Chile, fasc. 2: 84, figs. 37, 38 (alimentary tract), pi. 32 (skull), pi. 33 (skin) — CHILE: type locality, Parinacota, puna of Tarapaca. Mann, 1950, Investigaciones Zool. Chilenas, fasc. 2: 5 — CHILE: Tarapaca. Phyllotis darwini chilensis, Pearson, 1958, Univ. California Publ. Zool., 56: 415 — revision; osgoodi Mann a synonym; distribution [see p. 244 for locality records]. Phyllotis Darwini, Cabrera, 1905, Rev. Chilena Hist. Nat., 9: 16 — part, lanatus Philippi and glirinus Philippi in synonymy. Ph[yllotis] darwini, Thomas, 1912, Ann. Mag. Nat. Hist., (8), 10: 406— PERU: Arequipa. Phyllotis (Phyllotis) darwinii subsp., Pearson, 1951, Bull. Mus. Comp. Zool., 106: 142— PERU: Puno (Santa Rosa; Caccachara; Juli; Pairumani). Phyllotis darwini subsp., Thomas, 1921, Ann. Mag. Nat. Hist., (9), 8: 216— ARGENTINA: San Juan (Tontal). Phyllotis sp., Zuniga, 1942, Bol. Mus. Hist. Nat. "Javier Prado," Ano 6, no. 20: 393-396— PERU: Lima (Lomas near Lima). HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 305 Phyllotis Wolffsohni, Thomas (not Thomas, 1902), 1926, Ann. Mag. Nat. Hist., (9), 17: 321— BOLIVIA: Tarija (Sama). Phyllotis daru'inii limatus, Thomas (part, not Thomas), 1912, Ann. Mag. Nat. Hist., (8), 10: 407— PERU. Phyllotis jranthopygus, Thomas and St. Leger (not Water-house) 1926, Ann. Mag. Nat. Hist., (9), 18: 635 — "jranthopygus . . . (may) grade into the original Ph. dancini; ARGENTINA: Neuqucn (Chos Malal; Las Lajas). Thomas, 1927, op. cit., (9), 20: 200— ARGENTINA: Nruquen (Cotton Cura; Quilquihue). Phyllotis dancini xanthopygus, Osgood (part, not Waterhouse), 1943, Field Mus. Nat. Hist., Zool. Ser., 30: 208 ARGENTINA: specimen from Collon Cura, Neuqucn, only. Pearson, 1958, Univ. California Publ. Zool., 56: 420— part, ARGENTINA: Neuqufn (Collon Cura; Las Lajas; Quilquihue). Phyllotis nogalaris, Mann (not Thomas), 1950, Investigaciones Zool. Chilenas, fasc. 2: 5— CHILE: Tarapacd (Esquina; Putre). [Phyllotis dancini] osilae, Osgood (not J. A. Allen), 1944, Field Mus. Nat. Hist., Zool. Ser., 29: 194— classification. Plhyttotis] dancini donrini, Pearson (part, not Waterhouse), 1958, Univ. Cali- fornia Publ. Zool., 56: 421 (in text), Appendix 1 (measurements) Anto- fagasta (Paposa); Atacama (Vallenar); Santiago (Rfo Blanco). Type.— Skeleton only, presumably in Museum National d'His- toire Naturelle, Paris; collected 27 July, 1836, by M. Gaudichaud, during the "Voyage autour du Monde sur la Corvette 'La Bonite.' ' Type locality. — High mountains of Cobija, Antofagasta, northern Chile; specimens collected by Sanborn on the Rio San Pedro, a trib- utary of the Loa, are regarded as typical by Osgood (1943, p. 207) and Pearson (1958, p. 417) and the type locality is here restricted accordingly. Distribution (figs. 70, 73-75). Southern Peru, southwestern Bo- livia, northern Chile and northwestern Argentina, as follows: from the departments of Arequipa, Puno, Moquegua and Tacna in Peru, eastward into western Bolivia in the departments of La Paz, Oruro, Potosi and adjacent parts of Cochabamba, Chuquisaca and Tarija, thence southward into northern Argentina in the departments of Jujuy and western portions of Salta, Catamarca, La Rioja, San Juan, Mendoza and Neuqucn; in Chile from Tarapacd to northern Coquimbo thence southward through the Andes into Santiago. The range includes desert, xerophytic scrub country and grassland or puna; altitudinal range from approximately 500 to 5000 meters above sea level or to the upper limits of plant growth. Characters. Palest and smallest race of P. darwini; dorsal sur- face buffy, finely ticked to heavily lined with the dark brown guard hairs; head and shoulders usually paler than dorsum; bright ochra- ceous lateral line generally present and often continuing around base 306 FIELDIANA: ZOOLOGY, VOLUME 46 of rump; underparts sharply to moderately well defined whitish or pale gray with slaty basal portions of hairs showing through, a buffy wash often present; pectoral streak usually absent (present and well defined in less than 10 per cent of specimens examined) ; tail averaging longer than combined head and body length (tail: head and body = 81-132: 100), bicolor, well haired to nearly bushy, pencil well de- fined; auditory bullae moderately inflated, average antero-posterior length, less tubes, about 5.1 mm. (4.3-5.9, 345 specimens) and usu- ally less than alveolar length of molar row; bullar tubes rarely elon- gate; mesopterygoid fossa measured at base of hamular processes usually narrow but sometimes (particularly in Bolivia and Chile) as wide as parapterygoid fossa measured at same plane; posterior border of palate rounded, notched or with a pair of processes which some- times fuse into a single median spine; postero-lateral palatal pits usually situated anteriad to posterior border of palate, infrequently posteriad (Chile, Bolivia, Puna, Peru); supraorbital edge sharp or rounded, pinched anteriorly, rarely extended posteriorly as narrow shelves or ledges; fron to-parietal suture narrowly V-shaped to broadly arced, proximal ends of nasals pointed to truncate and ter- minating on a line with, often behind, rarely in front of, fronto- maxillary suture; alveolar length of molar row, in adults, 4.6-6.7 or 15-20 per cent of greatest skull length. Variation. — A number of individual, local and specific variables in rupestris are discussed in the descriptions of neighboring races and, particularly, the sympatric forms of the sibling species Phyllotis osilae (q.v.). Color phases in rupestris are more apparent than real. Whether pale gray, dark gray, buff, ochraceous or cinnamon all individuals in any large series are fully intergrading. Samples in different stages of molt usually differ markedly in color from each other, the older pelage being either redder, i.e., more ochraceous, or darker, more gray. Palest populations occur in southwestern Arequipa. Slightly darker populations with more gray or ochraceous inhabit northern Chile, western Bolivia and northern Argentina. Darkest represent- atives of the race are from the upper altitudinal limits of the range in the Peruvian departments of Moquegua, Tacna and Puno. Humid- ity rather than altitude appears to be the controlling factor in color, the paler series having been taken either during dry season or in year round arid localities. A direct correlation between altitude, i.e., temperature, and de- gree of intensity of pigmentation is not evident. An increase in the HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 307 thickness and length of guard hairs at high altitude, however, does result in a darker-appearing coat. This can be illustrated with the series collected by Pearson on a mountain slope north of Tarata, Tacna, and listed in the following table. TABLE 22.' — -P. d. rupestris from successively higher altitudes north of Tarata, Tacna Locality 2 km. north 3 km. north 4 km. north 4.5 km. north 5 km. north 20 km. north 25 km. north Altitude in meters 3600 3812 4000 4125 4220 4562 4750 Date collected Total Classification specimens ex Pearson (1958) 25 Jan. 1952 4 P d. nipexlris 25 Jan. 1952 •> P d. rupcstris 27 Jan. 1952 5 I> d. rhilrnw's 26 Jan. 1952 1 l> d. r/ii7rH.sv'.x- 26 Jan. 1952 •> I> d. chilenxis 28 Jan. 1952 5 I> d. chilcnsis 29 Jan. 1952 1 P d. chile nsis 1 Compare with Table 18. In the Tarata series the density of the guard hairs becomes greater with increasing altitude although the difference between series of successive stations is slight. The cumulative effect, however, is that the six specimens from the two highest stations appear to be notably darker than the six from the two lowest. Real differences in color, however, are practically nil. The tone of the guard hairs, the buffy bands of the cover hairs and the lateral line are virtually the same at all stations. A heavier coat at higher, colder localities, is also noted in mice of other areas including Rio Torata (not Tarata'), Moquegua. Here specimens taken at 2940 and 3500 meters were classified by Pearson as rupestris while those with the more promi- nent dark guard hairs taken three miles away at 4280 meters were determined as chilensis. Color difference between these is no more real than in the Tacna series. Tacna mice referred to rupestris by Pearson are smaller but also appear to be younger than his Tacna chilensis. The material from Moquegua is wholly inadequate for an accurate appreciation of size differences, if any. A remarkable fea- ture of Pearson's Tacna and Moquegua rupestris as compared with his chilensis of the same mountain sides, is the higher incidence of persistence in the former of a well-defined first primary fold in m* and, to a lesser degree, in m:y, in even well-worn teeth. This char- acter, however, is highly variable and certainly related to fodder differences at different stations. A well-defined pectoral streak is present in less than 10 per cent of all specimens examined. The character appears most frequently among specimens from the Department of Puno, with the greatest 308 FIELDIANA: ZOOLOGY, VOLUME 46 number (about 15 per cent of the total) from the area of sympatry with the sibling species Phyllotis osilae in which a pectoral streak is generally present. Four specimens, including one Juvenal, from the upper Rio Maule, Talca, are exceptionally large and more nearly resemble some of the darker Peruvian rupestris (cf. Asilla, Puno) than the generally pale Argentine or Chilean members of the race. The series was collected by Colin C. Sanborn in a mining camp 14 kilometers above Curil- linque. The oldest individual, a male, is larger than any other P. darwini examined by me. Its fractured skull (CNHM 50545), repre- sented by anterior half and fragments of posterior portions, could have exceeded 36 mm. in greatest length. Its molar row of 6.7 mm. is exceeded only by that of a specimen of P. darwini magister with an alveolar length of 6.8 mm. Evidently living conditions for Phyllotis in the mining camp were extremely favorable. The dis- cordant characters of the Rio Maule series are, possibly, manifesta- tions of niche variation. Coats of a series of 11 specimens from Salinas, Arequipa, Peru, collected by Colin C. Sanborn and associates, vary strikingly from new and bright in color to old and somber. The darkest (CNHM 49487) collected 3 October, 1939, is not quite as heavily lined with black as the paler of two specimens in prime pelage collected by 0. P. Pearson in the same locality at nearly the same altitude but on December 17 and 18, 1951. The difference in color led Pearson to suggest that Sanborn's series was "probably altered by treatment with borax or alum." This condemnation is unwarranted. The two Salinas series vary in the same way as, for example, the allochronic series of P. darwini limatus collected in Chosica, Lima, and Atico, Arequipa (cf. p. 300). Sanborn's series is no more or less discolored than his other material which includes the bulk of our southern Peruvian Phyllotis. All were prepared in a uniform manner with borax used as preservative and all, except the Salinas specimens, served without qualifications as bases for Pearson's fine discrimina- tions between species and subspecies. That Sanborn's Salinas series averages rupestris-\ike in color rather than "chilensis"-\ike, as Pear- son would desire, cannot be dismissed as an artefact. The 11 speci- mens were taken in a different year and season at a different locale and a different elevation from the two specimens collected by Pear- son. Sanborn's series is paler because of a paucity of blackish guard- hairs. It is also appreciably smaller in body size, a character which cannot be dismissed as extraneous. The differences between the HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS M9 Sanborn and Pearson series may be examples of crop variation or may point to different genotypes. The Salinas population sampled by Pearson is certainly not the same and need not be a direct de- scendant of, the Salinas population sampled by Sanborn 12 years earlier. Another consideration is that if Pearson had collected more specimens, complete intergradation between the two allochronic series might be demonstrated. Indeed, a series of 24 specimens from Cailloma, Arequipa (4330 meters), shows an even greater range of variation in color. The palest is as pale as any of the Sanborn Salinas series and the darkest (CNHM 49510, 49519) are more heav- ily lined with black and more warmly colored ochraceous than Pear- son's Salinas specimens. A short but well-defined first primary fold is normally present in m-, and usually m:t, of nearly all juvenals and many subadults. As a result of wear, the fold becomes reduced or completely erased in all adults of most series. In certain localities, however, a well- defined first primary fold persists in a large proportion of adults. The highest incidence of persistence appears in Arequipa (Arequipa; Yura; Balneario de Jesus), Moquegua (Torata at 9,400 and 11,200 feet), Tacna (below 12,800 feet), northern Chile (San Pedro; Ojos San Pedro; "Pocas") and in the vaccarutn of Pearson (Paihuano). The incidence of persistence of a well-defined pf 1 in adult m* is lower in series of Pearson's chilensis from the Arequipa localities Cailloma, Salinas and Sumbay and the Chilean locality Putre. Per- sistence of the fold elsewhere within the broad range here assigned to rupestris is insignificant and sporadic. The data, taking into account juvenals, show a tendency for the character to persist in areas from the western portion of the range of rupestris southward through the range of darwini. Eastward across southern Peru, northern Chile and Bolivia, pf 1 in adult m^ becomes weaker and virtually disap- pears at the eastern limits of the range. The character then suddenly reappears in adults of the bordering races caprinus and wolffsohni. It is significant that the highest incidence of occurrence of the first primary fold in m'-1 ^ is in specimens sympatric with magister in which the same character is found. An analogous condition, already noted, is the frequent appearance of a pectoral streak among the rupestris sympatric with Phyllotis osilae. Taxonomy. — Mus rupesfris Gervais, originally described from the arid parts of Antofagasta, northern Chile, was determined by Osgood (1943, p. 205) as a pale race of Phyllotis darwini. The type of Hes- peromys lanatus Philippi was examined by (Xsgood (op. cit., p. 208: 310 FIELDIANA: ZOOLOGY, VOLUME 46 in the National Museum of Santiago de Chile, and pronounced identical with rupestris. The same authority states that the types of the northern Chilean Mus capita Philippi and Mus glirinus Philippi are not now in the Santiago Museum but judged by the original descriptions, each may be regarded as a member of the common race. In the absence of evidence to the contrary, Osgood's conclusions are accepted. The original description of Phyllotis arenarius Thomas agrees with that of rupestris as here defined. The type locality, Uyuni, Potosi, Bolivia, lies within the same faunal area as that of rupestris. Specimens at hand from Lipez, Bolivia, practically topotypes of arenarius, are not distinctive and were already assigned to rupestris by Osgood in 1943 (p. 206). Pearson (1958, p. 417) agreed. Phyllotis darwini vaccarum Thomas, from Las Vacas (Punta Vacas), Mendoza, Argentina, judged by three topotypes at hand, should not be separated from Chilean rupestris. Osgood (loc. cit.) believed vaccarum differed from rupestris by its long tail and larger ears. Measurements were not given and none taken from the fresh skin are available for ears of specimens examined by Osgood. Tail lengths show no consistent difference. Osgood also distinguished vaccarum from typical darwini by its paler color. This character barely separates average rupestris from average darwini. Cutting this distinction finer for admitting another subspecies hardly im- proves the classification of these difficult rodents. Pearson also recognized vaccarum but on the basis of size alone, the "head and body, skull and tooth row [being] longer than in P. d. darwini, rupestris and ricardulus." Pearson examined 127 speci- mens of vaccarum from nine widely distributed localities in Argentina and seven in Chile but gives measurements for only 10 to 18 speci- mens from four closely grouped Argentine localities (Punta Vacas, nearby Puente del Inca and not very distant Pedernal and Los Sombreros). Specimens from all other localities are treated by Pearson as intergrades between vaccarum and one or another of the other races mentioned. The type of vaccarum, collected in 1901 by P. 0. Simons, deviated from Pearson's concept of the sub- species and was replaced by a "plesiotype" collected 20 years later by Budin. It appears that the "plesiotype" series of 21 specimens forms the chief if not the only basis for Pearson's characterization of vaccarum. This series and others from nearby localities average larger in size than Pearson's restricted rupestris but with such wide overlap in measurements that subspecific distinction is hardly in- HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 311 dicated. Like northern chilensis, from which it is indistinguishable, vaccarum represents the opposite end of the gradient from rupestris to larger posticalis in the north and xanthopygus in the south. Phyllotis ricardulus Thomas, described from Otro Cerro, Cata- marca, Argentina, was distinguished from vaccarum by its "much smaller" ears. It was also compared with wolffsohni and tucumanus (=P. osilae). Pearson (1958, p. 418) characterized the type of ricardulus "as an example of the species darwini with large ears." He distinguished it from rupestris by its darker color, longer tail and larger molars and from vaccarum by smaller body, skull and tooth row. In effect ricardulus differs from rupestris by being more like vaccarum and from vaccarum by being more like rupestris. Clearly, ricardulus is nothing more than part of the cline leading from smaller rupestris to larger vaccarum. Published measurements of the type of Phyllotis oreigenus Cab- rera, from Laguna Blanca, Catamarca, and the figure of its upper molars with well-marked first primary folds point only to a Juvenal of the common species. Persistence of a rudimentary mesoloph in the first and second upper molars of the type as described and figured by Cabrera, is also a character of sporadic occurrence in P. darwini or any phyllotine rodent, particularly the young. The measurement 13 mm. for zygomatic breadth of the type skull is right for a Juvenal. Braincase width of 13.2 for the same skull, however, would be anomalous in Phyllotis. This last measurement must be either a typographical error or is faulty because of the damage to the occipital bone mentioned by Cabrera. Of four topo- types at hand, one is among the palest, the other among the darkest of all specimens here assigned to rupestris. Pearson (1958, p. 419) includes oreigenus, together with wolffhuegeli, in the synonymy of vaccarum. The type specimen of Phyllotis wolffhuegeli Mann, from Boca- toma, Santiago, altitude 1800 meters, on the Chilean side of the Andes, is listed as an adult female in the introduction to the original description but identified as an immature male in the table of measurements in plate 9 of the original work. Accompanying figures of the skull of the type show the third molar recently erupted, a character of the suckling young in Phyllotis. All other characters said to be distinctive of wolffhuegeli are those distinguishing juvenal from adult rupestris. Phyllotis osgoodi and P. arenarius chilenxis Mann, both from Parinacota, Tarapaca, Chile, are juvenal and adult, respectively. > ^JS >J N: co os os oo co - -. . w. ~v . -. -. •^"^ m^flti *O J0, •«? id ^* " 00 T-H 00 oo 10 10 10 -^^t id co o o id id •* .2 » l> 5 ^c »-i o o 05 -0 »«-Jeo»o co fir- o t- o ~5~~ OC OC CD C£> oo §00^^ Sc^oo oo 53 a E 5, % O IO OS JQ ^ os' oo CD OQ Q$ S5^i oo'os'oc T-^ 00 00 CO s W J oo t> ^ "^ ^ s.^^g oc c- CO i u coco j! ^J *^> rvi t** IT T°l "l ^ •? 0 W o$ co co is" "^ oo -^ t^- CO 00 CO CO ^c^oo oo 00 a oio'^"'^' i CD Jo'J5't-"'oc S'S'o ooS'co os us" CO OJ ^O CO ' CO CO CO CO -r 16 rj< o COCO?^ 1 o3 ^ 3 IO OO ''l1^* CD c^^^"^oo CO CO OJ CO CO CO c- id" CO CO ': r- a 00 CD ?J?- S>^iOO i— i CD CO a < OO CO OS OO CO 00 -«S< . • oo oo co t" ^(D IO ^j CO -H CO iH •g ••* 1 O o — os *^ 7 -L ^ 222|§eie^l2lll §«^^£« -i-(-!il)a:c'>— i '3 _e ^5 2 C s — a ttT:Mfj ol ^•o _ro _ro ,30 ^20^30 312 •..- £ ca o o IS" > be stt -* ; 1- q iooo"co id id id id lo'ididkdkdid >o "5 »dioioidid »dididw kO iO I" t~ • t- 10 'd CJ '^ r* »• «? oo § §£ 5 *r id J2 id ^ ^ H i lO || co" *~! c~" t^ co o ^ «-• i< Tr SM*"* kd to " ^ "i" id "5 to id to' N"° |2 - S of co "'.^ co"-^ _ t-" ko"« > co t-^t^ ^' of o - •^ id i" kd ^'"(•^fto'id co "^ 10 id id id id kOidkOw i kO ^ : *j -H' 52. 00 t- IO OJ co ~ oj ^ ^~^ . — . ^j K oi °. o = > ** ** " - :«: • « C3 M « o S- S ^^ 7 * 8 c ^C 71] . O • CD t^ ^ J "— ' — O ^SOS^J j — - OJ OJ ^ oj 23 j^ S o — ^ K -I ?> rr ' QO —' : jg gj j M * r- io"qi: t, 0- ^a i vi co 01 '•^ oj co g 8 ?, 8 ««"SS3S ^ S l^^^SS g«« > 5r "* t- co 10 «' i- ^ 2 ^~. rvi co ^ 1 T.J CO | K 8 s i i s ^ s «s §}«» 1 s ?O ^3 CO ^5 O^ t'1* *-j* co irt rts ^.7 u » cb"to id"t-"co" 1 0101 Ol^^i > 3 51 * ; £ be ? '53 t- co" So gg t. S T T 5? •«* ~ co 10 «, jT CO_ t « w «: i<5 * t^ OJ oococo^ to°°kO_j ^ < ««T - • - CO kOCO-Tl" ^Ji CQ 1^5 ^J , 5 kO | Jg 1 B £ co 10 ^ Qot^ ^ jo 1 § K-l id" S§ *1 t~ 2 1 ? « £~ ' T3 t- 3 C OJ coio ?°coio J?coji * ^ t- x os co" to - ~ IO" J HH OJ t^» OO IO t*- t^- OS CO ^. ^j ^x t— CO I { OJ O-J O-J O J £-.J OJ O-J rs.j ^v j ^fi ^ i O J i ^ 2?5tO*X to"io"iO*c ' *J oj oj oj ri oj c oo - —i cf- * ^, 9 ~- «•? o ^\ i ^ ^ 1 J o-i co »-< ^ - - OJ ^i 3 ^ s - ^ ^ kO ^-T oo" co i* o-i — ^ oj o — cofi "T 5 CO ° ""- OI I^Os 0 — ' 0 to ~ H Of o i* t~ co o^^'oTojoT ^ w *"••' ?§tx 1.1id. O I-M •— * o o co *"^'"^O*-H^ »— • ^ i^ *— 1 >— ' ^-( I-H r-l ^^ r— 1 -^OJ-H ojos — - ^ ~ iO a 0 t~ 10 OS J 3 8 i vii £ s. — ^H t- r - -•» — iO T3 — 1 0 0 1 2 t- — 2 1 «« o - - §. S 2 - J2- I o r~ 12 !2 "^ " 1 £ K °» 2 o o o 222^^-2-2 ^ o *? -os^Tfl oft. id t- 1^ — 01 O to O " cS o p 1*"? 1— <>5 L 13 «S r ?i C_ C_ H|!i*"!2"! 2-S2aSSII *** l ^ 11 2"3 "~" ™ * fc^ „, o • — • -* 2 § ~ J2 2. .=. g 3^ S C 7. "3 C. UO53OK -^C-^ • ^*^ *-• **^ 9~i *T •"• lO *^ *^ *^" w- ^^ """^ P" K — . f 1 ~ ^ * ~ 'G —* ?3 —. & c^W^-^rl^ "* 3 ^] ^ :^t5**«2 £*" e ^^-r-^.z^f- £ffiO c 313 • •* * }00 us ol -t us US ic US 1C ic ic US US CD — r. t- OJ CNJ CD OJ 1C US US US "3 -3j2 05 *J g-o o S •^f OS _ ^^-^ ^ j ^ r-Tco" ot- U3 uj 1C -* t- -* OS o • °?» ! oo'o t- o H ^i c^ I CO •~ ^ S3 g K lt co J (M (M — ' O rH O rH ^ os ss ^ CD rffoo" > •o O rH. ^) x-x os OO rH rH CD rH rH rH OS rH o oo oo" rH rH >M •g ^ >> g^S.^CSOrtgijjWc «o8 1 IM Cd Ctf ! §S 03 B'0 "°PHPH PH (§J j^s y S •-C-3rtC^CKcK5 (X, C C C C o .S« KH rt os os ,c PH W Ol CQ 02 t 5 ss- *- i—i ri c < 3 (^ P H PnOO 1 OS oS 02 w M W i O 314 CD os oo T -H o CD .-< os •^< CD os oo I1 oo r- u o oo" eg to oo eg eg coto "fl -G to « 2 to •<*' to to ^td id ^ i «* bC i "i oo oo to * £^ * * qojos t- eg ^f co 4; 2 to OS N-5 i-* eg co eg" ^jT o' S^S'toS ^^-^or^ j-^- to' ^. •~1CO oo ^ ^ 22J2 "^ "9 j* ^ ^ ^"* 1" to -t -^ id -t 1- '" 1 f— 1 •S -^ 2 oo ^ «o os oo ^,00 ^ C w to ^2 co eg co o^co^P o u rt ^ o;^: °P c^ e-i e-i o eg co os' '« _- g ^ esjt- to co "x co ^ "^ "5 co" co co CD" co oo -< as co i— c os oo >-f 5^ oo' S co t--' as' t-' ad o i oo os co &M 3 e-i co co eg ^^ ^M >^^ eg vli- ^^^ eg o-i e~J eg e*< eg 01 O w ac co" eoco"co" eg' os co o t~ o os co TJ"OO TJI tr r^? os't- ei rj a n oo' o ^H — o os eg co co co co eg os' t-' oo o t-' oo' oo' t-' os' os' oo' eg eg eg co eg eg eg eg eg e i e j C 1 CD s. a TJ< to os oo ^* 1-1 t^ k. *° 1 19 *Ota*t**nir CO >f— CO eg Tf eg eg Sj e-j ei i; ' eg •S « eg CO tO ^ — CO ~« \f\ -« rvi \r \/-\ i s w e\j eg eg^to e^ 7,3 ^ ^j eg e^i e-i egeg^ 1 10* "^ 1 1 I eg eg ' ?g e^j eg eg eg e^j eg e~i ^j T-J eg ej of to' e-j eg 1 4J co oo SJ JS co co a> b£ I i oo t- eg ^ Tf eg eg LE 23. —Measurements il Hind foot CD" CD i— ( CO 00 t- 00 eg eg eg o -r eg op co i CO <^i eg co eg o" 1 1 eg eg eg S^iof. ^j" co 2 CO ^ CO T 1 — o> 1 - 1 oo S" T U ti7 u?- t~ ej I oo « ^t o i^^ltO to" to no e^e^e-j^ i ei Vi ri i1' t— CD" e-j i co -r '""Iti ei ?j S » « *t o *-i ei —i ^ co o o — — f-H -^ 00 gj ~ 2 H 0 t-» CD CD* t--' ei eg cpi ^-^. . / oo « fot^'o"^* 7J2 i T ^" ^ CD o "t3 , — , os t^, ~f ^~ t— O ^•vs oo , — v ~f CO -» ^-^ ^ — ' "^ /^ —I vV co i~- CO *— ~ ^-^ ,^— ^ cc ^ i j , _ to Vf > be « gT3 p bfl t. •§ eo eo 2 (M i— i rt *rt IO (M 1/5 of C> 1 o Chumbicha San Antonio < C$ C 3 1 •4- S J d (5 Laguna Blanca27 03 g II Otro Cerro28 Sombreros Punta Vaca25 Punta Vaca3t Puente Inca Chos Malal Collon Cura3 316 "8 C fc^ o 2 >> - 2^sorsJfc5fc5 «f?* - kT oj bfl 0) C OO Cy C Q OW W ^ w W V Q, •* *•* fl> M •— ta W 9 •« u CbflajMCbfibCbcbflbc OObatj-agcicSs o c >,£ o c c c c c0--c.c c^2 s v B -5 "8 « — 03^— 00000^!«!':0^§>^Oi-§^ |- 3 11 .g d in tt -r- £ fefell s: M ! fli^-Sj^icf.*^- r*. K^KT-T- ™ b£s^"X3 u 5 .? £ S •* * .5 h c be M .-« « c c fe ^^ -2 2 H J5 cxo.rzi D.Q.O :* o.~rri 0.0^73 P-p'f^Ji H H H H E- ^ C H H H H H H H H H w 318 FIELDIANA: ZOOLOGY, VOLUME 46 of Phyllotis darwini rupestris. Pearson, however, recognized chi- lensis (with the prior name osgoodi a synonym) as a valid sub- species of darwini "intermediate in size of most parts and in color between posticalis and rupestris." Inasmuch as most populations are also geographically intermediate and others wholly encompassed by or cutting completely across the range of rupestris, Pearson's chilensis may be better described as a heterogeneous combination of intergrading populations between posticalis and northern rupestris and between eastern rupestris and western rupestris and isolated peak populations wholly derived from rupestris stock. Moquegua and Tacna chilensis of Pearson average slightly darker than most rupestris to the east and west but are hardly if at all distinguishable from northern Chilean specimens (Ojos de San Pedro; San Pedro; Tatio Geysers) regarded by Pearson (1958, p. 417) as typical rupestris. Pearson (1958, p. 416) himself experienced difficulty in attaching "the name chilensis to a subspecies that ranges primarily in southern Peru [and with a] type specimen [which] is not typical of the sub- species." Actually, the type of chilensis is indistinguishable from typical representatives of rupestris. If a southern Peruvian race of darwini were indeed recognizable, the name chilensis could not be used for it. Forty specimens from Paposa, Antofagasta, and two from Vallenar, Atacama, included with darwini by Pearson (1958, p. 421), were described by him as "pale-colored but relatively large-eared," and considered "to be intergrades with rupestris and vaccarum." I have not seen the Paposa and Vallenar specimens but their geo- graphic distribution, color and size, as described, point to the small, pale rupestris rather than the larger, warmly colored darwini. Speci- mens of similar color and size from localities in the same area (Domeyko, Atacama; "east of Taltal," Antofagasta) are identified as rupestris by Pearson and in this paper. No published measure- ments are available for the Vallenar specimens but for the Paposa series Pearson (1958, Appendix I) includes head and body 105 (90-125, 26 specimens), ears 26 (19-30, 28 specimens). These com- pare with the following of darwini from "Santiago, Valparaiso and Aconcagua," according to the same authority: head and body 120 (107-140, 29 specimens), ear 25.9 (21-32). Relative to head and body, the ears of the Paposa series average appreciably larger than those of central Chilean darwini. On the other hand, they can be more nearly matched in size with ears of a number of popula- tions of rupestris (including vaccarum) scattered throughout the HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 319 range of the subspecies (cf. Table 23). Actually, ear measurements of fresh material and visual comparisons of ears in dry skins show considerable variation and give little or no basis for determination of the systematic position of any one population of rupestris or of darwini. The foregoing geographical and zoological considerations indicate the assignment of the Paposa and Vallenar series to rupestris, not darwini. It is noted that Pearson excludes Antofagasta and Atacama from the range of P. d. darwini. Remarks. — At several points along the periphery of its range in southwestern Peru, Phyllotis darwini rupestris contacts but, pre- sumably, does not interbreed with, P. d. ntagister. On the north- eastern border of its range, in Puno, Peru, and, probably, in La Paz, northern Bolivia, rupestris lives side by side with representatives of the strikingly similar Phyllotis osilae. The ranges of rupestris and P. osilae continue southward through the Andes of Bolivia and northern Argentina, their common borders overlapping or inter- digitating at several points. In some parts of the same sections of Bolivia and Argentina P. darwini caprinus appears. Whether or not caprinus actually contacts rupestris is moot. Observations made by Pearson indicate that the two may occur in the same general area but are ecologically segregated. For further discussions on the relationship of rupestris to caprinus and other forms, see the corresponding species and subspecies accounts. Measurements. — See Table 23. Specimens examined. — 473. ARGENTINA. Cataniarca: Chum- bicha, 2 (CNHM); Laguna Blanca, 4 (CNHM, 2; MACN, 2). Jujuy: La Laguna, Sierra de Zenta, 2 (CNHM, 1; MACN, 1); Tilcara, 1 (AMNH); Tilcara, \$ mi. E, 7 (MVZ); Tres Cruces, 2 (CNHMi. La Rioja: Puesto Viejo, 1 (MACN). Los Andes: San Antonio de los Cobres, 4 (CNHM, 1; MACN, 3). Mendoza: Puente del Inca, 4 (AMNH); Punta Vuca, 3 (CNHM). Neuquen: Chos Mulul, 2 (CNHM); Collon Cura, 1 (CNHM). Salta: G. M. Sola, 2 (C> San Juan: Angaco Sud, 1 (AMNH); Los Sombreros, 1 (CNHMi. BOLIVIA. Cochabamba: Tin Tin, 1 (CNHM). La Pa:: Espe- ranza, 3 (CNHM); Estacion Perez, 11 (CNHMi; La Pa/., 20 mi. ! 4 (MVZ). Oruro: Oruro, 40 mi. S, 2 (MVZ); Sajama (Mt.) (CNHM). Potosl: Lipez, 3 (CNHM); Potosi, 20 mi. S, 6 (MVZn Uyuni, 10 (MVZ); Villazon, 6 (MVZ). Tarija: Camataqui, J (MVZ). CHILE. Antofagasta: Muelle de Piedra, E of Taltal, 1 (AMNH) ; Ojos de San Pedro, 4 (MVZ) ; Pocos, 1 (M VZ) ; San Pedro, 16 (CNHM) ; Tatio Geysers, 1 (MVZ) ; Toconce, 3 (MVZ). Atacama: 320 FIELDIANA: ZOOLOGY, VOLUME 46 Domeyko, 2 (CNHM). Coquimbo: Banos del Toro, 2 (CNHM); Pai- guano, 43 (CNHM, 41; USNM, 2). Santiago: Los Valdes, 5 (USNM, including type of wolffhuegeli Mann) ; Rio Blanco, 1 (USNM). Talca: Arroyo del Valle, 1 (CNHM); Rio Maule, 4 (CNHM). Tarapacd: Caritaya, 6 (MVZ); Parinacota, 1 (USNM, type of chilensis Mann); Putre, 8 (CNHM); Timar, 1 (USNM). PERU.— Arequipa: Are- quipa, 2 (CNHM); Arequipa, 7 km. E, 7 (MVZ); Arequipa, 12 km. SSW, 4 (MVZ); Balneario de Jesus, 13 (CNHM); Ongoro, Valle de Mages, 2 (CNHM); Yura, 29 (CNHM); Cailloma, 26 (CNHM); Huaylarco, 5 (MVZ); Imata, 3 (MVZ); Salinas, 13 (CNHM, 11; MVZ, 2); San Ignacio, 5 (MCZ); Sibayo, 7 (MCZ); Sumbay, 14 (CNHM). Moquegua: Rio Torata, 15 km. NE Torata, 6 (MVZ); Torata, 10 km. NE, 2 (MVZ); Tala, 2 (MVZ); Lago Viscacha, 1 (MVZ) ; Lago Suche, 3 (MVZ). Tacna: Tarata, 2 km. N, 4 (MVZ) ; Tarata, 3 km. N, 2 (MVZ); Tarata, 4 km. N, 5 (MVZ); Tarata, 4.5 km. N, 2 (MVZ) ; Tarata, 5 km. N, 2 (MVZ) ; Tarata, 20 km. NE, 5 (MVZ) ; Tarata, 25 km. NE, 1 (MVZ) ; Tarata, 8 mi. NE, 2 (MVZ). Puno: Arapa, 1 (MVZ); Asilla, 6 (CNHM, 5; MVZ, 1); Caccachara, 12 (MCZ, 11; MVZ, 1); Collacachi, 11 (CNHM); Chucuito, 8 (CNHM); Huacullani, 4 (CNHM); Have, 3 (MVZ); Juli, 11 (MCZ, 6; MVZ, 5); Juliaca, 4 (CNHM, 3; MVZ, 1); Mazocruz, 30 mi. S, 1 (MVZ); Occomani, 2 (CNHM); Pairumani, 3 (MVZ); Pampa de Ancomarca, 3 (MVZ); Pisacoma, 4 (CNHM); Pomata, 2 (MVZ); Puno, 8 (CNHM, 1; MVZ, 7) ; Puno, 82 km. W, 2 (MVZ) ; Rio Santa Rosa, 3 (MVZ) ; San Antonio de Esquilache, 2 (CNHM) ; Santa Rosa de Juli, 13 (MVZ); Santa Rosa, 2 mi. W, 2 (MVZ); Sorapa, 2 (AMNH); Tincopalca, 1 (MVZ); Umayo, 4 (MVZ); Yunguyo, 3 (CNHM). Phyllotis darwini darwini Waterhouse Mus (Phyllotis) Darwinii Waterhouse, 1837, Proc. Zool. Soc. London, 1837: 28. A/us Darwinii, Waterhouse, 1839, Zool. Voy. "Beagle," Mammalia, p. 64, pi. 23 (animal), pi. 34, fig. 17 (molars) — Coquimbo (Coquimbo). Gay, 1847, Hist. Chile, Zool., 1: 117. H[esperomys] darwini, Thomas, 1884, Proc. Zool. Soc. London, 1884: 449 — type of subgenus Phyllotis. Phyllotis darwinii, Thomas, 1902, Ann. Mag. Nat. Hist., (7), 9: 131— Val- paraiso (Valparaiso). Phyllotis Darwini, Cabrera, 1905, Rev. Chilena Hist. Nat., 9: 16— part, Val- paraiso (Quillota). HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 321 Phyllotis darunni, Wolffsohn and Porter, 1908, Rev. Chilena Hist. Nat.. 12: 79 — Valparaiso (Valparaiso); Must capito Philippi, a synonym. Thomas, 1927, Ann. Mag. Nat. Hist., (9), 19: 556 Santiago (Puente Alto, south- east of Santiago). Phyllotis darwini darwini, Osgood, 1943, Field Mus. Nat. Hist., Zool. Ser., 30: 200 — taxonomy; characters; distribution; Aconcagua (La Ligua; Longo- toma; Los Agostinos, Palomar; Papudo); Coquimbo (Romero); Santiago (San Cristobal); Valparaiso (Buen Retiro, Las Hijuelas; Las Rojas, Quill- ota; Olmue; Palmilla, La Cruz). Pearson, 1958, Univ. California Publ. Zool., 56: 421 — taxonomic revision; distribution (for locality records see p. 250). Mus melanonotus Philippi in Philippi and Landbeck, 1858, Arch. Naturg., (24), 1: 78— type locality, near Angostura, Santiago. Philippi, 1900, Anal. Mus. Nac. Chile, Zool., Entr., 14a: 43, pi. 18, fig. 1 (animal) — type speci- men lost. Mus dichrous Philippi, 1900, Anal. Mus. Nac. Chile, Zool., Entr., 14a: 14, pi. 3, fig. 2 (animal) — type locality, "Provincia Santiago, prope Peine." Wolffsohn, 1910, Bol. Mus. Nac. Chile, 2: 88— dichrous = Phyllotix d oo o CD eg i— i oo eg I-H i— i | ,-( ,-H .-H eg' o coco i co co co coco eg. • I I I I I £5?^ o eo^ oq oi 03 C •a .M o o .2 -g c »- 4J 4J ^ .& ° T3 c oocooioooit-ocDocDcsiciiciot^coeg cgegegegegegcocgcoegcgcgegcgcgcgeg - ^ *J'" H >> g ^3 £ CO «2 "S weg eg^-cx. 13 a g g S £ S « g » g 2 OrHO-^Oiegoot- 3 -^rnSoifciflti^^SS Td c eg -c S3 aJ a eg oco eg cg*o oo"oo *"! ,3 o •|ii|l|f.i|Il f?!fill|ps CL S -rS s? o «, ^ *^ pfl f 5? a s >> S - «» a "^ s PS. S1^ « o 0 0 0 0 0 00 oo 1 a a a a a a a a a a >>>>>>>>>>>>>>>>>>>> 324 HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 325 Santiago: San Cristobal, Santiago, 2 (CNHM). Valparaiso: Lu Calera, 2 (CNHM, 1; MCZ, 1); Las Rojas, 2 (CNHM); Olmue, 4 (CNHM); Palmilla, La Cruz, 2 (CNHM); Quilpue", 3 (AMNH, 2; MCZ, 1). Phyllotis darwini fulvescens Osgood Phyllotis darwini fulvescens Osgood, 1943, Field Mus. Nat. Hist., Zool. Ser., 30: 204. Pearson, 1958, Univ. California Publ. Zool., 56: 422 review. Type.- Adult female, Chicago Natural History Museum no. 50550; collected 8 November, 1939, by Wilfred H. Osgood. Type locality. — Piedra de Aiguilas [Aguilas?) near summit of Sierra Nahuelbuta, west of Angol, Malleco, Chile; altitude, 1250 meters. Distribution (figs. 70, 75). — Known only from the Araucaria forest in the Sierra de Nahuelbuta, Malleco, Chile. Characters. — Most warmly colored race of Phyllotis darwini; size approximately as in the nominate subspecies or perhaps smaller; back warm brown, underparts heavily washed ochraceous, with dark basal portion of hairs showing through; ochraceous lateral line broad and well defined; tail thinly haired, short (85 per cent [type], 92 per cent of head and body length) and bicolor except brown tip; cranial characters generally as in darwini; interorbital region nar- row, the sides pinched upward; proximal ends of nasals pointed and terminating slightly behind fron to-maxillary suture; the median nasal sulcus unusually deep; molar row 5.9 (2 specimens). Remarks. Phyllotis darwini fulvescens and P. osilae phaeus are the darkest members of the Phyllotis darwini group. Each form inhabits a region of high rainfall. The subspecies is represented only by the type and a very similar specimen from the same general region collected 6 January, 1930, by H. E. Anthony and G. Ottley. The subspecies is most nearly approached in color by an extremely dark series of P. d. xanthopygus from Laguna Lazo, Magallanes, Chile. The latter, however, is much larger, with hairy tail and lateral line subdued. The geographically nearer populations of darwini from Quirihue and Pilen Alto, are distinctly paler with smaller molars. Measurements.- Bee Table 25. Specimens examined. 2. CHILE. Malleco: Sierra Nahuelbuta, 1 (CNHM, the type); Araucaria Forest, Nahuelbuta, 1 (AMNH). £2 ~" t; Oi CS ~~ 10 us _ r; J^. O) »-i i— I v; r^. 'c S O "o 4-1 eo 1 x 5 J -S -5 326 HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 3-J7 Phyllotis darwini xanthopygus Waterhouse Mus (Phyllotis) tanthopygus Waterhouse, 1837, Proc. Zool. Soc. London, 1837: 28. Thomas, 1919, Ann. Mag. Nat. Hist., (9), 3: 202 type a true Phyllotis. Mus xanthopygus, Waterhouse, 1839, Zool. Voy. "Beagle," Mammalia, pt. 2, p. 63, pi. 22 (animal), pi. 34, fig. 16 (molars) ARGENTINA: .S«n/« Cruz (Port Desire; Santa Cruz). Hesperomys xanthopygus, Burmeister, 1879, Descr. Phys. Rep. Argentina, 3: 225- ARGENTINA: Santa Cruz (Rio Santa Cruz). Burmeister, 1891, Anal. Mus. Nae. Buenos Aires, 3: 314- ARGENTINA: Santa Cruz (Cap- rek-aik). H[esperomys] xanthopygus, Thomas, 1884, Proc. Zool. Soc. London, 1884: 449 subgenus Phyllotis. [Hesperomys (Phyllotis)} xanthopygos [sic], Trouessart, 1880, Bull. Soc. Etudes Scient., Angers, 1880, (1): 135- classification. Hesperomys (Phyllotis) jcanthopygus, Thomas, 1891, in Milne-Edwards, Mis- sion du Cap, 6: Mamm, p. A. 29, pi. 6, fig. 2 (animal)- ARGENTINA: Santa Cruz (Santa Cruz). Euneomys (Auliscomys) xanthopygus, Thomas, 1916, Ann. Mag. Nat. Hist., (8), 17: 143— classification. [Phyllotis] xanthopygus, Trouessart, 1897, Cat. Mamm., p. 534 classification. Phyllotis xanthopygus, Allen, 1905, Princeton Univ. Exped. Patagonia, 3, (1), Mamm., p. 58, pi. 13, fig. 1 (skull), pi. 14, figs. 2, 3 (molars) ARGEN- TINA: Santa Cruz (Rio Chico; Swan Lake; Basaltic Canyons). Thomas, 1919, Ann. Mag. Nat. Hist., (9), 3: 200- ARGENTINA: Rio Negro (Pil- caniyeu). Thomas, 1919, op. cit., (9), 3: 202- true I'hyllolis, not Kuneomys as classified in 1916. Thomas, 1927, op. cit., (9), 19: 549 lectotype, British Museum (Natural History) no. 55.12.24.185, from Santa Cruz, Patagonia, lectoparatype from Port Desire. Yepes, 1935, Anal. Mus. argentine Cienc. Nat., 38: 335- ARGENTINA: Chiibut (Rio Senguer). Phyllotis darwini xanthopygus, Osgood, 1943, Field Mus. Nat. Hist., Zool. Ser., 30: 208— part, ARGENTINA: Chubui (Tecka), Rio Negro (Huanuiuan; Pilcaneu); CHILE: Magallanes (Laguna Lazo, near Lago Sarmiento). Pearson, 1958, Univ. California Publ. Zool., 56: 420 part, ARGENTINA: Chubut (Pico Salamanca; Tecka), Rio Negro (Huanuiuan; Pilcaneu), Santa Cruz (Arroyo Aike; between Swan Lake and Arroyo Aike; Port Desire; Rfo Chico; Rfo Coy; Santa Cruz); CHILE: Magallanes (Laguna Lazo). Type.--Adu\t, skin and skull, British Museum (Natural History) no. 55.12.24.185; collected April, 1934, by Charles Darwin. Type locality. -Santa Cruz, coast of Santa Cruz, southern Ar- gentina. Distribution (figs. 70, 75). Patagonian region from western Rfo Negro and coastal Comodoro Rivadavia (Pico Salamanca), Argentina, south into Magallanes, Chile; altitudinal range, sea level to approxi- mately 1400 meters above. >J5 :* :&k lOooooust-t-^iooq £ £ ca-w P O.OJ ** (/- C 4J eS C _ 0> 3| C t^ §1 .. 0) S£ | ttj 4 ^-4 CO' co co oq 10 co rH r4 oo - a ^ 00 *H (M Jjj C S - 03 ,C O»T ?? S ft. I •5 a 2 §« OS £ "N N O T3 3 3 S . 35 C C 2 ^ •>: o o O ^d 4^ o o be C c PH OH n3 00 GO o d H g O 328 HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS U29 Characters. — Upper parts more warmly colored, underparls more heavily overlaid with buff or ochraceous and lateral line less defined than in rupestris; slightly paler throughout than fulvescens, tail more hairy; size as in neighboring populations of rupestris (raccarum)-, pectoral streak usually present and either well defined or confused with the buffy or ochraceous wash of underparts; well-haired tail bicolor except sometimes at tip, length to head and body combined 97 per cent (79-111 per cent, 37 specimens); auditory bullae mod- erately inflated, their antero-posterior length, less tubes, subequal to alveolar length of molar row; bullar tubes short; width of meso- pterygoid fossa, measured at base of hamular processes, subequal to or less than width of parapterygoid fossa at same plane; pos- terior border of palate rounded, square or angular and without median spine or process; postero-lateral palatal pits anteriad to or in line with posterior border of palate; supraorbital region elongate, the edges square or pinched upward; fron to-parietal suture narrowly rounded or V-shaped; proximal ends of nasals terminating from slightly in front of to slightly behind fronto-maxillary sutures; alve- olar length of molar row, 5.6 (5.1-6.0, 41 specimens), or 16 18 per cent of skull length. Variation. — The transition from pale rupestris in the north to the more saturate xanthopygus in the south is gradual and appears to be insensible between populations separated by relatively short dis- tances. Specimens from Rio Negro (Pilcaneu and Huanuluan) are somewhat more warmly colored on upper parts and underparts than representatives of rupestris from Neuque"n (Collon Cura; Chos Malal). A specimen of xanthopygus from Tecka, Chubut, agrees with the Pilcaneu series. The specimen from the high pampas and scrub country at the base of the Andes in Santa Cruz, first described by J. A. Allen (1905, supra cit.), are more brightly colored than the preceding. The type specimen from Santa Cruz proper on the coast is, according to Thomas (1919b, p. 200), even more buffy on the underparts. Two adults and four juvenals from Lago Lazo, Magal- lanes, Chile, are darkest and appear to grade into the extremely dark fulvescens farther north in Chile. In general, specimens of xanthopygus from the Argentine side of the Andes are notably paler than representatives of darwini, fulrescens and xanthopygus of the same latitudes on the Chilean side. Taxonomy. — P. d. xanthopygus is the weakest subspecies of those recognized here. The name represents the southernmost populations of the same comparatively dark ecotypes which range from southern 330 FIELDIANA: ZOOLOGY, VOLUME 46 Peru (chilensis of Pearson) through the Andes of Bolivia, Argentina (ricardulus and vaccarum of Pearson) and Patagonia (xanthopygus) to the Straits of Magellan. Specimens from Neuque"n (Collon Cura; Las Lajas; Quilquihue) identified as xanthopygus by Pearson, who, evidently, followed Thomas and Osgood, are here referred to rupestris. A specimen at hand from Collon Cura, southern Neuque"n, is paler than xanthopygus from the south but indistinguishable from two individuals of rupestris from Chos Malal, Neuque'n, and other localities to the north. Unless evidence indicating the contrary is produced, I prefer to regard the Phyllotis from the intermediate and ecologically similar localities, Las Lajas and Quilquihue, as likewise representative of rupestris (or vaccarum of authors) . Measurements. — See Table 26. Specimens examined. — 57. ARGENTINA. — Chubut: Tecka, 1 (CNHM). Rio Negro: Huanuluan, 30 (CNHM, 1; MCZ, 29); Pil- caneu, 3 (CNHM). Santa Cruz: Arroyo Aike, 4 (AMNH); Lago Viedma, 1 (MACN); Rio Chico, 6 (AMNH, 2; USNM, 4); Rio Coy, 5 (USNM); Santa Cruz, 30 miles south along coast, 1 (AMNH). CHILE. — Magallanes: LagunaLazo, Ultima Esperanza, 6 (CNHM). Phyllotis darwini caprinus Pearson Phyllotis wolffsohni, Thomas (not Thomas, 1902), 1913, Ann. Mag. Nat. Hist., (8), 11: 139— ARGENTINA: Jujuy (Maimara). Yepes, 1933, Rev. Chi- lena Hist. Nat., 37: 47— ARGENTINA: Jujuy (Maimara, 2230 meters, 2500 meters). Yepes, 1935, Anal. Mus. argentine Cienc. Nat., 38: 334— ARGENTINA: Jujuy (Maimara). Yepes, 1935, Rev. Inst. Bacteriol. Buenos Aires, 7: 223— ARGENTINA: Jujuy. Phyllotis ricardulus, Thomas (part, not Thomas, 1902), 1919, Ann. Mag. Nat. Hist., (9), 3: 493— ARGENTINA: Jujuy (Maimara). Thomas, 1921, Ann. Mag. Nat. Hist., (9), 8: 611— ARGENTINA: Jujuy (Alfarcito). Thomas, 1926, Ann. Mag. Nat. Hist., (9), 18: 194— part, BOLIVIA: Potosi (Yuruma). Phyllotis caprinus Pearson, 1958, Univ. California Publ. Zool., 56: 406 (fig., molars), 409 (key characters), 434, 435— ARGENTINA: Jujuy (Alfar- cito; Humahuaca; Maimara; Tilcara; Sierra de Zenta); BOLIVIA: Chuqui- saca (Camargo), Potosi (Yuruma), Tarija (Camataqui, 20 miles SSE). Type. — Adult female, skin and skull, Museum of Vertebrate Zoology, University of California no. 120210; collected 1 October, 1955, by Oliver P. Pearson. Type locality. — Tilcara, Jujuy, Argentina; altitude, 2500 meters above sea level. HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS :m Distribution (figs. 70, 74, 77).— Andes of south central Bolivia from western Chuquisaca, western Tarija and southeastern Potosi, southward into central Jujuy, Argentina; altitudinal range from approximately 2000 to 4650 meters above sea level. Characters. — Size intermediate between wolffsohni and rupeslris of Bolivia and northern Argentina; color more nearly as in irolff- sohni; underparts with more buff or ochraceous than rupestris; a well-defined pectoral streak present in majority of specimens exam- ined; tail consistently longer than head and body combined, average 122 per cent (101-139 per cent, 35 specimens), bicolor except at tip, short-haired as in wolffsohni to long-haired as in rupestris; auditory bullae comparatively little inflated, antero-posterior length, less tubes, 4.8 (4.6-5.4, 37 specimens) and approximately equal to or, usually, shorter than alveolar length of molar row (figs. 83, 84); bullar tubes short to moderately long; mesopterygoid fossa, measured at base of hamular processes, from narrower to as wide as parapterygoid fossa measured at same plane, sides parallel or slightly divergent; poste- rior border of palate notched or with a well-defined unnotched median process; postero-lateral palatal fossa poorly defined, pit small to moderately large and located on a line with or anteriad to pos- terior border of palate; interorbital region usually flattened as in wolffsohni, the edges nearly always square, the posterior halves slightly or not at all beaded, usually divergent posteriorly and sometimes forming narrow ledges; proximal ends of nasals truncate or pointed and terminating on a line with or slightly fore or aft pre- maxillary sutures; fron to-parietal sutures arced or forming a right angle at midline; molar rows from slightly divergent to slightly con- vergent posteriorly; enamel pattern of ma from modified sigmoid as in wolffsohni to 8-shaped as in rupestris (fig. 85, a c); m1 sigmoid, 8-shaped or with second primary fold forming an enamel island which may coalesce with major fold; m2 sigmoid to 8-shaped; sec- ond primary fold present or absent in moderately worn m, 2. Variation. — Color of upper parts and sides is fairly uniform throughout; underparts are lightly washed with ochraceous in the Tilcara series, most heavily in specimens from Sierra de ZenUi and Maimara; a well-developed pectoral streak is present in all Tilcara specimens examined and present in about one-half the specimens of other localities. Tails in the Camargo series of three speci- mens are extremely hairy, that of the single Tarija specimen the le;ist hairy while tails in the Camataqui series (21 specimens) show a wide variation but generally are as hairy as those of P. danrini n//>r.s-/n'.s-. - o eS eij ^•° CO — £ ° ol si o ° 3 " ' S 31 .. o m fe a> -to N" •r s 3 •S'SS S £ 0) O 00 « fi 332 tea _z q t o •a _ a O * I ~ s 333 334 FIELDIANA: ZOOLOGY, VOLUME 46 FIG. 85. — Upper molars of a-c, Phyllotis darwini caprinus; d-f, P. d. wolffsohni. The fronto-parietal suture is much more sharply V-shaped in the Camataqui than the Tilcara series (fig. 83, a-c). In the lone Ta- rija specimen (MVZ 120201) the suture is broadly rounded as in most wolffsohni, the supraorbital region is also more divergent- sided, the edges beaded and with narrow ledges posteriorly as in wolff- sohni. The enamel pattern of the second and third molars of caprinus is the least stable of that in any of the races of P. darwini. HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 335 Bacula (fig. 7) of the Tilcara series of caprinus are like those of P. darwini rupestris except that of one subadult which resembles that of adult wolffsohni. Bacula of the Bolivian specimens from Camargo, Camataqui and Tarija also agree with those of wolffsohni. Taxonomy. — Phyllotis caprinus Pearson was originally described as a full species on the basis of characters which appear to be vari- able and on geographic considerations which are subject to conflicting interpretations. Pearson (1958, pp. 434, 436) alludes to sympatry between caprinus and darwini (i.e., P. darwini rupestris of this paper). The actual circumstances of "contact" are described (op. cit., pp. 400 401) as follows: "At Tilcara in northern Argentina caprinus was cap- tured along brushy hedgerows and stone walls at the edge of the town and darwini was taken a few hundred yards east of the town on mountain slopes where tumbled rocks, cardon cactus, Opuntia, bro- meliads, and Ephedra form a more open habitat. Six miles east of Tilcara at Alfarcito where there is a mixture of brushy and open habitat, Budin captured both darwini and caprinus [see my com- ments below]. At another location, 25 miles south-southeast of Camataqui, southern Bolivia, the ichu grass of the altiplano meets thornbush, saguaro and other cacti at 11,500 feet. In this mixed habitat I captured many darwini but no caprinus. Five miles away at 8500 feet among scattered bushes and Opuntia I caught large numbers of caprinus and no darwini." Clearly Pearson's field obser- vations indicate that his darwini (i.e., my P. d. rupestris) and ca- prinus are completely segregated ecologically and geographically. Budin's nominally sympatric mice may have been taken at a zone of overflow during a population explosion or during a time of tem- porary mixing of mice driven out of their normally segregated local- ities by some environmental change such as flood, fire or cultivation. These are speculations based on a possibly incorrect assumption that Budin's Alfarcito representatives of caprinus and darwini were ac- tually taken in the same trap line or even where contact between the two forms was probable. The Alfarcito specimens were not exam- ined by me but available material indicates that distinction between caprinus and darwini (i.e., rupestris) may be arbitrary. All specimens of caprinus collected by Pearson intergrade mor- phologically with darwini, wolffsohni or both. Unfortunately, this fact is obscured in the original description. Of a total of 65 sf>eci- mens from eight localities examined by Pearson, measurements are given for only 29 or less, all lumped without breakdown to locality. Formal comparisons are made only between 11 topotypes of caprinus 336 FIELDIANA: ZOOLOGY, VOLUME 46 and 7 specimens of darwini, all nominally from Tilcara. The sam- ples of these two populations differ more from each other than from all others known within the geographic area concerned. Tilcara caprinus (8 topotypes at hand) averages larger throughout than Til- cara darwini (1 specimens at hand), tail less hairy and relatively longer (129 per cent [116-135 per cent] as compared with 110 per cent [98-116 per cent] in darwini}, pectoral streak consistently pres- ent (consistently absent in darwini), fronto-parietal suture narrowly arced but not distinctly V-shaped (more widely arced, less sharply V-shaped in darwini}, supraorbital region flattened, divergent-sided, edges square, beaded or with narrow ledges (less flattened, edges square or pinched upward in darwini}, zygomatic plate vertical or slanted forward (vertical or slanted back in darwini}, ma with or without an enamel island (island present in all except extremely worn Tilcara darwini}. In conclusion, only tail pilosity and pectoral streak are constant distinguishing characters, while the others are intergrading. Camataqui caprinus (21 specimens) and darwini (8 specimens) agree in color, size (average head and body, 105 versus 104; skull, 28.7 versus 28.3), tail equally hairy but longer relative to head and body (125 per cent versus 114 per cent), pectoral streak present or absent (consistently absent in darwini}', fronto-parietal suture nar- rowly to broadly V-shaped in most, arced in others (narrowly to broadly arced in all darwini}, supraorbital flattened, divergent-sided, edges square (less flattened, edges square to pinched upward), zygo- matic plate vertical or inclined backward in both; m^ with island or more or less 8-shaped in both. No clear distinction can be made here between the two series. It is significant that the larger the number of samples compared, the fewer and vaguer the differences. Considering that the Camataqui populations are not synecetic and perhaps are allopatric, it may be strained to treat each as a peripheral population of caprinus and darwini respectively, rather than as a pair of well-marked pocket populations of either subspecies. The relationship of caprinus to wolffsohni was described in con- tradictory terms by Pearson. Regarding caprinus, Pearson (1958, p. 435) states that "large size, sharp interorbital edges, frequent ab- sence of the island in m2- [italics mine], presence of a second outer entrant notch or angle in ma and m^, and occurrence in bush and thorn scrub habitat occupied by Graeomys griseoflavus are all charac- ters [sic] shared at least in part with Phyllotis wolffsohni. Geograph- ically they are complementary, and thus suggest status of races." HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 337 The same author then concludes that he hesitates "to put into the same species populations in which the skulls can be distinguished so easily (by shape of m^ anj interorbital region [italics mine]) and in which geographically intermediate populations do not show evidence of intergradation [!]." Inasmuch as Pearson has already described a condition of intergradation between caprinus and wolffsohni addi- tional evidence of the same is hardly needed here. A large old male from Tarija (MVZ 120201), collected 25 Sep- tember, 1955, by Pearson, and geographically referable to caprinus, is as much like wolffsohni in external and cranial characters (fig. 83, o as anything so identified by Pearson. Its second molars are as in wolffsohni and many caprinus but m-, evidently 8-shaped when less worn, is now divided into two laminas (fig. 85,c). The Tarija speci- men was not mentioned or listed by Pearson with his caprinus or any other form of Phyllotis. A few specimens at hand collected by E. Budin in the Maimara and Sierra de Zenta region also show full intergradation between caprinus, wolffsohni and rupestris. A female from Sierra de Zenta (CNHM 85849) taken 28 April, 1947, has the cranial characters of rupestris, molars of wolffsohni, color characteristics of old pelage "cinnamon phase" rupestris and tail as in rupestris or many caprinus. A male (CNHM 41287) from the same locality, collected 27 July, 1947, is in unworn buffy pelage as in average caprinus or Bolivian rupestris, its interorbital region is as in caprinus and the fron to- parietal suture and ma as in rupestris. A female from Maimara (CNHM 85847) agrees with the preceding cranially and dentally but is intermediate in color between it and the "cinnamon phase" specimen from the Sierra de Zenta. Individuals of caprinus with pectoral streaks and thinly haired tails may be indistinguishable externally from Phyllotis osilae. The interorbital region of osilae, however, is narrower, the sides pinched, postero-lateral pits generally located on the posterior border of the palate, first primary fold of second and third molars obsolete or absent in the adult and ma usually 8-shaped or with pf 2 reduced to a large island. In some instances, however, separation from Phyllotis osilae can be made only on the basis of subtle (or dubious?) differ- ences in the structure of the glans penis or baculum alone. Remarks. Phyllotis darwini caprinus is the morphological and geographic bridge between ivolffsohni and rupestris. Its greater re- semblance to the first may be the result of a similar response to a similar habitat or may be merely the expression of the same genetic J"8! us 5i co GO «> rH ^x: 5 -t-> •- us* i US US g. bfl cs ^Jl us o 03 rH^ »HO o" 00 rH US rH CO CO 0 S3 us ^j« us" cc us 1 us us' O) a rH rH i— ( T-H *""* i rH rH w i— 1 O5 c 00 CN! 0 BQ -u to t- ^ co ^2 (-1 O S-. rt a> 4J 0 rH CO !M- CO f ^ o (^ S5 t>^ 1 2" CO •* CD w M^ O3 fj 1— 1 CT3 <* S «A C .-T ^ ss ci d ^«te T5 "~^ CO t> O t> CM -M CO CO C | CQ oo B9 O S; u? CO* O ^ T3 ^vi IM N ^ r^ •— t | 1 53 o cS -^T co S C C 2 H °i £i 7M_ PH ^ 'Sb c CO •<* xf CD 1 CM kO CD £ *£^ i £ (M g 9 Pi _3 CM oo O CO S^ § ll 03" O V OS" O rH I— 1 00 05 cu "S •2-3 S! •^j k fS, C &j oo" 05 US CM CD O O (M rH rH rH H H & s £ at g H ±5 «S S 13 05 N ^ M 1 Tilcara2 Sus' 338 HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 33«.» constitution. The above synonymy shows that Thomas considered specimens later called caprinus by Pearson as indeed inseparable from wolffsohni. P. d. rupestris occupies a different kind of habi- tat (cf. Pearson, 1958, pp. 400, 435). Whether near or far from the area occupied by caprinus its populations do not appear to be correspondingly more or less like caprinus. The lability toward convergence resides with caprinus. Measurements. — See Table 27. Specimens examined. —40. ARGENTINA. Juju-y: Maimara, 2 (CNHM, 1; MACN, 1); Sierra de Zenta, 3 (CNHM, 2; MACN, 1); Tilcara, 9 (AMNH, 8; MVZ, 1). BOLIVIA.- Chuquisaca: Carnargo, 3 (MVZ). Potosl: Yuruma, 1 (MVZ). Tarija: Camataquf, 21 (MVZ);Tarija, 1 (MVZ). Phyllotis darwini wolffsohni Thomas Phyllotis Wolffsohni Thomas, 1902, Ann. Mag. Nat. Hist., (7), 9: 131. Phyllotis wolffsohni, Pearson, 1958, Univ. California Publ. Zool., 56: 394, 409 431, fig. 2 (molars), pi. 10, figs, c, d (skull) characters; comparisons BOLIVIA: Chitquisaca (Tomina), Cochabamba (Arani; Cochabamba Cuchicanchi; Parotani; Pocona; Punata; Tapacari; Taquifla; Tiraque Totora), Santa Cruz (Comarapa). Type.- Female, skin and skull, British Museum (Natural His- tory) no. 2.1.1.59 (1.1.1.59 in original description); collected 23 March, 1901, by Perry 0. Simons. Type locality. — Tapacari, west of Cochabamba, Cochabamba, Bolivia; altitude, 3000 meters above sea level. Distribution (figs. 70, 74, 77).- Andes of central Bolivia in the de- partments of Cochabamba, extreme western Santa Cruz and north- western Chuquisaca; altitudinal range 2200 to 3500 meters above sea level. Characters. Larger and darker, more brown, than /'. darwini rupestris, larger than sympatric Fhyllotis osilae; upper parts bufTy to ochraceous coarsely ticked with dark brown; face paler than up- per parts; bright ochraceous lateral line nearly always present; under- parts sharply defined pale gray to dark gray heavily washed with ochraceous; a well-defined pectoral spot, streak or patch present in majority of specimens examined; hairs of chin and throat not white to base; tail usually longer than head and body combined, average 103 per cent (93-128: 100), bicolor except at tip, short-haired with little or no pencil; auditory bullae comparatively little inflated and less than alveolar length of molar row (figs. 83, r/ e, 84, r/ e), average 340 FIELDIANA: ZOOLOGY, VOLUME 46 length, less tubes, 5.1 (4.7-5.5, 25 specimens); bullar tubes short to moderately elongate; mesopterygoid fossa, measured at base of ham- ular processes, usually narrower, sometimes as wide as parapterygoid fossa measured at same plane, the sides parallel or slightly diver- gent; posterior border of palate rounded or with a more or less de- veloped median unnotched process; postero-lateral palatal fossa poorly defined, the pit (or pits) small and located in line with or anteriad to posteriad of posterior border of palate; interorbital re- gion usually broad but width at midfrontal plane less than greatest width of rostrum (fig. 83,d-e); supraorbital edges sharp, sometimes pinched upward but usually with posterior halves moderately diver- gent and frequently beaded or forming narrow ledges; proximal ends of nasals usually truncate and terminating behind in a line with or slightly behind premaxillary sutures; molar rows parallel-sided to slightly divergent posteriorly (fig. 84,d-e) ; crown of moderately worn m- with a modified sigmoid pattern (fig. 85,d-/) ; anteroloph of m- well developed, the first primary fold distinct in unworn to moderately worn tooth but obsolete or absent in worn tooth; crown pattern of m^ usually more or less sigmoid with first primary fold weak and ob- solete in moderately worn tooth, second primary fold usually open to margin, sometimes forming an enamel island, major fold always an- teriad to pf 2 and sometimes becoming isolated in worn tooth; m^ sigmoid with minor fold well defined except in very worn tooth; second primary folds present in unworn to moderately worn mT_^; alveolar length of molar row, 5.9 (5.6-6.3, 33 specimens) or 19 per cent (18-21 per cent) of greatest skull length. Taxonomy. — Phyllotis wolffsohni Thomas was originally distin- guished from other members of the darwini group by "the striking difference in the shape of the interorbital region" which was de- scribed as "much more expanded with marked divergent supra- orbital ridges." In 1913, Thomas (p. 139) identified six specimens from Maimara, Jujuy, Argentina, as Phyllotis wolffsohni and observed that they were "closely similar to the typical specimens." Later, Thomas (1919a, p. 116) recorded 22 specimens from Chumbicha, Catamarca, Argentina, as "not distinguishable from Ph. wolffsohni" and concluded that "the presence of divergent supraorbital edges in the type of Ph. wolffsohni . . . appears to be abnormal, or due to great age, as specimens since received are like Ph. darwini in this respect." Evidently Thomas was reaching the decision that specific differences between wolffsohni and darwini did not exist. However, Pearson (1958, pp. 431-2, 434) recognized wolffsohni as a full species and restricted HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 341 the name to central Bolivian material. The species was diagnosed as "an unusually large, long-tailed Phyllotis with sharp interorbitul edges, heavy rostrum, large tooth row [and] M« and M2 frequently with two deep outer entrant angles." Pearson referred the Chum- bicha series identified as wolffsohni by Thomas to Phyllotis darwini ricardulus and the Maimara wolffsohni of Thomas he combined with his own Phyllotis caprinus. None of the characters adduced by Pearson for specific separa- tion of wolffsohni from darwini and caprinus bear scrutiny. Pearson himself emphasized that wolffsohni was completely allopatric to neighboring races of darwini. His measurements prove that wolff- sohni is no larger than such races of P. darwini as posticalis, definite, xanthopygus and magister, and of some populations of any other race including the smallest, rupestris. Pearson's measurements also show that wolffsohni compares in size with allopatric populations of Phyl- lotis osilae. Proportional differences in size are non-existent and the diagnostic cranial and dental characters of wolffsohni are also pres- ent as individual and local variables in other races of darwini, par- ticularly posticalis, magister and rupestris. Finally, Phyllotis darwini caprinus of southern Bolivia and northern Argentina shows complete gradation from wolffsohni to rupestris in all characters, particularly in the shape of the baculum (fig. 7). The penis bone of adult wolff- sohni resembles that of Juvenal P. darwini rupestris. That of adult caprinus is as in wolffsohni or rupestris. Variation. — Specimens from Comarapa, Santa Cruz (2345 meter's), on the eastern limits of the range are comparatively dark because of the greater density of their guard hairs; the subterminal band of the cover hairs, however, shows the same range from buffy to ochra- ceous as in other series, including one from Tapacari, Cochabamba (2800 meters), which appears to be palest. Pectoral streaks are better defined and underparts more heavily washed with buff or ochraceous in darker populations. The tail is thinly haired through- out as in Phyllotis osilae. The broad supraorbital region with finely beaded edges is well marked in most specimens. In others (e.g., MVZ 120184, Tapacari; MVZ 120197, Comarapa [fig. 8.VJj the area is narrower, the edges square or pinched as in representatives of other races (cf. magister or rupestris from Tarata, Tacna, Peru, and San Pedro, Antofagasta, Chile). The sigmoid pattern of m ' is more prevalent in wolffsohni than any other race of P. darwini. This pattern depends on the presence of a well-defined second pri- mary fold open at the margin and entirely posteriad to the major S3 o o CO o> II be Z CO rH O O5 OO rH CO CD CO IO CO CD CO E o JS co co I I rH IO rH rH CO* O •2 w C^ (M C^ 10 10 TJ!" id" ^jT (M CQ ~ e a 13 ^ co co 'i T1 T1 >o oj t> co us 10 rH CO O (M CO "H< a PQ 03 .^ £PJ r- Gj TJ "2 », _- •5 J § g B B •gi I 8 § S _ O 3 cS O 3 3 03 O O CM PL, PH CL, H a 3^ 2 I1| o. cr g o3 .-: _ _ H EH H O omi oma 342 HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 34.'} fold of the opposite side. In 3 of the 29 specimens of wolffsohni immediately available, m;1 lacks the usual sigmoid pattern. In the first (MVZ 120194, Comarapa [fig. 85,/|) with well worn molars the second primary fold of m:l is suppressed and the major fold appears as an enamel island. In the remaining two specimens (USNM 271401, Tomina; CNHM 51918, Taquina [fig. 85,p], with molars less worn, pf 2 is isolated or suppressed but the major fold is well developed and open to the margin. Remarks.— Phyllotis darwini wolffsohni is the most strongly marked race. Its distinctive characters are the broad supraorbital region and wide fronto-parietal suture, the divergent-sided and fre- quently beaded supraorbital edges, greater persistence in adults of the first primary fold and the sigmoid pattern of ma. The same combination of characters though developed to different degrees are also present in partially sympatric Phyllotis griseoflavus. On the other hand, sympatric members of Phyllotis osilae show none of these characters and are significantly smaller as well. In some local- ities (Pomata, Cochabamba) the underparts of sympatric P. osilae are also darker. Allopatric wolffsokni and osilae may be indistin- guishable in size and external characters but the cranial and dental differences appear to be universal. The same internal characters, larger size and less hairy tail separate wolffsohni from Bolivian rep- resentatives of rupestris. The single specimen of rupestris from the geographically nearest locality, Tin Tin, Cochabamba, however, is indistinguishable externally except for smaller size. Elsewhere, in Peru, Chile and Argentina, are populations of rupestris with skulls and bodies as large as those of wolffsokni. Certain individuals and even entire series of rupestris share with wolffsohni the sigmoid pat- tern of m- but sigmoid ma may be found only in juvenals of rupestris and sporadically in adults of other races except caprinus where it is frequently present though never as clearly defined as in most wolff- sohni. Conversely, nv1 of wolffsohni is not always clearly sigmoid and its pattern may be as in rupestris (cf. Comarapa MVZ 120194 [fig. 85,/j) or, more commonly, as in caprinus. In no rupestris does the in- terorbital region become quite as broad and square as in most wolff- sokni although the supraorbital ridges may be as well developed. The nature of the supraorbital region in phyllotines is described elsewhere (p. 57). Phyllotis darwini magister, described by Pearson (1958, p. 431) as the counterpart of wolffsohni on the western slope of the Andes, is virtually indistinguishable in size and external characters except for its slightly more hairy tail. The enamel pattern of m'J 344 FIELDIANA: ZOOLOGY, VOLUME 46 in some specimens of magister from Rio Torata, Moquegua, and Arequipa, Arequipa, is as in wolffsohni but m^ is not sigmoid in adults. The supraorbital region in some magister is broad, divergent- sided and beaded as in most wolffsohni, while in some wolffsohni, the supraorbital region is as in most magister. Pearson (1958, p. 432) mentions a slight average difference in the length of the nasals rela- tive to the premaxillae. This is perceptible but adds no weight to Pearson's assumption that "skull differences are great enough to make subspecific relationship [between magister and wolffsohni] un- likely." Comparison of wolffsohni with caprinus, its nearest relative, is made under the preceding subspecies heading. Measurements. — See Table 28 (p. 342). Specimens examined. — 33. BOLIVIA. — Cochabamba: Cocha- bamba, 2 (CNHM) ; Cuchicanchi, 1 (AMNH) ; Parotani, 1 (AMNH) ; Pocona, 1 (CNHM); Punata, 5 (MVZ); Tapacari, 7 (MVZ); Ta- quina, 6 (CNHM, 5; MACN, 1); Tiraque, 2 (CNHM, 1; MACN, 1). Chuquisaca: Tomina, 2 (USNM). Santa Cruz: Comarapa, 5 (MVZ); Guadalupe, 1 (USNM). Phyllotis osilae J. A. Allen. (Synonymy under subspecies headings.) Distribution (figs. 70, 74, 77). — Highlands of southern Peru in the departments of Cusco and Puno above 2700 meters southward through the highlands of Bolivia above 1700 meters and northern Argentina as far as southern Catamarca and possibly northern La Rioja; altitu- dinal range from approximately 500 to 4400 meters above sea level. External characters. — Size as in P. darwini of the same general region; tail from approximately 80 per cent to 130 per cent of head and body length; ears comparatively small, rarely exceeding 25 milli- meters measured from notch, and usually less than 23 per cent of combined head and body length ; upper parts of body buff y or ochra- ceous to dark brown, more or less mixed with blackish; underparts rarely sharply defined usually dark gray with plumbeous basal por- tions of hairs showing through, a thin, sometimes a heavy, buff to ochraceous wash often present; pectoral streak, patch or mid ventral line usually present; inner and outer sides of ears thinly haired, dark brown, tail short-haired with weakly developed pencil, sharply bi- color or with terminal one-half inch to two-thirds or more of entire length uniformly brown. Cranial characters (figs. 87, 88). — Interorbital region narrow, par- allel-sided or, usually, concave mesially, the borders square, frequently raised or pinched upward and inward, never beaded or forming ledges; HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 345 proximal ends of nasals usually pointed and terminating in line with or, usually, behind, rarely slightly in front of, fronto-premaxillary su- tures; posterolateral palatal fossa poorly defined, the pit (or pitsi variable in diameter but usually small and located posteriad to, often in line with, infrequently anteriad to, posterior border of palate; pos- terior border of palate usually rounded, sometimes square, rarely incised and often with an unnotched median process; mesopterygoid fossa broad, width at base of hamular processes subequal to width of parapterygoid fossa measured at same plane, sides parallel or slightly concave medially, sometimes slightly divergent; Eustachian tubes usually elongate. Dental characters. — As in Phyllotis darwini but less variable; m- more nearly S- than 8-shaped, first primary fold obsolete or absent in moderately worn tooth; pf 1 of m± absent; pf 2 of m:i nearly always an enamel island. Comparisons. — The distinctive characters of Phyllotis osilae are included in the comparisons made between members of the Phyllotis darwini complex and other species of the genus (p. 256). Differences between P. osilae and P. darwini are idealized in figure 86. The baculum is described and compared in another section (p. 58). No single character will distinguish Phyllotis osilae from other species of the P. darwini complex. The shape of the fully developed baculum will separate Phyllotis osilae from all adult P. darwini with which direct comparison must be made but not from Phyllotis an- dium. The baculum of P. haggardi has not been studied but the organ is probably like that of andium and osilae. Unfortunately, material for comparative studies of the external genitalia of phyllo- tines is woefully inadequate and most, if not all reliance is placed on other characters. Much of the geographic range of P. osilae overlaps that of P. dar- wini. Except for the structure of the male genitalia, the characters which separate P. osilae from P. darwini in the regions of sympatry are of the same kind and degree which distinguish allopatric popula- tions and subspecies of either osilae or darwini. This state of affairs was vaguely recognized and incorrectly interpreted by authors until Pearson (1958, p. 422) provided the necessary data and directions for distinguishing sympatric populations of P. osilae and P. darwini. The characters adduced by Pearson are listed and discussed below. Those which prove to be of value in separating two or more sym- patric populations of osilae and darwini are summarized in Table 30 and shown in detail in Tables 31^8. The comparisons are based 346 FIELDIANA: ZOOLOGY, VOLUME 46 on sympatric series of osilae and darwini from Peru only. Available material from other areas of sympatry in Tarija, Bolivia (osilae- caprinus), Cochabamba, Bolivia (osilae-wolffsohni), and northern Argentina (osilae-rupestris) is insufficient for the form of analysis made in this part of the text. The particular specimens, however, are described or compared under the pertinent subspecies headings. A. EXTERNAL 1. Tail length. — According to Pearson (1958, pp. 423, 424), the tail of P. osilae compared with that of P. darwini is longer in the northern part of the range and shorter in the southern. Pearson's conclusions are based on average differences between actual tail lengths of se- lected individuals identified as "adults" from unspecified localities in each of three "regions of sympatry." Actual or absolute tail lengths may be misleading, and average differences are ordinarily not interpreted as specific differences. I have used the ratio of tail to head and body length for comparisons between individuals and populations of the same or different species. Furthermore only indi- viduals of sympatric series of osilae and darwini are compared in the following tables and only locally constant or non-intergrading differ- ences, not average differences, are regarded as valid. Tables 30-48 show that the tail of P. osilae is proportionately longer than that of darwini in 4 of the 18 contact points and shorter in 3. The small number of samples involved suggests that this variation in tail length is purely random. Differences in tail length in the remaining series are average and of no value for separating the species. 2. Tail pilosity (fig. 86,/). — The individual hairs are visibly shorter in all individuals of osilae in 12 of the 18 sympatric series, and the pencil is less if at all developed. The shorter-haired tail of osilae may be variously described in other parts of the text as less hairy, less pilose, etc. 3. Ventral color (fig. 86, e). — The color of belly and chest is apparently darker, i.e., with the dark basal portions of the hairs showing through more prominently in osilae than in darwini. The real difference between the sympatric populations is not in color but in length of individual hairs, those of osilae being shorter with the terminal whitish portion reduced. All individuals of 12 of 18 sym- patric series are separable on the basis of the apparent color of the underparts. HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS :U7 4. Pectoral streak (fig. 86,rf). A streak, patch, or mere fleck of buff or ochraceous on the midline of the chest is usually present in osilae and frequently absent in darwini. In some individuals or populations the streak may develop into a prominent midventral band or become confused with the wash or lining of buff or ochra- ceous extending over the entire ventral surface. Presence or absence of the streak distinguishes all samples of osilae and darwini in 10 of the 18 sympatric series and helps in sorting most samples of the remaining series. Pectoral spotting is usually associated with darker underparts (see 3 above) in both P. osilae and P. darwini. Clark (1938, p. 79) found pectoral spotting to be a simple Mendelian recessive character in the ecologically similar North American white-footed mouse Haplo- mylomys eremicus. This discovery was confirmed by Dice (1939, p. 14) who also noted that buffy pectoral spots occur more frequently on dark than on pale individuals. The shorter hair of the tail and underparts and the pectoral streak in osilae in localities where darwini occurs seem to be parts of a genetic complex independent of environment. Where the species are allopatric, shorter-haired populations occur at lower altitudes and darker ones in more humid localities. 5. Vibrissae and pelage (fig. 86,c). Pearson (1958, p. 423) notes that the vibrissae are "more slender in osilae"nnd the pelage "less lax and less fluffy, more evenly lined with black in osilae." I cannot appreciate any noteworthy or consistent difference in the thickness of the vibrissae but the pelage of the underparts is shorter, hence "less lax" in osilae than in darwini. These characters cannot be accurately measured or tabulated but may be considered as facets of the same genetic complex discussed in 3 and 4 above. 6. Ear size (fig. 86,a). The ear of osilae is smaller than that of darwini. Pearson illustrates the difference in size by absolute measurements and expresses it in terms of average differences. As in the case of tail lengths, absolute ear size is meaningless and usually misleading unless related to the overall size of the animal. The ears of osilae never become as large as those of large individuals of darwini. On the other hand, small darwini may have ears as small or smaller than average-sized or large osilae. Furthermore, ears are precociously large in young, and attain their maximum size in adults before maximum body size is reached. Thus, ears are disproportionately large in young individuals, and diapropor- 348 FIELDIANA: ZOOLOGY, VOLUME 46 tionately small in old individuals. In the tables both actual and proportional ear measurements are given but only the latter figures are used for comparative purposes. In either case, ear size proves to be a valid character for sorting nearly all sympatric individuals of osilae and darwini. 7. Ear color (fig. 86,6). — "Hairs clothing the inside of the ear pinna are frequently somewhat orange in darwini but never in osilae. . . ." (Pearson, 1958, p. 423.) This character is exceedingly fine and difficult to appreciate but, nevertheless, surprisingly con- stant in most sympatric series. The inner side of the ears of osilae are usually dark brown with or without a buffy or gray, but not orange or ochraceous, ticking or edging. B. CRANIAL 1. Nasals. — "Nasals in osilae come to a point posteriorly that almost always extends farther into the frontals than do the edges of the premaxillae." (Pearson, 1958, p. 423.) This is a valid descrip- tion but not a reliable distinguishing character. It has diagnostic value in only three pairs of sympatric series, each of which is com- posed of but a few specimens. Pearson also mentions that in osilae the rostrum becomes considerably broader in old age and that "osilae tends to have a conspicuous swelling on the side of the rostrum at the premaxillary-maxillary suture." This description applies to all other members of the P. darwini group. 2. Poster olateral palatal pits (fig. 86,h,l). — The pits, often set in fossae, are usually small in osilae and nearly always situated posteriad to, often in line with, infrequently anteriad to, posterior border of palate. The converse is true in darwini but the character is not constant enough one way or another in either species to be used alone for sorting samples of sympatric darwini and osilae. The character is further complicated by the fact that the pit of one side is sometimes anteriad while that of the other is in line with or posteriad to the posterior plane of the palate (fig. 86,0 • Again, there may be two or more pits of equal size variously located on each side of the palate. Position, size and number of postero- lateral palatal pits are determined by the number, position and diameter of blood vessels serving the affected area. In spite of its variability and complexity, the pit character is absolutely diag- nostic in 7 of the 18 pairs of sympatric series. a FIG. 86.— Idealized diagnostic characters of f'hyllolix dnrm'ni and /'. nuilar. For explanation of characters see text, pp. 346 348, 350. For thfir validity and relative constancy in sympatric populations, see Table 30. 34!) 350 FIELDIANA: ZOOLOGY, VOLUME 46 3. Mesoptery goid fossa (fig. 86,/). — The fossa in osilae is broad, frequently as wide as or slightly wider than the parapterygoid fossa measured at the anterior base of the pterygoid processes, the sides parallel or slightly concave medially, sometimes divergent. In sym- patric darwini, the width of the fossa at the anterior base of the pterygoid processes is usually narrower than the parapterygoid fossa measured at the same plane, the sides usually divergent or parallel. The character holds in the majority of the sympatric series examined. 4. Palatal spine (fig. 86, not labeled). — Posterior border of palate in osilae is usually rounded, sometimes square, rarely cleft in midline and often provided with an unnotched median process. In darwini, the posterior border may be rounded or square and often marked by a single unnotched median spine. The character holds in a number of the more poorly represented series. 5. Palatal width. — Pearson (1958, p. 423) finds a difference of .28 millimeters between the average width of "38 adult osilae" palates and the average of "20 adult darwini" palates. The iden- tities or localities of the specimens selected for measurement are not revealed. In any case, the character seems to be more directly related to the size of the individual skulls than to real differences between populations or species. 6. Bullar tubes (fig. 86,&). — The Eustachian tubes of the audi- tory bullae are usually elongate and tubular in osilae, short, often flattened and strongly ridged in darwini. Although its use requires some judgment by the observer, the character holds in nearly every instance where comparisons need be made. 7. Interorbital region. — "In osilae frequently more pinched than in darwini and never with sharp or square edges." Pearson (1958, p. 423). This character is descriptive rather than diagnostic. Pearson notes an average difference of .17 millimeters between the inter- orbital widths of 38 osilae and 47 darwini. C. DENTAL 1. Enamel pattern of m^. — There is little difference between sym- patric osilae and darwini in the enamel pattern of moderately worn m^. In unworn m^, the first primary fold is usually less developed in osilae than in darwini but the difference is not significant. In allopatric darwini, pf 1 of m^ may be well developed in the new HERSHKOVITZ: NEOTROPICAL CRICETINK RODENTS 351 tooth and persistent in the worn but it is always weakly developed or obsolete in osilae. Convergence in the enamel pattern of the molars appears to be the rule among closely related cricetines occupying the same or similar habitats. D. BACULUM The base of the baculum is convex to nearly plane in adult osilae and concave to nearly plane in darwini (figs. 6, 7, 86). The baculum is described in a separate section (p. 58). Variation. — Four geographic races, one nominal, are recognized. Phyllotis osilae osilae is the centrally located, most widely dispersed and most variable subspecies. It ranges from southern Peru through the Bolivian Andes into northern Argentina. Northward and east- ward across the divide into the upper Amazonian drainage in Puno, Peru, and La Paz, Bolivia, osilae grades into extremely dark pha-eus. Continuing southward in Bolivia and northern Argentina, osilae becomes progressively more warmly colored and merges with tucn- manus. Phyllotis osilae nogalaris is known only from the type lo- cality, on the eastern slope of the Andes, in Jujuy, Argentina. Its status is uncertain. Color appears to be the only trenchant character for distin- guishing the subspecies of P. osilae. As in P. darwini, and most mammals for that matter, the more intensely pigmented forms occur in the more humid regions, the more dilute or pale forms in the drier areas. In the case of P. osilae, the color cline is almost cer- tainly continuous. Additional material from intermediate collecting localities may prove that there is no real geographic line of separa- tion between the subspecies now admitted. Habits and habitat. The two original specimens of 1'hnllotis osilae were collected by H. H. Keays at the same localities as P. boliriensis. Keays noted that both species had the same habit of feeding on grass during the day. Pearson (1951, p. 144) reports, however, that "several of our specimens were captured at night." According to Pearson (1958, p. 399) "the long but narrow range of osilae will . . . prove to coincide with the distribution of ichu or ichu-like bunch grass . . . [and, p. 424) osilae does not enter the tola heaths nor does it live in the brushy and more arid western slopes of the Andes where ichu does not grow well." The altitudinal range of P. osilaf, from approximately 500 meters (Chumbicha, 352 FIELDIANA: ZOOLOGY, VOLUME 46 Catamarca) to somewhere between 4000 and 5000 meters ("Acon- quija 13,100-16,400 ft.," Catamarca, fide Pearson, 1958, p. 427) sug- gests, however, a much wider variety of habitats. The geographic area occupied by osilae in the Andes of northern Argentina incorporates the prepunena and puneha phytogeographic provinces defined by Ca- brera (1953, pp. 130, 145, fig. 12). The first extends from Jujuy to La Rioja between 2300 and 3400 meters above sea level but in some places descends to 800 meters. The province is dominated by bush- steppe and cirriform cacti. The parallel punena province lies between 3400 and 4500 meters above sea level. It is also marked by bush- steppe as well as herbaceous, halophytic and grassy steppes and meadows. The ground in both provinces is largely sand and rock. Altogether it appears that the prepunena and punena provinces com- bine all the elements said by Pearson to be favored by P. darwini rather than P. osilae. On the other hand, the next higher district, the altoandino septentrional, which lies, according to Cabrera (1953, p. 142), between 4500 and 5500 meters above sea level, is dominated by grasses, including Festuca and Stipa and should be the preferred habitat of P. osilae (fide Pearson). So far, the species has not been recorded from these altitudes. Reproduction. — Information gathered from notations on skin tags are tabulated below(Table 29) and summarized as follows: 1. Highest proportions of gravid and suckling females appear from April to June. 2. Most males are in breeding condition from December to May. 3. Lowest proportions of gravid and suckling females appear during September and October. 4. Most non-breeding males appear from July to October. 5. Means and extremes of the number of embryos are 4.4 (2-6) 8 specimens. 6. In 5 females for which data are available, the count of em- bryos in left and right uterine horns, respectively, is 1-5, 2-3, 3-1, 1-1, 2-3. 7. The rough correspondence between ostensible breeding cycles in P. osilae and P. darwini has been noted. Comments (p. 277) on the reproduction of P. darwini also apply to the present species. Remarks. — The difference in the shape of the base of the baculum and, inferentially, the structure of the soft parts of the glans penis, is slight but is all that stands between regarding osilae and darwini as sympatric subspecies. Many individuals and entire series of HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 353 Phyllotis darwini (sensu lato) are absolutely indistinguishable from osilae in the combination of their external, cranial and dental char- acters. Some of these are geographically intermediate. If the struc- ture of the penis were known from more samples and for all races of darwini and osilae, possibly its value as a diagnostic character would collapse like all the others. This is suggested by the fact that the bacula of juvenals of P. darwini and P. osilae are prac- tically indistinguishable. Phyllotis osilae and P. darwini rupestris resemble each other most where they are sympatric. The convergence appears to be restricted to the more widely distributed and plastic rupestris. Phyllotis osilae, on the other hand, does not tend to become rupestris-\\ke anywhere throughout its range. In spite of their similarities interbreeding or hybridization evidently does not take place between sympatric pop- ulations of osilae and rupestris. In sharp contrast to the above, Phyllotis osilae resembles certain allopatric populations of P. darwini posticalis more closely than it does sympatric populations of the same form. Thus, for example, sympatric series of osilae and posticalis are distinguishable by a large number of characters (Tables 44-48) while osilae from Choro, Bo- livia, is indistinguishable except in overall size from widely separated P. darwini posticalis of Palmira, Peru (Table 49). The bacula of the Palmira series, however, are of the P. darwini type. Those of the Choro osilae are unknown. The morphology of sympatric series of P. osilae, P. darwini rupestris and P. d. posticalis suggests that all are offshoots from an ancestral posticalis stock. Ostensibly, rupestris and posticalis have always maintained contact at various points while osilae became geographically and then reproductively isolated. Subsequent con- tact between osilae and rupestris was accompanied by convergence on the part of the more widely ranging and labile rupestris. Present overlap in the ranges of osilae and posticalis is also secondary but available material is too imperfect for determination of the relation- ship between the forms. On the other hand, contact between phaeus, an offshoot of osilae, and the peripheral Limbani populations of posti- calis is a meeting of two terminal twigs of the ancestral divergence (fig. 76). FIG. 87. — Skulls in dorsal and ventral view: a, Phyllotis osilae osilae from 3 miles west of Asillo, Puno, Peru; b, P. darwini rupestris from 5 miles south of Asillo. (About X 2.) 354 FIG. 88. - Skulls in dorsal and ventral view: a, I'hyllultx nnilar onilar from Hacienda Pairumani, Puno; b, /'. dartrini rnpc*trix from samp lorality. Note "osilae" type mesopterygoid fossa, bullar tubes and position of post errvlat oral palatal pits in P. danrini, and "darwini" positions of pits in /'. nuilac. i About • 'J. > •'i55 356 FIELDIANA: ZOOLOGY, VOLUME 46 TABLE 29. —Reproduction in Phyllotis osilae Data from skin tags of specimens collected by O. P. and A. K. Pearson, C. B. and M. Koford and E. Heller. See summary, p. 352. Month Locality1 January Juli, Puno Pomata, Puno February Pairumani, Puno Puno, Puno Vilque, Puno April Have, Puno May Calacala, Puno Macchu Picchu, Cusco June Arapa, Puno Calacala, Puno Macchu Picchu, Cusco Umayo, Puno August Limbani, Puno September Punata, Puno S. Rosa Ayaviri, Puno Tiraque, Cochabamba4 October Asillo, Puno Limbani, Puno Leon, Jujuy5 November Asillo, Puno December Puno, Puno 9 9: 0,0 0 5 J,0,0,5,6,L,L,L J2,J4,J5,L 4 4 JL 0,0 0,0,0,0 0 0,0 0 0 cf c?3 10 7,8,8,9,9.5,10 11,11 12 10 J4,J5,J5,J7,9,10,12,13 J5,10,10,12 2,3,4 4,7 J4,4 2,3 4,6 6,9 4,4,4 6 4,5,5 7 11 1 All localities in southern Peru unless otherwise indicated. - Each number (0-6) represents the number of embryos in each female. J= virgin or young; L=lactating. 3 Each number (4-12) represents length of testes in millimeters in each male; J= juvenal. 4 Bolivia. s Argentina. Footnotes and explanation of symbols in Table 30 1 Sympatric series from 13,000 feet. - The 8 specimens of the type series of P. osilae phaeus from 11,000 feet and 4 P. osilae phaeus from 11,500 feet, compared with 16 P. darwini posticalis from 1 5,000 feet which are not strictly sympatric. 3 Allopatric osilae and posticalis alike in all characters compared, with possible exception of baculum. X = Trait valid for distinguishing all samples of each pair of sympatric series. o= Phyllotis osilae osilae. ph= Phyllotis osilae phaeus. r= Phyllotis daricini rupestris. p= Phyllotis darunni posticalis. ?=Data unavailable or unknown. R = Reverse relationship of traits. Blank space= Character not valid for given locality. HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 357 TABLE 30. —Comparisons of sympatric series of Phyllotis osilae and P. danrini in southern Peru (See opposite page for explanation of symbols, and Tables 31 49 for individual variation in each locality.) 05 C >, u C C IT. CU CD _c g . = c s _c _^ c •5 1 Js '5? o o -o •J: ? '1 a "2 be § £ = | S.'s "o ~ 15 5 ^ ctt £ •3 •= * > C S- a; u O. L. fC ctt o ct •¥ "** JM — — . — - — 1* t- — C o> jj u 0 o "« "« ^ ! 1 P oo W H > W PQ OH 0, ft, H x cc H Arapa o-4 r-1 X X X X X X X X X X ? 104-1? Puno 0 1 r-2 X X X X X X X X X 9 Huacullani o a r 4 X X X X X X X X ? 84-1? Occomani o-2 r-2 X X X X X X X X 9 84-1? Puno o :{ r-3 X X X X X X X X 8 Juli o :} r-5 X X X X X X X X 8 Asillo o-3 r-6 X X X X X X X X 8 Umayo r 4 X X X X X X 9 64-1? Pomata o 7 r-2 X X X X X X 6 Chucuito 0-5 r-8 X X X X X 9 54-1? Yunguyo o 12 r :J X X X X 9 •14-1? Pairumani o 4 r :t 9 9 9 X X X X 9 J M? Pairumani o 18 r-3 X X X X 4 Ollantaytambo 0 1 p 1 X X X X XXX 9 7 fl? Huaracondo o :t P-1 X X X X 9 R X ? 54-'?4lR Urco o-4 P •» X X X R 9 3 4- 1 ? 4 1 R Limbani1 ph 1 P 2 X X X X X X R X R X 84'JR Limbani2 ph 12 p IB X X X X X X X i Choro, Bolivia3 o 5 9 04-1? Palmira, Peru3 p 5 C-' e- «- 71 71 *4 C- z '- a; " ^ Incidence of Validity (18=inflCK.i 4- ^r 4- e-o en + *— • CO 4- 4- — 358 FIELDIANA: ZOOLOGY, VOLUME 46 EXPLANATION OF SYMBOLS TABLES 31-49: Comparisons of Phyllotis osilae with sympatric Phyllotis darwini. For descriptions of traits see pp. 346-350 and fig. 86. A= American Museum of Natural History catalog number. C = Chicago Natural History Museum catalog number. H= Museum of Comparative Zoology catalog number. M = Museum of Vertebrate Zoology catalog number. U=U. S. National Museum catalog number. + = osilae trait. = darwini trait. o=trait intermediate or indeterminate. V= valid trait for distinguishing all samples of each sympatric pair of series. N= invalid trait. R= traits consistently reversed. ? =data unavailable or unknown. Blank space=data unavailable or unknown. s=subadult. j= juvenal. t=type specimen. e= measured on dry skin. iunpiot:g -f autds pm'P'd 4-4-4-4- I o* c 3 Q* ,-> J I (apis qot;a) c 4- 4- 4- I | .£ Joioo IBJJIMA £ > 4-4-4-4- I > A")iso|id Jin,!, 4- 4- + 4- I > jojoo ji;y 4-4-4-4- I > - -, - _ - A'poq pun pwoq :at:y 71 •"••! -M -M ?'i C "Z -M 05 — »0 z -r A'noq l)Ui; DBOU :IIBI — o ~- o» _K # — — • — •5 a ~ - .o x „ Xpoq pui: pnon 2 ^ J o ec "o PH ~^* ^C 7C 7 1 ^5 ]in>jS oi os o t-' x :;• 71 71 77 71 71 ^ C C 359 B w o* 3 I -8 so I '5 "s> g ^ E c '— lunjnoBg + 1 > M)3ue, IE8BN + + + 1 1 + 1 1 !£ saqnj JBjjng + + 1 1 1 1 1 1 1 £ q^piM BSSOJI + + + 1 1 1 1 1 1 > auids IBIBJBJ + + + 1 1 1 + 1 1 £ (apis qaBa) s^td IB^BJBJ + + + + + + 1 1 1 + 1 1 ^ \ 1 1 + 1 \ A ,0,0, WueA + + + 1 1 1 1 ! 1 > **, W* H — ho i i • i i ii A!Sot!dl!Ei + + + 1 1 1 1 1 1 > JOJOO JB[J + + + 1 1 1 1 1 1 > lUB pB3l{ :JBJ[ os oo i i-H I— 1 1 1 1 1 1 1 1 > Xpoq . O IIBiL Xpoq puB os od o > t- Ui «— 01 C rh ^ I ^ B X^isojid IJBJ, -f 4-4- +4- || +lllll/>: •-8 »- *» s -S -5 jo[oo jB3 4. + + 4- 4. || I? | °* Xpoq pus pBaq :jBg 4. 4. 4. 4. 4. | | •2 'C * £ Q >. •*< K Apoq put? pnaq IJIBJ, § put; o — ri 3fil iot- rc-r t «o x o o — ?i »o OCX OO1CT> — IO ^O i O IO • ^ ^O ^O to ^O &O fcO to to CX 71 M 71 71 «• 71 71 71 71 71 71 71 71 X >O »O >O >O ~% »O iO iO »O iO »O iO iO f,J OO OO a.OO OOOOOO umjnoBg c- [BSBN + + + I I I I > saqm J^Iing + + I I I I > 3 u^piM BSSO^ + o + 1 1 1 o jz 1* <1> $4 O auids {B^BiBj + o + 1 I 1 1 ;z; 3 "S J3 (apis qoBa) o 1 o 1 1 1 1 ^. S^ld IB^BIBJ O | | 1 1 1 1 "3 u S3 3 £M cc joioo iBJ^uaA + + + 1 1 1 1 > O £ »-i S c 0 }jBaj)s jBJO^oaj -\ — | — 1_ 1 1 1 1 [> QJ 05 3 1 <» • ^ S ^ '§ JOIoo .reg + + + | | | | > % ^ ^ 1 ^ ^poq pus pBaq :jBg ooo ^^HIMO^ .80 £ •2 cs ^ a "S> £ jprr CO (M CO Id -r)< t- 00 JBid > l-H ~— — T— ( 0 V. C rt a 00 i— ( CD CD IO M5 "^ ?5 £ IIB.J, T— I^HrH Oi— IT— li— t JJj 6 §0 I C8 1 a •^ co ^poq pus pBajj 2Io^ o2^^ c H tn 3 "o H qj3uaj Ijnjjg 1 06 oi o oo" oi > (M OJ CO auids jBiqBj + + + I I I I I > o 3 (X (apis ipBa) o + o I I I I I c + o I I I | | E O .cc J0|00 JRJ1U3A + + + I + I + + + + + I I I I I > r o> ^ ° < -S %> 1-5 51 A}ISO[ld [JBJ^ -f _(_-)- I I I I I > J £ •f«i «s> § .5 2 \ J°Io:) Jt?3 + + + I I I I I > PBaq : * o. ON C J«3 pus Xpoq pus p«3H o -r 10 71 T7 —i i- >c oo E ~ 00 OS t- 00 0> i- i- i- oooooo iO iO iO x irt »C lO lO 1C ^^n^ 5nz^~zi~ 363 + + saqni + + I I 1 £ rj - c O auids (apis ' JOJOD + + + + o + + 4- + + ' ' I > Xpoq puB pBaq Xpoq pus :JIBX X ~? Xpoq PUB pBajj S 00 CO oo r O 0 a 00 364 mnjnoBg e- SBN + + + + II I > saqm JBung + 4- 4- + II I > tpptAv BSSOJ 4- + + + II I > atlldS IB1BJBJ I+I + II I Z (apis qoBa) 4- 1 4- 4- o | | | ^ £ sVid IBIBJBJ 4- + + 4- II 14-' N E « joioo iBJ^uaA *" 2 '* •a | - ^ jfBaais iBjo^Daj 4-4-4-4- II I > C W JOJOO JBg ^ -I « S 'O O) O) O) *~* *~^ ^? — ^ OO OOOOOO.?£U5iO O) Q. ^^* ^f^Tf^ccco f [jj -^ — ~ ~ _ £.. _ _ £ 365 tunjnoBg + -f + saqm auids (apis t|.)i:,i ) |oo + o||+ |+o I o ooo + oo I |+o+ |+o c 03 I •- .t > « fe PH 0 C &D o *C *. - _r? jojoo S S . .S « 1 ~S Q* jojoo DUB PB3U IJB'J 00000000000,-H t---iC^ ,-l <-HOJi~i'-iT-iT-l••-> W COO'— I MOO 366 qtfuaj |BSBN + + + + + + I++Z saqnj JBjjng + + + + + + I I I > auids pnBjBa + + + ill | + + (apis qoBa) + c o | + + 0 o •3 ~ JOJOD irj;uaA + + + + + + I I I > I I 4- + + I + + 111^; "-» v. ^ 'f^ •2 .« '?• 5 § ^ + + + + + + I I I > Xpoq pun pnaq t § B» jt;q M M -H ri« M ^ N 04 04 Ol C^ •~ 'c 5 a Xpoq puB pBaq :ji I EM . Irt r-» \ \ I e o |+ | i o + +i o '8 aojoo DUB psaq : + + + + + ++ oot-oooot- ^H I I > + I 55 O J3 c --S Xpoq PUB PBaq rji HBJ, t- C- IO 05 00 T}< (O «£> t- t- oo oo oo oo oo oo oo o so t- t- oo oo 368 o uinjnoBg -f | c- qi3ua| IBSBN + + + Z saqnj JBnng + 1 1 > p qjpiM BSSOJ + 1 1 > 1 s. auids (B4B|Bj + 1 1 > 2 0) PL, o" (apis qoBa) S^Id JB1BJBJ o o [ > 3 PH d" C X jojoo jBj^ua^ + 1 1 > JS "So ~ . s j«a + i i > Xpoq pus psaq =JBg ™ Xpoq pu UBJL o Xpoq puB M co w oo q^uai [[n>is ^ ^ «j •< CO CO HIP!^ ESSOjI + + + I I I > auids JBIBJBJ + + + I I I > (apis ipna) oil | | I | O + + III C .£ JOJOO £ > C i I O i .« c S = s . .8 jopo JBg H — | — \- I I I > o Xpoq ptn? 42 f O >> . . Xpoq * ' O. I Xpoq pus E-1 t~ ^t< ^D o Tt eo HB»L »-• o oj o ^H lM BSSOJ + + + I I I I > euids JB^BIBJ [ | | I I I I 2 e CL, (apis qoBa) l+o I | I I >, sjid IB^BIBJ + + c I o | | CO jojoo jBJiuaA + + + I I I I > c rt •i: > c ** ^ O JJt?8JJS IBJOpaj + + + + + I + ^ < -c < 1 Xiisojid |IBX + + + I I I I > ^> w *> . rS S JOJOO JBg + + + I I I I > » "° "5 Xpoq pUB pBeq IJIBJ, ooo c£ c£ o o> Z I » 00 N TO M TO OS 7-J "5 O I!KX O «-i «-i -HOO--H TO Xpoq pUB J flQ U3 Oi *j ^ TO W ^i n t~ — -c >« to O O) O O) O^ O) 9) co to ?o co co co co cococo _« cococcco ' 371 tunjnoBg saqn; BSSO.J + + + + + + + ++ + I I E*3 2 g *c O — r CO "5 'C autds (apts I I I > o+ + o |+ + o + |+ |+ 1+ o + + I + I I o + 1+0 + o + o 1 I JOJOD -i ~ ^poq S5 a, pus pBaq :jBg •2 'C •3 *-> -2 O, j^rrr t- O5 O5 O •5 -g ^p°q c -g PUB psaq :|i o ^ 0 O Xpoq puB IM t- (M O T-H UJ T-H T-l T-H CO CO O U5 O5 t— o o o TJ< 10 oo t-' 06 oo* Cvj c^ N ••-i •,-> -m VI t- O5 O (M (M «D t- saqn; JBjing 4- 4- + I > !M BSSOJ 4-4- + 1 auids JB^BJBJ | + 4- 4-Z 0 o •o (apis qoua) 1 4- o 1 ^ 1 W^vd | + 14- \ " 1 1 aojoo jBJ^ua^ 4- 4- 4- + Z S o 0 JC r i >|r,u is (BJO^aj + 4- 4- + Z, C to o> US — O < -I X^isojid jtBj^ 4- 4- + \ > *^ o >-3 '« <» -S ^ | jojoo JBg + 4- + \ > ® "O "fc .2 tt. Xpoq puB psaq:jBg ^ & 2 53 > § -c .00 *-> ^; ^ ^^^ ^^^ tfi .s I JBg ^ S N CM s ^ O <3> 00 ••; *-> i " Xpoq puB pBaq IJIBJ^ o o o 1— 1 JH tf •c OO o. n^j, ^ ^ ^ "^1* CO E 1-1 ^ *H 1-1 0> 0 _* 0 a l_ Xpoq PUB pB3H oo 0 »o c S 0 •f tn S o X) » us ^> oo °. £ < ui3uai [inMS 06 «5 fj "* >> CO OJ ^H oo t— O & & | 373 q;3uai (BSBN + + £ mpiAv Bsso^i + + auids o" (apis sid c = a J°I00 O S ^ ^ 3fBaj;s [Bjoioaj + + ^ I 1 A-}Tsond HBJ, + I > •2 fe 'S •§ « K. Xpoq pUB pBaq qiBj, S o > c oo 'C IO t- U3 X JIBJ^ -H CO S* 0> £ M o « Xpoq pus pBajj •C •— , 05 — O R. a W 374 08 •O •r tunjnoBg e- qi3ua[ {BSBN + + -f + I I + + £ saqm JKnng -f o + o 00115?; « mP!-* vssod + + + + + + + + £ PH auids IBIBJBJ III I 111+?: O (apis qoBa) + 0 + | + I I | +00 + I I I J°Io:) l«-HuaA 4- + 4- + I I I I > £ O E- JB3JS KJODa _|_ + _|_ 4. KOU IJBX + + + + I I I I > B k. I •§ Xpoq pUB pBaq :jBg jBg 03 <-i fS -' o od od 06 o» I w >» 375 O c8 09 O< I «>• I \ saqn; \ z \ \ auids Pi (apis s^jd C co 1 1 £ .CO O i-3 £ 8 I •§ ». n' jojoo I0 l!BlL JOJOD JBg •sj •& (^ .*J Xpoq PUB pBaif t w ^r ^ rH rH (£ •c ?3 a o O IFJL N Apoq pus CO O be rt a fl o £ o a 5 3. w 376 o § o ° ° It* • S -Q 3 S3 ff " saqni JBjjng I I I I I I I I I I I I iM BSSOJ ++++++++++++ autds JBIBJBJ I I I I | I I I I I I I (apis sijd aojoo 8? | JOJOO r. "^ 1 A"poq PUB psaq : •s I •^ R. 03 a I O 00 /._„ Apoq pus pBaq nir» pUB m «£> t- oo I 377 ft, a o 1 c 3 ^ &« £ c" § S3 O_ "p t5 'S II •C •"" o c o c bC C a g e o "a. ^ •3 3 lunjnoBg III > «BN I I I I I I I I I I I I I I I I > saqn^ JBjing I I I I I I I I I I I I I I I I ^ IAV BSSO j 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 o | | I | | I I I I I I I I I I 9UldS (apis qoBa) sjtd JOJOO 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 I I I I I I I I I I I > JOJOO JBg I I + I I I I I I I I I I I I I Xpoq PUB pBaq :JBg t—^^<'CT5OSt-OOOO OiOi =s .«. Xpoq pus psaq I* Xpoq puB x eo ~. ~. i— i CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD 03 "3 ZH^^-^IiH^^H^^H^rH^^H^H^^H >* 378 + + o o + o + o o o + 0 | + | o o o o o BSSOJ auids T « •° 3 j= fc (aP!s MDBa) f o o I o O OH Slid JBJB|B(J + O O | -f- O I I + O 5; fi •2 'i O a £ ^ 1 rt MBaJ^S IBJOpaj + + + I I + + I + I £ •^J I— C j| s "3 <1 ^ nt>an • IP^ CO Oi •— i O T-H i— i 00 O CO T-I i^- y H Apoq put; pi;eq .JKJ c^-— lojojoj OJI-HOJO-JC^^ r^ oo O "Q so '^s T-H ^5 ^i <^ ^^ t^- OO CO IO t^~ '•C5 "S J^S iM O) "-H O-J IM OJ OJ O-J CO Z O **• o . • .20, U30000 OCDO^g b I'^X CS OJ O Ol >— i jt5 l-s & § * a O O"3OOU3 OOCOOSiOOl Xpoq puB psajj ojoosooi ^J2^^SJ UO O t^ CO CD t~ O '30 ^O O CD Oi ^O OO t~** *~H c^ *~H ^^ ^5 OJ Ol CJ OJ *M CO CO CO CO CO c _0 .-, ea — ^ ce n OOO-HOICO 00050^-01 oo oo 00 QO oo i2 t- t- t- I 379 380 FIELDIANA: ZOOLOGY, VOLUME 46 Phyllotis osilae osilae J. A. Allen Phyllotis osilae J. A. Allen, 1901, Bull. Amer. Mus. Nat. Hist., 14: 44— PERU: Puno (Osila [ = Asillo], type locality; Tirapata). Goodwin, 1953, Bull. Amer. Mus. Nat. Hist., 102: 323 — type history; measurements. Phyllotis osilae subsp., Pearson, 1951, Bull. Mus. Comp. Zool., 106: 144— PERU: Puno (Pairumani). Phyllotis osilae osilae, Pearson, 1958, Univ. California Publ. Zool., 56: 422, 426, fig. 6 (penis), pi. 9, figs, c, d (skull); pi. 13, fig. / (penis bones) — revision; ARGENTINA: Jujuy; BOLIVIA: Cochabamba; PERU: Cusco; Puno [for individual locality records see p. 252]. Phyllotis lutescens Thomas, 1902, Ann. Mag. Nat. Hist., (7), 9: 131— BOLIVIA: Cochabamba (type locality, Choro, 4 km. northeast of Cocapata, Ayopayo Prov., 3500 meters; Choquecamate); La Paz. Thomas, 1926, Ann. Mag. Nat. Hist., (9), 17: 321— BOLIVIA: Tarija (Pino [ = Pifias]). [1]Phylloti8 arenarius, Thomas (part, not Thomas), 1902, Ann. Mag. Nat. Hist., (7), 9: 224— BOLIVIA: Potosi (Lagunillas; El Cabrado). Phyllotis darwini tucumanus, Thomas (part, not Thomas, 1912), 1918, Ann. Mag. Nat. Hist., (9), 1: 187— ARGENTINA: Jujuy (Leon). Phyllotis darwini posticalis, Thomas (part, not Thomas), 1920, Proc. U. S. Nat, Mus., 58: 230— PERU: Cusco (Huaracondo [part]; Macchu Picchu; Occobamba Valley; Ollantaytambo [part]). Type. — Adult female, American Museum of Natural History no. 16503; collected 17 October, 1900, by H. H. Keays. Type locality. — Asillo ("Osila"), Puno, Peru; altitude, 3023 meters. Distribution (figs. 70, 74, 77). — From southern Cusco, Peru, through the Lake Titicaca drainage basin in Puno and western Bolivia, south- ward through the Andes of La Paz, Cochabamba, eastern Potosi, pre- sumably in western Chuquisaca, Tarija and into the department of Jujuy, northern Argentina; altitudinal range from approximately 1200 to at least 4200 meters above sea level (Hacienda Pairumani, 13,500 feet, O. P. Pearson) and possibly to 4375 (Macchu Picchu, 12,000-14,000 feet, E. Heller). Characters. — Palest race of P. osilae; dorsal surface buffy to tawny finely ticked with dark brown, posterior half of back often more warmly colored than anterior half; lateral line usually ill de- fined or absent; underparts gray with dark basal portions of hairs showing through; tail longer or shorter than head and body com- bined, bicolor except at tip. Variation.— Color of upper parts and sides of osilae from Puno, Peru, are virtually identical with those of sympatric P. darwini rnpestris and vary individually and locally in the same way. Cusco HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 381 osilae resembles darker P. darwini posticalis of the same region. Specimens of osilae from northern Bolivia (Choro) are darkest and intergrade with extremely dark P. o. phaeus. Central and Bolivian osilae are more warmly colored, like sympatric wolffsohni. Farther south osilae grades into the richly colored brown tucumanus of north- ern Argentina. P. o. osilae also tends to match its sympatriots in size and bodily proportions. The enamel pattern of the molars of osilae is comparatively simple with insignificant individual and local differences. Nevertheless, the tendency for convergence also appears here. The first primary fold of m- is obsolete or absent as in sym- patric P. darwini rupestris but the tooth is more nearly S- than 8- shaped as in sympatric P. d. wolffsohni and P. d. caprinus to the south. Other external and local variables are itemized in Tables 30-49 and discussed under the species heading above. Taxonomy. — The only known character which consistently dis- tinguishes osilae from sympatric races of darwini, is the shape of the penis bone (and associated soft parts). Any other character which separates sympatric populations of one locality may be a variable in either of the sympatric populations of other localities. Peripheral and allopatric populations of which the genitalia are un- known are assigned to osilae or darwini primarily on the basis of geography and analogy. Knowledge of the microhabitat of each sample may also aid in the specific determination but the available data are incomplete, generally vague, sometimes subjective, in part contradictory and mostly untested. Pkyllotis lutescens Thomas from Choro, northern Bolivia, de- scribed as "a southern representative of the Ph. Haggardi of Ecua- dor" was believed to differ from its nearest geographic ally, Phyllotis osilae, by "its shorter tooth-row." The alveolar length of the molar row of the type of osilae is 5.1 or 5.2 mm., the crown length about 4.7 mm., not 5.7 mm., as given in the original description. The crown length of the molar row of the type of lutescens is given as 4.6 mm., or virtually the same as in osilae. On the other hand, avail- able topotypes of lutescens are darker throughout than the type and topotypes of osilae. Pearson, nevertheless, regards the two named forms as subspecifically inseparable and I accept this solution. Ac- tually, lutescens is an intergrade between osilae and phaeus and could just as well be referred to the one subspecies as to the other. Measurements. — See Table 50. Sol 83 Z > bo rt etf-w §•§ N 5 a. H -= £ -r U PH id -*f c IO .X ,X "3 >X "3 1O 1O 1 "1 "1 I "1 I 1 1 O ^ OlO 1O rHCOOCM^OrH rH UJ 3 IO I •T co t>" ' d id J2- *d id idid^id^JS- i£- "•* ^oj^oio^iococor-rco^rHco^ioio co id" 5 10 10 1 0 rn ' D1O1O1O1O1O1O1O1O1O1O1O1O1O1O id 1O 1O T}< T}< t~ OO 1O _. ^ ,_ O t^ 1O • CO ^T • O • • • 1O • • 1O • CO 1O CO 1 *"! "~1 '"t l ' T1 H t— ^ O] CO 1O OO ^ 1O ^3 lO rH t^ O T^ "^J* 1O CO 1O 1O ""^ "^f* ^3 "^ 1O 10 3" c o~oot-'ooo'coid'coarG>'cM-^t''as^i| oo t-" f ;J<^Jj co co | 5 TC os co" oo T)^ c— >-j of OJ t> co 01 jj O t-O5 O oioOSOJOOO OS rH S3 CO C^J OJ CO CM CO CM c^l CO O4 Oj CO rtTrHrHKioco-^'ioco'cooo't--'coid'coco oo cT 0 OO CO < J OJ Ol < 1 CO CM. -] r-T ( g CM_ < H rH OJ" ( •J OJ OJ < 10 D t-^ g o^ -T 10 id" ^ OJ OJ rH •5 CO" g co lOOOOSrHOSOiCMOrHOOOOOSOOrHO O rH MOjoacocMOJcococoojojojoacoco co co co t- oo 10 ,, os os 10^° 1—l O to to i ^}* 1O t>* 1O OS co" 1O OO ^ OO |O^Os"b- 0]t-"cr US ] -*f 00 \ 'lOCOCO 1OCOT)< CO'lO IO co 10 t> oo OJ w "^ °° O3 OJ OJ | OJ_ OJ_ 'Hill M 10 ^-* ojrH •>-|OO!:"!£]CM H. ^1 2J-C<1 ^^ c^^,5iSJ-cf id"|-^<-^oi rHo"r-T r-TocM co of N Ol'OJOJOl CMCMCM OaCMOI OJ Ol co fT T)< 10 t- oo 03 0? oT P CO o t- £i t^T 10 -HT-^ooco"t-J^ •* co M OJ 1OCM OJ OJOjCMcM53-Si-$i-c<1 ^Tiot^'ioidt^'ioco'co-Hr-^osid'ooco us co" MCMOJOJOJCMCMOJOJOaOJOJOJOjOJ OJ OJ CO T)< U3 t- 00 •^ o oo t>- o co J* co 25 eo T* co rH t-xl rH '•^ rH rH rH 1 "~1 ® '"I 1 1 1 O CO O^-T o '"I OOrjJ'^lOOoO •H CO ^OJ CO CO CO '— ' OJ 1~l '— 1 rH o" 10 co" <1 CM CM ^3"coid"-«tt~Co"ojrH't-t~'o^CMC-^OOOJ CO O~ -ICOOJOJOJOJrHOJrHrHOCMrHCMCO OJ OJ rH D •"*" ^J rH co -3< 10 t- oo IO _, CO ^, CO t^ rH oj^ o oj^coojco rH1-1 .„ -u' I IS 1 •§ \ "a °§ ^§ ill 11 ill § §Il« iill l9 e8W»Nrt»«fi«sSF«-iaai3O«oJ W-etJ-^-^O-^tDOPHPH^-Ol-^pHpHPH PL, 382 5 jg -= I < •< >•» a S a ±i fS O CJ bfi *— >>-Q N a> - a> lO 10 "5 lO t^ U3 lO t- lO O5 T!« O> id id TI- co' -^1 Tf id i- o» oo »- C S 7 i- t- n MM M M W M M ^ o id id "*? oo id id Oi O o co t >» - l§i I§ 13 Ti — ' •« Ti a t— o • - co o >> 1-g c n "2 M "2 " S M rt i £ c 5 >> .£ c — — S3 ri bo be 5 bit- *f! rt » -s 2 — • "^- -y «M y. V, V, » » •/; _o QJ ^ rzz ^ rzz "^ •« ^ ex 7^ « 383 384 FIELDIANA: ZOOLOGY, VOLUME 46 Specimens examined. — 128. ARGENTINA. — Jujuy: Leon, 1 mile west, 4 (MVZ); mountains west of Yala, 4 (CNHM). BO- LIVIA.— Cochabamba: Choro, Ayopaya, 5 (CNHM); Colomi, 2 (CNHM, 1; MACN, 1); Punata, 4 (MVZ); Tiraque, 7 (MVZ); Vacas, 3 (CNHM). La Paz: Achacachi, Omapuyas, 1 (USNM). Tarija: Tarija, 10 miles NW, 1 (MVZ). PERU.— Cuzco: Huara- condo, 3 (CNHM) ; Macchu Picchu, 3 (USNM) ; Ollantaytambo, 1 (USNM); Tocopoqueyu, Occobamba Valley, 2 (CNHM, 1; USNM, 1); Hacienda Urco, 4 (CNHM). Puno: Asillo, 3 (AMNH, the type; MVZ, 2) ; Arapa, 3 (MVZ) ; Azangaro, 1 (CNHM) ; Hacienda Cala- cala, 10 (MVZ) ; Chucuito, 5 (CNHM) ; Juli, 3 (MVZ) ; Huacullani, 3 (CNHM) ; Occomani, 2 (CNHM) ; Hacienda Pairumani, 13,500 feet, 4 (MCZ); Hacienda Pairumani, 13,000 feet, 16 (MVZ); Hacienda Pairumani, 12,800 feet, 2 (MVZ); Pomata, 7 (MVZ); Puno, 1 (CNHM); Puno, 5 km. W, 1 (MVZ); Puno, 15 km. W, 3 (MVZ); Santa Rosa de Ayaviri, 2 (MVZ) ; Tirapata, 1 (AMNH) ; Hacienda Umayo, 3 (MVZ); Vilque, 2 (MVZ); Yunguyo, 12 (CNHM). Phyllotis osilae phaeus Osgood Phyllotis phaeus Osgood, 1944, Field Mus. Nat. Hist., Zool. Ser., 29: 193. Sanborn, 1950, Mus. Hist. Nat., "Javier Prado," (A), Zool., no. 5: 5— PERU: Puno (Limbani); BOLIVIA: La Paz (Pongo; Alaska Mine; Rio Aceramarca). Phyllotis osilae phaeus, Pearson, 1958, Univ. California Publ. Zool., 56: 427 — taxonomic revision; distribution (for individual collecting localities see p. 254). Type. — Male, skin and skull, Chicago Natural History Museum no. 53177; collected 23 February, 1942, by Colin C. Sanborn. Type locality. — Limbani, eastern slope of Cordillera Oriental, Puno, Peru; altitude, approximately 3500 meters above sea level. Osgood gives "about 9,000 feet" as the altitude, but this may have been an educated guess. Sanborn collected in the immediate vicin- ity of and above Limbani, which is at approximately 11,000 feet. Distribution (figs. 70, 74, 79). — Highlands of southern Peru and northwestern Bolivia in the upper Rios Inambari and Beni (Rios Mapiri-Coroico-Boopi) drainage systems; altitudinal range, 3400 to 4300 meters above sea level. The range occupies a narrow belt of heavy rainfall on the Amazonian side of the divide of the Cordil- lera Real. Characters. — Darkest member of the Phyllotis darwini group; dorsal surface of body heavily lined with dark brown of guard- HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 385 hairs; a blackish mid-dorsal stripe often present; lateral line poorly defined or absent; underparts from pale gray faintly washed with buffy to plumbeous heavily washed with brownish; tail usually longer than head and body combined, never wholly bicolor, pencil and at least terminal one-half inch of underside brown like upper side, sometimes nearly entirely brown. Variation. — The type series of phaeus from Limbani, Puno, con- sists of 8 specimens including the type (not 9 as stated by Osgood) representing two color phases. Four individuals are extremely dark with plumbeous underparts heavily overlain with brown, tail nearly uniformly brown; the remaining four are less brown, with under- parts dark gray thinly washed with brown, tail bicolor except at tip. • Four of the specimens were taken from 26 September to 1 October, 1941, the others from 22 23 February, 1942. Both color phases in prime to moderately worn pelage are included in both seasonal crops. A topotype series of 4 specimens collected by C. B. Koford, 11 August to 24 October, 1951 at 3530 meters, also includes both color phases and pelage types. A fifth specimen, taken 24 October, 1951, at 4060 meters, is quite like others from the lower altitudes. All five are practically indistinguishable from the type series except for their longer tails. In the 5 topotypes, the means and extremes of the tail relative to combined head and body length are 115% (104% 127%) and in the type series, 106% (97%-114%). A thin brown median ventral line is present on the tail of all but one of the Limbani speci- mens and also of some individuals from Pongo, Bolivia and sporadi- cally on the tails of other members of the P. darwini group. The palate of the Limbani series is short, its posterior border being on a plane with the posterior borders of m:i :t in nearly all samples. The posterolateral palatal pits are posteriad in all five specimens of the topotype series, in six of the type series, and posteriad on the right side and anteriad on the left side in the two remaining specimens. In all, the bullar tubes are short, the proximal ends of the nasals terminat- ing behind the frontomaxillary sutures, the mesopterygoid fossa well opened and the median spinous process usually absent. Seventy-nine specimens from Pongo in the Cordillera Real, La Paz, Bolivia, collected by G. H. H. Tate, agree with the less saturate color phase of the Limbani series. Their well-defined underparts vary from soiled white to nearly plumbeous; a more or less defined buff or ochraceous pectoral streak is present in about two-thirds of the specimens; in one, the streak is white. Specimens from Rio Aceramarca and Alaska Mine agree with those from nearby Pongo. 386 FIELDIANA: ZOOLOGY, VOLUME 46 Four specimens from Tacacoma, La Paz, are paler but each is marked with a well-defined black median dorsal band. Three indi- viduals from Sorata, farther south, are palest, with mid-dorsal sur- face of body lacking a dark line, and tail sharply bicolor for nearly entire length. In the La Paz material, the posterolateral palatal pits occupy all possible positions relative to the anterior border of the mesoptery- goid fossa. In some, the pits are anteriad, in others posteriad. In still others the pit or pits of one side of the palate may be anteriad, while on the other side they are posteriad to the anterior border of the mesopterygoid fossa. The mesopterygoid fossa may be wide or narrow, the median palatal spine present or absent. The nasals terminate behind the fronto-maxillary suture in all except two indi- viduals where they are on a line with the suture and in one where they terminate in front. Taxonomy. — The extremely dark Phyllotis osilae phaeus is the most differentiated member of the group. The separation of south- ern Peruvian phaeus from pale P. osilae osilae is sharp. In northern Bolivia, however, less saturate individuals of phaeus from Pongo and nearby localities in La Paz, merge with topotypes of lutescens Thomas from Choro, La Paz. Pearson (1958, p. 426) refers the type and Choro series of lutescens to typical osilae and his classification is followed here. The same material, however, could just as well be treated as extremely pale representatives of phaeus. In this case, the name lutescens would take precedence over phaeus. Sympatric and quasi-sympatric populations of P. osilae phaeus and P. darwini posticalis are easily distinguished from each other by a number of mensurable and cranial characters, as well as the marked difference in color. This is in sharp contrast with the P. osilae osilae- P. darwini rupestris relationship where differences between sympatric populations are, on the average, not as clear as between allopatric populations. Contact between posticalis and phaeus is the closing of the ends of a broken chain which began with the former and ended with the latter after separation between intermediates osilae and rupestris took place (fig. 76). Measurements. — See Table 51. Specimens examined. — 109. BOLIVIA. — La Paz: Alaska Mine, 5 (AMNH); Pongo, 79 (AMNH); Rio Aceramarca, 5 (AMNH); Sorata, 3 (AMNH); between Sorata and Tacacoma, 2 (AMNH); -* t- *i oo oo 10 co 01 ll SB'S o So O ^ ~j 00 P 0 T3 ~°"$ ». id ^ »d "i I *° f O > ^JiS IO *G ~ -*f co oj co •a T! ~ o id & "t "1 id, id id c — £ co t- CO CO i-7 •*?" t- co CD" co' p a id id id id 10 10 id id id id 0 0 — . "O rt co co »-H — f» ^ co q IO CO Ol ^ _« co S2 10 IO O "^J* 2 ^: "S-c co 2 ~ »« CO f* CO — — — fH > H 11 T ^ ^-^ 1 10 CD OS t^ bfl O id id id ^ 10 Tf t ^ >. . — N-° 10 OS OS oT co" ^ »-< os oo co T3 r c id id id IO IO "* co id id id O r> — — 'H B _c o co O -~^ "* t- ^JJ 10 co ^H C ~ 13 •o "* 10 -H "" o oj co o-i co co co •o s r' c | 1 \ I f O Oe bfl O o" ffs co co ^ •• • — T3 S (D O 3 — 00 OS o 00 Si " ^l OS O OS '3 i - EH C eg " "rr> us CO eo CO CO OC os t- t- oo H 1 - a c« « | os' o co CO O 05 t~ CO (M *s S o r , C ' .3 °? 7 V ~ ' — ' 4s "I S M 2 " r T3 r £ _5 o .2 T?" CO Ss oT r i" I O — -r T3 •C OJ • — •—1 1 ' 1 •" c C _Q T3 ^ •^ C3 5 Zf C ctf t3 s) TJ X T3 OS •S 1 T3 03 w B JS "o ' •*j _) • ' • OJ . 1 ~ 0 0 1C [ «! CO f- ^ — vC 3 s. £, ^ 10 co 01 CO ^-1 CO o w* rt -/. 09 03 co 2 o" o i-H Tf •Tt co io 10 Ol "TT ed "o *o EH ' = f^ 1*-t i .[ ( — fc H Ol oo t~-* oo" "-1 Ol Ol •-• co 01 Ol Ol O o T) V tT a 0) > . O) B ^ c ^1 00 ^f os -r o t~ 08 t : o CO '•'• 01 co co c c _E ^^ >, ^ ^ o t>- IO •** Ol S o M s 8 M O CO J £ o ^ P 03 oj co 01 4? "= e3 ja i _i "O « 2 CO ^1" os 10 o 0 — .- .1 \ la '3 i 'S tn 1 w' 00 eo 7 1 10 of -^ Ol Ol O -t t^ IO Os Ol •-> Ol -H 2 cc ^ ^ £ 73 P B r :• 0) .£ 1 gji H M jl Limbani1 ] •^ p S d 8 « 1 § SB I— 1 -** O Aceramarca 1 Alaska Mine ] Pongo c? cf ] Pongo 9 9 1 "s. S ® -M rf £ 'C ^J ^ liiiH a « a so ^ >. £ >• c s H *s E- HH w "3 < "-: 'B i a to e • — • 'C 01 oi (M ^> CTS — B < ~ >> •o o J2 -C tO t— -t <3i O O O-J O C'l *M 1 C ;a o a o .S H b T3 i- ^ rt .2i J? > cd i. •o -a ww slSI co 2 E-1 H :=• o 3 .- cr o c * 8 — rt < > 3 1- »3 O> 5 O- 389 390 FIELDIANA: ZOOLOGY, VOLUME 46 ever, I have examined in the American Museum of Natural History two specimens including a Juvenal from Tafi del Valle (2250 meters) that are markedly paler than four warm brown adults from the same named locality but at a higher elevation (3000 meters). The pectoral streak is barely indicated in both, the position of the posterolateral palatal pits anteriad to the posterior border of the palate in the juvenal but not certainly determinable in the damaged skull of the adult. Possibly, these specimens would be better referred to P. darwini rupestris. Measurements. — See Table 52. Specimens examined. — 12. ARGENTINA. — Catamarca: Acon- quija, 3 (CNHM, 1 ; MACN, 2) ; Rio Vallecito, Andalgala, 2 (MACN) ; Cerro San Javier, 1 (MACN); Tafi del Valle, 2 (AMNH); Tafi del Valle, above, 4 (AMNH). Phyllotis osilae nogalaris Thomas Phyllotis nogalaris Thomas, 1921, Ann. Mag. Nat. Hist., (9), 8: 611. Phyllotis osilae nogalaris, Pearson, 1958, Univ. California Publ. Zool., 56: 427 —ARGENTINA: Jujuy (Higuerilla) ; taxonomy. Type. — Male, skin and skull, British Museum (Natural History) no. 21.11.1.22; collected 29 May, 1921, by Emilio Budin. Type locality. — Higuerilla, Jujuy, Argentina; altitude, 2000 meters above sea level. Distribution (figs. 70, 74, 77) . — Known only from the type local- ity on the eastern slope of the Andes. Characters. — Known only from the type (and topotype) the original description of which follows: "A large species of rather dark colour. "Size large, only exceeded by Ph. magister. General colour of the usual grizzled greyish, but darker than usual, about as in dark specimens of Ph. tucumanus. Under surface dull whitish, slightly tinged with buffy, which may form a definite patch on the chest. Ears rather smaller than usual, dark brown. Hands and feet white. Tail long, blackish above and at the end, dull white for the proximal half below. "Skull nearly as large as that of magister, its upper outline evenly convex, the middle part of the skull high. Interorbital region very narrow, with well-marked vertically projecting edges, which add to the vertical height of the skull; these, however, have no resemblance to the divergent horizontal beads characteristic of the HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 391 genus Graomys. Zygomatic plate with sharply cut straight front edge. Bullae proportionately rather small. "Dimensions of the type: — "Head and body 147 mm.; tail 164; hind foot 30; ear 23. "Skull; greatest length 35; condylo-incisive length 32.5; zygo- matic breadth 19; nasals 14.5 x 5; intertemporal breadth 3.3; breadth of brain-case 15; height of crown from alveolus of m:t 10.6; palatilar length 16; palatal foramina 8.5; upper molar series 5." In addition Pearson (1958, p. 427) noted that the type had "small bullae, long tubular bullar tubes, lateral [i.e., posterior] postero- palatal pits, and nasals projecting behind premaxillae as in [Phyllotis] osilae tucumanus." Taxonomy. — The characterizations and geographic positions de- tailed above indicate and Pearson (1958, p. 427) affirms that the "two available specimens [of nogalaris] are big, old dark members of the osilae group" and that their great size "justifies their separa- tion as species." Remarks. — According to Budin, the collector, this is a "rata de los nogales," or rat of the walnut groves. Measurements. — See Characters above. Specimens examined. — None. Phyllotis micropus Waterhouse Mus micropns Waterhouse, 1837, Proc. Zool. Soc. London, 1837: 17. Water- house, 1839, Zool. Voy. "Beagle," Mamm., p. 61, pi. 20 (animal), pi. 34, fig. 13 (molars)— ARGENTINA: "interior plains of Patagonia, in lat. 50°, near the banks of the Santa Cruz." Mus (Ab[rothrix]) micropns, Waterhouse, 1837, Proc. Zool. Soc. London, 1837: 21. Hesperomys micropus, Burmeister, 1879, Descr. Phys. Rep. Arg., 3: 217— ARGENTINA: southern Patagonia. Euneomys micropus, Thomas, 1929, Ann. Mag. Nat. Hist., (10), 4: 39— ARGENTINA: Magallanes (Puerto Prat); Santo Cruz (Alta Vista, Lago Argentine). [Euneomys (Auliscomys)] micropus, Thomas, 1916, Ann. Mag. Nat. Hist., (8), 17: 143- classification. E[uneomys\ micropus, Thomas, 1919, Ann. Mag. Nat. Hist., (9), 3: 202 — ARGENTINA: Comodoro Rivadavia (Koslowsky). Phyllolis micropus, Allen, 1905, Patagonia Exped., Princeton Univ., 3, 1, Mamm., p. 60, pi. 12, fig. 13 (skull, molars), pi. 14, fig. 1 (molars) — ARGENTINA: Santo Cruz (Swan Lake; Basaltic Canyons; Rio Chico Valley); CHILE: Pacific slope. Pearson, 1958, Univ. California Publ. 392 FIELDIANA: ZOOLOGY, VOLUME 46 Zool., 56: 452— ARGENTINA: Chubut; Neuquen; Rio Negro; Santa Cruz; CHILE: Aisen; Cautin; Chiloe; Llanquihue; Magallanes; Malleco. Phyllotis (Auliscomys) micropus micropus, Osgood, 1943, Field Mus. Nat. Hist., Zool. Ser., 30: 211— revision; ARGENTINA: Rio Negro (Bariloche, Nahuel Huapi); Neuquen (San Martin de los Andes; Sierra de Pilpil); Chubut (Valle del Lago Blanco); CHILE: Magallanes (Casa Richards, Rio Nirehuao; Laguna Lazo; Puerto Natales; Punta Arenas); Llanquihue (La Picada; Puella, Lago Todos Santos); Cautin (Lonquimai; Rio Lolen); Aisen (Rfo Coihaique); Malleco (Sierra Nahuelbuta). Euneomys micropus alsus Thomas, 1919, Ann. Mag. Nat. Hist., (9), 3: 302 — ARGENTINA (type locality, Maiten, Chubut, 700 meters). Thomas, 1927, op. cit., (9), 19: 653— ARGENTINA: Neuquen (San Martm de los Andes; Sierra de Pilpil) ; Chubut (Epuyen; Barranca, near Tecka). Thomas, 1927, op. cit., (9), 20: 201— ARGENTINA: Neuquen (Quilquihue). Auliscomys micropus alsus, Yepes, 1935, Anal. Mus. argentine Cienc. Nat., 38: 335— ARGENTINA: Neuquen (Collon Cura). Phyllotis (Auliscomys) micropus fumipes Osgood, 1943, Field Mus. Nat. Hist., Zool. Ser., 30: 214— CHILE: Chiloe (type locality, Quellon, Chiloe Island; Rio Inio, Chiloe Island). [Phyllotis (Loxodontomys)] micropus, Osgood, 1947, Journ. Mammal., 28: 172 — characterization. Type. — Adult, British Museum (Natural History) no. 55.12. 24.179; collected April 1834, by Charles Darwin. Type locality. — According to Darwin (in Waterhouse, 1839, supra cit.) the type was "caught in the interior plains of Patagonia in lat. 50° near the banks of the Santa Cruz." Darwin relates in Chapter 9 of his "Journal of Researches" (any unabridged edition) that the "Beagle" anchored within the mouth of the Rio Santa Cruz in April 13, 1834. On the 18th a party that included Darwin and Captain Fitzroy as commander, started up the river in boats. Toward the end of the month the expedition came within a few miles of the Cordilleras and then turned back. The type of micropus was taken at some point along the route, perhaps in the vicinity of the present locality of La Argentina, province of Santa Cruz, southern Argentina. Distribution (figs. 57, 89). — Southern Chile and southern Argen- tina from 70° W to the Pacific and from 37° S to the Straits of Magel- lan; not known from Tierra del Fuego; altitudinal range between sea level and 2000 meters above. External characters. — Size large, body heavy; tail always shorter than combined head and body length, color of upper side not sharply defined from under side, the terminal portion nearly uniform, pencil rarely developed; hind feet normal, soles lightly scutulated, naked except on heel, plantar tubercles as in darwini; ears small, length FlG. 89. Collecting localities of I'hyllntix micropitx. See following p:ige for explanation. EXPLANATION OF FIGURE 89 Phyllotis micropus: Collecting localities and collectors. Type localities in boldface. ARGENTINA (1-12) 1. Quilquihue, Neuquen. — E. Budfn at 800 meters. 1. San Martin de los Andes, Neuquen. — E. Budfn at 1400 meters. 2. Sierra de Pilpil, Neuquen.— South of San Martfn; E. Budfn at 1200-2000 meters. 2. Collon Cura, Neuquen. 3. Bariloche=San Carlos de Bariloche. 3. San Carlos de Bariloche, Rio Negro. — E. Budfn at 600 meters; J. L. Peters at 2500 feet. 4. Maiten, Chubut. — Type locality of alsus Thomas; E. Budin at 700 meters. 5. Epuyen, Chubut.— H. E. Box. 5. Leleque, Chubut (not shown, 35 kilometers southeast of Epuyen). — H. E. Box. 6. Barrancas, near Tecka, Chubut.- — E. Budfn. 7. Valle del Lago Blanco, Comodoro Rivadavia. — J. Koslowsky. 8. Koslowsky, Comodoro Rivadavia (567 meters). — J. Koslowsky. 9. Swan Lake=Laguna del Cisne, Santa Cruz. — O. A. Peterson. 10. Rfo Chico Valley, Santa Cruz.— O. A. Peterson. 11. Basaltic Canyons, Santa Cruz. — O. A. Peterson. 12. Alta Vista, Lago Argentine, Santa Cruz. — E. Budfn at 600 meters. 12. La Argentina, Santa Cruz. — C. Darwin. CHILE (13-26) 13. Araucaria forest, Nahuelbuta, Malleco.— H. E. Anthony and G. Ottley. 13. Nahuelbuta (Sierra de), Malleco. — C. C. Sanborn. 14. Pinares, Lonquimai, Cautfn. — D. S. Bullock at 1600 meters. 14. Rfo Lolen, Cautfn.— C. C. Sanborn at 3600 feet. 14. Araucaria forest, west of Lonquimai, Cautfn. 15. Peulla, Lago Todos Santos, Llanquihue. — W. H. Osgood; J. M. Schmidt. 16. La Picada, Volcan Osorio, Llanquihue. — C. C. Sanborn. 17. Quellon, Chilo6. — Type locality of fumipes Osgood; W. H. Osgood; C. C. Sanborn. 18. Inio (Rfo), Chiloe".— W. H. Osgood; C. C. Sanborn. 19. Casa Richards, Rfo Nirehuao. — See Nirehuao (Rfo). 19. Nirehuao (Rfo), Aisen.— W. H. Osgood. 20. Coihaique (Rfo), Aisen.— H. E. Anthony and G. Ottley. 21. Laguna Lazo, Magallanes. — W. H. Osgood; C. C. Sanborn; J. M. Schmidt. 22. Lago Sarmiento, Magallanes. — J. M. Schmidt. 23. Rfo Verde, Magallanes.— W. H. Osgood; C. C. Sanborn. 24. Puerto Natales, Magallanes.- — C. C. Sanborn. 25. Puerto Prat, Magallanes.— E. Budfn. 26. Punta Arenas, Magallanes. — W. H. Osgood; C. C. Sanborn; J. M. Schmidt. 394 HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 395 from notch one-fourth to one-third less than that of hind foot with claw; postauricular patches reduced or absent, pelage thick, lax, lusterless, the individual hairs finely crinkled; upper parts brown mixed with black and buff, sides slightly paler; underparts not defined from sides, dark gray or plumbeous with an ochraceous suffusion especially on anterior half and mid-line. Cranial characters (figs. 58-60, 90). — Interorbital region narrow, the sides concave or parallel, borders square or raised, but never ridged or beaded; upper outer angle of fron to-parietal suture with a well-developed burr; zygomata markedly expanded posteriorly, forming a triangular outline when seen from above; greatest zygo- matic breadth in adults approximately equal to distance between posterior tips of nasals and anterior border of supraoccipital (usually less in subadults); anterior border of zygomatic plate vertical or sometimes slightly concave, the upper corner rounded or square, never produced as a spinous process; incisive foramina compara- tively well opened; posterior border of palatal bridge with a short blunt median process, never square or rounded; bullae moderately inflated, their antero-posterior length, exclusive of tubes, less than, sometimes barely equal to, alveolar length of molar row; outer side 3f mandible with projecting capsule for encasing base of incisor. Dental characters (figs. 67, 68, 91). — Incisors extremely thick, Dpisthodont, ungrooved; molar rows parallel-sided or divergent pos- teriorly; eruption of third molar retarded as compared with that of P. darwini; upper first molar 4-rooted, lower 3-rooted; molars hypso- iont, crowns of uppers plane tending to laminate obliquely, those of owers plane, sometimes slightly terraced, and tending to involution (fig. I7,c-d); cusps ovate; paracone and metacone of unworn and moderately worn m1 ~* inflected forward, metaconid and entoconid 3f mr 5 directed backward, the primary folds forming an angle of ibout 45° to long axis of teeth; with wear, angle of primary folds becomes progressively wider, approaching a right angle; crown of moderately worn m'± with modified sigmoid pattern; minor fold rf ma moderately developed, of m- faintly indicated in unworn or slightly worn tooth; procingulum of m^ - well developed, the first primary fold persistent except in extremely worn teeth; postcingulum rf m^ ^ well defined in unworn teeth, the second secondary folds present as enamel islands; vestigial mesoloph sometimes present; minor fold of m.2 moderately developed, of m^ obsolete; postcingulum Df mT j well developed and persistent except in old teeth; crown pattern of mff sigmoid. FIG. 90.—Phyllotis micropus. Dorsal and ventral aspects of skulls (X 2). ,396 FIG. 91. — Phyllotix micropmt. Right upper and left lower molars of «, Juvenal; 6, subadult; c, adult. 397 398 FIELDIANA: ZOOLOGY, VOLUME 46 Comparisons. — The only comparable species within the geographic range of Phyllotis micropus is P. darwini. The two resemble each other superficially in form and size, general structure of skull, and gross outline of molars. Important diagnostic characters that permit ready separation of one species from the other are found in the length, color and hairiness of the tail, size of ears relative to body and hind foot, spread and outline of zygomatic arches, thickness of incisors, degree of development of procingula of upper molars and postcingula of lower molars. Thomas (1919b, p. 202) was so far misled by the superficial resemblance between P. micropus and Mus xanthopygus Waterhouse (a race of Phyllotis darwini) that he included the latter along with the former in the genus Euneomys. Phyllotis micropus differs radically in external and cranial char- acters from the slender, silky furred, white bellied, long hairy- tailed P. griseoflavus with divergent-sided supraorbital region. The differences between the two species were used by Osgood (1947, p. 173) to enhance the claims of micropus for subgeneric rank. Paradox- ically, the greatest resemblance between micropus and griseoflavus lies in the dental characters (fig. 67) described by Osgood as "unique" in the former. The development of the procingulum and persistence of the first primary fold in the second upper molar is about the same in both species. The forward inflection of the metacone is also similar in both but the paracone, always oblique in unworn to moderately worn molars of P. micropus, is not consistently flexed to the same extent in comparable molars of P. griseoflavus. Similar oblique positions of cusps and primary folds have been noted in unworn to slightly worn molars in some forms of Phyllotis darwini, particularly P. d. wolffsohni, in Phyllotis pictus and in all but greatly worn molars of Euneomys chinchilloides (fig. 123). In P. pictus the procingulum of m- is as well developed as in micropus; in P. darwini wolffsohni it is less well developed and in Euneomys the procingulum is considerably reduced, and obsolete or absent in the worn tooth. The incisors of pictus are orthodont, weakly grooved and more delicate than in micropus; in Euneomys they are opisthodont and strongly grooved, though quite as robust as in micropus. Both Phyllotis micropus and Euneomys chinchilloides are heavy bodied, lax furred, dark bellied, short tailed, and small eared. The tail of Euneomys, however, is shorter (always less than 60 per cent of combined head and body length in adults) and sharply bicolor throughout its length. Cranially, Phyllotis micropus and Euneomys are widely separated (cf. p. 496). HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 399 Variation, — Individual, local and seasonal differences in P. mi- cropus are slight. Geographic variation, if it exists, is not appre- ciable. The geographic area occupied by the species is comparatively small, its climate uniform. The left upper incisor of a specimen (CNHM 50615) from Punta Arenas was described by Osgood (1943, p. 211) as "distinctly grooved." From this, Osgood inferred that the relationship between micropus and the normally grooved-tooth pictus was intimate and he in- cluded both in his genus Auliscomys. This classification was later rejected by Osgood (1947, p. 172). An examination of the incisor in question reveals the "groove" as a traumatic condition. Habits and habitat. — Mr. 0. A. Peterson, who collected Phyllotis micropus in Patagonia, records (in Allen, 1905, p. 61) that "this is the most common species met with on the 'Rio Chico Cordillera, especially in the heavy grass near water; but was also caught in timber at some distance from water.' ' Allen (supra cit.) added that "it appears to have been rare further north in the more open country where Mr. Colburn worked, he securing only four specimens during six weeks of constant trapping." Judged by the nature of the collecting localities, Phyllotis mi- cropus is essentially a forest glade dweller, but maintains itself in high grass and brush on the banks of streams in deforested country as well. The retarded eruption of the third molar in P. micropus is generally characteristic of forest dwelling cricetines. In deforested areas, micropus and invading representatives of P. darwini may be found in the same locality but not necessarily in the same habitat. Taxonomy. — Loxodontomys Osgood, 1947, was proposed as a sub- genus of Phyllotis to contain only Mus micropus Waterhouse. This classification would be justified, as indeed it was, on the scale of values that determines each aggregation of subspecies and their synonyms as a "species group," and each well defined species as a subgenus or genus. The "Phyllotis group," as understood by Osgood, and by Thomas as well, is nothing more than the P. darwini complex of this paper. Graomys, as conceived by the same authorities, is merely P. griseoflarus. Paralomys is P. gerbillus. Rejection of this inflated classification eliminates the need for the name Lo.ro- dontomys and leaves its species occupying exactly the same phylo- genetic position relative to all other species of the genus Phyllotis. Euneomys micropus alsus Thomas was sunk into the synonymy of Phyllotis micropus by Osgood. No particular reason was given for this action, but material at hand shows no basis for subdivision o — IO H os " °° os *— ^ x — ^ j 0 c x •^t 1— 1 ^ X ;"5 t- id id_ i *? id_ » bcj2 1 x' x — id"«o o Q ?C c<; TjT 5 § c tc id ^ tc id jd^u 5 IO c^ 10 ~~ E i x cTt- -H ce~t- O^O Os't^oTu 5 o^ eg tc 10 o x" id (D 10 «0 50 10 X X — t- •— ' _w 0 CO uj 0 eg __ ^ • OS ~ •*•* x' ^ X S _ x 11 cs oT '"I 1 ac co x rc ^ - X x >S.c t-^ W5 c-' os' t-' x' y : «o t-' fsj ' ^ """^ x o ~* eg os CM •**• o t> " 5 OS X OS IO iC C— 1 t> X X O-' X OS X X X t- - t- cc x x t~ t- cs •U X X ^ £« (V +J ^ ^~ ;g eg rt r" • ^ >\i ^ ^* ^ "is Cg cc ^J cc ^s 10 cT ^* ! x eg eg ^ e j os =: — ^ o S =2 ^; r: - •i O X j* eg re" ce -*" OS O i— ' ^t T ^" re x x" tc cs re CO i oj ,_: _" „" cc cc cc c*? eo eg i— i eg - - eg cs re eg — r r^ eg re re re 0 LH ^i OJ ,_( " w °i CO XXX OS II 1 "o 1— 1 1 1— 1 T-H »-H T-H 1 1 1 CM os o •U — x eg C o X M ^ i-2 eg ••* os" ^ 10 « c •r c x x" ^ ^ x o cTx o eg ~5 S ^ ? S t~ •%! o j eg^io re re •J £ t~ os oT t- c^ t~-^ CD eg ^ TJ ?j c x" cs' »-T eg cs -T ^ Tt t-T : o o eg •— os o i ^ i "S ^ cs o «. LH as o •o — oo eg 1 i o s i -5 5 s JE ^S . •c * ' i-H " - i-H ^^ 2 5! fj t^ t-^ '"I 1 ; ^ os — " III a 0) g§ssss"?g T^t ^5 *\J 'VJ f^. ^- * oC "^* ^^ ^^ ^^ *^3 *-£? |^.| •H § S Locality RGENTINA o N U C eg-i. -g- 5s=i=ls c ^o Sc'3.a'5 c ^2 jS£c£co § x o = J -g £ S ^j QJ Q < ~ u '£ C = -2:^ ^ ^•:: K -3 > -5 " •" J S o o OQ J ^ _ 3 *J »— . . C c 0, pi, Ci _3 J L i-? J: c "5 L c s:~ ^ =5S^I^1 •^£ g^^^ > £i U (X, J Ci £ !3 ^i *•* ^ ^ 5 *c *E "o a; o> a> c. c. c. >>>>>> r- r- E- O 400 HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 401 of the species into races. The distinctive color attributed to alsus, as compared with Rio Chico, Santa Cruz representatives of mi- cropus, is on the same order as that which distinguishes one specimen from another of the same locality. Phyttotis micropus fumipes Osgood, from Chiloe" Island, is based on four specimens "all of them somewhat immature." Their char- acters are said to be "practically as in micropus except that the hands and feet are darker." The four members of the type series are better described as juvenals, the type being the oldest. In these, upper surface of feet is dark because the brown skin is thinly covered with short silvery hairs. Feet of mainland individuals of comparable age are similar. In most adults, however, the feet are more thickly haired, hence less brown in appearance. Measurements. — See Table 53. Specimens examined. —103. ARGENTINA.— Chubut: Valle del Lago Blanco, 2 (CNHM). Neuquen: San Martin de los Andes, 1 (CNHM); Sierra de Pilpil, 1 (CNHM). Rio Negro: San Carlos de Bariloche, 7 miles east, 3 (CNHM). Santa Cruz: Rio Chico, Cordil- lera, 5 (AMNH, 4; CNHM, 1). CHILE.— Aisen: Coihaique, 10 (AMNH). Cautin: Rio Loten, 1 (CNHM); Pinares, Lonquimai, 1 (CNHM); Araucaria forest west of Lonquimai, 1 (AMNH). Chiloe; Rio Inio, 3 (CNHM); Quellon, type of fu wipes Osgood (CNHM). Llanquihue: La Picada, Volcan Osorio, 1 (CNHM); Puella, Lago Todos Santos, 2 (CNHM); Rio Nirehuao, 24 (CNHM ). Magallanes: Laguna Lazo, 9 (CNHM ) ; Lago Sarmiento, 3 (CNHM ) ; Puerto Natales, 1 (CNHM); Punta Arenas, 21 (CNHM); Rio Verde, 7 (CNHM). Malleco: Araucaria forest, Nahuelbuta, Malleco, 3 (AMNH); Sierra de Nahuelbuta, 3 (CNHM). EXPLANATION OF FIGURE 9l> Phylhtix picln*: Collecting localities and collectors. Type localities in boldface. I'hyllolis pictm (1-35) PERU 1. Quilcayhuanca, Ancash. C. Kalinowski at 4000 meters. '2. Catac (Hacienda), Ticapampa, Ancash. C. Kalinowski at 4000 meters. Chipa, Pasco. H. Watkins at 12,400 meters. Carhuamayo, .Junfn. C. C. Sanhorn at 14,500 feet. San Bias, Junfn (approximately 4000 meters). J. Kalinowski (father of C. Kalinowski), at "18,000 feet," an impossible altitude for the region. 6. Junin, Junin. M. P. Anderson at 14,000 14,500 feet. 402 FIELDIANA: ZOOLOGY, VOLUME 46 Explanation of Figure 92 (continued) 1. Pachacayo, Junfn.— C. H. T. Townsend at 12,000 feet. 8. Lachocc, Huancavelica. — A. R. G. Morrison at 12,500 feet. 9. Yaurichucchu, Huancavelica.— C. Kalinowski at 4200 meters. 10. "San Jenaro," Santa Ines, Huancavelica.— C. Kalinowski at 4760 meters. 11. Polanco, Tambo, Ayacucho. — C. Kalinowski. 12. Occobamba Pass, Cusco.— E. Heller at 12,500-13,000 feet. 13. Ollantaytambo, Cusco.— E. Heller at 13,000 feet. 14. Huaracondo, Cusco.— E. Heller at 11,000 feet. 15. Ccolini, Marcapata, Cusco. — C. Kalinowski at 3900 meters. 16. La Raya Pass, Cusco.— E. Heller at 14,000 feet. 17. Puerto Arturo, Puno.— C. C. Sanborn at 13,750 feet. 18. Crucero, Puno. — H. H. Keays at 15,000 feet; P. 0. Simons. 19. Limbani, Puno. -C. C. Sanborn at 11,000 feet. 20. Picotani, Puno. — C. C. Sanborn at 14,000 feet. 21. Posoconi, Puno. — C. C. Sanborn at 13,500 feet. 22. San Antonio, Puno. — H. H. Keays at 12,600 feet. 23. Tirapata, Puno.— H. H. Keays at 16,000 feet. 24. Santa Lucia, Puno.— M. R. Portugal. 25. Occomani, Puno. — E. Zuniga at 12,700 feet. 26. Chuquito, Puno. -C. C. Sanborn at 12,600 feet. 27. Collacachi, Puno. -C. C. Sanborn at 12,800 feet, 28. Yunguyo, Puno. -C. C. Sanborn at 13,000 feet. 29. Huacullani, Puno. — C. C. Sanborn at 12,700 feet. 30. Pairumani, Puno. —0. P. Pearson at 13,000 feet, 31. Santa Rosa, Puno. -O. P. Pearson at 14,000 feet. 32. Cailloma, Arequipa. — C. C. Sanborn and J. M. Schmidt at 14,500 feet. 32. San Ignacio, Arequipa. —13,500 feet. 33. Puquio, Ayacucho. —13,500 feet. 34. Chalhuanca, Apurimac. —14,500 feet. BOLIVIA 35. Pongo, La Paz. — G. H. H. Tate at 12,000 feet. ADDITIONAL RECORDS FOR PERU LISTED BY PEARSON (1958, p. 449) Sibayo, Arequipa, 11,500 feet. Hacienda Chacayani, Puno, 13,200 feet. Can-can, Junfn. Juliaca, Puno, 12,600 feet. Galera, Junfn, 15,800 feet. Mazocruz, Puno, 13,000 feet. Hacienda Atocsaico, Junin, 13,500 feet. Potone, Puno, 15,800 feet. Maraynioc, Junfn, 15,000 feet. Puerto Arturo, Puno, 13,000 feet Tarma, Junin, 10,100 feet (CNHM). (CNHM). Ninacaca, Pasco, 13,500 feet. Puno, Puno, 12,500 feet. Arapa, Puno, 12,600 feet. Santa Rosa de Juli, Puno, 14,000 feet. Cotacollo, Platerfa, Puno. Tincopalca, Puno, 13,500 feet. Hacienda Calacala, Puno, 13,000 feet. Vilque, Puno. FIG. 92. — Collecting localities of the subspecies of I'hyllotix piciitx. See page 401 for explanation. 403 404 FIELDIANA: ZOOLOGY, VOLUME 46 Phyllotis pictus Thomas Reithrodon pictus Thomas, 1884, Proc. Zool. Soc. London, 1884: 457, pi. 43, fig. 2 (animal), pi. 44, fig. 19 (hind foot), figs. 20-21 (skull). Thomas, 1897, Ann. Mag. Nat. Hist., (6), 20: 550 (in text). Thomas, 1900, op. cit., (7), 6: 469 — Simons' Cailloma and Arequipa collection. Rhithrodon [sic] pictus, Thomas, 1893, Proc. Zool. Soc. London, 1893: 337 — PERU: Junin (San Bias). E[uneomys] pictus, Thomas, 1901, Ann. Mag. Nat. Hist., (7), 8: 254 —classi- fication. E[uneomys] (Auliscomys) pictus, Thomas, 1916, Ann. Mag. Nat. Hist., (8), 17: 143 — classification. Euneomys (Auliscomys) pictus, Thomas, 1920, Proc. U. S. Nat. Mus., 58: 232 —PERU: Cusco (Huaracondo; Ollantaytambo; Occobamba Pass; La Raya Pass). Auliscomys pictus, Thomas, 1926, Ann. Mag. Nat. Hist., (9), 17: 317— PERU: Junfn. Thomas, 1927, op. cit., (9), 19: 550 —selection of lectotype. [Phyllotis (Auliscomys)} pictus, Osgood, 1915, Field Mus. Nat. Hist., Zool. Ser., 10: 191 — type of subgenus Auliscomys. Osgood, 1947, Journ. Mammal., 28: 172— characters. Phyllotis (Auliscomys) pictus, Sanborn, 1950, Mus. Hist. Nat., "Javier Prado," (A), Zool., no. 5: 6— BOLIVIA: La Paz (Pongo); PERU: Junin (Carhua- mayo; Pachacayo); Huancavelica (Lachocc); Cusco (Ollantaytambo; Hua- racondo); Puno (Limbani; Puerto Arturo; Picotani; Posoconi; Chuquito; Collacachi; Occomani; Santa Lucfa; Yunguyo; Huacullani); Arequipa (Cailloma). Pearson, 1951, Bull. Mus. Comp. Zool., 106: 146— PERU: Puno (Pairumani; Santa Rosa). Phyllotis pictus Sanborn, 1950, Mus. Hist. Nat., "Javier Prado," (A), Zool., no. 5: 6 — decoloratus Osgood a synonym. Pearson, 1958, Univ. California Publ. Zool., 56: 448— BOLIVIA: La Paz; PERU: Ancash; Apurimac; Arequipa; Ayacucho; Cusco; Huancavelica; Junin; Pasco; Puno. Phyllotis boliviensis, J. A. Allen (not Waterhouse), 1901, Bull. Amer. Mus. Nat. Hist., 14: 44— PERU: Puno (Tirapata; San Antonio). Phyllotis (Auliscomys) decoloratus Osgood, 1915, Field Mus. Nat. Hist., Zool. Ser., 10: 191— PERU: Puno (type locality, Tirapata, 3860 meters alti- tude). Thomas, 1920, Proc. U. S. Nat. Mus., 58: 232— doubtfully dis- tinct from pictus. Type. — Male, in spirits, skull separate, British Museum (Natural History) no. 85.4.1.34; collected by C. Jelski; (lectotype selected by Thomas, 1927, supra cit.). Type locality. — Junin, Junin, Peru; altitude about 4300 meters above sea level. Distribution (figs. 57, 92) . — Highlands of Peru and Bolivia, from the department of Ancash, Peru, south into the department of La Paz, Bolivia; altitudinal range from approximately 3400 to 4700 HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 403 meters above sea level and possibly higher, to limits of perpetual snow. The locality "San Bias, Cordillera, 18,000 feet [ = 5600 m.]" recorded for the species by Thomas (1893, p. 337) is almost certainly the San Bias west of Lake Junin on the Junin to Huallay road, altitude about 4000 meters. External characters. — Size as in the larger species of Phyllotis; bicolor tail less than equal but more than one-half combined head and body length, tip not conspicuously tufted; hind feet com- paratively short, upper surface gray more or less washed with ochra- ceous, plantar pads normal, soles finely scutulated (fig. 3) ; ears small, inner posterior surface ochraceous, pale postauricular tufts usually present; head and, usually, shoulders grizzled, contrasting with brownish back and sides of trunk; rump, base of tail and outer sides of thighs with more tawny or ochraceous than back; prominently projecting guard hairs of rump tipped black, or brown to pure white; ochraceous lateral line not always well defined, frequently interrupted in mid portion; underparts gray to dirty white with dark basal portions of hairs showing through; a small ochraceous pectoral streak sometimes present. Cranial characters (figs. 58-60, 93). — Supraorbital region of skull narrow, the sides concave or parallel-sided, edges square or slightly raised, never ridged or beaded; zygomata moderately expanded or nearly parallel-sided, their greatest breadth approximately equal to, usually greater than, distance between posterior tips of nasals and anterior border of supra-occipital; anterior border of zygomatic plate plane, slightly concave or slightly convex, the upper anterior corner rounded or pointed but not produced as a spinous process; interparietal well developed; pitted posterolateral palatal excava- tions usually large and deep; bullae little inflated, their an tem- pos terior length, less tubes, shorter than alveolar length of molar row; outer side of mandible without capsular projection for en- casing base of incisor. Dental characters (fig. 67,e). Incisors moderately heavy, generally orthodont, varying from slightly opisthodont to slightly p rood on t, the anterior face of each marked by weakly defined grooves with the outermost barely visible to the unaided eye. Molar rows parallel- sided or slightly divergent posteriorly; upper and lower first molars 4-rooted; molars sub-hypsodont; unworn crowns of upper molars slightly terraced and tending to laminate, little to moderately worn crowns plane and tending to involution and development of a modi- fied sigmoid pattern; cusps more or less ovate in outline and, in 406 FIELDIANA: ZOOLOGY, VOLUME 46 FIG. 93. — Phyttotis pictus. Dorsal and ventral aspects of skull, showing variation in size. (Slightly more than X 1M-) unworn or slightly worn teeth, with a forward flexion; procingulum of m1 biconulate in unworn, uniconulate and ovate in outline in slightly to extremely worn tooth; procingulum of m2- well developed, the first primary fold persistent in all but considerably worn tooth; minor fold of m^ obsolete or poorly developed; vestigial mesoloph often present in m^2-; postcingulum of m^ relatively well developed, HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 407 of m5 weak; first minor fold of m? persistent in well worn tooth; crown pattern of m5 sigmoid. Comparisons. — Phyllotis pictus is distinguished from the sym- patric species sublimis and boliviensis by generally more saturate coloration, by relative length of tail, hind foot and ear; from dark representatives of the Phyllotis darwini complex of the same geo- graphic area, by lax pelage, ochraceous base of tail, grizzled head contrasting with brownish trunk, unpencilled tail less pilose, less sharply bicolor, and consistently shorter than combined head and body length. Phyllotis micropus, the only other species comparable in size and appearance is allopatric, more or less uniformly dark brown, and without the markedly contrasting colored portions of the body that distinguish P. pictus. Cranially, pictus differs from boliviensis chiefly by smaller bullae, from sublimis by overall larger size, from the P. darwini complex by more widely expanded zygomatic arches, more angular braincase. Molars of pictus show the same tendency, though to a lesser degree, for forward flexion of the cusps, noted in Phyllotis micropus. Other- wise, the cheek teeth are like those of other species of the genus characterized by well developed procingulum and persistent first primary fold in m^. Minor fold of m1 is also persistent in pictus except in the extremely worn tooth. Variation. — In this, and other species of Phyllotis, there are the usual intergrading dark and pale color phases. The well-marked grizzled anterior portion of the body may be restricted to the muzzle or may extend over head and anterior half of trunk; in some speci- mens this area may be marked by large whitish patches. As in Phyllotis darwini, a pectoral streak is sometimes present, but there is no consistent relationship between the marking and position of the posterolateral palatal pits. In a series of 11 specimens from Tambo Polanco, Ayacucho, 5 show the pectoral streak but with the posterolateral palatal pits ranging in position from behind, to in front of, the posterior border of the bony palate. Position of the pits in the remaining, unmarked, individuals is similarly variable. A trio of specimens from two localities (Quilcayhuanca; Hacienda Catac) in the department of Ancash, with rump and base of tail hardly differentiated from back, are darkest of the species. Groov- ing of their incisors is hardly visible to the unaided eye. The two specimens from Quilcayhuanca are extremely old. A skin only from Limbani, Puno, a locality in Amazonian drainage, agrees with Puno material from the Lake Titicaca basin except that the upper surface 408 FIELDIANA: ZOOLOGY, VOLUME 46 of its hind foot is nearly uniformly ochraceous. A young individual from Puerto Arturo, Puno, near the Amazonian side of the divide, shows nearly as much ochraceous. The same character appears in specimens from Junin and Cailloma. A subadult from Pongo, Bo- livia, Amazonian drainage, is indistinguishable from the young Puerto Arturo individual. In contrast, specimens of Phyllotis osilae from the same areas in the Lake Titicaca and the Amazonian drainage, represent, respectively, the well-marked subspecies, pale P. osilae osilae and blackish P. osilae phaeus. Habitat and habits. — The following account is from Pearson (1951, p. 146): " A[uliscomys\ pictus seems to prefer grassy places, especially those near water, but it also lives in stone walls and in places far from water. One was caught in an unoccupied Indian hut in which Akodon jelskii pulcherrimus was also living, but its usual associates are Akodon boliviensis and A. amoenus. It is active day and night. I was unable to distinguish between the skull fragments of pictus and sublimis in the Pairumani owl pellets, but these two species combined provided the owls with 54 of their 153 meals. By weight, this would be the most important item in the diet of the owls. "None of the females caught between July 14 and 30 at Pairu- mani and Santa Rosa were pregnant, and none of the 8 males had testes larger than 4 mm. A specimen in greyish juvenile pelage caught on December 22 suggests that breeding begins in September or October as in many of the other species of mice in the region. The ratio of sexes was 17 males to 16 females." A female taken April 13, 1915 by Edmund Heller in Huaracondo, Cusco, Peru, is labelled as having been gravid with 5 embryos. Taxonomy. — Auliscomys Osgood was proposed as a subgenus of Phyllotis to contain pictus Thomas (type), boliviensis Waterhouse, and sublimis Thomas. The original definition of Auliscomys was based on a comparison of pictus with Phyllotis darwini and Euneomys and does not apply in every case to included species boliviensis and sublimis. The upper incisors in Auliscomys, said to be marked by "slight but distinct grooves," refers to pictus but not always to 60- liviensis and sublimis. The greater hypsodonty attributed to Aulis- comys as compared with P. darwini, is not confirmed in present material. Retention in m^ of the first primary fold ascribed by Osgood (in different terminology) to Auliscomys, serves to distin- guish pictus and boliviensis from some members of the P. darwini complex but not from other species of the genus Phyllotis. Later, .^ — . 01 Q> V9 • J2 ° £ ^ i e? ^i °°. 30 «> Ol" oq *? i (3 "? i 1 1 U3 i > be js oo •* oo" oo »« us co oc co T 5^ -r ^-1 £%»«§! ~" S" -H 01 ^ (0 Tf w"^^ ; 1/3 U3 1/3 W njf -4 OS ' ' T i ~ i T i OO U3 5 10 "• _o 1/300 oo* 5J o7 co ^4 p «-> J2 t— • t^- t— t- d "— < t-' 0 S °. °°. « . I 1 > O ~H t- 1 1" Mk tJ v t » t--' «D t-J lO (^ to «£5 ^ «C M ^- ^* H »"^ •o ^^ *^ ^* *~^ ^^ co eo«oo jwoTwc >»"c >^ CD t- t- b- t ~ oo oo t~- t~ to t- C~. {£> « 5 «D <£> <£> t- - «£> « 5 «D co 4J "*1* »H Ol Ol U3 U3 05 1 ^ o ^T 0? 01 oo oo ^ I O EH 1^ 5' " C "5 <=> . — ".£ —i o < vf ^ 1 1 d t- c d oi co eg 3 O U5 d CO 1 o oi 0} ^ "3 °^. S2-s 1°°. 5i Dt-S"»« >I% eg f o s. M ddoo'doio^oooiooioioooooioioioioo'o x 2 1-H I— 1 2. | "S *J £ c^ i? f d. K o, 1 CO O 001 f "o e-j^ oc eg^ e^ 1 V. •<-> e i eg .-i " ~« co co ^H ^H o ^'o' "III I I c 0) S t- ^i o^ Oi U3 0) *•• t— ' •« ^H CO t— 00 t*^ 1/3 easur o eg 'H £Q *^3 eg * * *jL 2 eg c-J i oo" eg U3 S 1 .S eg_ ^H »5^ eg eg £J_ e^j »o eg 1/31/3 ^i O-l D t- S 00 u i ^. ^H oo j t < 71 0^ Oi 1/3 rt^ 1/3 - Tf H 13 t- « < EH ° ua ' •o"t~ "5 OS ^f t^ Ol Ol Ol 1 oo ^*~ co ^e» co ^ OO Oi e- • ^4 oo c 2 S ^ S ! »S S5 ol SS 01 oo o o t^ oo 01 r - 0? c i oo : ^ •g •o 5=.! «r ^H OJ Ol U3 ^ U3 ~ oo 3 ® eg eg eo •» co - -O CO ' co co c^> , t ~ 1 5 oc 2| •* ^ ^ fi i T *] t~ r? e-i t- > •* 1 „! ^1 o| -L; C CD § od 3 O __._•• - «c 2 3 i. 15 >> « o ^^ J5 § o ii1 .« g • c S 5 3 rt •* i -3' M O CC O "r- e* 3 O TJ .^ c ^ *-* C o * n *t* *^ w ^< J O i c* • O u ed M M £ ,-^1 *3 C ,2 J*j C **^ — ^ C CQ 2» •• J •= tS c c * * c i "3 3 — 8 ^ ? 2 8 § •« ^ J - 3 co Cu DL, w O O >• 409 410 FIELDIANA: ZOOLOGY, VOLUME 46 Thomas (1916a, p. 143) reclassified Auliscomys as a subgenus of Eune- omys. He distinguished Auliscomys by the "mainly vertical" portion of the maxillo-premaxillary suture from Euneomys in which the suture is "bowed forward below." The same character separates Phyllotis from Euneomys and Thomas proved nothing by the trans- fer of Auliscomys. Thomas modified his views later (1926a, p. 317) and proposed that Auliscomys and other subgenera of Euneomys, be treated as full genera. This arrangement, he averred, was "far more convenient both for labelling and reference." Osgood (1947, p. 172) with more concern for biological realities restored Auliscomys to the genus Phyllotis. Basing comparisons on a single topotype of true pictus, Osgood first described Phyllotis (Auliscomys) decoloratus as "smaller and paler" than the single topotype of P. pictus with which it was com- pared. Additional material has shown these characters to be illusory. The type of decoloratus, judged by measurements, is a Juvenal, and all but one of the remaining 10 specimens of the original series of decoloratus (Tirapata, 4; San Antonio, 5; Crucero, 1) are juvenals and subadults. Measurements. — See Table 54. Specimens examined. — 127. BOLIVIA. — La Paz: Pongo, 1 (AMNH). PERU.— Ancash: Quilcayhuanca, 2 (CNHM); Haci- enda Catac, Ticapampa, 1 (CNHM); Junin: Junin, 11 (CNHM); Carhuamayo, 9 (CNHM); near Junin, 2 (CNHM); Chipa, 3 (AMNH). Huancavelica: Lachocc, 2 (CNHM); "San Jenaro," Santa Ine"z, 4 (CNHM); Yaurichucchu, 1 (CNHM). Ayacucho: Polanco, Tambo, 11 (CNHM); Cusco: Huaracondo, 1 (CNHM); Ollantaytambo, 2 (CNHM); Ccolini, Marcapata, 2 (CNHM). Are- quipa: Cailloma, 19 (CNHM). Puno: Limbani, 1 (CNHM); Puerto Arturo, 3 (CNHM) ; San Antonio, 5 (AMNH) ; Tirapata, 4 (AMNH) ; Crucero, 1 (AMNH); Picotani, 10 (CNHM) ; Posoconi, 12 (CNHM); Santa Lucia, 1 (CNHM) ; Occomani, 4 (CNHM) ; Hacienda Colla- cachi, 6 (CNHM) ; Chuquito, 1 (CNHM) ; Yunguyo, 6 miles S, 6 (CNHM); Huacullani, 2 (CNHM). Phyllotis boliviensis Waterhouse. (Synonymy under subspecies.) Distribution (figs. 57 and 94). — The altiplano of southern Peru, northern Chile, and northwestern Bolivia; altitudinal range from approximately 3500 meters above sea level to near limits of per- petual snow. HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 411 75 70 65 10 15 20 Phyllotis boliviensis flavidior boliviensis £ i.ll& (tf 25 15 20 25 75 70 65 FIG. 94. — Collecting localities of the subspecies of Phyllotis boliviensis. See explanation below. EXPLANATION OF FIGURE 94 PhyllotiK boliviensis: collecting localities and collectors. Type localities in boldface. Phyllotis boliviensis flavidior (1-5) PERU 1. Bateas, Cailloma, Arequipa. B. Hunt at 4500 meters. 1. Cailloma, Arequipa.- C. C. Sanborn and J. M. Schmidt at 14,500 feet; P. O. Simons. 412 FIELDIANA: ZOOLOGY, VOLUME 46 External characters. — Moderately large; tail from one-half to nearly as long as combined head and body length, its upper sur- face ochraceous with terminal portion darker, underside whitish or buffy basally becoming ochraceous terminally, tip un tufted; hind foot comparatively large and strongly developed, upper surface gray or buffy, an ochraceous metatarsal patch often present, plantar pads normal, soles coarsely granulated; large ears with prominent ochra- ceous preauricular tufts and smaller whitish or buffy postauricular tufts; upper surface of head and body with coarse mixture of gray, brown, and buff or ochraceous, sides paler, muzzle whitish; ochra- ceous lateral line not always well developed; underparts whitish or pale gray, more or less washed with buff; an ochraceous pectoral streak sometimes present. Cranial characters (figs. 59, 60, 95). — Skull more angular, less elongate than in other Phyllotis; supraorbital region as in P. pictus, zygomata more bowed, their breadth approximately equal to, or greater than, distance between posterior tips of nasals and anterior border of supraoccipital; anterior border of zygomatic plate more or less plane, never markedly concave, the anterior corner usually rounded, sometimes pointed but not produced; interparietal better developed, especially longitudinally, than in P. pictus; bullae well inflated, thin-walled, their antero-posterior length, less tubes, sub- equal to alveolar length of molar row; distance across mastoid proc- esses in fully adult specimens approximately one-half or more greatest skull length; outer side of mandible without projecting capsule for encasing base of incisor. Explanation of Figure 94 (continued). 2. Salinas, Arequipa. — C. C. Sanborn at 14,100 feet. 3. Arequipa, Arequipa. —P. 0. Simons. 4. San Antonio de Esquilache, Puno.— C. C. Sanborn at 15,000 feet. 5. Caccachara, Puno.— O. P. Pearson at 16,000 feet. Phyllotis boliviensis boliviensis (6-11) CHILE 6. Puna de Tarapaca, Tarapaca. — G. Mann. 7. ChoquelimpiS, Tarapaca.— C. C. Sanborn at 15,000 feet. BOLIVIA 8. Esperanza, Pacajes, La Paz. — F. Steinbach at 4200 meters. 9. Chuquisaca= Sucre, Chuquisaca. — Castelnau and Deville Expedition. 10. Potosi, Potosi (3945 meters).— T. Bridges; P. O. Simons. 11. Livichuco, Oruro. — P. O. Simons at 4500 meters. HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 413 FIG. 95. — Phyllotix boliviensia. Dorsal and ventral aspects of skulls, showing variation in size. (Slightly more than X 1 '2-) Dental characters (fig. 96).— Incisors generally proodont, some- times orthodont, their anterior face smooth or marked by faint shallow striae usually not visible to the unaided eye. Molar rows parallel-sided or slightly bowed; upper first molar with 3 or 4 roots, lower 3-rooted; hypsodont crowns plane or very slightly terraced in unworn upper molars, more persistently terraced in lower molars; un- worn or slightly worn crowns involuted, outline of cusps more or less FIG. 96. — Phyllotis boliviensis. a, right upper molars; b, moderately worn left lower molars; c, well worn left lower molars. 414 HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 415 ovate; modified sigmoid pattern of moderately worn m- becoming 8-shaped with increased wear; procingula of m4 * well developed, the first primary folds persistent in all but extremely worn teeth; minor fold of ma not developed; vestigial mesolophs often present in m- -; postcingulum of m2 well developed, of m^ reduced; minor fold of m.2 absent in moderately worn tooth. Comparisons. — In pale coloration, Phyllotis bolinensis resembles sym patriots P. sublimis and P. darwini rupestris. It is separable from P. darwini (and P. osilae) by shorter, untufted, less completely bi- color tail, bright preauricular tufts, and, in some populations, by the ochraceous metatarsal patch. Superficial distinction from sublimis is made by greater size, proportionately longer tail and other char- acters given in the descriptions. The proodont and practically smooth-faced incisors of boliriensis distinguish it from all other spe- cies of Phyllotis; molars generally as in pictus, but cusps not flexed forward, minor folds of m^ more reduced, anterior cingulum of m- and vestige of mesoloph larger, anterior fold of mT obsolete. Variation. — Most variation in color is the result of molt. Old grizzled pelage of head and shoulders is defined by a molt line. Old pelage mid-dorsally appears as a gray or plumbeous band. Molt area on rump converges to a concentric line. Condition of pelage in a series of 22 specimens from Bolivia taken from October 9 to Novem- ber 15 ranges from prime to extremely old. A series collected in Cailloma, Peru, in August, also includes individuals in old and new pelage, but in molting individuals the mid-dorsal area is gray not plumbeous as in the Bolivian mice. A more or less well-defined pectoral streak is present in many individuals. No relationship be- tween this marking and cranial characters is apparent. Posterolateral palatal pits of each side are usually paired, with one situated anteriad of median posterior border of palate, the other anteriad or posteriad of border. Remarks. Because of its greater angularity, the dorsal outline of the skull of P. boliviensis more nearly resembles that of microtines than do skulls of other species of the genus Phyllotis. Habitat and habits.— According to Pearson (1951, p. 145), who ob- served the mammals of the Lake Titicaca region in Puno, Peru, the species "lives at high altitudes in many kinds of habitats, such as rock slides, stone walls, and even in abandoned tuco-tuco \Ctenomyx} burrows out in open country. Its food must also be varied, for twice it was seen feeding on lichens in rocky places, yet no lichens were available to those living in the tuco-tuco diggings. It is chiefly 416 FIELDIANA: ZOOLOGY, VOLUME 46 diurnal. . . . We saw many during the day, and a few were trapped at night. The owls at Pairumani caught none, which confirms the evidence that these mice are chiefly diurnal. Because they live in so many kinds of places, their animal associates include most of the species living at high altitudes. Most remarkable, however, is their close relationship with viscachas. I found them living in the midst of a colony of viscachas, and two of these Auliscomys [i.e., Phyllotis boliviensis] habitually sunned on the rocks within a foot or two of sunbathing viscachas. When the viscachas moved down out of the rocks to feed, Auliscomys went along, too, and grazed like an elf among grownups. I have seen an Auliscomys that was feeding be- side a viscacha scurry for shelter when a distant viscacha sounded the alarm whistle. The nearby viscacha in this case did not retreat. The nervous, droll movements of these Auliscomys are always enter- taining, sometimes startling. Instead of peering cautiously out of tuco-tuco burrows, they pop out, ears, head and shoulders, with paws folded neatly on their breasts." Regarding the breeding among these mice, Pearson (op. cit. p. 146) observed that "a male at Pairumani on July 19 had small testes, several males at Caccachara had large testes (up to 10 mm. long) after September 15, and at Cailloma on December 6. The number of embryos varied from three to four. Juvenile individuals were seen on November 7 and 20." The type specimen of boliviensis was described by its collector, Thomas Bridges (in Waterhouse, supra cit.) as a "charming little animal . . . found in the same locality as the above ["Ctenomys Bra- ziliensis"], inhabiting the abandoned caves of the former species. It makes its appearance in the afternoon, when the sun is nearly on the horizon, to feed on grass, and is often seen sitting on its hind legs; and it then presents its pretty white abdomen and erect ears. In this position it has the appearance of a rabbit in miniature. The natives call it 'Achohalla/ pronounced 'Ha-cho-ha-ya'." This re- markably tame mouse not only permits itself to be observed at close range but also, according to Mann (1945, p. 81) does not attempt to bite when captured, is gentle when handled and fearlessly accepts food from its captor. Phyllotis boliviensis boliviensis Waterhouse Hesperomys Boliviensis Waterhouse, 1846, Proc. Zool. Soc. London, 1846: 9. Waterhouse, 1846, Ann. Mag. Nat. Hist., (1), 17: 483. Gervais, 1855, Castelnau Exped. Ame>ique Sud, pt. 7, Zool., Mamm., p. Ill — BOLIVIA: Chuquisaca (Chuquisaca [= Sucre]). HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 417 H[esperomys (Phyllotis}] boliriensis, Thomas, 1884, Proc. Zool. Soc. London, 1884: 449 — classification. [Phyllotis] boliriensis, Trouessart, 1897, Cat. Mamm., p. 534- classification. Phyllotis boliriensis, Thomas, 1902, Ann. Mag. Nat. Hist., (7), 9: 223 BOLIVIA: Oruro (Livichuco); Potosi (Potosi). Thomas, 1902, op. cit., (7), 10: 249- BOLIVIA: Potost. Mann, 1945, Biol6gica, Trab. Inst., Biol. Univ. Chile, fasc. 2: 78, figs. 29, 30, 31, 36, pis. 29, 30 - CHILE: Tara- pacd (Puna). Euneomys (Auliscomys) boliriensis, Thomas, 1916, Ann. Mag. Nat. Hist., (8), 17: 143 — classification. [Phyllotis (Auliscomys)] boliriensis, Osgood, 1915, Field Mus. Nat. Hist., Zool. Ser., 10: 191 — classification. Phyllotis (Auliscomys} boliriensis, Osgood, 1943, Field Mus. Nat. Hist., Zool. Ser., 30: 210, fig. 29 (skull)— part, CHILE: Tarapaca (Choquelimpie). Sanborn, 1950, Mus. Hist. Nat. "Javier Prado," (A), Zool., no. 5: 9 •- BOLIVIA: La Poz (Esperanza); CHILE: Tarapaca (Choquelimpie^. Phyllotis boliriensis boliriensis, Pearson, 1958, Univ. California Publ. Zool., 56: 452— BOLIVIA: La Paz (Esperanza); Oruro (Livicucho [ = Livichuco?!); Potosi (Potosf; Potosf Peak); CHILE: Tarapacd (Choquelimpie^. [Hesperomys (Hesperomys)] Waterhousii Trouessart, 1881, Bull. Soc. d'e"tudes Sci. d'Angers, 10, (1): 138 — new name for boliriensis Waterhouse, a sec- ondary homonym of Akodon boliriensis Meyen. [Phyllotis] Waterhousii, Trouessart, 1897, Cat. Mamm., p. 534 — a synonym of Phyllotis boliriensis Waterhouse. Type. — Adult, skin and skull, British Museum (Natural History) no. 45.11.18.9; collected by Thomas Bridges. Type locality. --"A few leagues south of Potosi, at an elevation of 12,000 feet," Potosi, Bolivia (Bridges, in Waterhouse, supra cit., pp. 8-9; see also Thomas, 1902, Ann. Mag. Nat. Hist, (7), 10: 249). Distribution (fig. 94).— Highlands of western Bolivia, in depart- ments of Chuquisaca, Potosi, Oruro and La Paz, west into the high- lands of Tacna, Peru (no records), and Tarapaca, Chile; altitudinal range from 3750 meters (type locality) to 4700 meters (Choquelimpie', Tarpaca, Chile) above sea level. Characters. — Those of the species, upper surface of hind foot usually without well-defined ochraceous metatarsal patches. Remarks. Specimens here referred to boliriensis are from local- ities about midway between the type localities of boliriensis and flavidior. Measurements.- See Table 55. Specimens examined.— 29. BOLIVIA. La Paz: Ksperanza, Pa- cajes, 4200 meters, 22 (CNHM). CHILE. -Tarapacd: Choque- limpie', 4700 meters, 7 (CNHM). Sol 85^ > bflj« «sf t- rH 05 CO -o io CO O 00 00 IO CO MD IO Zygomatic breadth 05 So I co co t-^ t-' o N" oo oo 71 71 rH t— CD CD rH rH OO rH rH CO CO t~~ O3 t>~ a h o -C Phy • •+-* &c 7 0) 3~ -* 02 |M n H e j- 4J O o -o u: 0 v^. co c CO t> rH f — ' co » oc t- rH CO 2 CO U3 CO OJ O 0 (M CO CO 1 1 1 or centimeter. w CD CO •^ w 00 t- oo oo os 0) (M (M (M CO CO QJ .£ ~ t- j—4 c 9 ^^ S" O *s O ^~ EH rH O M -C oo 03 oo o 03 oo os oo eg 0 M c S >j j-^ .2 a O o" D. •E 5 T3 rH o to c C L o> i a O T3 £ "2 rH 13 cu S CO co' CO U5 CD c -. W co rH CO (M CO 'So -^ 'C o c i \q) S •f 09 'S, o c "t. a S «»-. c3 to ^ £ "-H 9 J ** 0 1 _ O - S H flJ {Tj & ^ w S 3 cr o u oT Q c f "c O PC O rj H c £ o •c s 1 E J M w co co oo o o co' o co co co co OS S 00 ^ CD o o rH rH I I co o oo os CO 00 U5 oo 'C o - s o o «tl •tH fi oT PQ O 418 HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 419 Phyllotis boliviensis flavidior Thomas Phyllotis boliviensis, Thomas, 1897, Ann. Mag. Nat. Hist., (6), 20: 550 in text; PERU: Puno. Thomas, 1900, op. cit., (7), 6: 469-PERU: Arequipa (Cailloma; Arequipa). Phyllotis (Auliscomys) boliriensis, Osgood, 1943, Field Mus. Nat. Hist., Zool. Ser., 30: 210— part; PERU: Puno. Phyllolis boliviensis flavidior Thomas, 1902, Ann. Mag. Nat. Hist., (7), 10: 248. Pearson, 1958, Univ. California Publ. Zool., 56: 452— PERU: Arequipa (Cailloma; Huaylarco; Salinas; Sibayo); Moquegua (Lago Suche; Pampa Huaitire; Titire); Puno (Caccachara; Pairumani; Pampa de Ancomarca; Pampa de Capazo; Pasto Grande; San Antonio de Esquilache). [Euneomys (Auliscomys) boliviensis} flavidior, Thomas, 1916, Ann. Mag. Nat. Hist., (8), 17: 143— classification. [Phyllotis (Anliscomys) boliviensis} flavidior, Osgood, 1915, Field Mus. Nat. Hist., Zool. Ser., 10: 191— classification. Phyllotis (Auliscomys) boliviensis flavidior, Sanborn, 1950, Mus. Hist. Nat. "Javier Prado," (A), Zool., no. 5: 9 — PERU: Arequipa (Cailloma; Salinas) ; Puno (San Antonio de Esquilache). Pearson, 1951, Bull. Mus. Comp. Zool., 106: 144— PERU: Puno. Type. — Female, skin and skull, British Museum (Natural His- tory) no. 2.7.2.5, collected 19 May 1901, and presented by Bernard Hunt. Type locality. — Bateas, about 15 kilometers west of Cailloma, Arequipa, Peru; altitude 4500 meters above sea level. Distribution (fig. 94). — Altiplano of southern Peru in depart- ments of Arequipa and Puno; known altitudinal range from 4400 meters (Salinas, Arequipa) to 5300 meters (Caccachara, Puno) above sea level. Characters. — Like typical boliviensis but with a well-marked meta- tarsal patch usually present. Remarks. — In the original description a number of differences in coloration are noted between boliviensis and flavidior. In present material from three relatively widely separated localities, only the character of the metatarsal patch is valid. Measurements. — See Table 56. Specimens examined. 36. PERU. Arequipa: Cailloma, 26 (CNHM); Salinas, 1 (CNHM). Puno: San Antonio de Esquilache, 9 (CNHM). Phyllotis sublimis Thomas. (Synonymy under subspecies.) Distribution (figs. 57, 97). Highlands of southern Peru in the departments of Ayacucho, Apurimac, Arequipa and Puno, thence 420 EXPLANATION OF FIGURE 97 Phyllotis sublimis: collecting localities and collectors. Type localities in boldface. Phyllotis sublimis sublimis PERU 1. Chalhuanca, Apurimac, 25 miles SW. — O. P. Pearson at 14,500 feet. 2. Puquio, Ayacucho, 9 miles NE. — O. P. Pearson at 14,500 feet. 3. Puquio, Ayacucho, 21 miles ENE.— O. P. Pearson at 14,600 feet. 4. Rinconado Malo, Arequipa. P. O. Simons at 5500 meters. 5. Sumbay, Arequipa. — C. C. Sanborn at 13,500 feet. 6. Salinas, Arequipa. — C. C. Sanborn and J. M. Schmidt at 14,000 feet. 7. Huaylarco, Arequipa.— O. P. Pearson or associates at 15,000 feet. 8. Limbani, Puno. -C. B. Koford at 13,800 feet and 15,000-15,500 feet. 9. Santa Lucfa, Puno.— M. R. Portugal, at 13,300 feet. 10. San Antonio de Esquilache, Puno. — C. C. Sanborn at 15,000 feet. 11. Puno, Puno. —Variously collected by O. P. and A. K. Pearson and C. B. Koford at 3.1 miles NE., 13,200 feet; 3.1 miles W., 13,000 feet; 48 miles W., 13,500 feet; 51 miles W., 14,000 feet. 12. Pairumani, Puno.-O. P. Pearson at 13,000 feet. 13. Yunguyo, Puno.— C. C. Sanborn at 12,700 feet. 14. Caccachara, Puno.— O. P. Pearson at 15,500 feet. 15. Huacullani, Puno.— C. C. Sanborn at 12,700 feet. 16. Rfo Santa Rosa, Puno.— C. B. Koford at 14,400 feet. 17. Mazocruz, Puno. — Specimens from 17 miles S. at 13,500 feet and 30 miles S. and 3 miles W. at 14,300 feet. 18. Pampa de Ancomarca, Puno. — O. P. Pearson at 13,800 feet. 18. Pampa de Capazo, Puno.— Specimens collected at 14,300 feet. 19. Pampa de Titiri, Tacna. — Specimens collected at 14,600 feet. BOLIVIA 20. El Cumbre, La Paz. -G. H. H. Tate, at 15,200 feet. 21. La Paz, La Paz, 20 miles S.— O. P. Pearson at 13,000 feet. 22. Cosmini, Cochabamba.— P. O. Simons at 4300 meters. 23. Oruro, Oruro, 20 miles NE.— O. P. Pearson at 14,000 feet. 24. Livichuco, Oruro. - P. 0. Simons at 4500 meters. CHILE 25. Toconce, Antofagasta. — K. Koford at 4700 meters. Phyllotis sublimis leucurux BOLIVIA 26. Lipez, Potosf. ARGENTINA 27. La Lagunita, Sierra de Zenta, Jujuy. E. Budin at 4500 meters. 27. Sierra de Zenta, Jujuy. 421 422 FIELDIANA: ZOOLOGY, VOLUME 46 south through the highlands of northeastern Chile, western Bolivia and northern Argentina; altitudinal range from 4000 meters above sea level to limits of perpetual snow. External characters. — Smallest of the short-tailed species of Phyl- lotis; tail from one-third to less than two-thirds combined head and body length, white, gray or buff, with or without a brown mid-dorsal line on upper surface, tip un tufted; hind foot proportionately small, well haired above, naked below except on heel, soles smooth or lightly scutulated, plantar pads relatively large; ears not conspicuously large, pale postauricular patches sometimes present; upper surface of head and body buffy finely mixed with black, the anterior portion grizzled, posterior portion, especially rump, more saturate, sides of body paler, cheeks whitish, lateral line usually indistinct or inter- rupted; underparts sharply defined whitish to pale gray with plum- beous of basal portions of hairs showing through. Cranial characters (figs. 58-60, 98, 109, 110). — Supraorbital region of skull as in pictus, but borders hardly or not at all raised; zygomata more or less evenly bowed, greatest breadth approximately equal, usually more than, distance between posterior tips of nasals and an- terior border of supraoccipital; zygomatic plate as in boliviensis; in- terparietal extremely reduced; bullae moderately inflated, their antero-posterior length, less tubes, not equal to alveolar length of molar row; outer side of mandible without capsular projection en- casing base of incisor. Dental characters (figs. 67, 99, 111). — Incisors orthodont or slightly opisthodont, their anterior face nearly always marked by faint, shal- low grooves hardly or not at all visible to the unaided eye. Molar rows parallel-sided; upper first molar 4-rooted, lower 3-rooted; sub- hypsodont crown plane, or slightly terraced in unworn teeth; enamel of crown involuted, forming a modified sigmoid pattern in m^; out- line of cusps subovate; outline of procingulum of m- ovate or sub- cordate; procingulum of m^ weakly developed, the first primary fold hardly or not at all persistent in worn tooth; procingulum of m^ reduced; first minor fold of m^ slightly or not developed; vestige of mesoloph sometimes present in m1-'2-; postcingulum of m^ moder- ately developed, of m7 reduced; minor fold well developed and per- sistent in niTj. Comparisons. — Small size, pale coloration, thick, soft fur, ex- tremely short, whitish, thinly haired tail, and soles bare except at heel, are external characters that distinguish Phyllotis sublimis from all other phyllotine rodents. There is striking superficial resem- HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 423 Flo. 98.- Phylloiis xublimix. Dorsal and palatal aspects of skull (X 2). blance between P. sublimis and the vole-like Galenomys garlcppi in size, color, character of pelage and proportional lengths of tail, limbs and ears. The latter is separated from the former by thickly haired tail and soles, arched dorsal contour of skull, proodont incisors and enlarged first molars. The skull of sublinris agrees with those of pictus and boliviensis in general characters, but is smaller than either and less angular than that of boliviensis. Angle of projection of inci- 424 FIELDIANA: ZOOLOGY, VOLUME 46 FIG. 99. — Phyllotis sublimis. Right upper and left lower molars of a, Juvenal; b, adult. sors, relative to long axis of skull, is intermediate in position between the more proodont incisors of boliviensis and the more orthodont in- cisors of pictus. Molars of sublimis are essentially as in boliviensis except for the weaker procingulum of m^. Variation. — Differences in size, notably of the molar row, be- tween the mice of such nearby localities as Sumbay and Salinas in HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 425 Arequipa, Peru, are of approximately the same magnitude as those given by Pearson (1958, pp. 446, 466) for separating leucurus of the southern half of the specific range from sublimis of the northern half. Underparts appear to be whitish or gray depending on the propor- tional length of the whitish-tipped terminal portions of the hairs to the plumbeous basal portions. A series from the humid locality, El Cumbre, La Paz, Bolivia, on the eastern slope of the Andes, is darker than specimens from the comparatively dry altiplano of southern Peru. Habitat and habits. — In describing Phyllotis sublimis, Thomas (1900, p. 468) observed that the species "lives at the highest altitude [type locality, 5500 meters] from which mammalian life has been recorded in the New World, and in the Old is only surpassed in this respect by a few of the Himalayan species." Actually, a number of other species of rodents, carnivores and ungulates live at the same elevations as P. sublimis. The determin- ing factor in these cases is not altitude per se but the points at or below the limits of perpetual snow where suitable food is found. Permanent snow line in Peru varies locally from under 5000 to 5800 meters above sea level. Phyllotis sublimis is, according to Pearson (1951, p. 147), "one of the commonest mice of the altiplano of southern Peru. It is noc- turnal and seems to prefer the shelter of rocks in rather open, grassy places. A stone corral in a field of ichu grass is an ideal habitat, but sublimis will also live in abandoned tuco-tuco diggings, in rock piles, in stunted thorn and sage ... or among boulders QY jar eta. It shares these habitats with one or more of the following: Phyllotis darwini subsp. [=rupestris], P. boliviensis flavidior, Hesperomys, Neotomys, Akodon jelskii pulcherrimus, and A. andinus lutescens. Because it prefers grassy places, it is more frequently found in valleys than on barren or rocky hilltops. It seems to be gregarious, for many times I have caught several at the same place, and Thomas [supra cit.] re- ported that the type and 8 topotypes were all dug out of one burrow. "The most puzzling thing about these mice is their disappearance during October, November, and December. They were found in considerable numbers at Pairumani and Santa Rosa during July, and at Caccachara in August and early September. Thirty-four were trapped at Caccachara from August 16 to September 15. No more traps were set until October 4, whereupon considerable trap- ping in likely terrain was carried out until early December, yet not one more sublimis was caught during this time. Nor were any caught 426 FIELDIANA: ZOOLOGY, VOLUME 46 at Cailloma, Pairumani, and Caccachara in December of 1939. It seems likely that either the population was greatly reduced by dis- ease in September in 1939 and 1946 ... or that sublimis remains below ground, possibly hibernating during much of October, Novem- ber, December, and perhaps subsequent months. We saw no evi- dence of an epizootic nor of hibernation (except that one specimen was recorded as fat on July 15) but none of the later specimens were skinned and, therefore, subcutaneous fat deposits may have been overlooked. Sanborn has kindly given me additional dates of cap- ture of sublimis. These also suggest a disappearance during the wet season; his latest capture was October 7 and the earliest, April 30." Pearson suggests the possibility that " 'Hibernation' during the southern hemisphere summer would not be totally unexpected in the case of P. sublimis, because this mouse is nocturnal. Its un- usually long, soft, thick fur would be good protection in the cold dry nights of winter; but might wet easily in the warmer nights of sum- mer when the ground is frequently covered with wet snow. Such an escape (if it occurs) from wet and snow in summer is unique, and should, of course, be considered a form of estivation." In conformity with Pearson's observations, I have noted also that with the onslaught of a rainy season, after prolonged drought, there is a dramatic disappearance of certain species of usually noc- turnal mice. This phenomenon is general throughout the tropical zone from sea level to the highest altitudes inhabited by mammals. The disappearance, or radical reduction in numbers, however, is not real. It is apparent only in trapping records. Non-breeding mice reduce their geographic radius of activity and the frequency of forays concurrently with the revival of plants and the sprouting of succu- lent grasses and herbs stimulated by the rains. In the midst of an abundance of familiar, desirable foods in natural situations, mice become trap shy and lose much of their susceptibility to bait. As Pearson points out, no specimens of Phyllotis sublimis were taken during the rainy season. Conversely, most successful trapping oc- curs toward the end of the dry season, when food plants are scarce and low in food value. In the case of P. sublimis, this period coin- cides with the recorded breeding season when sexually stimulated mice are extremely active, least wary, and most prone to being trapped. Reproduction. — The following is quoted from Pearson (1951, p. 147) : "Males at Pairumani in mid-July had small testes (3 mm.), whereas those at Santa Rosa at the end of July were in breeding con- HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 427 dition (testes 7 to 8 mm.). Males at Caccachara, however, did not reach breeding condition until September 3. None of the females con- tained embryos through September 15, when the last capture was made. It is interesting to speculate upon the time of reproduction if they estivate at the season when the other mice are reproducing. The ratio of sexes was 30 males to 21 females." Later Pearson (1958, p. 447) observed that "young females are sexually precocious. One specimen in gray juvenile pelage and with head and body only 86 mm. long contained a vaginal plug; another specimen with head and body 89 mm. and skull 23.8 mm. long was in an early stage of pregnancy." Sexual precocity in phyllotines has already been noted in P. darwini and P. osilae. Phyllotis sublimis sublimis Thomas Phyllotis sublimis Thomas, 1900, Ann. Mag. Nat. Hist., (7), 6: 467. Koford, 1954, Invest. Zool. Chilenas, 2: 95— CHILE: Antofagasta (Toconce). E[uneomys] sublimis, Thomas, 1901, Ann. Mag. Nat. Hist., (7), 8: 254 — classi- fication. Euneomys sublimis, Thomas, 1902, Ann. Mag. Nat. Hist., (7), 9: 134— BO- LIVIA: Cochabamba (Cosmini). Thomas, 1902, Ann. Mag. Nat. Hist., (7), 9: 226— BOLIVIA: Oruro (Livichuco). [Euneomys (Auliscomys)} sublimis, Thomas, 1916, Ann. Mag. Nat. Hist., (8), 17: 143 — classification. [Phyllotis (Auliscomys)] sublimis, Osgood, 1915, Field Mus. Nat. Hist., Zool. Ser., 10: 191 — classification. Phyllotis (Auliscomys) sublimis, Sanborn, 1950, Mus. Hist. Nat. "Javier Prado," (A), Zool., no. 5: 1- — PERU: Arequipa (Salinas; Sumbay); Puno (Yunguyo). Pearson, 1951, Bull. Mus. Comp. Zool., 106: 147- PERU: Puno (Pairumani; Santa Rosa; Caccachara). Phyllotis sublimis sublimis, Pearson, 1958, Univ. California Publ. Zool., 56: 447 — BOLIVIA: La Paz; Oruro; PERU: Apurimac; Arequipa; Ayacucho; Puno; Tacna (for specific localities see explanation of fig. 97, p. 421 ); revision. Phyllotis (Auliscomys) sublimis leucurus, Sanborn (not Thomas) 1950, Mus. Hist. Nat. "Javier Prado," (A), Zool., no. 5: 8- BOLIVIA: La Paz (El Cumbre). Phyllotis sublimis leucurus, Pearson (part, not Thomas), 1958, Univ. California Publ. Zool., 56: 448- CHILE: Antofagasta (Toconce). Hesperomys lepidus montanus, Sanborn (part, not Sanborn), 1950, Mus. Hist. Nat. "Javier Prado," (A), Zool., no. 5: 4— BOLIVIA: La Paz (El Cumbre, juvenals only). Type. — Female, skin and skull, British Museum (Natural His- tory) no. 0.10.1.60; collected 18 June, 1900, by Perry 0. Simons. Type locality. — Rinconado Malo mountain pass, above Cailloma, on road to Sumbay, Arequipa, Peru; altitude said to be 5500 meters above sea level. 428 FIELDIANA: ZOOLOGY, VOLUME 46 Distribution (fig. 97). — Highlands of southern Peru, northeastern Chile and western Bolivia from southern Ayacucho and Apurimac through Arequipa and Puno, in Peru, into Antofagasta, Chile, and La Paz and Oruro, Bolivia; altitudinal range approximately from 3800 to 5625 meters above sea level. Characters. — A small, pale, short-tailed, thickly furred vole-like mouse, as described for the species. Taxonomy. — Undoubted representatives of typical sublimis are from the Pacific and Lake Titicaca drainage basins of southern Peru. Specimens from El Cumbre, Bolivia, referred to Phyllotis sublimis leucurus and Hesperomys lepidus montanus by Sanborn, are from the head of the Rio La Paz in Amazonian drainage. They are darker throughout than true sublimis. Pearson (1958, pp. 447-448) also treats them as inseparable from typical sublimis. On the other hand he notes that "three unusually short-tailed specimens from the Prov- ince of Antofagasta, northern Chile, are somewhat paler than typical sublimis sublimis and have relatively large tooth rows but white bellies" and he refers them to leucurus. Pearson's measurements for the Toconce specimens are reproduced in Table 57. If, as Pear- son affirms, the only significant differences between sublimis and leu- curus are in the length of the molar row and color of the belly, the Toconce series is certainly indistinguishable from typical sublimis. Measurements. — See Table 57. Specimens examined. — 33. PERU. — Arequipa: Salinas, 6 (CNHM); Sumbay, 6 (CNHM). Puno: Huacullani, 5 (CNHM); San Antonio de Esquilache, 3 (CNHM); Yunguyo, 6 mi. south, 1 (CNHM); Santa Lucia, 1 (CNHM). BOLIVIA.— La Paz: El Cumbre, 11 (AMNH). Phyllotis sublimis leucurus Thomas Euneomys (Auliscomys) leucurus Thomas, 1919, Ann. Mag. Nat. Hist., (9), 4: 129, footnote. Thomas, 1921, op. cit., (9), 8: 612— ARGENTINA: Jujuy (Sierra de Zenta; La Lagunita). [? \Aidiscomys leucurus, Thomas, 1926, Ann. Mag. Nat. Hist., (9), 18: 194 — BOLIVIA: Potosi (Lipez). Phyllotis (Auliscomys) sublimis leucurus, Sanborn, 1950, Mus. Hist. Nat. "Javier Prado," (A), Zool., no. 5: 8 —part, classification. Phyllotis sublimis leucurus, Pearson, 1958, Univ. California Publ. Zool., 56: 448 -part, ARGENTINA: Jujuy (La Lagunita; Sierra de Zenta); BO- LIVIA: Potosi (Lipez). Type. — Adult female, body in alcohol, skull separate, British Museum (Natural History) no. 19.7.10.3. IO .-<* b- co U3 •*!•' 10' £T3 1 1 "1 o S *-* o eg ••* be C co' •"* 10' 10 y. a N 10 ^" ^ £• Hr £ •*!• IO Tf -rf IO O J= ~ to US X — • co eg 10" co 10" 8, T. D 7 V — .» rt W o co »o —~ ; ; r 1 o co oi o Ol Ol B 5 ^ ^^ -~] ~ i e» j '-i co" CO .2 S 3 J •o £ D co CO I - 5 of " of 1 01 co" oj^ co CO" — '• "a t 5 - 01 Ol Ol ~- g ___, o 13 8 kO I—I ^-i co 5 "8 TJ >^ - 1C • •— 1 t "1 IO 1 oo i 0. •g '5 •c o r^ 7 1 CO ^ i ~ r oo 10 Sfi '~ •^ tf L "I w E 9 _, IO CO — IO co oo ~ c \ IO IO If ) -r 10 -r 10 c •;: a c ^ 1 to g ~ co 'B . 'C "" tc "i" ^1* ° co •o 1 -' # ^^ Ol i 'S, to c - •^ 1 01 o £ d c "S — [ ' r "^ ocT •^-^ Ol E IO 0 t - o> 5. oc - - »" 0 ~ 0) 01 0 0 ."<" iO S O5 - 3 '~ - _ _0 J w eg 1 | . -S I 'c > JS w 'c : :? g i c , •_:• •^ 3 1 d t S Q. o> Q 1 ^ L, _5 _C c 1 — i < 5 g^ g § S J 0 L. >, ^S j •O *^* -/; ~ c/5 U E^ 5 J3 429 430 FIELDIANA: ZOOLOGY, VOLUME 46 Type locality. — La Lagunita, east of Maimara, Sierra de Zenta, Jujuy, Argentina; altitude, 4500 meters above sea level. Distribution (fig. 97). — High Andes of Jujuy, northern Argentina and Potosi, southwestern Bolivia; altitudinal range between 4500- 4700 meters above sea level. Characters. — Originally said to differ from typical sublimis "by the greater size of its skull;" according to Pearson (1958, p. 447) "specimens from southern Bolivia and northern Argentina have longer tooth rows and grayer bellies but otherwise agree well with typical sublimis." Remarks. — External measurements of the type were taken from the shrunken specimen preserved in spirits. Cranial dimensions, however, indicate an adult, probably very old, and larger than any of the available specimens of P. sublimis sublimis. Besides the type, Thomas recorded 4 paratypes, 5 specimens from Sierra de Zenta, and 8 from Lipez, without data. Most of these were examined by Pearson (1958, p. 448) and found to be "gray bellied and [with] large teeth, hence . . . placed with leucurus." As noted under the species heading, differences in molar row length and color of underparts between northern and southern pop- ulations of the species are not significantly greater than differences between local populations in Arequipa Department, Peru. Pearson (1958, p. 448) lists 13 specimens from southern Bolivia (Lipez, 8) and northern Argentina (La Lagunita, 2, and Sierra de Zenta, Jujuy, 3) but lumps measurements for only 5 from "Lipez (Bolivia) and Prov- ince of Jujuy." Measurements. — See Table 57. Specimens examined. — None. Phyllotis gerbillus Thomas Phyllotis gerbillus Thomas, 1900, Ann. Mag. Nat. Hist., (7), 5: 151— PERU: Piura (Piura; Catacoas). Thomas, 1900, op. cit., p. 356— comparison. Paralomys gerbillus, Thomas, 1926, Ann. Mag. Nat. Hist., (9), 17: 315 — generic type. Type. — Adult female, skin and skull, British Museum (Natural History) no. 0.1.1.30; collected 12 April, 1899, by Perry 0. Simons. Type locality. — Piura, Piura, northwestern Peru; altitude 50 meters above sea level. Distribution (figs. 57, 100). — Known only from the neighbor- hood of Piura and nearby Catacoas, both localities in the Sechura HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 431 Desert, Piura, northwestern Peru. The range of Phyllotis gerbillus is less than that of any other phyllotine and is one of the most re- stricted of South American mammals. No other cricetine has been recorded from the same area. External characters. — Smallest member of the Phyllotis section, and smaller than most Calomys; coloration gerbil-like; tail subequal to combined head and body length, thinly haired white throughout, sometimes with an occasional dark hair or patch of hairs; integument of terminal one-fourth or one-third of tail dark brown, contrasting sharply with whitish basal portion, pencil rudimentary; hind foot (fig. 3) broad and short (19-21 mm.), sole sparsely set with whitish bristles except on six well-developed pads; ear comparatively small, its length measured from notch, appreciably less than that of hind foot, a white preauricular tuft usually present; upper parts of head and body ochraceous or salmon-color finely peppered with dark brown; upper half of sides of body like back but hairs without dark tipping, lower half wholly white like underparts; hairs of limbs, cheeks, sides of rostrum white to roots. Cranial characters (fig. 101). — Sides of supraorbital region di- vergent, the edges square, without beading or ledges; zygomatic arches more or less parallel-sided, greatest breadth between them less than distance from posterior tips of nasals to anterior border of supraoccipital; braincase not inflated; interparietal well devel- oped; bulla moderately large, its antero-posterior length, less tube, equal to or more than, length of molar row; outer side of mandible with capsular projection encasing base of incisor. Dental characters (figs. 101, 102).- Incisors ungrooved, opistho- dont; molar rows parallel-sided ; upper and lower first molars 4-rooted ; molar crowns low, slightly terraced; second and third molars with tendency to involute or laminate, a sigmoid or figure 8 crown pat- tern resulting; upper cusps ovate, lower subtriangular; procingulum of ma moderately developed, the first primary fold well defined and persistent; minor fold of m2 present. Comparisons. The bright sandy-colored upper parts of head and body of Phyllotis gerbillus contrasting sharply with the white of lower half of sides, limbs, cheeks and white-haired tail is strikingly similar to the color pattern of northern Argentine representatives of Klig- modontia typus. The first is readily distinguished from the other by its narrower, less hairy feet with the three postdigital plantar pads not coalesced to form a hairy cushion. The short-tailed, compara- tively small-eared North American Peromyscus polionatus is also 432 FIELDIANA: ZOOLOGY, VOLUME 46 similarly colored, but is too distantly related to require more than incidental mention. Nearest geographic, as well as phyletic ally is the slightly larger, allopatric Phyllotis amicus, with comparatively dark upper parts, sides of body and limbs, much larger ears and longer, sharply bicolor, more hairy tail. Cranially and dentally, gerbillus retains the more primitive features of less expanded brain- case, smaller auditory bullae and less planed molars. However, the greater tendency for lamination in the molars of gerbillus as com- pared with those of amicus, is an advanced condition. Variation. — All known specimens of Phyllotis gerbillus are from the region about Piura. Individual variation, the only kind of vari- ation, except possibly crop variation, that can occur here, is remark- ably slight. In some specimens the dorsal surface is nearly uni- formly ochraceous, in others the dorsum is sandy in appearance because of the greater amount of dark brown tipping to the cover hairs. These two conditions are equivalent to the fully intergrading pale and dark color phases of normal occurrence in mice. In one juvenal hair bases of outer sides of thighs are plumbeous, in all others the hairs are wholly white. Habits and habitat. — The notation on the original label of a topo- type collected by Perry 0. Simons is "in pile of brush." Nothing more is known regarding the habits of the Piura desert mouse. Seventeen specimens of Phyllotis gerbillus collected by Celestino Kalinowski, 5 kilometers east of Piura, were secured in 1954 dur- ing the course of three trap nights (July 11, 18, 19). No other mammal was taken in the locality and no other cricetine rodent has ever been recorded from the Piura coast. Remarks. — The striking coloration of Phyllotis gerbillus Thomas, its desert habitat and uncertainty regarding its exact generic posi- tion induced Thomas (1926a, p. 315) to regard the mouse as rep- resenting a new genus he named Paralomys. It was said to be nearest related to Eligmodontia. Ellerman (1941, pp. 446, 448) referred Para- lomys to Hesperomys (=Calomys) but his definition of Hesperomys excludes it. Osgood (1947, p. 171) restored the Peruvian desert mouse to Phyllotis but as a subgenus. He observed, however, that "no very unusual generic or subgeneric characters for it [Paralomys] have been pointed out." Measurements. — See Table 58. Specimens examined. — 23. PERU. — Piura: Piura, 6 (CNHM, 1; USNM, 5); 5 kilometers east of Piura, 17 (CNHM). CS ° O "o-c o o T •-T <5 c 3J2 3""S| a -^ I- II N"° ^^ CO I I oo ic os t>- co o co 2 I I I I ' I I os oq IM° i-H co co oTo" co' ic (M (M I I o co co co CO 1C (M_ t-_ co oj c os x' c> 0) O g "«.— >, « , §«§ * Illilllll 450 HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 451 2. Most females are gravid in summer (December March), none in winter (June-August [September?]). 3. The sexes are apparently equal on the basis of a nine-month capture count. Differences in monthly captures between males and females vary from a minimum of QAC[ more females in March, to 22.4% more males in August. The great excess of males over females in August may be attributed, in part, to wider ranging, trap-prone males which were inaugurating the spring breeding season. According to Kiihlhorn (1952, p. 120) females of P. griseoflarus from southern Mato Grosso were without young during April and May of 1938. This indicates a reproductive pattern similar to that of the species in Argentina. In some parts where their geographic ranges overlap, Phylhlis griseoflarus, Phyllotis darwini and P. osilae are found side by side. Other nearly related species occurring in the same areas are Calomys laitcha, Calomys carillus, and Eligmodontia typns. Sylvatic plague in these mice has been recorded by Uriarte and Morales (1935, p. 200) and by de la Ban-era (1936, p. 439; 1939, p. 441; 1940, p. 565). Phyllotis griseoflavus griseoflavus Waterhouse Mus (Phyllotis) griseo-flarus Waterhouse, 18.37, Proc. Zool. Soc. London, 1837: 28. Mus griseo-flatus, Waterhouse, 1839, Zool. Voy. "Beagle," Mamm., p. 62, pi. 21 (animal), pi. 34, fig. 15 (molars) type redescribed. Phyllotis griseoflarus, Thomas, 1897, Ann. Mag. Nat. Hist., (6), 20: 215 ARGENTINA: Salta (upper Rfo Cachi and lower Rio Cachi); Catamarra. Thomas, 1898, Bol. Mus. Zool. Anat. Comp. Univ. Torino, 13, no. 315: 3 —PARAGUAY: Chaco (San Francisco Mission). Hesperomyx grixeoflarux, Burmeister, 1879, Descr. Phys. Rep. Argentine, 3: 219- ARGENTINA: Rfo Chubut to Port Desire. H[esperomys (Phyllotis)} grixeoflarux, Thomas, 1884, Proc. Zool. Soc. London, 1884: 449- classification. Eligmodontia griseoflara, Thomas, 1898, Proc. Zool. Soc. London, 1898: 210 ARGENTINA: Chubut; classification; habits. Allen, 1905, Exped. Patag. Princeton Univ., 3, Mamm., p. 55 review. Graomys griseo-flarus, Thomas, 1916, Ann. Mag. Nat. Hist., (8), 17: 142 type of Graomys. Graomys griseoflarus griseoflarux, Uriarte and Morales, 1935, Rev. Inst. Bac- teriol., 7, (2): 200, pi. 2 (col.) ARGENTINA: La I'ampa (Villasuso); bubonic infection. De la Barrera, 1936, op. rit., pp. 458, 501. fig. 13 (animal), figs. 14-23 (habitats: tree trunks, birds' nests, bushes, etc.) 452 FIELDIANA: ZOOLOGY, VOLUME 46 ARGENTINA: La Pampa; bubonic infection. De la Barrera, 1939, op. cit., p. 441, figs. 3, 8, 9 (habitats)— ARGENTINA: Mendoza (northeastern part) ; bubonic infection. G[raomys] griseoflavus griseoflavus, Yepes, 1935, Rev. Inst. Bacteriol., 7, (2): 224, pi. 4, fig. 3 (animal)— ARGENTINA: from La Pampa to Santa Cruz. Hesperomys Bravardi Burmeister, 1879, Descr. Phys. Rep. Argentine, 3: 228 —ARGENTINA: (type locality, Buenos Aires [Pleistocene]). Phyllotis chacoensis J. A. Allen, 1901, Bull. Amer. Mus. Nat. Hist., 14: 408— PARAGUAY: Chaco (type locality, Waikthlatingwayalwa, northwest of Asuncion); ARGENTINA: Salta (lower Rfo Cachi). E[ligmodontia] chacoensis, Thomas, 1902, Ann. Mag. Nat. Hist., (7), 9: 132 — classification, comparison. Graomys chacoensis, Thomas, 1916, Ann. Mag. Nat. Hist., (8), 17: 142 — clas- sification. Phyllotis cachinus Allen, 1901, Bull. Amer. Mus. Nat. Hist., 14: 409— ARGEN- TINA: Salta (type locality, upper Rio Cachi). E[ligmodontia] cachinus, Thomas, 1902, Ann. Mag. Nat. Hist., (7), 9: 132, footnote — classification. Graomys cachinus, Thomas, 1916, Ann. Mag. Nat. Hist., (8), 17: 142 — classi- fication. Thomas, 1919, op. cit., (9), 3: 494— ARGENTINA: Catamarca (Chumbicha); "Rioja" (Otro Cerro). Thomas, 1920, op. cit., (9), 6: 116 —ARGENTINA: Catamarca (near Tinogasta). Thomas, 1920, op. cit., (9), 6: 417— ARGENTINA: La Rioja (Potrerillo, Cordillera de Famatina). Eligmodontia griseoflavus centralis Thomas, 1902, Ann. Mag. Nat. Hist., (7), 9: 240— ARGENTINA: Cordoba (type locality, Cruz del Eje, 600 meters). Graomys griseoflavus centralis, Thomas, 1916, Ann. Mag. Nat. Hist., (8), 17: 142 — classification. Graomys sp., Thomas, 1919, Ann. Mag. Nat. Hist., (9), 3: 116— ARGEN- TINA: Catamarca (Chumbicha). Graomys medius Thomas, 1919, Ann. Mag. Nat. Hist., (9), 3: 494— ARGEN- TINA: Catamarca (type locality, Chumbicha, 600 meters). Thomas, 1926, op. cit., (9), 17: 603— ARGENTINA: Tucumdn (Tapia). Yepes, 1945, Rev. Argentina Zoogeogr., 4: 66— ARGENTINA: Santiago del Estero (Tacafiitas). Phyllotis griseoflavus medius, Cabrera, 1961, Rev. Mus. argentine Cienc. Nat. "Bernardino Rivadavia," 4: 495 — classification; synonyms, edithae Thomas, hypogaeus Cabrera. Graomys lockwoodi Thomas, 1918, Ann. Mag. Nat. Hist., (9), 1: 187 — AR- GENTINA: Salta (type locality, Manuel Elordi, Bermejo, 500 meters). Thomas, 1920, op. cit., (9), 5: 191— ARGENTINA: Jujuy (Villa Caro- lina). Thomas, 1926, op. cit., (9), 18: 194— BOLIVIA: Potosi (Tupiza, 2000 meters). Graomys taterona Thomas, 1926, Ann. Mag. Nat. Hist., (9), 17: 320— BO- LIVIA: Tarija (type locality, Tablada, north of Tarija, 2000 meters). [l]Graomys sp., Klihlhorn, 1952, Zeitschr. Sauget., 18: 120— BRAZIL: Mato Grosso (Rio Ivinheima). HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 453 Type.— Male, British Museum (Natural History) no. 55.12.24. 184; collected August 1833 by Charles Dai-win. Type locality. — Near mouth of Rio Negro, Province of Rio Negro, southern Argentina. Distribution (fig. 103).— From Santa Cruz, Argentina, northward through the Pampas and Chaco into the Chaco region of Paraguay and Bolivia and possibly southern Mato Grosso, Brazil; west into the western foothills of the Andes and Sierras Pampeanas in Argentina; in the departments of Tarija and southern Potosi in Bolivia; alti- tudinal range from sea level to approximately 2000 meters above. Characters. — Upper parts of adults buffy to tawny, basal portions of hairs of underparts white to plumbeous; bullae well inflated. Remarks. — Original measurements for combined head and body length (6" 8'" = 169 mm.) and for tail length (5" 6"' = 140 mm.) of the type of M us griseoflavus Waterhouse are obviously incorrect and may have been transposed. In this species, the tail is always longer than combined head and body length. Taxonomy. — Phyllotis chacoensis Allen is based on three specimens from the Paraguayan Chaco. They were compared with the type of griseoflavus and found to differ by brighter color and hairs of ventral surface wholly white, not "pale plumbeous" basal ly, as in griseoflavus. In 12 adults and subadults from Rio Negro, base of hairs of ventral surface vary from white to plumbeous; in 5 juvenals from the same province, the hairs are dark gray basally. In 15 skins from the Paraguayan Chaco, basal portions of hairs of ventral surface are dark gray in one subadult and one adult, pale gray in three adults, and white in ten specimens, including a Juvenal and a fully adult cotype of chacoensis. Paraguayan Chaco specimens are slightly more warmly colored on dorsal surface than Rio Negro representa- tives of griseoflavus. The prominent hair tuft at base of ears de- scribed for the holotype of chacoensis is absent in the cotype and all other specimens examined. Tail of cotype is well-haired, but with- out conspicuous pencil. Phyllotis cachinus Allen from the upper Rio Cachi, Salta, was found to be "similar to griseoflarus in general coloration but with the pelage of the ventral surface pale grayish at extreme base, in- stead of deep plumbeous [italics mine] for the basal half or more." Variability in color of underparts has already been dealt with. It is of interest to add, nevertheless, that the basal portions of the hairs of the underside of the type of griseoflavus appeared to Allen to be "deep plumbeous" in comparison with cachinus, and "pale plumbe- 454 FIELDIANA: ZOOLOGY, VOLUME 46 ous" in comparison with chacoensis. The specimen that was to be- come the type of cachinus Allen, and another from the lower Rio Cachi, Salta, referred by Allen to chacoensis had been examined by Thomas (1897, p. 215) who concluded "after the most careful com- parison of these Salta specimens with the type [of Phyllotis griseo- flavus Waterhouse] and with a skin from Catamarca . . . [that he was] still of the opinion . . . that the northern and southern forms cannot be separated." Eligmodontia griseoflavus centralis Thomas from Cordoba Prov- ince, Argentina, was distinguished from all previously described Argentine forms by smaller molars (4.5 mm.) and smaller bullae (6.0 mm.). Size of bullae of centralis is less than average, but still well within the range of variation for Argentine P. griseoflavus. The molars appear to be extremely small, but measured on the crowns they agree with those of an old adult from Cordoba (MACN No. 29.22). The alveolar length of the molar row is 5.5 mm. in the latter. This specimen also agrees in all other respects with the de- scription and measurements of the type of centralis. It cannot be distinguished from northern or southern Argentine representatives of griseoflavus. Phyllotis lockwoodi Thomas from Salta, Argentina, was described as a species with incisors more recurved than in related forms, hind foot larger than in domorum, and bullae larger than in domorum, smaller than in chacoensis and cachinus. The form of the incisors is too variable to merit consideration. According to measurements provided by Thomas, hind foot of lockwoodi is 32 mm., that of do- morum, 31. Auditory bullar length of lockwoodi, 6.7 mm. according to Thomas, lies well within the limits of variation of specimens from the type region of cachinus (6.0-7.3) and from the type region of chacoensis (5.9-7.1). It seems that the status of lockwoodi is no dif- ferent from that of other Salta specimens (see cachinus, p. 453), so emphatically referred to typical griseoflavus by Thomas. A series of thirty specimens collected by E. Budin in Chumbicha, Catamarca, Argentina, was first recorded as representative of an un- identified species "most nearly allied to G. chacoensis and lockwoodi." Later, Thomas (1919c, p. 494) discovered to his "surprise . . . that those specimens belong to no less than three species — large, middle, and small, — distinguishable almost entirely by size, though, as is not unusual in such cases, the development of ears, tail, and tail-tufts are in proportion to the general size, the larger species being finer animals throughout, with more handsomely tufted tails. There are no speci- HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 455 mens of a size to make determination difficult, and the series of the three species contains examples with fully worn teeth. The largest species, G. cachinus, has a skull-length of 33.5-35 mm." The middle- sized species was described as Graomys mediits. Its skull length is given as 31.2 mm. The smallest species, with skull length 28.5, was called edithae. Of the original thirty specimens from Chumbicha, ten were identified as mediits, six as edithae. The remaining 14 speci- mens were assigned, presumably, to cachinus. In segregating three species on size, Thomas relied on the mis- leading criterion of a constant and direct relationship between size and age as determined by molar abrasion. The type of variation in- volved has been discussed under the heading of Crop and Niche Variation (pp. 36, 38). It will be noted in any case that size difference, admittedly the only difference, between medius and specimens Thomas assigned to cachinus, allows for practically no individual variation and is, rather, what is expected between individuals of a series as large and repre- sentative as the one from Chumbicha. Measurements given for edithae, on the other hand, are surprisingly small for an adult and even for a subadult, of the common species. The "supraorbital edges without beading" is also a noteworthy character, whether or not the type of edithae is fully adult. Whereas medius and cachinus are un- hesitatingly assigned to the synonymy of griseoflarits, judgment on the status of edithae must be deferred until the type can be examined. At present, it is listed as a species perhaps nearest related to I'hyl- lotis hypogaeus Cabrera. Since the above was written (prior to 1957) Cabrera (1961, p. 495) has included edithae Thomas and hypogaeus Cabrera in the synonymy of Phyllotis griseoflarus medius on the basis of the following opinion, freely translated: "The small difference in size used by Thomas as basis for distinguishing three species [edithae, medius, each in its] in a series of otherwise quite similar specimens does not seem to be suffi- cient grounds for the establishment even of subspecies. The differ- ence seems to be a common population characteristic of frequent occurrence in other kinds of rodents. My Graomys hypogaeus, from Corral Quemado, Catamarca, 2250 meters altitude, is indistinguish- able from some specimens from a lower elevation in southern Cata- marca and which are undoubtedly referable to medius." As already noted, I agree with Cabrera regarding the insignifi- cant difference in size between the Chumbicha specimens determined as medius and cachinus by Thomas. On the other hand, cdithae is «! «• cc be 0) C42 N 1 CD ji> g ^O^ CO CO t-. _ «H oo "^* Oi .2 " t-* CD_ CD CO CO 13 ~* oT 10" eg" i 10 IO i— 1 Od i I i O O3 0 O « CD CD CD id id CD id 1 ' t> id co' t> eg 1 O3 CJ CD* 10" id '"^5 »-< t-' i-H_ 2 ^ •^f 03 03 10" C c3 oo CD I* t~-' co O >42 T-H IO 03 eg t*~ O3 O3 N OO' CD i— 1 id oo co i-H T-H i-H cd id 1 O3 i CD I 1-1 r-l ' T-t ' ,-H 1 ^ 1 4J DO o> t- 1 co" 03 ^43 CO CO i— 1 CO 2 "S) co" t- oo" r-i 10 .g t-* co O Tjl eg 3 oo" Tt" o" oq 10 eg" cT t~; 10 eg GQ CO CO co oo T-3 co co co co ^-3 1 oo i i •«* i-J co co ' co 1 co co co ^* co co W CO I < •d a O>42 w •!T OS £ 2 :r "3 W PH O O 4= W il § 8 § ^ 43 E oj O O hJ PQ 456 **• «-^ "~ v,^; «v Q} < § 2 "5 «= 10 ~ C 10 "T t- '. ^ICO lOTfC^r-i TfOi IO CO to " OiO lOio'iOiO lo'iO J= 0 o •M IB ^ TJjjl C-Q oo of co a> 3 r>^ X co "5" -5 5 °° ~* co" IO 3 P 5 So1 . -«- «v* . . » frH OCOIOCO t^t- •^ W CO t- ^ C< 3 CO CO t'™ CO CO to CO O-1 o to »H o> -H ""l. co" •~J J3 CO 06 o £12 t-' .2 1 •£ O § -i « ryf IO 5 >i-Q 1 ^ ^ t2 rt s; •*? t-T * s • < tO O Tt co' oo' t— cr 5 CO t"~ t^ 00 OO ~ oo 0 •£ «. 2 ^ r_- t"J o *-> . * , co .* g) *j co co ^ &£? rjT co co" CO CO oo -J1 a> oi co co Oi ^ — ~" co co co ^1 g ^ co" 10" co" o 1 CO O CO C^J CO ce C/2 <-i 10 co — < 9-* \ » .T ^J* w o> •M 'C . C • ' o» o • r* V C < co coTj«iooi iot-- c ?: v- a c •— Ol M OJ OJ W Ol Ol O -2 * tC O C Q. **^ 1- '-5 a. "^ o- '-^ 2 'C O 00 2 o ^i F «2 ^ ^1 a> -I -1 ^ f 'Sc|f.S-^ co w & «c v*' o "^ E "2 ?> 1 2P iO "C 2_"CaI":8 a! .S I . ^1 "— «« PS MM "^ C Npl OJ [ f- t- 3 * 01 r -M 0 00 COOt— t^ Ol^l1 e c£.£ M 01 0^1 01 01 co co .- 'C, c 5 .~ ^ -r .Sf -n i 'C S .E: c o> o" C £ . c 'C 5 _ | 'So to" o ° a- o T £ s? i ijliM. S "^. ^1 O-O5 *~" ""*••*•*£ ^ t^» » X) «-H O U —i IO t^ IO co-oioito 010 • •/' i - •— 2 •«*• coiot-co loco g > "> e 9 * ^ n ""1= p:Sc£SJ .** — o> oo* P-H ^ Tf i— ( < J x £ J 5 E- c ^] " _^ T- 10" 0>" ™ o ^ *™^ »2 sl^^ si 0- C . S ^~ $ K * * 01 |" co ri 01 S2 co i- -r co ~ o, S S i 1 "S "f -S "I 01 ococooi coio -*c ?4-<*~~5 = 7s & OgCSCCCGCC "S •£ ^ c o o )5 •— 2 .1, c cx5?'&c.Q.c.cxc. - a - 6 S ^^23= E ± p rt ^ Q- "3 "2 •« ££\£££££££ "«»u«4i20a< .T. § C §<§a 3 Q o« O a> JC 'OK c— •• S OcS-pCsW": i-3 w U OQ O j E 5 S J5 S J 457 458 FIELDIANA: ZOOLOGY, VOLUME 46 distinguished not only by an appreciable difference in size, but by a cranial character which should not be dismissed without good reason. I am also unwilling to accept at face value Cabrera's identification of his hypogaeus with Phyllotis griseoflavus. Some explanation of what would now be regarded as serious errors in the description of the type and depiction of its molars (cf. fig. 107) is in order. Graomys taterona Thomas from the Bolivian headwaters of the Rio Bermejo is indistinguishable in external characters from Rio Pilcomayo mice already described as chacoensis (= griseoflavus) . The bullae of the type of taterona were said to be smaller than in lockwoodi and chacoensis, larger than in domorum. No measurements were given in support of the statements. Two specimens of the ten com- posing the original type series are at hand. Their skulls are badly damaged but bullae, intact in one, a fully developed adult, measure 6.0 mm. The length indicates an intermediate position between griseoflavus and domorum. On a geographic basis, however, the type of taterona is best assigned to griseoflavus. Measurements. — See Table 61. Specimens examined. — 78. ARGENTINA. — Chubut: Rawson, 1 (MACN); Rio Negro: Chimpay, 9 (CNHM); Choele-Choel, 5 (MACN, 2; CNHM, 3); Pichi Mahuida, 1 (CNHM); Buenos Aires: Algarrobo, 1 (MACN); Cordoba: La Paz, 2 (MACN); San Juan: Media Agua, 3 (AMNH); La Rioja: General Roca, 2 (MACN); Tucuman: Tapia, 1 (MACN); Catamarca: Chumbicha, 1 (CNHM); Puntilla, Tinogasta, 2 (CNHM); Bel<§n, 6 (CNHM); Santiago del Estero: Lavalle, 16 (AMNH) ; Satia: El Desmonte, Metan, 4 (MACN, 3; CNHM, 1) ; Macapillo, Metan, 2 (MACN); La Represa, Metan, 1 (CNHM); Chaco: Avia Terai, 1 (AMNH); Las Palmas, 1 (MACN). PARAGUAY.— Chaco: La Urbana, 1 (CNHM); Jesematathla, 1 (CNHM); Colonia Fernheim, 2 (CNHM); Guachalla, 5 (CNHM); Puerto Casado, 3 (CNHM); Waikthlatingwayalwa, 1 (AMNH); Waikthlawaya, 1 (AMNH). BOLIVIA.— Santa Cruz: Guanacos, 2 (CNHM); Tarija: Tablada, 2 (CNHM). ?Department: Rio Pilco- mayo, 1 (AMNH). Phyllotis griseoflavus domorum Thomas Eligmodontia domorum Thomas, 1902, Ann. Mag. Nat. Hist., (7), 9: 132. Thomas, 1902, op. cit., (7), 9: 223— BOLIVIA: Sucre (Sucre); Potosi (El Cabrado). E[ligmodontia] g[riseoflavus] domorum, Thomas, 1902, Ann. Mag. Nat. Hist., (7), 9: 240 — comparison. HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 459 Graomys domorum, Thomas, 1916, Ann. Mag. Nat. Hist., (8), 17: 142 -classi- fication. Phyllotis (Graomys) domorum, Osgood, 1916, Field Mus. Nat. Hist., Zool. Ser., 10: 207— BOLIVIA: Parotani, Cochabamba. Type. — Adult male, British Museum (Natural History) no. 1.1.1.47; collected 23 March, 1901, by Perry O. Simons. Type locality. — Tapacari, at headwaters of Rio Rocha, about 50 kilometers southwest of Cochabamba, Department of Cochabamba, Bolivia; altitude, approximately 3000 meters above sea level. Distribution (fig. 103). — Bolivian highlands, from the head- waters of Rio Arque in the Lake Poopo drainage basin eastward into the Rio Grande (upper Rio Mamore") drainage basin, departments of Cochabamba and Santa Cruz; altitudinal range, from approximately 2000 to about 3000 meters above sea level. Characters. — Color of upper parts of adults from olivaceous, as in juvenals of typical griseoflavus, to rufous; hairs of ventral surface dark basally; bullae comparatively small, their average antero-poste- rior length, less tubes, approximately equal to average length of molar row. Variation. — There are no significant differences between speci- mens from the Rio Grande on the Amazonian slope of the Cordillera Real and specimens from the upper Rio Arque in the Lake Poopo basin across the divide. Taxonomy.- The original series of domorum was distinguished from other representatives of the species by "less bushy tail and smaller bullae." Degree of hairiness of tail varies individually and seasonally. On the other hand, bullae in 3 specimens from Parotani, a locality in the type region, are less inflated than in representatives of the nominate race. In the vast majority of specimens of P. g. domo- rum, the antero-posterior length of bullae, less tubes, is less than alveolar length of molar row. In connection with his description of domorum, Thomas remarked that "the most interesting point about this species is its extraordinary resemblance to Phyllotis Wolffsohni, caught at the same localities. Except that the Eligmodontia [i.e., domorum] has larger and clumsier feet, it is almost impossible to distinguish the two without examining the skulls. This resemblance in itself is a point in favour of the two animals being really generically different!!]" Measurements.— See Table 62. 311 , A 3s 3 ikull, greatest Zygomatic Bullar length breadth less 1 TjJ OO T-t t^s ui S^ «o ui id »— t s~ 1— 1 IO id i— i t- <^5 1 rH 1 (M i— I ao *4 CO 0 S2, co' w S3 T— 1 t-T T-) 00 oq 1 o C/J to co co eg co B 00 co co eo co 3) & U5 H •I u S S W O «o ' eg "M > 13 > 2S~ ^^lO o CM i— 1 o E- H S rO 0 cgiH us to 0 s ^ I "<*< CO CO imj 0 r. . — . ^H rH — J^ In 0) £ 1 O 1 Q a a rt § .s •^ C— i 1 i '5 .S" omarapa oT o, EH 1 H <; O 460 HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 461 Specimens examined. — 43. BOLIVIA. — Cochabamba: Parotuni, 33 (AMNH, 31; CNHM, 2); Tin Tin, 5 (CNHM); Aiquile, 3 (CNHM); Santa Cruz: Comarapa, 1 (CNHM); Florida, 1 (CNHM). Phyllotis edithae Thomas Graomys edithae Thomas, 1919, Ann. Mag. Nat. Hist., (9), 3: 495. Type.~Ma\e, British Museum (Natural History), no. 19.2.7.34; collected 26 September, 1918, by E. Budin. Type locality. — Otro Cerro, a collecting locality about 18 kilo- meters NNW of Chumbicha, Catamarca, Argentina; altitude, 3000 meters above sea level. Distribution (figs. 57, 108). — Known only from the type locality in the Andes of northern Argentina. Characters. — The original description follows: "Size again smaller than in G. medius, making it the smallest known species of the genus [Graomys}. Colour about as in that animal, a bufify wash on the sides rarely present. Under surface white, the hairs either slaty basally or white to their roots. Tail shorter than in medius, and less heavily haired terminally; brown above, white on sides and below. "Skull a miniature of that of the other species; supraorbital edges without beading. "Dimensions of the type:— "Head and body 108 mm.; tail 127; hind foot 25; ear 20. "Skull: greatest length 28.5; condylo-incisive length 26.5; zygo- matic breadth 15; nasals 10.5; interorbital breadth 4.5; breadth of braincase 13.5; palatilar length 12.8; palatal foramina 6.7; length of bullae 6; upper molar series 4.7." Remarks. — P. edithae was described as the smallest species of the group typified by P. griseoflavus. As a species of Phyllotis, however, it is larger than P. hypogaeus, gerbillus and amicns. Absence of supraorbital beading in the skull of edithae is also a character of gerbillus and amicus. The status of edithae must remain obscure until the enamel pattern of its molars can be studied. In addition to the type from Otro Cerro, Thomas recorded 6 specimens from Chumbicha, Catamarca. They were taken together with representatives of cochin us and medius, both synonyms of /'. griseoflavus griseoflavus. The possibility that edithae may also be a synonym is discussed above (footnote 1, p. 448, and p. 455). Specimens examined. — None. 462 FIELDIANA: ZOOLOGY, VOLUME 46 Phyllotis hypogaeus Cabrera1 Graomys hypogaeus Cabrera, 1934, Notas Prelim. Mus. La Plata, 3, (la): 124, fig. (molars). Type. — Adult male, Museo de La Plata, Universidad de La Plata, Buenos Aires; collected 25 February, 1930, by Angel Cabrera. Type locality. — Corral Quemado, Bele"n, Catamarca; altitude 2250 meters above sea level. Distribution (figs. 57, 108). — Known only from type locality in the Andes of northern Argentina. Characters. — The original description, translated from the Span- ish, follows: "Size less than in any other known species of Graomys; pelage of upper parts pinkish buff with a grizzling on the back and forehead, the hairs slaty basally; underparts and feet white, the hairs white to the roots; a white postauricular patch present as in Eligmo- dontia typus and Hesperomys bimaculatus. Pencilled tail white with tawny-olive stripe on upper surface, the stripe sharply defined, nearly black at terminal portion, but hardly distinguishable basally. "Skull with interorbital region expanded behind, the edges finely ridged ; anterior border of zygomatic plate slightly concave, the upper corner projecting. Molars similar to those of Phyllotis darwini. "Measurements: Head and body, 80 mm.; tail, 114; hind foot, 23; ear, 18; greatest length of skull, 23; condylobasal length, 22; zygo- matic breadth, 13.5; interorbital constriction, 4; palatilar length, 10.3; palatine foramina, 5.5; nasals, 8; auditory bulla, 4.5; diastema, 6; upper molar row, 3.6; lower molar row, 4. "This mouse, apparently the smallest species of the genus Gra- omys, resembles G. edithae Thomas of southeastern Catamarca (type locality, Otro Cerro, eastern slope of the Sierra de Ambato) but is smaller, its sides with a buffy wash not present in edithae, and tail with a well developed pencil. It resembles Eligmodontia typus super- ficially, but differs in the structure of the molars, and by its naked soles and pencilled tail." Habits and habitat. — The type of P. hypogaeus was discovered by Cabrera living among the large stones inside the tombs of pre- historic Indians. Nests of the mice were nearly spherical in shape and formed of dry grasses. The natives call the little animal raton punero. The region around Corral Quemado is described as arid, the sandy soil supporting xerophytic vegetation with many car- 1 See footnote, p. 448. HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 4ti3 dones. Phyllotis griseoflavus and Eligmodontia typus live in the same locality. B FIG. 107.— Molars of Phyllotis hypo- gaeus (after Cabrera): A, left upper; B, left lower. Remarks.— The mouse described by Cabrera may be unique. The enamel pattern of the molars (fig. 107) of P. hypogaeus is like that of comparably worn molars of Phyllotis darwini. The described cranial characters, however, agree with those of P. griseoflavus, or possibly P. amicus. Measurements indicate hypogaeus may be the smallest species of Phyllotis. Characters of color of sides of body and degree of hairiness of tail adduced by Cabrera for distinguishing hypogaeus from edithae are individual variables and conflict with statements in the original description of the latter. On the other hand, edges of supraorbital region of hypogaeus are described as finely ridged or beaded, while those of edithae are said to lack beading. Taxonomy. In his posthumously published second part of the Catdlogo de los mamiferos de America del Sur, Cabrera (1961, p. 495) disposes of his hypogaeus (and edithae Thomas) in the synonymy of Phyllotis griseoflarus medius. This action carried with it no explana- tion for certain diagnostic characters attributed to hypogaeus which are opposed to those of Phyllotis griseoflavus. For additional com- ments see Phyllotis griseoflarus griseoflavus. Specimens examined. None. 464 FIELDIANA: ZOOLOGY, VOLUME 46 Genus Galenomys Thomas Galenomys Thomas, 1916, Ann. Mag. Nat. Hist., (8), 17: 143 — subgenus of Euneomys Coues. Gyldenstolpe, 1932, K. Sv. Vet. Akad. Handl., 11: 96— full genus. Ellerman, 1941, The families and genera of living rodents, 2: 451,452,455 — subgenus of Phyllotis. Osgood, 1947, Journ. Mamm., 28: 171 — subgenus of Phyllotis. [Galenomys], Thomas, 1926, Ann. Mag. Nat. Hist., (9), 17: 317— former sub- genera of Euneomys regarded as full genera. Type species. — Phyllotis garleppi Thomas, by original designation. Included species. — Galenomys garleppi Thomas. Distribution Figures 52, 108 The altiplano of western Bolivia, southern Peru and extreme northeastern Chile. Characters External. — General appearance vole-like, color pale, size medium, body stout, tail short; upper parts of trunk buffy thinly lined with brown, sides clearer, rump ochraceous, head paler than trunk; under- parts and legs sharply defined white, the individual hairs wholly white on legs, feet, hands and chin to anterior part of chest, slaty basally on remaining underparts; ears large, well haired on outer side; fore and hind feet white above, hind feet (fig. 3) beneath thinly cov- ered with long white hairs except on tubercles and terminal pha- langes; tail from approximately one-third to nearly one-half combined head and body length, thickly haired white above and below, the scales not visible, terminal tuft not conspicuous. Cranial (figs. 61-63, 109, 110). — Dorsal contour of skull extremely arched, rostrum inflected downward, its dorsal plane forming an angle of about 30 degrees with plane of frontals; supraorbital re- gion narrow, constricted mid-frontally, the edges square, not ridged or beaded, interparietal moderately developed; zygomatic arches deli- cate, as widely expanded anteriorly as posteriorly and with a pro- nounced downward inflexion; fron to-parietal sutures crescentic in outline, not forming an angle at midline, distance across sutures on dorsal surface more than greatest width across fronto-maxillary sutures and more than alveolar length of molar row; anterior zygo- matic plate nearly as wide as least interorbital breadth, the anterior border straight and slanted backward, upper anterior corner rounded; palate arched in conformity with dorsal curvature of skull, its width between first molars more than length of m1; incisive foramina com- Galenomys garleppi Phyl/of/s hypogaeus edithae 20 Fie. 108. -Type localities of Galenomyx garleppi, I'hylloti* hitpogarnn and P. edithae: I, Esperanza, Mount Sajama, La Paz, Bolivia; '1, Corral Qwmado, Belen, Catamarca, Argentina; '\, Otro Cerro, Catamarca, Argentina. 465 466 FIELDIANA: ZOOLOGY, VOLUME 46 paratively well opened, especially posteriorly; pitted posterolateral palatal depressions shallow; mesopterygoid fossa narrow, its width at anterior base of pterygoid processes less than width of m^ or about one-half width of parapterygoid fossa measured at same plane; bullae moderately inflated, length, less tubes, less than alveolar length of molar row. Dental (figs. 62, 63, 69, 111). — Incisors delicate, ungrooved, for- ward projecting, the lower remarkably procumbent; upper molar rows bowed outwardly; upper and lower first molars 4-rooted; crowns low, slightly terraced, cusps ovate; second and third upper molars each at least as wide as long, anteroloph of second moderately devel- oped, of third weak; postcingulum absent in worn upper molars, present in lower; minor fold present in lower first and second molars, absent in worn uppers. Comparisons Superficially, Galenomys most nearly resembles pale individuals of Phyllotis sublimis. It is distinguished by relatively shorter, more thickly haired tail, and stouter hind feet with hairy soles. In the smaller, longer-tailed Eligmodontia, the thickly haired part of the sole is restricted to the surface of the three coalesced middle post- digital tubercles. In contrast the plantar tubercles of Galenomys are bare and not fused. Galenomys also resembles Phyllotis sub- limis in cranial characters, but the downwardly flexed rostrum and low-slung zygomatic arches of the first occur in no other phyl- lotine. The skull of the sigmodont Reithrodon is comparable in outline but the dorsal contour is evenly convex, or leporine, not abruptly bent downward at rostrum; the zygomatic arches are almost as low as in Galenomys, but slightly more expanded posteriorly than anteriorly, the orbits much larger. Because of the inward rotation of m-, the appearance of outward bowing of the upper molar rows is extreme in Galenomys. Otherwise, the molars are similar to those of P. sublimis, but relatively wider. The lower incisors of Galenomys attain a degree of procumbency not equaled in any other living cricetine. However, in Bensomys Gazin of the Kansas Pleistocene, the lower incisors are similarly procumbent, and the lower molars show the same enamel pattern as in Galenomys. The equally short-tailed, vole-like Punomys lemminus Osgood bears a superficial resemblance to Galenomys garleppi. Both species inhabit the altiplano and possibly may occur in the same parts. Punomys is larger, darker in color, its hind feet with but a few scat- FIG. 109. Dorsal and ventral aspects of skulls of a, Galrnomys gnrlrppi; b, Phyllotis sttblimis. (about X 2.) 467 468 FIELDIANA: ZOOLOGY, VOLUME 46 tered hairs between the plantar tubercles, ears considerably shorter, tail bicolor. Cranially, Punomys is intermediate between Phyllotis and sigmodont rodents; the enamel pattern of its molars, however, is unique. Taxonomy The vaulted skull, low slung, anteriorly expanded zygomata, en- larged first upper molars, uniquely procumbent lower incisors, extremely hairy soles and very short, hairy tail are individually distinctive characters that together justify generic rank for Gale- nomys. None of the species heretofore referred to as Auliscomys, erstwhile subgenus of Phyllotis, show such absolute distinctions. Each of them is distinguished from all other species of Phyllotis by relative differences and by the manner in which common char- acters are combined. Phyllotis sublimis appears to be nearest related to Galenomys. When first described, Galenomys garleppi was doubtfully referred to Phyllotis. It was compared with grasshopper mice of the genus Onchomys because of a superficial resemblance. In 1916, Thomas reclassified some South American Muridae and erected Galenomys as one of three subgenera of Euneomys. In 1926, however, Thomas decided that the "three subgenera into which Euneomys has been divided, largely on Mr. Osgood 's work, being unnatural groups, I propose to consider them as full genera, as this course is so far more convenient both for labelling and reference" [italics mine], Ellerman (1941, supra cit.) and Osgood (1947, supra cit.) preferred to regard Galenomys as a subgenus of Phyllotis. Galenomys garleppi Thomas Phyllotis (?) Garleppii Thomas, 1898, Ann. Mag. Nat. Hist., (7), 1: 279. Euneomys (Galenomys) garleppi, Thomas, 1916, Ann. Mag. Nat. Hist., (8), 17: 143 — type of subgenus. Galenomys garleppii, Gyldenstolpe, 1932, K. Sv. Akad. Handl., 11: 96 — de- scription. Phyllotis [(Galenomys)] garleppii, Ellerman, 1941, The families and genera of living rodents, 2: 452, 455. Osgood, 1947, Journ. Mamm., 28: 171. Phyllotis (Galenomys) garleppi, Pearson, 1957, Breviora, Mus. Comp. Zool., no. 73: 2, fig. 1 (skull) — PERU: Puno (Pichupichuni, 5 miles northwest of Huacullani, 12,600 feet; Pampa de Ancomarca, 76 miles south of Have, 13,700 feet). Type. — Adult male, skin and skull, British Museum (Natural History), no. 98.3.16.5; collected May 20, 1897, by Gustav Garlepp. HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 469 FIG. 110. — Side view of skull of a, Galenomys garleppi; b, Phyllotis sublimis. (about X l}4.) Type locality. — Esperanza, Mount Sajama, Pacajes Province, La Paz, Bolivia; altitude, 4000 meters above sea level. Distribution (figs. 52, 108). — The puna zone in the depart- ments of La Paz, Bolivia, Puno, Peru, and bordering parts of Tara- paca, Chile; altitudinal range from approximately 3800 to 4500 meters above sea level. Characters. — The characters of the species are those of the genus. The mammae of a female at hand are not sufficiently developed for determination of their number. The type is said to be a male, but Gyldenstolpe (supra cit.) gives the mammary formula for Galenomys as 2-2=8. Remarks. — The procumbent lower incisors of Galenoniys recall those of insectivores and caenolestids while the downwardly bent rostrum is comparable to the condition in the cuniculoid Reilhrodnn as well as to that in leporids in general. Nothing is known of the habits of G. garleppi. Galenomys garleppi is known from the type, another Bolivian specimen in the British Museum (Natural History), the specimen at hand in Chicago Natural History Museum, and two specimens in the Museum of Vertebrate Zoology, Berkeley, California. The latter were collected and recorded by Pearson (supra cit.) while this paper was in press. FIG. 111. — Right upper and left lower molars of a, Galenomys garleppi; b, Phyllotis sublimis. 470 HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 471 Measurements. — Those of the type, a male, from original descrip- tion followed by those of a female topotype: head and body, 123 (from dry skin), 105; tail, 38 (from dry skin), 45; hind foot, 25 (dry), 25 (dry, with claw); greatest length of skull, — , 30.2; zygomatic breadth, 17.8, 16.7; nasals, 13.4, 13.1; interorbital constriction, , 3.9; brain- case, — , 13.9; zygomatic plate, width, ,3.3; bulla, length less tube, — , 5.0; incisive foramina, 7.2, 6.8; diastema, 8.7, 8.6; alveolar length of upper molar row, 5.1, 5.3. Specimens examined. — 1. BOLIVIA. La Paz: Esperanza, 1 (CNHM). < ..,..-. BOLJ VIA / ARGENTINA • Andimonys edax edax * " fineicaudatus FIG. 112. — Collecting localities of the subspecies of Andinomys edax. See opposite page for explanation. 472 HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 473 Genus Andinomys Thomas Andinomys Thomas, 1902, Nature, 65: 431 (6 March) abstract. Thomas, 1902, Proc. Zool. Soc. London, 1902: 116 (June, apud Duncan, 1937, Proc. Zool. Soc. London, (A), 107: 75). Yepes, 1935, Anal. Mus. argentino Cienc. Nat., 38: 333— revision. Osgood, 1947, Journ. Mammal., 28: 173 char- acters, comparisons. Type species. — Andinomys edax Thomas, by original designation. Included species. — Andinomys edax Thomas. Distribution Figures 52, 112 Highlands of southern Peru, Bolivia and northwestern Argentina. EXPLANATION OF FIGURE 112 Andinomys edax: collecting localities and collectors. Type localities in boldface. Andinomys edax edax (1-13) PERU (1-2) 1. Juli, Lake Titicaca, Puno. — O. P. Pearson at 12,500 feet. 2. Yunguyo, 6 miles south, Puno.- C. C. Sanborn at 13,000 feet. BOLIVIA (3-8) 3. Choro, Cochabamba. — J. Steinbach at 3500 meters. 4. El Cabrado. P. O. Simons at 3700 meters. 5. Sama, Tarija. — E. Budfn at 4000 meters. 6. Pinos, Tarija.— E. Budfn. 7. Tupiza, Potosf. — E. Budfn at 2000 meters. 8. Yuruma (Estancia), Potosf. —E. Budfn at 2200 meters. ARGENTINA (9-13) 9. Cerro Casabindo, Jujuy. - E. Budfn at 4800 meters. 10. Sierra de Zenta, Jujuy.- E. Budfn at 4500 meters. 10. La Laguna, Sierra de Zenta, Jujuy. —E. Budfn at 4500 meters. 11. Maimara, Jujuy (2230 meters).— E. Budfn. 11. Alfarcito, near Maimara, Jujuy.- E. Budfn at 2600 meters. 12. Cerro La Lagunita, Maimard, Jujuy. E. Budfn at 4500 meters. 13. Yala, mountains west of, Jujuy. — W. H. Osgood. Andinomys edax linficaudalns (14-19) ARGENTINA 14. Norco, Vipos, Tucuman. E. Budfn at 2500 meters. 15. Cerro San Javier, Tucuman. E. Budfn at 2000 meters. 16. Sierra de Tafi Viejo, Tucuman. E. Budfn at 2000 meters. 17. Aconquija, Tucuman. J. Mogensen at 3000 meters. 18. Rfo Vallecito. Catamarca. J. Crespo. 19. Otro Cerro, Catamarca. E. Budfn at 3000 meters. 474 FIELDIANA: ZOOLOGY, VOLUME 46 Characters External. — Large, heavy-bodied, pelage soft, lax, upper parts and sides of body drab, underparts gray, the slaty bases of the hairs showing through, tail from 60 per cent to approximately equal to head and body length, sharply bicolor, thinly haired and without terminal brush; fore and hind feet well developed, white above; heel well haired, plantar surface bare, the tubercles greatly enlarged but not fused; ears comparatively small, usually with a fine gray edging; mammae, 2 — 2=8. Cranial (figs. 61, 63, 113). — Dorsal surface of skull flattened; supraorbital region narrow, parallel-sided, the edges square or slightly raised; a slit-like fontanelle present between anterior halves of frontals. A second fontanelle sometimes present between poste- rior ends of frontals; temporal ridges obsolete or absent; interparietal well developed; zygomata moderately expanded, markedly wider posteriorly than anteriorly; fron to-parietal sutures meeting at mid- line to form a right or acute angle, distance across them on dorsal surface less than greatest width across fron to-maxillary sutures and less than alveolar length of molar row; nasals broad posteriorly, well expanded anteriorly; anterior border of zygomatic plate straight or slightly concave, the upper border pointed or projecting as a short process; incisive foramina comparatively well-open, elliptical in out- line; bony palate narrow anteriorly, and barely extending behind to posterior plane of third molar; greatest width between inner borders of first molars less than greatest length of m1; pitted posterolateral palatal depressions troughlike, continued as gutters on anterolateral portion of palate; mesopterygoid fossa comparatively broad, width at anterior base of pterygoid processes nearly equal to width of parapterygoid fossa measured at same plane; parapterygoid fossae shallow, especially anteriorly; bullae moderately inflated, length, less tubes, less than length of molar row. Dental (figs. 19, D, 63, 69, 114-116). — Upper incisors opisthodont, broad, ungrooved, external surface flexed inwardly toward midline to form a broad median valley, cutting edges meeting to form an approx- imately 90° angle, upper molar rows divergent posteriorly; upper first molar 3-rooted, lower 2-rooted; molars large, prismatic, flat-crowned; cusps of m|I-| triangulate; crown of ma involuted in unworn, roughly 8-shaped in worn, of m^ fused on lingual side; procingulum of m1 subtriangular, the primitive trilobate pattern hardly or not at all indicated; first primary fold of m^ well developed and persistent in adult, of m3- reduced to an enamel island in newly erupted and obso- FIG. 113. — Andinomys rdnr. Dorsal and ventral aspect,1* of skull. Not*- frontal fontanelles. (About X 2.) 475 476 FIELDIANA: ZOOLOGY, VOLUME 46 lete in slightly worn tooth; second primary fold of m-~^ extending little or not at all beyond longitudinal midline of molars; postcingu- FIG. 114.- — Right upper molars of a, Phyllolis griseoflavus; b, Andinomys edax. Note normal plane crown surfaces. lum obsolete in m1^, vestigial in Juvenal m-; first minor fold indi- cated in m3-, slightly more developed in m-; inner and outer cusps of first and second molars subequal; procingulum of mT trilobate in juvenal, bilobate with anterior median fold persistent in adult; first minor fold well developed in m^, shallower in m^; apices of major and first primary folds of mT^ touching; second primary fold of m^ short, extending about one-fourth distance across molar crown; first primary fold of m7 nearly obsolete, second primary absent. Comparisons The inwardly flexed upper incisors with cutting edges forming a sharp angle distinguish Andinomys from all other phyllotines. The interfrontal fontanelles are larger and more consistently pres- ent in Andinomys than in other cricetines. External appearance of Andinomys is not striking. The animal compares with Phyllotis griseoflavus and P. micropus in size, but resembles the latter in color, texture of pelage and, except for its slightly broader hind feet, in limb proportions. Enamel pattern of upper molars of Phyl- lotis griseoflavus is essentially as in Andinomys; the teeth themselves, HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 477 however, though high crowned are not prismatic, the cusps less sharply angulate, mr with procingulum simplified as contrasted with the advanced trilobate pattern of the same part of the tooth in Andi- nomys. Interfrontal fontanelles occur sporadically in a number of cricetines. They are common, though rarely well developed, in Phyllotis micropus, and nearly as prevalent in the sigmodont red- nosed vole Neotomys as in Andinomys. Also, in Neotomys, the high median ridge of the posterior portion of the palate, and the inwardly flexed incisors are as in Andinomys. The incisors of Neotomys, how- ever, are grooved, their cutting edges straight or forming a very obtuse angle, the molars notably compressed antero-posteriorly, ma longer than wide, larger than m^, and with second secondary and second minor folds present and well developed in adult. Andinomys is readily distinguished from Neotomys by these and any number of other external, cranial and dental characters. Molars of the phyllo- tine Chinchillula are as hypsodont and flat crowned as in Andinomys but their enamel pattern has been simplified to an 8-shaped figure in adult upper and lower second and third molars. The color pattern of Chinchillula is unique and cannot be confused with Andinomys. The prismatic crowns and triangulate cusps of the molars of An- dinomys recall those of the North American wood rat Neotoma (figs. 115, 116) but there are trenchant differences between the two genera in number and disposition of cusps and orientation of the folds. In Neotoma, apex of each of first and second primary folds of ml * ex- tends well across crown and nearly, or quite, touches enamel wall of opposite side; in Andinomys only the first primary fold extends nearly to enamel of opposite side; first minor fold of m- :l, absent in Neo- toma, is present, though weakly defined in Andinomys; m3 with three subequal lophs on outer side in Neotoma has but two subequal lophs in Andinomys; procingulum of m,, never trilobate in Neotoma, is trilobate in Juvenal Andinomys; outer main cusps of m, 2 one-half or less size of analogous cusps of inner side in Neotoma are subequal in Andinomys; in mr .j of Neotoma apices of first primary and first minor folds touch, and apices of second primary and major folds touch while in Andinomys only apices of second primary and major folds nearly or quite touch, the first primary being reduced; in Neo- toma m5 is 8-shaped, in Andinomys roughly C-shaped. Molars of Hodomys are slightly less specialized than those of Neotoma, m, with anterolophid present in Juvenal, m.T S-shaped. The marked tend- ency toward lamination, especially in m3, present in neotomyine 478 -r O •s§ ^ 's *4 E. _ I* 479 480 FIELDIANA: ZOOLOGY, VOLUME 46 rodents, is absent in Andinomys. Furthermore, no neotomyine shows incisors flexed inwardly as in Andinomys. Cranial characters dis- tinguishing Andinomys from Neotoma are too obviously different to require analysis here. Gross differences in the structure of the glans penis and baculum are discussed elsewhere (cf. p. 20). Variation The lone species of the genus is pale in the Lake Titicaca re- gion and becomes darker, as well as larger, southward into north- western Argentina. A Juvenal from Choro, Cochabamba, Bolivia, in the upper Amazonian drainage, is darker throughout than com- parable individuals from other regions. Except for a narrowing of the braincase and interorbital region, the general proportions of the skull are not notably modified by increasing age; the interf rental fontanelles are as prominent in old as in young individuals. Because of the high crowns, the enamel pattern of the molars varies widely with wear. Nevertheless, the diagnostic characters of the dentition are not obscured except in extremely worn teeth. Taxonomy Molars of Andinomys, and even more so those of Chinchillula, represent the extreme of hypsodonty in phyllotine rodents. Among long palate tetralophodont cricetines, hypsodonty with attendant prismatic shape of crown elements has gone even farther in the sigmodont coney rat, Reithrodon. A comparable evolution from bunobrachyodonty to columnar platyhypsodonty in cricetines with simple, or short, palates can be traced from Haplomylomys (ex Pero- myscus) to its culmination in Neotoma. The process of triangulation in Andinomys is repeated in several phyllotines, but in none are all triangular shaped elements closed. Triangulation is more advanced in neotomines and is extreme in microtines. In all forms considered the triangular outline of the dental elements appears in the unworn tooth but becomes progres- sively rounded during the course of wear. Four extinct genera from the Pleistocene of eastern Argentina indicated by Osgood (1947, p. 173) as resembling Andinomys were de- scribed by Ameghino in 1889 (pp. 111-120). They are Ptyssophorus and Tretomys, both synonyms of Reithrodon; the third genus, Bothri- omys is a Euneomys, and the fourth, Necromys, is the same as Calo- mys. Merriam (1894, p. 225) included the first two genera in his "subfamily" Neotominae. HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 481 Habits and Habitat Andinomys lives in temperate zone puna, or paramo, and scrub country together with Phyllotis darwini. It is terrestrial, but may be scansorial as well. Emilio Budin, who collected most of the specimens recorded by Thomas, noted (in Thomas, 1913, p. 140) that Andinomys "lives in branches of trees where it makes its nest." It is very probable that Budin's field notes referred to another animal. There is nothing in the structure of Andinomys, and very little in its habitat, to indicate arboreal habits. Thirteen years later, Budin observed (in Thomas, 1926b, p. 603) that the rat "lives in round holes, quite clean, and the inside is carpeted with very fine straw; feeds on green herbs." Whether or not the burrow was of the rat's own making was not determined. During an intervening excursion, Budin (in Thomas, 1919d, p. 129) found the Chozchorito, the name he gave Andinomys, "among the rocks on the bank of a stream." The peculiar upper incisors of Andinomys indicate some special feeding or food-seizing habits (cf. p. 106). Of three female Andinomys collected December 18, 1939, by Pearson (1951, p. 152) one contained three large embryos. Measure- ments given for the females, said to be adults, seem to be, rather, those of young individuals. Andinomys edax edax Thomas Andinomys edax Thomas, 1902, (6 March) Nature, 65:431. Thomas, 1902, (March), Ann. Mag. Nat. Hist., (7), 9: 225— BOLIVIA: Poto»\ (El Ca- brado, 3500 meters). Thomas, 1902, Proc. Zool. Soc. London, 1902: 116, pi. 9, figs. 1-3 (skull), figs. 4-6 (molars). Thomas, 1913, op. oil., (8). 11: 140— ARGENTINA: Jujuy (Maimara; Cerro de Lagunita; Maimara). Thomas, 1919, op. cit., (9), 4: 129— ARGENTINA: Jujuy (Cerro Casa- bindo). Thomas, 1921, op. cit., (9), 8: 611— ARGENTINA: Jujuy (Al- fareito; Sierra deZenta). Thomas, 1926, op. cit., (9), 17:320 BOLIVIA: Tarija (Pinos; Sama). Thomas, 1926, op. cit., (9), 18: 194 BO- LIVIA: Potoni (Tupiza; Yuruma). Gyldenstolpe, 1932, K. Sv. Vet. Akad. Handl., 11: 97, pi. 11, fig. 2 (skull of type), pi. 18, fig. 3 (upper molars of same). Andinomys edax edax, Yepes, 1935, Anal. Mus. argentino Cienr. Nat., 38: 342, fig. 1 (skull), figs. 2, lib r (molars), fig. 4a (tail) ARGENTINA: Jujuy (Maimara; La Laguna; Sierra de Zenta). Yepes, 1938, Rev. Centro Estud. Doct. Cienc. Nat. Buenos Aires, 2: 24 distribution. Sanborn, 1950, Mus. Hist. Nat. "Javier Prado," (A), Zool., no. 5: 10 PEKt": Puno (Yunguyo). Pearson, 1951, Bull. Mus. Comp. Zool., 106: 152 PERU: Puno (Juli, Lake Titicaca). Type.— Old female, skin and skull, British Museum (Natural History) no. 2.2.2.15; collected 20 September, 1901, by Perry O. Simons. 482 FIELDIANA: ZOOLOGY, VOLUME 46 Type locality. — El Cabrado, between Potosi and Sucre, Bolivia; altitude, 3700 meters above sea level. Distribution (fig. 112). — From the Lake Titicaca region in Peru southward through the Bolivian highlands into the Andes and Si- erras Pampeanas of Jujuy, northern Argentina and, probably, ad- jacent parts of the Argentine province of Salta; altitudinal range from 2000 to 5000 meters above sea level. Characters. — Those of the genus; tail brown above, sharply de- fined white beneath. Measurements. — See Table 63. Specimens examined.— 11. PERU. — Puno: 6 miles south of Yunguyo, 5 (CNHM). BOLIVIA— Potosi: Estancia Yuruma, 2 (CNHM); Cochabamba: Choro, 1 (CNHM). ARGENTINA.- Jujuy: Mountains west of Yala, 3 (CNHM). Andinomys edax lineicaudatus Yepes Andinomys edax, Thomas, 1919, Ann. Mag. Nat. Hist., (9), 3: 492— ARGEN- TINA: Catamarca (Otro Cerro, "Rioja"). Thomas, 1926, op. cit., (9), 17: 603— ARGENTINA: Tucumdn (Norco, Vipos). Andinomys edax lineicaudatus Yepes, 1935, Anal. Mus. argentine. Cienc. Nat., 38: 345, fig. 46 (tail)— ARGENTINA: Tucumdn (Cerro San Javier; Aconquija; Sierra de Tafi Viejo). Yepes, 1938, Rev. Centre Estud. Doct. Cienc. Nat. Buenos Aires, 2, (4): 24 (separate) — distribution. Type. — Adult female, skin and skull, Museo Argentine de Cien- cias Naturales, Buenos Aires, no. 26147; collected 17 August, 1926, by Emilio Budin. Type locality. — Cerro San Javier, Tucuman, Argentina; altitude, 2000 meters above sea level. Distribution (fig. 112). — Sierras Pampeanas of Tucuman and Catamarca (Otro Cerro, "Rioja," is in Catamarca) ; altitudinal range, 2000 to 3000 meters above sea level. Characters. — Tail with brown midventral line. Measurements. — See Table 64. Specimens examined. — 1. ARGENTINA. — Rio Vallecito, Cata- marca, 1 (MACN). HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 483 TABLE 63. —Measurements of Andinomys edax edax Thomas Head and Locality body Tail Hind foot Ear Skull, Zygo- Alveolar greatest matic length of length breadth molar row 107 31 33.6 18.7 7.5 132 145 26 25 37 Yala 182 — 32 — 36.6 20.8 7.1 Yala 172 135 32 — 35.4 19.5 7.8 Maimara4 172 134 28* 24 36 21 7.5 Maimara4 162 135 28s 27 — — — Maimara4 155 160 285 22 34 19 7.0 1 Juvenal. 'Dry. 3 Type, from original description. 4 Ex Yepes (1935, An. Mus. Arg. Cienc. Nat., 38: 347). 5 Without claw. TABLE 64. —Measurements of Andinomys edax lineicaudatiis Yepes Head and Locality body Tail Hind foot Ear Skull, greatest length Zygo- matic breadth Alveolar length of molar row San Javier1 171 130 30 25 37 20 8 San Javier2 156 144 31 26 33 18 8 Tafi Viejo2 176 146 34 26 38 20 7 Aconquija2 170 140 32 22 36 20 7 Vallecito 145 142 35' 34.5(ro.) 18.5 Of type, from original description. Of paratypes, from original description. Dry. 7.8 • Chinchillula sahamae FIG. 117.' — Collecting localities of Chinchillula sahamae. See opposite page for explanation. 484 HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 483 Genus Chinchillula Thomas Chinchillula Thomas, 1898, Ann. Mag. Nat. Hist., (7), 1 : 280. Thomas, 1902, Proc. Zool. Soc. London, 1902: 116, pi. 9, figs. 5, 7 (molars). Osgood, 1947, Jour. Mamm., 28: 174. Type species. — Chinchillula sahaniae Thomas by original desig- nation. Included species. — Chinchillula sahaniae Thomas. Distribution Figures 52, 117 From southern Cuzco, Puno and Arequipa departments in south- ern Peru, southward into Tarapaca, northern Chile and the depart- ments of La Paz and Oruro, western Bolivia; altitudinal range, approximately 3500 to 5000 meters above sea level. Characters External. — Largest of the phyllotines and most distinctively col- ored New World cricetine; pelage long, silky, upper parts of body buffy to tawny overlaid with blackish mid-dorsally; upper half of sides like back, except for large white hip patch continuous with EXPLANATION OF FIGURE 117 Chinchillula sahamae: Collecting localities and collectors. Type locality in boldface. PERU (1-10) 1. Cordillera de Sicuani, Cusco. I. Ceballns. 2. Cailloma, Arequipa. P. O. Simons; C. C. Sanborn and J. M. Schmidt at 14,500 feet. 3. Arequipa, Arequipa. P. O. Simons. 4. Sumbay, Arequipa. C. C. Sanborn at 13,500 feet. 5. Puerto Arturo, Puno. C. C. Sanborn. 6. Posoconi (Hacienda), Puno. C. C. Sanborn. 7. Picotani (Hacienda), Puno. C. C. Sanborn. 8. Collacachi (Hacienda), Puno. C. C. Sanborn. 9. Pairumani, Puno. O. P. Pearson at 13,000 feet. 10. Caccachara, Puno. O. P. Pearson at 16,000 feet. BOLIVIA 11. Esperanza, La Paz. G. Garlapp at 4200 meters. CHILE 12. Parinacota, Tarapaca. G. Mann. 486 FIELDIANA: ZOOLOGY, VOLUME 46 white of lower half of sides and underparts; a dark brown lateral line often present between upper and lower halves of posterior por- tion of sides; dark brown to black patch on lower outer side of rump extended over outer side of thighs; inner sides of rump white, base of tail like back; white tips of guard hairs of rump and thighs projecting well beyond cover- hairs; brown of sides continuing over sides of arms; muzzle white; ears brown with white patch behind; large preauricular tufts white; fore and hind legs and feet white, digital tufts usually concealing the comparatively weak claws, plantar surface smooth, bare, the tubercles large; tail approximately one- half to three-fourths length of head and body combined, white with a brown mid-dorsal stripe or band, well haired and terminating in a thin pencil; mammae, 2 — 2=8. Cranial (figs. 61-63, 118, 119). — Dorsal surface of skull flattened, supraorbital region narrow, parallel-sided, the edges square or raised; temporal ridges obsolete or absent; nasals comparatively slender, except at flared anterior portion; interparietal well developed antero-posteriorly; zygomata well expanded, evenly bowed; fronto- parietal sutures meeting at midline to form a right or acute angle; distance across them on dorsal surface less than greatest width across fron to-maxillary sutures and less than alveolar length of molar row; anterior border of zygomatic plate sloped backward from below up- ward, and often with a slight tubercle on middle portion, upper an- terior corner rounded; incisive foramina well opened; bony palate always extending well behind posterior plane of last molars; greatest width between inner borders of first molar equal to, or more than, greatest length of m1; postero-lateral palatal depression shallow; mesopterygoid fossa broad, width at anterior base of pterygoid proc- esses more than width of parapterygoid fossa measured at same plane; parapterygoid fossa deep in adults; bullae moderately in- flated, less than length of molar row. Dental (figs. 62, 63, 69, 119, 120).— Upper incisors long, orthodont or slightly opisthodont, anterior face smooth or, sometimes, weakly grooved; upper molar rows parallel-sided or slightly divergent pos- teriorly; upper first molar three-rooted, lower two-rooted; molars large, prismatic, crowns normally flat but sometimes secondarily terraced in old worn teeth; cusps sub triangular or subovate in outline; enamel pattern essentially 8-shaped; apices of opposing enamel folds never touching, each extending nearly to, rarely be- yond, midline of molars; m- with procingulum subtriangular in adult, a subovate subcylindrical lamina in Juvenal; postcingulum HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 48? FIG. 118. — Chinchillnla sahamae. Dorsal aspect of skulls, showing variation in size (less than X FIG. 119.— Chinchillula sahamae. Palatal aspect of skulls, showing variation in size (less than X I1 2). absent; m- - with pro- and postcingula absent, each molar with only a second primary and major fold; m , 3 without postcingula; prooin- gulum of mr usually with an anterolingual stylid fused with a low, rudimentary anterolophid and nearly, or quite, isolating minor fold from outer margin of tooth; m2 with minor fold indicated by a style- like anterolophid; first primary fold absent in m5 3. 488 FIELDIANA: ZOOLOGY, VOLUME 46 Comparisons Among American cricetines, only the sympatric Akodon jelskii pulcherrimus shows some resemblance to Chinchillula in color pat- FIG. 120. — Chinchillula sahamae. Right upper molars, showing longitudinal sulcus and secondary terracing resulting from extreme wear. tern. The Akodon has similar white auricular tufts, sharply defined white lower halves of head, sides, rump, and dark outer sides of upper arms and thighs. It is, however, considerably smaller in size, with tail dark, nose reddish, and without black and white markings on posterior portion of body. The Eurasian hamster, Cricetus, offers some interesting points of resemblance to Chinchillula in its white auricular patches and pale sides of head and body. The unique arrangement of two white patches separated by the dark rump patch in Chinchillula is, in Cricetus, translated to the fore part of the body where two pale patches are separated by a dark gular band. The most remarkable resemblance in color, however, occurs in the unrelated North American kangaroo rats of the genus Dipo- domys. Except for the absence of the dark rump patch in Dipodomys, the pattern is essentially the same in both genera. The skull of Chinchillula most nearly resembles that of Andinomys, but is larger throughout, nasals more tapered posteriorly, posterolateral palatal depressions not furrowed, mesopterygoid fossa wider, parapterygoid fossa deeper, interfrontal fontanelles absent, etc. Molars of Chin- chillula are as hypsodont as in Andinomys but more simplified in enamel pattern than in any other phyllotine. Variation The distinctive blackish and white markings of head, hips, rump and thighs are consistent, even in nestlings. There is some individ- HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 489 ual variation in ground color of back and upper half of sides, as well as in the intensity of the blackish overlay of back. Color and pelage of juvenals are precisely as in adults. Braincase and in- terorbital region are proportionately narrower in older than younger individuals. Faintly to moderately defined grooves on outer face of incisors are present in about 10 per cent of specimens exam- ined. Molars of a Juvenal with skull length 31.6, are unworn, the third just erupted. Larger specimens, including a Juvenal with skull length 32.7, have fully functional and abraded molars. The interparietal of a subadult (CNHM 49410) is paired, the median suture being complete, well defined and in line with the suture separating the parietals. Taxonomy Chinchillula is the best defined and most strikingly colored phyl- lotine. Its extreme specializations are large size, unique color pat- tern, angular fron to-parietal sutures, sloping anterior zygomatic plate and simplified enamel pattern of the molars. A tendency to specialization of the incisors is noted in the occasional presence of grooves; on the other hand, the hard palate and postpalatal fossae of Chinchillula are not greatly differentiated from the prim- itive phyllotine condition. Greater resemblance of Chinchillula to Andinomys in size, shape of skull and height of molars does not indicate a nearer relationship to it than to other phyllotines such as Phyllotis boliviensis and P. micropus. In external and dental charac- ters Chinchillula and Andinomys are widely divergent. Habits and Habitat Field observations on Chinchillula of the Lake Titicaca region in southern Peru were made by Pearson (1951, p. 148). Accord- ing to this authority, the rat lives in "rocky places at high alti- tudes. It is most often caught among boulders and along stone walls. The surrounding vegetation may be grass, tola, yareta, or many of the other plants common in the region. ... It is nocturnal and herbivorous. At times, its enormous stomach is distended by more than 12 grams of green, finely ground, vegetable matter. "Chinchillula is one of the rats known and distinguished by the natives, not so much because of its striking color pattern, but be- cause its exceedingly deep and fine fur makes it valuable commer- cially. The skins are used as trim, or made into robes containing more than 150 skins. I never heard of a person who could be con- 490 FIELDIANA: ZOOLOGY, VOLUME 46 sidered a professional Chinchillula trapper, but the skins were being sold by the natives at Santa Rosa [Puno] and Juli [Puno] for about 20 centavos apiece — the cost of a half-pound of rice. . . ." The Chilean naturalist, Guillermo Mann (1945, p. 87) discovered Chinchillula in the puna zone of the northern Chilean province of Tarapaca. He expressed concern that persecution of the rodent for its fine pelt might lead to extermination of the species. The pelage of other cricetines of the altiplano of Peru, Bolivia and the temperate zones of Argentina and Chile is as fine as, or finer than, that of Chinchillula. However, the larger size of the latter and, possibly, the greater ease with which it may be captured, may make its pelt a more profitable item of commerce. Chinchillula sahamae Thomas Chinchillula sahamae Thomas, 1898, Ann. Mag. Nat. Hist., (7), 1: 280. Thomas, 1900, op. cit., (7), 6: 469— PERU: Arequipa (Cailloma; Are- quipa). Thomas, 1902, Proc. Zool. Soc. London, 1902: 117, pi. 9, figs. 5, 7 (molars) — PERU: Arequipa (Cailloma; Arequipa). Mann, 1945, Bio- logica, Trab. Inst, Biol. Univ. Chile, fasc. 2: 87, figs. 39-41 (alimentary tract), pi. 33, fig. 6 (skin), pi. 34 (skull)— CHILE: Tarapaca (Parinacota). Sanborn, 1950, Mus. Hist. Nat. "Javier Prado," (A), Zool., no. 5: 10— PERU: Arequipa (Sumbay; Cailloma); Puno (Hacienda Collacachi; Haci- enda Posoconi; Hacienda Pico tani; Puerto Arturo). Mann, 1950, Invest. Zool. Chilenas, 1, no. 2: 5— CHILE: Tarapaca. Pearson, 1951, Bull. Mus. Comp. Zool., 106: 148 — PERU: Puno (Caccachara; Pairumani). Type. — Juvenal male, skin and skull, British Museum (Natural History) 98.3.16.6; collected June 25, 1897, by Gustav Garlepp. In the original description, the type was said to be "slightly immature." Measurements indicate it to be a nestling (cf. Thomas, 1902, Proc. Zool. Soc. London, p. 116, footnote). Type locality. — Esperanza, about 50 kilometers north of Mt. Sa- jama, Pacajes, southwestern La Paz, Bolivia; altitude, 4200 meters above sea level. Distribution (fig. 117). — As for the genus. The locality records for Bolivia and Chile are in the Lakes Titicaca-Poopo drainage basin. Pearson (supra cit.) includes "northern Argentina" in the range, but I find no published locality record for this country. Characters. — Those of the genus. Measurements. — See Table 65. Specimens examined. — 35, all in Chicago Natural History Mu- seum. PERU. — Cusco: Cordillera de Sicuani, 1; Puno: Collacachi, 4; Picotani, 9; Posoconi, 2; Puerto Arturo, 1; Arequipa: Cailloma, 15; Sumbay, 3. t- *^ ^ —- c. > fcijs eg boC >.J2 s: ** * QJ "3"" _li w 0} — i •— I t~. t— |^ ad ao" CT> o ad CD CD' oo • - CO B^ 01 CD ~f rr o D r~ E g •J ^ U w £ •- c - •£ H "S j: £ £: r g 1 x | = ^ S. - ^ PQ 491 FIG. 121. — Type localities of nominal species and subspecies of Euneomys chinchilloides. 492 HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 493 Genus EUNEOMYS Coues Reithrodon, Waterhouse, 1839, Zool. Voy. "Beagle," p. 68— part, ft. chinchil- loides Waterhouse, only. Euneomys Coues, 1874, Proc. Acad. Sci. Philadelphia, 26: 185 -subgenus of Reithrodon; type by monotypy, R. chinchilloides Waterhouse. Coues, 1877, Monogr. N. A. Rodents, pp. 118, 119. Thomas, 1901, Ann. Mag. Nat. Hist., (7), 8: 254— genus. J. A. Allen, 1903, Bull. Amer. Mus. Nat. Hist., 19: 194— taxonomic history. J. A. Allen, 1905, Princeton Univ. Exped. Patagonia, 3, (1): 66 — taxonomic history. Thomas, 1916, Ann. Mag. Nat. Hist., (8), 17: 142- — characters; classification. Bothriomys Ameghino, 1889, Act. Acad. Nac. Cienc. Rep. Argentina, 6: 118 — type, B. catenatus Ameghino (Pleistocene). [!}Chelemyscus Thomas, 1925, Ann. Mag. Nat. Hist., (9), 15: 584-585— type, Reithrodon fossor Thomas. Type species. — Reithrodon chinchilloides Waterhouse. Included species. — Euneomys chinchilloides Waterhouse, K. noei Mann, E. mordax Thomas, E. catenatus Ameghino, E. fossor Thomas. Distribution Figure 121 Andes of Chile and Argentina from approximately 33° south lati- tude to Cape Horn. EXPLANATION OF FIGURE 121 Type localities of species and subspecies of Euneomys. Euneomys fossor Thomas 1. "Salta Province, N. Argentina." Precise locality is unknown but arbi- trarily shown in southwestern Salta. It is doubtful if the range of Euncomy* extends into this province. Euneomys catenatus Ameghino 2. Ccrdoba, C6rdoba, Argentina. The type and only specimen is a Pleistocene fossil. Living representatives of Euneomys are not known to occur in C6r- doba Province. Euneomys noei Mann 3. Valle de la Junta, Canyon of the Rfo Volcan, Santiago, Chile. Euneomys mordax Thomas 4. San Rafael, Mendoza, Argentina. Euneomys chinchilloides petersoni J. A. Allen 5. Upper Rfo Chico, near the Cordilleras, Santa Cruz, Argentina. Euneomys chinchilloides chinchilloides Waterhouse 6. Straits of Magellan, near eastern entrance, Tierra del Fuego, Chile. 494 FIELDIANA: ZOOLOGY, VOLUME 46 Characters External. — Size large, body heavy, vole-like, tail from 40 to 70 per cent of head and body length, brown above, white beneath, thinly haired, without well defined terminal tuft; ears moderate; hind foot broad, white above, fifth hind toe, less claw, extending to base of second phalanx of fourth toe, first hind toe to slightly be- yond base of second; heel thickly haired, plantar surface of sole bare, lightly scutulated, with 6 well developed pads, claws weak; pelage thick, lax, the hairs crinkled; upper parts of body mixed buff and black, gray underparts not sharply defined, washed with buff; an ochraceous lateral line sometimes present. Cranial (figs. 61-63, 122). — Dorsal contour of skull convex; su- praorbital region narrow, parallel-sided, the edges square or raised, never beaded; interparietal well developed; zygomatic arches broadly expanded posteriorly, greatest width across them more than dis- tance from posterior tips of nasals to anterior border of supra- occipital; anterior edge of outer zygomatic plate straight, the upper corner rounded; portion of maxillo-premaxillary suture below ant- orbital foramen directed forward in a line approximately parallel to longitudinal plane of palate; incisive foramina narrow, pointed pos- teriorly and extending behind anterior plane of first molars; posterior half of palate marked by a pair of large pitted depressions; meso- pterygoid fossa one-half or less width of parapterygoid fossa, measured at same plane; parapterygoid fossa moderately deep, subrectangular in outline, anterior wall sloped, not undercut; outer side of mandible with capsular projection strongly pronounced at base of coronoid process. Dental (figs. 62, 63, 69, 123). — Upper incisors opisthodont, their anterior faces marked by a well defined groove near midline; upper molar rows slightly divergent posteriorly; upper first molar 3-rooted, lower 2-rooted; crowns high, their occlusal surface plane, cusps loz- enge-shaped, the intervening enamel folds greatly compressed antero- posteriorly; enamel pattern of each worn upper and lower second and third molars distinctly S-shaped; procingulum of mT cylindrical in form, its occlusal surface cordate in outline and isolated from pro- toconid and metaconid, except in extremely worn tooth, by conflu- ence of minor and first secondary folds; first primary fold of m^~^ reduced in Juvenal, obsolete or absent in adult; postcingulum of upper molars vestigial, the second secondary fold reduced to a small enamel island in juvenals and persistent only in young adult m-; minor folds absent in mfl^; procingula of m^_^ obsolete; second primary fold FIG. l22.—Euneomya rhinchilloidfx. Dorsal and ventral aspects of skulls, and mandibles, showing variation in size (about X 495 496 FIELDIANA: ZOOLOGY, VOLUME 46 of m^ present in all but old adults, of m7 indicated only in unworn tooth; anterolabial stylid usually present and well developed. Comparisons In color and bulk of body, Euneomys resembles Andinomys and Phyllotis micropus, but the tail of the first is much shorter. Cra- nially, there is some resemblance to Phyllotis pictus, but the dorsal contour of the skull of Euneomys is more convex, though not arched as in Galenomys. Characters distinguishing Euneomys from all phyl- lotines include the nearly horizontal portion of the maxillo-pre- maxillary suture below the infraorbital foramen, antero-posterior compression of the enamel folds of the molars, lozenge-shaped cusps and procingulum, laminate and cylindrical procingulum of mT, distinct S-shaped pattern of moderately worn crowns of upper and lower second and third molars, and deeply grooved upper incisors. Involution of molars into a sigmoid pattern has gone farther in Euneomys than in true sigmodont rodents (Sigmodon, Holochilus, Reithrodon, Neotomys). In the latter, the pure sigmodont pattern is clearly defined only in m^. Euneomys is further separated from sigmodonts by less specialized hind feet, unmodified anterior border of zygomatic plate, comparatively shallow parapterygoid fossa, and the highly modified procingulum of mT. The procingulum of mT in Sigmodon also tends to lamination, and in the newly erupted tooth of some individuals may be laminate, but its outline is triangular and an anterior median fold is lacking. Taxonomy Euneomys is closely related to both phyllotine and sigmodont rodents without being a member of either group. The rat is a highly specialized form that retains many phyllotine characters, but goes far beyond all other cricetines in development of the sig- moid pattern in virtually all molars. The cylindrical, laminate procingulum of the first lower molar of Euneomys is also unique among cricetines. Inclusion of a revision of the genus in this mono- graph is justified by historical relationships, and the probability that students may, with reason, look for it here. The mandible and lower molars of Bothriomys catenatus Ame- ghino (1889, atlas, pi. 4, fig. 13,d) show all essential characters of Euneomys chinchilloides. Discrepancies between the figured molars of Bothriomys and molars of Recent cricetines with which compari- sons have been made are certainly attributable to artistic liberties. Fie. 123. Eitncomy* chinchiUoidcx, right upper and left lower molars, a, un- worn uppers, oblique view; 6, same, nearly full crown view; e, lowers of same; c, f, Juvenal; d, g, adult. 497 498 FIELDIANA: ZOOLOGY, VOLUME 46 Resemblances between Bothriomys and Andinomys noted by Osgood (1947, p. 173) are of a superficial order. The genus Chelemyscus was erected by Thomas to contain Reith- rodon fossor Thomas. The type species is known from only a single specimen represented by a skin indistinguishable from that of Noti- omys macronyx and a skull like that of a Euneomys chinchilloides. The possibility that the type might be a composite of an improperly associated skin and skull was suggested by Thomas in the following footnote to the original description (1899, p. 281). "The skull should be taken as the type if it were hereafter shown not to belong to the skin; but it was extracted in the Museum on arrival, so that any mistake seems quite impossible." Osgood (1943, p. 164), however, took exception. "Such mistakes," he declared, "are quite possible, as has been shown in other cases and, in view of the failure of collectors to obtain the species again during the past thirty years of activity, it seems more than probable that the type of fossor is composite, the skin being Notiomys and the skull Euneomys. The type is an old specimen 'presented by the La Plata Museum through Dr. F. P. Moreno' and said to have proceeded from Salta Province, Argentina, without exact locality. It is perhaps ungracious to discredit it with- out actual examination of the specimen, but the general evidence seems very much against it. The genus Chelemyscus, which is based exclusively upon it, has no characters except those that might be the result of a transposed skull; that is, Chelemyscus has no characters of its own, its external characters being strictly those of Notiomys and its cranial characters those of Euneomys. Unless additional specimens are forthcoming, therefore, this genus is suspect and de- serves no better position than in a 'hypothetical' list." Cabrera (1961, pp. 499-500), without allusion to its clouded status, recognized Chelemyscus as a valid subgenus of Euneomys. He gave no indication of having examined the type species or any bona fide specimen of the taxon. Pending further investigation, Chelemyscus is here treated as an absolute synonym of Euneomys on the basis of published descriptions of its cranial and dental characters. Nominal Species and Subspecies Euneomys chinchilloides chinchilloides Waterhouse Reithrodon chinchilloides Waterhouse, 1839, Zool. Voy. "Beagle," p. 72, pi. 27 (animal), pi. 34, fig. 20 (skull, molars). Gay, 1847, Hist. Chile, Zool., 1: 120. Burmeister, 1879, Descr. Phys. Rep. Argentina, 3: 231. Thomas, HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 499 1891, in Milne Edwards, Miss. Scient. Cap, vol. 6, Mamm., p. A 29, pi. 3, fig. 2 (animal) — CHILE: Magallanes (Orange Bay, Cape Horn). R[eith rodon] chinchilloides, Thomas, 1881, Proc. Zool. Soc. London, 1880: 696 —comparison; measurements. Reithrodon (Euneomys) chinchilloides, Coues, 1877, Monogr. N. A. Rodents, 1: 119 — characters. E[uneomys] chinchilloides, Thomas, 1901, Ann. Mag. Nat. Hist., (7), 8: 254 • — classification. Euneomys chinchilloides, J. A. Allen, 1905, Princeton Univ. Exped. Patag., 3, (1): 67. Thomas, 1916, Ann. Mag. Nat. Hist., (8), 17: 142— classification. Euneomys chinchilloides chinchilloides, Osgood, 1943, Field Mus. Nat. Hist., Zool. Ser., 30: 214, fig. 31 (skull) — characters; distribution. Euneomys ultimas Thomas, 1916, Ann. Mag. Nat. Hist., (8), 17: 185— type locality, St. Martin's Cove, Hermite Island, Cape Horn Islands, Tierra del Fuego, Chile. Euneomys chinchilloides ultimus, Osgood, 1943, Field Mus. Nat. Hist., Zool. Ser., 30: 216 — classification. Type. — Juvenal, British Museum (Natural History), no. 55.12. 26.111; collected by Charles Darwin. The type was originally "pre- served in spirits, and has since been mounted." Type locality. — Straits of Magellan, near eastern entrance, Tierra del Fuego, Chile. Euneomys chinchilloides petersoni J. A. Allen Euneomys petersoni J. A. Allen, 1903, Bull. Amer. Mus. Nat. Hist., 19: 192. J. A. Allen, 1905, Princeton Univ. Exped. Patagonia, 3, (1): 68, pi. 13, fig. 4 (skull), pi. 14, figs. 6-7 (molars)- ARGENTINA: Santa Cruz (upper Rfo Chico). Osgood, 1943, Field Mus. Nat. Hist., Zool. Ser., 30: 216— synonymy; distribution. Euneomys dobbenei Thomas, 1919, Ann. Mag. Nat. Hist., (9), 4: 127— ARGENTINA: Santa Cruz (type locality, Lago Viedma). Type. — Adult female, skin and skull, U. S. National Museum, no. 84198; collected February 10, 1897, by O. A. Peterson. Type locality.- Upper Rio Chico, near the Cordilleras, Santa Cruz, Argentina. Euneomys noei Mann Euneomys noei Mann, 1944, Biologica, Univ. Chile, fasc. 1: 95, fig. 1 (skull), fig. 2 (carpalia), pi. 1 (skin), pis. 2, 3 (skulls), pi. 4 (molars), pi. 5 (skele- ton), pi. 6 (embryo), pi. 7 (limb bones), pi. 8 (digestive tract). Type. — Male, skin and skull, Guillermo Mann Collection no. 179- II; collected 29 February, 1944, by Guillermo Mann. 500 FIELDIANA: ZOOLOGY, VOLUME 46 Type locality. — Valle de la Junta, Canyon of the Rio Volcan, Santiago, Chile; altitude, 2,400 meters. Remarks. — Doubtfully separable from Euneomys mordax Thomas. Euneomys mordax Thomas Euneomys mordax Thomas, 1912, Ann. Mag. Nat. Hist., (8), 10: 410. Type. — Adult female, skin and skull, British Museum (Natural History), no. 55.12.24.199; collected by Thomas Bridges. Type locality. — San Rafael, Mendoza, Argentina. Remarks. — Regarded by Cabrera (1961, p. 500) as a species of the dubious Chelemyscus. There is a greater probability that mor- dax may prove to be no more than a subspecies of Euneomys chin- chilloides. Euneomys catenatus Ameghino Bothriomys catenatus Ameghino, 1889, Act. Acad. Nac. Cienc. Rep. Argentina, 6: 118, atlas, pi. 4, figs. 13-13e (mandible), fig. 13d (lower molars). Type. — Left mandible with complete dentition. Type locality. — Vicinity of Cordoba, Cordoba, Argentina, "Piso bonaerense de la formacion pampeana (plioceno superior)," now re- garded as Pleistocene. PEuneomys fossor Thomas Reithrodon fossor Thomas, 1899, Ann. Mag. Nat. Hist., (7), 4: 280. [Euneomys} fossor, Thomas, 1916, Ann. Mag. Nat. Hist., (8), 17: 143— classification. [Chelemyscus} fossor, Thomas, 1925, Ann. Mag. Nat. Hist., (9), 15: 585— type of genus. "Reithrodon'' fossor, Osgood, 1943, Field Mus. Nat. Hist., Zool. Ser., 30: 164 — type skin a Notiomys macronyx, type skull a Euneomys. Euneomys fossor, Cabrera, 1961, Rev. Mus. argentine Cienc. Nat. "Bernar- dino Rivadavia," 4: 500. Type. — Skin and skull, British Museum (Natural History) no. 99.2.22.25; presented by the Museo de La Plata, Argentina, through Dr. F. P. Moreno; "the skull should be taken as type if it were here- after shown not to belong to the skin" (Thomas, 1899, footnote, p. 281). Type locality. — "Salta Province," Argentina. Remarks. — Present allocation of "Reithrodon" fossor to Euneomys is based on cranial and dental characters only, as suggested by HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 501 Thomas in the original description. Regarding external characters, Thomas (op. cit., p. 282) states that the "resemblance to A[kodon, i.e., Notiomys] macronyx, inhabiting the same districts, amounts practically to identity, there being absolutely no single character, of size, proportions, or colour, which would make the keenest-eyed 'splitter' suppose that the skin of R. fossor did not belong to Akodon [= Notiomys]." Contrary to the belief expressed by Thomas, the supposed type locality of "Retihrodon" fossor, in Salta Province, Argentina, lies at least 500 miles north of the nearest locality record for living repre- sentatives of either Notiomys or Euneomys. Z fi % e rt c 5 S ^ u s I K £ W N^ £ £ CO _ TJ C S T J§ § "o ob TjT co m - r- •* r* ^3] •i O CO OS O C ~" £ in* CD in* CD C£ 5 cd in in CD £ C m 3 PH TJI £ .H °°' c ° £ i C^3 00? «-• « « C 03 on oa> £j 2 "§ - ^.a o o" CD m m in c-^ co g "O N 06 *H in o c > oo t> oo' o — • jS .-H (M »-i CO C- ] T-t 1-H 1— < CO _ >> cc 0 "S o 21— ' 1) ^ o 3 - • — o " ^ w. T3 O O> -u ?o ^H rj ^0 be 2P I-H 0 g _r § oo ~"-" CO t-TcO O m oTo ,_T _bd os ^j« in m" i- , ro <= o co ^ Q j/2 cococococooocococo c g 0 «* i *rl ^ -* OS C ffi H Sn 3 -V ] ^H ^ ^7 1 " "o o rH «i. oo C8 o c jH »^H ^_^ -g ^ o *- 1 | 0) Win co ^ in 01 CO CO CO CO O jooSc^cc £ 02 ^ os •«3 3 -^ ^i S " 3 5g •5 co" H ^ oo ->^ «. s •£ ° c T3 £ o S on S irt ^ ^ § CD S • ^ g JB S5 fc g O) H •O •2 P C ^ o OO 1 S ^S o 0 C cq cs m i-J C oj X (3 'b ^ I 0 ^ « OS .S 11 | w 1 "b^ c '•boo-b 1 c§2 1 l§ Isl- lift 2 . 32 95 OB D C *5 •§•§ S c £ M » 'S 'S .w .-, « = 2 ^ £ ?H^ S H S 8 co S^ fe U rt W4 — *""* C 2 3 C Ig'il X r\ -i i (H w o o o ctf -§.«Svg §ws «eiT«" .-••••». ^ ssji£-stfe.| £ £ •w-rj-O^Qg^-^^S ft) <-] v> v> a *»~, ft S 55. 55, S tn O 502 Literature Cited ALLEN, J. A. 1891. Notes on new or little known North American mammals, based on recent additions to the collections of mammals in the American Museum of Natural History. Bull. Amer. Mus. Nat. Hist., 3: 263-310. 1905. Mammalia of southern Patagonia. Reports of the Princeton University Expeditions to Patagonia, 1896-1899, 3, Zoology, pt. 1: 210 pp., 29 pis. AMEGHINO, F. 1889. Contribucion al conocimiento de los mamfferos fosiles de la Republica Argentina. Act. Acad. Nac. Cienc. Rep. Argentina, Cordoba, 6: xxxii + 1027 pp. and atlas. ANDERSON, SYDNEY 1960. The baculum in microtine rodents. Univ. Kansas Publ., Mus. Nat. Hist., 12, (3): 181-216, 49 figs. AZARA, F. 1801. Essais sur 1'histoire naturelle des quadrupedes de la province du Para- guay. 2: 499 pp. Paris. 1802. Apuntamientos para la historia natural de los quadrupedos del Paraguay y Rfo de la Plata. 2: i-iii+328 pp. Madrid. BADER, ROBERT S. 1959. Dental patterns in Peromyscus of Florida. Jour. Mammal., 40: 600-602, 1 table. BAIRD, SPENCER F. 1859. Mammals of North America, pp. i xxxiv+764, 87 pis. Philadelphia. BARRERA, J. M., DE LA 1936. Peste rural. Rev. Inst. Bacteriologico, Buenos Aires, 7: 439 506. 1939. Contribucion al conocimiento de la peste selviitica en la Argentina caracteres del brote de Mendoza en 1937. Rev. Inst. Bacteriologico, Buenos Aires, 8:423-454. 1940. Estudios sobre peste selvatica en Mendoza. Rev. Inst. Bacteriologico, Buenos Aires, 9: 565 586. BIGGERS, J. D., M. R. ASHOUB, ANNE MCLAREN, and DONALD MICHIK 1958. The growth and development of mice in three climatic environments. Jour. Experiment. Biol., 35: 144-155. BIRABEN, M., and M. I. BIRABEN 1937. Viaje alrededor de Santa Cruz. Rev. Mus. La Plata, (n.s.), sec. ofirial, 1936, pp. 93-164, 31 figs. BLAIR, W. FRANK 1948. Population density, life span, and mortality rates of small mammals in the blue-grass meadow and blue-grass associations of southern Michigan. Amer. Midi. Nat., 40: 395-419. 503 504 FIELDIANA: ZOOLOGY, VOLUME 46 BROWN, W. L., JR., and E. O. WILSON 1956. Character displacement. Syst. Zool., 5: 49-64, 6 figs. BURMEISTER, H. 1854. Systematische Uebersicht der Thiere Brasiliens. Saugethiere, Berlin, pp. i-x+342. BURT, WILLIAM HENRY 1960. Bacula of North American mammals. Univ. Michigan Mus. Zool., Misc. Publ. no. 113: 76 pp., 15 pis. BUTTERWORTH, BERNARD B. 1960. The cricetid mouse, Calomys, from Venezuela. Jour. Mammal., 41: 517-518. CABRERA, A. L. 1953. Esquema fitogeografico de la republica Argentina. Rev. Mus. La Plata, (n.s.), 8, Botanica, pp. 87-168, 12 figs. 1961. Catalogo de los mamiferos de America del Sur. Rev. Mus. argentine Cienc. Nat. "Bernardino Rivadavia," 4, (2): 309-732. CALHOUN, J. B. 1949. The study of wild animals under controlled conditions. Ann. New York Acad. Sci., 51: 113-122. CLARK, FRANK H. 1938. Inheritance of pectoral buff spotting in the cactus mouse, Peromyscus eremicus. Jour. Heredity, 29: 79-80. CONSTANTINE, DENNY G. 1957. Color variation and molt in Tadarida brasiliensis and Myotis velatus. Jour. Mammal., 38: 461-466. 1958. Bleaching of hair pigment in bats by the atmosphere in caves. Jour. Mammal., 39: 513-520. COSTA LIMA, A. DA, and C. R. HATHAWAY 1946. Pulgas. Bibliografia, catalogo e hospedadores. Mem. Inst. Oswaldo Cruz, 4: 522 pp. COUES, E. 1874. Synopsis of the Muridae of North America. Proc. Phila. Acad. Nat. Sci., pp. 173-196. CRESPO, JORGE A. 1944. Relaciones entre estados climaticos y la evolution de algunos roedores de campo. Rev. Argentina Zoogeogr., 4: 137-144, 2 pis. DERBY, ORVILLE A. 1879. On the plague of rats in Brazil. Nature, 20: 65. DICE, LEE R. 1939. Variation in the cactus mouse, Peromyscus eremicus. Contr. Lab. Vert. Gen. Univ. Michigan, no. 8: 27 pp. DIDIER, ROBERT 1959. Note sur les os peniens des rongeurs collected au Perou par J. Dorst. Mammalia, 23: 172-179, 8 figs. DORST, J. 1958. Contribution a I'Stude £cologique des rongeurs des hauts plateaux du Pe>ou Meridional. Mammalia, 22: 547-565. HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 505 ELLERMAN, JOHN R. 1941. The families and genera of living rodents. British Museum (Natural History), i-xii +690 pp., 50 figs. FERRIS, F. F. 1919-1935. Contributions to a monograph of the sucking lice. Stanford Univ. Publ., Univ. Ser. Biol. Sci., 2: 634 pp. (appeared in 8 parts). FERRIS, G. F., and C. J. STOJANOVICH 1951. The sucking lice. Mem. Pacific Coast Ent. Soc., 1 : 320 +ix pp., 124 figs. FITCH, HENRY S., and DENNIS G. RAINEY 1956. Ecological observations of the woodrat, Neotoma floridana. Univ. Kansas Mus. Nat. Hist., 8: 499-533, 3 figs. GILMORE, RAYMOND M. 1947. Cyclic behavior and economic importance of the rata muca (Oryzomys) in Peru. Jour. Mammal., 28: 231-240. GIOVANNANI, M., L. G. C. VELLOZO, and G. V. L. KUBIAK 1946. Sobre as "ratadas" do primeiro planalto Paraense. Arq. Biol. Tec., Curitiba, Parana, 1: 185-195, pis. 52-60. GYLDENSTOLPE, NILS 1932. A manual of Neotropical sigmodont rodents. Kungl. Svensk. Vetenskaps- akad. Handl., (3), 11: 164 pp., 18 pis. HARPER, JOHN L., J. N. CLATWORTHY, I. H. MCNAUGHTON, and G. R. SAGAR 1961. The evolution and ecology of closely related species living in the same area. Evolution, 15: 209 227.' HARRIS, NORMAN O., ROY B. MEFFERD, JR., and SALVATORE R. RESTIVO 1960. Dental changes induced in rats by chronic exposure to adverse environ- ment. American Jour. Physiol., Washington, 198: 476-480, 2 figs., 1 table. HARRISON, G. AINS\VORTH 1958. The adaptability of mice to high environmental temperatures. Jour. Experiment. Biol., 35: 892-901. HATT, ROBERT T. 1932. The vertebral column of ricochetal rodents. Bull. Amer. Mus. Nat. Hist., 63: 599-714, 27 figs. HERSHKOVITZ, PHILIP 1944. A systematic review of the Neotropical water rats of the genus Ncctomyi* (Cricetinae). Univ. Michigan Mus. Zool., Misc. Publ. no. 58: 101 pp., 5 figs., 2 maps, 4 pis. 1955. South American marsh rats, genus Holochtlux, with a summary of sig- modont rodents. Fieldiana: Zool., 37: 639 687, figs. 139 144, pis. 17 29. 1960. Mammals of northern Colombia, preliminary report no. 8: Arboreal rice rats, a systematic revision of the subgenus Occttmys, genus Oryzomy*. Proc. U. S. Nat. Mus., 110: 513-568, 6 figs., 12 pis. HINTON, M. A. C. 1926. Monograph of the voles and lemmings (Microtinae), British Museum (Natural History). 488 pp., 110 figs., 15 pis. HOFFMEISTER, DONALD F. 1951. A taxonomic and evolutionary study of the pinon mouse, /Vrom//xrM.< truei. Illinois Biol. Monogr. 21, no. 4: 104 pp., 24 figs., 5 pis. 506 FIELDIANA: ZOOLOGY, VOLUME 46 HOOPER, EMMET T. 1952. A systematic review of the harvest mice (genus Reithrodontomys) of Latin America. Univ. Michigan Mus. Zool., Misc. Publ. no. 77: 255 pp., 24 figs., 9 pis. 1957. Dental pattern in mice of the genus Peromyscus. Univ. Michigan Mus. Zool., Misc. Publ. no. 99: 55 pp., 24 figs., 4 tables. 1959. The glans penis in five genera of cricetid rodents. Univ. Michigan Mus. Zool., Occ. Pap. no. 613: 11 pp., 5 pis. 1960. The glans penis in Neotoma (Rodentia) and allied genera. Univ. Michi- gan Mus. Zool., Occ. Pap. no. 618: 21 pp., 11 pis. HOPKINS, G. H. E. 1949. The host-associates of the lice of mammals. Proc. Zool. Soc. London, 119, (2): 387-604. HOPKINS, G. H. E., and M. ROTHSCHILD 1953-1956. An illustrated catalogue of the Rothschild collection of fleas (Sipho- naptera) in the British Museum (Natural History), with keys and short de- scriptions for the identification of families, genera, species and subspecies of the orders. 1, Tungidae and Pulicidae: xv+361 pp., 465 figs., 45 pis., 1 map, 1953. 2, Coptopsyllidae, Vermypsyllidae, Stephanocircidae, Ischnopsyllidae, Hypsophthalmidae and Xiphiopsyllidae: xi+445 pp., 707 figs., 32 pis., 1 map, 1956. HUMMELINCK, P. WAGENAAR 1940. Studies on the fauna of Curacao, Aruba, Bonaire and the Venezuelan Islands. 1: 1-130, 16 pis. The Hague. HUSSON, A. M. 1960. Studies on the fauna of Curacao and other Caribbean Islands, 10, (43): 33-40, fig. 7, pis. 6-7. JOHNSON, PHYLLIS T. 1957. A classification of the Siphonaptera of South America, with descriptions of new species. Mem. Ent. Soc. Washington, no. 5: 798 pp., 114 plates in text. JORDAN, DAVID STARR 1888. A manual of the vertebrate animals of the northern United States. Ed. 5, 375 pp. KUHLHORN, FRIEDRICH 1952. Beobachtungen liber die Biologie von Hesperomys musculinus Thomas. Zeitschr. Sauget., 18: 116-124. 1954. Saugetierkundliche Studien aus Siid-Mattogrosso. 2. Teil: Edentata; Rodentia. Saugetierk. Mitt., 2: 66-72. LANDRY, S. 0., JR. 1957. Factors affecting the procumbency of rodent upper incisors. Jour. Mam- mal., 38: 223-234, 5 figs. LAURIE, E. M. O. 1946. The reproduction of the house mouse (Mus musculus) living in different environments. Proc. Roy. Soc. London, (B), 133: 248-281. MANN, F. G. 1945. Mamiferos de Tarapaca. Biologica, fasc. 2, pp. 23-98, 44 figs., 37 pis. MARTIN, EDWIN P. 1956. A population study of the prairie vole (Microtus ochrogaster) in north- eastern Kansas. Univ. Kansas Mus. Nat. Hist., 8: 361-416, 19 figs. HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 507 MERRIAM, C. HART 1894. A new subfamily of murine rodents — the Neotominae — with description of a new genus and species and a synopsis of the known forms. Proc. Phila- delphia Acad. Nat. Sci., pp. 225-249, 5 figs. ODUM, E. P. 1955. An eleven year history of a Sigmodon population. Jour. Mammal., 36: 368-378, 4 figs., 1 pi. OGLE, CORDELIA 1934a. Adaptation of sexual activity to environmental stimulation. American Jour. Physiol., 107: 628-634. 1934b. Climatic influence on the growth of the male albino mouse. American Jour. Physiol., 107: 635-640. OSGOOD, W. H. 1916. Mammals of the Collins-Day South American Expedition. Field Mus. Nat. Hist., Zool. Ser., 10: 199-216, pis. 6, 7. 1943. The mammals of Chile. Field Mus. Nat. Hist., Zool. Ser., 30: 268 pp., 33 figs., 10 maps. 1944. Nine new South American rodents. Field Mus. Nat. Hist., Zool. Ser. » 29: 191-204. 1947. Cricetine rodents allied to Phyllotis. Jour. Mammal., 28: 165-174. PACKARD, ROBERT L. 1960. Speciation and evolution of the pygmy mice, genus Baiomys. Univ. Kansas Publ., Mus. Nat. Hist., 9: 579-670, 4 pis., 12 figs. PEARSON, 0. P. 1951. Mammals in the highlands of southern Peru. Bull. Mus. Comp. Zool., 106: 117-174, 8 pis. 1958. A taxonomic revision of the rodent genus Phyllotis. Univ. California Publ. Zool., 56, (4): 391-496, pis. 6-13, 21 figs. PEREIRA, C. 1941. Sobre as "ratadas" no sul do Brasil e o ciclo vegetativo das Taquaras. Arq. Inst. Biol. Sao Paulo, 12: 175-196, 4 pis. PHILIPPI, F. 1879. A plague of rats. Nature, 20: 530. PHILIPPI, R. A. 1900. Figuras i descripciones de los murideos de Chile. An. Mus. Nac. Chile, Zool. Entr., 14a, 70 pp., 25 pis. PHILIPPI, R. A., and L. LANDBECK 1858. Beschreibung einiger neuen chilenischen mause. Arch. Naturg., 24, 1 : 77-82. RABOR, D. S. 1955. Notes on mammals and birds of the central northern Luzon highlands, Philippines. Silliman Jour., 2: 193-218. ROONWAL, M. L. 1949. Systematics, ecology and bionomics of mammals studied in connection with tsutsugamushi disease (scrub typhus) in the Assam-Burma War Theatre. Trans. Nat. Inst. Sci. India, 3: 67-122, 3 figs., 6 pis., 15 tables. SCHAUB, S. 1925. Die hamsterartigen Nagetiere des Tertiars und ihre lebenden Verwand- ten. Abh. Schweiz. Pal. Ges., 45: 114 pp., 15 figs., 5 pis. 508 FIELDIANA: ZOOLOGY, VOLUME 46 SCHAUB, S., and H. ZAPFE 1953. Die Fauna der miozanen Spaltenfiillung von Neudorf an der March (CSR). Sitzb. Akad. Wiss., Wien, 162: 181-215, 5 figs., 3 pis. SCHEFFER, V. B. 1955. Body size with relation to population density in mammals. Jour. Mam- mal., 36: 493-515. SCHNEIDER, CARLOS O. 1946. Catalogo de los mamfferos de la provincia de Concepcion. Bol. Soc. Biol. Concepcion [Chile], 21: 67-83. SEEVERS, C. H. 1955. A revision of the tribe Amblyopinini: staphylinid beetles parasitic on mammals. Fieldiana: Zool., 37: 211-264, figs. 28-44. SIMPSON, G. G. 1936. Studies of the earliest mammalian dentitions. Dental Cosmos, Aug., Sept., pp. 1-24 (separate), 10 figs. SNYDER, D. P. 1956. Survival rates, longevity and population fluctuations in the white-footed mouse, Peromyscus leucopus, in southeastern Michigan. Univ. Michigan Mus. Zool., Misc. Publ. no. 95: 33 pp. STEHLIN, H. G., and S. SCHAUB 1951. Die Trigonodontie der simplicidentaten Nager. Schweiz. Pal. Abh., Basel, 67: 385 pp., 620 figs. STIRTON, R. A. 1935. A review of the Tertiary beavers. Bull. Geol. Sci., Univ. California Publ., 23: 391-458, 142 figs., 1 map, 2 charts. TATE, G. H. H. 1932a. The South American Cricetidae described by Felix Azara. Amer. Mus. Nov., no. 557: 1-5. 1932b. The taxonomic history of the South American cricetid genera Euneomys (subgenera Euneomys and Galenomys), Auliscomys, Chelemyscus, Chinchillula, PhyUotis, Paralomys, Graomys, Eligmodontia and Hesperomys. Amer. Mus. Nov., no. 541: 1-21. THOMAS, OLDFIELD 1884. On a collection of Muridae from central Peru. Proc. Zool. Soc. London, 1884: 447-458, pis. 42-44. 1888. On a new and interesting annectant genus of the Muridae with remarks on the relation of the Old- and New- World members of the family. Proc. Zool. Soc. London, 1888: 130-135, pi. 5. 1893. On some mammals from central Peru. Proc. Zool. Soc. London, 1893: 331-341, pis. 28-29. 1896. On the genera of rodents: an attempt to bring up to date the current arrangement of the Order. Proc. Zool. Soc. London, 1896: 1012-1028. 1897. On some small mammals from Salta, N. Argentina. Ann. Mag. Nat. Hist., (6), 20: 214-218. 1898a. On some mammals obtained by the late Mr. Henry Durnford in Chubut, E. Patagonia. Proc. Zool. Soc. London, 1898: 210-212. 1898b. On the small mammals collected by Dr. Borelli in Bolivia and northern Argentina. Boll. Mus. Zool. Anat. Comp., Torino, 13, no. 315: 1-4. 1899. Descriptions of new neotropical mammals. Ann. Mag. Nat. Hist., (7), 4: 278-288. HERSHKOVITZ: NEOTROPICAL CRICETINE RODENTS 509 1900. New Peruvian species of Conepatus, Phyllotis and Akodon. Ann. Mag. Nat. Hist., (7), 6: 466-469. 1902. On mammals collected by Perry 0. Simons in the southern part of the Bolivian plateau. Ann. Mag. Nat. Hist., (7), 9: 222-230. 1912. New bats and rodents from South America. Ann. Mag. Nat. Hist., (8), 10: 403-411. 1913. On small mammals collected in Jujuy by Senor E. Budin. Ann. Mag. Nat. Hist., (8), 11: 136-143. 1916a. On the grouping of the South American Muridae that have been referred to Phyllotis, Euneomys and Eligmodontia. Ann. Mag. Nat. Hist., (8), 17: 139-143. 1916b. Notes on Argentine, Patagonian and Cape Horn Muridae. Ann. Mag. Nat. Hist., (8), 17: 182-187. 1918. A new species of Eligmodonlia from Catamarca. Ann. Mag. Nat. Hist., (9), 2: 482-484. 1919a. On small mammals from Catamarca. Ann. Mag. Nat. Hist., (9), 3: 115-118. 1919b. On small mammals collected by Sr. E. Budin in northwestern Patagonia. Ann. Mag. Nat. Hist., (9), 3: 199-212. 1919c. On small mammals from "Otro Cerro," northeastern Rioja, collected by Sr. E. Budin. Ann. Mag. Nat. Hist., (9), 3: 489-500. 1919d. List of mammals from the highlands of Jujuy, north Argentina, collected by Sr. E. Budin. Ann. Mag. Nat. Hist., (9), 4: 128-135. 1926a. The Godman-Thomas expedition to Peru. I. On mammals collected by Mr. R. W. Hendee near Lake Junfn. Ann. Mag. Nat. Hist., (9), 17: 313-318. 1926b. The Spedan-Lewis South American Exploration — III. On mammals collected by Sr. Budin in the province of Tucuman. Ann. Mag. Nat. Hist., (9), 17: 602-609. 1927. The Octodon of the highlands near Santiago. Ann. Mag. Nat. Hist., (9), 19: 556-557. TRAUB, R. 1952. Records and descriptions of fleas from Peru. Proc. Ent. Soc. Washing- ton, 54: 1-22. URIARTE, L., and N. MORALES V. 1935. A proposito de peste en roedores agrestes de la Argentina. Rev. Inst. Bacteriologico, 7: 185-212, pis. 2-3, map. VANZOLINI, P. E., and L. R. GUIMARAES 1955. Lice and the history of South American land mammals. Rev. Brasil. Ent., 3: 13-46, 4 figs. VORONTSOV, N. N. 1957. Structure of the stomach and the relative development of parts of the gut in hamsters (Cricetinae, Rodentia, Mammalia) of the Palaearctic and the New World. Doklady, Akademia Nauk, SSSR, 117, (3): 526 529 (in Russian; English translation in "A translation of Doklady" published by the American Institute of Biological Sciences, Nov. Dec., 1957). 1959a. The systematics of hamsters (Cricetinae) in the sphere of the world fauna and their phylogenetic relations. Bull. Mosk. Ochestv. Ispyt. Prirody Otdel Biol., 64, (5): 134-137 (in Russian; English abstract in Biol. Abstr., 36: 42974, July 1, 1961). 1959b. The geographical distribution of hamsters (Cricetinae) and some prob- lems concerned with the zoogeography of the New World. Bull. Mosk. Ochestv. Ispyt. Prirody Otdel Biol., 64, (5), 137 139 (in Russian). 510 FIELDIANA: ZOOLOGY, VOLUME 46 1960a. The rate of evolution in Cricetinae and certain agents by which it is de- termined. Doklady, Akademia Nauk, SSSR, 133, (4): 980-983, 1 fig. (in Russian; English translation in "A translation of Doklady," published by the American Institute of Biological Sciences, Jan.-Feb., 1961). 1960b. The ways of food specialization and evolution of the alimentary system in Muroidea. International Symposium Theriologicum, Brno, Acad. Scient. Bohemosloveniae, pp. 155-158 (mimeograph in English). VIEIRA, C. 1953. Roedores e lagomorfos do estado de Sao Paulo. Arq. Zool. Sao Paulo, 8: 129-168. WAGNER, J. A. 1843. Die Saugthiere in Abbildungen nach der Natur mit Beschreibungen. Supplement, 3: xiv+614. WATERHOUSE, C. R. 1839. The zoology of the voyage of H.M.S. Beagle, under the command of Captain Fitzroy, R.N., during the years 1832 to 1836. Part 2. Mammalia, i-ix+i-v+97 pp., 35 pis. WHARTON, GEORGE W. 1948. Four new Peruvian chiggers (Acarina-Trombiculidae). Psyche, 55: 87- 100, 4 figs. WHITE, THEODORE E. 1959. The endocrine glands and evolution, no. 3: os cementum, hypsodonty, and diet. Contrib. Mus. Paleo. Univ. Michigan, 13: 211-265. WINGE, H. 1887. Jordfundne og nulevende Gnavere (Rodentia) fra Lagoa Santa, Minas Geraes, Brasilien. E. Mus. Lundii, 1, (3), 200 pp., 8 pis. WOOD, A. E. 1947. Rodents- — a study in evolution. Evolution, 1: 154-162, 5 figs. 1955. A revised classification of the rodents. Jour. Mamm., 36: 165-187. WOOD, A. E., and R. W. WILSON 1936. A suggested nomenclature for the cusps of the cheek teeth of rodents. Jour. Pal., 10: 388-391, 2 figs. INDEX Phyllotines are indexed by their respective subspecific, specific, generic or supergeneric names. Non-phyllotines are listed by their respective generic names. Page references to the main account of each phyllotine taxon or special subject are shown in boldface type. Page references to illustrations are italicized. abrocodon, Phyllotis, 282 taxonomy, 285 measurements, 287 Abrocoma bennelti, commensalism, 277 molars, 94 sociability, 41 Abrofhrix, 129 commensalism, 277 longipilis, 41 sociability, 41 Acomys, penis, 19 adaptive radiation, 25 Aepeomys, 84 Agastopsylla nylota, 48 pearsoni, 50, 51 Akodon, 84 aerosus, 201 amoenns, 408 arviculoides, 201 benefactus, 45 bolitiensis, 408 commensalism, 277 comparisons, 201, 203, 206, 488, 501 dorsalis, 84 habitat, 408, 425 habits, 408, 425 hibernation, 46 lutescens, 425 macron yx, 501 nigrita, 45 obscurus, 201, 206 olivaceus, 41, 277 pukherrimus, 408, 425, 488 ratadas, 43, 45 sociability, 41 varius, 201 akodonts, comparisons, 203 interrelationships, 22 molars, 84 pastoral, 21, 22 algarrobo, 449 alimentary tract, 17 allopatry (see sympatry and allopatry) alsus, Auliscomys, 392 Euneomys, 392 measurements, 400 taxonomy, 399 Amblyopinus, 50, 51 ancashi, 48 amicus, Phyllotis, 126, 438 baculum, 60 characters, 119 (key), 222 comparisons, 126, 140, 222, 236, 432, 446, 461 distribution, 218, 435 ectoparasites, 50 foot, 52 molar roots, 101 molars, 181 (planation), 232, 436, 437 skull, 224, 437 sympatry and allopatry, 36 tail, 54 taxonomy, 448 Andinomys, 473 classification, 22, 24 comparisons, 220, 323, 488, 489, 496 distribution, 209, 472 foot, 53 incisors, 103 interrelationships, 22 key characters, 120 molar patterns, 96 palate, 55, 56 andiiim, Phyllotis, 217, 219, 256, 260 baculum, 60, 63, 64, 68 comparisons, 258, 263, 283, 285, 297, 345, 439 distribution, 235, 239, 240 ectoparasites, 48 habitat, 302 key characters, 256 sympatry, 32, 36 Anotomys, incisors, 106 anterior internal fold, 71, 72 anterocone, 75 anteroconule (-id), 71, 74, 75 anterolabial conule (-id), 71, 74, 75 anterolabial style (-id), 71, 74, 75 anterolingual conule (-id), 71, 75, 76 anterolingual style (-id), 71, 74, 75 anteroloph (-id), 71, 74, 75 anteromedian style (-id), 71, 75 anticlines in molars, 84 Apodcmus, penis, 19 511 512 FIELDIANA: ZOOLOGY, VOLUME 46 Aporodon, comparisons, 126 dental characters, 19 molars, 78, 79 palate, 56 penis, 19, 20 relationships, 78 tenuirostris, molars, 79 arenarius, Phyllotis, 303, 380 measurements, 317 taxonomy, 310 argurus, Calomys, 164 distribution, 159 Hesperomys, 164 Arvicanthis, penis, 19 Arvicola amphibius, hypsodonty, 91 arvicoloides, Hesperomys, 204 Auliscomys, 217 habitat, 416 taxonomy, 223, 408, 468 baculum, 19, 58, 59, 60, 63, 64, 351 complex, 19, 58 simple, 19, 58 Baiomys, baculum, 59 classification, 85 comparisons, 24, 126, 127, 143 molars, 135 musculus, 143 palate, 56, 134 penis, 20, 59 skull, 133 taylori, 133, 134, 135, 143 bamboo, fruiting of, 42 Barreropsylla excelsa, 51 basal-incisive plane, 107, 108 basal-mandibular plane, 107, 108 bats, hibernation, 46 hypsodonty, 89 bi-level molars, 87, 181 bilophodont molars, 23, 76, 77, 86, 98 bimaculatus, Calomys, 123 comparisons, 462 Eligmodontia, 150 habitat, 147 Hesperomys, 123, 150 measurements, 190 Mm, 123, 129, 150, 154 sociability, 41 taxonomy, 155 bianco debaxo, characters, 154 Blarinomys, adaptive radiation, 26 foot, 53 boedeckeri, Phyllotis, 321 Mus, 321 taxonomy, 323 boliviae, Eligmodontia, 172 Hesperomys, 172 measurements, 192 taxonomy, 173 boliviensis, Phyllotis, 219, 404, 410 adaptive radiation, 26 Auliscomys, 417 boliviensis, 416 characters, 121 (key), 221, 222 comparisons, 221, 222, 236, 237, 407, 422, 423, 489 distribution, 218, 411 ectoparasites, 49 Hesperomys, 417 incisors, 104, 106 mandible, 226 molar roots, 101 molars, 414 skull, 225, 413 sociability, 41 supraorbital region, 58 tail, 54 Bolomys, adaptive radiation, 26 bonariensis, Calomys, 152 Hesperomys, 152 measurements, 190 taxonomy, 155 Bothriomys, 493 taxonomy, 480, 497 brachyodont molars, 56 brachyurus, Zygodontomys, 206 distribution, 197 Hesperomys, 206 measurements, 207 taxonomy, 202 bravardi, Hesperomys, 452 breeding precocity, 27 brevicauda, Zygodontomys, 205 characters, 203 distribution, 197 ectoparasites, 48 measurements, 207 molars, 1 70, 231 palate, 231 ratadas, 45 skull, 136, 200 brunneus, Zygodontomys, 204 distribution, 197 measurements, 207 bunobrachyodonty, 480 bunodont molars, 56 cachalote, 449 cachinus, Phyllotis, 448, 452 Eligmodontia, 217, 452 Graomys, 452 measurements, 457 sympatry, 461 taxonomy, 448, 453 caenogenesis in molars, 101 calden, 449 callidus, Hesperomys, 172 measurements, 192 taxonomy, 173 Callomys, 123, 129, 130 callosa, Eligmodontia, 171 callosus, Calomys, 124, 141, 165, 171 callosus, 171 characters, 27 (Juvenal), 128, 137 (key) skull, 168 INDEX 513 commensalism, 15 comparisons, 147, 199, 201, 211, 445 dispersal, 14 distribution, 125, 167 ectoparasites, 48 Hesperomys, 123, 171, 172 mammae, 128 measurements, 192 molars, 169, 170, 214 Mus, 171 Oryzomys, 171 palate, 169 skull, 131, 132, 136, 168, 212 sociability, 41 supraorbital region, 58 sympatry, 36 tail, 54, 128 Calomys, 123 ancestral characters, 24 classification, 22, 24 comparisons, 182, 211, 220, 431 dental characters, 117 distribution, 14 (of Section), 122, 125 dubius, 153, 154 ectoparasites, 47 (of Section) foot, 52, 53 incisors, 106 interrelationships, 22 key characters, 118, 119 molar roots, 101 molars, 117 palate, 56 penis, 20 ratadas, 42 Section, 24, 118, 123, 211 sympatry, 36 Calomyscus, stomach, 18, 21 campestris, Mus, 321 Cansumys, hypsodonty, 88, 91 capita, Mus, 303 measurements, 317 taxonomy, 310 caprinus, Phyllotis, 330 baculum, 61, 63 comparisons, 237, 290, 341, 343 distribution, 235, 245, 253 molars, 309, 334 reproduction, 279 skull, 332, 333 sympatry, 32, 34, 319 variation, 270 carilla, Eligmodontia, 164 Hesperomys, 123, 164 carillus, Calomys, 164 color phase, 161 distribution, 159 measurements, 190 catenatus, Euneomys, 493, 500 Bothriomys, 500 distribution, 492 taxonomy, 496 Cavia, sociability, 41, 277 caviomorphs, molars, 89, 93 cavy, incisors, 104 Celaenomys silaceus, incisors, 104 cenlralis, Eligmodontia, 452 Graomys, 452 measurements, 456 taxonomy, 454 chacoensis, Eligmodontia, 452 Graomys, 452 measurements, 456 Phyllotis, 452 taxonomy, 448, 453 Chalcomys, 18 chanar, 449 character displacement, 34, 35, 353 Chelemyscus, 493, 498 cherriei, Zygodontomys, 203 distribution, 197 measurements, 207 Oryzomys, 203 chilemis, Phyllotis, 270, 304 distribution, 307 sympatry, 276 taxonomy, 311 chinchilloides, Euneomys, 498 chinchilloides, 498 distribution, 492 ectoparasites, 51 incisors, 106 mandible, 229 molar patterns, 96 molars, 233, 398, 497 Reithrodon, 493, 498 skull, 227, 228, 495 supraorbital region, 58 Chinchillula, 485 classification, 24 comparisons, 220, 221, 323, 477 distribution, 209 foot, 53 interrelationships, 22 key characters, 120 molars, 477, 480 palate, 56 Chrotomys whiteheadi, incisors, 104 cingulum, 74 Clat-iglis, incisors, 106 Cleopsylla townsendi, 49, 50, 51 coalescence of dental elements, 97, 98 cochineal, bone stain, 285 color and humidity, 222, 258, 270, 271, 284, 306, 347, 357, 425 complex, molars, 85 compression of dental elements, 97, 98 confluence of enamel folds, 98 conules of molars, 69, 71, 74 conulids of molars, 71, 75 cordobenxis, Hesperomys, 152 cordovensis, Calomys, 152 Hesperomys, 152 measurements, 190 taxonomy, 156 coruro, incisors, 104 cortensis, Calomys, 152 514 FIELDIANA: ZOOLOGY, VOLUME 46 Hesperomys, 152 measurements, 190 taxonomy, 155 Craneopsylla minerva, 48, 50 wolffhuegeli, 48, 50 cranial measurements, 109 Crateromys schadenbergi, molars, 9b crested molars, 87, 181 crests, of molars, 76 Cricetinae, classification, 19, 129 cricetines, adaptive radiation, 25 ancestral characters, 16, 21 comparisons, 17 evolution, 16 food, 86 genera, 21 habitats, 21 interrelationships, 22 Old World, 21, 23 pastoral, 16, 21, 22, 23, 25, 26 pentalophodont, radiation of, 22 South American, origin of, 21 sylvan, 16, 21, 22, 23, 26 Cricetodon, molars, 84 Cricetulus, 26 penis, 19 Cricetus, color pattern, 488 comparisons, 488 molars, 130 penis, 19 crop variation, 36, 266, 271, 296, 302, 308, 309 Cryptomys, incisors, 104 Ctenomys, braziliensis, 416 habitat, 415, 416 niche variation, 39 sociability, 41 cusps, nomenclature, 70, 71 cylindrification of molars, 97, 98 cylindriform molars, 76, 77, 86, 97 cylindrodont molars, 23 dabbenei, Euneomys, 499 Dactylomys dactylinus, molars, 96 Dactylomyinae, adaptations, 25 darwini complex (Phyllotis), 13, 219, 234 characters, 120 (key), 222 comparisons, 345, 407 distribution, 218, 285 species, 256 sympatry and allopatry, 31 darwini, Phyllotis, 129, 217, 257, 269, 305 baculum, 59, 60, 63, 349 characters, 120 (key), 256 (key), 3^9 (diagnostic) commensalism, 15, 39 comparisons, 263, 290, 340, 344, 398, 407, 463 darmni, 320 distribution, 14, 235, 2^9 ectoparasites, 49 foot, 52 glans penis, 59 habitat, 481 Hesperomys, 320 incisors, 103 mandible, 226, 229 molar roots, 101 molars, 23 1, 232, 233, 275 Mus, 217, 320 niche variation, 39 palate, 231 pocket populations, 30 ratadas, 43, 45 reproduction, 277 skull, 221*, 225, 227, 228, 230, 231, 272, 273, 271t (abnormal) sociability, 41 supraorbital region, 57 sympatry and allopatry, 33-36, 276, 277 tail, 53 taxonomy, 13 decoloratus, Phyllotis, 404 taxonomy, 410 deer, crop variation, 38 definitus, Phyllotis, 270, 296 comparisons, 235, 243, 263, 283, 290, 292 distribution, 235, 2^3 Delomys, 84 Delostichus phyllotis, 49 Deomys, hypsodonty, 89, 91 dichrous, Mus, 321, 323 Dinomys branicki, molars, 96 brannicki (see Dinomys branicki) Dipodomys merriami, molars, 97 domorum, Phyllotis, 458 comparisons, 446 distribution, MS Eligmodontia, 217, 458 Graomys, 459 habitat, 463 taxonomy, 448 Dromiciops australis, hibernation, 46 dubius, Calomys, 153, 154 ducilla, Eligmodontia, 163 Hesperomys, 123, 163 ducillus, Calomys, 157, 163 color phase, 161 distribution, 159 habits, 162, 163 Hesperomys, 163 Dysmicus barrerai, 50 budini, 50 claviger, 48 hapalus, 50 Echiothrix leucura, incisors, 102 Ectinorus disjugis, 50 ineptus, 50 onychius, 51 ectolophid, 71 ectoparasites, 47 INDEX 515 ectostylid, 71 edax, Andinomys distribution, 209, 472 ectoparasites, 50 edax, 481 incisors, 103 mandible, 229 molar roots, 101 molars, 233, 476, 478, 479 skull, 227, 475 sociability, 41 tail, 54 Rdentata, molars, 99 edithae, Phyllotis, 219, 461 comparisons, 463 distribution, 218, 465 Graomys, 448, 455, 461 key characters, 119 elegans, Calomys, 123, 184 Eligmodontia, 184 Hesperomys, 184, 257 measurements, 194 Mitf, 184 taxonomy, 185 elegantulus, Phyllotii*, 257, 258 taxonomy, 259 Eligmodon, 175 Kligmodonlia, 130, 175, 217 classification, 18, 24 comparisons, 181, 220, 466 distribution, 122 foot, 51, 52 incisors, 106 interrelationships, 22 key characters, 118 molars, 181 nest, 148 palate, 56 tail, 54 Elimodon, 175 Kilobit, incisors, 104, 105 Elygmodovtia, 175 enamel folds, nomenclature, 69, 71 evolution, 100 enemies of phyllotines, 46 enteroloph, 71 enterostyle, 71 entoconid, 71, 80 entolophulid, 71, 80, 81 ppsilon molar pattern, 94 Erioryzomys, 84 Esperomys, 157 estivation, 426 Eumys, 18, 84 Euneomys, 217, 493 distribution, 492 incisors, 106 key characters, 119 material, 15 molars, 96, 98, 497 taxonomy, 408, 468, 480 Etisrhonygastia phijllnli, 49, 51 expulsus, Calomys, 123, 174 distribution, 167 ectoparasites, 48 Hesperomys, 123, 174 measurements, 193 Mas, 174 feciindus, Hesperomys, 172 measurements, 192 taxonomy, 173 flaridior, Phylloiis, 419 Auliscomys, 419 distribution, 411 Euneomys, 419 habitat, 425 folds, enamel, 69, 71 anterior labial, 71, 72 anterior lingual, 71, 72 anterior median, 71, 72 anterior primary, 71, 72 anterior secondary, 71, 72 internal, 71, 72 major, 69, 71, 73 minor, 69, 71, 73 primary, 69, 71, 73 secondary, 71, 72, 73 supernumerary, 71, 73 fontanelle, interfrontal, 475, 476, 477 food storage, 17 fossor, Euneomys, 493, 500 Chelemyscns, 500 distribution, 492 incisors, 106 Reithrodon, 500 foxes, predatory, 46 fralerculns, Zygodontomyx, 204, 207 frida, Calomys, 137 Hesperomys, 137 measurements, 189 fruiting cycles, 42 frtixtrator, Akodon, 205 Zygodontomys, 205, 207 fruticicollis, Phyllotis, 262 taxonomy, 266 fulpfucens, PhyUotis, 325 comparisons, 322, 329 distribution, 235, 249 habitat, 276 variation, 270 ftimipex, Phyllotix, 392 Auliscomys, 392 measurements, 400 taxonomy, 401 fnxcinHfi, Zygodontomyx, 205 distribution, 197, 198 measurements, 207 /Mxrux, PhyllotiK, 257, 258 comparisons, 260 taxonomy, 259 fusion, dental, 97, 99 Galea. 41 Galenomys, 464 classification, 24 516 FIELDIANA: ZOOLOGY, VOLUME 46 comparisons, 160, 220, 496 distribution, 209 incisors, 104, 107 interrelationships, 22 key characters, 120 garleppi, Galenomys, 464, 468 adaptive radiation, 26 comparisons, 423 distribution, 465 Euneomys, 468 foot, 52, 53 mandible, 229 molar roots, 101 molars, 238, 470 Phyllotis, 464, 468 skull, 227, 228, 467, 469 sociability, 41 tail, 54 garleppii, Phyllotis, 468 gerbil, 181 Gerbillus, 181, 188 gerbillus, Phyllotis, 217, 219, 430 baculum, 60 characters, 219 (key), 221 comparisons, 181, 222, 438, 439, 461 distribution, 218, 435 foot, 52, 53 habits, 42 Hesperomys, 123 molar roots, 101 molars, 436, 437 Paralomys, 430 pocket populations, 30 skull, 436 glans penis, 19 glirinus, Hesperomys, 303 measurements, 317 Mus, 303 taxonomy, 310 gopher, incisors, 104 Gourliea, 449 gracilipes, Calpmys, 123 Eligmodontia, 151 Hesperomys, 123, 151 measurements, 190 Mus, 151 taxonomy, 155 Graomys, 217 taxonomy, 223, 399, 448 griseoflavus, Phyllotis, 217, 219, 441 adaptive radiation, 25 baculum, 60 characters, 119 (key), 221, 222 comparisons, 127, 128, 182, 201, 211, 222, 236, 237, 343, 398 commensalism, 15 dispersal, 14 distribution, 218, 443 ectoparasites, 50 Eligmodontia, 451 Graomys, 451 griseoflavus, 451 Hesperomys, 451 hibernation, 46 Juvenal characters, 26, 27 mandible, 226 molar roots, 101 molars, 170, 232, 447, 476, 478, 479 Mus, 217, 321, 451 skull, 136, 224, 225, 444, 445 sociability, 41 supraorbital region, 58 sympatry, 36 tail, 54 griseus, Zygodontomys, 204, 207 Grisonella, predator, 46 growth, allometric, 27 gymnuromyines, molar pattern, 76 Gymnuromys, molars, 80, 88, 94 roberti, 94 haggardi, Phyllotis, 217, 219, 256 comparisons, 222, 263, 265 distribution, 235, 239, 240 ectoparasites, 48 key characters, 256 skull, 272, 273 sympatry and allopatry, 32, 265, 319 hamster, 17 Haplomylomys, 85 californicus, 21 molar evolution, 480 palate, 56 stomach, 21 Hectopsylla stomis, 50 Heligmodontia, 175 Hesperomys, 217 bravardi, 452 habitat, 415 history, 123, 129, 130 renggeri, 204 taxonomy, 432 Heterocricetodon, molars, 84 Heteromys anomalus, ratada, 45 gaumeri, molars, 97 hibernation, 17, 46, 148, 425, 426, 449 hirtipes, Eligmodontia, 186 measurements, 195 Phyllotis, 186 taxonomy, 188, 195 hispidus, Sigmodon, 37, 129 crop variation, 37 ratadas, 45 stomach, 21 Hodomys, molars, 478, 479 Holochilus, 216 comparisons, 213, 496 convergence, 25 leucogaster, 45 magnus, 58 pastoral, 18, 23 ratadas, 45 supraorbital region, 58 Hoplopleura affinis, 48, 49, 50 hesperomydis, 48 nesoryzomydis, 48 INDEX 517 reducta, 49, 50 humidity and color, 222, 258, 270, 271, 284, 306, 347, 357, 425 hummelincki, Baiomys, 152 characters, 143, 146, 148, 156 measurements, 190 taxonomy, 156 hydrocephaly, 265 Hydromys, relationships, 104 hypocone(-id), 71 hypogaens, Phyllotis, 219, 462 comparisons, 461 distribution, 218, 465 Graomys, 462 key characters, 119 molars, 463 taxonomy, 448, 455 hypolophule, 71 Hypsimys, 19 hypsodonty, 88, 99, 480 coronal, 88, 90 cusp, 89 root, 92 seleno-, 90 tooth-base, 92 tubercular, 89, 90 ichthyomyines, 21 interrelationships, 22 Ichthyomys, dispersal, 18 hypsodonty, 89 incisors, 106 molars, 19 relationships, 19, 24 illapellinus, Mus, 321 incisors, 101, 107 degenerative, 102, 105 flexed, 102, 106, 476, 477 grooved, 106, 399, 407, 408 indices, 107, 108 lower, 107 opisthodont, 102, 103 orthodont, 102, 103 proodont, 102, 103 seizer-digger, 102, 103, 105 triturator, 104, 105 upper, 101 Inomyx, 84 insectivores, 14 hypsodonty, 89 internal folds, 98 interparietal, paired, 489 involution, 93, 94, 96, 100 Irenomys, 56 incisors, 106 molars, 94 isolation in molars, 97, 98 Ixodev, 50 andinus, 47, 48 jacunda, Eligmodontia, 186 measurements, 195 taxonomy, 188 juvenal characters, 27, 58 Lagomorpha, molars, 98 Lagostomus, incisors, 104 lamina, of molars chevron, 93 oblique, 93 transverse, 92 lamination, molar, 79, 92, 94, 96 lanatus, Hesperomys, 303 measurements, 317 taxonomy, 309 Langsgrat, 80 lasiurus, Zygodontomys, 206 characters, 203 distribution, 197, 198 ectoparasites, 48 Habrothrix, 206 measurements, 207 Mus, 206 taxonomy, 202 laucka, Calomys, 41, 123, 124, 142, 171 characters, 126, 128, 137 (key), H6 (external) classification, 24 commensalism, 15, 39 comparisons, 126, 127, 140, 182 dispersal, 14 distribution, 125, 145 ectoparasites, 47 Eligmodontia, 150 feet, 146 head, 146 Hesperomys, 123, 150 hibernation, 46, 154 interrelationships, 18, 22 laucha, 149 mammae, 128 measurements, 190 molars, 155 Man, 150 niche variation, 39 Oryzomyx, 148, 150 palate, 134 relationships, 18 skull, 151, 132, 133 sociability, 41 supraorbital region, 57, 58 sympatry and allopatry, 36 tail, 54, 128 Leidymys, 18 lenguarum, Zygodontomyx, 206 distribution, 197, 198 measurements, 207 Icpida, Eligmodontia, 162 lepidti*, Calomys, 124, 143, 157, 160 adaptive radiation, 26 allopatry, 26 characters, 128, 137 (key) distribution, 125, 159 ectoparasites, 48 Hexperomys, 123, 157, 162 lepidus, 162 518 FIELDIANA: ZOOLOGY, VOLUME 46 mammae, 128 measurements, 191 molars, 135, 136 skull, 131, 132, 161 supraorbital region, 57, 111 sympatry and allopatry, 36 tail, 54, 128 Lepus, arcticus, 27 leucurus, Phyllotis, 427, 428 Auliscomys, 428 distribution, 420 Euneomys, 428 limatus, Phyllotis, 288, 299, 305 baculum, 65 comparisons, 263, 283, 292 distribution, 235, 240, 243, 253, 258 hydrocephaly, 265 molars, 298 reproduction, 278 skull, 274 (abnormal), 298 sympatry and allopatry, 32 variation, 270, 308 lineicaudatus, Andinomys, 482 distribution, 472 lockwoodi, Graomys, 452 measurements, 457 taxonomy, 454 locomotor organs, pastoral modification, 17 longevity, 37 longitudinal ridge, of molars, 83 Lophiomys, incisors, 106 lophs, molar, 69, 71 lophules, molar, 69, 71 Loxodontomys, 217 taxonomy, 223, 399 lutescens, Phyllotis, 380 taxonomy, 381 magister, Phyllotis, 288 baculum, 61, 63, 65 comparisons, 283, 297, 308, 343, 344 distribution, 235, 243, 253 reproduction, 278 sympatry and allopatry, 278 variation, 270 marcarum, Calomys, 162 Hesperomys, 162 measurements, 190 marica, Eligmodontia, 186 habits, 183 measurements, 195 taxonomy, 188 maritimus, Phyllotis, 438 measurements, 440 taxonomy, 439 Marmosa elegans, commensalism, 277 hibernation, 46 pusilla, 46 sociability, 41 marmosets, hibernation, 46 marsupials, hibernation, 46 hypsodonty, 89 medius, Phyllotis, 452 Graomys, 452 measurements, 457 sympatry, 461 taxonomy, 455, 463 Megadontomys, classification, 84 convergence, 25 Megalomys, classification, 21, 84 megalotis, Mus, 321 melanius, Phyllotis, 262 taxonomy, 266 Melanomys, 84 melanonotus, Mus, 321 melanotis, Mus, 321 Meriones, 19 Merostachys fistulosa, 44 Mesocricetus, 19 mesoloph (-id), 71, 76, 78, 79, 81, 85, 86, 100 false, 80 mesoloph-mesostyle (-id), 79, 80, 86 mesostyle (-id), 71, 76, 77, 78, 79, 81,85, 86, 100 metacone (-id), 71, 76, 77 metalophule (-id), 71, 82 Microcavia australis, 41 Micromys, 19 micropus, Phyllotis, 219, 391 Auliscomys, 391 baculum, 60 characters, 119 (key), 222 comparisons, 222, 236, 407, 489, 496 dispersal, 14 distribution, 218, 393 ectoparasites, 49 Euneomys, 391 fontanelles, 477 Hesperomys, 391 incisors, 106 Loxodontomys, 392 mandible, 226 molar roots, 101 molars, 96, 232, 233, 397 Mus, 217, 391 skull, 284, 225, 396 sympatry, 36 tail, 54 Microryzomys, 84 microtines, 21, 23, 76 microtinus, Zygodontomys, 205 characters, 207 distribution, 197, 198 taxonomy, 203 Microtus ochrogaster, 37 crop variation, 37 longevity, 37 pennsylvanicus, 37 Microxus, 18 miurus, Calomys, 189 Hesperomys, 137 molars, bi-level, 87, 181 complex, 76 crested, 87 INDEX 519 cylindriform, 100 enamel pattern, 71 evolution, 82 (dynamics), 99 (differ- ential rates), 100 (reversal of) Juvenal, 27 of pastoral cricetines, 19 pentalophodont, 16 plane, 87, 108, 181 proportional growth, 284 roots, 101 simple, 85 terraced, 87, 181 third, 100 mole rat, 104 molilor, Hesperomys, 123 mollia, Mus, 321 montanus, Phyllotis, 438 Calomys, 190 Hesperomys, 165, 427 measurements, 191, 440 taxonomy, 439 Moniicaria, 104 mordax, Euneomys, 493, 500 distribution, ^92 moreni, Eligmodon, 186, 189 measurements, 195 taxonomy, 188 morenoi, Eligmodontia, 186 morgani, Eligmodontia, 185 measurements, 194 taxonomy, 185 mure, 80, 81 muriculus, Hesperomys, 172 measurements, 192 taxonomy, 173 Muridae, 19, 89 molars, enamel pattern, 71 murids, 89 mitrillHs, Calomys, 148, 152 Hesperomys, 148, 152 hibernation, 46 measurements, 190 taxonomy, 155 murinae, 19 Mus, 129 cochineal stain, 285 commensalism, 15, 41 comparisons, 126, 146 crop variation, 39, 40 dubius, 154 miisculus, 126, 127, 173 niche variation, 39 penis, 19 musculina, Eligmodontia, 152 muscitlinns, Calomys, 152 Hesperomys, 152, 171, 172 measurements, 190 taxonomy, 155 muskrat, 38 Miistelafrenata, 46 Myotis telifer, 39 Mystromys albicaudatiis, 24 comparisons, 221 incisors, 107 mandible, 226 molars, 231, 232 palate, 231 skull, 230, 231 Neacomys, 21, 84 Necromys, 123, 480 conifer, 172 Nectomys, 84, 213, 216 adaptation, 25 squamipes, 25 stomach, 18, 21 Nekonia, 20, 85 Neocometes, 80 brunonis, 9-4 molars, 80, 92, 9^ Neotoma, 129 alleni, molars of, 1>78, ^79 cinerea, molars of, 1^78, ^79 classification, 85 comparisons, 477 crop variation, 37 floridana, 37 incisors, 106 mexicana, molars of, ^78, 1*79 molars, 19, 96, ^78, 1>79, 480 palate, 56 penis, 20 relationships, 19 stomach, 21 Neotominae, 89, 480 Neotomodon, 20, 85 Neotomys, comparisons, 477 fontanelles, 477 Nectomys, comparisons, 496 ebriosus, 58 habitat, 425 relationships, 23 supraorbital region, 58 Neotyphloceras chilensis, 49 crassispina, 48, 49, 50, 51 hemisus, 50 Nesokia, 19 Nesoryzomys, 21, 85 niche variation, 38, 271, 308 Nosppsyllus londiniensis, 48 noei, Euneomys, 493, 500 distribution, 492 nogalaris, Phyllotis, 305, 390 distribution, 235, 2^5, 253 Notiomys, adaptations, 26 comparisons, 501 foot, 53 marronyx, 498, 501 taxonomy, 498, 501 Nyctomys, 21 arboreal, 25 classification, 85 supraorbital region, 58 Ochrotomys, 84 penis, 20 520 FIELDIANA: ZOOLOGY, VOLUME 46 Octodon, commensalism, 277 sociability, 41 Oecomys, 84 adaptation, 25 supraorbital region, 58 olfactory lobes, 17 Onychomys, 85 hypsodonty, 89 penis, 20 oreigenus, Phyllotis, 304 measurements, 317 taxonomy, 311 Ornithodores, 49 Orthogeomys cuniculus, 97 oryzomyines, 84 interrelationships, 22 molars, 76, 84, 88 Oryzomys, 21, 84, 130 adaptive radiation, 25, 35 alfaroi, 59 baculum, 59 bicolor, 35 caliginosus, 25, 45, 203 commensalism, 277 comparisons, 212 concolor, 25, 35 dispersal, 18 eliurus, 45 galapagoensis, 129 ica, 45 laticeps, 45 longicaudatus, 18, 41, 43, 277 melanotus, 59 molars, 19, 214 nigripes, 43, 45 obtusirostris, 203 palustris, 26, 210 (foot), 212, 213 (skull), 21 It (molars) ratadas, 43, 45 sympatry and allopatry, 35 xantheolus, 45 osgoodi, Phyllotis, 304 measurements, 317 taxonomy, 311 osilae, Phyllotis, 217, 219, 256, 305, 344 baculum, 60, 61, 63, 349 characters, 256 (key), 849 comparisons, 263, 283, 308, 341, 343, 349, 408, 415 distribution, 235, 245, 252, 253 ectoparasites, 49 osilae, 380 pocket population, 30 reproduction, 282 skull, 354, 355 sympatry and allopatry, 33-35, 276, 319 otomyines, molars, 76 Otomys orestes, 94 Otonyctomys, 21, 85 Ototylomys, 20, 58, 85 oven-bird, 449 oxymycterines, 21 Oxymycterus, 129 adaptations, 26 foot, 53 rufus, 21 stomach, 21 palatal pits, 348 palate, 22, 54, 55 palingenesis in molars, 101 pamparum, Eligmodontia, 185 measurements, 194 taxonomy, 186 paracone, 71, 76, 77, 80 Paralomys, 123, 217, 223, 399, 432 paralophule, 71, 80 pastoral cricetines, 13 stomach, 18 habits, 86 peak populations, 318 pectoral streak, 347 pelts, commercial, 489 penis, types, 19, 22 pentalophodont molars, 16, 21, 22, 55, 76, 77, 84, 85 pericotes, 43, 449 peromyscines, 84 habits, 84 molars, 76, 84 origin, 21 radiation, 23 Peromyscus, 84 adaptations, 25 characters, 17 comparisons, 431 gilberti, 81 leucopus, 129 longevity, 37 maniculatus, 37 molar, 19, 81 palate, 56 penis, 20 polionotus, 431 pseudomesoloph, 80, 81 stomach, 21 truei, 80, 81 petersoni, Euneomys, 499 distribution, 492 Phaenomys, 21, 84 phaeus, Phyllotis, 384 baculum, 61, 63 comparisons, 325, 381, 408 distribution, 235, 245, 252, 253 sympatry and allopatry, 33, 276 phyllotines, 21, 23, 116 adaptive radiation, 26 baculum, 58, 60 characters, 23, 27, 116, 118 (key) classification, 24 crop variation, 37 distribution, 14 ecology, 13 ectoparasites, 47 enemies, 46 INDEX 521 feet, 51 hibernation, 46 interrelationships, 22 material, 15 molars, 69, 84 niche variation, 38 origin, 18 palate, 54 ratadas, 42 reproduction, 28 sociability, 41 supraorbital region, 57 sympatry and allopatry, 31 tail, 53 Phyllotis, 208, 217 adaptive radiation, 26 baculum, 58 characters, key, 118 classification, 24 comparisons, 182, 468 distribution, Ik, 209, 218 ectoparasites, 48 foot, 53 incisors, 106, 107 interrelationships, 22 molars, 99, 182, 2S2 palate, 55 pocket populations, 30 ratadas, 42 reproduction, 28 Section, Ik, 24, 118, 208 sociability, 41 taxonomy, 13 (revision by Pearson), 448 pictus, Phyllotis, 404 Auliscomys, 404 baculum, 60 characters, 120 (key), 121, 222 comparisons, 222, 237, 412, 423, 496 distribution, 21 8, kOS ectoparasites, 50 Euneomys, 217, 404 foot, 52 incisors, 106 mandible, 226 molar roots, 101 molars, 232, 398 Reithrodon, '111, 404 skull, 22k, 225, 406 supraorbital region, 58 tail, 54 pixiina, Zygodontomys, 206 characters, 202 distribution, 197, 198 measurements, 207 Plagiodontia hylaeum, molars, 9k plague (see ratada) sylvatic, 449 planation, 86, 87, 99, 181 plane molars, 87, 181 platyhypsodonty, 480 platytarsiis, JMus, 321 plication of molars, 83 Plocopsylla achilles, 49 chins, 49 enderleini, 50, 51 inli, 49 pallas, 48 icolffsohni, 49, 51 pocket populations, 28 Podomys, palate, 56 Podoiomys, 18 Polygenis bohlsi, 48 byturus, 49 cisandinus, 50 occidentalis, 50 puelche, 50 rimatus, 48 thurmani, 49 populations, 318 ecological types, 28-30 postcingulum, 71, 74, 75, 76, 100 posteroconule (-id), 71, 75 posteroloph (-id), 71, 75 postero-median hypolophule (-id), 71 postero-median protolophule (-id), 71 posterostyle (-id), 71, 75 posticalis, Phyllotis, 262, 270, 282, 302, 380 baculum, 61, 63, 65 cochineal bone stain, 285 comparisons, 263, 289, 297, 311, 381 distribution, 235, 240, 2kS, 252, 253 ectoparasites, 48 measurements, 264, 383 reproduction, 278 sympatry and allopatry, 32, 33, 36, 276 primary fold, 98 procingulum, 71, 74, 75, 76, 100 Proechimys gttyannensis, 45 Prosopsis alba, 449 algarrobilla, 449 protocone (-id), 71 protolophule (-id), 71 pseudomesoloph (-id), 80, 81, 85 Pseudoryzomys, 208 adaptive radiation, 26 characters, 119 (key) comparisons, 201 distribution, 209 foot, 51, 210 interrelationships, 22 Pseudoseirus, 449 Ptyssophorus, 480 pueruliw, Eliffmodontia, 175, 183, 186 distribution, 177 Hesperomyx, 186 measurements, 195 M ii8. 186 Puler irritanx, 50 punctulalus, Zygodontomyx, 203, 204 Acodon, 204 distribution, 197, 198 measurements, 207 Punomys, 276, 468 lemminus, 466 522 FIELDIANA: ZOOLOGY, VOLUME 46 pusillus, Akodon, 151, 152 Hesperomys, 151 Mus, 151 raccoon, crop variation, 38 rat, brown (see Rattus) ratadas, 42, 277 rata laucha, 149 rata-muca, 45 rata pitoco, 45 Rat septieme, 149 Rattus, commensalism, 277 crop variation, 38 niche variation, 40 norvegicus, 38 penis, 19 ratadas, 43 rattus, 129, 277 sociability, 41 Reithrodon, 23, 493 comparisons, 469, 496 incisors, 106 physodes, 58 supraorbital region, 58 taxonomy, 480 Reithrodontomys, 85 fulvescens, molars, 79 incisors, 106 megalotis, 21 molars, 78, 79 palate, 56 penis, 20 stomach, 21 renggeri, Hesperomys, 204 reproduction, 28 reversal in molar evolution, 100, 102 Rhagomys, 106 Rhambomys, 19 Rheomys, 18, 24, 89 Rhipidomys, 21, 25, 84 palate, 55 Rhizomys pruinosus, 106 incisors, 105 Rhopaplopsyllus cacicus, 50 Rhynchomys soricoides, 102 incisors, 105 ricardulus, Phyllotis, 270, 304, 330 measurements, 317 sympatry, 276 taxonomy, 292, 311 roots, molar, 101 rupestris, Phyllotis, 32-34, 302 Abrothrix, 303 Akodon, 302 baculum, 61, 63, 65 comparisons, 283, 289, 296, 299, 322, 329, 331, 337, 339, 343, 415 distribution, 243, 245, 249, 253, 294, 330 ectoparasites, 49 habitat, 425 Mus, 302 reproduction, 278 skull, 272, 273, 354, 355 sympatry and allopatry, 32-34, 243, 276, 290, 319, 335 variation, 270 sahamae, Chinchillula, 490 adaptive radiation, 26 distribution, 209, 484 ectoparasites, 51 incisors, 106 mandible, 229 molar roots, 101 molars, 233, 488 sociability, 41 skull, 227, 228, 487 tail, 54 Salsola, 449 saltatorial rodents, 54 sanctaemartae, Zygodontomys, 204 distribution, 197 measurements, 207 Oryzomys, 204 Scapteromys, 53, 129 Schaubeumys, 84 Scolomys, 21, 84 Scotinomys, 20, 85 secondary folds, 98 segethi, Mus, 321, 323 selenodont, 88 sense organs, 17 seorsus, Zygodontomys, 204 characters, 202 distribution, 197 measurements, 207 sigmation, 79 Sigmodon, 18, 19, 23, 26, 29, 121, 496 alstoni, 106 Sigmodontes, 129 Sigmodontomys, 84 sigmodonts, 21, 23 characters, 25 hypsodonty, 89 interrelationships, 22 molars, 84 palate, 56 sigmoid molar pattern, 93 simplex, Hesperomys, 123 skull, murid, 111-113, 115 measurements, 109 types, 24 sloths, hibernation, 46 snake, predator, 46 sociability, 41 sorella, Eligmodontia, 137 sorellus, Calomys, 124, 137 baculum, 60 characters, 137 (key) classification, 24 cochineal bone stain, 285 comparisons, 147, 160 distribution, 125, 139 ectoparasites, 47 Hesperomys, 123, 137 INDEX 523 mammae, 128 measurements, 189 molars, 135 palate, 134 skull, 131, 132, 1^1 supraorbital region, 57 sympatry and allopatry, 36 tail, 54, 128 Spalacopus, 104 Sphinctopsylla inca, 48, 49, 50 squirrel, gray, 38 stellae, Zygodontomys, 205 distribution, 797 measurements, 207 slenops, Phyllotis, 262 taxonomy, 266 stomach, 18, 20 styles (-id), 69, 71, 74, 75 sublimis, Phyllotis, 419 adaptive radiation, 26 Auliscomys, 427 characters, 120 (key), 121, 221-223 comparisons, 160, 222, 237, 407, 415, 466 distribution, 218, ^20 ectoparasites, 50 Euneomys, 217, 427 habitat, 219 hibernation, 46 incisors, 104, 106 mandible, 226 molar roots, 101 molars, 232, 424, 470 predation, 408 skull, 22k, 225, ^23, 467, J^69 sociability, 41 sublimis, 427 tail, 54 supernumerary crests, 75, 76 supraorbital region, 57, 111 sylvatic plague, 449 Symidon, 129 sympatry and allopatry, 31, 147, 243, 263, 290, 319, 335, 343, 345, 353, 357 379, 381, 386, 399 syncline, molar, 84 synecetic, 31, 32, 336 Tachyoryctes, 39 Tadarida brasiliensix, 39 tail, 53 tamborum, Phyllotis, 260, 262 tapirapoanus, Zygodontomys, 207 distribution, 197 taquara, 43, 44, 45 tarapacensis, Eligmodonlia, 187 measurements, 195 taxonomy, 188 taterona, Graomys, 452 measurements, 456 taxonomy, 458 tener, Calomys, 157 Hesperomys, 123, 157 rat ;uias. 45 tenuirostris, Aporodon, molars, 79 Teonoma, molars, 1*78, 479 terraced molars, 87, 88 tetralophodont molars, 22, 23, 55, 76, *v> O E / / , OO Tetrapsyllus bleptus, 48, 49, 50 rhombus, 51 Thalpomys, 154 Thaptomys, 26, 127 Theridomya, molars, 84, 85 thistle, 449 thomasi, Zygodontomys, 205 distribution, 197 measurements, 207 thomasomyines, 22, 76, 84, 88 Thomasomys, 84 ancestral characters, 17, 20, 21 incisors, 104 interrelationships, 22 molars, 19 palate, 55, 56 ratadas, 45 Thomomys, 39 bulbivorus, 104 Tiamastus subtilis, 49 tobagi, Zygodontomys, 205 distribution, 197 measurements, 207 Tretomyx, 480 triangulation in molars, 93, 95, 100, 477 tribosphenic molar, 69, 73 trilophodont, 23, 86, 98 Tsaganomys, 99 altaicus, molars, 97 tuco-tuco, 183, 415 lucumanw, Phyllotis, 380, 388 comparisons, 311 distribution, 2S5, 21t5, 253 sympatry, 276 Tylomys, 25, 85 comparisons, 126 penis, 19, 20 supraorbital region, 58 typiix, Eligmodontia, 175, 176 adaptive radiation, 26 comparisons, 431, 462 distribution, 177 foot, 51, 52 habitat, 463 habits, 42, 182 measurements, 183 molar roots, 101 molars, 180 palate, 179 skull, 178, 179 typux, 184 raccarum, Phyllolis, 270, 303, 322 comparisons, 318 measurements, 317 molars, 309 taxonomy, 292, 310 524 FIELDIANA: ZOOLOGY, VOLUME 46 ventriosus, Zygodonlomys, 204 distribution, 197 measurements, 207 venustus, Calomys, 172, 192 Hesperomys, 123, 141, 172 measurements, 192 Oryzomys, 172 taxonomy, 173 vertical incisive plane, 107, 108 viscacha, 41, 104, 416 vole, crop variation, 38 waterhousii, Hesperomys, 417 Phyllotis, 417 wavrini, Pseudoryzomys, 126, 208, 215 comparisons, 126 foot, 210 molar roots, 101 molars, 214. Oryzomys, 208, 215 skull, 212, 213 wild cat, predator, 46 Wilfredomys, 21, 84 wolffhuegeli, Phyllotis, 304 measurements, 317 taxonomy, 311 wolffsohni, Phyllotis, 305, 330, 339 baculum, 61, 63, 67 comparisons, 211, 222, 237, 290, 311, 331, 335, 336, 381, 445, 459 distribution, 235, 245, 253 molars, 237, 309, 334, 398 reproduction, 279 skull, 332, 333 sympatry, 34 variation, 270 xanthopygus, Phyllotis, 129, 217, 305, 327 Auliscomys, 327 comparisons, 290, 311 distribution, 235, 249 ectoparasites, 49 Euneomys, 327 habitat, 276 Hesperomys, 327 Mus, 327 variation, 270 Xenomys, 20, 56, 85 Zygodontomys, 18, 196 adaptive radiation, 26 baculum, 60 characters, 27 (Juvenal), 118 (key) classification, 24 comparisons, 127, 128, 168, 211, 220, 221, 445 dispersal, 14 distribution, 122, 197 foot, 51 incisors, 106 interrelationships, 22 material, 15 molar roots, 101 molars, 170, 181, 182, 231 palate, 56 prey, 46 pseudomesoloph, 80 ratadas, 14, 45 skull, 136 supraorbital region, 58 tail, 54 Publication 962 UNIVERSITY OF ILLINOIS-URBANA