2 g m* 
 
 THE UNIVERSITY 
 
 OF ILLINOIS 
 LIBRARY 
 
 CD 
 
 UJ 
 
 U. 
 
 ^■■ 569.3 
 
 Go 
 
 qEOLOGV UBRtf* 
 
GEQlJG* library 
 
 Return this book on or before the 
 Latest Date stamped below. 
 
Digitized by the Internet Archive 
 in 2019 with funding from 
 University of Illinois Urbana-Champaign 
 
 https://archive.org/details/cenozoicgravigraOOstoc 
 


 
 
 
 
 
 
 
 
 
STOCK 
 
 FRONTISPIECE 
 
 Nothrotherium shastense Sinclair. 
 
 Restoration by Charles-R. Knight under direction of Chester Stock. 
 
CENOZOIC GRAVIURADE EDENTATES OK WESTERN NORII1 AMERICA 
 
 WITH SPECIAL REFERENCE TO 
 
 THE PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE 
 
 OF RANCHO LA BREA 
 
 BY 
 
 CHESTER STOCK 
 
 THE IIBKAKT P' tni 
 
 MAY 11 |S25 
 
 UNIVERSITY Of 8LL*NO 
 
 Published by the Carnegie Institution of Washington 
 Washington, January, 1925 
 
CARNEGIE INSTITUTION OF WASHINGTON 
 
 Publication No. 331 
 
 PRESSES OF 
 
 JUDD & DETWEILER, INC. 
 WASHINGTON, D. C. 
 
St £ c 
 
 CONTENTS. 
 
 a) 
 
 <J9 
 
 CS 
 
 V 
 
 J 
 
 A 
 
 & 
 
 £V> 
 
 to 
 
 <s 
 
 
 a 
 
 -K 
 
 V 
 
 Page 
 
 Introduction. 3 
 
 Acknowledgments. 4 
 
 Review of the literature on the gravigrade edentates. 6 
 
 Megalonychinae. 6 
 
 Mylodontidae. 10 
 
 Geologic range and geographic distribution of the 
 
 Megalonychinae. 14 
 
 Occurrence and description of remains probably 
 belonging to members of the Megalo¬ 
 nychinae in Tertiary beds of western 
 
 North America. 18 
 
 Rattlesnake formation, Eastern Oregon. 18 
 
 Pinol Tuff-Orinda-Siesta deposits, middle California. 20 
 
 San Timoteo and Eden beds, southern California. 21 
 
 Habitat of the Megalonychinae as suggested by 
 
 occurrence. 22 
 
 Geologic range and geographic distribution of the 
 
 Mylodontidae. 24 
 
 Habitat of the Mylodontidae as suggested by 
 
 occurrence. 26 
 
 Occurrence of ground-sloths at Rancho La Brea.... 30 
 
 Probable appearance, habits, and some factors 
 conditioning the life of the Rancho La 
 
 Brea ground-sloths. 33 
 
 Classification and relationship of the gravigrade 
 
 edentates of Rancho La Brea. 36 
 
 Megalonychinae of Rancho La Brea. 39 
 
 Characters of subfamily Megalonychinae. 39 
 
 Characters of genus Nothrotherium. 39 
 
 Description of material. 40 
 
 Cranium. 40 
 
 Mandible. 47 
 
 Dentition. 49 
 
 Endocranial cast. 53 
 
 Elements of hyoid arch. 54 
 
 Vertebrae. 55 
 
 Haemapophyses. 62 
 
 Costal ribs. 64 
 
 Sternal ribs. 64 
 
 Sternum. 66 
 
 Clavicle. 67 
 
 Scapula. 68 
 
 Humerus. 69 
 
 Ulna. 70 
 
 Radius. 70 
 
 Scaphoid. 71 
 
 Trapezium and metacarpal 1. 72 
 
 Metacarpal II. 73 
 
 Phalanx II, digit II, manus. 74 
 
 Phalanx III, digit II, manus. 74 
 
 Phalanx II, digit III, manus. 75 
 
 Phalanx III, digit III, manus. 76 
 
 Phalanx I, digit IV or digit III, manus.... 76 
 
 Phalanx II, digit IV, manus. 76 
 
 Phalanx III, digit IV, manus. 77 
 
 Metacarpal V. 77 
 
 Comparison of Manus. 78 
 
 Pelvis. 79 
 
 Femur. 80 
 
 Patella. 82 
 
 Tibia. 83 
 
 Fibula. 84 
 
 Calcaneum. 84 
 
 Astragalus. 86 
 
 Page 
 
 Megalonychinae of Rancho La Brea ( Continued ). 
 Description of material ( Continued ). 
 
 Cuboid. 88 
 
 Navicular. 89 
 
 Ectocuneiform. 90 
 
 Entocunciform and metatarsal I. 91 
 
 Metatarsal II. 91 
 
 Phalanx I, digit II, pes. 93 
 
 Phalanx II, digit II, pes. 93 
 
 Phalanx III, digit II, pes. 93 
 
 Metatarsal III. 95 
 
 Co-ossified phalanges I and II, digit III, pes. 95 
 
 Phalanx III, digit III, pes. 96 
 
 Metatarsal IV. 97 
 
 Metatarsal V. 99 
 
 Comparison of pcs. 99 
 
 Characters of genus Megalonyx. 101 
 
 Description of material. 101 
 
 Mandible. JO! 
 
 Dentition. 10! 
 
 Vertebrae. 103 
 
 Humerus. 104 
 
 Ulna. 105 
 
 Unciform. 105 
 
 Patella. 106 
 
 Calcaneum. 106 
 
 Metatarsal Ill and digit III. 107 
 
 Metatarsal IV. 108 
 
 Metatarsal V. 109 
 
 Occurrence and description of Megalonychinae in the 
 
 Pleistocene of western North America. 110 
 
 Great Basin Province. 110 
 
 Description of ground-sloth material from 
 White Bluffs near Hanford, Wash¬ 
 ington. 110 
 
 Pacific Coast Province. 112 
 
 Mylodontidae of Rancho La Brea. 120 
 
 Characters of family Mylodoniidae. 120 
 
 Characters of genus Mylodon. 120 
 
 Description of material. 120 
 
 Dermal ossicles. 120 
 
 Cranium. 121 
 
 Mandible. 127 
 
 Dentition. 128 
 
 Endocranial cast. 131 
 
 Elements of hyoid arch. 132 
 
 Vertebrae. 133 
 
 Haemapophyses. 140 
 
 Costal ribs. 141 
 
 Sternal ribs. 141 
 
 Sternum. 142 
 
 Scapula. 145 
 
 Clavicle. 146 
 
 Humerus. 146 
 
 Ulna. 147 
 
 Radius. 147 
 
 Falciform bone of manus. 147 
 
 Scaphoid. 148 
 
 Lunar. 150 
 
 Cuneiform. 150 
 
 Pisiform. 151 
 
 Unciform. 151 
 
 Magnum. 152 
 
 Trapezoid. 154 
 
 m 
 
 
IV 
 
 CONTENTS. 
 
 Page 
 
 Mylodontidae of Rancho La Brea ( Continued ). 
 
 Description of material ( Continued ). 
 
 Trapezium and metacarpal 1. 154 
 
 Co-ossified phalanges I and II, digit I, 
 
 manus. 156 
 
 Phalanx III, digit I, manus. 156 
 
 Metacarpal II. 157 
 
 Phalanx I, digit II, manus. 158 
 
 Phalanx II, digit II, manus. 159 
 
 Phalanx III, digit II, manus. 159 
 
 Metacarpal III. 160 
 
 Sesamoid bones for metacarpal III. 162 
 
 Phalanx I, digit III, manus. 162 
 
 Phalanx II, digit III, manus. 163 
 
 Phalanx III, digit III, manus. 163 
 
 Metacarpal IV. 164 
 
 Sesamoid bones for metacarpal IV. 165 
 
 Phalanx I, digit IV, manus. 166 
 
 Rudiment representing phalanges II and 
 
 III, digit IV, manus. 167 
 
 M etacarpal V. 167 
 
 Proximal rudimentary element, digit V, 
 
 manus. 168 
 
 Comparison of manus. 168 
 
 Innominate bone. 169 
 
 Femur. 171 
 
 Patella. 172 
 
 Tibia. 172 
 
 Fibula. 173 
 
 Falciform bone of pes. 174 
 
 Calcaneum. 174 
 
 Page 
 
 Mylodontidae of Rancho La Brea ( Continued). 
 Description of material ( Continued ). 
 
 Astragalus. 175 
 
 Cuboid. 176 
 
 Navicular. 177 
 
 Ectocuneiform. 178 
 
 Mesocuneiform. 178 
 
 Metatarsal II. 179 
 
 Co-ossified phalanges I and II, digit II, 
 
 pes. 180 
 
 Phalanx III, digit II, pes. 181 
 
 Metatarsal III. 181 
 
 Phalanx I, digit III, pes. 182 
 
 Phalanx II, digit III, pes. 183 
 
 Phalanx III, digit III, pes. 184 
 
 Metatarsal IV. 185 
 
 Sesamoid bones for metatarsal IV. 186 
 
 Phalanx I, digit IV, pes. 187 
 
 Rudiment representing phalanges II and 
 
 III, digit IV, pes. 188 
 
 Metatarsal V. 188 
 
 Proximal rudimentary elements, digit V, 
 
 pes. 189 
 
 Comparison of pes. 190 
 
 Mylodon harlani tenuiceps Stock. 192 
 
 Occurrence and description of Mylodontidae in the 
 
 Pleistocene of western North America. 194 
 
 Great Basin Province. 194 
 
 Pacific Coast Province. 199 
 
 Literature. 204 
 
ILLUSTRATIONS. 
 
 PLATES. 
 
 Frontispiece. Nothrotherium shastense Sinclair. Restoration by Charles R. Knight under direction of Chester Stock. 
 
 Plate 1a. 
 
 View looking north showing Los Angeles Museum pits 61 and 67, Rancho La Brea. Photograph by L. E. Wyman. Courtesy 
 of Los Angeles Museum. 
 
 1b. 
 
 View looking northeast showing final stage in removal of wall separating Los Angeles Museum pits 61 and 67, Rancho La Brea. 
 Photograph by L. E. Wyman. Courtesy of Los Angeles Museum. 
 
 Plate 2. 
 
 Nothrotherium shastense Sinclair. 
 
 Figs. 1 and 2. Skull, No. 1800-11, dorsal and ventral views; X0.44. 
 
 Figs. 3 and 4. Skull and mandible, No. 1800-11 and No. 1801-7, anterior and posterior views; X 0.44. 
 
 Los Angeles Museum Collection. Rancho La Brea Pleistocene. 
 
 Plate 3. 
 
 Nothrotherium shastense Sinclair. 
 
 Figs. 1, 2, and 3. Endocranial cast. Fig. 1, dorsal view; fig. 2, lateral view; fig. 3, ventral view; X 0.75. 
 Los Angeles Museum Collection. 
 
 Hapalops sp. 
 
 Figs. 4, 5, and 6. Endocranial cast. Fig. 4, dorsal view; fig. 5, lateral view; fig. 6, ventral view; X 0.75. 
 Los Angeles Museum Collection. 
 
 Plate 4. 
 
 Nothrotherium shastense Sinclair. 
 
 Fig. 1 . Skull and mandible, No. 1800-11 and No. 1801-7, lateral view; X 0.44. 
 
 Fig. 2. Mandible, No. 1801-7, dorsal view; X 0.44. 
 
 Figs. 3 and 4. Cervical vertebrae I-VII inclusive, lateral and dorsal views; X 0.25. 
 
 Los Angeles Museum Collection. Rancho La Brea Pleistocene. 
 
 Plate 5. 
 
 Nothrotherium shastense Sinclair. 
 
 Figs. 1 and 2. Thoracic and lumbar vertebrae, dorsal and lateral views; X 0.25. 
 
 Fig. 3. First thoracic vertebra, No. 1809-1, anterior and ventral views; X 0.25; az., anterior zygapophysis; c./., facet for 
 capitulum of rib; t. /., facet for tuberculum of rib. 
 
 Fig. 4. Third thoracic vertebra, No. 1811-1, anterior view; X 0.25. 
 
 Fig. 5. Last thoracic vertebra. No. 1825-1, anterior view; X 0.25. 
 
 Fig. 6. First lumbar vertebra, No. 1826-1, anterior view; X 0.25. 
 
 Fig. 7. Third lumbar vertebra, No. 1828-1, anterior view; X 0.25. 
 
 Los Angeles Museum Collection. Rancho La Brea Pleistocene. 
 
 Plate 6. 
 
 Nothrotherium shastense Sinclair. 
 
 Figs. 1 and 2. Caudal vertebrae, dorsal and lateral views; X 0.25. 
 
 Fig. 3. First caudal vertebra, No. 1830-1, anterior view; X 0.25. 
 
 Fig. 4. Sixth caudal vertebra, No. 1835-1, anterior view; X 0.25. 
 
 Fig. 5. Eleventh caudal vertebra. No. 1840-1, anterior view; X 0.25. 
 
 Los Angeles Museum Collection. Rancho La Brea Pleistocene. 
 
 Plate 7. 
 
 Nothrotherium shastense Sinclair. 
 
 Fig. 1 . Sternal segments and sternal ribs, ventral view; X 0.33. I and VI, first and sixth sternal elements; II (?) and VIII, 
 sternal ribs. 
 
 Fig. 2. Costal ribs, posterior view; X 0.33. I, VI, X, XIII, XVII, first costal rib fused with first sternal rib, sixth, tenth, 
 thirteenth, seventeenth costal ribs. 
 
 Los Angeles Museum Collection. Rancho La Brea Pleistocene. 
 
 Plate 8. 
 
 Nothrotherium shaste7ise Sinclair. 
 
 Figs. 1, la. Right humerus, No. 1874-R-l. Fig. 1, anterior view; fig. la, view of distal end; X 0.50. 
 
 Figs. 2, 2a. Right humerus, No. 1874-R-l. Fig. 2, inner view; fig. 2a, view of proximal end; X 0.50. 
 
 Figs. 3, 3a. Right scapula, No. 1868-R-2. Fig. 3, outer view; fig. 3a, view of proximal end; X 0.25. 
 
 Fig. 4. Right clavicle, No. 1878-R-2, anterior view; X 0.50. 
 
 Los Angeles Museum Collection. Rancho La Brea Pleistocene. 
 
 Plate 9. 
 
 Nothrotherium shastense Sinclair. 
 
 Figs. 1, 2, and 3. Left ulna, No. 1873-L-l. Fig. 1, inner view; fig. 2, outer view; fig. 3, radial view; X 0.50. 
 
 Figs. 4 and 5. Left radius without distal epiphysis, No. 1872-L-l. Fig. 4, anterior view; fig. 5, inner view; X 0.50. 
 
 Los Angeles Museum Collection. Rancho La Brea Pleistocene. 
 
 v 
 
VI 
 
 ILLUSTRATIONS. 
 
 Plate 10. 
 
 Nothroiherium shastense Sinclair. 
 
 Figs. 1 and 2. Pelvis, No. 1892-1. Fig. 1, posterior view; fig. 2, dorsal view; X 0.22. 
 
 Los Angeles Museum Collection. Rancho La Brea Pleistocene. 
 
 Plate 11. 
 
 Nothrotherium shastense Sinclair. 
 
 Figs. 1 and 2. Pelvis, No. 1892-1. Fig. 1, lateral view; fig. 2, anterior view; X 0.22. 
 
 Los Angeles Museum Collection. Rancho La Brea Pleistocene. 
 
 Plate 12. 
 
 Nothrotherium shastense Sinclair. 
 
 Figs. 1, la, and 11?. Left femur, No. 1871-L-l. Fig. 1, anterior view; fig. la, view of proximal end; fig. 11?, view of distal 
 end; X 0.50. 
 
 Figs. 2 and 3. Left patella, No. 1S95-L-1. Fig. 2, femoral view; fig. 3, anterior view; X 0.50. 
 
 Los Angeles Museum Collection. Rancho La Brea Pleistocene. 
 
 Plate 13. 
 
 Nothrotherium shastense Sinclair. 
 
 Figs. 1 and 2. Left femur, No. 1871-L-l. Fig. 1, posterior view; fig. 2, inner view; X 0.50. 
 
 Los Angeles Museum Collection. Rancho La Brea Pleistocene. 
 
 Plate 14. 
 
 Nothrotherium shastense Sinclair. 
 
 Figs. 1, la, and 11?. Right tibia, No. 1870-R-l. Fig. 1, anterior view; fig. la, view of proximal end; fig. 16, view of distal 
 end; X 0.50. 
 
 Fig. 2. Right tibia, No. 1870-R-l, inner view; X 0.50. 
 
 Fig. 3. Right fibula, No. 1869-R-l, tibial view; X 0.50. 
 
 Los Angeles Museum Collection. Rancho La Brea Pleistocene. 
 
 Plate 15. 
 
 Nothrotherium shastense Sinclair. 
 
 Figs. 1 and 2. Right pes. Fig. 1, dorsal view; fig. 2, inner view; X 0.50. 
 
 Los Angeles Museum Collection. Rancho La Brea Pleistocene. 
 
 Plate 16. 
 
 Nothrotherium shastense Sinclair. 
 
 Mounted skeleton, viewed from right side; X 0.094. 
 
 Los Angeles Museum Collection. Prepared by J. W. Lytle. 
 Angeles Museum. 
 
 Plate 17. 
 
 Photograph by L. E. Wyman. 
 
 Megalonyx jeffersonii californicus Stock. 
 
 Figs. 1 and 2. Mandible, No. 21429. Fig. 1, dorsal view; fig. 2, view of left side; X 0.50. 
 University of California Collection. Rancho La Brea Pleistocene. 
 
 Courtesy of Los 
 
 Plate 18. 
 
 Megalonyx jeffersonii californicus Stock. 
 
 Figs. 1, la, and 2. First thoracic vertebra, No. 22111 U. C. C. Fig. 1, anterior view; fig. la, ventral view; fig. 2, lateral 
 view; X 0.33. az, anterior zygapophysis; c. /., facet for capitulum of rib; t. /., facet for tuberculum of rib. 
 Figs. 3 and 4. Anterior thoracic vertebra, No. 24243 U. C. C. Fig. 3, anterior view; fig. 4, lateral view; X 0.33. 
 
 Figs. 5 and 6. Posterior thoracic vertebra, No. 24244 U. C. C. Fig. 5, anterior view; fig. 6, lateral view; X 0.33. 
 
 Figs. 7, 7a, 8, and 8a. Two caudal vertebrae, Nos. 6003-1, 6003-2 L. A. M. anterior and dorsal views; X 0.33. Specimens 
 referred tentatively to M. j. californicus. 
 
 Los Angeles Museum and University of California Collections. Rancho La Brea Pleistocene. 
 
 Plate 19. 
 
 Megalonyx jeffersonii californicus Stock. 
 
 Figs. 1, la, and 16. Left humerus, No. 21003. Fig. 1, anterior view; fig. la, view of proximal end; fig. 16, view of distal 
 articulation; X 0.50. 
 
 Fig. 2. Left humerus, No. 21003, inner view; X 0.50. 
 
 University of California Collection. Rancho La Brea Pleistocene. 
 
 Plate 20. 
 
 Megalonyx jeffersonii californicus Stock. 
 
 Figs. 1, 2, and 3. Left ulna, No. 23192. Fig. 1, outer view; fig. 2, inner view; fig. 3, radial view; X 0.50. 
 University of California Collection. Rancho La Brea Pleistocene. 
 
 Dermal ossicles; X 1. 
 
 Los Angeles Museum Collection. 
 
 Plate 21. 
 
 Mylodon harlani Owen. 
 Rancho La Brea Pleistocene. 
 
ILLUSTRATIONS. 
 
 vii 
 
 Plate 22. 
 
 Mylodon harlani Owen. 
 
 Skull, No. 1717-32, dorsal view; X 0.40. 
 
 Los Angeles Museum Collection. Rancho La Brea Pleistocene. 
 
 Plate 23. 
 
 Mylodon harlani Owen. 
 
 Fig. 1. Skull, No. 1717-32, ventral view; X 0.40. 
 
 Fig. 2. Premaxillaries, ventral view; X 0.40. 
 
 Fig Jth* Right incus; X 2. 
 
 Fig. Left malleus; X 2. 
 
 Los Angeles Museum Collection. Rancho La Brea Pleistocene. 
 
 Plate 24. 
 
 Mylodon harlani Owen. 
 
 Skull and mandible, No. 1717-32 and No. 1718-1, lateral view; X 0.40. 
 
 Los Angeles Museum Collection. Rancho La Brea Pleistocene. 
 
 Plate 25. 
 
 Mylodon harlani Owen. 
 
 Figs. 1 and 2. Skull and mandible, No. 1717-32 and No. 1718-1. Fig. 1, anterior view; fig. 2, posterior view; X 0.40. 
 
 Los Angeles Museum Collection. Rancho La Brea Pleistocene. 
 
 Plate 26. 
 
 Mylodon harlani Owen. 
 
 Figs. 1, 2, and 3. Endocranial cast. Fig. 1, dorsal view; fig. 2, lateral view; fig. 3, ventral view" X 0.75. 
 
 Los Angeles Museum Collection. Rancho La Brea Pleistocene. 
 
 Plate 27. 
 
 Mylodon harlani Owen. 
 
 Fig. 1. Mandible, No. 1718-1; dorsal view; X 0.40. 
 
 Fig. 2. Hyoid arch, lateral view; X 0.40. Sh, stylohyal; eh, epiphyal; ch, ceratohyal; hh , basihyal; th, thyrohyal; tc, ossified 
 thyroid cartilage. 
 
 Figs. 3 and 4. Cervical vertebrae I to VII. Fig. 3, dorsal view; fig. 4, lateral view; X 0.25. 
 
 Los Angeles Museum Collection. Rancho La Brea Pleistocene. 
 
 Plate 28. 
 
 Mylodon harlani O-wen. 
 
 Thoracic vertebrae; X 0.20. 
 
 Los Angeles Museum Collection. Rancho La Brea Pleistocene. 
 
 Plate 29. 
 
 Mylodon harlani Owen. 
 
 Caudal vertebrae and haemapophyses. Figs. 1 and 2, X 0.20; figs. 3 to 7; X 0.50. 
 
 Los Angeles Museum Collection. Rancho La Brea Pleistocene. 
 
 Plate 30. 
 
 Mylodon harlani Owen. 
 
 Sternal segments and sternal ribs; X 0.25. 
 
 Los Angeles Museum Collection. Rancho La Brea Pleistocene. 
 
 Plate 31. 
 
 Mylodon harlani Owen. 
 
 Figs. 1 and la. Right scapula, No. 1715-R-15. Fig. 1, outer view; fig. la, view of proximal end; X 0.33. 
 
 Fig. 2. Right clavicle, No. 1738-R-l, anterior view; X 0.33. 
 
 Los Angeles Museum Collection. Rancho La Brea Pleistocene. 
 
 Plate 32. 
 
 Mylodon harlani Owen. 
 
 Figs. 1 and la. Right humerus, No. 1712-R-l. Fig. 1, posterior view; fig. la, view of proximal end; X 0.33. 
 
 Figs. 2 and 2a. Right humerus, No. 1712-R-l. Fig. 2, anterior view; fig. 2a, view of distal end; X 0.33. 
 
 Los Angeles Museum Collection. Rancho La Brea Pleistocene. 
 
 Plate 33. 
 
 Mylodon harlani Owen. 
 
 Figs. 1, 2, and 3. Right ulna, No. 1714-R-27. Fig. 1, outer view; fig. 2, inner view; fig. 3, radial view; X 0.33. 
 
 Figs. 4, 4a, and 46. Right radius, No. 1713-R-3. Fig. 4, anterior view; fig. 4a, view of proximal articulation; fig. 46, view 
 of distal end; X 0.33. 
 
 Fig. 5. Right radius, No. 1713-R-3, ulnar view; X 0.33. 
 
 Los Angeles Museum Collection. Rancho La Brea Pleistocene. 
 
 Plate 34. 
 
 Mylodon harlani Owen. 
 
 Fig. 1. Right manus, dorsal view; X 0.33. Cu, cuneiform; lu, lunar; un, unciform; mg, magnum; tr, trapezoid; tin, trape¬ 
 zium; me I, first metacarpal; I, II, III, IV, V, digits. 
 
 Fig. 2. Right manus, palmar view; X 0.33. Pi, pisiform; s and s', sesamoid bones of the third and fourth metacarpals. 
 Fig. 3. Right manus, inner view; sc, scaphoid; remaining letters and numerals as in other figures; X 0.33. 
 
 Los Angeles Museum Collection. Rancho La Brea Pleistocene. 
 
ILLUSTRATIONS. 
 
 viii 
 
 Pelvis, No. 1719-1, anterior view; X 0.20. 
 
 Los Angeles Museum Collection. 
 
 Plate 35. 
 
 Mylodon harlani Owen. 
 Rancho La Brea Pleistocene. 
 
 Pelvis, No. 1719-1, posterior view; X 0.20. 
 Los Angeles Museum Collection. 
 
 Plate 36. 
 
 Mylcdon harlani Owen. 
 Rancho La Brea Pleistocene. 
 
 Plate 37. 
 
 Mylodon harlani Owen. 
 
 Figs. 1 and 2. Pelvis, No. 1719-1. Fig. 1, lateral view; fig. 2, dorsal view; X 0.20. 
 Los Angeles Museum Collection. Rancho La Brea Pleistocene. 
 
 Plate 38. 
 
 Mylodon harlani Owen. 
 
 Figs. 1, la, and lb. Right femur, No. 1704-R-15. Fig. 1, anterior view; fig. la, view of proximal end; fig. 16, view of distal 
 end; X 0.33. 
 
 Figs. 2 and 3. Right patella, No. 1710-R-lo. Fig. 2, femoral view; fig. 3, anterior view; X 0.33. 
 
 Los Angeles Museum Collection. Rancho La Brea Pleistocene. 
 
 Plate 39. 
 
 Mylodon harlani Owen. 
 
 Figs. 1 and 2. Right femur, No. 1704-R-lo. Fig. 1, posterior view; fig. 2, inner view; X 0.33. 
 
 Los Angeles Museum Collection. Rancho La Brea Pleistocene. 
 
 Plate 40. 
 
 Mylodon harlani Owen. 
 
 Figs. 1, la, and 16. Right tibia, No. 1702-R-5. Fig. 1, anterior view; fig. la, view of proximal end; fig. 16, view of distal 
 end; X 0.33. 
 
 Fig. 2. Right tibia, No. 1702-R-5, inner view; X 0.33. 
 
 Figs. 3 and 4. Right fibula, No. 1703-R-7; fig. 3, tibial view; fig. 4, outer view; X 0.33. 
 
 Los Angeles Museum Collection. Rancho La Brea Pleistocene. 
 
 Plate 41. 
 
 Mylodon harlani Owen. 
 
 Figs. 1, 2, and 3. Right pes. Fig. 1, dorsal (outer) view; fig. 2, inner (dorsal) view; fig. 3, plantar (inner) view; X 0.33. 
 
 Ca, calcaneum; as, astragalus; cu, cuboid; na, navicular; ec, ectocuneiform; me, mesocuneiform; ml. II, ml. V, 
 second and fifth metatarsals; II, III, IV, V, digits. 
 
 Los Angeles Museum Collection. Rancho La Brea Pleistocene. 
 
 Plate 42. 
 
 Mylodon harlani lenuiceps Stock. 
 
 Fig. 1. Ventral, posterior, and dorsal views of skull, No. 1716-1. 
 
 Mylodon harlani Owen. 
 
 Figs. 2, 3, and 4. Same views as in fig. 1. Fig. 2, No. 1717-1; fig. 3, No. 1717-8; fig. 4, No. 1717-22. 
 
 Los Angeles Museum Collection. Rancho La Brea Pleistocene. Photographs by L. E. Wyman. Courtesy of 
 Los Angeles Museum. 
 
 Plate 43. 
 
 Mylodon harlani Owen. Figs. 1 to 4. Mandibles, dorsal and lateral views. Fig. 1, No. 1718-20; fig. 2, No. 1718-21; fig. 3, 
 No. 1718-29; fig. 4, No. 1718-3. 
 
 Mylodon harlani tenuiceps Stock. Fig. 5, skull, No. 1716-1, lateral view. 
 
 Mylodon harlani Owen. Figs. 6, 7, and 8, skulls, Nos. 1717-1, 8, 22, lateral views. 
 
 Los Angeles Museum Collection. Rancho La Brea Pleistocene. Photographs by L. E. Wyman. 
 
 Plate 44. 
 
 Mylodon harlani Owen. 
 
 Mounted skeleton, viewed from right side; X 0.077. 
 
 Los Angeles Museum Collection. Prepared by J. W. Lytle. Photograph by L. E. Wyman. Courtesy of Los 
 Angeles Museum. 
 
 Plate 45. 
 
 Mylodon harlani Owen. 
 
 Restoration by Charles R. Knight under the direction of Chester Stock. 
 
 Plate 46. 
 
 Mylodon harlani Owen. 
 
 Replica of footprints in Pleistocene deposits at the Nevada State Penitentiary, Carson City, Nevada. Diagram indicates 
 size of imprints, length of stride, and width of straddle, a, possible impression of ungual phalanx, digit III, 
 manus. Cast prepared by J. W. Lytle. Los Angeles Museum Collection. 
 
 Plate 47. 
 
 Mylodon harlani Owen. 
 
 Figs. 1 to 6. Footprints on floor of tunnel in east wall of prison yard, Nevada State Penitentiary, Carson City, Nevada. 
 
ILLUSTRATIONS. 
 
 IX 
 
 TEXT-FIGURES. 
 
 Fig. 1.—Chart showing geologic and geographic distribution of the Megalonychinae and Mylodontidae. 
 
 Fig. 2.— A to G, ground-sloth remains from Tertiary beds of Western North America. A, B, and C, ungual phalanx, No. 
 1096 U. C. C., from Rattlesnake formation, Eastern Oregon; D and E, tooth, No. 22110 U. C. C. from Pinole Tuff-Orinda- 
 Siesta deposits, middle California; F and G, metapodial, No. 22900 U. C. C., from Rattlesnake formation, Eastern Oregon. 
 All figures X 1. 
 
 Fig. 3. —Topographic map of Rancho La Brea showing location of principal excavations of the University of California and 
 Los Angeles Museum. 
 
 Fig. 4.— A, Nothrotherium shastense Sinclair. A 1 , Hapalops sp. Skeletons shown in lateral view and to same scale. A 1 
 after Matthew. 
 
 Fig. 5.— Nothrotherium shastense Sinclair. Skull, dorsal view. X 0.33. », nasal; mx, maxillary; l, lachrymal; If, lachrymal 
 
 foramen; fr, frontal; ma, malar; sq, squamosal; pa, parietal; so, supra-occipital. Rancho La Brea Pleistocene. 
 
 Fig. 6.— Nothrotherium shastense Sinclair. Skull, ventral view. X 0.33. mx, maxillary; anf, antorbital foramen; ma, 
 malar; pi, palatine; pic, palatine canal; sq, squamosal; pt, pterygoid; s, pterygoid sinus: opt, orifice of pterygoid sinus; 
 per, periotic; bs, basisphenoid; bo, basioccipital; Jim, foramen lacerum medum; flp, foramen lacerum posterius; stp, 
 stylohyal process; 2, 5, second and fifth superior teeth. Rancho La Brea Pleistocene. 
 
 Fig. 7.— Nothrotherium shastense Sinclair. Skull, lateral view. X 0.33. 3, third superior tooth; fr, foramen rotundum; 
 
 fc, foramen ovale; remaining letters as in other figures. Rancho La Brea Pleistocene. 
 
 Fig. 8.— Nothrotherium shastense Sinclair. Skull, posterior view. X 0.33. Letters as in other figures. Rancho La Brea 
 Pleistocene. 
 
 Fig. 9.— A, Nothrotherium shastense Sinclair. A 1 , Hapalops longiceps Scott. Skull and mandible, lateral view. X 0.33. 
 A 1 adopted and reversed from Scott. 
 
 Fig. 10.—A, Nothrotherium shastense Sinclair. A 1 , Pronothrotherium typicum Ameghino. A 2 , Hapalops longiceps Scott. 
 Ventral view of skull. A 1 after Rovereto; A 2 , after Scott. 
 
 Fig. 11.— Nothrotherium shastense Sinclair. Vertical longitudinal section of skull, No. 1806. X 0.33. n, nasal; fr, frontal; 
 s, frontal sinus; pa, parietal; so, supraoccipital; per, periotic; bo, basioccipital; bs, basisphenoid; ps, presphenoid; pt, 
 pterygoid; opt, orifice of pterygoid sinus; vo, vomer; pi, palatine; mx, maxillary, rnxl, maxillo-turbinal; et, ethmo-tur- 
 binals; ix, x, xi, foramen lacerum posterius; xii, condylar foramen. 
 
 Fig. 12.—A, Nothrotherium shastense Sinclair. A 1 , Pronothrotherium typicum Ameghino; A 2 , Hapalops longiceps Scott. 
 Dorsal view of mandible. X 0.33. A 1 , after Rovereto; A 2 , after Scott. 
 
 Fig. 13.— Nothrotherium shastense Sinclair. Superior dentition. Lateral and occlusal views. X 1. 
 
 Fig. 14.— Nothrotherium shastense Sinclair. Inferior dentition. Lateral and occlusal views. X 1. 
 
 Fig. 15.—A, Nothrotherium shastense Sinclair; A 1 , Hajmlops sp. Views showing relation of size of endocranium to that of 
 skull. X 0.33. 
 
 Fig. 16.— Nothrotherium shastense Sinclair. A, B, C, outer, anterior, and inner views of right half of second haemapophysis, 
 No. 1919-R-l. D, E, F, outer, anterior, and inner views of right half of fourth haemapophysis, No. 1921-R-l. X 0.50. 
 Rancho La Brea Pleistocene. 
 
 Fig. 17.— Nothrotherium shastense Sinclair. Fifth haemapophysis, No. 1922-1. X 0.50. A, anterior view; B, outer view; 
 C, posterior view; D, dorsal view; E, ventral view. Rancho La Brea Pleistocene. 
 
 Fig. 18.— Nothrotherium shastense Sinclair. Sixth haemapophysis, No. 1923-1. X 0.50. A, outer view; B, posterior view. 
 Rancho La Brea Pleistocene. 
 
 Fig. 19.— Nothrotherium shastense Sinclair. Ninth haemapophysis, No. 1926-1. X 0.50. A, dorsal view; B, outer view; 
 C, ventral view. Rancho La Brea Pleistocene. 
 
 Fig. 20.— Nothrotherium shastense Sinclair. Sternal segments; anterior, ventral, and right side views. X 0.50. A, B, C, 
 second segment, No. 1896-1; D, E, F, third segment, No. 1897-1; G, H, I, fifth segment, No. 1899-1; J, K, L, sixth 
 segment, No. 1900-1. Rancho La Brea Pleistocene. 
 
 Fig. 21.— Nothrotherium shastense Sinclair. Left scaphoid, No. 1944-L—1. A, distal view; B, dorsal view; C, proximal view. 
 X 0.50. Rancho La Brea Pleistocene. 
 
 Hapalops sp. Left scaphoid, No. 15171 P. U. C. A 1 , B l , C 1 , same views as in A, B, C. X 1.0. Santa Cruz 
 Miocene. 
 
 Fig. 22. Nothrotherium shastense Sinclair. Right trapezium and metacarpal I, Nos. 1912-R-l, 1913-R-l. A, dorsal view; 
 B, proximal end; C, distal end; D, inner view. X 0.50. Rancho La Brea Pleistocene. 
 
 Fig. 23.— Nothrotherium shastense Sinclair. Left metacarpal II, No. 1914-L-l. A, inner view; B, dorsal view; C, outer 
 view; D, proximal end; E, distal end. X 0.50. Rancho La Brea Pleistocene. 
 
 Hapalops sp. Left metacarpal II, No. 15171 P. U. C.; same views as in A, B, C, D, E. X 1.0. Santa Cruz 
 Miocene. 
 
 Fig. 24.— Nothrotherium shastense Sinclair. Second phalangeal elements of manus; proximal, lateral, and distal views. 
 X 0.50. a, No. 1930-L-l, digit II; b, No. 1929-L-l, digit III; c, No. 1928-L-l, digit IV. Rancho La Brea Pleistocene. 
 
X 
 
 ILLUSTRATIONS. 
 
 Fig. 25.— Nothrotherium shastense Sinclair. Ungual phalanges of manus, lateral views with cross-sections through claw- 
 processes. X 0.50. a, digit II; b, digit III; c, digit IV. Rancho La Brea Pleistocene. 
 
 Fig. 26. — Nothrotherium shastense Sinclair. Right metacarpal V, No. 1917-R-l. A, dorsal view; B, inner view; C, outer 
 view. X 0.50. Rancho La Brea Pleistocene. 
 
 Hapalops sp. Left metacarpal V, No. 15171 P. U. C. A 1 , dorsal view; B 1 , inner view. X 1.0. Santa Cruz Miocene. 
 
 Fig. 27. — Megalonyx, probably wheatleyi Cope. Right manus, Amer. Mus. Spec., dorsal view'. X 0.50. Port Kennedy 
 fissure, Pennsylvania, Pleistocene. 
 
 Fig. 28.- — Nothrotherium maquinense (Lund). Right manus, dorsal view, after Winge; approx. X 0.66. Brazil, Pleistocene. 
 
 Fig. 29.— Nothrotherium shastense Sinclair. Right posterior limb. A, lateral view; B, anterior view. X 0.166. Rancho 
 La Brea Pleistocene. 
 
 Fig. 30.— Nothrotherium shastense Sinclair. Right calcaneum, No. 1877-R-l. A, inner view; B, view of articulating end; 
 C, dorsal view. X 0.50. Rancho La Brea Pleistocene. 
 
 Hapalops elongatus Ameghino. Right calcaneum, No. 15155, P. U. C. A 1 , B l , C 1 , same views as in A, B, C. X 1-0. 
 Santa Cruz Miocene. 
 
 Fig. 31.— Nothrotherium shastense Sinclair. Right astragalus, No. 1875-R—1. A, tibial view; B, inner view; C, calcaneal 
 view; D, lateral view’. X 0.50. Rancho La Brea Pleistocene. 
 
 Megalonyx, probably wheatleyi Cope. Right astragalus, Amer. Mus. Spec. A 1 , B l , C 1 , D', same views as in A, B, C, D. 
 
 X 0.50. Port Kennedy fissure, Pennsylvania, Pleistocene. 
 
 Hapalops sp. Right astragalus, No. 15594, P. U. C. A 2 , B 2 , C 2 , D 2 , same views as in A, B, C, D. X 1.0. Santa 
 Cruz Miocene. 
 
 Fig. 32.— Nothrotherium shastense Sinclair. Right cuboid, No. 1885-R-l. A, distal view; B, dorsal view; C, proximal view. 
 X 0.50. Rancho La Brea Pleistocene. 
 
 Hapalops indifferens Ameghino. Right cuboid, No. 15110, P. U. C. A 1 , B 1 , C 1 , same views as in A, B, C. X 1.0 
 Santa Cruz Miocene. 
 
 Fig. 33.— Nothrotherium shastense Sinclair. Right navicular, No. 1891-R-l. A, distal view; B, view of cuboid face; C, 
 dew of astragalar face. X 0.50. Rancho La Brea Pleistocene. 
 
 Hapalops elongatus Ameghino. Right navicular, No. 15545, P. U. C. A 1 , B l , C 1 , same views as in A, B, C. X 1.0. 
 Santa Cruz Miocene. 
 
 Fig. 34.— Nothrotherium shastense Sinclair. Left ectocuneiform, No. 1943-L-2. A, navicular view; B, metatarsal view. 
 X 0.50. Rancho La Brea Pleistocene. 
 
 Hapalops elongatus Ameghino. Right ectocuneiform, No. 15160 P. U. C. A 1 , B 1 , same views as in A, B. X 1.0. 
 Santa Cruz Miocene. 
 
 Fig. 35.— Nothrotherium shastense Sinclair. Right entocuneiform and metatarsal I, No. 1880-R-2. A, ventral view; 
 B, external view; C, dorsal view; D, proximal view showing these elements in articulation with mesocuneiform and 
 metatarsal II. X 0.50. Rancho La Brea Pleistocene. 
 
 Fig. 36. — Nothrotherium shastense Sinclair. Right mesocuneiform and metatarsal II, No. 1881-R-2. A, outer view; B, 
 dorsal view; C, inner view; D, proximal end; E, distal end; F, proximal end of metatarsal II, No. 1881-R-l. 
 X 0.50. Rancho La Brea Pleistocene. 
 
 Hapalops elongatus Ameghino. Right metatarsal II, No. 15545 P. U. C. A 1 , B l , C l , D 1 , E l , same view's as in A, B, 
 C, D, E. X 1.0. Santa Cruz Miocene. 
 
 Fig. 37.— Nothrotherium shastense Sinclair. Right phalanx I, digit II, pes, No. 1886-R-l. A, proximal view; B, lateral 
 view; C, distal view. X 0.50. 
 
 Fig. 38.— Nothrotherium shastense Sinclair. Phalanx II, digit II, pes. No. 1887-R-l; proximal, lateral, and distal views. 
 X 0.50. Rancho La Brea Pleistocene. 
 
 Fig. 39.— Nothrotherium shastense Sinclair. Right metatarsal III, No. 1882-R-l. A, distal view; B, inner view; C, outer 
 view; D, proximal view. X 0.50. Rancho La Brea Pleistocene. 
 
 Hapalops elongatus Ameghino. Right metatarsal III, No. 15545 P. U. C. A 1 , B 1 , C 1 , D\ same view's as in A, B, 
 C, D. X 1.0. Santa Cruz Miocene. 
 
 Fig. 40.— Nothrotherium shastense Sinclair. Co-ossified phalanges I and II, digit III, pes. A, B, C, No. 1889-L-2 with inner 
 sesamoid bone, proximal, lateral, and distal views. X 0.50. D, No. 1889-L-l, dorsal view'. X 0.50. Rancho La 
 Brea Pleistocene. 
 
 Fig. 41. — Nothrotherium shastense Sinclair. Ungual phalanges, right pes. Lateral view. X 0.50. a, digit III, No. 1890- 
 R-l; b, digit II, No. 1888-R-l. Rancho La Brea Pleistocene. 
 
 Fig. £2.^Nothrotherium shastense Sinclair. Right metatarsal IV, No. 1883-R-l. A, outer view; B, dorsal view; C, inner 
 view; D, proximal end; E, distal end. X 0.50. Rancho La Brea Pleistocene. 
 
 Hapalops elongatus Ameghino. Right metatarsal IV, No. 15545 P. U. C. A 1 , B\ C L , D 1 , E', same views as in 
 A, B, C, D, E. X 1.0. Santa Cruz Miocene. 
 
 Fig. 43.— Nothrotherium shastense Sinclair. Right metatarsal V, No. 1884-R-l. A, dorsal view; B, inner view; C, ventral 
 view. X 0.50. Rancho La Brea Pleistocene. 
 
 Hapalops elongatus Ameghino. Right metatarsal V, No. 15545 P. U. C. A 1 , B l , C l , same views as in A, B, C. 
 X 1.0. Santa Cruz Miocene. 
 
 Fig. 44.— Megalonyx, sp. A and B, fifth superior tooth, No. 6000-1. X 1.0. Rancho La Brea Pleistocene. 
 
ILLUSTRATIONS. 
 
 xi 
 
 Fig. 45. — Megalonyx jeffersonii californicus Stock. Right unciform, No. 22775 U. C. C. A, proximal view; B, dorsal view; 
 C, ventral view; D, distal view. X 0.50. Rancho La Brea Pleistocene. 
 
 Fig. 46.— Megalonyx jeffersonii californicus Stock. Patella, No. 22776 U. C. C. A, anterior view; B, femoral view. 
 X 0.50. Rancho La Brea Pleistocene. 
 
 Fig. 47.— Megalonyx jeffersonii californicus Stock. Left calcaneum, No. 20095, U. C. C. A, outer view; B, view of articulat¬ 
 ing end; C, inner view. X 0.33. Rancho La Brea Pleistocene. 
 
 Fig. 48. — Megalonyx jeffersonii californicus Stock. Right metatarsal III, No. 20001 U. C. C. A, proximal view; B, outer 
 view; C, inner view; D, distal view. X 0.50. Rancho La Brea Pleistocene. 
 
 Fig. 49. — Megalonyx jeffersonii californicus Stock. Right metatarsal IV, No. 6002-1. A, inner view; B, dorsal view; C, 
 outer view; D, proximal view. X 0.50. Rancho La Brea Pleistocene. 
 
 Fig. 50.— Megalonyx jeffersonii californicus Stock. Right metatarsal V, No. 22773 U. C. C. A, dorsal view; B, inner view 
 X 0.50. Rancho La Brea Pleistocene. 
 
 Fig. 51. —Map showing occurrences of Pleistocene Megalonychinae in western North America. 1, Rancho La Brea; 2, 
 Alameda County; 3, Hawver Cave, Eldorado County; 4, Potter Creek Cave and Samwel Cave, Shasta County; 5, 
 Bautista Creek Badlands, Riverside County; 6, White Bluffs near Hanford, south central Washington; 7, Murphys, 
 Calaveras County; 8, Douglas City, Trinity County; 9, San Pedro, Los Angeles County. 
 
 Fig. 52.— Megalonyx sp. Skull fragment with superior cheek-teeth of right side. X 1.0. White Bluffs, near Hanford, 
 Washington. 
 
 Fig. 53.— Nothrotherium shastense Sinclair. Right metacarpal IV, No. 9443 U. C. C. A, inner view; B, dorsal view; C, 
 outer view; D, proximal end; E, distal end. X 0.50. Samwel Cave Pleistocene, California. 
 
 Hapalops sp. Left metacarpal IV, No. 15171 P. U. C. A 1 , B l , C 1 , D l , E l , same views as in A, B, C, D, E. X 1.0. 
 Santa Cruz Miocene. 
 
 Fig. 54.— Nothrotherium shastense hawveri Stock. Left unciform, No. 22899 U. C. C. A, proximal view; B, dorsal view; C, 
 ventral view; D, distal view. X 0.50. Hawver Cave Pleistocene, California. 
 
 Fig. 55.— Mylodon harlani Owen. Section through middle of a dermal ossicle showing entrance of a vascular canal and inner 
 honeycombed cavity. X 2.0. 
 
 Fig. 56.— Mylodon harlani Owen. Skull, dorsal view. X 0.33. n, nasal; mx, maxillary; l, lachrymal; fr, frontal; sq, squa¬ 
 mosal; pa, parietal; so, supraoccipital. Rancho La Brea Pleistocene. 
 
 Fig. 57.— Mylodon harlani Owen. Skull, ventral view. X 0.33. mx, maxillary; pi, palatine; pi, pterygoid; sq, squamosal; 
 per, periotic; bo, basioccipital; 6s, basisphenoid; flm, foramen lacerum medium; flp, foramen lacerum posterius; fc, 
 condylar foramen; 2, second superior tooth; 5, fifth superior tooth. Rancho La Brea Pleistocene. 
 
 Fig. 58.— Mylodon harlani Owen. Skull, lateral view. X 0.33. os, orbitosphenoid; as, alisphenoid; eo, exoccipital; If, 
 lachrymal foramen;/Za, fo, combined optic foramen and foramen lacerum anterius; fro, combined foramen rotundum and 
 foramen ovale; remaining letters as in other figures. Rancho La Brea Pleistocene. 
 
 Fig. 59.— Mylodon harlani Owen. Skull, posterior view. X 0.33. Letters as in other figures. Rancho La Brea Pleistocene. 
 
 Fig. 60.— Mylodon harlani Owen. Vertical longitudinal section of skull, No. 1717-2. X 0.33. mxt, maxillo-turbinal; et 
 ethmo-turbinals; s, sinus in frontal, parietal, and basisphenoid; O? V 1 , canal for first branch of trigeminal nerve and 
 possibly the optic nerve; V 2 , V 3 , canal for second and third branches of the trigeminal nerve: IX, X, XI, exit for ninth, 
 tenth, eleventh cranial nerves; XII, exit for twelfth nerve; remaining letters as in other figures, Rancho La Brea Pleis¬ 
 tocene. 
 
 Fig. 61.— Mylodon harlani Owen. Skull fragment with right superior teeth of young individual, No. 1717-35; inferior and 
 lateral views. X 0.50. Note small size of first superior tooth. Rancho La Brea Pleistocene. 
 
 Fig. 62.— Mylodon harlani Owen. Occlusal views of second superior tooth, showing variation in shape of crowns. X 0.50. 
 Rancho La Brea Pleistocene. 
 
 Fig. 63.— Mylodon harlani Owen. Occlusal views of left fourth superior tooth, showing variation in shape of crowns. X 
 0.50. Rancho La Brea Pleistocene. 
 
 Fig. 64.— Mylodon harlani Owen. Occlusal views of right fifth superior tooth showing variation in shape of crowns. X 0.50. 
 Rancho La Brea Pleistocene. 
 
 Fig. 65.— Mylodon harlani Owen. Occlusal views of left third inferior tooth showing variation in shape of crowns. X 0.50. 
 Rancho La Brea Pleistocene. 
 
 Fig. 66.— Mylodon harlani Owen. Occlusal views of right fourth inferior tooth, showing variation in shape of crowns. 
 X 0.50. Rancho La Brea Pleistocene. 
 
 Fig. 67.— Mylodon harlani Owen. Cervical vertebra VII, No. 1573-6 with cervical rib, No. 1737-1; anterior view. X 0.33. 
 cr, cervical rib; ccr, facet for capitulum of cervical rib; fcr, facet for tuberculum of cervical rib; vac, vertebrarterial 
 canal; tp, transverse process; az, anterior zygapophysis. Rancho La Brea Pleistocene. 
 
 Fig. 68.— Mylodon harlani Owen. Thoracic vertebra XV, No. 1588-4, anterior and posterior views of right half. X 0.33.; 
 az 1 , lateral anterior zygapophysis; az 2 , medial anterior zygapophysis; pz 1 , lateral posterior zygapophysis; pz 2 , medial pos¬ 
 terior zygapophysis; m, metapophysis; tp, transverse process; cr, facet for capitulum of rib; fr, facet for tuberculum of 
 rib. Rancho La Brea Pleistocene. 
 
Xll 
 
 ILLUSTRATIONS. 
 
 Fig. 09.— Mylodon harlani Owen. A, B, and C, views of falciform bone of manus, No. 1700-19. X 0.50. Rancho La Brea 
 Pleistocene. 
 
 Fig. 70.— Mylodon harlani Owen. Right scaphoid, No. 24261 U. C. C. A, distal surface with facet for trapezoid at right 
 and facet for magnum at left; B, proximal surface with articulation for radius; C, dorsal view with facet for trapezium 
 at right; D, dorso-external view. X 0.50. Rancho La Brea Pleistocene. 
 
 Fig. 71.— Mylodon harlani Owen. Right lunar, No. 24262 U. C. C. A, proximal view of articulating surface for radius; B, 
 distal view 7 with surface for unciform in middle, a small portion of facet for magnum shown on left and facet for cunei¬ 
 form on right; C, dorsal view; D, inner view with surface for scaphoid above and that for magnum below; E, outer view. 
 X 0.50. Rancho La Brea Pleistocene. 
 
 Fig. 72.— Mylodon harlani Owen. Right cuneiform, No. 24263 U. C. C. A, proximal view of ulnar surface; B, dorsal view; 
 C, distal view of surface for unciform; D, palmar view 7 showing facet for pisiform; E, inner view with facet for lunar. 
 X 0.50. Rancho La Brea Pleistocene. 
 
 Fig. 73.— Mylodon harlani Ow 7 en. Right pisiform, No. 24264 U. C. C. .4, outer distal view; B, view of surface for cuneiform; 
 
 C, proximo-internal view. X 0.50. Rancho La Brea Pleistocene. 
 
 Fig. 74.— Mylodon harlani Owen. Right unciform, No. 24265 U. C. C. A, dorsal view; B, inner view; C, outer view showing 
 surface for cuneiform; D, proximal view; E, distal view showing surfaces for metacarpals IV and V. X 0.50. Rancho 
 La Brea Pleistocene. 
 
 Fig. 75.— Mylodon harlani Owen. Right magnum, No. 24266 U. C. C. A, proximal view; B, distal view; C, dorsal view; D, 
 inner view 7 ; E, outer view. X 0.50. Rancho La Brea Pleistocene. 
 
 Fig. 76.— Mylodon harlani Owen. Right trapezoid, No. 24267 U. C. C. A, proximal view; B, inner view showing below the 
 facet for metacarpal II; C, dorsal view; D, outer view showing above deflected surface for scaphoid and below facet for 
 magnum; E, distal view showing from left to right the facets for magnum, metacarpal III, metacarpal II. X 0.50. 
 Rancho La Brea Pleistocene. 
 
 Fig. 77.— Mylodon harlani Owen. A to D, right metacarpal I and trapezium, No. 24268 U. C. C. A, dorsal view; B, 
 proximal view 7 ; C, outer view showing on left a small facet for metacarpal II; D, distal view showing on right surface for 
 digit I and on left the distal portion of facet for metacarpal II. X 0.50. E, left metacarpal I, No. 1482-L-21, dorsal view; 
 F, trapezial moiety, dorsal view; G, trapezial moiety showing contact surface for metacarpal II; H, metacarpal I, No. 
 1482-R-ll, dorsal view; I, No. 1482-R-ll, inner view showing facet for trapezial moiety. X 0.50. Rancho La Brea 
 Pleistocene. 
 
 Fig. 78.— Mylodon harlani Owen. Co-ossified phalanges I and II, digit I, right manus, No. 24269 U. C. C. A, proximal 
 view; B, outer view; C, dorsal view; D, distal view 7 . X 0.50. Rancho La Brea Pleistocene. 
 
 Fig. 79.— Mylodon harlani Owen. Ungual phalanx, digit I, right manus, No. 1471-R-l. A, dorsal view; B, outer view; C, 
 ventral view. X 0.50. Rancho La Brea Pleistocene. 
 
 Fig. 80.— Mylodon harlani Owen. Right metacarpal II, No. 24270 U. C. C. A, outer view; B, dorsal view; C, inner view 7 ; 
 
 D, proximal view; E, distal view. X 0.50. Rancho La Brea Pleistocene. 
 
 Fig. 81.— Mylodon harlani Owen. Phalanx I, digit II, right manus, No. 24271 U. C. C. A, proximal view; B, inner view; 
 
 C, distal view. X 0.50. Rancho La Brea Pleistocene. 
 
 Fig. 82.— Mylodon harlani Owen. Phalanx II, digit II, right manus, No. 24272 U. C. C. A, outer view; B, distal view; C‘ 
 proximal view; D, dorsal view 7 . X 0.50. Rancho La Brea Pleistocene. 
 
 Fig. 83.— Mylodon harlani Owen. Ungual phalanx, digit II, right manus, No. 24273 U. C. C. A, outer view with cross- 
 section of claw 7 -process; B, dorsal view 7 ; C, ventral view. X 0.50. Rancho La Brea Pleistocene. 
 
 Fig. 84.— Mylodon harlani Owen. Right metacarpal III, No. 24274 U. C. C. A, outer view; B, dorsal view; C, inner view 7 ; 
 
 D, proximal view; E, distal view; F, proximal view of another specimen, No. 23148 U. C. C., showing confluence of 
 facets. X 0.50. Rancho La Brea Pleistocene. 
 
 Fig. 85.— Mylodon harlani Owen. Phalanx I, digit III, right manus, No. 24275 U. C. C. A, proximal view; B, outer view; C, 
 distal view. X 0.50. Rancho La Brea Pleistocene. 
 
 Fig. 86.— Mylodon harlani Owen. Phalanx II, digit III, right manus, No. 24276 U. C. C. A, proximal view; B, dorsal view; 
 
 C, outer view; D, distal view 7 . X 0.50. Rancho La Brea Pleistocene. 
 
 Fig. 87.— Mylodon harlani Owen. Ungual phalanx, digit III, right manus, No. 24277 U. C. C. A, outer view with cross- 
 section of claw 7 -process; B, dorsal view; C, ventral view 7 . X 0.50. Rancho La Brea Pleistocene. 
 
 Fig. 88.— Mylodon harlani Owen. Right metacarpal IV, No. 24278 U. C. C. A, inner view; B, dorsal view; C, outer view; 
 
 D, proximal view; E, distal view. X 0.50. Rancho La Brea Pleistocene. 
 
 Fig. 89.— Mylodon harlani Owen. A, B, and C, outer sesamoid, metacarpal IV, left manus, No. 1462-L-4. X 1.0. Rancho 
 La Brea Pleistocene. 
 
 Fig. 90.— Mylodon harlani Owen. A, B, and C, inner sesamoid, metacarpal IV, left manus, No. 1468-L—5 X 1.0. Rancho 
 La Brea Pleistocene. 
 
 Fig. 91.— Mylodon harlani 0w 7 en. Phalanx I, digit IV, right manus, No. 24279 U. C. C. A, proximal view; B, outer view; 
 C, distal view 7 . X 0.50. Rancho La Brea Pleistocene. 
 
 Fig. 92.— Mylodon harlani Owen. Terminal phalanx, digit IV, right manus, No. 1484-R-2. A, proximal view; B, lateral 
 view; C, dorsal view. X 0.50. Rancho La Brea Pleistocene. 
 
ILLUSTRATIONS. 
 
 xiii 
 
 Fig. 93.— Mylodon harlani Owen. Right metacarpal V, No. 24280 U. C. C. A, outer view; B, dorsal view; C, distal view; 
 D, inner view; E, ventral view. X 0.50. Rancho La Brea Pleistocene. 
 
 Fig. 94.— Mylodon harlani Owen. A to D, phalanx I, digit V, right manus, No. 24281 U. C. C. E and F, phalanx I, fused 
 with rudimentary phalanx II, No. 1485-R-5. A, proximal view; B, outer view; C, inner view; D, distal view; E, proxi¬ 
 mal view; F, outer view. X 0.50. Rancho La Brea Pleistocene. 
 
 Fig. 95.— Mylodon harlani Owen. A and B, falciform bone of pes, No. 1701. X 0.50. Rancho La Brea Pleistocene. 
 
 Fig. 96.— Mylodon harlani Owen. Left calcaneum, No. 24282 U. C. C. A, ventral view; B , outer view; C, dorsal view; D, 
 view of articular end. X 0.33. Rancho La Brea Pleistocene. 
 
 Fig. 97.— Mylodon harlani Owen. Left astragalus, No. 23133 U. C. C. A, tibial view; B, lateral view; C, calcaneal view; 
 D, distal view. X 0.33. Rancho La Brea Pleistocene. 
 
 Fig. 98.— Mylodon harlani Owen. Left cuboid, No. 23134 U. C. C. A, proximal view; B, dorsal view; C, distal view; D, 
 outer or calcaneal view; E, inner view. X 0.50. Rancho La Brea Pleistocene. 
 
 Fig. 99.— Mylodon harlani Owen. Left navicular, No. 23135 U. C. C. A, astragalar or proximal view; B, lateral view 
 showing astragalar and cuboidal surfaces; C, distal view. X 0.50. Rancho La Brea Pleistocene. 
 
 Fig. 100.— Mylodon harlani Owen. Left ectocuneiform, No. 23136 TJ. C. C. A, proximal view; B, inner view; C, distal 
 view; D, dorsal view. X 0.50. Rancho La Brea Pleistocene. 
 
 Fig. 101.— Mylodon harlani Owen. Left mesocuneiform, No. 23137 U. C. C. A, outer view showing at right facet for ecto¬ 
 cuneiform; B, proximal view; C, inner view; D, distal view. X 0.50. Rancho La Brea Pleistocene. 
 
 Fig. 102.— Mylodon harlani Owen. Left metatarsal II, No. 23138 U. C. C. A, proximal view; B, outer view; C, inner view; 
 D, distal view. X 0.50. Rancho La Brea Pleistocene. 
 
 Fig. 103.— Mylodon harlani Owen. A and B, inner and dorsal views, right mesocuneiform and metatarsal II, No. 22771 
 U. C. C. C and D, dorsal and inner views of co-ossified mesocuneiform and metatarsal II, No. 22772. X 0.50. Rancho 
 La Brea Pleistocene. 
 
 Fig. 104.— Mylodon harlani Owen. Co-ossified phalanges I and II, digit II, left pes, No. 23139 U. C. C. A, proximal view; 
 B, inner view; C, distal view; D, dorsal view; E, ventral view. X 0.50. Rancho La Brea Pleistocene. 
 
 Fig. 105.— Mylodon harlani Owen. Ungual phalanx, digit II, left pes, No. 23140 U. C. C. A, dorsal view; B, ventral 
 view; C, inner view with cross-section of claw-process. X 0.50. Rancho La Brea Pleistocene. 
 
 Fig. 106.— Mylodon harlani Owen. Left metatarsal III, No. 23141 U. C. C. A, dorsal view; B, outer view with large 
 articulating surface for metatarsal IV; C, inner view with large surface for ectocuneiform; D, proximal view with large sur¬ 
 face on right for ectocuneiform and small surface of articulation on left for cuboid. X 0.50. Rancho La Brea Pleistocene. 
 
 Fig. 107.— Mylodon harlani Owen. Phalanx I, digit III, left pes, No. 22769 U. C. C. A, proximal view; B, inner view; C. 
 distal view. X 0.50. Rancho La Brea Pleistocene. 
 
 Fig. 108.— Mylodon harlani Owen. Phalanx II, digit III, left pes. A to D, No. 22769 U. C. C E to H, No. 23147 U. C. C. 
 A and E, proximal views; B and F, dorsal views; C and G, inner views; D and H, distal views. X 0.50. Rancho La Brea, 
 Pleistocene. 
 
 Fig. 109.— Mylodon harlani Owen. Ungual phalanx, digit III, left pes, No. 23142 U. C. C. A, dorsal view; B, ventral 
 view; C, inner view with cross-section of claw-process. X 0.50. Rancho La Brea Pleistocene. 
 
 Fig. 110.— Mylodon harlani Owen. Left metatarsal IV, No. 23143, U. C. C. A, inner view; B, dorsal view; C, outer view; 
 D, distal view. X 0.50. Rancho La Brea Pleistocene. 
 
 Fig. 111.— Mylodon harlani Owen. A, B, and C, outer sesamoid bone, digit IV, left pes, No. 1493-L-l. X 1-0. Rancho La 
 Brea Pleistocene. 
 
 Fig. 112.— Mylodon harlani Owen. A and B, inner sesamoid bone, digit IV, pes No. 1444-L-l. X 1.0. Rancho La Brea 
 Pleistocene. 
 
 Fig . 113.— Mylodon harlani Owen. Phalanx I with outer sesamoid, digit IV, left pes, No. 1453-L-15. A, proximal view; B, 
 distal view. X 0.50. Rancho La Brea Pleistocene. % 
 
 Fig. 114.— Mylodon harlani Owen. Phalanx I, digit IV, left pes, No. 23144 U. C. C. A, proximal view; B, outer view; C, 
 distal view. X 0.50. Rancho La Brea Pleistocene. 
 
 Fig. 115.— Mylodon harlani Owen. Rudiment representing phalanges II and III, digit IV, left pes, No. 1450-L-2. A proxi¬ 
 mal view; B, inner view; C, dorsal view, X 0.50. Rancho La Brea Pleistocene. 
 
 Fig. 116.— Mylodon harlani Owen. Left metatarsal V, No. 23145 U. C. C. A, dorsal view;B, inner view; C, ventral view 
 X 0.50. Rancho La Brea Pleistocene. 
 
 Fig. 117.— Mylodon harlani Owen. A to D, proximal phalanx, digit V, left pes, No. 23146 U. C. C.; E, co-ossified proximal 
 and terminal rudiments, digit V, left pes, No. 1458-L-3. A, proximal view; B and E, outer views; C, inner view; D, 
 distal view. X 0.50. Rancho La Brea Pleistocene. 
 
 Fig. 118.—Map showing occurrences of Pleistocene Mylodontidae in western North America. 1, Rancho La Brea; 2, Mc- 
 Kittrick, Kern County; 3, San Francisco peninsula; 4, Petaluma, Sonoma County; 5, Contra Costa County; 6, Marys¬ 
 ville, Yuba County; 7, Hawver Cave, Eldorado County; 8, Carson City, Nevada; 9, Fossil Lake region, Oregon, 10, 
 Dayton, Oregon; 11, Willamette Valley, Oregon; 12, Washtuckna Lake, Washington; 13, Klamath River; 14, Point 
 Conception, California. 
 
 Fig. 119.— Mylodon near harlani Owen. Left ramus of mandible, No. 3859 U. S. Nat. Mus. X 0.33. fin 1 , fin 2 , mental 
 foramina; 1, 2, 3, 4, inferior alveoli and teeth. Button Ranch, near Christmas Lake, Oregon. 
 
 Fig. 120.— Mylodon near harlani Owen. Incomplete left manus, No. 3859 U. S. Nat. Mus.; dorsal view. X 0.33. Button 
 Ranch, near Christmas Lake, Oregon. 
 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA, 
 
 WITH SPECIAL REFERENCE TO 
 
 THE PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE 
 
 OF RANCHO LA BREA. 
 
 RY 
 
 Chester Stock. 
 
 l 
 
INTRODUCTION. 
 
 A study of the gravigrade edentates from the Pleistocene asphalt deposits 
 of Rancho La Brea, near Los Angeles, California, was suggested to the writer, 
 then a student at the University of California, by Professor John C. Merriam. 
 The investigation, based originally on materials collected by the Department of 
 Palaeontology, University of California, was gradually enlarged to bring also under 
 consideration the vast collection of edentate remains in the Los Angeles Museum of 
 History, Science, and Art. The extensive series of specimens in the Los Angeles 
 Museum, the accessibility of the material for a detailed review, and the adequate 
 facilities for preparation of mounted individuals made it especially desirable to 
 direct attention to the Los Angeles collections. 
 
 While emphasis has been placed very largely on the collections in the Los 
 Angeles Museum of History, Science, and Art, an attempt has been made to 
 review the edentate assemblage from Rancho La Brea, as it is also represented in 
 the Museum of Paleontology, University of California. Fortunately, the Univer¬ 
 sity collection supplements those in southern California, particularly in offering 
 further specimens of the genus Megalonyx. As a result of the studies that have 
 been pursued, several preliminary papers, concerned for the most part with special 
 topics, have appeared between 1913 and 1920. 
 
 The large amount of material available from Rancho La Brea affords an 
 admirable basis for examination of ground-sloth remains occurring not only in 
 other Pleistocene deposits of Western North America, but in Tertiary beds as 
 well, and the present contribution endeavors to survey this portion of the field 
 and its bearing on the history of the gravigrade edentates. 
 
 The study of extinct North American edentates may perhaps be regarded as 
 having inaugurated palaeontologic research on this continent, for in 1797, in a 
 paper read before the American Philosophical Society, Thomas Jefferson described 
 some bones from a limestone cavern in western Virginia, as belonging to an animal 
 which he called Megalonyx. Since the pioneer investigations of Jefferson, pub¬ 
 lished in 1799, our fund of knowledge of these mammals has been greatly augmented. 
 The association of Jefferson’s name with a species of ground-sloth now widely 
 recognized in Pleistocene faunas and distinguished by possessing an extensive 
 geographic range on the North American continent, reaching from the Atlantic to 
 the Pacific Coast, is a humble but fitting tribute expressed by paleontology to one of 
 America’s greatest statesmen and scholars. 
 
 In broader palaeontological studies involving faunas of the two continents 
 of the western hemisphere, the order Edentata has long been recognized as one 
 which occupies an important position and comprises a number of unusual types. 
 In the New World the edentates have undergone the major course of then evolu¬ 
 tion during Tertiary time in South America, but are also intimately connected 
 with the Cenozoic history of North America. Highly characteristic members of 
 this group are living to-day in the Neotropical region. It is, however, during the 
 
 3 
 
4 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 Pleistocene that the edentates were well represented in the northern continent, 
 principally by the pilosic division of the Xenarthra, including those forms known 
 commonly as ground-sloths. The glyptodonts and the armadillos were relatively 
 less important as members in the Pleistocene fauna of North America, judging 
 from the infrequent record of their occurrence and their limited distribution, but 
 it is an interesting fact that the armadillos are now the only forms that have 
 survived in this region. 
 
 The great group of sloths, of which the living tree-sloths of the forested regions 
 of Central and South America are but the diminutive representatives, include 
 during the Pleistocene an array of forms. Structurally, the ground-sloths will 
 always remain among the most unique mammals that were evolved in the Tertiary 
 and Pleistocene of the western hemisphere. Especially to the student of mam¬ 
 malian evolution, as staged during the later Cenozoic in North America, will the 
 structural characters possessed by these forms appear in many instances anomalous, 
 pointing unmistakably to an earlier racial history in which biotic associations 
 and physical conditions were decidedly different from those encountered by the 
 indigenous land mammals of North America. 
 
 Within the group of ground-sloths represented in the American Pleistocene, 
 there are recognized three divisions, commonly regarded as of family rank. These 
 are (1) Megatheriidae, (2) Megalonychidae, and (3) Mylodontidae. In North 
 America the family Megatheriidae, of which the gigantic ground-sloth Megatherium 
 is typical, seems to be confined to the Pleistocene of the southeastern United States. 
 This genus is apparently absent in Pleistocene deposits of western North America, 
 and we are not particularly concerned with it in the present memoir. In western 
 deposits Nothrotherium and Megalonyx are the representatives of the megalonychid 
 ground-sloths, while Mylodon is the only known genus belonging to the Mylo¬ 
 dontidae. 
 
 ACKNOWLEDGMENTS. 
 
 My foremost acknowledgment and very grateful appreciation go to Professor 
 John C. Merriam. Whatever value may be ascribed to the present research rests 
 largely on his unstinted, ever friendly advice and his assistance given on many 
 occasions. 
 
 For the unexcelled opportunity to enlarge the observations on the edentate 
 remains from the asphalt beds by a study of the material in the Los Angeles Museum 
 of History, Science, and Art, and for the privilege to incorporate the results thus 
 obtained with those derived from an examination of the university collection, I 
 should like to express my deep obligation and sincere thanks in memory of the late 
 director of the museum, Dr. Frank S. Daggett. The keen interest shown by Dr. 
 Daggett in the progress of the research and the personal services rendered so 
 generously by him have done much in promoting the best results. Acting Director 
 Howard Robertson and the present director, Dr. Wm. Alanson Bryan, have 
 continued the enthusiastic support given by Dr. Daggett. For the liberality shown 
 in the furtherance of the work on the Rancho La Brea collections I wish to thank 
 also the officials of the county of Los Angeles. I desire also to acknowledge my 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. .) 
 
 great indebtedness to the University of California and to the Carnegie Institution 
 of Washington for support given in the study of the Los Angeles collections. 
 
 Other members of the museum staff in Los Angeles have rendered meritorious 
 service in the progress of the research. The assistance given by Mr. James W. 
 Lytle, assistant vertebrate palaeontologist, has been especially valuable, furnish¬ 
 ing important results. To Mr. Lytle has fallen the task of assembling the osteo- 
 logical materials for the mounted skeletons, a task which he has accomplished with 
 much proficiency. Mr. Lytle has also faithfully constructed the mounted specimens 
 of Notlirotherium and Mijlodon under my direction. Mr. L. E. Wyman furnished 
 valuable information regarding the occurrence of ground-sloth remains at Rancho 
 La Brea and a number of photographs. The restorations of Nothrotherium, and 
 Mylodon are by Mr. Charles R. Knight. 
 
 Most of plates and a number of the line drawings used as text-figures were 
 executed by Mr. John L. Ridgway, of the Carnegie Institution of Washington. 
 Many of the text-figures are the work of Mrs. Louise Nash and of Miss Frieda 
 Lueddemann. 
 
 In the course of the research on the edentates I have had opportunity to visit 
 a number of institutions in the East and to study collections. For many favors 
 bestowed during this visit I wish to extend my thanks particularly to Professor 
 Henry Fairfield Osborn, to Dr. W. D. Matthew, and to Mr. Walter Granger, of 
 the American Museum of Natural History; to Mr. C. W. Gilmore and Mr. J. W. 
 Gidley, of the U. S. National Museum; to Dr. W. J. Holland and Mr. 0. A. Peter¬ 
 son, of the Carnegie Museum; to Dr. Glover M. Allen, of the Harvard Museum of 
 Comparative Zoology; to Professor R. S. Lull, of Yale University; and to Professor 
 W. B. Scott and Dr. W. J. Sinclair, of Princeton University. Pleistocene ground- 
 sloth remains were kindly loaned for study by the American Museum of Natural 
 History and by the U. S. National Museum. Several loans of valuable specimens 
 from the Patagonian Miocene, now in the collections of Princeton University, 
 were generously granted by Dr. W. J. Sinclair. 
 
 Lastly, those friends of my shelf and study—the contributions of Richard 
 Owen, Joseph Leidy, Richard Lydekker, Herman Burmeister, Florentino Ameghino, 
 William B. Scott, Herluf Winge, and others—have laid the foundations upon 
 which it has been possible to build. 
 
REVIEW OF THE LITERATURE ON THE GRAVIGRADE EDENTATES. 
 
 MEGALONYCHINAE. 
 
 Following the early description of Megalonyx by Thomas Jefferson, further 
 remains of this animal were discovered which led to a somewhat better under¬ 
 standing of the form. The specimens collected in West Virginia, Tennessee, 
 Kentucky, Mississippi, and Alabama in the period following the publication of 
 Jefferson’s report and prior to 1855, and described or discussed by Cuvier, Owen, 
 Harlan, Jeffries Wyman, and William Cooper, formed the basis of an excellent 
 monograph by Joseph Leidy (1855). 
 
 While research on the North American edentates was in progress, the extinct 
 mammalian life of South America was also subjected to searching inquiry. The 
 vast stretches of the pampas of Argentina and the limestone caverns of Brazil 
 offered opportunities of collecting in Pleistocene deposits comparable to those in 
 North America. In the early part of the nineteenth century we note the investiga¬ 
 tions of the Danish naturalist P. W. Lund in the Brazilian caverns which gave the 
 first information regarding the very important Pleistocene fauna now known from 
 these deposits. Ground-sloth remains described by Lund (1836) from the cave 
 deposits were at first regarded as belonging to a small megathere. Lund later 
 recognized the distinctness of the form and established the genus Coelodon. This 
 is the first indication of a megalonychid ground-sloth later recognized in Pleistocene 
 faunas of North America. In the report on the ground-sloth remains collected by 
 Charles Darwin during the voyage of the Beagle, Owen (1840, p. 99, pi. 29) 
 described and figured a fragmentary mandible which was referred to Megalonyx. 
 This specimen was reviewed by Leidy, who correctly determined it as not pertaining 
 to the North American megalonychid genus. 
 
 In the great development of palaeontologic research in North America, 
 witnessed by the latter half of the nineteenth century, our knowledge of the Pleis¬ 
 tocene Megalonychinae was increased by descriptions of fragmentary materials 
 pertaining to species of Megalonyx other than M. jeffersonii and of specimens, in 
 part heretofore unknown, assigned to the latter form. These North American 
 remains were described principally from localities east of the Mississippi River. 
 Thus, Leidy (1860) published a description of additional skeletal elements giving 
 more complete information regarding the structure of the feet in Megalonyx. 
 
 E. D. Cope (1871) published a preliminary report of the Pleistocene fauna 
 from the famous Port Kennedy fissure of Pennsylvania, in which were described 
 several species of Megalonyx, based chiefly upon parts of the dentition. Twenty 
 years after the publication of Cope’s paper, E. W. Claypole (1891) reported upon 
 the occurrence of Megalonyx in Ohio and listed a number of skeletal structures 
 which were referred to M. jeffersonii. In the following year, J. M. Safford (1892) 
 announced the discovery of a fairly complete pelvis of Megalonyx and of other 
 
 bones in Big Bone Cave, Tennessee. No detailed description and comparison of 
 
 6 
 
 i 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 7 
 
 the pelvis were given by Safford. This specimen is now in the collections of 
 Vanderbilt University, Nashville, Tennessee. 
 
 The fourth annual report of the Texas Geological Survey contains a description 
 of fragmentary skull material and a single tooth referred by Cope (1892) to the 
 species Megalonyx leptostomusJ The specimens are from the Blanco Upper Pliocene 
 and occur in association with Pliohippus simplicidens, P. cumminsii, and the 
 carnivore Borophagus. This is the first record of a megalonychid ground-sloth in 
 North America from deposits older than Pleistocene. Whether the species belongs 
 to Megalonyx is perhaps still a matter of conjecture. 
 
 Lindahl^ (1893) published the description of the species Megalonyx leidyi 
 based upon a well-preserved skull without mandible from the Pleistocene of Kansas. 
 In the final report of the Pleistocene fauna from the Port Kennedy fissure by Cope 
 (1899)* published after his death, the four species Megalonyx wheatleyi,' M. loxodo?i, 
 M. tortulusj and M. scalper were recognized as valid forms. Megalonyx splienodon 
 (also established in the earlier paper on the Port Kennedy fauna) was no longer 
 considered specifically separable from M. wheatleyi, and M. dissimilis (established 
 previously by Leidy) was regarded as synonymous with M . jeffersonii. 
 
 During the latter half of the nineteenth century a number of paperr appeared 
 which treated of the systematic position, structural characters, and relationships 
 of the South American genus Coelodon (Nothr otherium ). It is, howeves, particu¬ 
 larly from the excellent account of Coelodon by Reinhardt (1878), based on materials 
 of a young individual previously collected by Lund, that we secure considerable 
 information regarding this genus. Reinhardt clearly indicated for the first time 
 the close agreement that exists between Megalonyx and Coelodon in many charac¬ 
 ters. In 1889, Lydekker called attention to the preoccupation of the generic name 
 Coelodon and proposed to replace the name used by Lund for this ground-sloth by 
 Nothrotherium. 
 
 Several years after the publication of Reinhardt’s paper, Burmeister (1882) 
 described the form Nothr opus prisons from a fragmentary right ramus of the 
 mandible found presumably in Pleistocene deposits along the Rio Carcaranal, i-n 
 the province of Santa Fe, Argentina. The specimen was compared principally 
 with the living tree-sloths by Burmeister. So far as comparison can be made, 
 Nothr opus agrees with Nothr otherium in shape of inferior cheek-teeth, but is less 
 specialized in the retention of the first or foremost tooth as indicated by the presence 
 of an alveolus. F. Ameghino considers Nothr opus prisons as directly ancestral to 
 N othr otherium and as occurring at a lower stratigraphic horizon than the latter. 
 
 The monumental work of Sco tt (1903-4) on the edentates of the Upper 
 Miocene Santa Cruz deposits of Patagonia gave, for the first time, to students of 
 Tertiary and Pleistocene mammalogy very full information relating to these types. 
 In the family Megalonychidae previously established generic and specific types 
 were reviewed and new forms were described. Among the latter was the genus 
 Meg along chother ium, considered to be related to Megalonyx in the same way that 
 Hapalops\ also from Santa Cruz deposits, is related to Nothr otherium. Hapalops 
 is represented by a number of species, but the most complete material described 
 
8 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 by Scott was grouped under H. longiceps and H. elongatus. Frequent comparisons 
 of structure were made with the Pleistocene Megalonyx and Notlirotherium. In the 
 present study of the ground-sloths from Rancho La Brea this contribution to the 
 history of the American edentates by Scott has been the source of much valuable 
 data. 
 
 It was not until the early years of the present century, when the limestone 
 caverns of northern California were systematically explored by the Department 
 of Palaeontology, University of California, and the Pleistocene faunas from the 
 deposits were studied in detail, that Nothrotherium was tentatively recorded by 
 Sinclair (1905) as a constituent of the North American Pleistocene fauna as well 
 as of the South American assemblage. It was recognized in Potter Creek Cave, 
 Shasta County, in association with remains of Megalonyx. In the same paper 
 Sinclair described the species Megalonyx sierrensis, based upon a lower jaw and 
 skeletal material secured from the Pleistocene of Mercer’s Cave, near Murphys, 
 Calaveras County, California. Ground-sloth remains similar to those described 
 by Sinclair from Potter Creek Cave were collected by E. L. Furlong in Samwel 
 Cave, Shasta County. The specimens were tentatively regarded by Furlong 
 (1906) as belonging to Megalonyx. In the following year appeared a valuable 
 paper by Florentino Ameghino (1907), in which mammalian remains were described 
 from the limestone grotto of Iporanga in the State of Sao Paula, Brazil. Ameghino’s 
 report is devoted in part to the description of a damaged skull and dentition 
 of N othr other ium maquinense, and is followed by a discussion of the relationship 
 and parentage of the genus. A bibliography on N othr other ium accompanies the 
 paper. 
 
 Nothr other ium was definitely recognized by Stock (1913) in the Rancho La 
 Brea fauna, where it is found associated with Megalonyx. Since then the presence 
 of the genus has been established in the Pleistocene fauna of Samwel Cave and in 
 that of Hawver Cave (Stock, 1918) in Eldorado County, California, occurring 
 always with Megalonyx. The presence of N othr other ium in Pleistocene fissure depos¬ 
 its of California is therefore directly comparable to its occurrence in the Brazilian 
 caverns. Aside from the localities in California, the genus is now known to have 
 lived in Texas® during the Pleistocene, for a damaged skull of N othr other ium reported 
 to have come from “a well, at a depth of 40 feet, in Wheeler County,” was 
 described by 0. P. Hay (1916). 
 
 A year previous to the appearance of Hay’s paper there was published the 
 important report by the veteran zoologist Herluf Winge (1915). In this volume 
 Winge surveys all the edentate materials collected by Lund in the limestone caverns 
 of Lagoa Santa, Minas Geraes, Brazil, and now in the collections of the Zoological 
 Museum at the University of Copenhagen. Among the illustrations should be 
 noted the first views that have been made available of the greater portions of the 
 feet of Nothrotherium. 
 
 “In a discussion of the quicksilver deposits of the Terlingua district, Brewster County, Texas, H. W. Turner makes 
 mention (Economic Geology, vol. 1, p. 275, 1906) of the finding of ground-sloth remains in a fissure deposit. The material 
 was submitted to Professor J. C. Merriam, who identified several teeth as belonging to Nothrotherium. The specimens were 
 placed in the Museum of the California Academy of Sciences in San Francisco, where they were destroyed by the fire of 
 April 18, 1906. 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 9 
 
 Iii recent surveys of Pleistocene faunas of North America by O. P. Hay a an 
 attempt is made to determine the position of Meg atony x in the Pleistocene with 
 reference to the stages of advance and retreat of the continental ice-sheets during 
 that period. The charts employed by Hay to illustrate the geographic distribution 
 of Megalonyx clearly demonstrate the extensive range of this genus. With the 
 exception of the occurrences in California, the megalonychid forms are at present 
 rarely noted in Pleistocene deposits located in the region west of the Wasatch. 
 Recently Merriam and Buwalda 6 reported the presence of a species of Megalonyx 
 in a fauna from the White Bluffs exposed along the Columbia River in south- 
 central Washington. 
 
 One of the very interesting and important problems in the study of the mega¬ 
 lonychid ground-sloths arises from the discovery of remains of these forms in 
 Pleistocene cave, fissure, and spring deposits on islands of the Greater Antilles. 
 The presence of Megalocnus in the Pleistocene of Cuba was early recognized by 
 De Castro. The genus was described by Leidy (1868), who regarded it as closely 
 related to the North American Megalonyx. Anthony (1918) has described the 
 megalonychid genus Acratocnus from remains found in deposits on the island of 
 Porto Rico. Dr. W. D. Matthew (1919) has recently reviewed the discoveries of 
 extinct mammals in the West Indies and has added three distinct megalonychid 
 genera to the Pleistocene assemblage known from Cuba. 
 
 Wilhelm Freudenberg c has described from the Pleistocene of Mexico the 
 species Nothrotherium mexicanum, based on a fragmentary ramus of the mandible 
 of a young individual. 
 
 Recent descriptions of fragmentary remains of megalonychid ground-sloths 
 from southern California by Childs Frick d should be cited, for they bring to view 
 forms from two distinct Pliocene horizons. Information relating to the Tertiary 
 forms of North America is much desired, for the types are still known only by 
 very incomplete specimens. 
 
 MYLODONTIDAE. 
 
 The earliest descriptions of mylodont ground-sloth remains from North America 
 are those published between 1831 and 1835 by Richard Harlan (1835) and based 
 upon incomplete materials, including a fragmentary right ramus of the mandible. 
 In referring them to Megalonyx laqueatus, Harlan failed, however, to recognize the 
 generic characteristics of the form. The incomplete mandible described by him 
 is regarded as having been found at Big Bone Lick, Kentucky. 
 
 The recognition of mylodont ground-sloths in the past mammalian life of 
 America occurred later than that of Megatherium, for it was not until the epoch- 
 making voyage of the Beagle that a sufficient collection was secured by Charles 
 Darwin in South America to establish the group as distinct from other ground- 
 
 ° O. P. Hay, The recognition of Pleistocene faunas, Smithson. Misc. Coll., vol. 59, No. 20, 1G pp., 1912. 
 
 h J. C. Merriam and J. P. Buwalda, Age of strata referred to the Ellensburg formation in the White Bluffs of the 
 Columbia River. Univ. Calif. Pubh, Bull. Dept. Geol., vol. 10, pp. 255-266, plate 13, 1917. 
 
 c Wilhelm Freudenberg, Geologic von Mexiko dargestellt nach der Literatur und nach eigenen Forschungen, Born- 
 traeger, Berlin, pp. 140-141, and accompanying plate, fig. 3, 1921. 
 
 d Childs Frick. Extinct vertebrate faunas of the badlands of Bautista Creek and San Timoteo Canon, Southern 
 California. Univ. Calif. Publ., Bull. Dept. Geol., vol. 12, pp. 320-321, 349-350, 1921. 
 
10 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 sloths. Sir Richard Owen (1840), in the report on the fossil mammalian remains 
 collected by Darwin, described three genera. It is in this report that Owen also 
 recognized the mylodont characters exhibited in the lower jaw described by Harlan 
 and designated the specimen as the type of the species Mylodon harlani. 
 
 In 1841, the Royal College of Surgeons of London acquired a skeleton of a 
 mylodont sloth collected in Pleistocene beds 7 leagues north of Buenos Aires. The 
 specimen formed the basis of a profound study by Owen (1842), whose memoir on 
 Mylodon robustus remains to-day a source of very complete information regarding 
 the osteology of the genus. Owen’s statement of the probable habits of Mylodon 
 is a splendid example of astute deductive reasoning from materials available to a 
 palaeontologist. 
 
 The earliest record of remains of mylodont ground-sloths in western North 
 America is that by Perkins (1842), based on specimens found at a depth of 12 feet 
 in the Willamette Valley, in latitude 44° north, western Oregon. The tooth and 
 humerus described by Perkins were mentioned by Owen in a letter to the American 
 Journal of Science, dated 1842, in which he referred the specimens to Mylodon 
 harlani. The remains of Mylodon recovered in Benton County, Missouri, and 
 described by Harlan (1843) under the name Orycterotherium missouriensis, were 
 also considered by Owen as belonging to Mylodon harlani. 
 
 Leidy (1855, pp. 47-49, plate 16), in the valuable memoir on the. Pleistocene 
 sloth tribe, described and figured the type specimen of Mylodon harlani and reviewed 
 several other collections of mylodont remains. In addition to the occurrences at 
 the type locality and in Oregon and Missouri, there were noted those at Ashley 
 River, South Carolina, Mammoth Ravine, Mississippi, and at Natchez, Mississippi. 
 
 In the period of 10 or 15 years following the publication of Leidy’s memoir, 
 little if any information regarding the North American Mylodontidae was added 
 to the knowledge already in the possession of the palaeontologist. Leidy (1870) 
 indicated the presence of a mylodont ground-sloth in Central America. Cope 
 (1871), in the preliminary report on the Pleistocene vertebrates from the Port 
 Kennedy fissure, recorded questionably the occurrence of Mijlodon. Seven years 
 later Cope (1878), on the basis of fragmentary specimens, established the species 
 Mylodon sodalis, stating that the material came from the Pliocene of Oregon. 
 The form is a member of the important Pleistocene assemblage known from the 
 Fossil Lake beds of the Great Basin region. Cope (1880) also mentions the occur¬ 
 rence of Mylodon in auriferous gravels of the Klamath River, near Yreka, Cali¬ 
 fornia. In the early eighties we note the first statement (Le Conte, 1882, 1883) 
 regarding the “human” footprints in the Pleistocene beds near Carson City, 
 Nevada, in which their origin is attributed to the ground-sloth. In 1889, Leidy 
 recorded Mylodon in the Pleistocene of Louisiana, describing specimens from the 
 salt mine of Petite Anse. Cope (1889) reviewed the edentates of North America 
 and in the nineties he made shorter contributions to the Mylodontidae (Cope, 
 1893, 1895, 1899) from specimens collected on the Staked Plains of Texas, from 
 Petite Anse, and from the Port Kennedy fissure. S. W. Williston (1895) also 
 described a specimen from the Pleistocene of Kansas. 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 11 
 
 Since mylodont ground-sloths were native to South America and occur in 
 Pleistocene beds much more frequently than in deposits of similar age in North 
 America, it is but natural to find a greater number of papers dealing with these 
 forms. In the latter half of the nineteenth century appeared first the description 
 of Scelidotherium leptocephalum by Owen (1857). This important report is followed 
 by Gervais’s account (1860) of the skeletal structures of Lestodon armatus. Several 
 years later, Gervais (1869) and Pouchet (1868-69) discussed the endocranial casts 
 of Mylodon and of allied forms. In the seventies appeared two noteworthy con¬ 
 tributions by Reinhardt (1875, 1879), in which were given the structural characters 
 of Lestodon armatus and of Grypotherium darwinii. In the period 1864 to 1881 a 
 number of valuable additions to our knowledge of the South American gravigrade 
 edentates were made by Herman Burmeister. Special notice should be given to 
 Burmeister’s papers of 1879 and 1881, wherein important data are recorded on 
 the osteology and probable habits of the mylodont ground-sloths, and a statement 
 of the types known from South America. Lydekker (1886) reviewed the characters 
 of Scelidotherium leptocephalum and described two other species of this genus. In 
 the early eighties a number of genera and species of mylodont ground-sloths were 
 described by Florentino Ameghino, but many of these have since been referred to 
 previously established forms. In a later work, Ameghino (1889) gathered together 
 information of known types and added descriptions of others. In the valuable 
 and very useful repprt on the extinct edentates of Argentina, Lydekker (1894) 
 recognizes, as valid genera in the group of mylodont ground-sloths, Mylodon, 
 Glossotherium, Scelidotherium, and Catonyx. The fine plates published with this 
 memoir illustrate not only the skull, but also the mounted skeletons of mylodont 
 genera. Lydekker critically examined the basis for establishing genera and species 
 within the group and pointed out the great amount of variation that exists in the 
 skull and dentition. Many previously described genera and species were regarded 
 as invalid types. 
 
 The closing years of the nineteenth century and the opening years of the 
 twentieth witnessed the discovery and announcement of the remarkably preserved 
 mylodont ground-sloth remains in a large cavern near Consuelo Cove, Last Hope 
 Inlet, Patagonia. Most interesting specimens of not only skeletal structures, but 
 of the horny claws and skin as well, were collected by the explorer Dr. Otto 
 Nordenskjold in 1896 and by Dr. F. P. Moreno in 1897. The ground-sloth materials 
 from the Patagonian cave were described in detail by Lonnberg (1899), Smith 
 Woodward (1899, 1900), and by Roth and Lehman-Nitsche (1901). Erland 
 Nordenskiold (1900) also described remains of the ground-sloth and of associated 
 forms from this locality. Interesting results of studies of the skin and hair of the 
 Patagonian ground-sloth were published by Ridewood (1901) and by Jacob (1902). 
 The discovery of specimens showing a remarkable state of preservation and yet 
 belonging to types characteristic of the Pleistocene of the New World aroused 
 scientific discussion, as well as much popular interest. The presence of ground- 
 sloths in the Recent fauna of the unexplored regions of South America was regarded 
 as so probable that an expedition was actually sent from England to hunt these 
 
12 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 animals. Failing in its primary purpose, the expedition furnished opportunity to 
 make further observations on Patagonia and its inhabitants. 0 
 
 Noteworthy remains of mylodont ground-sloths were described by Gervais 
 (1873) and by Burmeister (1879, 1881 6) from the valley of the Rio de la Tarija, 
 Bolivia, and additional observations on specimens from this locality were made 
 by Nordenskiold (1901) and by Ameghino (1902). Philippi had directed attention 
 to the mylodont forms occurring at Ulloma, Bolivia. Later, Sefve (1915) described 
 skulls of Scelidotherium collected at Tarija and Ulloma. In the memoir by Scott 
 (1903-4), already referred to, we find (under the discussion of the Mylodontidae 
 from the Santa Cruz, Upper Miocene beds of Patagonia) the recognition of the 
 two genera Analcitherium and Nematherium. The small representation of the 
 Mylodontidae in the Santa Cruz stands in noticeable contrast to that of the 
 Megalonychinae. Interesting comments on the family as developed at this stage 
 of geologic time are made by Scott. Rautenburg (1906) discussed the structure 
 of the skull and of the skeletal elements of Pseudolestodon ( Mylodon?) hexaspondylus. 
 
 Meanwhile, our knowledge of the Mylodontidae in North America was 
 increased much more slowly, chiefly because the occurrences were infrequent and 
 the collections were oftentimes in a poor state of preservation. J. C. Merriam 
 (1899, 1900) discussed the presence of ground-sloths in the Pleistocene of Cali¬ 
 fornia. These papers were followed by Brown’s description (Brown, 1903) of the 
 skull and skeletal structures of a ground-sloth which he described as Paramylodon 
 nebrascensis. The specimens were obtained in the Pleistocene of Hay Springs, 
 Nebraska. Three years later, Merriam (1906) announced the discovery of Pleis¬ 
 tocene mammals in the now famous Rancho La Brea asphalt beds of southern 
 California. He indicated in this paper and in a later publication (Merriam, 1908) 
 the occurrence of mylodont ground-sloths. Cockerell (1909) described, from the 
 Pleistocene near Walsenburg, Colorado, a large skull of Mylodon. Dermal ossicles 
 of the Rancho La Brea species were described by W. J. Sinclair (1910). W. Freuden- 
 berg/' in 1910, figured a tibia from Mexico and listed the genus represented by this 
 element as Orycterotherium. Important information on the North American Mylo¬ 
 dontidae is furnished by Allen’s memoir (G. M. Allen, 1913), based on collections 
 made in 1880 by Samuel Garman in the Pleistocene deposits of the Niobrara River, 
 Nebraska. This type was regarded by Allen as distinct from known North American 
 forms and was described as Mylodon gcirmani. 
 
 The important Pleistocene mammalian assemblage known from Rock Creek, 
 Briscoe County, Texas, includes also Mylodon harlani, remains of which were 
 described by Lull (1915). In the year previous to the publication of Lull’s paper, 
 Stock (1914) reported on the skull and dentition of the mylodont sloths of Rancho 
 La Brea, and other papers appeared at a later date (Stock, 19176, 1917c, 1920). 
 
 Within recent years one of the important references to the mylodont ground- 
 sloths of South America is that by Rovereto (1914), who indicated the types 
 
 ° H. H. Prichard. Through the heart of Patagonia. D. Appleton & Co., N. Y., 346 pp., 1902. 
 
 6 W. Freudenberg, Die saugetierfauna des Pliocans und Postpliocans von Mexiko, I. Carnivoren. Goel. u. Palae 
 Abhandl., vol. 13, pi. 20, 1910. 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 13 
 
 occurring in later Tertiary horizons of the southern continent. In the comprehen¬ 
 sive account of the edentates from the Pleistocene cave deposits of Brazil, Winge 
 (1915) discusses the skull and parts of the skeleton of Mylodon, Catonyx, and 
 Scelidotherium. Recent important papers by Kraglievich (1921, 1922) deal with 
 the South American Mylodontidae. 
 
 PERIOD 
 
 MEGALONYCHINAE 
 
 NORTH 
 
 AMERICA 
 
 SOUTH 
 
 AMERICA 
 
 MYLODONTIDAE 
 
 NORTH 
 
 AMERICA 
 
 SOUTH 
 
 AMERICA 
 
 Recent 
 
 Glosso thenum 
 
 Megalonyx 
 
 Mylodon 
 
 Nothrotherluw 
 
 Pleistocene 
 
 Nothropus 
 
 Mylodon 
 
 Lestodon 
 
 Catonyx 
 
 Seelidoth’m 
 
 Pliocene 
 
 Pronothroth’m 
 
 Glossotherium 
 
 Scelidoth’m 
 
 Eumylodon 
 
 Pleurolestodon 
 
 N eoneaia thermal 
 
 Miocene 
 
 Hapalops 
 Eucholoeops 
 Hyperleptus 
 Pelecyodon 
 Sehismoth’m 
 Analcimorphus 
 1 Megalonychoth’m 
 
 Xematherium 
 
 Aaalcitlierium 
 
 Hapalops 
 
 Oligocene 
 
 Eocene 
 
 Fk;. 1.—Chart showing geologic and geographic distribution of the Megalonychinae and Mylodontidae. 
 
GEOLOGIC RANGE AND GEOGRAPHIC DISTRIBUTION OF THE 
 
 MEGALONYCHINAE. 
 
 In contrast with the Mylodontidae and the Megatheriinae, the subfamily 
 Megalonychinae is known by more abundant materials from Tertiary deposits of 
 America, and of particular interest is the occurrence of members of this group to 
 the apparent exclusion of the two other families in Pliocene deposits of the northern 
 continent. 
 
 The megalonychid ground-sloths are definitely recorded from the Deseado 
 Oligocene of South America, but it is quite unusual that they have here but a 
 sparse representation, a fact which appears to be true for the entire edentate 
 group at this period in the known Tertiary of the southern continent. Ground- 
 sloths occurring in the Deseado formation have been referred by Ameghino to 
 the genus Hapalops, a form found commonly in the following period. Loomis 
 (1914) reaches the conclusion, from an analysis of the Deseado fauna, that the 
 small percentage of edentates present in this assemblage of mammals is not to 
 be attributed to an initial appearance of the group, for they are apparently close 
 in stage of evolution to types of the Miocene and probably indicate in their structure 
 an extended history antecedent to the Deseado accumulation. Loomis believes 
 that the explanation for their meager representation should be sought rather in 
 environmental factors less favorable than those extant during the deposition of 
 the Santa Cruz beds. 
 
 The megalonychid ground-sloths were by far the most abundant forms of gravi- 
 grade edentates in the Santa Cruz mammalian assemblage of Miocene age. It is 
 in this group that we secure the first adequate evidence as to the progenitors of Pleis¬ 
 tocene types. According to Scott (1903-4), at least 7 genera are clearly distinguish¬ 
 able, this number being in decided contrast to the two incompletely known mylodont 
 forms from these beds. The generic names are: Hapalops , Hyperleptus, Eucho- 
 loeops, Megalonychotherium, Analcimorphus, Schismotherium, Pelecyodon. 
 
 Hapalops is known by very complete and varied palaeontologic materials. 
 Worth emphasizing is the great development of megalonychid forms during the 
 Miocene, just antedating the earliest appearance of these ground-sloths in North 
 America. It suggests that we should look to the Megalonychinae and not to the 
 Mylodontidae or to the Megatheriinae as a probable source from which have 
 been derived the gravigrade edentates appearing in the later Tertiary of the 
 northern continent. The presence of megalonychid ground-sloths in Cenozoic beds 
 in North America older than the Pleistocene or late Pliocene lends considerable 
 support to this suggestion. Thus we may regard these early forms as the harbingers 
 of the many ground-sloths that found their way northward during the Pleistocene. 
 As suggested by Matthew (1915, p. 261), a possibility remains that the later 
 Tertiary forms of North America are survivors from a primitive Paleocene-Eocene 
 stock lingering in this region while the major evolution of the Xenarthra was taking 
 place in the southern continent. It should be noted, furthermore, that, if this is 
 actually the case, the later Tertiary ground-sloths of the northern continent would 
 not show close relationship to either the middle Tertiary forms of South America 
 
 14 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 15 
 
 or to those types reaching North America during the Pleistocene. Scott (1904, 
 p. 361) has already pointed out that the Upper Miocene Santa Cruz forms of 
 Patagonia are less like the early Tertiary Ganodonta than are the Pleistocene 
 genera, and are closely related to the latter types. 
 
 No remains of mylodont ground-sloths have been recorded from deposits 
 older than Pleistocene in North America. In contrast to this record, several occur¬ 
 rences of megalonychid forms in Pliocene deposits are known, but unfortunately the 
 remains are still very incomplete and fragmentary. Future search in the Lower 
 and Middle Pliocene of North America will probably yield much better preserved 
 material, thereby adding a very important chapter to the history of the group. 
 
 The presence of edentate-like remains in Middle Tertiary deposits of North 
 America was announced by Dr. W. J. Sinclair (1906). The specimen described 
 by Sinclair is a fragmentary ungual phalanx said to have been collected in the 
 Middle Miocene Mascall deposits of eastern Oregon. Since the publication of 
 this description doubt has been expressed as to the exact horizon from which the 
 specimen was obtained. J. C. Merriam and W. D. Matthew (1915, p. 261) are 
 inclined to the view that the ungual phalanx came originally from the Rattlesnake 
 formation (Lower Pliocene). It is believed that in the course of erosion, now 
 actively removing materials in the John Day region, the specimen was washed from 
 Rattlesnake deposits and came to rest upon an exposed surface of the underlying 
 Mascall beds. Ameghino (1912, p. 55) unhesitatingly accepted the occurrence of 
 gravigrade edentates in the Mascall and regarded the fragmentary claw described 
 by Sinclair as the type of the genus and species, Sinclairia oregoniana. On the 
 basis of this occurrence, Ameghino postulated a land connection between North 
 America and South America during Upper Miocene time. Reference to the ground- 
 sloth specimen from the Rattlesnake beds will be made again in another section. 
 
 Megalonyx and Megatherium have been recorded from the Alachua formation 
 in Florida. The two genera are known also from the Pleistocene of that State. 
 In view of the fact that other ground-sloth remains in Lower Pliocene deposits 
 of North America represent types with characters somewhat different from those 
 of the Pleistocene genera, it appears rather doubtful whether the two forms men¬ 
 tioned actually ranged from the Pleistocene into the Lower Pliocene. Furthermore, 
 Sellards 0 has remarked: 
 
 “The vertebrate fauna of the Dunnellon formation and the Alachua Clays is practically the same, including 
 the mastodon, rhinoceros, camel, and early horse. With these late Miocene or early Pliocene forms are found 
 Odocoileus, Tapirus, Megatherium, and other late Pliocene or Pleistocene types. The mixing of fossils results 
 unavoidably from tfth manner of accumulation of the deposits, for while the formation is believed to have 
 accumulated chiefly during the Pliocene, yet sink formation having continued, later forms were carried into the 
 deposits and mixed with the earlier.” 
 
 Leaving aside the doubtful occurrences of ground-sloths in the Mascall deposits 
 and in the Alachua Clays, there still remains the presence of possible megalonychids 
 in the Snake Creek beds of Nebraska, in the Rattlesnake beds of eastern Oregon, 
 and in the Orinda of Middle California. Ground-sloths definitely related to the 
 Megalonychinae are known also from the Eden and San Timoteo horizons of 
 
 ° E. H. Sellards, The relation between the Dunnellon formation and the Alachua clays of Florida, 6th Ann. Rpt. Fla. 
 State Geol. Surv., p. 162, 1914. 
 
16 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 Southern California and from the Blanco of Texas. The authentic records of 
 megalonychid ground-sloths in the Pliocene of North America seem, then, to 
 favor the view that these forms are derivatives from a South American center of 
 dispersal. If we assume this to be true, we are led to believe that at least two 
 incursions from South America have taken place during the latter half of the 
 Cenozoic, one during the early Pliocene or just prior to this period, the other in 
 the Pleistocene. 
 
 Members of the Megalonychinae from Pliocene deposits of Mexico, should 
 they be found in beds of that age, would undoubtedly aid in establishing more 
 definitely the relationship of the North American and South American ground- 
 sloths of the later Tertiary. 
 
 Table 1 . —Known occurrences of later Tertiary ground-sloths of North America. 
 
 Form. 
 
 Material. 
 
 Deposit. 
 
 Age. 
 
 Edentate-like. 
 
 Ungual phalanx. 
 
 Mascall(?).... 
 Alachua. 
 
 Middle Miocene. 
 
 Lower Pliocene. 
 
 Lower Pliocene. 
 
 Lower Pliocene. 
 
 Lower Pliocene. 
 
 Lower Pliocene. 
 
 Lower Pliocene. 
 
 Lower Pliocene. 
 
 Lower Upper Pliocene. 
 Upper Pliocene. 
 
 Megatherium. 
 
 
 Megalonyx sp. 
 
 Toe-bone. 
 
 Alachua. 
 
 Megalonychid (?). 
 
 Claw, navicular. 
 
 Snake Creek. . 
 Rattlesnake... 
 Orinda. 
 
 Megalonychid (?). 
 
 Metapodial. 
 
 Megalonychid (?). 
 
 Tooth. 
 
 Nothrotherium? or Pronothrotherium? sp. . . . 
 Megalonyx sp. 
 
 Tooth. 
 
 Eden. 
 
 Tooth. 
 
 Eden. 
 
 Megalonyx? sp. 
 
 Teeth. 
 
 San Timoteo.. 
 Blanco. 
 
 Megalonyx leptostomus Cope. 
 
 Skull fragments and tooth. . 
 
 
 
 In the Pliocene of South America the record of the Megalonychinae is still 
 fragmentary, the material described at present being much less abundant than 
 that of the Santa Cruz stage. Rovereto (1914) has reviewed the occurrence of 
 mammals in various deposits of the later Tertiary, and from this study one may 
 infer that these ground-sloths were not so well represented in the Pliocene as 
 were members of the Mylodontidae. They have been obtained only from one 
 horizon, namely, the Araucanian, at Catamarca, northwestern Argentina. The 
 types found are Pronothrotherium typicum Ameghino and Pyramiodontherium 
 dubium described by Rovereto. The former genus is currently regarded as occupy¬ 
 ing a position in the phylogenetic series between Hapalops of the Miocene and 
 Ndthrotherium of the Pleistocene. Whether or not this form is related to any of 
 the Pliocene megalonychid ground-sloths of North America is still a matter of 
 conjecture, but future discovery may demonstrate such a relationship. 
 
 During the Pleistocene the North and South American members of the Mega¬ 
 lonychinae show interesting resemblances and differences. The genus Nothrothe- 
 rium is now known to have ranged from Brazil northward into Texas and California. 
 It has been found in cave deposits in the Brazilian region and in similar accumula¬ 
 tions in the mountain provinces of California. In western North America Nothro- 
 therium seems to have been especially characteristic of the more elevated forested 
 regions, but such a range did not necessarily preclude its presence in particularly 
 favored localities, as at Rancho La Brea. The distribution of the genus Megalonyx 
 stands in contrast to that of Nothrotherium. This form is found mainly in asso- 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 17 
 
 ciation with forest faunas of the Pleistocene and seems to have been indigenous 
 to the North American continent, where it has undergone considerable specific 
 differentiation. The distribution of Megalonyx is rather unusual, for although the 
 genus enjoys as extensive a range in North America as Mylodon and exceeds 
 Nothrotherium in this respect, it still remains completely unknown in South American 
 faunas. In the southern continent, during the Pleistocene, other megalonychid 
 ground-sloths appear, as shown by the genus N othr opus and by the unidentified form 
 recorded from Peru by Nordenskiold (1908). The distribution of Megalonyx should 
 be credited with due consideration, for it may indicate that the genus has evolved 
 on the North American continent from some Pliocene descendant of the Santa Cruz 
 Megalonychinae and has resided here for a longer period than Nothrotherium. 
 
 Attention should now be directed to the presence of very interesting types 
 of ground-sloths in Cenozoic deposits of the Greater Antilles. Remains of these 
 mammals from the island of Cuba were recognized by Leidy as belonging to a form 
 closely related to the North American Megalonyx and were referred to under the 
 generic name Megalocnus. Several years ago, Anthony (1918) described another 
 tj'pe, Acratocnus, from the island of Porto Rico. Recently, Matthew (1919) 
 has added the three genera Mesocnus, Miocnus, and Microcnus to the mammalian 
 assemblage known from cave, fissure, and spring deposits of Cuba. 
 
 All the ground-sloths from the Greater Antilles are of small size and are 
 undoubtedly related to the still smaller forms from the Santa Cruz Miocene of 
 South America. To account for their presence on the islands of the West Indies, 
 we are led to postulate either a migration of types during the later Tertiary from 
 the continental areas to the south or west or a conveyance of these forms on natural 
 rafts, as suggested by Matthew. Geological and faunal evidence gives no support 
 to the possibility of a Tertiary land connection between the Antillean land-mass 
 and North America. The Pleistocene ground-sloths found in the Greater Antilles 
 must then be regarded as representatives of culminating stages in a development 
 under the influence of conditions of insular isolation operating for a long period 
 of time, as suggested by Matthew (1915, p. 204). 
 
 When due allowance is made for the structural features acquired as a result 
 of isolation, Megalocnus, Acratocnus, and possibly the three genera described by 
 Matthew possess certain characters in common and form a group which should be 
 distinguished from Nothr other ium and related types. This evidence may be cited, 
 perhaps, in support of a belief that Megalonyx, Acratocnus, Megalocnus, and 
 other Cuban ground-sloths are Pleistocene derivatives from a common ancestry, 
 some members of which during the Pliocene invaded North America from the 
 southern continent by way of Central America, while others found their way 
 into the Antillean land-mass. Thus would be explained incidentally the absence 
 of Megalonyx in the Pleistocene of South America, a fact which stands in decided 
 contrast to the range of N othr otherium. With an earlier entrance into North America 
 Megalonyx became better adjusted to the northern biotic and physical environments 
 than did N othr other ium, thereby accentuating their structural differences and per¬ 
 mitting greater specific differentiation within the former genus. 
 
OCCURRENCE AND DESCRIPTION OF REMAINS PROBABLY 
 BELONGING TO MEMBERS OF THE MEGALONYCHINAE IN 
 TERTIARY BEDS OF WESTERN NORTH AMERICA. 
 
 RATTLESNAKE FORMATION, EASTERN OREGON. 
 
 Mention has been made of the edentate-like ungual phalanx described by 
 Sinclair (1906) from the Mascall Middle Miocene of the John Day region, eastern 
 Oregon. The specimen (No. 1096 U. C. C.) is shown in figure 2, a, b, c. Much 
 additional information of the later Tertiary beds and faunas of the John Day 
 region is now available. It is generally regarded that the ungual phalanx came 
 originally from the Rattlesnake Lower Pliocene deposits that overlie unconformably 
 the Mascall beds. The mammalian fauna known from the Rattlesnake 0 is listed 
 as follows: 
 
 ?Canis davisi Merriam. 
 
 Amphicyon, near amnicola Matthew and Cook. 
 Mustela sp. a. 
 
 Mustela sp. b. 
 
 Felis, large sp. 
 
 Indarctos? oregonensis Merriam, Stock, and Moody. 
 Lepus? sp. 
 
 Dipoides? sp. 
 
 Megalonychid? gen. and sp. indet. 
 
 Prosthennops sp. 
 
 Alticamelus altus (Marsh). 
 
 Pliauchenia sp. 
 
 Procamelus sp. 
 
 Sphenophalos, near nevadanus Merriam. 
 Ilingoceros? sp. 
 
 Teleoceras fossiger (Cope). 
 
 Hipparion sinclairii Wortman. 
 
 Hipparion, near occidentale Leidy. 
 Hipparion, near anthonyi Merriam. 
 Pliohippus, near fairbanksi Merriam. 
 Pliohippus spectans (Cope). 
 Tetrabelodon? sp. 
 
 One of the results of the University of California expedition to the John Day 
 region in 1916 was the discovery of a single fragmentary metapodial (No. 22900 
 U. C. C.) of a gravigrade edentate in deposits of Rattlesnake age. Unfortunately, 
 the specimen has suffered injury during preservation and contributes little to our 
 knowledge of the skeleton of North American ground-sloths of the later Tertiary. 
 It is important, however, in establishing definitely the occurrence of these forms 
 in the Lower Pliocene of eastern Oregon. 
 
 In the metapodial (No. 22900 U. C. C., figure 2, f, g,) from the Rattlesnake 
 beds, one of the sides and the greater part of the proximal end have been broken 
 away. On the side that is best preserved the facet for the adjoining metapodial 
 is intact, while only a small portion of the mesopodial surface remains. At the 
 distal end the carina is preserved, as well as an articulating offset along one of 
 the sides. 
 
 No. 22900 is short and stout. The lateral wall of the shaft is at right angles 
 to that portion of the dorsal surface which still remains. It indicates that the 
 metapodial possessed a relatively deep shaft and that the latter was presumably 
 quadrilateral in section. The lateral facet for adjoining metapodial is flat and oval 
 and is directed outward and downward. The proximal margin of this facet is not 
 closely approached by the mesopodial surface, but this may be due in part to the 
 worn condition of the specimen. The remaining portion of the mesopodial surface 
 
 ° From unpublished manuscript of Merriam, Stock, and Moody. 
 
 J. C. Merriam, Relationships of Pliocene mammalian faunas from the Pacific Coast and Great Basin provinces of 
 North America. Univ. Calif. Publ., Bull. Dept. Geol., vol. 10, No. 22, p. 428, 1917. 
 
 18 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 19 
 
 is but slightly concave transversely and does not reach the dorsal margin of the 
 proximal end. Although an offset flanks the lower portion of the distal carina on 
 one side, there was apparently not an extensive offset present on the opposite side. 
 
 The Rattlesnake specimen resembles the third metacarpal of Hapalops in 
 quadrilateral cross-section of shaft. It differs, however, from this element in 
 having a relatively shorter and deeper shaft. In Hapalops the proximal facet of 
 the inner side is more extensive dorso-vent rally, is directed for the most part 
 proximo-internally, and is of an entirely different shape than in No. 22900. The 
 dorsal margin of the proximal end in the latter specimen was not deeply indented 
 
 Fig. 2. —A to G, ground-sloth remains from Tertiary beds of Western North America. A, B, and C, ungual phalanx, 
 
 No. 1096 U. C. C., from Rattlesnake formation, Eastern Oregon; D and E, tooth No. 22110 U. C. C. from 
 Pinole Tuff-Orinda-Siesta deposits, middle California; F and G, metapodial, No. 22900 U. C. C., from Rattle¬ 
 snake formation, Eastern Oregon. All figures X 1. 
 
 in middle line as in the Miocene form. The distal carina in Hapalops is slightly 
 more prominent dorsally and the inner offset is also more extensive above than is 
 the offset preserved in the Pliocene metapodial. 
 
 Metacarpal III of Megalonyx jeffersonii is approximately twice as long as No. 
 22900 from the Pliocene. The shaft in the Pleistocene genus is relatively shorter 
 than in Hapalops, but not as short as in the Rattlesnake specimen. 
 
 With metatarsal II of H. elongatus certain differences and resemblances are 
 to be noted. No. 22900 is relatively shorter and the shaft deeper; the lateral 
 facet is small and flattened, while in Hapalops the facet for metatarsal I is relatively 
 large and concave dorso-ventrally; the offset, which is preserved at the distal end, 
 
20 
 
 CENOZOIC GRAYIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 is less extensively developed. In the metatarsal of Hapalops (No. 15545 Princeton 
 Univ. Coll.) under observation, the lateral facet articulating with the first meta- 
 podial has much the same position and is directed downward in much the same 
 manner as the facet in No. 22900. Below this facet in the Rattlesnake specimen 
 there is some indication of a small knob or boss, as in the second metatarsal from 
 the Santa Cruz beds. Dr. W. J. Sinclair, who has kindly examined a plaster cast 
 of the Rattlesnake specimen, is of the opinion that it represents a second metatarsal 
 and refers it to the right pes. Dr. Sinclair remarks that the proximal surface is 
 flatter than in Hapalops, and in metatarsal II of the latter genus and of Analci- 
 morphus this facet and the anterior (dorsal) surface meet in a markedly concave 
 margin, which is not the case in the Pliocene form. If the specimen does represent 
 the second metatarsal, then the inner facet (fig. 2, f) is for the first metatarsal 
 and the proximal surface for the mesocuneiform, as stated by Dr. Sinclair. 
 
 Presuming that the claw-phalanx described by Sinclair did come from the 
 Rattlesnake, it is still difficult to make a satisfactory comparison with the second 
 specimen now at hand. In a consideration of size the metapodial appears to 
 pertain to an individual larger than that to which the claw belonged, although 
 sufficient information concerning structure of the feet in Pliocene ground-sloths 
 of North America is not available to warrant even such comparison. 
 
 Table 2. —Comparative measurements (in millimeters). 
 
 
 Length 
 
 through 
 
 middle. 
 
 Depth of 
 distal 
 carina. 
 
 Least 
 depth 
 of shaft. 
 
 No. 22900, Rattlesnake. 
 
 l 52 
 
 46.8 
 
 30.3 
 
 Hapalops elongatus, metatarsal II, No. 15545, Santa Cruz. 
 
 20.6 
 
 16.2 
 
 7.6 
 
 Analcimorphus, sp. 2 metatarsal II, Santa Cruz. 
 
 30.3 
 
 7.6 
 
 15 
 
 Approximate. 2 Measurements taken by W. J. Sinclair. 
 
 PINOLE TUFF-ORINDA-SIESTA DEPOSITS, MIDDLE CALIFORNIA. 
 
 A single tooth (No. 22110 U. C. C.), presumably of a megalonychid ground- 
 sloth, has been found at University of California locality 2572, in the Pinole Tuff- 
 Orinda-Siesta deposits exposed along San Pablo Bay, near the station of Pinole, 
 California. The locality occurs in beds immediately overlying the Pinole tuff. 
 The deposits have been folded and are regarded as of Lower Pliocene age. The 
 mammalian fauna collected at locality 2572 consists of the following forms: 0 
 
 Tephrocyon sp. Megalonychid, probably n. g. and n. sp. 
 
 Martes sp. Antelope, near Sphenophalos. 
 
 Hypolagus sp. Pliohippus, near fairbanksi Merriam. 
 
 No. 22110 (fig. 2, d, e) is smaller than teeth of most species of Megalonyx and 
 approaches in its dimensions the teeth of Nothrotherium. It does not agree with 
 any of the teeth of the latter genus in shape or in wear of occlusal surface. It 
 lacks also the distinct longitudinal grooves on the lateral sides, a character occurring 
 commonly in teeth of Pleistocene megalonychids. Cheek-teeth of Megalonyx 
 
 a J. C. Merriam and C. Stock, manuscript in preparation. 
 
 J. C. Merriam, Relationships of Pliocene mammalian faunas from the Pacific Coast and Great Basin provinces of 
 North America. Univ. Calif. Publ., Bull. Dept. Geol., vol. 10, p. 425, 1917. 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 21 
 
 wheatleyi from the Pleistocene of Port Kennedy fissure are, according to measure¬ 
 ments given by Cope, relatively wider transversely than the Pinole specimen. 
 
 The tooth from near Pinole is well rounded on one of the lateral sides, with 
 the opposite side nearly flat. The anterior and posterior faces are flattened or 
 broadly convex and meet the flattened lateral face in well-rounded angles. On 
 the occlusal surface of No. 22110 the flattened lateral side has suffered least wear 
 and is presented as a strong ridge. The rounded side, on the other hand, is most 
 worn, and the anterior and posterior sides are beveled. The middle of the occlusal 
 surface is deeply excavated, but it is not quite certain how much of this was acquired 
 during the life of the animal. The crown of the Pinole tooth is approximately of 
 same diameter throughout its length. The transverse diameter is 17 mm., the 
 antero-posterior diameter 12.3 mm. 
 
 SAN TIMOTEO AND EDEN BEDS, SOUTHERN CALIFORNIA. 
 
 The important Cenozoic mammalian faunas from southern California described 
 by Mr. Childs Frick® include members of the ground-sloth group. Megalonychid 
 types are recorded from two Pliocene horizons. The assemblage from the San 
 Timoteo beds is regarded by Frick * * 6 to be close in age to the Blanco fauna of Texas 
 and consists of the following forms: 
 
 Megalonyx? sp. Cervid (?), medium, n. sp. 
 
 Pliauchenia? sp. Pliohippus francescana Frick. 
 
 Camelid, small sp. Pliohippus francescana minor Frick. 
 
 One of the two ground-sloth specimens described by Frick from the San 
 Timoteo is a last upper tooth, No. 23373 U. C. C., of small size and similar in 
 shape to corresponding teeth of Megalonyx. 
 
 The Eden assemblage is older than that from the San Timoteo and is regarded 
 by Frick as representing a late faunal phase of the Lower Pliocene. The list of 
 mammals is given in table 3. 
 
 Table 3. —List of mammals. 
 
 Canis? sp. Platygonus? sp. 
 
 Felis? sp. Pliauchenia merriami Frick. 
 
 Smilodon? sp. Pliauchenia sp. A. 
 
 Hyaenarctos gregoryi Frick. Procamelus edensis edensis Frick 
 
 Hypolagus edensis Frick. Procamelus edensis raki Frick. 
 
 Nothrotherium? or Pronothrotherium? sp. Procamelus, n. sp. 
 
 Megalonyx? sp. Cervid sp 
 
 Prosthennops edensis Frick. Merycodus? or Ilingoceros? 
 
 Antilocapra? sp. 
 
 Pliohippus osborni Frick. 
 
 Pliohippus osborni subform A. 
 Pliohippus edensis Frick. 
 
 Pliohippus edensis subform A. 
 Pliohippus edensis subform B. 
 Trilophodon shepardi edensis Frick. 
 
 The megalonychid remains from the Eden consist of fragmentary teeth. No. 
 23374 U. C. C. is referred tentatively by Frick c to Nothrotherium or Pronothrothe¬ 
 rium. It is remarkably similar to teeth of the Pleistocene Nothrotherium from 
 Rancho La Brea, and clearly indicates the presence of a nearly related form in the 
 Pliocene. 
 
 ° C. Frick. Extinct vertebrate faunas of the badlands of Bautista Creek and San Timoteo Canon, Southern Cali¬ 
 
 fornia. Univ. Calif. Publ., Bull. Dept. Geol., vol. 12, No. 5, pp. 277-424; plates 43-50, 1921. 
 
 6 Ibid., p. 320, fig. 29a. 
 c Ibid., pp. 349-350, fig. 56a. 
 
HABITAT OF THE MEGALONYCHINAE AS SUGGESTED BY 
 
 OCCURRENCE. 
 
 When the history of the Megalonychinae is viewed in its entirety a noticeable 
 change is to be observed in the relation which members of this group bear to their 
 environment, a change undergone during their period of evolution from the Miocene 
 to the Pleistocene. The difference in ecologic relationships witnessed in this sub¬ 
 family of ground-sloths exemplifies again the vicissitudes attending the history 
 of the mammalia throughout the Cenozoic, and some of the agencies which underlie 
 these changing conditions are of particular significance in the modification of 
 mammalian types. It is evident that adaptability to new sets of conditions per¬ 
 mitted the megalonychid ground-sloths to extend their geographic range consider- 
 abU, and the seeming readiness with which this has been accomplished probably 
 explains, or at least is connected suggestively with, the somewhat different history 
 of the group in North America in contrast to the Mylodontidae. 
 
 It is an unusual fact that the ground-sloths, while they have a remarkable 
 development during the Santa Cruz period in South America, are but meagerly 
 recorded from the older stage, the Deseado. Loomis (1914) attributes this sparsity 
 of edentates in the Deseado fauna to an environment which was not favorable to 
 the group. The Deseado mammalian fauna has, however, much in common with 
 that of the Santa Cruz in its biotic and physical relationships and is presumably 
 not distantly removed from the latter assemblage in stage of evolution. In a 
 discussion of the Deseado fauna as collected by the Amherst Patagonian expedition, 
 Loomis (1914, p. 24) states: 
 
 “Of our fauna, 11 per cent were flesh or insect eating, and for the purpose of determining the type of country 
 may best be omitted. The rodents could have been either forest or open-country forms. Of the remaining 54 
 per cent, the typotheres, the litopternas, the Rhynchippidae, the Leontinidae, the nesodonts and the birds 
 (46 per cent) were distinctly adapted to live on hard ground, the other 8 per cent being evidently suited to 
 living near a river. All 54 per cent ate either grass or browse. The litopternas are grass eaters; the typotheres 
 were specialized to eat grass or bark; nesodonts, Leontinidae, and Rhynchippidae are grass- and browse- 
 eaters. Even the Pyrotherium has a pair of gnawing tushes. The picture arising from these considerations 
 is a bush-covered prairie, a country not unlike the upland bush pampas of Patagonia to-day.” 
 
 According to Loomis, the edentates form 8 per cent of the Deseado fauna. 
 The t} 7 pes found here are but slightly less advanced than those known from the 
 Santa Cruz beds. As a matter of fact, the entire Deseado mammalian fauna 
 bears considerable similarity to that of the Santa Cruz and indicates that it is 
 not separated from the latter by a very great time-interval. 
 
 In the Santa Cruz beds the edentates comprise over 50 per cent of the mamma¬ 
 lian fauna, and of this group the ground-sloths, particularly members of the Meg¬ 
 alonychinae, are very numerous. The animals associated with the megalonychid 
 forms during Santa Cruz time are presented in table 4, compiled from the works 
 of Scott and others, wherein families are the smallest units given. 
 
 The type of environment prevailing during the period of accumulation of the 
 Santa Cruz beds, suggested by the deposits and their faunal content, was apparently 
 not greatly different from that found on the prairies or pampas now present in 
 the Patagonian region. J. B. Hatcher, 0 in a discussion of the geography of southern 
 
 a J. B. Hatcher. Rpts. Prin. Univ. Exp. to Patagonia, 1896-1899, vol. 1, Narrative and geography, p. 235, 1903. 
 
 22 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 23 
 
 Patagonia published with the narrative of the expeditions to Patagonia, has 
 commented on this subject in part as follows: 
 
 “During and after the deposition of the Santa Cruzian beds in Miocene and early Pliocene times, this 
 region was a broad but little-elevated plain, with numerous and extensive marshes, but no real mountains then 
 existed where now are the Andes.” 
 
 Our knowledge concerning the Pliocene ground-sloths, both of the northern 
 and southern continents, is still far less complete than that of the forms from the 
 preceding or following geologic periods. No extended effort is made at present to 
 segregate the various American mammalian faunas of the Pliocene on the basis 
 of environmental facies represented, primarily because of the sparsity of materials. 
 It is on this account that our concept of the true ecologic relationships of the 
 Pliocene gravigrade edentates must remain for a time in the formative stage. The 
 record of megalonychid forms in such North American Pliocene assemblages as 
 the Rattlesnake, Eden, San Timoteo, Blanco, and others may be interpreted, 
 perhaps, as pointing toward a biotic association and a physical environment 
 somewhat different from those in which members of the Megalonychinae were 
 found during the Pleistocene. 
 
 Table 4. —List of mammalian groups represented in Santa Cruz beds. 
 
 Marsupalia. 
 
 Polyprotodontia. 
 Thylacynidae. 
 Didelphyidae. 
 Diprotodontia. 
 Caenolestidae. 
 Insectivora. 
 
 Necrolestidae. 
 
 Rodentia. 
 
 Hystricomorpha. 
 
 Octodontidae. 
 
 Erethizontidae. 
 
 Chinchillidae. 
 
 Caviidae. 
 
 Edentata. 
 
 Dasypoda. 
 
 Tatuidae. 
 
 Dasypodidae. 
 
 Peltephilidae. 
 
 Glyptodontia. 
 
 G lyptodontidae. 
 ?Doedicuridae. 
 Gravigrada. 
 
 Megalonychidae. 
 
 Planopsidae. 
 
 Mylodontidae. 
 
 Toxodonta. 
 
 Toxodontidae. 
 
 Notohippidae. 
 
 Entelonychia. 
 
 Typotheria. 
 
 Interatheriidae. 
 
 Hegetotheriidae. 
 
 Litopterna. 
 
 Proterotheriidae. 
 
 Macrauchenidae. 
 
 With the coming of the Glacial Epoch a decided change is to be noted in the 
 ecologic relationships of the megalonychid division of the ground-sloth group. 
 Cope, long ago, made the fundamental distinction between the mammalian assem¬ 
 blages of this period on the basis of plains and forest habitat. The group of 
 mammals from the Equus beds of the western United States, which included 
 Mylodon and typified the former environment, was contrasted by Cope with the 
 fauna secured from eastern cave deposits. The latter assemblage was characterized 
 by the presence of Megalonyx, with which, however, Mylodon and Megatherium 
 were occasionally associated. The importance of the former genus in this fauna 
 is revealed by the designation Megalonyx beds, which Cope applied to the horizons 
 containing the forest fauna. With the palaeontological investigation of cave 
 deposits in California, mammalian assemblages were disclosed that possess much 
 in common with those obtained in similar beds of the eastern United States. A 
 somewhat similar contrast can probably be made between the mammalian 
 aggregates that lived in the forested regions and on the prairies of South America 
 during the Pleistocene. Branco,“ in an interesting and valuable study, has con- 
 
 ° W. Branco. Uber eine fossile saugethier-fauna von Punin bei Riobamba in Ecuador. Palae. Abhdl., vol. 1, pp. 
 185-198, 1.883. 
 
24 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 trasted the late Cenozoic faunas of South America with those of North America. 
 Branco concerned himself particularly with the determination of age of these 
 horizons, but from his studies there becomes apparent a certain degree of similarity 
 of relationship when the mammalian faunas from pampa deposits and from the 
 Brazilian caves are compared to the corresponding Equus-Mylodon fauna and 
 Megalonyx fauna of the northern continent. 
 
 Herluf Winge’s researches on the collections obtained by Lund from the 
 Brazilian caverns resulted in the recognition of a long list of mammals known from 
 these accumulations. Nothrotherium is a typical member of this assemblage. It 
 may be readily conceded that in such a region as Brazil, where probably during 
 the Pleistocene, as at the present time, open savannahs were associated with 
 dense forests or jungles, there would of necessity be a representation of some 
 mammalian types characteristic of the pampas. The horses, camels, litopterns, 
 and some edentates may perhaps be cited as belonging to this group. In 
 California similar associations of forest and plains mammals are known. At 
 Hawver Cave, in Eldorado County, the Pleistocene mammalian assemblage includes 
 also types living normally under the environment found in the Great Valley of 
 California. An association of mammalian forms living usually in separate environ¬ 
 ments is likewise quite as possible where accumulation is occurring on open stretches 
 of country in proximity to forested areas and where a certain amount of inter¬ 
 mingling of faunas takes place. Under these conditions it is but natural to expect 
 the presence of occasional forest types in the fossil record. Such association of 
 types has probably taken place to some extent in the Rancho La Brea beds. In 
 a survey of the cave deposits of California and their contained Pleistocene faunas 
 it has been suggested (Stock, 1918) that only three genera of the larger mammals 
 so far known, namely, Megalonyx, Nothr otherium, and Euceratherium, a are to be 
 regarded as specially characteristic of the eastern mountain belt of the State. 
 
 GEOLOGIC RANGE AND GEOGRAPHIC DISTRIBUTION OF THE 
 
 MYLODONTIDAE. 
 
 Among the earliest gravigrade edentates, known as yet only by fragmentary 
 remains from the Deseado formation of South America, presumably of Oligocene 
 age, are forms which in some characters bear resemblance to mylodont ground- 
 sloths, but with the incompleteness and paucity of materials available it is hazardous 
 to assert their true relationships. Such, for example, is the genus Octodontotherium 
 described by Ameghino (see Loomis, 1914). 
 
 Remains of ground-sloths undoubtedly belonging to the family Mylodontidae 
 have been described, principally by Ameghino, Lydekker, and Scott, in studies of 
 the Upper Miocene vertebrate fauna from the Santa Cruz beds of Patagonia. 
 The fauna from these deposits, in comparison to that from the Deseado, includes 
 the ground-sloths as an integral part. It is a curious fact, however, as noted by 
 Scott, that the family Mylodontidae, in contrast to the Megalonychinae, is 
 represented in the Santa Cruz beds by relatively few types. Only two genera are 
 
 ° Preptoceras should probably be placed with this group. 
 
PLEISTOCENE MEGALONYCHTNAE AND MYLODONTIDAE OF RANCHO LA BREA. 25 
 
 recognized by Scott, namely, Nematherium and Analcitherium. It is, of course, 
 possible that, when palaeontologic investigations have been extended to other 
 Middle Tertiary deposits of South America, this rare occurrence of mylodont 
 ground-sloths during accumulation of the Santa Cruz sediments will lose the 
 significance which may be attached to it at present. 
 
 Mylodont ground-sloths have been described from several Pliocene horizons in 
 South America, but the family is not known to be represented in beds of that 
 period in North America. In the light of recent investigations of Pliocene mam¬ 
 malian faunas of the northern continental region, it seems reasonable to suppose 
 that the mylodont ground-sloths did not reach North America until Pleistocene 
 time.“ The sparse representation of the Mylodontidae in South America during 
 the Miocene may be responsible for the absence of the group in the Pliocene of 
 North America. 
 
 Recently the Pliocene faunas of South America have been considered in detail 
 by Rovereto (1914), who distinguished four divisions or stages. These are, pro¬ 
 ceeding from youngest to oldest: 
 
 1. Chapalmalian. 2. Hermosian. 3. Araucanian. 4. Rionegrian. 
 
 From the type locality of the Chapalmalian the following members of the 
 Mylodontidae are listed by Rovereto: 
 
 Glossotherium sp. Scelidotkerium chapalmalense Ameghino. 
 
 Scelidodon rothi Ameghino. Eumylodon sp. 
 
 The Pliocene fauna from the Hermosian at the type locality, Monte Hermoso, 
 includes: 
 
 Scelidodon patrius Ameghino. Scelidodon pendolai Rovereto. 
 
 The following mylodont forms are recorded from the type locality of the 
 Araucanian in the valley of Santa Maria, province Catamarca: 
 
 Scelidodon almagroi Rovereto. Pleurolestodon avitus Rovereto. 
 
 Sphenotherus zavaletianus Ameghino. Pleurolestodon macrodon Rovereto. 
 
 Pleurolestodon acutidens Rovereto. 
 
 Finally, Rovereto lists the genus Neonematherium from deposits assigned to 
 the Rionegrian stage as exposed in the valley of the Rio Negro and occurring also 
 at various localities in Patagonia. 
 
 The present writer is not in a position to review critically the status of the 
 forms recorded above from the Pliocene of South America. The genus known 
 from the earliest stage is related to a preceding type found in the Santa Cruz, 
 while in the latest stage of the Pliocene, as recognized by Rovereto, several of the 
 mylodont ground-sloths are generically the same as Pleistocene forms. It may 
 be also noted that in general the Pliocene types are transitional in character of 
 size, occupying a position between the mylodonts of the Miocene and those of the 
 Pleistocene. 
 
 During the Pleistocene the Mylodontidae enjoyed their greatest geographic 
 distribution. Having occupied South America since pre-Miocene time, it is not 
 
 a The genus Mylodon has been recorded from the Blanco Pliocene deposits of Texas, but this statement is asserted 
 by W. D. Matthew to be an error. 
 
26 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 surprising to find the group ranging over the southern continent toward the close 
 of the Tertiary. With establishment of a land connection between the two 
 continents at about this time, these ground-sloths would perhaps be among the 
 first of the South American fauna to avail themselves of the opportunity to migrate 
 northward. Their entry into North America probably occurred by way of Central 
 America and the Isthmus of Panama, but little mylodont material has been 
 recorded as yet from these regions and from Mexico. Aside from an occurrence of 
 a fragmentary arm-bone of Mylodon in Central America, noted by Leidy (1870), 
 and the presence of a mylodont tibia in the Mexican Pleistocene," no other remains 
 are known to the writer. In the Antillean region, where the megalonychid ground- 
 sloths are surprisingly well represented in Pleistocene deposits, the mylodont types 
 are unknown. 
 
 While no less than five valid mylodont genera have been described from the 
 Pleistocene of South America, only one genus, Mylodon, is known from deposits of 
 similar age in North America. Although the physical environment in North Amer¬ 
 ica may not have offered serious disadvantages to an extensive deployment of the 
 group, a more potent factor in limiting the range of the mylodont ground-sloths 
 would seem to be the efficacy of the biotic barrier as established by the carnivore 
 group. In this regard it should be noted that just prior to the entrance of Mylodon 
 into North America there were present in the late Pliocene mammalian faunas of 
 this continent several formidable types, as, for example, Ischryosmilus, Felis, 
 Hyaenognathus, and Borophagus. Mylodon is exceptionally well represented in 
 western North America, an indication perhaps of closer proximity of this province 
 to the avenue of migration from the south. With the advent of Mylodon into 
 North America during the Pleistocene, the distribution of the Mylodontidae was 
 greatly extended. This is particularly well shown by the latitudinal range of the 
 family, which reached from approximately 51° south to approximately 47° north of 
 the equator. No other group of the larger mammals indigenous to the South 
 American continent ranged over so vast a territory during the Pleistocene. 
 
 HABITAT OF THE MYLODONTIDAE AS SUGGESTED BY 
 
 OCCURRENCE. 
 
 Obviously, an attempt to portray the habits of members of the Mylodontidae 
 can be materially aided by a consideration of the structural characteristics pos¬ 
 sessed by these forms, and by an investigation of their habitat. A correct inter¬ 
 pretation of the physical environment in which the mylodont sloths maintained 
 themselves is more likely to be derived when both the mode of occurrence and 
 particularly their biotic associations are carefully examined. In viewing such a 
 mammalian assemblage it may be readily conceded that a better understanding 
 of the activities and adaptations of the ground-sloths is reached, for the position 
 held by any animal in the economy of Nature can be judged correctly only in the 
 
 a W. Freudenberg. Op. cit., plate 1 (20), fig. 2, 1910. The tibia figured by Freudenberg is referred to Oryctero- 
 therium. This generic name is synonymous with Mylodon. 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 27 
 
 light of the internal and external factors which more or less control or influence 
 its existence. 
 
 Possessed of many unusual features in skeletal structures, these ground-sloths 
 must have offered a pronounced contrast to the more progressive types of mammals 
 indigenous to the North American continent with which they were associated 
 during the Pleistocene. Since the mylodont sloths form an integral part of the 
 Pleistocene fauna of South America, and as there is apparently no reason for 
 believing that they occupied a physical environment decidedly different from that 
 in which they moved in North America, the information relating to these southern 
 forms should render much aid in our discussion of the northern types. As a matter 
 of fact, Branco® indicated a number of years ago the fundamental similarity which 
 exists between the Pleistocene mammalian faunas of North and South America 
 so far as the broader zonal distribution of the mammalia during this period is 
 concerned. 
 
 That Mylodon and related genera were inhabitants of the great stretches of 
 open country, the pampas or steppes, in South America seems definitely estab¬ 
 lished by their occurrence and faunal association at several of the most note¬ 
 worthy localities. Mylodont genera appear also to have been restricted less in 
 their altitudinal range than other types of the larger herbivorous mammals. Let 
 us briefly review some of these occurrences. 
 
 Argentina. —The discovery of large Pleistocene quadrupeds, including a num¬ 
 ber of gravigrade edentates, in the Pampean formation of Argentina, by Charles 
 Darwin, is recorded in the chronicles of his researches while on the voyage of the 
 Beagle. From alluvial deposits exposed at Punta Alta, in Bahia Blanca, remains 
 of edentates belonging to the Megatheriidae, Mylodontidae, and Glyptodontidae 
 were collected by Darwin 6 in association with fossil material of a horse, Macrau- 
 chenia, and Toxodon. The fauna is typically one which characterized the South 
 American pampas during the Pleistocene. 
 
 Patagonia. —The occurrence of the mylodont sloth Grypotherium in the great 
 cavern located at Ultima Esperanza, Last Hope Inlet, in magellanic South America, 
 has attracted considerable attention because of the valuable materials that have 
 been preserved. It is desirable to review at this time certain of the distinctive 
 features of the find. Not only were the bones of Grypotherium recovered, but integ¬ 
 umentary parts and specimens of the droppings of this animal were also secured. 
 It has been found possible, therefore, to reach a more specific inference as to the 
 probable appearance and activity of the form than could otherwise be obtained 
 from examination of only skeletal material, and this in turn aids in an understanding 
 of the environment in which the animal lived. 
 
 The mammals found associated with Grypotherium consist of the following types: 
 
 Felis sp., extinct. Ctenomys magellanicus, living. 
 
 Felis concolor, living. Macrauchenia sp., extinct. 
 
 Arctotherium sp., extinct. Onohippidium saldiasi Roth, extinct. 
 
 Lyncodon patagonicus, living. Lama huanacos, living. 
 
 Megamys sp., extinct. 
 
 ° W. Branco. Op. tit., 1883. 
 
 b C. Darwin. Journal of researches into the natural history and geology of the countries visited during the voyage 
 of H. M. S. Beagle around the world, under the command of Capt. Fitz Roy, R. N., Appleton & Co. 1896. 
 
28 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 According to some investigators, man should be included in this fauna, or 
 perhaps some division of it. 
 
 In this faunal assemblage we note several types indicative of a plains environ¬ 
 ment, or at least one in which considerable stretches of open country were available. 
 Confirmatory evidence that Grypotherium lived under conditions such as are 
 suggested by its faunal association is offered also by an examination of materials 
 constituting the specimens of excrement of this animal. To quote from Smith 
 Woodward (1900): 
 
 “Mr. Spencer Moore has kindly examined them from the botanist’s point of view and reports that they 
 are composed ‘in large part apparently of grasses, as the haulms, leaf-sheaths, fragments of leaves, etc., of 
 these plants are frequent in the mass. A spikelet, almost entire, of what seems to be a species of Poa, and the 
 flowering glume of another grass, probably avenaceous, have also been found. Besides these there are at least 
 two dicotyledonous plants, one herbaceous and the other almost certainly so, the latter having a slender 
 greatly sclerotized stem. Unfortunately, as no leaves have hitherto been observed attached to the fragments of 
 stem, their affinities are altogether doubtful. There are numerous siliceous particles in the excrement, and 
 there are many pieces of the underground parts of plants, suggesting that they have been pulled out of the 
 ground. A few pieces of stems are sharply cut, not bruised or torn at the end.’” 
 
 Similar material has been carefully examined by Richters (1904), who reports 
 as follows: 
 
 “Die Hauptmasse des pflanzlichen Materials, aus deni der Dung besteht, sind Teile von Gramineen: 
 Halm-Bruchstucke, Blatt- und Spelzenreste: Epidermisgewebe deuten sowohl auf glatte wie behaarte Formen 
 him Haufig kommen durch Maceration freigewordene, derbe, verastelte Gefassbtindel eines, wie es scheint, 
 kahnformigen Blattgebildes, andererseits derbe, durch die Verdauung kaum beeinflusste Blatter mit dornigem 
 Rande vor. Ein durch seine maanderformigen Zellgrenzen ausgezeichnetes Epidermisgewebe mit ringformigen 
 Wulsten, auf denen Haare standen, deutet auf eine Dicotyledone hin: ebenso eine kleine Frucht, die etwa an 
 ein Chenopodium oder einen Rumex erinnert. Von Kryptogamen wurdo ein Zweigstuck eines Lebermosses und 
 durchaus unverkennbare Reste einer flachenhaft wachsenden Alge, Prasiola, gefunden. Auffallig an diesen 
 Prasiola-Resten ist der Gehalt ilirer Zellen an noch griinem Chlorophyll; im Dunkel der staubtrockenen Hohle 
 kan die Prasiola nicht gewachsen sein; vielleicht schiitzte der glatte Algenschleim das Chlorophyll vor den 
 verdauenden Saften und forderte die Stiicke schnell durch den Darmtraktus, von dessen Schleimhaut uns 
 haufig Fetzen in mikroskopischen Gesichtsfeld aufstossen.” 
 
 Bolivia. —From the Pleistocene deposits exposed in the valley of the Rio de 
 Tarija in southern Bolivia, a great collection of vertebrate remains has been 
 secured. Material from this region has been described by Gervais and by 
 Burmeister. Pertinent to the present discussion are the remarks of Erland 
 Nordenskiold (1901), who has examined the Tarija badlands: 
 
 “Mir erscheint es als das glaublichste, dass das Tarijathal ehemals in seinen engen Grenzen diese an 
 Individuen und Arten so reiclie Fauna beherbergt habe, dass das Klima aber trockener als jetzt gewesen sei, 
 so dass die Landschaft steppenartig war mit einzelnen durch die blind endenden Bache gebildeten Siimpfen. 
 Steppen, wo der Guanaco, von dem man zahlreiche Knochen einer identischen oder nahestehenden Form findet, 
 und wo die kleinhufigen Pferde geeignete Naturverhaltnisse ftxr sich fanden, und einzelne Trtimpel wo Hydro- 
 choerus cabibara und Myopotamus copyus leben konnten. Man findet nahmlich Knochen von diesen beiden 
 letzten selir nahestehenden Formen. 
 
 “Dass die Riesenthiere, von denen die wichtigsten liier gefundenen Mastodon andium , Megatherium ameri- 
 canum und Scelidotherium capellini-tarijensis?, Glyptodon clavipes, Lestodon armatus, Mylodon robustus und 
 Macrauchenia patagonica sind, gerade in solchem steppenartigen Boden giinstige Lebensverhaltnizze getroffen 
 liatten, ist nach dem, was wir liber die gegenwartige Verbeitung der grossen Saugetiere wissen, sehr wahrsch- 
 einlich.” 
 
 Another occurrence of Pleistocene mammalian remains in Bolivia, of consider¬ 
 able importance, is that near Ulloma. The fauna includes not only Mylodon(?), 
 but Scelidotherium and Megatherium, species of Hippidium, Onohippidium, Mac- 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 29 
 
 rauchenia, and a mastodon. In commenting on the habitat of Mastodon bolivianus } 
 a member of this assemblage, Pompeckj,® states as follows: 
 
 “Ob die Vegetation damals aber eine ganz wesentlich andere gewesen sein muss als heute, darf dahingestellt 
 bleiben. Gerade in der Gegend des Rio Desaguadero und Mauri ist die Vegetation heute gar nicht so sparlich. 
 Weite Strecken sind ziemlich dicht bestanden mit Biischen der Baccharis tola. Hier und in dem grasreicheren 
 Desaguaderotal finden selir sattliche Herden von Llamas reichliche Nahrung. Die Hippidionarten und Mac- 
 rauchenia brauchten wold kaum eine sehr viel reichlichere Vegetation als die heute dort lebenden Llamas. 
 Mastodon bolivianus pflugte mit scinen Oberkieferstosszahnen den Boden auf und hat wohl in Grasern und 
 Strauchern seine Nahrung gefundcn, ohne reichlicheren Baumwuchses zu bediirfen. Dass die schwerfalligeren 
 Megatherien und Scelidotherien auf reichen Baumwuchs angewiesen waren, das anzunehmen sind wir nicht 
 gezwungen, da die heute lebenden Xenarthra ja keineswegs ausgesprochene Pflanzenfresser sind. 
 
 “Zur Diluvialzeit, oder wenigstens wahrend des Abschnittes derselben, als die Fauna mit Mastodon 
 bolivianus existierte, hat die bolivianische Hochebene in der Hohe von ca. 3800-4000 m wahrscheinlich den 
 Charakter einer Steppe getragen, ahnlich wie heute, aber reicher an Niederschlagen und darum wohl auch 
 reicher an Graswuchs und an Strauchern als heute.” 
 
 Ecuador .—An interesting occurrence of Pleistocene vertebrates at Punin, 
 near Riobamba, was described a number of years ago by W. Reiss and the fauna 
 was discussed by Branco. 6 This region lies in one of the intermontane valleys of 
 the Cordilleras of Ecuador, between 1° and 2° 5' south latitude, at an elevation 
 of 2,600 to 2,800 meters. Reiss believed that the tuff which forms the deposit in 
 this valley was erupted during the Pleistocene and that the volcanic activity 
 extended into Recent time. Reiss was also of the opinion that the topographic 
 and climatic features of the region during the Pleistocene were largely comparable 
 to those of the country today. Numerous vertebrate remains were secured from 
 the Pleistocene tuffaceous beds, and the following iorms were determined by 
 Branco: 
 
 Mylodon sp. Cervus cf. chilensis Gay. 
 
 Equus andium (A. Wagner) Branco. Mastodon andium Cuvier. 
 
 Protauchenia reissii Branco. Smilodon cf. neogaeus (Lundl. 
 
 Cervus, 3 sp. 
 
 Peru .—One of the results of the Yale Expedition of 1912 was the discovery of 
 early Pleistocene vertebrates at Ayusbamba, Peru. The fauna was secured from 
 a lacustrine deposit exposed in the heart of the Peruvian Andes at an elevation of 
 approximately 12,400 feet. The assemblage of mammals described by Eaton 
 (1914) comprises the following forms: 
 
 Mylodon sp. Odocoileus brachyceros. 
 
 Parahipparion sp. Dibelodon bolivianus (Philippi). 
 
 Llama sp. 
 
 This ancient body of water, in whose deposits the above mammals were 
 entombed, has been designated Lake Ayusbamba by H. E. Gregory. 0 The origin 
 and character of the lake are described by that author as follows: 
 
 “From the sections and descriptions given above it is evident that we are dealing with deposits which in 
 part are truly lacustrine, material laid down in a body of quiet water which existed for a relatively long period 
 of time. On the assumption that each layer of sand ancl of clay represents the amount deposited during a 
 
 ° J. F. Pompeckj. Mastodon-Reste aus dem interandinen Hochland von Bolivia. Palaeontographica, vol. 52, pp- 
 17-56, plates 3 and 4, 1905. 
 
 b W. Branco. Uber eine fossile Saugethier-Fauna von Punin bei Riobamba in Ecuador. Palae. Abhandl., Bd. 1, 
 Hft. 2, pp. 39-204, plates 6-24, 1883. 
 
 c H. E. Gregory. Geologic reconnaissance of the Ayusbamba (Peru) fossil beds. Amer. Jour. Sci., vol. 37, pp. 
 125-140, 1914. 
 
30 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 single rainy season, approximately 100,000 years would be required for the accumulation of the materials ex¬ 
 posed in the present fragmentary sections. It is probable that the lake beds had greater thickness. That they 
 formerly extended much farther northward is shown by an unprotected, truncated section including 185 feet of 
 strata perched high on the valley side overlooking the Chipura River. The containing wall of rock is complete 
 except on the north side, where it has been entirely removed. Unlike the remaining portions of the rim, the 
 north wall probably consisted of fluvial and glacial debris washed from the highlands. Moraines extend to the 
 edge of the present lake deposits and ma 3 r have formed the original barrier. During the life of the lake, fans 
 from the high ridge at the south encroached upon its waters and separated the original sheet into more or less 
 detached bogs. This process, combined with the development of a channel through the unconsolidated northern 
 barrier, led to the extinction of the water body. Judging from the physical data at hand, this mountain tarn 
 may have beautified the landscape of the late Pliocene or any portion of the early Pleistocene epoch—a conclu¬ 
 sion which is in harmony with the palaeontological evidence.” 
 
 As indicated above in the present discussion, the mylodont ground-sloths of 
 North America apparently dwelt in surroundings similar to those in which the 
 animals lived in South America, but their mammalian associates in the northern 
 region during the Pleistocene were quite different from those found with the forms 
 in the southern continent prior to Pleistocene time. 
 
 OCCURRENCE OF GROUND-SLOTHS AT RANCHO LA BREA. 
 
 Table 5 is a list of the Pleistocene mammalian fauna found at Rancho La Brea: 
 
 Table 5. 
 
 Insectivora. 
 
 Chiroptera. 
 
 Carnivora. 
 
 Arctotherium californicum Merriam. 
 Ursus sp. 
 
 Aenocyon dims (Leidy). 
 
 Aenocyon milleri (Merriam). 
 
 Cams occidentalis furlongi Merriam. 
 Canis ochropus orcutti Merriam. 
 Canis andersoni Merriam. 
 
 Urocyon californicus Mearns. 
 Taxidea, possibly n. sp. 
 
 Mephitis occidentalis, n. subsp.? 
 Spilogale sp. 
 
 Mustela sp. 
 
 Smilodon californicus Bovard. 
 
 Felis atrox bebbi Merriam. 
 
 Felis daggetti Merriam. 
 
 Lynx, near rufus californicus Mearns. 
 
 Edentata. 
 
 Mylodon harlani Owen. 
 
 Mylodon harlani tenuiceps Stock. 
 Megalonyx jeffersonii californicus Stock. 
 Nothrotherium shastense Sinclair. 
 Artiodactyla. 
 
 Camelops hesternus (Leidy). 
 Capromeryx minor Taylor. 
 
 ?Antilocapra americana (Ord) 
 
 Odocoileus sp. 
 
 Bison antiquus Leidy. 
 
 Platygonus, possibly n. sp. or n. subsp. 
 Perissodactyla. 
 
 Equus occidentalis Leidy. 
 
 ??Tapirus sp. 
 
 Proboscidea. 
 
 Mammut americanum (Kerr)? 
 
 Elephas imperator Leidy? 
 
 Lagomorphs and many rodents. 
 
 An interesting feature of the deposits at Rancho La Brea, and one which does 
 not readily submit to a satisfactory explanation, is that which centers in the 
 occurrence of Pleistocene mammals in the various pits excavated by the Los 
 Angeles Museum of History, Science, and Art and by the University of California. 
 The ground-sloths forming an important element in the Pleistocene assemblage 
 occurring in the asphalt deposits possess their full share of this peculiarity. Of 
 the 96 pits sunk by the Los Angeles Museum at Rancho La Brea, only 30 were 
 productive of osseous material in sufficient quantity to encourage pursuit of exca¬ 
 vating work for any length of time. 
 
 As recorded in the field-notes of Mr. L. E. Wyman, in charge of the excavations 
 at Rancho La Brea for the museum, ground-sloth remains were secured in approx¬ 
 imately 19 of these excavations. Only a relatively small number of the remaining 
 pits yielded Pleistocene bones to the apparent exclusion of edentate remains, 
 while in a number only Recent forms were found. Pit 10, conspicuous in the 
 
PLEISTOCENE MECIALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 31 
 
 annals of the Rancho La Brea investigations because of the presence of remains 
 of man, yielded in addition a large number of Recent types and only the extinct 
 Teratornis of the truly Pleistocene assemblage. During the excavations conducted 
 by the museum, the largest amount of material belonging to the genus Mylodon 
 was secured from pits 3, 4, 13, 60, 61, 67, and 77 (see fig. 3). An estimation of 
 the number of individuals preserved in these pits collected by the museum may 
 be reached by a survey of the series of skeletal elements belonging to this animal. 
 Results of this survey indicate the presence of at least 55 individuals of Mylodon 
 in the collections of the Museum. 
 
 Fig. 3. —Topographic map of Rancho La Brea showing location of principal excavations of the University of 
 
 California and Los Angeles Museum. 
 
 Prior to the excavations by the Los Angeles Museum, the University of 
 California obtained a smaller collection of vertebrate remains from the asphalt 
 deposits, principally at localities 2050 and 2051, the latter afterwards representing 
 pits 1 and 2 of the museum catalogue. At the university diggings not less than 
 23 individuals of the genus Mylodon are represented. 
 
 The genus Mylodon may then be considered as a form occurring always with 
 the Pleistocene assemblage at Rancho La Brea whenever this is typically rep¬ 
 resented. Such an assemblage would include Aenocyon, Canis, Smilodon, Bison, 
 Equus, and Camelops, and undoubtedly a number of other large and small forms 
 
32 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 of mammalia. The true concept of what this faunal complex was like will be 
 obtained only when the last word concerning Rancho La Brea has been written. 
 It is pertinent to any faunal study involving types occurring in the asphalt beds 
 to ascertain whether or not a fundamental or basal faunal complex can be recog¬ 
 nized in these deposits, and which in turn would reflect the type of environment 
 present in the vicinity of Rancho La Brea during the period of accumulation of 
 materials. It is of great importance also in any attempt to correlate this fauna 
 with other Pleistocene assemblages living under conditions of environment essen¬ 
 tially different from those prevailing at the asphalt beds. 
 
 Considering only the ground-sloth group, the recognition of a faunal substratum 
 would materially aid in explaining the infrequent occurrence of the genera Mega- 
 lonyx and Nothrotherium as contrasted with that of Mylodon. In the subfamily 
 Megalonychinae the genus N othr other ium is by far better represented in the pits 
 than is Megalonyx —both genera occurring always in association with remains of 
 Mylodon. No less than 12 examples of N othr other ium are known and were obtained 
 in pits 3, 4, 61, 67, 77, and 91. The nearly complete skeleton of N othr other ium 
 was exhumed in pit 3. During the first examination of the asphalt deposits in 
 1905, Professor Merriam recovered a tooth of this ground-sloth, now in the col¬ 
 lections of the Museum of Paleontology, University of California. The specimen 
 was found close to the surface in thoroughly indurated asphaltic material that 
 apparently had been discarded during the early mining operations for asphaltum. 
 
 If the excavations of the museum are considered for the moment alone, it 
 will be noted that in comparison to the ground-sloth Mylodon, which is represented 
 in 19 pits, Nothr other ium occurs in only 5 or at most 6 pits and is particularly well 
 represented in 1 (pit 67).° The evidence seems to indicate that N othrotherium was 
 not an habitual resident of the region about the petroleum pools, as was Mylodon f 
 and the record of its sporadic occurrence is perhaps best interpreted if the form 
 is considered as a visitant. 
 
 In a consideration of the remaining member of the Megalonychinae known 
 to be present at Rancho La Brea—namely, the genus Megalonyx —an occurrence 
 similar to that of Nothrotherium is noted, with, however, the added peculiarity that 
 it is present in fewer numbers than the latter form. This fact is even more striking 
 when we recall the extensive geographic distribution of Megalonyx in North America 
 during the Pleistocene and its restriction to that continent as contrasted with 
 Nothrotherium. At Rancho La Brea the genus is represented by comparatively 
 little material. During the first excavations conducted by university parties, 
 some material referred to this form was obtained, which included structures of the 
 feet. Later, the university secured at locality 2051 additional remains consisting 
 of a well-preserved lower jaw, elements of the anterior and posterior extremities, 
 and of the manus and pes, and indicating perhaps the presence of not more than 
 one or two individuals. Curiously enough, the remains of Megalonyx secured by 
 
 ° It should be noted, however, that as the excavations progressed in the field it was found that pits 61 and 67, which 
 originally were opened in close proximity, later developed into one and the same pit. It is possible that during the period 
 of active trapping of these animals in the tar both pits were open at the surface in the form of pools, either separated or 
 connected, and had their origin in the same underground phenomena which resulted ordinarily in the formation of such 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 33 
 
 the Los Angeles Museum of History, Science, and Art are far fewer than those 
 obtained by the university, notwithstanding the fact that the excavations were 
 conducted on a much more extensive scale. In the museum collections there is 
 a fourth metatarsal and a patella from pit 3, a peculiar tooth which has con¬ 
 siderable resemblance to Megalonyx dissimilis, and several phalanges belonging 
 probably to the same individual or individuals secured by the University of 
 California from locality 2051. Thus, from the known materials not more than 4 
 or 5 individuals of Megalonyx can be estimated as occurring at Rancho La Brea. 
 
 Fig. 4.—A, Nolhrotherium shastense Sinclair. A 1 , Hapalops sp. Skeletons shown in lateral view and to same scale. 
 
 A 1 , after Matthew. 
 
 PROBABLE APPEARANCE, HABITS, AND SOME FACTORS CONDITION¬ 
 ING THE LIFE OF THE RANCHO LA BREA GROUND-SLOTHS. 
 
 Nolhrotherium is smaller than Mylodon, but is distinctly larger than the great 
 ant-eater Myrmecophaga. Compared with nearly related forms, N othr other ium is 
 seen to be as large as some contemporary species of Megalonyx of the Pleistocene 
 and probably three times as large as the ancestral Hapalops of the Miocene (see 
 fig. 4). Nothr other ium possesses a relatively small head, which tapers forward to 
 the prehensile lips, and a rather long and flexible neck. A narrow skull and the 
 
34 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 spout-like symphyseal region of the mandible seem to imply a long, narrow tongue 
 capable of protrusion. Nothrotherium lacked the caniniform teeth, so prominent a 
 feature in other ground-sloths. The cheek-teeth grew from persistent pulps and 
 their crowns were usually worn into anterior and posterior transverse crests with 
 intervening valleys. These teeth were of service in mincing vegetable food. 
 
 The anterior appendages are long, Nothrotherium in this character reminding 
 one, to some extent, of the tree-sloths. The individual elements of the limbs are 
 more lightly constructed than in Mylodon, and this is true also when comparison 
 is made with Megalonyx. The hand with its three claw-bearing digits found use 
 in serving the head and in defense. The curiously flattened claw of the second 
 digit, in contrast to those of the remaining fingers, indicates perhaps some special 
 function. 
 
 While the pelvic girdle is fairly broad, Nothr otherium was not so heavily con¬ 
 structed posteriorly as Mylodon. The elements of the hind-leg are more slender 
 than those in the latter genus and the fore-leg in proportion to the thigh was 
 longer in Nothr otherium. The peculiar posterior foot in this ground-sloth seems 
 to indicate that the animal was incapable of prolonged walking on the ground. 
 The hind foot was less adapted to this purpose than that in Mylodon and certainly 
 less so than that in either Hapalops or Myrmecophaga. 
 
 Nothr otherium (see frontispiece) probably had a heavy coat of hair, as is com¬ 
 monly found among the pilosic Xenarthra, but we can not of course be certain of the 
 exact nature of the external covering. While the animal probably defended itself 
 chiefly by use of the anterior extremities, there appears to be good reason for be¬ 
 lieving that it carried no protective armor of ossicles within the skin, as did Mylodon. 
 
 The characters thus indicated for Nothr otherium are interpreted as suggesting 
 an animal living almost habitually in the dense jungles and forests or in regions 
 having a heavy stand of timber. In this environment the animal availed itself 
 not only of fronds, herbs, mosses, nuts, fruits, and berries, but clambering at the 
 base of trees it would feed also on the foliage of the lower branches within reach. 
 In such surroundings the ground-sloth was perhaps safe from the attacks of the 
 larger carnivores frequenting the plains, as the saber-toothed tiger, the great lion, 
 and the dire wolves. Excessive drought, glaciation with a lowering of the snow¬ 
 line in the mountainous regions or the presence of other unfavorable conditions 
 may have driven Nothr otherium occasionally from its natural habitat, forcing the 
 animal to compete at a disadvantage with other mammals in a new and temporary 
 environment. The occurrence of the ground-sloth at Rancho La Brea suggests 
 that it did not live habitually in the immediate vicinity of the petroleum pools. 
 
 Megalonyx is unfortunately not well enough represented by materials in the 
 Rancho La Brea beds or in other Pleistocene deposits of western North America 
 to permit a satisfactory restoration. The ground-sloth is undoubtedly related to 
 Nothr otherium, but distinct structural differences, as well as resemblances, are 
 noted when the two forms are compared. Megalonyx was more robust than 
 Nothr otherium, in which character it makes some approach to Mylodon . The 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 35 
 
 similarities in structure that exist between Megalonyx and Nothrotherium pre¬ 
 sumably point to habits of life somewhat comparable to those of Nothr otherium. 
 
 In the Rancho La Brea mammalian assemblage, Mylodon harlani, as a quad¬ 
 rupedal form (plate 45), was much smaller than the mastodon, but, like the latter, 
 possessed considerable bulk and weight. With ponderous structure and perhaps 
 an inherent trait of lethargic movement, Mylodon typifies an animal which should 
 be regarded as the antithesis of a fleet-footed ungulate. Mylodon frequented the 
 plains or dwelt at least in open stretches of flat or rolling country and occasionally 
 ranged into elevated and perhaps more rugged regions. On the plains it may have 
 resembled the rhinoceros in being more dependent on the olfactory sense than on 
 keenness of sight in the recognition of approaching danger. Mylodon was well 
 represented during the Pleistocene in the region of Rancho La Brea, where its 
 associates among the large mammals were the horse ( Equus occidentals), the bison 
 (Bison antiquus), the camel ( Camelops hesternus), and members of the Proboscidea. 
 Its most formidable enemies among the carnivores of the Rancho La Brea 
 mammalian fauna were the saber-toothed tiger (Smilodon californicus), the great 
 lion (Felis atrox), and the packs of dire wolves ( Aenocyon dirus). In this 
 assemblage the saber-toothed tiger, one of the most ferocious of the predatory 
 beasts evolved during the later Cenozoic, and remarkably adjusted in structural 
 characters to an association with the thick-skinned, slow-moving mammals of the 
 North American Tertiary, may have found in the ground-sloths an added if not 
 an easier prey. Counteracting the offensive warfare waged by the Carnivora 
 were several characters in Mylodon of considerable importance. The strongly 
 built chest and the powerful anterior limbs suggest great crushing strength. The 
 latter character and the large claws of the hand were undoubtedly of great service 
 to Mylodon in combating the attacks of its enemies. We note here a defensive 
 struggle comparable to that which the great ant-eater and tree-sloths engage in 
 with their enemies to-day. Mylodon was also protected against the deadly stabs 
 of the truculent saber-tooth by the thick skin with heavy coat of coarse hair, 
 and particularly by a dermal armor of pebble bones. The great tenacity of life 
 exhibited presumably by the ground-sloths, even when severely injured, and still 
 a notable feature of the tree-sloths and ant-eaters of South America, was undoubt¬ 
 edly of considerable benefit to these animals in the struggle for existence. 
 
 Mylodon may have wandered in small herds, feeding largely on grasses, 
 shrubbery, fruit, berries, roots, and herbs. The short, thick neck with rather 
 closely interlocking vertebrae did not possess much range of movement. The 
 shortness of the neck was compensated by the long skull. The head terminated 
 rather bluntly in front with thick broad lips that were somewhat prehensile. 
 Mylodon, in lacking incisor teeth, probably did not graze by shearing the blades 
 of grass after the manner of ungulates, but often tore the plants from the ground. 
 The caniniform teeth may have been used in gnawing roots and the bark of trees, 
 but the rather weak construction of these teeth and their frequent absence in the 
 upper jaw suggest that they were not efficient structures in this service. These 
 teeth may have found more adequate employment in cutting small branches or 
 
36 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 twigs as the animal browsed. The relatively long jaws and the lobate teeth growing 
 from persistent pulps gave a fairly good grinding battery. 
 
 In Mylodon, not only are the anterior limbs heavy, but the femur is particularly 
 massive, while the tibia shows the characteristic shortening found in animals of 
 slow movement. The writer can not agree with the suggestion made by Abel, 
 that Mylodon may have been bipedal. It is true that in this genus the pelvis is 
 large, with ilia greatly expanded. This type of pelvis is essential, however, for 
 the support of the heavy viscera as well as for the attachment of powerful muscles 
 controlling the movements of the femur. Mylodon, in resting, may have sat on its 
 haunches, after the manner of bears. 
 
 The presence of a well-formed clavicle not only facilitates the use of the 
 anterior appendages in serving the head and in permitting the arms and hands to 
 act better as defensive organs, but the element was retained perhaps as a result 
 of the slow movement of the quadruped. In walking the hands rested on the ends 
 of digits 4 and 5, and we notice here a decided reduction in digital elements. The 
 large claws carried by digits 3 and 2 may have been used also for digging. 
 
 As we have seen from the discussion of the habitat occupied by the ground- 
 sloths, with which the present study is particularly concerned, Nothrotherium, 
 Megalonyx, and Mylodon were not subjected always to similar environmental 
 conditions. The differences in habitat noted between the megalonychid and 
 mylodont ground-sloths tend to strengthen the interpretation of the structural 
 characters possessed by these forms. 
 
 CLASSIFICATION AND RELATIONSHIP OF THE GRAVIGRADE 
 
 EDENTATES OF RANCHO LA BREA. 
 
 A survey of the classification of the ground-sloths as made by previous authors 
 indicates that there has not been uniformity of opinion regarding the systematic position 
 of these mammals. Under the group of gravigrade edentates several divisions have 
 been recognized, having either subfamily or family rank. For example, Zittel divides 
 the Gravigrada into three subfamilies, namely, (1) Megatheriinae, (2) Megalonychinae, 
 and (3) Mylodontinae. Under the Megatheriinae, Zittel lists, among other forms, the 
 large Megatherium and the genus Nothrotherium. Zittel also regards the Planopsinae, a 
 group recognized originally by Scott as of family rank, as ancestral to the Megatheriinae. 
 In the subfamily Megalonychinae are placed such genera as the Pleistocene Megalonyx 
 and the Miocene Hapalops, Eucholoeops, Megalonychotherium, and other forms from 
 the Santa Cruz beds. In the Mylodontinae are included Mylodon and related genera. 
 
 Scott (1903-4), in the report on the edentates from the Santa Cruz deposits, 
 regarded the genus Planops as rather closely related to the megatheres, but sufficiently 
 distinct to be given recognition with other forms under a family, namely, the Planop- 
 sidae. In a later work, however, Scott (1913) has simplified the classification of the 
 Gravigrada and lists but three families, namely, the Megatheriidae, Megalonychidae, 
 and the Mylodontidae. 
 
 The present study of the gravigrade edentates has convinced the writer that a 
 somewhat different grouping of the forms should be made. It appears evident that the 
 members of the Megatheriidae and the Megalonychidae stand closely related in Miocene 
 time and that their common ancestry was not distantly removed in time from this 
 period. This seems to be clearly indicated in the fundamental structures of the skull 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 37 
 
 arid dentition of the types as represented by the specimens from the Santa Cruz beds of 
 Patagonia. In both forms the dentition in the cheek region consists of teeth that are 
 nonlobate. In the Pleistocene megatheres and megalonychids these teeth become 
 quadrate or semi-quadrate in cross-section. The last lower tooth in the Miocene planop- 
 sids and megalonychids exhibits also the peculiarity of having the principal transverse 
 axis oblique to the antero-posterior axis of the jaw. While the Pleistocene Megatherium 
 exhibits a last lower tooth similar in general shape and in position to the anterior cheek¬ 
 teeth, the corresponding tooth in Megaionyx and in Nothrotherium differs in shape from 
 the anterior teeth and retains the obliquity of its transverse axis. Moreover, the Miocene 
 planopsids and megalonychids are both characterized by the presence of caniniform 
 teeth anterior to the cheek-teeth. These teeth are lost in Megatherium and in Nothro- 
 therium, but are very prominent in Megaionyx. 
 
 Fig. 5. — Nothrotherium shastense Sinclair. Skull, dorsal view. X 0.33. n, nasal; nix, maxillary; l, lachrymal; If, lachrymal 
 foramen; fr, frontal; rna, malar; sq, squamosal; pa, parietal; so, supra-occipital. Rancho La Brea Pleistocene. 
 
 Fig. 6. — Nothrotherium shastense Sinclair. Skull, ventral view. X 0.33. mx, maxillary; anf, antorbital foramen; ma, malar; 
 pi, palatine; pic, palatine canal; sq, squamosal; pt, pterygoid; s, pterygoid sinus; opt, orifice of pterygoid sinus; per, 
 periotic; bs, basisphenoid; bo, basioccipital; flm, foramen lacerum medium; ftp, foramen laccrum posterms; stp, stylohyal 
 process; 2, 5, second and fifth superior teeth. Rancho La Brea Pleistocene. 
 
 From the above considerations it would appear that the differences in the dentition 
 between the megatheres and the megalonychid ground-sloths are not of sufficient 
 importance to merit the recognition of the groups as distinct families. Likewise, such 
 differences as may appear in the structure of the skull and skeleton, particularly in 
 the Pleistocene forms, may be in part due to a difference in size and not necessarily 
 indicative of a distant relationship. The writer would, therefore, group all these 
 forms under the Megatheriidae, recognizing within this family the two subfamilies, 
 Megatheriinae and Megalonychinae. 
 
38 
 
 CENOZOIC GRAYIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 Standing in decided contrast to the Megatheriidae is the great division of mylodont 
 ground-sloths, including such forms as Mylodon, Scelidotherium, Nematherium, and 
 others. Because of distinctive characters in the dentition, skull, and skeleton, this 
 group is regarded of family rank, for which the name Mylodontidae should be retained. 
 
 The mylodont ground-sloths are not only distinguished by teeth that are lobate, but 
 the later types are known definitely to have carried in their skin a dermal armor of 
 pebble-like bones. In these two characters the Mylodontidae differ from the Mega¬ 
 theriidae and are to be compared rather with the Glyptodontidae. The Pleistocene 
 glyptodonts are characterized by a peculiar type of teeth, from which the group derives 
 its name. Among the glyptodonts of the Miocene, forms are present in which the 
 trilobed teeth were not distributed throughout the length of the jaws, but the anterior 
 teeth were of more simple pattern. An interesting comparison can be made between 
 the anterior teeth of Eunicepeltus and the teeth present in Mylodon. The last lower 
 tooth in Mylodon exhibits particularly some characters strikingly similar to those in 
 the trilobed teeth of the glyptodonts. On the other hand, it is recognized that a compar¬ 
 ison between the Miocene mylodonts and glyptodonts would be more appropriate. 
 
 Fio. 7 .—Nothrotherium shastense Sinclair. Skull, lateral view. X 0.33. 3, third superior tooth; fr, foramen rotundum; 
 
 fo, foramen ovale; remaining letters as in other figures. Rancho La Brea Pleistocene. 
 
 Fia. 8 .—Nothrotherium shastense Sinclair. Skull, posterior view. X0.33. Letters as in other figures. Rancho La Brea 
 
 Pleistocene. 
 
 Unfortunately, the former group is not known by much material from the Santa Cruz 
 beds. Concerning the dentition in the Miocene Mylodontidae, Scott (1904, p. 345) has 
 remarked as follows: 
 
 “The family is most distinctively characterized by the teeth, which in the upper jaw are triangular, and 
 in the lower jaw lozenge-shaped, while the last lower tooth is bilobate; the valleys are central and are usually 
 enclosed in a continuous wall of the harder dentine.” 
 
 In such a genus as Nematherium, the dentition shows no greater similarity to the 
 anterior teeth of the Miocene glyptodonts than does that of Mylodon. As a matter of 
 fact, in this form the last lower tooth has a more simple pattern than in the Pleistocene 
 Mylodon. Nevertheless, the lobate character is already defined and, in view of the 
 very scant material available of these Tertiary forms, still furnishes a basis for comparison 
 with the glyptodont types. 
 
 We might conceive of the Mylodontidae as arising from an early stock of glyptodonts 
 or nearly related forms in which simple lobate teeth were present throughout, or nearly 
 throughout, the length of the jaws. Such armor as may have been present in the earliest 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 39 
 
 forms was subsequently reduced until in the Miocene and later types only rudimentary 
 dermal bones remain. It is possible also that in the earliest mylodonts the exoskeleton 
 was already rudimentary. It would be interesting to compare the Mylodontidae and 
 the Glyptodontidae of the South American Oligocene and Eocene were these groups 
 known by complete types. Perhaps a detailed comparison of the later forms, with due 
 allowance for changes in the skull and skeleton of the glyptodonts necessitated by the 
 great development of the dermal armor, would yield data pertinent to the present 
 discussion. The origin of the Mylodontidae must have antedated considerably the 
 period of the Santa Cruz deposition. 
 
 The tree-sloths are more closely related to the Megatheriidae than to the Mylodon¬ 
 tidae. The origin of the Tardigrada is still obscure, for, unfortunately, these forms are 
 not known in the palaeontologic record of early or middle Tertiary time. 
 
 MEGALONYCHINAE OF RANCHO LA BREA. 
 
 CHARACTERS OF SUBFAMILY MEGALONYCHINAE. 
 
 Hairy ground-sloths with apparently no dermal ossicles. Skull elongate and cylin¬ 
 drical ( Hapalops and Nothrotherium), or short, broad, and deep ( Megalonyx); pre- 
 maxillaries reduced. Neck and trunk more elongate than in Mylodontidae; lumbar 
 vertebrae not ankylosed with sacral series. Dentition flf. First tooth, when present, 
 is caniniform and in cross-section is round, oval, or meniscoid, with an inner bulge. First 
 tooth separated by long diastema from cheek-teeth. Cheek-teeth oval or semi-quadrate 
 in cross-section. Last upper tooth compressed antero-posteriorly; last lower tooth 
 with principal transverse axis oblique to fore-and-aft axis of tooth-row. In Pleistocene 
 forms last lower tooth with convex inner face and flattened outer face. Humerus with 
 entepicondylar foramen; femur with third trochanter; pelvis with ilia not so expanded 
 as in Mylodontidae. Claw-processes of ungual phalanges usually laterally compressed; 
 largest ungual phalanx in digit III, pes. 
 
 CHARACTERS OF GENUS NOTHROTHERIUM. 
 
 Skull elongate and cylindrical; lachrymal prominent; malar vertically expanded 
 with inferior process slender; frontal with large sinus; pterygoid usually with large 
 posterior sinus opening inward. Lower jaw slender with spout-like predental region. 
 Dentition |. Last superior tooth compressed antero-posteriorly and semi-quadrate in 
 cross-section. Anterior caniniform teeth absent. Haemapophyses in middle part of 
 tail are X-shaped. Scapula relatively small. Manus with metacarpals longer and more 
 slender than in Megalonyx; pollex reduced and digit V rudimentary. Ungual phalanx 
 of digit II, manus, broad with convex dorsal surface. Wing-like posterior process of 
 calcaneum longer, but not as broad as in Megalonyx. Astragalus resembles that of 
 Mylodon and is more specialized than in Hapalops or in Megalonyx in adjustment to 
 rotation of pes to a position of rest on outer side. Metatarsal IV larger than in Mega¬ 
 lonyx; metatarsal V with lateral wing-like process. Hallux reduced; ungual phalanges 
 in digits II and III; digits IV and V rudimentary in pes. 
 
40 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 Genus NOTHROTHERIUM Lydekkeb. 
 
 Coelodon. P. W. Lund, Ann. Sei. Nat., ser. 2 (Zool.), vol. 11, p. 220, 1839. 
 
 Nothrotherium. R. Lydekker, Nicholson and Lydekker, Manual of Palaeontology, cd. 3, vol. 2, p. 1299, 1889. 
 Hypocoelus. F. Ameghino, Rev. Arg. Hist. Nat., vol. 1, p. 250, 1891. 
 
 Nothrotherium?. W. J. Sinclair, Univ. Calif. Publ., Bull. Dept. Geol., vol. 4, p. 153, 1905. 
 
 Nothrotherium. F. Ameghino, Rev. do Museu Paulista, vol. 7, p. 74, 1907. 
 
 Nothrotherium. C. Sfock, Univ. Calif. Publ., Bull. Dept. Geol., vol. 7, p. 342, 1913. 
 
 Coelodon. H. Winge, E. Museo Lundii, vol. 3, pt. 2, p. 287, 1915. 
 
 Nothrotherium. O. P. Hay, Proc. U. S. Nat. Mus., vol. 51, p. 116, 1916. 
 
 Nothrotherium. C. Stock, Univ. Calif. Publ., Bull. Dept. Geol., vol. 10, p. 137, 1917. 
 
 Nothrotherium shastense Sinclair. 
 
 Nothrotherium graciliceps Stock, Univ. Calif. Publ., Bull. Dept. Geol., vol. 7, p. 342, 1913. 
 
 Nothrotherium shastense Sinclair, Univ. Calif. Publ., Bull. Dept. Geol., vol. 10, p. 137, 1917. 
 
 Description of Material. 
 
 CRANIUM. 
 
 The largest available skull (No. 1800-5)° of Nothrotherium from Rancho La Brea is 
 approximately as long as a skull of Megalonyx jeffersonii described by Leidy (1855). 
 With this exception the specimens from the asphalt beds are all somewhat shorter than 
 either M. jeffersonii or M. leidyi. In marked contrast to Megalonyx, however, is the 
 
 a! 
 
 Fig. 9. — A, Nothrotherium shastense Sinclair. A 1 , Hapalops longiceps Scott. Skull and mandible, 
 lateral view. X 0.33. A i adopted and reversed from Scott. 
 
 relative slenderness of skull in Nothrotherium. In this respect the latter genus more 
 closely resembles the Miocene Megalonychinae and Megatheriinae (Scott, 1903, p. 164, 
 fig. 16) from which it has also deviated less than Megalonyx in general shape of skull. 
 Contrasted with Planops, one of the largest of the Santa Cruz Gravigrada, the skull 
 of the Pleistocene genus is distinctly larger. 
 
 ° All numbers used in the following comparisons in this paper, unless otherwise stated, are the catalogue numbers of 
 specimens in the Los Angeles Museum of History, Science, and Art. 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 41 
 
 Nothrotherium differs from Megalonyx and agrees with Hapalops in the long and 
 slender muzzle (plates 2 and 4, and fig. 9). In No. 1800-3, the anterior ends of the 
 maxillaries are noticeably flared. This character is less apparent in No. 1800-6 and may 
 be but little developed as in No. 1800-5. The nasals are long and may be more decidedly 
 arched in some skulls than in others. In No. 1800-3 the fronto-nasal suture is convex 
 posteriorly, thus resembling most species of Hapalops. This suture may, however, vary 
 in Nothrotherium. It may be decidedly V-shaped as in No. 1800-6, but in this specimen 
 the original suture appears to have been somewhat in advance of the present separation. 
 In No. 1800-6 the nasals have apparently fused with the frontals, a secondary or pseudo 
 fronto-nasal suture developing posterior to the original suture. A similar abnormality 
 in the separation between nasal and frontal has been noted in skulls of Mylodon harlani 
 from Rancho La Brea (Stock, 1914, p. 322). 
 
 In some specimens the lachrymal is not as prominent in superior view as in species 
 of Hapalops , but its canal is very much in evidence. In Megalonyx the lachrymal and 
 its canal are apparently not to be seen in dorsal aspect. The postorbital processes of 
 the frontals may be very blunt in some specimens of Nothrotherium, thus being more as 
 in Megalonyx and differing from Megatherium. 
 
 Just behind the postorbital process is a small depression, seen best in lateral view 
 of skull (plate 4, fig. 1), at the posterior end of which lies the exit of a small canal. This 
 canal extends backward and downward in the lateral wall of the frontal for a distance 
 of approximately 45 mm., then turns inward and opens into the brain-case, the perfora¬ 
 tion being situated beyond the upper outer margin of the orifice through which the 
 olfactory lobe extends. Between and slightly posterior to the supraorbital borders, 
 the frontals sag in the middle line. This sag is perhaps most evident in the largest 
 specimen, No. 1800-5, and in skull 1800-7. In skull 1800-13 it is practically absent. 
 
 The temporal ridges are but faintly marked in all specimens from Rancho La 
 Brea, Nothrotherium thus resembling the Miocene Megalonychinae. At the coronal 
 suture, where the crests approach each other most closely, the distance between them 
 may vary. There is never a pronounced sagittal or lambdoidal crest present, as in 
 Megalonyx. In front of the weak lambdoidal crest the dorso-lateral surface of the parietal 
 may sag slightly. This sag is distinctly developed in 1800-6, where it reaches the median 
 dorsal suture of the cranium, or it is hardly discernible, as in skulls 1800-3 and 1800-5. 
 In all skulls of Nothrotherium the supraoccipital is more prominent in superior view than 
 in Megalonyx. 
 
 The Brazilian species described by Reinhardt (1878) differs noticeably from the 
 Californian form in dorsal contour of skull. It is possible that this difference is in part 
 due to age, but it is probably a specific distinction. In Hapalops longiceps the frontal 
 is more elevated than the parietal, while in H. vulpiceps and H. gracilidens the reverse 
 is true, the parietal being much more highly vaulted than the frontal. In the elevation 
 of the frontal the Rancho La Brea species of Nothrotherium approaches Planops most 
 closely. Megalonyx differs from its Pleistocene contemporary in the straight dorsal 
 contour of the skull. 
 
 The zygomatic process of the squamosal is long and slender. It is relatively longer 
 than in H. longiceps. In skull 1800-13, used as the type in the first description (Stock, 
 1913) of the species from Rancho La Brea, this process is lacking. The malars have 
 been preserved in several skulls, notably in No. 1800-11, where they are firmly attached 
 at the base to lachrymal and maxillary. The malar resembles that in Hapalops in 
 slenderness, and, as in the Miocene genus, is not fused with the zygomatic process of 
 the squamosal. It differs from H. longiceps in (1) its relatively shorter antero-posterior 
 extent, (2) an absence of a marked outward flare of the infraorbital border, (3) a shorter- 
 dorsal prong, but deeper notch for the zygomatic process of the squamosal, and (4) a 
 relatively longer and more slender ventral prong. 
 
42 
 
 CENOZOIC GRAYIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 In Megalonyx the zygomatic process of the squamosal is much heavier than in 
 Nothrotherium. In M. leidyi (Lindahl, 1893) and in Megatherium americanum this process 
 has completely fused with the malar. The postorbital border of the malar is decidedly 
 more curved in M. leidyi than in Nothrotherium. The ventral prong is also relatively 
 more slender at the base in M. leidyi. 
 
 The parietal sends downward and forward a narrow extension which usually meets 
 the lateral portion of the alisphenoid. Occasionally this extension does not reach the 
 alisphenoid, but is interrupted by the frontal. The foramen ovale and the foramen 
 rotundum are separate and are of approximately equal size. In Hapalops, according to 
 Scott, the exits for the second and third branches of the trigeminal nerve are combined. 
 The anterior lacerate foramen is situated anterior to the foramen rotundum at a distance 
 slightly greater than that between the foramen rotundum and the foramen ovale. The 
 exit for the ophthalmic branch of the trigeminal nerve lies at the posterior end of a groove 
 which widens anteriorly, and may be somewhat obscured by an overhanging ridge 
 formed by the frontal. Posteriorly this ridge straightens and in some specimens forms 
 
 Fig. 10.—A, Nothrotherium shastense Sinclair. A 1 , Pronothrotherium typicum Ameghino; A 2 , 
 Hapalops longiceps Scott. Ventral view of skull. A 1 , after Rovereto; A 2 , after Scott. 
 
 the division between frontal and alisphenoid to the anterior end of the squamosal. 
 The optic nerve enters the orbit from this groove. It is not clear whether the second 
 cranial nerve passes through the cranial wall in the canal occupied by the first branch 
 of the trigeminal or whether it passes through a distinct canal. A small opening has 
 been observed on the inner wall of the canal for the first branch of the fifth nerve close 
 to the exit. The orbitosphenoid is partly defined by suture in some specimens. 
 
 The premaxillaries, although originally present in the Rancho La Brea skulls, have 
 unfortunately not been found in the collections. Their articulating surfaces on the 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 43 
 
 maxillaries reveal some interesting peculiarities. On each side of the small median 
 V-shaped notch the maxillary is thickened, especially the palatal portion. This thickening 
 of the maxillary extends well up along the sides of the external narial opening, in some 
 skulls reaching nearly halfway to the top of the orifice. In the Brazilian species described 
 by Reinhardt, the premaxillary is shown as a Y-shaped element. In Reinhardt’s figure 
 (Reinhardt, 1878, plate 1, fig. 3) the median arm of the posterior end is seen to be short, 
 while the lateral arm is longer and appears to curve upward. This apparently, 
 corresponds with the thickening of the maxillary at the inferior external angle of the 
 narial opening. The anterior arm of the premaxillary was presumably not very long 
 in Nothrotherium from the asphalt beds. 
 
 In Hapalops a similar Y-shaped premaxillary is present. The articulation of this 
 element with the maxillary may somewhat resemble that in the Rancho La Brea genus. 
 The articulation between premaxillary and the end of the palatal part of the maxillary is 
 not as extensive in Nothrotherium as in Hyperleptus. In the Miocene Planops, according 
 to Scott (1904) “the premaxillae were evidently large, the facets for them upon the 
 maxillaries being of uncommon size, while the median notch for the spines is a charac¬ 
 teristically deep and narrow V.” In Megalonyx jeffersonii, Leidy (1855) describes the 
 premaxillaries as “simple, oblong, quadrilateral plates, a little more than two inches 
 in depth, and three quarters of an inch in breadth.” 
 
 Owing to the slenderness of the muzzle, the anterior aspect of the skull of Nothrothe¬ 
 rium is very unlike that in Megalonyx. In M. jeffersonii, according to Leidy, “the end 
 of the face is relatively narrower and higher than in Mylodon; and in outline is more 
 vertically oblong quadrilateral, with the upper margin convex and the sides nearly 
 vertical.” In Nothrotherium, where the first or caniniform tooth is absent, the anterior 
 end of the skull is nearly round or oval. The external narial opening is relatively wide 
 transversely as contrasted with the height. The lateral walls show no indication of 
 notching as in Megatherium. With the disappearance of the caniniform tooth in 
 Nothrotherium, the depression of the lateral wall of the maxillary anterior to the first 
 (2) upper cheek-tooth becomes nearly obsolete. This depression is better developed 
 in Hapalops and is very pronounced in Megalonyx. 
 
 The palate (plate 2, fig. 2) anterior to the first upper cheek-tooth (2) may be 
 approximately twice as wide as between the tooth-rows. The lateral borders of the 
 edentulous portion of the palate may be nearly straight, as in No. 1800-3, or they may 
 become convex outward, as in skull 1800-5. In the outwardly convex borders of the 
 palate, No. 1800-5 resembles greatly species of Hapalops in which the first or caniniform 
 tooth is some distance posterior to the end of the skull. In particular does it resemble 
 a skull of Planops longirostris illustrated by Scott (1904, plate 59, fig. la). In Megalonyx 
 the palate in front of the first upper tooth (2) is relatively shorter and the sides diverge 
 anteriorly more than in Nothrotherium. 
 
 The palate is flat or somewhat convex. It is considerably pitted with large and 
 small foramina, resembling Hapalops in this respect more than Planops. On either side 
 of the raised median suture and between the tooth-rows extends a series of large foramina. 
 From the most anterior of these, which occupy a position approximately at the middle 
 of the predental portion of the palate, widening grooves extend forward to nearly the 
 front end of the skull. Scott describes the edentulous space in H. longiceps as having 
 “a well-defined groove on each external side, with raised outer border.” In No. 1800-5 
 the anterior openings of the palatine canals lie between the points of greatest outward 
 curvature of the margins of the palate. In skull 1800-2 they open on the surface of the 
 palate much more posteriorly, or nearer the first upper cheek-tooth, than in the remaining 
 skulls. 
 
 The palatine suture, which for some distance parallels the margin of the postpalatine 
 notch, meets the middle line of the palate between or behind the posterior ends of the 
 
44 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 third upper cheek-tooth. The postero-internal border of the alveolus for the fourth 
 upper cheek-tooth (5) may be somewhat elevated, causing the palate to become 
 transversely concave in this region. The posterior opening of the palatine canal opens 
 at the anterior end of an elongate pit. The latter is situated on the palatine shelf 
 bordering the postpalatine notch, and its anterior border is usually posterior to the last 
 superior tooth. The pit may, however, extend well along the inner side of the fourth 
 upper cheek-tooth (5). At the posterior end of the elongate pit a second opening leads 
 to the lateral cranial wall. The large foramina, extending in a row on either side of 
 the median suture of the palate, open directly into the palatine canals. 
 
 The postpalatine notch is always more acute than in the skull described by Reinhardt, 
 and apparently reached slightly farther forward in the Rancho La Brea species. Its 
 position is, however, well posterior to the fourth upper cheek-tooth. In Planops magnus, 
 Scott (1904, p. 326) states that “the posterior nares do not extend so far forward as in 
 most of the contemporary Gravigrada, the front margin being a little behind 5 . . .” 
 The pterygoid plates continue backward from the shelf, formed by palatine and in part 
 also by maxillary, along either side of the postpalatine notch. On these plates the inferior 
 border is broadly rounded, while in Megatherium it is angulate. In the Rancho La Brea 
 Nothrotherium the plates are very similar to those in Reinhardt’s specimen, and posteriorly 
 are seen to enlarge into great bullae, as in the latter form. These bullae differ from 
 those shown by Reinhardt in being pierced along their inner ventral wall by a large and 
 elongate orifice. Between the outer border of this opening and the inferior border of the 
 pterygoid plate a broad groove extends from the outer posterior to the inner anterior side, 
 diagonally across the inflated portion. The posterior border of the pterygoid may approach 
 the occipital condyles very closely, thus differing decidedly from Megalonyx. 
 
 The inflated pterygoids are perfectly preserved in skull 1800-11 (plate 2, fig. 2), in 
 which specimen they swell outward somewhat below the base of the zygoma and behind 
 the glenoid fossa. They are much better preserved than in the skull No. 1800-13 
 originally described. The skull in the Brazilian species described by Reinhardt possesses 
 also pterygoids that are inflated. In the Brazilian species, judging from the figures given 
 by Reinhardt, the inflated pterygoids do not possess internally an orifice, as in the 
 skulls of the Californian species. The pterygoid bullae in all skulls except one are 
 large. In No. 1800-4, which is the smallest of the 8 nearly complete skulls contained in 
 the Los Angeles Museum of History, Science, and Art, and evidently belongs to a 
 somewhat younger individual, the posterior inflation of the pterygoid is exceedingly 
 small and is restricted to the extreme posterior end of the pterygoid. The ventral 
 opening of the bullae is also small. Between the cavity and the descending plate the 
 pterygoid is thickened in this specimen. 
 
 In Hapaiops longiceps, according to Scott (1903, p. 185), “the pterygoids are narrow, 
 inconspicuous plates, closely applied to the descending processes of the alisphenoids, 
 and not in the least like the swollen, bulla-like bones of Nothrotherium, which Reinhardt 
 has described (’78, 336).” Structures comparable to the pterygoid bullae of Nothro¬ 
 therium and occupying a similar position are absent in Megalonyx. Lindahl (1893, 
 p. 7), however, records that in M. leidyi “capacious air sinuses extend backward in the 
 root of the pterygoid (pi. iii), and branch off from there forward into the alveolar 
 wall of the maxilla.” An extensive inflation of the pterygoid in skulls of the tree-sloths 
 Choloepus and Hemibradypus have been noted by Anthony. a With reference to the air 
 sinuses in the skull of the former genus, he states: 
 
 “Le sinus sphenoidal s’etend lateralement a l’int&ieur des pterygoides, se prolonge dans le temporal, 
 envahissant la base de l’apophyse zygomatique et venant presque au contact des prolongements lateraux du sinus 
 frontal. Le crane osseux pr6sente un orifice situ 6 a la face interne et anterieure de chaque pt6rygoide et qui 
 conduit dans ce sinus.” 
 
 ° It. Anthony. Itecherches anatomiques sur les bradypes arboricoles, etc. Ann. Sci. Nat. s6r. 9 (Zool.), vol. 9, pp. 
 157-285, plates 2-6, 1909. 
 
PLEISTOCENE MECALONYCIIINAE AND MYLODONTIDAE OF ItANCIIO LA BREA. 45 
 
 Noihrotherium therefore resembles Choloepus very closely, not only in the large size 
 of the pterygoid inflation, but also in the presence of an orifice perforating the inner 
 wall of the pterygoid. 
 
 The alisphenoid forms not only the roof of the pterygoid bulla, but the dorso-internal 
 wall as well. In some skulls an internal continuation of the vacuity of the pterygoid 
 extends toward the middle line. In skull 1800-5 this extension, lying above the ventral 
 surface of the basisphenoid, is separated from its fellow of the opposite side by a median 
 partition only 2 mm. in thickness. 
 
 The basisphenoid is concave transversely and tapers toward the anterior end, 
 where the pterygoid bullae approach each other most closely. 
 
 The basioccipital in most skulls is short, relatively much more so than in Megalonyx. 
 The tympanic, as seen in No. 1800-11, resembles Hapalops in the incomplete irregular 
 circle of bone which incloses ventrally the large external auditory meatus. Anteriorly 
 the tympanic may touch the dilated wall of the pterygoid. Posterior to the tympanic 
 is the stylo-hyal process with its ovoid depression. It is situated much closer to the 
 occipital condyle than in Megalonyx. The small condylar foramen lies in a pit and may 
 be entirely hidden by the border of the occipital condyle. In Hapalops, and especially 
 in Megalonyx, this foramen is well in front of the occipital condyle. 
 
 Fig. 11 .—Noihrotherium shas- 
 tense Sinclair. Vertical long¬ 
 itudinal section of skull No. 
 1800-6. X 0.33. n, nasal; 
 fr, frontal; s, frontal sinus; 
 pa, parietal; so, supraoccipi- 
 tal; per, periotic; bo, basiocci¬ 
 pital; bs, basisphenoid; ps, 
 presphenoid; pt, pterygoid; 
 opt., orifice of pterygoid sinus; 
 vo, vomer; pi, palatine; mx, 
 maxillary; mxt, maxillotur- 
 binal; et, ethmo-turbinals; 
 IX, X, XI, foramen lacerum 
 posterius; XII, condylar 
 foramen. 
 
 The paroccipital process, as in Hapalops, is inconspicuous, being much less prominent 
 than in Megalonyx. Between it and the stylohyal process is a deep groove, which reaches 
 the occiput and extends for a short way along its lateral border. The position of the 
 occipital condyles with reference to the pterygoid bullae has already been noted. The 
 condyles in ventral aspect are not obliquely placed, as in Megalonyx, and project 
 posteriorly much less than in that genus or in Megatherium. They may be widely 
 separated, as in No. 1800-4, or much closer together, as in No. 1800-7. By far the 
 greater articulating surface of the condyles is directed downward, forward, and outward. 
 Leidy states that in Megalonyx the articulating surface of the condyles is bent at about 
 the middle. In No. 1800-6 the dorsal portion of the outer border of the condyle is 
 curiously flattened by a rough and rather extensive surface. The foramen magnum 
 may be transversely oval in shape. It opens more downward than in Megalonyx. 
 
 The base of the zygoma is long, and the glenoid fossa is somewhat concave trans¬ 
 versely. The zygomatic process is quite like that in Hapalops, and differs from that in 
 Megalonyx in not projecting so far from the side of the skull. 
 
 As in Miocene genera, the occiput of the Rancho La Brea Noihrotherium is nearly 
 vertical or inclined slightly beyond the occipital condyles. The supraoccipital forms 
 usually an extensive forwardly inclined surface. In posterior view (plate 3, fig. 4) the 
 condyles are seen to be as a rule more widely separated than in Hapalops. In No. 
 1800-7 their position is more as in the latter genus. The posterior face of the condyles 
 is relatively smaller than in Hapalops. The dorsal lip of the foramen magnum is notched 
 
46 
 
 CENOZOIC GRAYIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 at the middle only in skulls 1800-13 and 1800-2. The median vertical ridge of the 
 occiput does not usually reach the foramen magnum. It is sharp below, but greatly 
 broadens dorsally on the surface of the supraoccipital. Where the vertical surface of 
 the occiput meets the forwardly inclined surface, a heavy, rugose, transverse ridge is 
 formed, as in Megalonyx. This ridge swings outward and downward and in some skulls 
 is directly connected with the posterior surface of the stylohyal process. The crest of 
 the occiput is not prominent, except along its lower end. 
 
 Skull 1800-6 is shown (fig. 11) in vertical longitudinal section. In contrast to the 
 skull of Mylodon, that of Nothrotherium shows the presence of few sinuses. This is 
 particularly noticeable in the parietal roof of the brain-chamber. A rather capacious 
 sinus (s) is formed in the frontal and extends from the median septum laterally and 
 ventrally. In the posterior region of the frontal the sinus overlies the front end of the 
 brain-case. Here it has considerable dorso-ventral extent, but narrows toward the outer 
 side. Farther forward the sinus reaches along the side of the ethmoids. The inflation 
 of the pterygoid has already been mentioned. 
 
 Table 6. —Measurements (in millimeters ) of cranium of Nothrotherium shastense. 
 
 
 No. 
 
 1800-2 
 
 No. 
 
 1800-3 
 
 No. 
 
 1800-4 
 
 No. 
 
 1800-5 
 
 No. 
 
 1800-6 
 
 No. 
 
 1800-11 
 
 Length from anterior end of maxillaries to posterior 
 
 
 
 
 
 
 
 end of occipital condyles. 
 
 320 
 
 313.6 
 
 302 
 
 330 
 
 
 309 
 
 Length from anterior end of nasals to posterior end of 
 
 
 
 
 
 
 
 supraoccipital. 
 
 339.8 
 
 333.3 
 
 320.4 
 
 352 
 
 339.5 
 
 336.3 
 
 Length of palate, from anterior end of maxillaries to 
 
 
 
 
 
 
 
 postpalatine notch. 
 
 141.7 
 
 131 
 
 127 
 
 144 
 
 
 x 135.4 
 
 Greatest width of palate anterior to first upper cheek- 
 
 
 
 
 
 
 
 tooth (2). 
 
 45.8 
 
 42.9 
 
 42.9 
 
 48 
 
 
 46.2 
 
 Width of palate between alveoli of second upper cheek- 
 
 
 
 
 
 
 
 tooth (3). 
 
 22 
 
 20.6 
 
 21.4 
 
 21.8 
 
 24 
 
 23.6 
 
 Greatest width of pterygoid bulla. 
 
 41.4 
 
 *37.9 
 
 17.4 
 
 *41.6 
 
 37.4 
 
 43.2 
 
 Greatest length of internal orifice of pterygoid bulla.... 
 
 37.6 
 
 36.7 
 
 21.7 
 
 
 30.5 
 
 38.2 
 
 Greatest width of internal orifice of pterygoid bulla... . 
 
 
 
 *8.5 
 
 
 12 
 
 14.4 
 
 Least distance between pterygoid bullae. 
 
 
 23.3 
 
 
 21.8 
 
 21.3 
 
 
 Mastoid width above stylohyal processes. 
 
 112.8 
 
 114 
 
 110.5 
 
 120.6 
 
 114 
 
 118.4 
 
 Greatest width across occipital condyles. 
 
 75.4 
 
 74.9 
 
 78.4 
 
 81.1 
 
 78.4 
 
 78 
 
 Transverse diameter of foramen magnum, internal 
 
 
 
 
 
 
 
 measurement. 
 
 33.5 
 
 32 
 
 30 
 
 36 
 
 31.8 
 
 29.6 
 
 Dorso-ventral diameter of foramen magnum, internal 
 
 
 
 
 
 
 
 measurement. 
 
 28.7 
 
 24.2 
 
 26.3 
 
 29.5 
 
 24.2 
 
 24.9 
 
 Height of occiput from plane of basioccipital to lamb- 
 
 
 
 
 
 
 
 doidal suture. 
 
 79.6 
 
 78.5 
 
 80.7 
 
 85 
 
 76.5 
 
 87.7 
 
 Greatest width of muzzle at anterior end. 
 
 70 
 
 73.9 
 
 65.5 
 
 73.4 
 
 73.4 
 
 78.2 
 
 Length of nasals. 
 
 114.5 
 
 111.9 
 
 130.8 
 
 122.4 
 
 134.3 
 
 114.7 
 
 Greatest width above orbits. 
 
 103.2 
 
 104.3 
 
 98.3 
 
 112.9 
 
 102.9 
 
 104.9 
 
 Least width behind orbit in region of coronal suture. . . 
 
 78.2 
 
 78.5 
 
 82 
 
 83.8 
 
 77 
 
 80 
 
 Width, measured across anterior ends of zygomatic 
 
 
 
 
 
 
 
 processes of squamosals. 
 
 153.8 
 
 
 
 
 
 153.5 
 
 Least distance between outer sides of pterygoid plates.. 
 
 61 
 
 
 
 
 
 56.4 
 
 Greatest height from ventral border of pterygoid plate 
 
 
 
 
 
 
 
 to highest elevation of frontals. 
 
 134.7 
 
 
 bso.s 
 
 
 
 139.9 
 
 
 
 
 
 
 
 
 1 Approximate. 
 
 The longitudinal section through the skull brings to view also the structures lying 
 in the nasal region. The mesethmoid forms a thin plate which divides the nasal cavity 
 into two halves. The mesethmoid extends dorsally over the ethmoidal bones of each 
 side. Posteriorly the mesethmoid is fused above with that portion of the frontal lying 
 below the frontal sinus. The mesethmoid is joined below with the vomer. Seven 
 endoturbinals ( et ) are found in Nothrotherium shastense. The first endoturbinal forms 
 a large scroll which extends forward in the nasal cavity and covers anteriorly the maxillo- 
 turbinal. 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 47 
 
 MANDIBLE. 
 
 The most striking characteristic of the lower jaw of Nothrotherium is its great 
 similarity to that of Hapalops. The former genus in this regard is decidedly closer to 
 the Miocene form than to its Pleistocene contemporary Megalonyx. In general structure 
 of mandible N othr other ium bears the same relation to Hapalops as Megalonyx does to 
 Eucholoeops. The lower jaw of an apparently adult individual (No. 1801-7) of the 
 Rancho La Brea species is approximately twice as large as that in Hapalops longiceps . 
 Although the mandible may very closely approach Megalonyx in total length, it differs 
 from that genus in its very decided slenderness. 
 
 Table 7. —Measurements (in millimeters ) of mandible 1 of N othrotherium shastense. 
 
 
 No. 
 
 1801-5 
 
 No. 
 
 1801-7 
 
 No. 
 
 1801-1 
 
 No. 
 
 1801-2 
 
 No. 
 
 2 1801-3 
 
 No. 
 
 1801-R-4 
 
 No. 
 
 1801-L-4 
 
 Length from anterior end of symphysis to 
 
 
 
 
 
 
 
 
 posterior end of condyle. 
 
 275 
 
 277.1 
 
 3 260.7 
 
 284.4 
 
 3 267 
 
 .... 
 
 .... 
 
 Distance from anterior end of symphysis 
 
 
 
 
 
 
 
 
 to anterior border of alveolus for 2. 
 
 100.9 
 
 106.2 
 
 3 96.8 
 
 104.7 
 
 100.3 
 
 • • • • 
 
 .... 
 
 Distance from posterior border of alveolus 
 
 
 
 
 
 
 
 
 for 4 to posterior end of angle . 
 
 121.2 
 
 134.6 
 
 3 120 
 
 124.3 
 
 117.1 
 
 .... 
 
 .... 
 
 Greatest length of symphysis. 
 
 90.4 
 
 90.3 
 
 3 82.9 
 
 90.2 
 
 87 
 
 .... 
 
 .... 
 
 Greatest depth between 2 and 3 measured 
 
 
 
 
 
 
 
 
 normal to tooth row. 
 
 56.6 
 
 60.6 
 
 54.2 
 
 56.1 
 
 53 
 
 61.3 
 
 61.6 
 
 Least height posterior to 4. 
 
 43 
 
 47.5 
 
 42 
 
 44.8 
 
 38.6 
 
 46.8 
 
 47.5 
 
 Greatest height from angle to coronoid 
 
 
 
 
 
 
 
 
 process. 
 
 106.6 
 
 121.4 
 
 
 3 113.8 
 
 3 101.7 
 
 
 
 Height of condyle above ventral border of 
 
 
 
 
 
 
 
 
 angular process. 
 
 79.7 
 
 
 
 
 
 26.4 
 
 26.8 
 
 Thickness of horizontal ramus at 3. 
 
 27.2 
 
 29.3 
 
 26.7 
 
 28.2 
 
 25.3 
 
 
 
 Distance between inner alveolar borders 
 
 
 
 
 
 
 
 
 of fourth cheek-teeth. 
 
 33 
 
 30.7 
 
 
 24.3 
 
 
 .... 
 
 .... 
 
 1 Where association of mandible with cranium is in doubt, the former is listed under a separate number. 
 
 2 This mandible may belong to cranium 1800-4. 
 
 3 Approximate. 
 
 In both Hapalops and Nothrotherium the predentary portion of the jaw is produced 
 anteriorly into a long beak, which in the latter genus is also deeply concave. The 
 edentulous beak or spout (plate 4, fig. 2, and fig. 12a) is much more elongate than in 
 Megalonyx. The anterior end may be rather acutely tapering, as in No. 1801-7, or 
 rounding, as in No. 1801-2. 
 
 As pointed out by Scott, the lower border of the ramus in Nothrotherium is less 
 sinuous than in Hapalops. In the former genus it forms an even convexity which reaches 
 its maximum development below the third inferior tooth. This border is more strongly 
 convex than in Megalonyx, but decidedly less so than in Megatherium. The inferior 
 border posterior to the tooth series and reaching to the end of the angular process is 
 less decidedly concave in Nothrotherium than in Hapalops. 
 
 The symphyseal keel is long and sharp, but not as prominent as in Megalonyx. 
 Usually but a single mental foramen is present. In No. 1801-2, however, a small opening 
 is situated slightly above and posterior to the mental foramen of the right side. The 
 horizontal limb of the ramus (plate 4, fig. 1) is noticeably of less height, especially in 
 the anterior region, and less robust than in Megalonyx. With the disappearance of the 
 first tooth in Nothrotherium, the deeply concave_outer wall of the mandible between the 
 caniniform tooth (I) and the first cheek-tooth (2) occurring in Hapalops and Megalonyx 
 becomes greatly reduced and may be entirely absent, as in No. 1801-5. There are 
 frequently two foramina representing the postero-external opening of the dental canal 
 in the Rancho La Brea species. In No. 1801-7 two foramina are present on the left 
 ramus and one on the right, an occurrence similar to that in a mandible of Hapalops Ion - 
 
48 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 giceps (No.45523, Prin. Univ. Coll.), while in No. 1801-2, from Rancho La Brea, there 
 are two foramina on each side. The anterior foramen is situated on the outer surface 
 of the ramus, opposite the last inferior tooth, and the second opening is at or slightly 
 above the base of the ascending process. 
 
 Fig. 12. — A, Nothrotherium shastense Sinclair. A 1 , Pronothrotherium typicum 
 Ameghino. A 2 , Hapalops longiceps Scott. Dorsal view of mandible, 
 
 X 0.33. A 1 , after Rovereto; A 2 , after Scott. 
 
 The coronoid process arises well to the outer side of the last cheek-tooth (4), as in 
 Hapalops, but is relatively smaller than in the latter. It may be broad, as in No. 1801-2, 
 or rather narrow. The sigmoid notch, due to the high position of the condyle above the 
 tooth row in Nothrotherium, is relatively much smaller than in the Miocene genus. In 
 
 Table 8.— Measurements {in millimeters) of superior dentition of Nothrotherium shastense. 
 
 
 No. 
 
 1800-2 
 
 No. 
 
 1800-3 
 
 No. 
 
 1800-4 
 
 No. 
 
 1800-5 
 
 No. 
 
 1800-6 
 
 No. 
 
 1800-9 
 
 No. 
 
 1800-10 
 
 No. 
 
 1800-11 
 
 No. 
 
 1800-12 
 
 Greatest length of superior molar 
 
 
 
 
 
 
 
 
 
 
 series alveolar measurements. . 
 
 61.6 
 
 57.5 
 
 61.5 
 
 65.4 
 
 62.3 
 
 .... 
 
 . . . . 
 
 62.8 
 
 
 2, greatest antero-posterior diam- 
 
 
 1 12.3 
 
 
 
 
 
 
 
 
 eter. 
 
 12 
 
 12.1 
 
 12.4 
 
 
 9.9 
 
 10.4 
 
 12 
 
 9.7 
 
 2, greatest transverse diameter. . 
 
 12.9 
 
 x 13.7 
 
 13 
 
 14.3 
 
 
 11.9 
 
 10.5 
 
 13 
 
 11.2 
 
 3, greatest antero-posterior diam- 
 
 
 J 14 
 
 
 
 
 
 
 
 
 eter. 
 
 
 13 
 
 13.8 
 
 
 .... 
 
 . . . . 
 
 13.3 
 
 11 
 
 3, greatest transverse diameter. . 
 
 
 1 16.8 
 
 16.9 
 
 18.9 
 
 
 .... 
 
 , , . . 
 
 16.8 
 
 14.4 
 
 4, greatest antero-posterior diam- 
 
 
 
 
 
 
 
 
 
 
 eter. 
 
 13.2 
 
 13 
 
 12.8 
 
 12.1 
 
 
 . . . . 
 
 11.2 
 
 13 
 
 10.8 
 
 4, greatest transverse diameter. . 
 
 17 
 
 17 
 
 17 
 
 18.3 
 
 
 
 14.4 
 
 17.2 
 
 13.5 
 
 5, greatest antero-posterior diam- 
 
 
 *7.8 
 
 
 
 
 
 
 
 
 eter. 
 
 8.7 
 
 7.3 
 
 8.8 
 
 
 7.5 
 
 7.3 
 
 7.6 
 
 7.2 
 
 5, greatest transverse diameter. . 
 
 13.6 
 
 ] 12.9 
 
 13.8 
 
 16 
 
 
 11.3 
 
 11.5 
 
 14.4 
 
 10.5 
 
 1 Approximate. 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 49 
 
 No. 1801-7 the condyle in dorsal view has much the shape seen in Hapalops longiceps. 
 In No. 1801-2 and No. 1801-1 the condyle has a greater oblique diameter and is more 
 slender than in No. 1801-7. The ventral notch between condyle and angle is relatively 
 wider than in Hapalops , while the angular process in some specimens from the asphalt 
 extends posteriorly as a more slender hook than in H. longiceps. The angle of the jaw 
 is much more slender in Nothrotherium than in Megalonyx. In M. jeffersonii the coronoid 
 and angle are more nearly equally spaced on either side of the condyle than in the Rancho 
 La Brea genus. The inferior border of the angle is deflected inward, forming a more 
 pronounced shelf than in Megalonyx. 
 
 Table 9. —Measurements (in millimeters) of inferior dentition of Nothrotherium shastense. 
 
 
 No. 
 
 1801-5 
 
 No. 
 
 1801-7 
 
 No. 
 
 1801-1 
 
 No. 
 
 1801-2 
 
 No. 
 
 1 1801-3 
 
 No. 
 
 1801-R-4 
 
 No. 
 
 1801-L-8 
 
 Greatest length of inferior molar series alveolar 
 measurements. 
 
 57 
 
 54.7 
 
 51.7 
 
 59 
 
 51.7 
 
 49.4 
 
 48.5 
 
 2, greatest antero-posterior diameter. 
 
 13.6 
 
 
 13.9 
 
 13.6 
 
 
 2 12.5 
 
 hi.8 
 
 2, greatest transverse diameter. 
 
 16.5 
 
 
 15.9 
 
 17.6 
 
 • • • • 
 
 2 16.3 
 
 h6.2 
 
 3, greatest antero-posterior diameter. 
 
 14.2 
 
 13.5 
 
 2 14.7 
 
 14.1 
 
 • • • • 
 
 13.3 
 
 13.4 
 
 3, greatest transverse diameter. 
 
 17 
 
 17.5 
 
 16 
 
 18.4 
 
 • • • • 
 
 16.4 
 
 16.4 
 
 4, greatest antero-posterior diameter. 
 
 14 
 
 15 
 
 
 13.4 
 
 14 
 
 14.5 
 
 x 15 
 
 4, greatest transverse diameter. 
 
 15.6 
 
 15.8 
 
 
 16.2 
 
 15 
 
 15 
 
 14.8 
 
 1 This specimen may belong to cranium 1800-4. 1 Approximate. 
 
 DENTITION. 
 
 Dental formula: 3 _ 3 
 
 In the loss of the anterior or caniniform teeth, Nothrotherium differs markedly from 
 known Miocene Megalonychinae and Megatheriinae as well as from the Pleistocene 
 genus Megalonyx. All the teeth are considerably smaller than in Megalonyx. The 
 external layer of cement on the teeth may be only as thick relatively as in Hapalops. 
 
 Superior series .—The tooth-rows extend nearly parallel, the palate increasing 
 slightly in width between the fifth cheek-teeth. The teeth are more closely spaced 
 than in Miocene genera. In the Rancho La Brea species the largest interspace occurs 
 between the alveoli of the third (4) and fourth (5) cheek-teeth. Reinhardt states that 
 in the Brazilian species of Nothrotherium the teeth are equally spaced. 
 
 As in the various species of Santa Cruz ground-sloths, the differentiation of the 
 superior molariform teeth into a small 2, much larger 3 and 4, and a 5, greatly compressed 
 antero-posteriorly, occurs also in Nothrotherium, though the differences between 2 and 
 3 or 4 is not as marked in the former as in the latter. The teeth in Nothrotherium are 
 noticeably more complex. This is indicated by the presence of distinct, median vertical 
 grooves on the outer faces of 2, 3, and 4, and to a less extent also on the inner faces. 
 Even 5, which is much more compressed antero-posteriorly than the other superior 
 teeth, shows a distinct groove on the outer face and a much more faintly developed one 
 on the inner. Nothrotherium, therefore, differs from Megalonyx, in which such distinct 
 lateral grooves are lacking on these teeth. In Hapalops indifferens the vertical grooving 
 on the outer faces of 2 , 3, and 4 has been noted by Scott in some specimens. In 
 describing the corresponding teeth, as well as 5, in H. elongatus, Scott (1903, p. 218) 
 states that “the vertical grooving of these teeth appears to be very capriciously present 
 or absent. ” Judging from the appearance of the occlusal surface, the outer compact 
 layer of dentine in all the superior teeth of Nothrotherium is best developed on the ante¬ 
 rior and posterior surfaces and least developed on the outer and inner sides. In 2, 3, 
 and 4, this layer is slightly thicker on the outer than on the inner surface. In 5 the 
 layer is thickest on the posterior face, while on the outer and inner faces it is of equal 
 thickness. 
 
50 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 The first cheek-tooth (2) is smaller than 3 and 4. Contrasted with the latter 
 teeth, 2 is relatively narrow transversely in Nothrotherium. It is relatively longer antero- 
 posteriorly than in the Hapalops. The tooth may be trapezoidal in shape, with ante¬ 
 rior and posterior faces parallel, as in No. 1800-4, or these faces may converge to the outer 
 side, as in skull 1800-11. The anterior and posterior faces are nearly flat and the angles 
 are well rounded; 2 thus differs from the corresponding tooth in Megalonyx, which Leidy 
 (1855, p. 17) has described as follows: 
 
 “The second upper molar in section (PI. xvi, Fig. 9, d ), is quadrate with rounded angles. Its inner and 
 posterior sides are the greater, and are nearly equal; and the remaining sides are also nearly equal, and are 
 planes. The inner side is slightly convex, and the posterior side is nearly a plane and is directed obliquely 
 outward.” 
 
 Fig. 13 .—Nothr other ium shastense Sinclair. Superior Fig. 14 .—Nothr other ium shastense Sinclair. Inferior 
 dentition. Lateral and occlusal views. X 1. dentition. Lateral and occlusal views. X 1. 
 
 On the occlusal surface the transverse valley between the anterior and posterior 
 crests opens most broadly on the inner side in No. 1800-11. In No. 1800-4, on the other 
 hand, where the transverse crests are parallel, the valley is of nearly the same width 
 throughout. 
 
 Table 10 .—Measurements (in millimeters ) of endocranial casts. 
 
 
 Nothrotherium 
 No. 1800-1. 
 
 Hapalops. 
 
 Greatest length from anterior end of cerebrum to posterior 
 end of cerebellum. 
 
 102.4 
 
 67.7 ' 
 
 Greatest transverse diameter of anterior end of cerebrum. . . 
 
 67 
 
 x 39.6 
 
 Greatest transverse diameter of posterior end of cerebrum. . 
 
 84.7 
 
 x 49.2 
 
 Transverse diameter of cerebellum. 
 
 66.6 
 
 40 
 
 Height from infundibulum to dorsal surface of cerebrum.... 
 
 61.8 
 
 *32.4 
 
 1 Approximate. 
 
 The second and third superior cheek-teeth (3 and 4) are very similar in general 
 shape; 3 may be slightly larger than 4, but the corresponding measurements of these 
 teeth are usually very close. In the greater number of species of Hapalops, 3 is the largest 
 of the molariform series, and this is true in general for the Santa Cruz Megalonychinae. 
 In both teeth the anterior face is broadly convex and the posterior face concave. The 
 anterior face rounds gradually to the external side; in other words, does not form a 
 distinct angle with the outer side, as in other cheek-teeth. The teeth do not narrow so 
 much externally as in Megalonyx jeffersonii, in which respect they are more like M. 
 leidyi. They lack the typical quadrate cross-section characteristic of the corresponding 
 teeth of Megatherium. In 3 and 4 the anterior crest of the occlusal surface is decidedly 
 beveled in front, and the cutting-edge of this chiseled surface is usually worn into a 
 crescent. The transverse valley is worn most broadly on the inner side. In both teeth 
 the postero-internal angle of the occlusal surface is subjected to greatest wear. In 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 51 
 
 several loose teeth of Nothrotherium from the asphalt beds, referable to 3 or 4, the outer 
 face is seen to be noticeably concave in its longitudinal extent. This character apparently 
 distinguishes the teeth from 3 and especially from 2. 
 
 Table 11. —Vertebral formulae. 
 
 
 C. 
 
 Th. 
 
 L. 
 
 S. 
 
 Cd. 
 
 Nothrotherium shastense. 
 
 7 
 
 17? 
 
 3 
 
 5 
 
 21 
 
 Hapalops longiceps (after Scott). 
 
 7 
 
 22 
 
 3 
 
 6 
 
 20 ± 
 
 Hapalops elongatus (after Scott). 
 
 7 
 
 21-22 
 
 3-4 
 
 5-6 
 
 20 ± 
 
 Megatherium americanum (after Flower). . . 
 
 7 
 
 16 
 
 3 
 
 5 
 
 17 
 
 Myrmecophaga jubata (after Flower). 
 
 7 
 
 15 
 
 3 
 
 5 
 
 29 
 
 The fourth or last cheek-tooth (5) resembles in its antero-posterior compression 
 the corresponding tooth in Hapalops and in Megalonyx. It is approximately only one- 
 half as long in that direction as the remaining superior cheek-teeth of Nothr otherium. 
 The posterior face of 5 in the species from the asphalt beds is very broadly concave 
 and the concavity may become quite deep, as in a young individual (No. 1800-12). 
 The nearly flat anterior face, which is of less width than the posterior face, may have 
 a median strip defined by a slight groove on each side. The inner face may have a faint 
 vertical groove near the anterior border, while the outer face is characterized by a well- 
 marked median groove. On the occlusal surface there is no distinct valley formed, 
 but the anterior edge is worn much lower than the posterior. This tooth differs decidedly 
 from 5 in Megalonyx jeffersonii, which Leidy (1855, p. 18) describes as follows: 
 
 “The last molar is a smaller tooth than the preceding pair, but has nearly the same form in a reversed 
 position, the base of the triangular section (PL xvi, Fig. 9, a) being outward. Its posterior side is transverse 
 and is slightly concave; and the anterior side is convex and directed obliquely inward.” 
 
 Inferior series. —The tooth-rows are relatively as far separated as in Hapalops 
 longiceps. They are very nearly parallel or diverge slightly posteriorly. In the 
 Pleistocene genus the characteristic features of the three posterior teeth in Hapalops 
 have been accentuated. _ Scott (1903, p. 182) describes these teeth in H. longiceps as 
 follows: “ ... 2 and 3 are transverse and rectangular, and their raised margins are 
 quite deeply notched on the inner and outer sides by the transverse valley; 4 is sub- 
 cylindrical, as in most of the other species.” 
 
 In the first and second inferior cheek-teeth (2 and 3), the denser, outer layer of 
 dentine of the tooth-crown is least developed along the outer and inner sides. It is 
 perhaps less developed on the outer than on the inner side. The layer is thicker on the 
 
 Table 12. —Measurements {in millimeters ) of atlas. 
 
 
 No. 
 
 1802-1 
 
 No. 
 
 1802-2 
 
 No. 
 
 1802-3 
 
 No. 
 
 1802-4 
 
 No. 
 
 1802-5 
 
 Greatest transverse width across lateral wings. 
 
 140.4 
 
 153.3 
 
 145.8 
 
 130 
 
 145.3 
 
 Greatest antero-posterior diameter. 
 
 63.3 
 
 65.4 
 
 65.5 
 
 63 
 
 69.7 
 
 Antero-posterior diameter of dorsal arch over neural canal 
 
 
 
 
 
 
 along median line. 
 
 33.8 
 
 31.9 
 
 41.4 
 
 29.2 
 
 41.2 
 
 Antero-posterior diameter of ventral arch under neural canal 
 
 
 
 
 
 
 along median line. 
 
 24.4 
 
 23.6 
 
 25.3 
 
 21 
 
 24.6 
 
 Distance from antero-external border to lateral facet for axis. 
 
 43.7 
 
 48.3 
 
 45.7 
 
 39.4 
 
 43.6 
 
 Greatest transverse distance between posterior borders of 
 
 
 76 
 
 
 
 
 facets for axis. 
 
 71.1 
 
 78.8 
 
 69.5 
 
 74.7 
 
 Least transverse distance between anterior borders of facets 
 
 
 
 
 
 
 for axis. 
 
 30.8 
 
 31.3 
 
 33.8 
 
 28.4 
 
 27.9 
 
 Greatest height of neural canal at posterior end, internal 
 
 
 
 
 
 
 measurement. 
 
 44 
 
 44 
 
 51.2 
 
 43.2 
 
 46.8 
 
52 
 
 CENEZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 anterior and posterior faces of the crown, and appears slightly better developed on the 
 posterior face. The thickness of this more compact layer of dentine determines in 
 part the structure of the occlusal surface. The outer and inner walls, which are thinnest, 
 resist wear less, and are worn down more rapidly than the anterior and posterior walls. 
 The latter are therefore prominent on the wearing surface as two transverse ridges with 
 intervening valley excavated in the softer inner material of the crown. The type of 
 wearing surface thus obtained resembles very much that in the larger teeth of 
 Megatherium, but in that genus the outer compact dentine is thinner and the cement 
 (on the anterior and posterior faces) relatively much thicker. In the Rancho La Brea 
 Nothrotherium the cement is more evenly distributed on the crowns of the teeth, with a 
 moderate increase on the anterior and posterior faces in some specimens. In Megatherium 
 there is an excessive thickening of the cement on the anterior and posterior faces. 
 
 Table 13 .—Measurements {in millimeters ) of axis. 
 
 
 No. 
 
 1803-1 
 
 No. 
 
 1803-2 
 
 No. 
 
 1803-3 
 
 No. 
 
 1803-4 
 
 No. 
 
 1803-5 
 
 No. 
 
 1803-6 
 
 No. 
 
 1803-7 
 
 Greatest length along median ventral line. 
 
 70.9 
 
 69.8 
 
 81.6 
 
 82 
 
 82.6 
 
 *72.8 
 
 75.3 
 
 Greatest vertical height. 
 
 73.3 
 
 77 
 
 81.4 
 
 77.6 
 
 76 
 
 80.4 
 
 73.7 
 
 Least width behind articulating surfaces for atlas. . . 
 
 58.4 
 
 60.8 
 
 65 
 
 *65 
 
 *60 
 
 66.4 
 
 63 
 
 Dorso-ventral diameter of centrum across posterior 
 
 
 
 
 
 
 
 
 face. 
 
 21.6 
 
 22 
 
 24.8 
 
 22.8 
 
 23 
 
 29.5 
 
 23.7 
 
 Greatest transverse width of centrum across poste- 
 
 
 
 
 
 *34.5 
 
 
 
 rior face. 
 
 37.2 
 
 34.4 
 
 39.3 
 
 39.2 
 
 37.3 
 
 35.6 
 
 Greatest transverse diameter of neural canal at ante- 
 
 
 
 
 
 
 
 
 rior end. 
 
 35.8 
 
 30.6 
 
 35.3 
 
 35.4 
 
 35.3 
 
 35.5 
 
 31.8 
 
 Greatest width across posterior ends of transverse 
 
 *82.8 
 
 
 *98.2 
 
 ] 93 
 
 
 404.6 
 
 
 processes. 
 
 89 
 
 
 96.7 
 
 Distance from anterior end of neural spine to poste- 
 
 
 
 
 
 
 
 
 rior end of neural arch. 
 
 57.2 
 
 49 
 
 68.6 
 
 66.6 
 
 67 
 
 62 
 
 57.7 
 
 Width across postzygapophyses. 
 
 62.4 
 
 58.9 
 
 71.1 
 
 65.2 
 
 66.3 
 
 66 
 
 70.3 
 
 Least distance from antero-lateral border of neural 
 
 
 
 
 
 
 
 
 canal to notch below postzygapophysis. 
 
 24.9 
 
 20 
 
 26 
 
 23.7 
 
 26.3 
 
 22.6 
 
 27.3 
 
 Greatest width across posterior ends of lateral facets 
 
 
 
 
 
 
 
 
 for atlas. 
 
 70.9 
 
 69.3 
 
 72 
 
 73 3 
 
 72 4 
 
 75.7 
 
 71.9 
 
 Distance from anterior border of lateral facet for 
 
 
 
 
 
 
 
 
 atlas to posterior end of transverse process. 
 
 68 
 
 66.8 
 
 76.3 
 
 76.9 
 
 
 66.3 
 
 72 
 
 1 Approximate. 
 
 The complete length of the crown of 2, 3, and 4 is exposed in the somewhat broken 
 specimen, No. 1801-1. The teeth are seen to reach the inferior walls of the ramus, which 
 is very thin. In this ramus the inferior teeth diverge upward in their course from the 
 base to the alveolar border. In longitudinal extent the crown of 2 is slightly concave 
 posteriorly; 3 is practically straight posteriorly, but slightly concave laterally; 4 evidently 
 sloped posteriorly; 3, and especially 2, exhibit, then, a different longitudinal curvature 
 from that shown by the third and fourth superior teeth. 
 
 In 2 (fig. 14) the transverse width of the anterior half is distinctly less than that 
 of the posterior half. The transverse axis of the tooth is oblique to the long axis of the 
 tooth-row. This tooth differs from the corresponding one in Hapalops and Megalonyx 
 in the presence of median vertical grooves on the inner and outer faces. The anterior 
 face is broadly concave, while the posterior face is correspondingly convex. On the 
 occlusal surface the anterior transverse ridge is nearly straight and may be beveled in 
 front, while the posterior is crescentic and is beveled behind. 
 
 The second inferior cheek-tooth (3) is slightly larger than 2, and narrows more 
 toward the inner side. It is transversely placed with reference to the long axis of the 
 tooth-row. In occlusal view it resembles somewhat the third and fourth superior teeth, 
 but is in reverse position. If the tooth is oriented so that the posterior margin of the 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 53 
 
 occlusal surface corresponds to the anterior margin of the third or fourth superior tooth, 
 the former can be distinguished by rounding less gradually toward the narrow lateral 
 side. In other words, the third inferior tooth has all four angles better defined than those 
 in the third and fourth superior teeth. This serves also to distinguish 2 from either 3 
 or 4. The vertical grooves of the lateral faces appear, as a rule, to be more sharply 
 defined than in 2, while the anterior half of the tooth is not as distinctly smaller trans¬ 
 versely as the posterior half. The second inferior cheek-tooth (3) differs from the 
 corresponding tooth of Hapalops and Megalonyx in the presence of well-marked lateral 
 grooves. The anterior and posterior faces resemble those in 2. 
 
 Table 14 .—Measurements (in millimeters ) of third cervical vertebra. 
 
 
 No. 
 
 1804-1 
 
 No. 
 
 1804-2 
 
 No. 
 
 1804-3 
 
 No. 
 
 1804-4 
 
 No. 
 
 1804-5 
 
 No. 
 
 1804-7 
 
 No. 
 
 1804-16 
 
 Length of centrum measured over ventral 
 surface. 
 
 46.7 
 
 49.7 
 
 48.3 
 
 
 
 46.8 
 
 
 Width of centrum measured over anterior 
 face and between inner borders of verte- 
 brarterial canals. 
 
 39.8 
 
 40.2 
 
 41.4 
 
 44.5 
 
 34.8 
 
 35.3 
 
 36.3 
 
 Depth of centrum measured over anterior 
 face and normal to dorsal surface. 
 
 17.1 
 
 15 
 
 17.8 
 
 
 17.6 
 
 14.9 
 
 17.3 
 
 Height from ventral border of posterior face 
 of centrum to tip of neural spine. 
 
 75.9 
 
 71.8 
 
 73.6 
 
 x 74 
 
 '73 
 
 71.3 
 
 x 67 
 
 Width across outer sides of prezygapophyses. 
 
 86.8 
 
 84.4 
 
 91.8 
 
 81 
 
 74.4 
 
 83 
 
 78.7 
 
 Greatest width across postzygapophyses.... 
 
 67.9 
 
 6.3 
 
 70.8 
 
 64.6 
 
 61 4 
 
 71 4 
 
 61 4 
 
 Greatest width of posterior side of neural 
 spine. 
 
 
 22 
 
 26.2 
 
 23.9 
 
 18.9 
 
 24.7 
 
 19.2 
 
 Greatest width across posterior ends of ven¬ 
 tral wings. 
 
 104.5 
 
 111.5 
 
 109.2 
 
 
 x 58.5 
 
 >6l" 
 
 ‘61 
 
 Length of ventral wing. 
 
 63 
 
 65,2 
 
 68.8 
 
 .... 
 
 1 Approximate. 
 
 The last inferior tooth (4), which in Hapalops is more or less cylindrical in shape, 
 retains the . rounded internal side in N othrotherium. It thus agrees with Megalonyx in 
 differing markedly in shape from either 2 or 3. This tooth in Megatherium , although 
 smaller and relatively longer antero-posteriorly than 2 and 3, is essentially of same 
 shape as the other inferior teeth. The transverse axis of 4 in Nothrotherium is directed 
 obliquely to the long axis of the tooth-row, the genus thus resembling Hapalops and 
 Megalonyx. On the outer flattened face of this tooth a median vertical groove may be 
 slightly defined, as in No. 1801-7, or well defined, as in No. 1801-2. In the presence of 
 this groove Nothrotherium differs from Hapalops and possibly from Megalonyx. The outer 
 anterior and posterior corners of 4 are well rounded. The denser layer of dentine is best 
 developed on the inner and posterior sides and least on the outer side. The border of the 
 occlusal surface is least worn at the middle of the inner side and at the postero-external 
 corner. Between the resulting prominences the border is beveled posteriorly. In Megal¬ 
 onyx the occlusal surface of the corresponding tooth is somewhat similarly worn. It is 
 described by Leidy (1855, p. 19) as having “a transverse valley, whose boundaries are 
 most prominent at the antero-internal and postero-external angles.” 
 
 ENDOCRAN1AL CAST. 
 
 In plate 3, figures 1, 2, and 3, are shown the dorsal, lateral, and ventral views 
 respectively of the endocranial cast of Nothrotherium shastense, and figures 4, 5, and 6 
 give the comparable views of the endocranial cast of Hapalops sp. In the preparation 
 of these casts and of that of Mylodon harlani, the cranium of each form was sectioned 
 in the sagittal plane and a glue cast made of the brain-chamber and of adjacent openings 
 for the principal cranial nerves and blood-vessels. A glue mold was then made from the 
 
54 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 glue cast and from this mold the final casts of plaster of paris were obtained. By the 
 application of this method satisfactory results were secured by Mr. J. W. Lytle at the 
 Los Angeles Museum. A skull kindly furnished by Dr. W. J. Sinclair was used for the 
 endocranial cast of Hapalops. The glue cast of the cranial cavity expresses details of 
 structure more clearly than the plaster cast and is therefore preferable to the latter in 
 a study of the surface features. 
 
 The endocranial cast of Nothrotherium is distinctly larger than that of Hapalops, but 
 is somewhat smaller than that of Mylodon harlani (compare plate 3 with plate 26). 
 The surface of the cerebrum is more richly convoluted in the Pleistocene megalonychid 
 ground-sloth than in the Miocene form. The forward portion of the cerebrum shows 
 also a fuller development than in Hapalops. When viewed from the dorsal side the 
 
 A' 
 
 Fig. 15. — A, Nothrotherium shastense Sinclair. A 1 , Hapalops sp. Views showing relation of size of 
 
 endocranium to that of skull. X 0.33. 
 
 cerebellum in both Nothrotherium and Hapalops is seen to be separated by a distinct 
 cleft or fissure from the cerebrum. The two lateral and the median lobes of the 
 cerebellum are discernible on the endocranial cast of the Pleistocene genus, but are not 
 so pronounced as in Hapalops. In a lateral view the cerebellum appears to be better 
 defined and relatively larger in Nothrotherium and Hapalops than in Mylodon harlani. 
 
 ELEMENTS OF HYOID ARCH. 
 
 Only the stylohyal and basihyal elements are available of the chain of bones in the 
 hyoid apparatus of Nothrotherium. 
 
 The stylohyal, as contrasted with that of Mylodon, is of course much more slender. 
 The shaft is straighter, with the lateral surface flattened and the median surface convex; 
 the proximal third is expanded, but the posterior end of this portion is not so prominently 
 developed above the anterior end and is relatively thicker than in Mylodon. 
 
 The basihyal, with which the thyrohyals are ankylosed, differs from the corresponding 
 element in Megalonyx jeffersonii in decidedly less prominent and more widely separated 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 55 
 
 surfaces for ceratohyals. In No. 1541 L. A. M. the diverging arms do not taper to the 
 free end, but broaden at the tips, and this broadening is even more distinct in Nothro- 
 therium than in Mylodon. The length of the arm is 51 mm. as contrasted with 
 approximately 68 mm. for a similar measurement in M. jeffersonii. The facets borne 
 by the arms are relatively large. When viewed from in front the inner surfaces are 
 seen to be directed more anteriorly than in the element of M. jeffersonii and also in that 
 of Mylodon harlani. 
 
 The measurements (in millimeters) are as follows: 
 
 Stylohyal No. 1540: Greatest length, 97.4; width of proximal expansion, 31.3. 
 
 Basihyal No. 1541: Greatest width across extremities of diverging arms, 61.8; greatest width over facets 
 for ceratohyals, 17. 
 
 Table 15 .—Measurements (in millimeters ) of fourth cervical vertebra. 
 
 
 No. 
 
 1805-1 
 
 No. 
 
 1805-2 
 
 No. 
 
 1805-3 
 
 No. 
 
 1805-4 
 
 No. 
 
 1805-5 
 
 No. 
 
 1805-6 
 
 Length of centrum measured over ventral surface. 
 
 47.5 
 
 48 
 
 J 39 
 
 To 
 
 49.4 
 
 49 
 
 Width measured over anterior face of cen¬ 
 trum and between inner borders of vertebrar- 
 
 
 
 
 
 
 
 terial canals. 
 
 39.7 
 
 39.3 
 
 37 
 
 36.7 
 
 42 
 
 40.8 
 
 Depth of centrum measured over anterior face 
 
 
 
 
 
 
 
 and normal to dorsal surface. 
 
 17.4 
 
 17.4 
 
 18 
 
 18 
 
 15.1 
 
 18.4 
 
 Height from ventral border of posterior face of 
 
 
 
 J 67 
 
 1 65 
 
 
 
 centrum to tip of neural spine. 
 
 72.8 
 
 78 
 
 76.9 
 
 78 
 
 Width across outer sides of prezygapophyses. . 
 
 94.3 
 
 89.4 
 
 79.2 
 
 85 
 
 87.8 
 
 X 91 
 
 Greatest width across postzygapophyses. 
 
 65.7 
 
 66.4 
 
 62.9 
 
 67.3 
 
 64.1 
 
 74 
 
 Greatest width of posterior side of neural spine.. 
 
 26.5 
 
 23.9 
 
 16.6 
 
 
 20.4 
 
 28.2 
 
 Greatest width across posterior ends of ventral 
 
 bio 
 
 
 
 ! 97 
 
 
 
 wings. 
 
 102.6 
 
 .... 
 
 111.4 
 
 109.8 
 
 Length of ventral wing measured along lower 
 
 
 
 
 
 
 
 border. 
 
 67.7 
 
 56.8 
 
 
 48.2 
 
 64.6 
 
 68 
 
 1 Approximate. 
 VERTEBRAE. 
 
 We note from table 11 that Nothrotherium resembles Hapalops in number of 
 vertebrae in the cervical, lumbar, sacral, and caudal regions, while the greatest difference 
 occurs only in the thoracic section. In both Nothrotherium and Hapalops the cervical 
 region is long (plate 4, figs. 3 and 4). In the former genus the length of the series exceeds 
 that of the series in Mylodon harlani. While no complete series of thoracic vertebrae 
 of a single individual has been obtained from the asphalt deposits, there appear to be 
 17 segments in this region of the column in Nothrotherium (plate 5). The reduction of 
 thoracic vertebrae from the number characteristic of Hapalops represents one of the 
 most fundamental changes in the skeleton that has occurred in the derivation of the 
 Pleistocene genus from its Miocene ancestor. The shortened trunk in Nothrotherium is 
 comparable to that seen in other Pleistocene ground-sloths. No ankylosis of lumbar 
 and sacral vertebrae is known to occur in Nothrotherium as in Mylodon. A greater num¬ 
 ber of vertebrae are present in the tail of the Pleistocene megalonychid than in that ot 
 Megatherium, but this number is distinctly less than that in the tail of the great ant-eater. 
 
 Cervical vertebrae. —The individual vertebrae are characterized by their broadness 
 and, in decided contrast to those of Mylodon, their greater antero-posterior extent. 
 They resemble the cervical vertebrae of Hapalops in both of these features, but with 
 the latter character somewhat less developed. The neural canal is noticeably less 
 capacious than in cervical vertebrae of the Miocene genus. The centrum in this region 
 of the column is depressed, relatively more so than in Hapalops elongatus. The anterior 
 face of the centrum is convex, the posterior face slightly concave. The dorsal surface 
 (within the neural canal) is perforated at the middle by two nutritive foramina. The 
 ventral surface is keeled, the median ridge tending to become obscure in posterior 
 
56 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 vertebrae. On either side of this ridge a rude protuberance on the ventral surface 
 extends beyond the border of the posterior face. These processes are most conspicuous 
 in the fourth and fifth vertebrae. The centra in the cervical region of the Pleistocene 
 form partake of the relative shortening which distinguishes the vertebrae from those 
 of Hapalops. 
 
 The atlas resembles considerably that of Hapalops. In anterior view it differs 
 from that of Megalonyx in a broader doming of the dorsal arch of the neural canal. 
 This arch bears above a rudimentary tubercle representing the spine. The articulating 
 facets for the occipital condyles are deeply concave and have their median internal walls 
 incised by a roughened surface. This indentation divides each surface into a large 
 ventral portion and a smaller dorsal portion corresponding to the two principal surfaces 
 of the occipital condyles. The ventral arch to the neural canal is deeply notched 
 anteriorly, as in Hapalops and differs thus from Megalonyx and from Mylodon. It 
 bears also on the lower side a hypapophysial tubercle as in the atlas of H. elongatus. 
 The articulating facet on the inner side for the odontoid process of the axis has the 
 long diameter transverse. 
 
 Table 16 .—Measurements (in millimeters ) of fifth cervical vertebra. 
 
 
 No. 
 
 1806-1 
 
 No. 
 
 1806-2 
 
 No. 
 
 1806-3 
 
 No. 
 
 1806-4 
 
 No. 
 
 1806-5 
 
 Length of centrum measured over ventral surface. 
 
 Width of centrum measured over anterior face and between 
 
 46.8 
 
 ! 43 
 
 46.6 
 
 .... 
 
 .... 
 
 inner borders of vertebrarterial canals. 
 
 Depth of centrum measured over anterior face and normal 
 
 39.5 
 
 37.8 
 
 43.7 
 
 44.6 
 
 44 
 
 to dorsal surface. 
 
 Height from ventral border of posterior face of centrum to 
 
 18 
 
 19.9 
 
 17.1 
 
 *75 
 
 16 
 
 tip of neural spine. 
 
 78.7 
 
 x 82 
 
 77.7 
 
 .... 
 
 Width across outer sides of prezygopophyses. 
 
 92.5 
 
 89.5 
 
 99 
 
 88.6 
 
 87.4 
 
 Greatest width across postzygapophyses. 
 
 65.6 
 
 65.6 
 
 73.2 
 
 70.6 
 
 61.9 
 
 Greatest width of posterior side of neural spine. 
 
 1 112 
 
 25.5 
 
 29 
 
 23.7 
 
 18.2 
 
 Greatest width across posterior ends of ventral wings. 
 
 *103 
 
 113 
 
 103 
 
 112.8 
 
 Length of ventral wing measured along lower border. 
 
 
 64 
 
 63.8 
 
 53.1 
 
 57.7 
 
 1 Approximate. 
 
 The lateral wings or processes of the atlas reach their greatest lateral extent at 
 approximately the posterior third of the vertebra. They are not broadly developed 
 anteriorly as in Mylodon , and are not deflected downward so extensively at the posterior 
 side, as in that genus. The posterior border of the lateral processes is not so deeply 
 notched, adjacent to the articulating surfaces for the axis, as in the first cervical vertebra 
 of Megalonyx. The course of the vertebrarterial canal is very similar to that in the 
 atlas of M. jeffersonii, which has been adequately described by Leidy (1855, pp. 20-21). 
 The various openings of this canal in Nothrotherium become successively smaller anteriorly. 
 
 The axis in lateral aspect is seen to differ noticeably from that of Megalonyx 
 jeffersonii in the following characters: (1) the lateral wall of the neural canal is relatively 
 much longer antero-posteriorly; (2) the greatest height of the neural spine is reached in 
 front of the posterior zygapophysis, while in Megalonyx the highest point lies above the 
 posterior zygapophysis; (3) the anterior end of the neural spine is pointed in Nothro¬ 
 therium; (4) the posterior border of the lateral wall of the neural canal is oblique; and 
 (5) the odontoid process projects relatively farther beyond the lateral facets for the 
 atlas than in Megalonyx. 
 
 The neural spine is wide, flattened above, and flares or widens considerably at the 
 posterior end. The centrum is depressed and carries a keel ventrally which widens 
 posteriorly, thus resembling Hapalops. The transverse processes are slender, as in the 
 second vertebra of the latter genus. The odontoid process may show but little inclination 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 57 
 
 upward, as in No. 1803-3. The facet on the ventral side of the odontoid process is oval 
 in shape, with greatest diameter antero-posteriorly. The posterior zygapophyses may 
 be flat, and may face downward or decidedly outward as well (1803-4). 
 
 Table 17 .—Measurements {in millimeters) of sixth cervical vertebra. 
 
 
 No. 
 
 1807-1 
 
 No. 
 
 1807-2 
 
 No. 
 
 1807-3 
 
 No. 
 
 1807-4 
 
 No. 
 
 1807-5 
 
 No. 
 
 1807-6 
 
 No. 
 
 1807-7 
 
 Length of centrum measured over ventral surface. . . 
 Width of centrum measured over anterior face and 
 
 *44 
 
 47.6 
 
 45 
 
 46.2 
 
 
 
 41.8 
 
 between inner borders of vertebrarterial canals . . . 
 Depth of centrum measured over anterior face and 
 
 41.2 
 
 42 
 
 46.6 
 
 44 
 
 47 
 
 41 
 
 37.4 
 
 normal to dorsal surface. 
 
 Height from ventral border of posterior face of cen- 
 
 17.2 
 
 18.2 
 
 20.1 
 
 17.9 
 
 *84 
 
 .... 
 
 17.4 
 
 trum to tip of neural spine. 
 
 *90 
 
 86 
 
 89 
 
 101 
 
 
 83.7 
 
 Width across outer sides of prezygapophyses. 
 
 *92 
 
 87.8 
 
 96.6 
 
 82.6 
 
 90.3 
 
 81.4 
 
 86.9 
 
 Greatest width across postzygapophyses. 
 
 76.6 
 
 69.3 
 
 74.3 
 
 58.7 
 
 
 61.2 
 
 63 7 
 
 Greatest width across ventral wings. 
 
 95 
 
 113.7 
 
 108.6 
 
 112.6 
 
 
 87.7 
 
 
 Greatest length of ventral wing. 
 
 J 63 
 
 66.2 
 
 78 
 
 56.8 
 
 59 
 
 46.7 
 
 67.3 
 
 1 Approximate. 
 
 Between the axis and the seventh vertebra the segments possess low neural spines, 
 while that of the seventh is decidedly higher. On the latter vertebra the spine is usually 
 relatively larger and heavier than in Hapalops. The vertebrae posterior to the axis 
 are also remarkable for their well-developed transverse processes and haemapophyses. 
 In these characters they show considerable likeness to the cervicals of Hapalops. With 
 the shortening of the vertebrae the anterior and posterior zygapophyses form relatively 
 much less extensive surfaces of articulation, the pedicles of the neural arch are narrower, 
 and the notch below the posterior zygapophysis is less conspicuous than in cervicals of 
 H. elongatus. 
 
 Cervical vertebrae 3 to 5 inclusive have the neural spine flaring or widening 
 posteriorly, a character which is possessed also by the spine of the axis. In the sixth 
 vertebra the spine is slightly higher than in preceding vertebrae and the posterior width 
 is not so great. The posterior border of the neural arch at the base of the neural spine 
 in these vertebrae is deeply notched. 
 
 In proceeding backward from the axis the simple transverse process is retained 
 for the most part in the third vertebra. It has flattened and broadened somewhat. 
 In the fourth it shows signs of bifurcation at the outer posterior end, and this division 
 becomes much more evident in the fifth. Here the lower part develops into a decided 
 
 Table 18 .—Measurements {in millimeters) of seventh cervical vertebra. 
 
 
 No. 
 
 1808-1 
 
 No. 
 
 1808-2 
 
 No. 
 
 1808-3 
 
 Length of centrum. 
 
 38.8 
 
 
 41.4 
 
 Width of centrum measured over anterior face . 
 
 43.5 
 
 40.8 
 
 41 
 
 Depth of centrum measured over anterior face. 
 
 22.4 
 
 24.2 
 
 22 
 
 Width across metapophyses. . . . 
 
 95.4 
 
 86.8 
 
 79.4 
 
 Greatest width across transverse processes. 
 
 136.1 
 
 145.4 
 
 128.8 
 
 Width across pleurapophyses. . . 
 
 l 77 
 
 62.8 
 
 64 
 
 Length of neural arch, measured from anterior end of prezygapophysis to posterior 
 end of postzygapophysis. 
 
 60.3 
 
 59.8 
 
 64.9 
 
 Width across postzygapophvses. 
 
 77.6 
 
 66.1 
 
 65.6 
 
 Height, measured from middle of ventral border of posterior face of centrum to end 
 of neural spine. 
 
 131.6 
 
 h34 
 
 138.6 
 
 Dorso-ventral diameter of neural canal measured at anterior end. 
 
 24 
 
 22 
 
 24.7 
 
 Transverse thickness of neural spine at middle. 
 
 10.5 
 
 14 
 
 13.7 
 
 
 
 
 
 1 Approximate. 
 
58 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 inferior plate which reaches its maximum size in the sixth vertebra. As in Hapalops, 
 it extends well below the ventral surface of the centrum. The diapophysis in the sixth 
 vertebra has shifted forward and may in turn be bifid, in which case the upper portion 
 is heavier than the lower. In the seventh cervical the diapophysis forms a decided 
 transverse process which extends well outward and downward, while the inferior lamella 
 lies much more to the ventral side of the centrum than it does in the sixth and has 
 become much reduced. There is, in other words, a rapid decrease in size of the inferior 
 lamella from the sixth to the seventh vertebra. 
 
 The much greater development of the lateral wings of the cervical vertebrae, 
 together with greater length of centra, are decided differences that exist between 
 Nothrotherium and Mylodon in this region of the column. An effective scalene musculature 
 is thus indicated in the former genus. Probably with the unusually developed inferior 
 lamellae is to be correlated a greater degree of vertical flexibility of the neck than exists 
 in Mylodon. The bending of the neck in this direction seems to have been facilitated 
 also by the deep notching of the posterior border of the neural arch. Length of 
 articulated cervical series, 1 to 7 inclusive, is 293 mm. 
 
 Table 19 .—Measurements {in millimeters ) of first thoracic vertebra of Nothrotherium shastense. 
 
 
 No. 
 
 1809-2 
 
 No. 
 
 1809-3 
 
 No. 
 
 1809-4 
 
 No. 
 
 1809-5 
 
 No. 
 
 1S09-6 
 
 Length of centrum. 
 
 40.9 
 
 42.6 
 
 
 
 
 Width of centrum measured over anterior face. 
 
 44.5 
 
 42.4 
 
 42.3 
 
 41.8 
 
 
 Depth of centrum measured over anterior face. 
 
 25.2 
 
 26.7 
 
 23.7 
 
 27.8 
 
 
 Width across metapophyses. 
 
 96.6 
 
 79.7 
 
 85.4 
 
 86.3 
 
 76.2 
 
 Greatest width across transverse processes. 
 
 Length of neural arch along middle and at base of neural 
 
 137.5 
 
 132.2 
 
 141 
 
 139.7 
 
 128.3 
 
 spine. 
 
 Height measured from middle of ventral border of posterior 
 
 56.2 
 
 56.9 
 
 51.3 
 
 *131.5 
 
 49.5 
 
 face of centrum to end of neural spine. 
 
 Dorso-ventral diameter of neural canal measured at ante- 
 
 137.9 
 
 141.5 
 
 
 .... 
 
 rior end. 
 
 23.7 
 
 24.9 
 
 23.3 
 
 21.7 
 
 20.5 
 
 Transverse thickness of neural spine at middle. 
 
 14 
 
 13.5 
 
 15 
 
 15.2 
 
 13.8 
 
 Transverse thickness of end of neural spine. 
 
 34.2 
 
 31.5 
 
 t 65 
 
 
 
 Antero-posterior extent of end of neural spine. 
 
 46 
 
 62.9 
 
 J 58 
 
 .... 
 
 1 Approximate. 
 
 Thoracic vertebrae. —Six specimens of the first thoracic vertebra (plate 5, figs. 2 a and 
 3) of Nothrotherium are available for study. In this vertebra the dorsal spine is very 
 broad antero-posteriorly, thus resembling the corresponding element in Megalonyx and 
 in Hapalops. In the Rancho La Brea specimens the spine is perpendicular, whereas 
 in a first thoracic of H. elongatus (No. 15545, Prin. Univ. Coll.), before the writer, the 
 principal axis of the spine is inclined somewhat away from the perpendicular. The 
 large spine in Nothrotherium., Megalonyx, and Hapalops is a remarkable feature and is 
 entirely different from the type of spine in the first thoracic of Mylodon. 
 
 The centrum, as in the seventh cervical, is relatively shorter when contrasted with 
 that of H. elongatus. It is compressed, with dorsal face within the neural arch perforated 
 by large foramina in the middle, and the ventral surface shows similar perforation. 
 Short parallel ridges may be present on the ventral surface, one along the outer side of 
 each foramen as in No. 1809-3, and are probably the remains of the inferior lamellae 
 of the cervical vertebrae. The posterior surface of the centrum is truncated at each 
 dorso-lateral angle by a small articulating facet for the capitulum of the second rib. 
 The anterior zygapophyses are relatively very much smaller than in Hapalops, and the 
 transverse processes are also smaller and more slender. Immediately adjacent to each 
 metapophysis is a small canal that pierces the edge of the transverse process and extends 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 59 
 
 from the anterior to the dorsal surface. Sometimes this canal is open. Viewed from the 
 ventral side (plate 5, fig. 3), the facet for the capitulum of the first rib is seen to be 
 smaller than that for the tubercle. Contrasted with similar facets in H. elongatus, 
 those of Nothrotherium are relatively smaller and more closely situated. The small facet 
 for the capitulum is situated on the ventral side of the transverse process or along the 
 outer side of the pedicle and does not extend downward along the lateral border of the 
 centrum so far as in the first dorsal of Mylodon harlani. 
 
 Specimen 1809-4 exhibits an interesting variation in that the metapophysis of the 
 right side is entirely dissociated from the articulating facet for the seventh cervical. 
 Usually the facet is concave and extends upward along the inner side of the metapophysis, 
 but in this specimen the surface extends downward and outward, having an appearance 
 like that in posterior thoracic vertebrae. 
 
 Table 20. —Measurements {in millimeters) of thoracic vertebrae of N. shastense, series 1809-1 to 1825-1. 
 
 
 No. 
 
 No. 
 
 No. 
 
 No. 
 
 No. 
 
 No. 
 
 No. 
 
 No. 
 
 No. 
 
 No. 
 
 No. 
 
 No. 
 
 No. 
 
 No. 
 
 No. 
 
 No. 
 
 No. 
 
 
 1 
 
 2 
 
 3 
 
 4 
 
 5 
 
 6 
 
 7 
 
 8 
 
 9 
 
 10 
 
 11 
 
 12 
 
 13 
 
 14 
 
 15 
 
 16 
 
 17 
 
 Length of centrum. 
 
 38 
 
 38.9 
 
 39.1 
 
 39 
 
 38.1 
 
 40.3 
 
 38.7 
 
 40.7 
 
 41 
 
 
 43.5 
 
 43.2 
 
 43.9 
 
 44.4 
 
 44.3 
 
 46.3 
 
 47.3 
 
 Width of centrum meas- 
 
 
 ured over anterior face.. 
 Depth of centrum meas- 
 
 42 
 
 42 
 
 40.9 
 
 42.3 
 
 42.2 
 
 43.8 
 
 45.2 
 
 47 
 
 48 
 
 *50 
 
 53.9 
 
 56.9 
 
 58.6 
 
 61.3 
 
 65.8 
 
 60.5 
 
 62.9 
 
 ured over anterior face.. 
 Greatest width. 
 
 24.1 
 
 25.6 
 
 28.3 
 
 113.7 
 
 31 
 
 110.5 
 
 31.3 
 
 107.9 
 
 33.5 
 
 105 
 
 34.5 
 
 113.5 
 
 36.5 
 
 115 
 
 38.2 
 
 >123.4 
 
 38.5 
 
 126 
 
 41.6 
 
 '120 
 
 42.2 
 
 '129.2 
 
 44.1 
 
 '132.6 
 
 44.7 
 
 ‘132.8 
 
 46.3 
 
 130 
 
 48.2 
 
 130.7 
 
 48.8 
 
 134.2 
 
 Length of neural arch along 
 
 
 
 middle at base of neural 
 spine. 
 
 53.3 
 
 
 49.8 
 
 52.3 
 
 52.8 
 
 53.4 
 
 56.9 
 
 61 
 
 61 
 
 
 61.9 
 
 58.2 
 
 63.4 
 
 60.8 
 
 63.7 
 
 62 
 
 68.7 
 
 Height measured from mid¬ 
 dle of ventral border of 
 posterior face of centrum 
 to end of neural spine... 
 
 Dorso-ventral diameter of 
 
 
 129.2 
 
 '128.6 
 
 119 
 
 120 
 
 119.6 
 
 120.1 
 
 121.2 
 
 124.5 
 
 '124.5 
 
 134.5 
 
 138 
 
 140.7 
 
 145 
 
 151.8 
 
 160.4 
 
 160.6 
 
 
 neural canal measured 
 at anterior end. 
 
 21 
 
 20.5 
 
 21.5 
 
 21.8 
 
 24 
 
 24.8 
 
 26.8 
 
 26 
 
 
 *36 
 
 35.2 
 
 33.7 
 
 35 
 
 35.6 
 
 35.8 
 
 35.8 
 
 35.8 
 
 Transverse thickness of 
 neural spine at middle.. . 
 Transverse thickness of 
 
 12 
 
 15.5 
 
 
 10.9 
 
 10.7 
 
 11 
 
 10.4 
 
 11.9 
 
 13.7 
 
 13.7 
 
 14 
 
 14.7 
 
 13.3 
 
 14.7 
 
 14.7 
 
 14 
 
 13 
 
 
 end of neural spine. 
 
 Transverse width across 
 
 35.4 
 
 . 
 
 16.4 
 
 15.3 
 
 15.7 
 
 16.8 
 
 19.2 
 
 18.9 
 
 19.7 
 
 22 
 
 22 
 
 23.6 
 
 23.8 
 
 24.1 
 
 26 
 
 25.3 
 
 23.6 
 
 anterior zvgapophyses.. . 
 
 
 ‘57.8 
 
 56.1 
 
 56.4 
 
 56.3 
 
 *55 
 
 59.7 
 
 60.6 
 
 
 59.7 
 
 60.5 
 
 62.1 
 
 63.7 
 
 66.4 
 
 70.4 
 
 x 76.4 
 
 78.7 
 
 
 
 
 
 1 Approximate. 
 
 In the thoracic series figured on plate 5, figures 1 and 2, all the vertebrae do not 
 belong to the same individual. Certain changes can be noted, however, in passing from 
 the first to the last of the series. In the third and succeeding vertebrae the tips of the 
 neural spines are much less thickened transversely than in No. 1. In the second vertebra 
 of this series the neural spine has been broken away. In the second thoracic and in the 
 following vertebrae the neural spines slope posteriorly. 
 
 Posterior to the second thoracic there is a gradual increase in transverse width 
 across the anterior zygapophyses. The upper facet for the tubercle of the rib is concave 
 in the first to sixth vertebra inclusive. In the seventh it is nearly flat, and in the following 
 vertebrae, to the end of the series, it is convex. From the second vertebra to presumably 
 the ninth inclusive, the dorso-external angle of the posterior face of the centrum is 
 truncated by a small articulating surface for a part of the capitulum of the rib. 
 
 The depth of the centrum is fully doubled from anterior to posterior end of the 
 series. In the same distance the length of the centrum increases 24 per cent, while the 
 width of the centrum increases 50 per cent. Throughout the thoracic series the ventral 
 surface of the centrum is pierced on either side of the median line by a foramen. These 
 foramina may become quite large posteriorly and an additional foramen may be added. 
 
60 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 Lumbar vertebrae. —The lumbar series in Nothrotherium consists of three vertebrae. 
 In Hapalops there are three or four vertebrae in this section of the column. There is a 
 slight increase in length of centrum, but a slight decrease in depth from anterior to 
 posterior end of the lumbar series. The width of the centrum increases from the first 
 to third vertebra inclusive. The vertebrae decrease in total height as they approach 
 the sacrum. 
 
 The lumbar vertebrae differ from the last thoracic in the development of a forward 
 and upward projecting process having its origin on the upper surface of the base of the 
 transverse process. The process is rather low in the first lumbar, becomes most prominent 
 in the third, meanwhile coming in closer relation with the anterior zygapophysis. In 
 the third lumbar the process is no longer posterior to the anterior zygapophysis, but 
 supports the latter along the inner side of its base. In other words, the anterior 
 zygapophyses now have the major part of their articulating surface directed inward 
 instead of upward. 
 
 The base of the transverse process on the right side of the first lumbar is pierced 
 by a canal. The forward-projecting process on this side is also pierced by a rather 
 large foramen near the anterior border. 
 
 Table 21. —Measurements (in millimeters) of lumbar vertebrae of N. shastense (series 1826-1 to 1828-1). 
 
 
 No. 1 
 
 No. 2 
 
 No. 3 
 
 Length of centrum . 
 
 49.3 
 
 51.5 
 
 63 8 
 
 Width of centrum measured over anterior face. 
 
 65.4 
 
 70.4 
 
 73.9 
 
 Depth of centrum measured over anterior face. 
 
 52.2 
 
 52.4 
 
 50.7 
 
 Width across supports for anterior zygapophyses. 
 
 86 
 
 91.9 
 
 57.5 
 
 Width across transverse processes. 
 
 135.7 
 
 164.9 
 
 147.3 
 
 Greatest width across posterior zygapophyses. 
 
 74 
 
 51.8 
 
 61 
 
 Length of neural arch along middle and at base of neural spine. 
 
 74 5 
 
 *64.6 
 
 76.6 
 
 Height measured from middle of ventral border of posterior face of centrum to end of 
 neural spine . 
 
 166.7 
 
 164.6 
 
 159.5 
 
 Dorso-ventral diameter of neural canal measured at anterior end. 
 
 37.9 
 
 39 
 
 40 
 
 Transverse thickness of neural spine at middle. 
 
 15.5 
 
 14.3 
 
 16 
 
 Transverse thickness of end of neural spine. 
 
 26.8 
 
 30 
 
 33.6 
 
 
 1 Approximate. 
 
 Sacral vertebrae. —In both of the more complete specimens of the pelvis of 
 Nothrotherium the sacrum consists of five vertebrae. These have been completely united 
 in No. 1892-1, both centra and arches having fused. Dorsally the spines form a 
 continuous crest. In the second specimen the individual centra are still distinguishable. 
 In this the dorsal spine of the fifth or last sacral vertebra has not fused with the spinous 
 crest in front of it. 
 
 According to Scott, the sacrum of Hapalops consists of five vertebrae, though in some 
 individuals six vertebrae have been noted. The number of sacrals in Nothrotherium 
 and in Hapalops is thus distinctly less than in Mylodon. Close resemblances between 
 the two megalonychids are seen in the structure of the centra and in the union of the 
 sacrum with the os innominatum. As in the sacrum of Hapalops longiceps, the middle 
 centra are very much depressed. The nutrient foramina piercing the under sides of 
 the centra tend to disappear in the middle vertebrae of the sacrum. The first three 
 sacral vertebrae join with the ilia, the last two with the ischia. 
 
 Caudal vertebrae. —Probably 21 vertebrae were present in the tail of N othr other ium 
 shastense (plate 6). In the most complete caudal series from the asphalt deposits 16 
 vertebrae are preserved. The first to the fourteenth inclusive form a consecutive series, 
 while the remaining 2 represent the sixteenth and eighteenth vertebrae. 
 
 The first caudal vertebra has a short but very broad centrum, with transversely 
 convex anterior face and concave posterior face. The anterior face is not distinct from 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 01 
 
 the base of the neural arch, as is the posterior face. Consequently the dorsal border of 
 the former is flat, not broadly convex as in the latter. This character is seen also in 
 vertebrae that follow the first. From the first to eighth inclusive the difference between 
 width and height of anterior face of centrum ranges approximately from 30 mm. to 
 10 mm. From the ninth to the fourteenth inclusive this difference ranges from 8 mm. 
 to approximately 5 mm. In the sixteenth and eighteenth vertebrae the difference is 
 approximately 5 mm. 
 
 The neural canal is semicircular when viewed from the front. In posterior vertebrae 
 the roof of the canal may become flatter. The canal is still completely arched in the 
 sixteenth vertebra. In the eighteenth the canal is open dorsally. The neural spine 
 is low and thick in the first segment and gradually decreases in height in those following. 
 It is just barely visible in the eleventh. 
 
 Table 22.— Measurements (in millimeters) of caudal vertebrae of N. shastense series 1830-1 to 1847-1. 
 
 
 No. 
 
 1 
 
 No. 
 
 2 
 
 No. 
 
 3 
 
 No. 
 
 4 
 
 No. 
 
 5 
 
 No. 
 
 6 
 
 No. 
 
 7 
 
 No. 
 
 8 
 
 No. 
 
 9 
 
 No. 
 
 10 
 
 No. 
 
 11 
 
 No. 
 
 12 
 
 No. 
 
 13 
 
 No. 
 
 14 
 
 No. 
 
 16 
 
 No. 
 
 18 
 
 Length of centrum. 
 
 Width of centrum meas- 
 
 51.3 
 
 50.2 
 
 48.9 
 
 47.5 
 
 48.4 
 
 48.3 
 
 48.2 
 
 48.9 
 
 47.9 
 
 46.3 
 
 44.6 
 
 42 
 
 39.3 
 
 35 
 
 33 
 
 27.3 
 
 ured over anterior face. 
 Depth of centrum meas- 
 
 74.1 
 
 70.7 
 
 68.4 
 
 64.6 
 
 61.4 
 
 59 
 
 56 
 
 51.6 
 
 47.6 
 
 45.2 
 
 41.6 
 
 39.9 
 
 36.6 
 
 33 9 
 
 31 
 
 27.3 
 
 ured over anterior face.. 
 Greatest width across 
 
 44.7 
 
 49 
 
 49.9 
 
 49.7 
 
 47 
 
 45.3 
 
 43.6 
 
 41 
 
 39 
 
 38 
 
 34.7 
 
 34 
 
 30.8 
 
 29 
 
 25.9 
 
 21.5 
 
 metapophyses. 
 
 Width across transverse 
 
 89.5 
 
 90.5 
 
 87.5 
 
 78.3 
 
 70.8 
 
 66.3 
 
 60.5 
 
 55.6 
 
 52.5 
 
 48.5 
 
 38.9 
 
 25.3 
 
 22 
 
 21.6 
 
 19 2 
 
 
 processes. 
 
 Greatest antero-posterior 
 diameter of transverse 
 
 214 7 
 
 212 
 
 176.4 
 
 154.2 
 
 154.3 
 
 144 9 
 
 145.3 
 
 137.3 
 
 126.5 
 
 111.9 
 
 92 
 
 76.5 
 
 68.8 
 
 59 
 
 51.6 
 
 41.7 
 
 process. 
 
 Greatest width across pos- 
 
 39.7 
 
 44 5 
 
 38 
 
 41.9 
 
 36 3 
 
 34.7 
 
 39.8 
 
 41 5 
 
 40.4 
 
 36.8 
 
 34.8 
 
 32.4 
 
 29.3 
 
 22.3 
 
 21.4 
 
 16.3 
 
 terior zygapophyses.... 
 
 Height measured from 
 middle of ventral border 
 of posterior face of cen¬ 
 trum to end of neural 
 
 55.8 
 
 53.3 
 
 48.7 
 
 43.8 
 
 46 
 
 46.2 
 
 45 2 
 
 44 3 
 
 42.3 
 
 32.3 
 
 21.7 
 
 16.2 
 
 12.8 
 
 16 
 
 13.6 
 
 
 spine. 
 
 97.2 
 
 89.7 
 
 84.6 
 
 81 
 
 77.9 
 
 72.7 
 
 66.9 
 
 62.8 
 
 56 
 
 52.8 
 
 46.5 
 
 41 
 
 37.6 
 
 37.3 
 
 28.7 
 
 
 The anterior and posterior zygapophyses in the anterior 11 vertebrae project in 
 front and in back of the centrum respectively. In the eleventh segment they reach 
 only to the limits of the anterior and posterior faces. In the first vertebra the anterior 
 zygapophysis is inclined steeply inward and is relatively narrow antero-posteriorly. In 
 following vertebrae the facet becomes wider. In the seventh and following vertebrae 
 the facet becomes less steeply inclined and ultimately comes to face distinctly upward. 
 The facet is functional to and including the eleventh vertebra. The posterior zyga¬ 
 pophyses are first steeply inclined and face directly outward, but in following vertebrae 
 the facets are gradually directed more downward as well as outward. Commencing with 
 the eighth vertebra and proceeding backward, the posterior border of the neural arch 
 between the posterior zygapophyses is decidedly notched. 
 
 The transverse process is long in the first two vertebrae and much shorter in the 
 third of the series and posteriorly. In the second segment it differs from that in the 
 first in having the outer border deflected dorsally. A distinct flange is found along the 
 posterior border of the process in the second, an indication of which is already present 
 in the first caudal. In the third vertebra the flange is transformed into a process and the 
 outer border is relatively thick. From the fourth to the thirteenth vertebra inclusive 
 a defined downward deflection occurs which reaches its maximum extent in the seventh 
 vertebra. The metapophysis of the first segment is short antero-posteriorly. It is low 
 and heavy, decreasing in size in following vertebrae. 
 
62 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 On each side of the ventral surface of the centrum and connected with the anterior 
 border, but separated from the posterior by a deep channel, is a knob-like prominence. 
 This is connected dorsally with the transverse process. Opposite these prominences and 
 on the posterior border are two facets articulating with the chevron-bone. In the second 
 vertebra the anterior pair of chevron facets occupies a position comparable to the 
 position of the knob-like prominences of the first caudal. The posterior pair of chevron 
 facets is smaller than the anterior pair. The latter difference between the two pairs 
 of facets is found also in the third vertebra. In the fourth and fifth vertebrae the two 
 pairs are of nearly equal size, while in the sixth to ninth inclusive the posterior pair is 
 decidedly the larger. From the tenth vertebra posteriorly the anterior pair of facets 
 is absent, while the posterior pair disappears in the twelfth. The interspace between 
 right and left sets of chevron facets becomes narrower from the first vertebra posteriorly. 
 In the first segment this area is perforated by two large foramina. In those following 
 the usual number is either 2 or 4. 
 
 Fig. 16.— Nothrotherium shastense Sinclair. A, B, C, outer, anterior, and inner views 
 of right half of second haemapophysis, No. 1919-R-l; D, E, F, outer, anterior, and 
 inner views of right half of fourth haemapophysis, No. 1921-R-l. X 0.50. Rancho 
 La Brea Pleistocene. 
 
 HAEMAPOPHYSES. 
 
 The available haemapophyses or chevron-bones of Nothrotherium pertain principally 
 to the forward and middle regions of the tail. Associated with the most complete caudal 
 series of vertebrae from Rancho La Brea are 5 haemapophyses. The material belonging 
 to this individual consists of the following specimens: 
 
 (1) Right half of second haemapophysis articulating with second and third vertebrae (fig. 16, A, b, c). 
 
 (2) Right half of fourth haemapophysis articulating with fourth and fifth vertebrae (fig. 16, d, e, f). 
 
 (3) Complete fifth haemapophysis articulating with fifth and sixth vertebrae (fig. 17, a to e). 
 
 (4) Complete sixth haemapophysis articulating with sixth and seventh vertebrae (fig. 18, A, b). 
 
 (5) Complete ninth haemapophysis articulating with ninth and tenth vertebrae (fig. 19, a, b, c). 
 
 The haemapophyses of Nothrotherium differ strikingly in shape from those of Hap- 
 alops. In the anterior portion of the tail, the elements when completely formed by 
 fusion of the right and left pieces are more X-shaped than Y-shaped. They are Y-shaped 
 in Hapalops and in Mylodon. As a descriptive term the designation “chevron” is 
 therefore not truly applicable to the ventral appendage of the caudal vertebra in Noth- 
 rotlierium. In the anterior four haemapophyses the two lateral pieces are not united, 
 for the facets on the caudal vertebrae articulating with these elements are widely sepa- 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF HANCIIO LA BREA. 63 
 
 rated transversely. Including and posterior to the fifth haemapophysis, the two parts 
 have united in median line. This union usually takes place at about the middle of the 
 element and may or may not extend completely to the distal end. 
 
 In the series available the haemapophyses increase in depth from the most anterior 
 element to at least the sixth. Posterior to the latter the decrease in depth is apparently 
 rapid. When right and left parts of a single haemapophysis are joined in middle line, 
 the proximal arms of the X inclose an oval-shaped trough. Each arm bears an anterior 
 and a posterior facet for the vertebral centra with which it articulates. In the first and 
 second haemapophyses the posterior facet is the larger, and in the third the anterior 
 and posterior facets are approximately of same size. In the haemapophyses behind 
 
 Fig. 17 .—Nothrotherium shastense Sinclair. Fifth haemapophysis, No. 1922-1. 
 
 X 0.50. A, anterior view; B, outer view; C, posterior view; D, dorsal view; 
 
 E, ventral view. Rancho La Brea Pleistocene. 
 
 and including the fourth, the anterior facet is the larger and lingers after the posterior 
 one has disappeared. The ninth haemapophysis is much reduced in dorso-ventral 
 extent and bears two large vertebral facets anteriorly, but only two tubercles posteriorly. 
 The distal surface of this element is broad and the haemal canal is relatively large. 
 
 In the fifth and sixth haemapophyses, where the two halves are present and united, 
 the lower or distal arms of the X are not only divergent from their median union, but 
 flare also posteriorly. In other words, the two distal arms of the X are not parallel in 
 their antero-posterior extent, as are the proximal arms. In the five haemapophyses 
 listed above, the distal arm is shortest in antero-posterior extent in the second and 
 
 Table 23. — Measurements (in millimeters) of haemapophyses of N. shastense. 
 
 
 No. 
 
 1919-R-l 2 
 
 No. 
 
 1921-R-l 4 
 
 No. 
 
 1922-1 5 
 
 No. 
 
 1923-1 6 
 
 No. 
 
 1926-1 9 
 
 Greatest depth. 
 
 57.9 
 
 56.2 
 
 55.1 
 
 64.1 
 
 29.8 
 
 Greatest transverse width at proximal end. 
 
 
 
 48 
 
 48.5 
 
 28.7 
 
 Least transverse width. 
 
 
 
 24.2 
 
 20.7 
 
 
 Greatest transverse width at distal end. 
 
 
 
 62.7 
 
 *61 
 
 16.4 
 
 Greatest antero-posterior length of proximal right limb. 
 
 43 
 
 43.7 
 
 39.8 
 
 40.6 
 
 32.4 
 
 Least antero-posterior length of right side of haemapophysis . 
 
 28.2 
 
 28.3 
 
 27.3 
 
 28.4 
 
 
 Greatest antero-posterior length of distal right limb. 
 
 39 
 
 58.4 
 
 53 
 
 2 49.5 
 
 
 Thickness of proximal end of right limb. 
 
 24.4 
 
 23.8 
 
 20.1 
 
 21.3 
 
 12.6 
 
 1 Approximate. 2 Left side. 
 
C>4 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 longest in the fourth. In several haemal elements of Nothrotherium from Rancho La 
 Brea, a canal extends through the fused median portion of the bone, reaching from the 
 haemal trough to between the diverging lower arms. 
 
 The peculiar shape of the haemapophyses in Nothrotherium is quite different from 
 that seen in other gravigrade edentates. It suggests a fleshy and massive tail. A great 
 development of caudal musculature is probably also indicated by the large size of the 
 principal caudal blood-vessel inferred from the capaciousness of the haemal canal. 
 
 COSTAL RIBS. 
 
 A number of costal ribs (plate 7, fig. 2) accompany the thoracic vertebrae from 
 Pit 3. In this series the first is fused with the sternal rib. In the second rib the 
 articulating face of the head is flat, while that of the tubercle is transversely concave. 
 The latter facet encroaches upon the posterior face of the rib. In the fourth rib the facet 
 of the tubercle is slightly convex, while that of the head is still flat. The tubercle is 
 prominent to about the ninth rib. In this region the articulating facet begins to sink 
 into the shaft and the tubercle is reduced. Distal to the tubercle the shaft in the first 
 
 Fiq. 18 .—N othrotherium shastense Sinclair. Sixth haemapophysis, No. 1923-1. X 0.50. A, outer view; B, posterior view; 
 
 Rancho La Brea Pleistocene. 
 
 Fio. 19 .—Nothrotherium shastense Sinclair. Ninth haemapophysis, No. 1926-1. X 0.50. A, dorsal view; B, outer view; 
 
 C, ventral view. Rancho La Brea Pleistocene. 
 
 and second ribs is quadrilateral in cross-section. In the two ribs which follow, the shaft 
 begins to broaden antero-posteriorly. The tenth, eleventh, twelfth, and thirteenth 
 ribs are wide, while the last four ribs are relatively thick. 
 
 STERNAL RIBS. 
 
 The sternal ribs (plate 7, fig. 1) in Nothrotherium resemble those typically developed 
 in ground-sloths and lend strength to the ventral side of the rib-basket. There were 
 presumably 10 ossified sternal ribs on each side of the sternum, a number comparable 
 to that found in other Pleistocene genera. The anterior 8 joined the costal ribs to the 
 sternum, while the ninth connected with the eighth and the tenth with the ninth. The 
 first sternal rib commonly fuses with the costal rib, and such fusion may occasionally 
 occur in the second. In Mylodon harlani and in Megatherium apparently only 7 sternal 
 ribs articulate directly with the sternum. The complete series is not known in Megalonyx 
 or in the Upper Miocene megalonychids from Patagonia. 
 
 The first sternal rib is plate-like and of trapezoidal shape. It articulates by a narrow 
 elongate facet with the side of the manubrium. The dorso-ventral extent of this rib 
 is greater than the transverse diameter. 
 
 The second and third sternal ribs are not definitely known. In the nearly complete 
 series from Pit 3 a sternal rib is present which possibly represents the second. It 
 apparently joined only with the ventral keel of the sternal segment, for an articulating 
 surface more dorsally situated is not present in this specimen as in posterior ribs. In 
 this respect the specimen under observation would probably answer best to the second, 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 65 
 
 for in the union of the second and first sternal segments there is no dorsal facet for the 
 second sternal rib. The rib possesses, however, an antero-internal articulating surface, 
 while no comparable facet is found on the manubrium. The shaft of the rib is compressed 
 antero-posteriorly. Toward the union with the costal rib the shaft widens and deepens. 
 
 Table 24. —Measurements (in millimeters') of costal ribs of N. shastense, 
 
 series 1850-1 to 1866-1. 
 
 
 Distance across 
 head and 
 tubercle. 
 
 Dorso-ventral 
 diameter of head. 
 
 Greatest trans¬ 
 verse width of 
 shaft. 
 
 Length along 
 greater curvature 
 from facet of 
 tubercle to dis¬ 
 tal end. 
 
 No. 1. 
 
 36 
 
 11.4 
 
 
 150 
 
 No. 2. 
 
 40.5 
 
 14 
 
 .... 
 
 
 No. 4. 
 
 J 42.4 
 
 .... 
 
 29.6 
 
 
 No. 5 . 
 
 • • • • 
 
 .... 
 
 35 
 
 280 
 
 No. 6. 
 
 53.7 
 
 17.8 
 
 37.4 
 
 330 
 
 No. 7. 
 
 59.7 
 
 21 
 
 *38 
 
 370 
 
 No. 8. 
 
 61 
 
 24.7 
 
 x 34.8 
 
 390 
 
 No. 9. 
 
 61.6 
 
 24.6 
 
 35 
 
 405 
 
 No. 10. 
 
 63.3 
 
 24.6 
 
 35.3 
 
 410 
 
 No. 11. 
 
 63 
 
 23.9 
 
 35.2 
 
 
 No. 12. 
 
 62.7 
 
 23.7 
 
 35.4 
 
 .... 
 
 No. 13. 
 
 66 
 
 23.3 
 
 39.4 
 
 .... 
 
 No. 14. 
 
 64 
 
 23.9 
 
 38.9 
 
 440 
 
 No. 15. 
 
 65.5 
 
 22.7 
 
 36.3 
 
 mo 
 
 No. 16. 
 
 66.3 
 
 20.7 
 
 32.5 
 
 400 
 
 No. 17. 
 
 65.1 
 
 19.6 
 
 32.4 
 
 1 360 
 
 1 Approximate. 
 
 The fourth sternal rib (No. 1905-L-l) is relatively thin and has a warped appearance, 
 the median half extending in dorso-ventral direction, the lateral half reaching more in 
 an antero-posterior direction. The ventral surface of the lateral half is concave. At 
 the middle of the posterior border of the lateral half this rib was attached to the fifth 
 sternal rib. A connection between the fourth and fifth sternal ribs is not known to 
 occur in Mylodon or in Megatherium. The lateral extremity is expanded for cartilaginous 
 connection with the costal rib. The medial articulation of the fourth rib with the 
 sternum consists dorsally of a single facet, which joins with the dorsal portions of the 
 third and fourth sternal segments. Ventrally the articulation consists of two facets, 
 one joining with the ventral head of the third sternal segment, the other with the ventral 
 head of the fourth sternal segment. The facets extend well to the anterior and posterior 
 sides of the medial extremity of the rib. 
 
 The fifth sternal rib (No. 1906-Lr-l) resembles the fourth, but is longer. It bears 
 along the anterior border a facet for the fourth rib which is directed downward and 
 forward. The rib is relatively thin and the ventral crest lies along the posterior border. 
 The vertebral end is expanded and divided into two parts, one of which is directed 
 anteriorly and connects with the costal rib, while the other is directed posteriorly and 
 joins with the sixth sternal rib. 
 
 The sixth sternal rib (No. 1907-L-l) does not possess the warped appearance of 
 the fourth and fifth, but the dorsal part of the shaft has broadened to form a distinct 
 base, from which is subtended a sharp ventral crest. The latter reaches from the head 
 to a point where the medial two-thirds and lateral one-third of the shaft meet. The 
 lateral one-third of the shaft carries a contact-surface on the anterior border for the 
 fifth sternal rib. The vertebral end joins in part with the costal rib and in part with 
 the seventh sternal rib. 
 
66 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 The seventh sternal rib (No. 1908-Lr-l) is heavier than the sixth, although the medial 
 head is no larger than that of the latter rib. The ventral crest which extends at first 
 along the middle of the shaft moves to the anterior border as the vertebral extremity 
 is approached. Here it carries an oval surface for the sixth sternal rib. Posterior to 
 this ridge the rib widens into a thin plate, but much of this and practically all of the 
 lateral extremity have been broken away in the specimen under examination. There 
 is a distinct upward curvature to the lateral portion of the dorsal surface and a forward 
 bending of the anterior border. The medial extremity articulates with the sixth sternal 
 segment and the xiphisternum. 
 
 The eighth sternal rib (No. 1909-R-l) also articulates with the sternum, in which 
 respect Nothrotherium differs from Megatherium and Mylodon. The shaft is much 
 curved. The medial extremity is tumid, with antero-posterior diameter smaller than 
 that in any of the four preceding ribs. The head bears medially a flattened facet, and 
 above this a small surface, both of which apparently articulate only with the xiphisternum. 
 
 D E 
 
 F 
 
 G H I J K. L 
 
 Fig. 20 .—Nothrotherium shastense Sinclair. Sternal segments; anterior, ventral, and right side views. X 0.50. A, B, C 
 second segment, No. 1896-1; D, E, F, third segment, No. 1897-1; G, H, I, fifth segment, No. 1899-1; J, K, L, sixth 
 segment, No. 1900-1. Rancho La Brea Pleistocene. 
 
 The ninth sternal rib (No. 1910-R-l) unfortunately lacks the medial stem. It is 
 broad, flattened, and plate-like where it joins with the costal rib, but tapers inward 
 towards the stem. The surface for attachment to the costal rib cuts obliquely across 
 the outer anterior corner of the sternal rib. 
 
 The tenth sternal rib (No. 1911-L-T) also lacks the medial stem. This rib is short, 
 considerably flattened, thin, and tapers towards the medial end. The surface of 
 attachment to the costal rib is comparable in position to that in the ninth sternal rib. 
 The difference in size between this specimen and the ninth rib is probably no greater 
 than that between the corresponding ribs of Megatherium. 
 
 STERNUM. 
 
 The sternum (plate 7, fig. 1; also fig. 20, a to l) has not been found completely 
 preserved in any one individual from the asphalt deposits, but the number of elements 
 available furnish some information regarding this structure. Presumably the sternum 
 of Nothrotherium consists of 7 segments, a number comparable to that in other Pleistocene 
 ground-sloths. 
 
 The dorsal or internal surface of the sternum is flattened, or may be slightly concave 
 transversely in mid-region. Behind the manubrium the individual segments increase 
 gradually in width to the posterior end of the fourth, when a decrease in width takes 
 place. The mesosternal elements are compact and pawn-like in appearance. They are 
 constricted laterally and, in contrast to those of Mylodon , do not have the ventral 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 67 
 
 head or keel so elongate in antero-posterior direction. This is particularly true of the 
 second, third, and fourth sternal segments. All the elements articulate laterally with 
 the heads of the sternal ribs. Those of the mesosternum possess for this articulation 
 4 facets on each side, 2 supported by the ventral process and 1 situated at each angle 
 and on the under side of the dorsal plate. 
 
 The manubrium approaches the triangular shape, with anterior end tapering. It 
 differs from that in Hapalops in lacking a large forward process. The dorsal surface 
 is for the most part smooth and concave. In Hapalops the dorsal concavity is much 
 deeper than in Nolhrotherium. Immediately behind the anterior projection the manu¬ 
 brium possesses the greatest dorso-ventral diameter. The posterior half of each lateral 
 margin supports a large and a small facet for the first sternal rib. 
 
 Table 25 .—Measurements (in millimeters) of sternal elements 1 of N. sliastense. 
 
 
 Manu¬ 
 brium 
 No. 1876-1 
 
 Manu¬ 
 brium 
 No. 1876-2 
 
 Manu¬ 
 brium 
 No. 1876-3 
 
 Second 
 segment 
 No. 1896-1 
 
 Third 
 segment 
 No. 1897-1 
 
 Fourth 
 segment 
 No. 1898 
 
 Fifth 
 segment 
 No. 1899-1 
 
 Sixth 
 segment 
 No. 1900-1 
 
 Greatest length through middle... 
 
 80.5 
 
 95.6 
 
 89.4 
 
 
 
 
 
 
 Greatest width. 
 
 69.2 
 
 82 
 
 67.7 
 
 
 
 
 
 
 Greatest thickness. 
 
 Width of facet for second sternal 
 
 31.3 
 
 36 
 
 30.3 
 
 
 
 
 .... 
 
 .... 
 
 segment. 
 
 33 
 
 31 
 
 27 
 
 
 
 .... 
 
 
 
 Greatest length of dorsal plate. . . 
 Greatest width of dorsal plate at 
 
 
 .... 
 
 .... 
 
 32.5 
 
 35.4 
 
 43.2 
 
 34.4 
 
 29.9 
 
 posterior end. 
 
 
 
 .... 
 
 35.8 
 
 2 38.5 
 
 52.8 
 
 41.9 
 
 39 
 
 Greatest depth. 
 
 Antero-posterior diameter of ven- 
 
 .... 
 
 .... 
 
 .... 
 
 36 
 
 35.8 
 
 38.8 
 
 37.9 
 
 37.4 
 
 tral process. 
 
 Transverse diameter of ventral 
 
 .... 
 
 .... 
 
 .... 
 
 27.6 
 
 26.2 
 
 31.4 
 
 19.2 
 
 15.8 
 
 process. 
 
 .... 
 
 .... 
 
 
 30.1 
 
 26.3 
 
 35.1 
 
 30.8 
 
 19.1 
 
 1 Mesosternal elements from several individuals. 2 Approximate. 
 
 At the middle of the posterior margin is a wide facet articulating with the second 
 sternal segment. This surface is concave transversely and may be incompletely divided 
 into two parts by a median notch, as in No. 1876-4. In five specimens from Rancho La 
 Brea the facet varies considerably in width. In the manubrium of Mylodon harlani 
 the articulating surface for the second sternal segment is directed posteriorly and the 
 ventral surface of the first element lies, therefore, in practically the same plane as that 
 of the mesosternal series. In Nothrotherium, however, the comparable facet of the 
 manubrium is directed for the most part not posteriorly, but decidedly more to the 
 dorsal side, as in Hapalops elongatus, and indicates that the first segment is sharply 
 deflected from the sternal series behind it. In Megatherium americanum the posterior 
 articulating surface is also directed more to the dorsal side, much as in N othr other ium, 
 although it differs in shape. 
 
 The second sternal segment differs from that in Mylodon in possessing a single 
 articulating surface for the manubrium, namely, along the anterior side of the dorsal 
 plate. In other words, there is no connection between the ventral process and the 
 element in front. The third sternal segment is relatively long. Both 2 and 3 increase 
 in width toward the posterior end. The fourth sternal segment is shorter than the 
 third, but relatively broader transversely. 
 
 CLAVICLE. 
 
 The clavicle (plate 8, fig. 4) is similar in appearance and in curvature to that of 
 Mylodon. The shaft is fairly uniform in width, but the sternal extremity is expanded 
 and relatively heavy. 
 
08 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 SCAPULA. 
 
 Three or four nearly complete scapulae (plate 8, figs. 3, 3a) of Nothrotherium are 
 available. These specimens are strikingly similar on the whole to the shoulder-blade 
 in Hapalops. The spine is high and, as in the Miocene genus, divides the outer face 
 into a large prescapular fossa and a much smaller postscapular fossa. In Megalonyx, 
 according to Leidy (1855, p. 25), “the dorsum (PL viii, Fig. 1), as in Mylodon, is nearly 
 equally divided by the spine into two deeply concave fossae, which are roughened with 
 reticular ridges.” On the inner surface of specimen 1868-R-2 there are two concavities, 
 the larger and broader being below the prescapular fossa, while the smaller is below 
 the postscapular fossa. The surface is marked by several longitudinal ridges and 
 reticulations. 
 
 Table 26. —Measurements (in millimeters) of clavicle of N. 
 
 shastense. 
 
 
 1878-R-2 
 
 1878-R-4 
 
 Greatest length. 
 
 140.6 
 
 
 Greatest width of outer end. 
 
 20.3 
 
 .... 
 
 Greatest width at middle of shaft. 
 
 18.2 
 
 17.5 
 
 Thickness at middle of shaft. 
 
 11.5 
 
 9.2 
 
 Greatest width of sternal end. 
 
 x 32.8 
 
 37.3 
 
 Thickness of sternal end. 
 
 24.6 
 
 20 
 
 1 Approximate. 
 
 The spine is connected anteriorly with the coracoid by the coraco-acromial arch. 
 In the extent to which the arch projects below the glenoid cavity Nothrotherium appears to 
 be more like Hapalops than like Choloepus. It differs from Hapalops in having a broader 
 coracoid portion. In the region of its greatest width the arch is relatively wider than in 
 Megalonyx. Along the lower portion of the inferior border of the arch there is a prominent 
 knob-like projection which bears on the anterior side an articulating surface for the clavi¬ 
 cle. In the scapula (1868-L-3) of a young individual that portion of the acromion which 
 possesses the facet for the clavicle is lacking. The appearance of the end of the scapular 
 spine and of the end of the coracoid to which the acromial arch ordinarily joins leads one 
 to believe that the intervening piece ossifies separately and only in the mature animal 
 does it fuse completely with the adjacent structures to form the acromial arch. The 
 coraco-scapular foramen is oval in shape, with the long axis extending dorso-ventrally. 
 
 Table 27. —Measurements (in millimeters) of scapula of N . shastense. 
 
 
 No. 
 
 1868-R-l 
 
 No. 
 
 1868-R-2 
 
 No. 
 
 1868-L-l 
 
 No. 
 
 1868-L-3 
 
 Length from superior border of glenoid cavity to suprascapular border and 
 along the spine. 
 
 *240 
 
 *236 
 
 *234 
 
 *234.5 
 
 Length from calvicular facet to suprascapular border, measured along spine. 
 Greatest width of blade, measured between ends of suprascapular border.. 
 Width measured below base of spine. 
 
 149.8 
 
 ! 311 
 l 27‘2.7 
 144.3 
 
 1 304.5 
 1 250.4 
 128.7 
 
 *249.6 
 142.5 
 
 Height from middle of inferior border of glenoid cavity to point directly 
 above on surface of acromian process. 
 
 J 95 
 
 100 
 
 ’90.5 
 
 J 97 
 
 Greatest antero-posterior diameter of glenoid cavity. 
 
 71 
 
 70 
 
 74 
 
 73.7 
 
 Greatest transverse diameter of glenoid cavity. 
 
 45.3 
 
 49.7 
 
 48 
 
 50.3 
 
 Greatest width of coraco-acromial arch. 
 
 61.4 
 
 J 54.8 
 
 Greatest diameter of coraco-scapular foramen. 
 
 31.7 
 
 29.7 
 
 1 Approximate. 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 69 
 
 It may be relatively large in some specimens, being comparable in size to the foramen 
 in the much larger scapula of Mylodon harlani. 
 
 In none of the specimens is the coraco-scapular suture visible. The extent of the 
 coracoid anterior to the coraco-scapular foramen is relatively as great as in the scapula 
 of Hapalops. In specimen 1868-L-3, although the suture between coracoid and 
 scapula has become obliterated, the epiphysis forming the glenoid cavity is still distinct. 
 The glenoid epiphysis exists separate from the coracoid, as in Megalonyx jeffersonii, 
 but there is no indication of a division of this epiphysis into two parts, as in the former. 
 The glenoid cavity is broadest opposite the base of the spine and narrows anteriorly. 
 
 HUMERUS. 
 
 Several specimens in the collections of the Los Angeles Museum represent the 
 upper arm-bone of Nothrotherium (plate 8, figs. 1, la, 2, 2a). These include a complete 
 humerus (No. 1874-R-l) and a larger specimen (No. 1874-R-2), without proximal 
 tuberosities and head, and evidently belonging to an older individual than the former. 
 There are also two specimens belonging to young individuals in which the epiphyses 
 are absent. 
 
 Table 28. —Measurements (in millimeters ) of humerus. 
 
 
 Nothrotherium 
 
 shastense, 
 
 No. 1874-R-l 
 
 Nothrotherium 
 
 shastense, 
 
 No. 1874-R-2 
 
 Hapalops sp., 
 No. 15532, Prin. 
 Univ. Coll. 
 
 Hapalops 
 longiceps 
 (after Scott). 
 
 Greatest length. 
 
 432.4 
 
 
 179 
 
 192 
 
 Width at proximal tuberosities. 
 
 106.2 
 
 
 50.8 
 
 
 Width of shaft at middle. 
 
 55.3 
 
 67.8 
 
 27 
 
 
 Least width of shaft. 
 
 50.3 
 
 55.6 
 
 19.8 
 
 
 Thickness of shaft at end of deltoid crest. .. . 
 
 39 
 
 49.3 
 
 29.8 
 
 
 Greatest width of distal expansion. 
 
 168 
 
 *192.4 
 
 73.5 
 
 79 
 
 Width of distal trochlea. 
 
 95.9 
 
 109.6 
 
 39.4 
 
 44 
 
 1 Approximate. 
 
 The striking difference between the humerus of Nothr other ium and that of Megalonyx 
 is the more slender construction of the element in the former genus. Although No. 
 1874-R-l is but a trifle shorter than the humerus, No. 21003 U. C. C., of Megalonyx 
 (compare with plate 19, figs. 1, la, 16, 2), and shorter than specimens of M. jeffersonii 
 described by Leidy, the shaft is comparatively slender and the distal third much less 
 expanded. Nothrotherium differs more from Hapalops in slenderness of shaft of humerus 
 than does Megalonyx. In anterior view the head of the humerus does not appear as 
 prominent as in the Miocene genus and the tuberosities are not separated so much and 
 are not quite so prominently developed. A difference between the upper arm-bone of 
 Nothrotherium and that of Megalonyx from Rancho La Brea is seen in the deep groove 
 which separates the lesser tuberosity from the head and extends outward and in front 
 of the head. This groove is also absent in the humerus of Hapalops. Between the 
 greater tuberosity and head no deep gutter exists, but the slight groove which is present 
 is more distinct than in the humerus of Megalonyx. Anterior to the middle of the head 
 the tuberosities are connected by a broader shelf than in Megalonyx. 
 
 A deep groove extending parallel to the shallow bicipital groove is present on the 
 anterior face of the lesser tuberosity. This channel is lacking in the humerus of 
 Megalonyx. No deep groove is present in the humerus of Hapalops available for 
 comparison, but in the specimen a broad, shallow groove or depression extends obliquely 
 downward and across the anterior surface of the lesser tuberosity from its upper border. 
 
 In No. 1874-R-l the pectoral and deltoid ridges are but feebly developed, while 
 in an older individual (No. 1874-R-2) they become more rugose. In neither of the 
 
70 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 two specimens from the asphalt beds do the ridges meet below in such a decided eminence 
 as in Hapalops. This eminence is also of less height than in the humerus of Megalonyx. 
 In the Miocene genus the pectoral ridge extends as a very pronounced crest along the 
 anterior face of the humerus from between the proximal tuberosities, while in Nothro- 
 therium (No. 1874-R-l) it is slightly closer to the inner border of the shaft and is by 
 no means so prominent. In a specimen from Hawver Cave (Stock, 1918, p. 504, figs. 
 24a, 245) the pectoral ridge forms the antero-internal border above the middle of the 
 shaft. The deltoid ridge in the young individual is very little developed and the mus- 
 culo-spiral course is hardly outlined, while in a mature form (No. 1874-R-2) these struc¬ 
 tures are much better indicated. The shaft in the older individual is more subquad¬ 
 rate or trapezoidal in cross-section. 
 
 The distal portion of the humerus in Nothrotherium is relatively not so expanded 
 transversely as it is in that of Megalonyx. The internal tuberosity is not so massive 
 as in the latter genus. An entepicondylar foramen is present, as in Hapalops and 
 Megalonyx. The bar over this foramen is slender. The ulnar articulation surface of 
 the distal trochlea differs from that in Megalonyx in being approximately flat transversely 
 and in sloping downward and inward, reaching a point at or slightly below the level of 
 the radial articulation. It is, however, distinctly convex antero-posteriorly, as in 
 Megalonyx. The radial surface is relatively smaller and the trochlea not thick antero- 
 posteriorly, as in Hapalops. The olecranon fossa resembles that in the humerus of 
 Hapalops and is deeper than in the specimen of Megalonyx from Rancho La Brea. 
 
 Table 29. —Measurements {in millimeters) of ulna of N. shaslense. 
 
 
 No. 
 
 1873-R-l 
 
 No. 
 
 1873-L-l 
 
 No. 
 
 1873-L-2 
 
 No. 
 
 1873-L-3 
 
 No. 
 
 1873-L-4 
 
 Greatest length. 
 
 
 448.3 
 
 442.5 
 
 
 
 Greatest width from summit of coronoid process. 
 
 94.3 
 
 93.5 
 
 92 
 
 94.6 
 
 83 
 
 Width of shaft at middle. 
 
 50.5 
 
 44.8 
 
 45.5 
 
 Thickness of shaft at middle. 
 
 34.1 
 
 31 
 
 30.9 
 
 
 
 Greatest width of shaft at distal end. 
 
 43 
 
 44.8 
 
 
 
 Thickness of distal articulating surface. 
 
 
 31.7 
 
 40 
 
 
 
 Distance from summit of coronoid process to end of olecranon. 
 
 122 
 
 114 
 
 124.8 
 
 114.5 
 
 105 
 
 ULNA. 
 
 The ulna (plate 9, figs. 1, 2, 3) differs decidedly from that of Megalonyx in the 
 much less antero-posterior diameter, but noticeably greater transverse thickness of 
 shaft. The element thus resembles the ulna of Hapalops more in lateral view. It 
 lacks, however, the distinct sigmoid curvature which the latter possesses, and does not 
 have so prominent an olecranon. The latter process is very short in the Pleistocene 
 genus and is expanded in transverse direction. Such expansion of the olecranon is 
 absent in the ulna of Hapalops , according to Scott. The coronoid process is prominent 
 in the ulna of Nothrotherium. A line drawn through this process and through the posterior 
 end of the sigmoid notch forms an acute angle with the main axis of the bone. In the 
 ulna of Hapalops a corresponding line is nearly parallel to the main axis. 
 
 A difference from the ulna of Megalonyx is seen in the absence or small development 
 of the surface articulating with the radius. In its place there is sometimes a deep 
 depression. The outer surface is plane or concave, and in specimen No. 1873-L-l is 
 marked by reticular ridges. The distal articulating surface is semioval, concave, and 
 is relatively not so extensive as that of Megalonyx. 
 
 RADIUS. 
 
 The radius (plate 9, figs. 4, 5) is not represented by complete material, but the 
 specimens available supply information for all parts of the element. The lower two-thirds 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 71 
 
 in one specimen from Rancho La Brea is relatively more slender than that in the radius 
 of either Hapalops or Megalonyx. The shaft of the radius is strongly curved. The 
 bicipital tubercle is large as in Hapalops and placed on the posterior side of the shaft. 
 It is not as prominent as in H. longiceps and is situated relatively not so far below the 
 head as in that species. In its proximal and distal thirds the posterior surface of the 
 shaft is convex, while the middle third is flatter. In No. 1872-R-l the latter surface 
 is marked by a number of reticular ridges. The anterior face of the shaft is also 
 
 Fig. 21 .—Nothrotherium shastense Sin¬ 
 clair. Left scaphoid, No. 
 1944-L-l. A, distal view; B, 
 dorsal view; C, proximal view. 
 X 0.50. Rancho La Brea 
 Pleistocene. 
 
 B 
 
 i 
 
 C 
 
 i 
 
 Hapalops sp. Left scaphoid, No. 
 15171 P. U. C. A 1 , B l , C 1 , same views 
 as in A, B, C. X 1.0. Santa Cruz 
 Miocene. 
 
 roughened and is traversed obliquely in the lower half by a broad groove. The surface 
 which articulates with the humerus is oval in shape and moderately concave. The 
 facet for the ulna is not large, the radius thus resembling that of Hapalops. 
 
 The distal extremity is wide and thick. It bears an undivided carpal facet which 
 is concave, widest at the ulnar border, and descending to the styloid process. It is of 
 greater antero-posterior extent toward the inner side than that in the radius of Megalonyx. 
 
 Table 30. —Measurements {in millimeters) of radius of 
 
 N. shastense. 
 
 
 No. 
 
 1872-R-l 
 
 No. 
 
 1872-L-l 
 
 Greatest length. 
 
 *415 
 
 
 Width at middle of shaft. 
 
 52 
 
 44.6 
 
 Thickness at middle of shaft. 
 
 30.8 
 
 21.6 
 
 Greatest width of distal end. 
 
 78.5 
 
 » • • • 
 
 Greatest thickness of distal end. 
 
 56.7 
 
 
 Greatest width of proximal end. 
 
 .... 
 
 48.9 
 
 Greatest thickness of proximal end. 
 
 
 44.6 
 
 1 Approximate. 
 
 SCAPHOID. 
 
 The scaphoid of Nothrotherium is known by two specimens from the Pleistocene of 
 California; the first specimen, No. 1944-L-l (fig. 21 a, b, c) from Rancho La Brea 
 and in the collections of the Los Angeles Museum, the second specimen, No. 9437, 
 Univ. Calif. Coll., from Samwel Cave. 
 
 The scaphoid of Nothrotherium is decidedly smaller than that of Megalonyx jeffersonii 
 and is also smaller than that of M. sierrensis (W. J. Sinclair, 1905, plate 20, figs. 7 and 8). 
 In proximal or radial view (fig. 21 c), the scaphoid differs noticeably from that of M. 
 jeffersonii in shape and in the relatively smaller extent of the radial surface. At the 
 
72 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 outer dorsal side there is a more pronounced extension of the scaphoid than in M. jeffer- 
 sonii. The element in M. sierrensis is somewhat more like that of Nothrotherium than is 
 M. jeffersonii, but also lacks the conspicuous projection of the outer dorsal border. 
 
 The outer dorsal projection of the scaphoid is much better developed than in 
 Hapalops (fig. 21, compare a, b, c with a 1 , b 1 , c 1 ), the latter form being in this respect 
 more like Megalonyx. The distal surface of this process in Nothrotherium shastense (fig. 
 21 a) is sigmoid transversely and articulates with the magnum. It is bordered on its 
 outer side by a small facet for the lunar. Along the inner side it is continuous with, 
 but also separated by a sharp angle from, an elongate facet for the trapezoid, as in M. 
 jeffersonii. 
 
 Table 31. —Measurements (in millimeters ) of scaphoid of 
 
 N. shastense. 
 
 
 No. 1944-L-l 
 
 No. 9437 
 
 
 L. A. M. 
 
 u. c. c. 
 
 Greatest dorso-palmar diameter. 
 
 49.5 
 
 49.2 
 
 Greatest transverse diameter. 
 
 54.5 
 
 51.7 
 
 Greatest proximo-distal diameter. 
 
 34 
 
 32 
 
 The facet for the trapezoid in the specimen from Rancho La Brea extends to the 
 palmar border, where it lies adjacent to a small flat articulating surface for the magnum. 
 In the scaphoid of Nothrotherium from Samwel Cave the surface for the trapezoid is 
 divided into two parts. In Hapalops (fig. 21 a 1 ) a second small facet for the trapezoid 
 is situated close to the palmar border. The absence of a distinct articulating surface 
 for the magnum on the scaphoid of Hapalops permits the dorsal trapezoid facet and 
 the lunar facet to meet in a ridge. The dorsal 
 surface for the magnum is well developed in both 
 Nothrotherium and Megalonyx. In M. jeffer¬ 
 sonii, as shown by Leidy (1855, plate 8, fig. 7), 
 there is present a second facet for the magnum 
 close to the palmar border and separated from 
 the first facet for the magnum by a transverse 
 gutter. The presence of a palmar facet for the 
 magnum is seen also in the scaphoid of Nothro¬ 
 therium from Rancho La Brea, but it is not 
 clearly indicated in the specimen from Samwel 
 Cave. 
 
 The facet for the trapezium is relatively smaller than in Hapalops. It is less 
 extensive in the specimen from Rancho La Brea than in that from Samwel Cave. As 
 will be shown below, the trapezium in Nothrotherium may join with metacarpal I to form 
 a single element and does not unite, therefore, with the scaphoid. There is reason for 
 believing that a similar union occurs in Megalonyx and the inner proximal carpal element 
 should not be designated “scaphotrapezium. ” 
 
 TRAPEZIUM AND METACARPAL I. 
 
 Only one specimen of a trapezium of Nothrotherium is known from the asphalt 
 deposits, and it is fused with metacarpal I to form a single element (fig. 22, a to d). 
 Such fusion commonly occurs in the manus of Mylodon, but the two elements are separate 
 in a small number of mylodont forms from Rancho La Brea. The presence in the 
 collections from the asphalt of a single metacarpal I, (No. 1912-L-l) of Nothrotherium 
 indicates also that the trapezium and the first metacarpal are not always joined in this 
 genus. In the individual of Nothrotherium described by Winge (see p. 78, fig. 28), 
 
 Fig. 22 .—Nothrotherium shastense Sinclair. Right 
 trapezium and metacarpal I, Nos. 1912-R-I, 
 1913-R-l. A, dorsal view; B, proximal end; C, 
 distal end; D, inner view. X 0.50. Rancho La 
 Brea Pleistocene. 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 73 
 
 the trapezium and the first metacarpal are fused and this element is in articulation with 
 the scaphoid. In the manus of Hapalops the two elements are not united, and presumably 
 the separation is the usual occurrence in the Miocene genus. The trapezium in this 
 genus articulates with the side of the second metacarpal. 
 
 At the proximal end of the trapezial portion of the compound element (No. 1912-R-l, 
 1913-R-l) is located a narrow articulating surface, elongated in dorso-ventral axis, 
 somewhat constricted at the middle, and deeply concave. The facet joins with the 
 scaphoid. On the inner side of the trapezium is a lunate articulating facet which joins 
 with the trapezoid. 
 
 Metacarpal I is much shortened and reduced. The proximal and distal articulating 
 surfaces in the single available specimen which is not fused with the trapezium, closely 
 resemble the corresponding faces in the first metacarpal of Mylodon. The former surface 
 in Nothrotherium is narrow, elongated in dorso-ventral axis, and constricted at the middle 
 with sigmoid curvature. The distal face presents a convexity with flattened base. In 
 Mylodon a similar convexity at the end of the first metacarpal carries a digit of two 
 segments, the distal one of which is an ungual phalanx. If the structure in Nothrotherium 
 is directly comparable to that in Mylodon, the ungual phalanx must have been very 
 small. 
 
 METACARPAL II. 
 
 If metacarpal II of Nothrotherium (fig. 23, a to e) is so oriented that the proximal 
 extremity corresponds in position to that of Hapalops, the distal trochlea is directed 
 inward much more obliquely (i. e., the lower end of trochlea is directed downward 
 and inward) with reference to a vertical plane passing through the metacarpal than 
 in the Miocene form. The trapezoid surface is somewhat more uniform than in Hapalops, 
 lacking the deep median vertical groove. Towards the external side this surface is 
 sigmoid in a dorso-ventral direction, while internally it is flatter. The dorso-external 
 process of the proximal end is not quite so prominent as in the metacarpal of the Miocene 
 form. In slenderness of the second metacarpal, and in the structure of its proximal 
 extremity, Nothrotherium shows greater change from Hapalops than does Megalonyx. 
 The character of rotated distal extremity of the metacarpal is possessed in greater 
 degree by both Pleistocene genera than by the Miocene form. 
 
 Table 32 .—Measurements {in millimeters) of N. shastense. 
 
 Fused trapezium and metacarpal I (No. 1912-R-l, 
 
 1913-R-l): 
 
 Greatest length. 52.7 
 
 Greatest depth of proximal end. 27.6 
 
 Width of proximal end. 20.3 
 
 Depth of distal end. 20.6 
 
 Width of distal end. 14.2 
 
 Metacarpal I (No. 1912-L-l): 
 
 Greatest length. 30.5 
 
 Depth of proximal end. 20.8 
 
 Width of proximal end. 18 
 
 Depth of distal end. 19.2 
 
 Width of distal end. 13.5 
 
 In Nothrotherium the facet on the outer side for metacarpal III is less extensive 
 and the facet on the inner side is also much smaller than the corresponding facets in 
 Hapalops. The latter facet is concave dorso-ventrally and articulates with the trapezium, 
 which in Nothrotherium has fused with the first metacarpal. Both inner and outer facets 
 are continuous with the trapezoid surface. The shaft, in contrast to that of metacarpal 
 II of the Miocene genus and of Megalonyx is less quadrate and more nearly oval in section. 
 At the distal end the carina is directed inward more than in Hapalops. A well-formed 
 offset is present along the inner side, and along the lower half of the outer side a much 
 narrower offset occurs. 
 
 The second metacarpal of Nothrotherium exhibits further differences from that of 
 Megalonyx in the less triangular shape of the trapezoid surface and in the smaller extent 
 of the outer facet for the median metacarpal. 
 
74 
 
 CENOZOIC GRAYIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 Table 33. —Measurements {in millimeters ) of metacarpal II. 
 
 
 N o throtherium 
 shastense, 
 
 No. 1914-L-l 
 
 N. shastense, 
 No. 1914-L-2 
 
 Megalonyx 
 jeffersonii, 
 (after Leidy). 
 
 M. jeffersonii, 
 (after Leidy). 
 
 M. jeffersonii, 
 (after Leidy). 
 
 Greatest length. 
 
 79 5 
 
 81.5 
 
 95.3 
 
 82.6 
 
 88.9 
 
 Depth of proximal end. 
 
 29 
 
 26.4 
 
 44.5 
 
 44.5 
 
 44.5 
 
 Width of proximal end. 
 
 24.6 
 
 25.5 
 
 
 Depth of distal end. 
 
 45.8 
 
 43.2 
 
 50.8 
 
 50.8 
 
 44.5 
 
 Width of distal end. 
 
 28.5 
 
 26.8 
 
 Least depth of shaft. 
 
 23.4 
 
 
 
 
 
 
 
 
 
 PHALANX II, DIGIT II, MANUS. 
 
 The second phalanx of the index-finger (No. 1930-L-l, fig. 24a) is long and slender, 
 exceeding in length the corresponding segment in digit III, manus, and the coossified 
 first and second phalanges in digit III, pes. The proximal end is not very deep and 
 supports two facets meeting in a thick median ridge. The shaft is distinctly constricted 
 dorso-ventrally immediately behind the distal trochlea. The two condyles are separated 
 by a median U-shaped groove. 
 
 Fig. 23. —Nothrotherium shastense Sinclair. Left metacarpal II, No. 1914-L-l. A, inner 
 view; B, dorsal view; C, outer view; D, proximal end; E, distal end. X 0.50. 
 
 Rancho La Brea Pleistocene. 
 
 Hapalops sp. Left metacarpal II, No. 15171 P. U. G.; same views as in A, B, C, D, 
 
 E. X 1.0. Santa Cruz Miocene. 
 
 The measurements (in millimeters) of phalanx II, digit II, manus of N. shastense, 
 are: length through middle, 66.5; width of proximal end, 25; approximate depth of prox¬ 
 imal end, 32; depth of distal end, 26.7. 
 
 PHALANX III, DIGIT II, MANUS. 
 
 Perhaps the most remarkable ungual phalanx of Nothrotherium, peculiar because of 
 its decided difference from all other terminal phalanges, is that borne by the second 
 digit of the hand. This specimen (No. 1931-L-l, fig. 25a) is identical with that shown 
 by Reinhardt (1878, plate 4, figs. 8 and 9) and by Winge (1915, plate 24). In the manus 
 of Hapalops the terminal phalanx of the second digit is no wider than other phalanges. 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 75 
 
 In Prepotherium, however, the corresponding ungual is broadened, but appears to have 
 a somewhat different shape from that in Nothrotherium. 
 
 Contrasted with the ungual phalanx of digit II, pes, of N othr other ium, the comparable 
 phalanx in the manus is somewhat longer and of less depth. Its most peculiar 
 characteristics are the wide and well-rounded claw-process and the great projection of 
 this process beyond the hood. The latter projection is equal to that in the ungual 
 phalanx of the median digit of the pes, although the latter segment is much larger. 
 The ventral surface of the process just anterior to the base of the hood is not so broad 
 as in the corresponding phalanx of the median digit, pes, but at the tip it is much broader. 
 
 There is no subungual process, the ventral surface of the hooded region being 
 nearly flat. A distinct overhanging process at the proximal end is absent and the two 
 articulating concavities are not directed downward as much as in other ungual phalanges 
 of Nothrotherhmi. The median ridge is heavy. 
 
 Fio. 24 .—Nothrotherium shastense Sinclair. Second 
 phalangeal elements of manus; proximal, 
 lateral, and distal views. X 0.50. a, No. 
 1930-L-l, digit II; b, No. 1929-L-l, digit III; 
 c, No. 1928-L-l, digit IV. Rancho La Brea 
 Pleistocene. 
 
 Fia. 25 .—Nothrotherium shastense Sinclair. Ungual pha¬ 
 langes of manus, lateral views with cross-sections 
 through claw-processes. X 0.50. a, digit II; b, digit 
 III; c, digit IV. Rancho La Brea Pleistocene. 
 
 The measurements (in millimeters) of No. 1931-L-l, phalanx III, digit II, manus, 
 of N. shastense are as follows: greatest length from posterior end of overhanging 
 process to tip of claw-process (approximate) 115.6; depth of proximal end 33.2; width 
 of proximal end 26.7; width of ventral surface of claw-process just anterior to hood 16. 
 
 PHALANX II, DIGIT III, MANUS. 
 
 This phalanx (No. 1929-L-l, fig. 24 b) is slightly longer and heavier than the 
 corresponding segment in the second digit, pes. It is shorter and stouter than phalanx 
 II, digit II, manus. The proximal end is deeply excavated where the facets articulate 
 with phalanx I. The median ridge is narrow at the middle, broadening above and more 
 so below. Dorsal and ventral margins are straight, except where they are notched in 
 median vertical axis. The distal trochlea is deep and relatively narrow as contrasted 
 with the distal end of phalanx II, digit II, manus. The median furrow in No. 1929-L-l 
 is deep and narrow. 
 
 The measurements (in millimeters) of No. 1929-L-l, phalanx II, digit III, manus, 
 of N. shastense are as follows: length through middle 56; width of proximal end 31; 
 depth of proximal end 34; depth of distal end 29.3. 
 
76 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 PHALANX III, DIGIT III, MANUS. 
 
 The ungual phalanx (No. 1890-Lr-l, fig. 256) is larger than any of the terminal 
 phalanges of the fore and hind feet, with the exception of phalanx III, digit III, pes. 
 Contrasted with the latter, No. 1890-L^l is considerably smaller. The phalanx is 
 relatively narrow. At the proximal end the articulating surfaces are subequal. Within 
 the hood the claw-process possesses a convex dorsal surface which is transformed 
 anteriorly into a sharp edge extending to the tip. The end of the claw-process reaches 
 well below the horizontal plane of the base, in which character No. 1890-L-l differs 
 from the terminal phalanges of other digits. A distinct tuberosity is situated at 
 approximately the middle of the ventral surface of the subungual base. On each side 
 posteriorly is a subungual foramen. 
 
 The measurements (in millimeters) of No. 1890-L-l, phalanx III, digit III, manus, 
 of N. shastense, are as follows: greatest length from proximal end of overhanging 
 process to tip of claw-process 127.5; greatest depth from dorsal surface of hood to ventral 
 tuberosity (approximate) 53; width of proximal end 28; length of subungual base 54.5; 
 width of ventral surface of claw-process just anterior to hooded region 12.5. 
 
 PHALANX I, DIGIT IV OR DIGIT III, MANUS. 
 
 A single specimen (No. 1932-1^1) appears to represent the proximal segment in 
 the annular finger, but may belong to the median digit. It is deeper than the 
 corresponding element (No. 1886-R-l) in the second digit, pes. A deep median furrow, 
 which narrows below, traverses the length of the proximal end. Along the lower half 
 of one side, presumably the inner side, an articulating offset occurs. Below this and 
 on the ventral side is a facet, transversely elongated, for a sesamoid. A larger facet 
 for a sesamoid is present on the opposite side. A greater extent of bone is present above 
 the distal articulation in No. 1932-L-l than in No. 1886-R-l. The distal articulation 
 is not oblique with respect to a median plane passing through the proximal groove. 
 
 The following are the measurements (in millimeters) of phalanx I, digit IV or 
 digit III, manus, of N. shastense, (No. 1932-1^1): depth 46.5; greatest length 26.9; 
 greatest width 35. 
 
 PHALANX II, DIGIT IV, MANUS. 
 
 This segment (No. 1928-L-l, fig. 24c) is the shortest and stoutest of the second 
 phalanges in the manus of Nothrotherium. The proximal end is greatly expanded dorso- 
 ventrally and the articulating facets are relatively large. At the distal end the two 
 condyles diverge ventrally, the outer condyle extending farther downward and having 
 a larger articulating surface than the inner condyle. 
 
 The following are the measurements (in millimeters) of phalanx II, digit IV, manus, 
 of N. shastense (No. 1928-L-l): length through middle 37; width of proximal end 
 33; depth of proximal end 39.5; depth of outer condyle 30.2. 
 
 Table 34. — Measurements {in millimeters) of 'phalanx III, digit IV, manus, of N. shastense. 
 
 
 No. 
 
 1933-R-l 
 
 No. 
 
 1933-R-2 
 
 No. 
 
 1933-L-l 
 
 Greatest length from posterior end of overhanging process to tip of claw-process. 
 
 101.6 
 
 100 
 
 *93 
 
 Greatest depth from dorsal surface of hood to discoid area on ventral surface. 
 
 43.8 
 
 44.8 
 
 41 
 
 Width of proximal end. 
 
 31.4 
 
 29.6 
 
 27 
 
 Length of subungual base. 
 
 .... 
 
 . . . . 
 
 47.6 
 
 Width of ventral surface of claw-process just anterior to hood. 
 
 12 
 
 11 
 
 11 
 
 ■* Approximate. 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 77 
 
 PHALANX III, DIGIT IV, MANUS. 
 
 The terminal phalanx (No. 1933-L-l, fig. 25c) of the annular finger is the smallest 
 of the series of functional ungual phalanges of the manus and pes. It approaches closest 
 in size the terminal phalanx (No. 1888-R-l) of digit II, pes, differing chiefly from this 
 segment in the character of proximal end and in the structure of the claw-process. 
 The overhanging process is slender. The surface articulating with the outer condyle of 
 phalanx II is narrow above and widens below. It reaches below the lower margin of 
 the inner articulating facet. The latter facet is deeply concave. A median plane passed 
 through the ridge between the two articulating facets is oblique to a median plane 
 through the phalanx along the fore-and-aft axis. 
 
 The anterior end of the claw-process does not extend below the horizontal plane 
 of the ventral surface of the subungual base as in No. 1888-R-l. The claw-process 
 is relatively thick, and the sides of that portion projecting from the bony hood meet in 
 a sharp edge. The inner of the two subungual foramina is the larger. The hood is 
 well developed. The phalanx may show slight dorso-ventral constriction above the 
 anterior border of the proximal articulation. For measurements see table 34. 
 
 Fig. 26 .—Nothrotherium shastense Sinclair. Right metacarpal V, No. 1917-R-l. 
 
 A, dorsal view; B, inner view; C, outer view. X 0.50. Rancho La Brea 
 Pleistocene. Hapalop-s, sp. Left metacarpal V, No. 15171 P. U. C. A 1 , dorsal 
 view, inner view. X 1.0. Santa Cruz Miocene. 
 
 METACARPAL V. 
 
 Only one specimen (No. 1917-R-l, fig. 26, a, b, c) of this metacarpal is available 
 for study. The metapodial lacks the distal epiphysis. Metacarpal V is a slender element, 
 much as in Hapalops. In Megalonyx the metapodial is somewhat heavier. The proximal 
 extremity is distinctly less expanded transversely than in Hapalops and does not possess 
 such an extensive articulating surface. The latter surface joins only with the unciform 
 of the carpal elements, as shown in Winge’s figure (Winge, 1915, plate 24) of the manus 
 of Nothrotherium maquinense. In Hapalops also the fifth metacarpal, as seen in specimen 
 15171, Princeton University Collection, does not touch the unciform. In the fifth 
 metacarpal of Megalonyx, however, a small facet is bent away from the unciform surface 
 and articulates with the cuneiform. Along the inner side of the proximal end is the 
 surface for metacarpal IV, which extends distally in its course from the ventral to the 
 dorsal side. It is flattened below and becomes broader and concave above. Dorsally, 
 the proximal extremity is grooved, while laterally it becomes a discoid flattened surface. 
 The shaft is cylindroid, with flattened inner side. The distal end of metacarpal V in 
 Nothrotherium is expanded in dorso-ventral extent as in that of Megalonyx, and the 
 principal axis of this extremity is oblique to the dorso-ventral plane of the shaft. In 
 Hapalops the metacarpal apparently carried a complete digit, but there is still doubt 
 as to the nature of this digit in the Pleistocene megalonychids. 
 
78 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 The measurements (in millimeters) of No. 1917—R—1, metacarpal V of N. shastense, 
 are as follows: Length (approximate), 105.7; transverse diameter of proximal extremity, 
 24.1; dorso-ventral diameter of proximal extremity, 24.9; width of shaft at middle, 
 13.4; depth of shaft at middle 16.2. 
 
 COMPARISON OF MANUS. 
 
 Unfortunately, the manus of Nothrotherium from Rancho La Brea is not known by 
 as complete material as the pes. Of the carpus, only the scaphoid and tiapezium, the 
 latter fused with metacarpal I, are available. The 
 metacarpal series includes the first, second, fourth, 
 and fifth, and the median three digits are fairly 
 complete. 
 
 Judging from the Californian material and from 
 the Brazilian collection described by Winge (1915), 
 the manus (fig. 28) is like that of Hapalops, but is 
 more specialized. Contrasted with the hand of H. 
 longiceps, that of Nothrotherium exhibits a greater 
 amount of interlocking of carpal elements. The 
 scaphoid has broader contact with the magnum and 
 the lunar apparently overlaps to some extent on 
 the unciform. The trapezium is not fused with 
 metacarpal I in Hapalops , while in Nothrotherium, 
 as in Megalonyx, such fusion occurs. 
 
 Fig. 27. — Megalonyx, probably wheatleyi Cope. Right manus, Amer. Mus. Spec., dorsal view. X 0.50. Port Kennedy 
 fissure, Pennsylvania, Pleistocene. 
 
 Fig. 28 .—Nothrotherium maquinense (Lund). Right manus, dorsal view, after Winge. Approx. X 0.66. Brazil, Pleistocene. 
 
 The four outer metacarpals are comparable in their relative lengths to those of 
 Hapalops and of Megalonyx. The metapodials are more slender than those of Mega¬ 
 lonyx (compare figures 27 and 28). This is particularly evident in the fifth metacarpal. 
 Metacarpal I is not so reduced in length as it is in Megalonyx. Metacarpal II differs 
 in shape more from that in Hapalops than does that of Megalonyx. 
 
 The first metacarpal may have carried a small functional digit, but as yet no remains 
 of it have been identified. The second, third, and fourth digits have each the full 
 complement of phalanges. In digit II the second phalanx is relatively very long and 
 slender and the terminal phalanx bears a claw-process which is much wider than in 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCIIO LA BREA. 79 
 
 Hapalops. The ungual phalanx of digit III is the largest of the series in the manus, 
 but is decidedly smaller than that in the third digit of the pes. In this respect Nothro- 
 therium resembles Megalonyx. In Hapalops, phalanx III, digit III, manus, is also 
 smaller than that in the pes, but the difference is not so marked as in the Pleistocene 
 genera. The ungual phalanx of digit IV shows less freedom of movement than in digits 
 III and II. No remains of digit V have been found. The digit may have been more 
 reduced than in Hapalops. 
 
 PELVIS. 
 
 A nearly complete pelvis (plates 10, 11) is available from Rancho La Brea, and a 
 second specimen not so well preserved as the first and belonging to a young animal is 
 also present. Besides this material, there are in the collections from the asphalt deposits 
 a few fragments which fortunately indicate some of the parts missing in the former speci¬ 
 mens. The pelvis of Nothrotherium exhibits many interesting resemblances to that of 
 the Miocene genus Hapalops, with some notable changes in structure from those found 
 in the latter form. 
 
 Nothrotherium is less like the Tardigrada than is Hapalops and resembles more the 
 large Pleistocene Gravigrada in the lateral expansion of the pelvis. The wing-like 
 development of the ilia in the Pleistocene megalonychid from Rancho La Brea is due, 
 as already pointed out by Scott (1903, p. 170) for Megatherium and Mylodon, to the 
 increase in size of the animal with accompanying enlargement of the viscera to be sup¬ 
 ported. In lacking an excessive expansion of the ilium and a great forward projection 
 of the superior border as in Mylodon, the pelvis of N othr other ium reflects the size of the 
 animal, for although the genus is decidedly larger and bulkier than the Miocene form, 
 it is smaller and considerably more slender than Mylodon. 
 
 The anterior or inner surface of the ilium is deeply concave in the area adjacent to 
 the sacrum and has a shallow depression near the outer border. Between these two 
 areas the middle upper portion of the inner surface is convex. A number of low ridges 
 on the inner surface of the ilium tend to converge from the anterior superior border 
 towards the acetabular border. The labium of the ilium is not such a prominent feature 
 as in the pelvis of Mylodon. It is bent forward, and for the greater part of its course is 
 narrow, excepting towards the sacrum, where it increases considerably in width. The 
 gluteal surface faces posteriorly and somewhat upward. This surface is concave over 
 its middle portion and is also traversed by numerous ridges for muscle attachment. 
 
 Nothrotherium has no oblique ridge connecting the labium with confluent articular 
 processes of the anterior sacrals, as in Mylodon robustus. In the former genus the artic¬ 
 ular processes of each of the five sacrals are distinct. The labium does, however, con¬ 
 nect posteriorly with the fused ends of the transverse processes. As seen from above, 
 the four foramina located between the transverse processes and giving exit to the divi¬ 
 sions of the sacral nerves are subcircular or oval in shape and increase but slightly in 
 size posteriorly. The second and third ventral openings for divisions of the sacral nerves 
 are situated about opposite the middle of the sacro-ischiadic foramen, while the first 
 is situated in advance of the forward border of the latter opening. Broad canals lead 
 from these ventral openings to the sacro-ischiadic foramen. The acetabular border is 
 ventral rather than external. The acetabulum is large, and although the sulcus for the 
 round ligament is prominent, the articulating surface in the socket is still quite exten¬ 
 sive. The sulcus incises the articulating surface from the lower posterior or obturator 
 border of the acetabulum. 
 
 In specimen 1892-1 the sacro-ischiadic foramen is of relatively small size. As in 
 Hapalops longiceps, the tuberosity of the ischium in Nothrotherium is large and promi¬ 
 nent, more so than in the pelvis of Mylodon. The descending bar of the ischium, extend¬ 
 ing ventral to the tuberosity and forming the posterior side of the obturator foramen, 
 
80 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 is flattened laterally and is comparatively thin. The bar forming the anterior side of 
 the foramen is rounded. Below the obturator foramina the sides inclosing the pelvic 
 opening meet in the pubic symphysis at an angle of approximately 80 degrees. The 
 obturator foramen is quite large, with its shortest border along the ischium between 
 acetabulum and ischial tuberosity, and the longest border along the pubic bar. As in 
 Mylodon, the upper border of the foramen, situated on the inner side of the acetabular 
 portion of the pubis, is notched by the canal for the obturator nerves and blood-vessels. 
 In posterior view, plate 10, figure 1, the pelvic opening in Nothrotherium is seen to be large 
 and narrows more sharply to the ventral side than in the pelvis of Mylodon. 
 
 Table 35. —Measurements (in millimeters ) of pelvis of N. shastense. 
 
 
 1892-1 
 
 1892-2 
 
 1892-3 
 
 Greatest transverse width across ilia. 
 
 1 846 
 
 
 
 Length from anterior border of ilium to tuberosity of ischium. 
 
 456 
 
 
 
 Width of ilial wing, measured parallel to posterior border. 
 
 318 
 
 
 
 Antero-posterior diameter of ilial wing through middle. 
 
 272 
 
 
 
 Length of posterior descending process of ischium, measured from superior border 
 
 
 
 
 to ventral symphysis. 
 
 285 
 
 
 
 Dorso-ventral diameter of pelvic opening, measured from posterior end of ischial 
 
 
 
 
 symphysis to ventral border of last sacral vertebra. 
 
 222 
 
 
 
 Greatest transverse diameter of pelvic opening, measured between inner surfaces 
 
 
 
 
 of ischia. 
 
 249 
 
 238 
 
 
 Greatest width of posterior end of pelvis, measured across posterior ends of 
 
 
 
 
 ischia. 
 
 346 
 
 1 346 
 
 
 Antero-posterior diameter of left acetabulum. 
 
 89.8 
 
 102.5 
 
 
 Length of fused sacral vertebrae, measured from middle of anterior face of first 
 
 
 
 
 sacral to middle of posterior face of last sacral. 
 
 231 
 
 254 
 
 
 Transverse distance between outermost border of sacro-ischiadic foramen of each 
 
 
 
 
 side, measured across ventral surface of sacrum. 
 
 227 
 
 263 
 
 
 Transverse diameter of centrum of first sacral, measured across anterior face.... 
 
 84.6 
 
 91.8 
 
 
 Depth of centrum of first sacral. 
 
 46.4 
 
 44 
 
 
 Transverse diameter of neural canal of first sacral, internal measurement. 
 
 57.6 
 
 62.4 
 
 
 Height of neural canal of first sacral, internal measurement. 
 
 39 
 
 43.4 
 
 
 Depth of centrum of last sacral. 
 
 40 
 
 40.2 
 
 
 Transverse diameter of last sacral, measured across posterior face of centrum.... 
 
 82 
 
 87 
 
 
 Greatest transverse width of sacrum, measured above sacro-ischiadic foramina... 
 
 *238 
 
 238 
 
 
 Length of crest formed by fused dorsal spines. 
 
 224 
 
 257.6 
 
 
 Greatest width of dorsal crest at anterior end. 
 
 36 
 
 
 
 Height of dorsal crest at anterior end, measured from base of centrum. 
 
 157 
 
 
 
 Height of dorsal crest at posterior end, measured from base of centrum. 
 
 102 
 
 105.7 
 
 
 First sacral, width across anterior end of neural arch, i. e., across supports for pre- 
 
 
 
 
 zygapophyses. 
 
 
 93.6 
 
 
 Last sacral, greatest width across postzygapophyses. 
 
 52 
 
 61 
 
 
 Longest diameter (dorso-ventral) of obturator foramen. 
 
 
 
 130.7 
 
 Greatest diameter taken normal to dorso-ventral diameter. 
 
 
 
 79.6 
 
 1 Approximate. 
 
 FEMUR. 
 
 A single complete femur (No. 1871-Lr-l, plates 12 and 13) from Rancho La Brea 
 is much larger than the specimen described by Reinhardt and belongs to an older indi¬ 
 vidual. Contrasted with that of Megalonyx, the femur of Nothr otherium is much smaller; 
 the third trochanter is situated relatively lower on the outer side, the latter border being 
 concave for a greater distance above it; the shaft is relatively narrower transversely 
 above the third trochanter, and the lesser trochanter is not so well developed. 
 
 The head is comparatively not so prominent, and the great trochanter does not 
 reach proximally so far as in the femur of Hapalops. Between them the superior surface 
 of the femur from Rancho La Brea is nearly flat and not concave, as in that of Hapalops. 
 The digital fossa is small. The pit for the ligamentum teres (plate 12, fig. 1a) is small, 
 semi-oval in shape, and entirely inclosed by the articular surface of the head, as in Hapa¬ 
 lops. The second or lesser trochanter is much less conspicuous in than the Miocene 
 
PLEISTOCENE MECALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 81 
 
 genus, having more the appearance seen in Megalonyx. It is less developed in Nothro - 
 therium shastense than in the Brazilian species. The femur of Nothrotherium, as that of 
 Megalonyx, possesses a third trochanter, which is less conspicuous than in Hapalops. 
 It is situated relatively lower on the shaft in Nothrotherium than in Hapalops. In the 
 
 Fig. 29 .—Nothrotherium shastense Sinclair. Right posterior limb. A, lateral view; B, anterior view. 
 
 X 0.166. Rancho La Brea Pleistocene. 
 
 Rancho La Brea specimen the outer border of the third trochanter is more directly con¬ 
 tinuous with that of the external tuberosity than in Hapalops. This causes the distal 
 half of the femur of the Pleistocene form to become relatively wider, and resembles 
 more that in Prepotherium. 
 
 The shaft is of least width immediately above the third trochanter and at approxi¬ 
 mately the middle. It is thickened in a line extending obliquely across from the great 
 
82 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 trochanter to the internal condyle. The lower half of the thigh-bone lies in a plane 
 which turns inward and is oblique to a vertical plane passing through the head and 
 great trochanter. This obliquity is not a noticeable feature in the femur of Hapalops 
 under observation. 
 
 At the distal end (plate 12, fig. lb) the two articulating surfaces of the condyles 
 and that for the patella are separate, in which respect the femur of Nothrotherium resem¬ 
 bles that of Hapalops and of Megalonyx and differs from that of Mylodon. The condyles 
 are entirely hidden in anterior view. The inner condyle has the larger articulating area 
 and is convex, while the outer one is flattened or slightly concave transversely. The 
 concave rotular surface is deeper in dorso-ventral extent, but not so wide transversely 
 as in Megalonyx. It is more closely approached by the articulating face of the inner 
 condyle in the femur of M. jeffersonii than in that of Nothr other ium. The rotular surface 
 overlaps the external condyle, although separate from it, decidedly more than in the 
 femur of the Brazilian species, and is like Megalonyx in this respect. No distinct supra- 
 patella fossa is present. Continuous with the articulating surface of the external con¬ 
 dyle, but on the proximo-posterior border, is a small facet for a fabella. 
 
 Table 36 .—Measurements {in millimeters ) of femur. 
 
 
 N othrotherium 
 shastense, 
 
 No. 1871-L-l 
 
 Morotherium 
 gigas, No. 11898 
 Yale Univ. Coll. 
 
 Megalonyx 
 jeffersonii, 
 (after Leidy). 
 
 Greatest length, measured from head. 
 
 397.7 
 
 
 546.1 
 
 Greatest length, measured from great trochanter. 
 
 386 
 
 415 
 
 
 Width, measured obliquely from head to great trochanter. 
 
 186 
 
 223 
 
 254 
 
 Thickness of shaft at middle. 
 
 49.3 
 
 51 
 
 
 Least width of shaft. 
 
 112 
 
 127.7 
 
 
 Greatest width across distal tuberosities (above condvles).. 
 
 196.3 
 
 
 254 
 
 Width across condyles. . . . 
 
 168 
 
 
 215.9 
 
 Width of rotular groove. 
 
 70 
 
 81.4 
 
 101.6 
 
 Least width of intercondyloid space. 
 
 35 
 
 J 40 
 
 
 Width of inner condyle. 
 
 74.7 
 
 
 
 Vertical extent of inner condyle. 
 
 84.9 
 
 
 
 Width of outer condyle. 
 
 58.3 
 
 76 
 
 
 
 
 
 
 1 Approximate. 
 
 PATELLA. 
 
 The patella of Nothr other ium (plate 12, figs. 2 and 3) is very close in general shape 
 and proportions to that of the Brazilian species figured by Reinhardt (1878, plate 4, 
 fig. 5). The femoral surface is relatively large, convex transversely and concave dorso- 
 ventrally, thus differing from Megalonyx, in which it is nearly flat in the latter direction. 
 The articulating surface is approximately of even dorso-ventral diameter throughout, 
 and below this the ventral half of the patella is perhaps a little more tongue-shaped than 
 in the Brazilian form. 
 
 The patella is somewhat shorter in dorso-ventral extent than the specimens of the 
 knee-cap belonging to Megalonyx and available from the asphalt deposits, but is 
 relatively much narrower. In latter character, as well as in shape of femoral surface, 
 Nothrotherium differs decidedly from Hapalops. The patella of the Miocene genus is 
 closer in shape to that of Megalonyx. The dorsal surface of the patella in Nothrotherium 
 is relatively much thicker than in that of Megalonyx. The anterior face is marked by 
 a number of longitudinal ridges. The measurements in millimeters of the patella of 
 N. shastense No. 1895-L-l, are as follows: Dorso-ventral diameter, 92.8; width, 61.3; 
 dorso-ventral diameter of femoral surface, 48.5; transverse diameter of femoral surface, 
 59.8; thickness through femoral surface, 35.2. 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 83 
 
 TIBIA. 
 
 The tibia (plate 14, figs. 1 , la, lb, 2) of Notlirotherium, contrasted with that of 
 Megalonyx, is smaller but quite similar in general shape. In the Rancho La Brea 
 specimens the superior border of the anterior face, viewed from above, extends practically 
 in a straight transverse direction. This is due to an extension of the anterior tuberosity 
 for attachment of the ligamentum patellae to the outer side. The expanse of bone 
 anterior to the outer femoral articulating surface is much greater than that in front of 
 the inner articulation, and reaches nearly to the outer margin of the surface. In 
 Megalonyx the greatest development of the tuberosity occurs in front of the gutter 
 separating outer and inner femoral surfaces. 
 
 Table 37. —Measurements {in millimeters ) of tibia. 
 
 
 N othrotherium 
 shastense, 
 
 No. 1870-R-l 
 
 N. shastense, 
 No. 1870-L-l 
 
 Megalonyx 
 jeffersonii, 
 (after Leidy). 
 
 Greatest length, measured along middle of shaft. 
 
 Greatest length, from middle of posterior border of inner femoral 
 
 *317 
 
 308 
 
 
 surface to most distal point on tendinal groove. 
 
 313 
 
 *292 
 
 381 (?) 
 
 Greatest transverse width of proximal end. 
 
 190.8 
 
 178 
 
 215.9 
 
 Least width of shaft. 
 
 67 
 
 56.4 
 
 *76.2 (?) 
 
 Greatest transverse width of distal end. 
 
 Thickness of proximal end, from anterior border to most backward 
 
 142.8 
 
 129.3 
 
 165.1 
 
 projecting part of outer femoral surface. 
 
 121.6 
 
 103.8 
 
 
 Greatest transverse diameter of inner femoral surface . 
 
 85.6 
 
 80.5 
 
 
 Thickness at middle of proximal end. 
 
 86.4 
 
 72.5 
 
 139.7 
 
 Thickness of shaft at middle. 
 
 50.6 
 
 46 
 
 
 Thickness of distal end. 
 
 73.2 
 
 63 
 
 88.9 
 
 1 Approximate. 
 
 The attachment of the patella ligament in Nothrotherium is, then, relatively farther 
 toward the outer side of the tibia than in Megalonyx. This appears to accompany the 
 position of the distal condyles of the femur situated in a transverse plane, which is 
 oblique to a plane passing transversely through the upper portion of the shaft of the 
 femur. The obliquity of the distal end of the femur with reference to the proximal 
 end causes the tibia to face inward as well as anteriorly and brings the fibula well around 
 to the lateral side of the lower leg. With the plane of the front face of the tibia directed 
 anteriorly, the fibula would occupy a position well posterior to the outer side of the tibia. 
 
 The anterior face of the tibia along the superior border is greatly roughened, and 
 the antero-internal border of the shaft is also very rugose. The shaft is thickest along 
 the inner side and narrows laterally. It is relatively narrower transversely than in 
 the tibia of Hapalops and is more compressed antero-posteriorly toward the outer side. 
 The inner femoral articulating surface is large, subcircular, and concave. The antero- 
 external portion of this face is steeply inclined to form the inner side of the dorsal spine. 
 The latter is relatively more prominent than in the tibia of Hapalops. The outer femoral 
 surface is for the most part nearly flat and extends well posterior to the inner femoral 
 surface (plate 14, fig. la). The fibular facet at the proximal end is more elongate 
 transversely than in the tibia of Megalonyx. It is confluent with the outer femoral 
 surface toward the inner posterior side, but forms a distinct angle with this surface. 
 
 The inner malleolus is relatively much broader transversely in Nothrotherium than 
 in Megalonyx and extends farther along the postero-inferior border of the tibia. It is 
 as prominent as in the tibia of Hapalops. It supports posteriorly two tendinal grooves 
 as in the Miocene genus, the external of which is quite broad, while the internal groove 
 is much narrower. The latter channel swings well around to the antero-internal side 
 at its lower end. In Mylodon only a single broad groove is present in this region of the 
 
84 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 tibia. The articulating surfaces for astragalus and fibula extend obliquely across the 
 distal end from in front backward and inward. The inner and outer surfaces for the 
 astragalus are separated by an obscure ridge. The fibular facet is wide in front and 
 narrows behind. The distal spine for the astragalus is well developed, as in Hapalops. 
 
 The ratio of tibial length to femoral length has been computed from specimens 
 not pertaining to the same individual. In Nothrotherium the ratio is: length of tibia to 
 length of femur as 0.796 to 1. Presumably the ratio is greater than this and is more 
 like that found in Megalonyx. In Hapalops longiceps the ratio is: length of tibia to length 
 of femur as 0.768 to 1. W. K. Gregory records the ratio 0.72 to 1 for Hapalops sp. 
 
 FIBULA. 
 
 The fibula (plate 14, fig. 3) is long and slender, with comparatively widely expanded 
 proximal and distal extremities. Compared with the fibula of Mylodon, that of Noth- 
 rotherium is very much more slender in the region of the shaft and is longer. While 
 the proximal extremity is greatest in fore-and-aft diameter, the distal extremity is of 
 greatest diameter transversely. The proximal articulation with the tibia is usually an 
 elongate and narrow surface, but may be short and rather broad. The small facet for 
 the fabella, having a postero-dorsal position on the proximal extremity of the fibula, 
 is not invariably present on fibulae from Rancho La Brea. 
 
 Table 38. —Measurements (in millimeters ) of fibula of N. shastense. 
 
 
 No. 
 
 1869-L-2 
 
 No. 
 
 1869-R-l 
 
 Length through shaft. 
 
 299 
 
 *324 
 
 Greatest antero-posterior diameter of proximal 
 end. 
 
 69.6 
 
 81.5 
 
 Transverse diameter of proximal end. 
 
 59.9 
 
 61 
 
 Antero-posterior diameter at middle of shaft.. 
 
 30 
 
 .... 
 
 Transverse diameter at middle of shaft. 
 
 23 
 
 • • • • 
 
 Antero-posterior diameter of distal end. 
 
 58.3 
 
 62.8 
 
 Greatest transverse diameter of distal end.... 
 
 85.3 
 
 87.9 
 
 t 
 
 1 Approximate. 
 
 The tibial or inner side of the shaft is flattened, while the remaining surface is 
 convex. The distal end is roughly pyramidal in shape, with the transverse diameter 
 greater than the antero-posterior diameter. Below the tibial facet the external malleolus 
 is massive and bears along the inner side a concave triangular, articulating surface for 
 the astragalus. The base of this facet is confluent with, but distinguished by a sharp 
 angle from, the tibial facet. The astragalar surface is directed downward as well as 
 inward, while the tibial facet faces upward and inward. 
 
 The fibula of Nothrotherium resembles greatly that of Hapalops. In proportion to 
 its length and to size of shaft the fibula of the Pleistocene genus has heavier extremities 
 than that of H. longiceps. 
 
 CALCANEUM. 
 
 A single calcaneum of Nothrotherium (No. 1877-R-l, fig. 30, a, b, c) is available 
 from Rancho La Brea and lacks only the posterior border of the tuber calcis. In length 
 and slenderness of neck the calcaneum of Nothrotherium differs decidedly from that of 
 Megalonyx (compare with fig. 47, p. 106). It differs also from Hapalops, in which form 
 the neck of the calcaneum is relatively broader than in that of Megalonyx. The poste¬ 
 rior end is not so expanded in Nothrotherium as in Hapalops. The dorsal side of the fan¬ 
 like expansion is concave, while the ventral side is broadly convex. The fibular border 
 of the neck is thick, the neck narrowing toward the inner side. 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 85 
 
 At the articulating end of the calcaneum the external astragalar facet is relatively 
 small, more prominently placed, and does not extend upon the neck so far as in Hapalops. 
 At its upper inner border a semi-cone-shaped protrusion of the facet articulates with 
 the concavity or channel described on the external calcaneal facet of the astragalus. 
 With the above exception the facet is nearly flat transversely and slightly convex 
 dorso-ventrally. 
 
 Fig. 30. —Nothrotherium shastense Sinclair. Right calcaneum, No. 1877-R-l. A, inner view; B, view of 
 articulating end; C, dorsal view. X 0.50. Rancho La Brea Pleistocene. 
 
 Hapalops elongatus Ameghino. Right calcaneum, No. 15155, P. U. C., A 1 , B l , C l , same views 
 in A, B, C; X 1.0. Santa Cruz Miocene. 
 
 The external astragalar facet is separated from the smaller internal facet and from 
 the cuboid surface by a channel which is deeper than in the calcaneum of the Miocene 
 genus and is continued farther up on the dorsal face as a shallow groove. The internal 
 astragalar facet of the calcaneum is nearly circular and is relatively smaller than in 
 Hapalops. It is situated less in front of and more to the inner side of the large astragalar 
 facet than in the latter. It is joined with the cuboid facet below by a narrow bridge of 
 
86 
 
 CENOZOIC GRAYIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 bone, thus differing from Megalonyx. The cuboid facet is somewhat larger than the 
 smaller of the two astragalar facets and is concave in its long diameter. The postero- 
 inferior side of this facet is produced outward into a more conspicuous, but relatively 
 more slender, plate-like process than in Hapalops. The latter process is absent in the 
 calcaneum of Megalonyx. 
 
 The calcaneum of H. longiceps differs from that of Nothrotherium, according to Scott, 
 in that “the external astragalar facet is smaller, more simply convex, and less reflected 
 upon the dorsal side; the sustentaculum is smaller and its facet is smaller and more 
 distal, less dorsal; the cuboid surface is more oval.” In the calcaneum of Hapalops 
 before me the external astragalar facet is relatively larger and more reflected upon the 
 dorsal surface than in No. 1877-R-l from Rancho La Brea. The sustentaculum is not 
 so prominent as in the latter, but the facet which it bears is relatively larger than that 
 in the calcaneum from the asphalt beds. 
 
 Contrasted with the fragment of a calcaneum (No. 19871, U. C. C.) of Nothrotherium 
 from the Pleistocene of Hawver Cave (Stock, 1918, p. 510, fig. 29), the Rancho La 
 Brea specimen exhibits some points of difference. In No. 1877-R-l the cuboid facet 
 is distinctly larger than the sustentacular facet, while in the cave specimen the cuboid 
 facet and the sustentacular facet are of nearly equal size, with the latter perhaps 
 slightly larger. The interspace between the two facets is also greater in the calcaneum 
 from the asphalt. The internal channel between cuboid facet and large external facet 
 for the astragalus is relatively wider in No. 1877-R-l. The lower end of the large 
 astragalar facet is narrower in the Rancho La Brea specimen. 
 
 Table 39. — Measurements (in millimeters) of calcaneum of N. shastense, No. 1877-R-l. 
 
 Least transverse width of neck. 39 
 
 Thickness of neck. 24.5 
 
 Distance from inner border of sustentaculum to infero-external prominence of articulating end. 77.7 
 
 Distance from dorsal border of external astragalar facet to ventral border of cuboid facet. 70.5 
 
 ASTRAGALUS. 
 
 In Nothrotherium the astragalus (fig. 31, a to d) is relatively shorter, wider, and 
 deeper than in Hapalops , with the tibial surface higher and more convex. The convexity 
 of this surface with the inner extension of an odontoid shape modifies the astragalus 
 considerably from that of Hapalops. These characters are much more marked than in 
 the astragalus of Megalonyx, the latter genus retaining rather the type of structure found 
 in Miocene forms (compare in fig. 31, A 1 to D l with A 2 to D 2 ). The astragalus of Nothro¬ 
 therium resembles more in general appearance that of Mylodon. The outer portion of the 
 tibial surface is not so sharply separated from the fibular surface as in Hapalops. At 
 the middle of the distal border of the tibial articulation is a V-shaped notch as in Hapa¬ 
 lops, ending in a depression that receives a process of the tibia. This notch may be quite 
 large. 
 
 The external calcaneal facet resembles that in the astragalus of Hapalops and differs 
 from that in Megalonyx in its greater concavity. It differs from that in Hapalops in the 
 development of a broad and deep channel on the internal side of the posterior half, 
 which meets the interarticular gutter opposite the facet for the sustentaculum. The 
 facet for the sustentaculum forms the posterior surface of the head and is less distinct 
 than in the Miocene form. 
 
 The neck is much better defined than in the astragalus of Hapalops. As in the 
 Miocene genus, the cuboid facet is on the palmar side of the head, at least in major part. 
 The head is less elongate and more rounded than in the astragalus of Hapalops. The 
 depression on its anterior surface for the mammillate surface of the navicular may vary 
 in depth. This surface is reflected to the inner side, corresponding with the concavity of 
 the navicular. The outer dorsal margin of the depressed area is quite prominent in some 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 87 
 
 I 
 
 C 
 
 2 
 
 Fig. 31 N othr other ium shastense Sinclair. Right astragalus, No. 1875-R-l. A, tibial view; B, inner view; C, calcaneal 
 view; D, lateral view. X 0.50. Rancho La Brea Pleistocene. 
 
 Megalonyx, probably wheatleyi Cope. Right astragalus, Amer. Mus. Spec. A 1 , B l , C 1 , D 1 , same views as in 
 A, B, C, D. X 0.50. Port Kennedy fissure, Pennsylvania, Pleistocene. 
 
 Hapaiops sp. Right astragalus, No. 15594, P. U. C. A 2 , B 2 , C 2 , D 2 , same views as in A, B, C, D. X 1.0. Santa 
 Cruz Miocene. 
 
88 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 specimens from Rancho La Brea. The dorsal border of the depressed navicular surface 
 may be separated from the anterior border of the tibial articulation by a rather broad 
 channel, thus defining the neck region, or may closely approach that border. 
 
 Table 40. —Measurements {in millimeters ) of astragalus of N. shastense. 
 
 
 No. 
 
 1875-L-2 
 
 No. 
 
 1875-R-2 
 
 No. 
 
 1875-R-3 
 
 No. 
 
 1875-R-4 
 
 No. 
 
 1875-R-5 
 
 Greatest antero-posterior diameter. 
 
 Greatest transverse diameter, measured at a right angle 
 
 102.4 
 
 100.7 
 
 105.2 
 
 91.3 
 
 102 
 
 to antero-posterior diameter. 
 
 97.8 
 
 104 
 
 106.3 
 
 102.6 
 
 108.2 
 
 Depth of fibular side at distal end. 
 
 64.5 
 
 62.7 
 
 69.6 
 
 61.2 
 
 67.5 
 
 Greatest diameter of head across navicular depression. 
 
 50 
 
 53.6 
 
 56 
 
 49.4 
 
 53.8 
 
 Greatest antero-posterior thickness of head. 
 
 42.7 
 
 44.6 
 
 50 
 
 42 
 
 44.4 
 
 CUBOID. 
 
 In Nothrotherium the pyramidal shape of the cuboid (fig. 32, a, b, c) is more markedly 
 defined than in Hapalops. This is brought about by a closer approximation of the astra- 
 galar facet distally to the plane of the metatarsal surface. Nothr other ium differs also from 
 Megalonyx in this respect. The calcaneal facet is convex and prominent as in Hapalops, 
 oval in shape, and extends across almost the entire length of the fibular side of the cuboid. 
 The astragalar facet is large and decidedly more concave than in the Miocene genus. It 
 meets the calcaneal facet in a sharp angle. 
 
 Fig. 32. —Nothrotherium shastense Sinclair. Right cuboid, No. 1885-R-l. A, distal view; B, 
 dorsal view; C, proximal view. X 0.50. Rancho La Brea Pleistocene. 
 
 Hapalops indifferens Ameghino. Right cuboid, No. 15110, P. U. C. A 1 , B l , C 1 , same 
 views as in A, B, C. X 1.0. Santa Cruz Miocene. 
 
 As in Hapalops and Megalonyx, the inner border of the cuboid in Nothrotherium is 
 formed by a narrow articulating face for the navicular. This surface is continuous with, 
 but sharply separated from, the astragalar facet. The navicular surface may be evenly 
 continuous with an ectocuneiform facet along its dorsal half, as in No. 1885-L-l, or it 
 may be separated from this facet by a distinct angle. In three specimens of the cuboid 
 of Megalonyx from the Pleistocene of the Port Kennedy fissure and in the collections of 
 the American Museum of Natural History no connection was noted between these two 
 surfaces. In the cuboid of Nothrotherium the articulating surfaces for metatarsals IV 
 and V are continuous and not sharply separated from each other. This combined sur¬ 
 face is distinctly more external to the ectocuneiform facet in Nos. 1885-R-2 and 1885- 
 L-l from Rancho La Brea than in the cuboid of Hapalops. It has a shape somewhat 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 89 
 
 like that in Megalonyx, but the inner border is not so evenly concave, as shown by Leidy 
 for that genus. The articulating surface for the metatarsals in Nothrotherium, as in Mega¬ 
 lonyx, is wider than deep, while in Hapalops the depth is greater than the width. 
 
 Table 41. — Measurements (in millimeters) of cuboid of N. shastense. 
 
 
 No. 
 
 1885-L-l 
 
 No. 
 
 1885-R-2 
 
 No. 
 
 1885-R-l 
 
 Greatest distance between facets for astragalus and metatarsal V, measured 
 across calcaneal facet. 
 
 43.6 
 
 42.8 
 
 44 
 
 Greatest depth, measured over navicular facet. 
 
 44.1 
 
 43.2 
 
 60.2 
 
 Greatest width of distal face, measured from middle of plane of navicular 
 facet to external end. 
 
 58.3 
 
 59.2 
 
 60.3 
 
 
 NAVICULAR. 
 
 The navicular of Nothrotherium (fig. 33, a, b, c) differs decidedly from that of Mega¬ 
 lonyx in its oval shape. In the latter genus, as is well shown by Leidy (1855, plate 13, 
 figs. 7 and 8) for M. jeffersonii and by W. J. Sinclair (1905, plate 20, figs. 5 and 6) for 
 M. sierrensis, the navicular is roughly triangular in form, the base of the triangle bor¬ 
 dering the concavity and apex situated above the convexity for the astragalus. The 
 form of the navicular in Hapalops is more like that in Megalonyx than like that in Noth- 
 
 i 
 
 Fig. 33. —Nothrotherium shastense Sin¬ 
 clair. Right navicular, No. 
 1891-R-l. A, distal view; B, 
 view of cuboid face; C, view of 
 astragalar face. X 0.50. Rancho 
 La Brea Pleistocene. 
 
 Hapalops elongatus Ameghino. 
 Right navicular, No. 15545, P. 
 U. C. A 1 , B l , C\ same views as 
 in A, B, C. X 1.0. Santa Cruz 
 Miocene. 
 
 rotherium. A navicular belonging presumably to a megalonychid ground-sloth and 
 described by Sinclair 0 from the Snake Creek Lower Pliocene of Nebraska approximates 
 more closely the corresponding element in Megalonyx and in Hapalops than it does that 
 of Nothrotherium from Rancho La Brea. 
 
 The mammiform surface of the proximal side (fig. 33, c), which articulates with the 
 concave surface on the head of the astragalus, may vary in degree of convexity, as it 
 does also in the navicular of Mylodon and in that of Megalonyx. The concavity on the 
 proximal side is much deeper than in the navicular of Megalonyx sierrensis. The cuboid 
 facet is narrow, flat or slightly convex, and confluent with the astragalar surface. It is, 
 however, distinctly separated from the concave portion of the latter surface by a sharp 
 angle, as in Hapalops. Along the outer half it is also confluent with the facet for the 
 ectocuneiform. 
 
 °W. J. Sinclair. Additions to the fauna of the Lower Pliocene Snake Creek beds (Results of the Princeton Univer¬ 
 sity 1914 Expedition to Nebraska), Proc. Amer. Philos. Soc., vol. 54, p. 83, fig. 7, 1915. 
 
90 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 The entocuneiform and mesocuneiform facets form a continuous articulating sur¬ 
 face which may be separated by a median channel from the facet articulating with the 
 ectocuneiform. These surfaces differ from those in Megalonyx and in Hapalops in the 
 lack of decided divergence. As in Megalonyx and in the Miocene genus, the ectocunei¬ 
 form facet has the greater area. 
 
 Table 42. — Measure?nents {in millimeters) of navicular of N. shastense. 
 
 
 No. 
 
 1891-R-l 
 
 No. 
 
 1891-R-2 
 
 No. 
 
 1891-R-3 
 
 No. 
 
 1891-R-4 
 
 No. 
 
 1891-L-l 
 
 No. 
 
 1891-L-2 
 
 Greatest diameter between cuneiform facets 
 (width). 
 
 64 
 
 61 5 
 
 66 9 
 
 64.6 
 
 65.8 
 
 65.2 
 
 Diameter taken at right angles to greatest diam¬ 
 eter (dorso-palmar diameter). 
 
 51.5 
 
 49 5 
 
 54 2 
 
 51 
 
 51.2 
 
 50 
 
 Greatest thickness through convexity articulat¬ 
 ing with astragalus. 
 
 33.7 
 
 28.8 
 
 31.7 
 
 28.4 
 
 32.2 
 
 32.5 
 
 
 ECTOCUNEIFORM. 
 
 In comparing the corresponding measurements, the ectocuneiform of Nothrotherium is 
 seen to differ decidedly from that of Megalonyx in its smaller size. The ectocuneiform 
 (fig. 34, a, b) is not deep and is relatively broader superiorly than that of Megalonyx. 
 It differs from that of Hapalops in possessing a more narrow ventral lobe or projection. 
 In this respect Nothrotherium is more like Megalonyx. The articulating surface on the 
 distal side conforms in shape to the proximal surface of metatarsal III. It is broad 
 
 Fig. 34 .—Nothrotherium shastense Sin¬ 
 clair. Left ectocuneiform, No. 1943- 
 L-2. A, navicular view; B, mettarsal 
 view. X 0.50. Rancho La Brea 
 Pleistocene. 
 
 Hapalops elongatus Ameghino. 
 Right ectocuneiform, No. 15160 P. 
 U. C. A 1 , B l , same views as in 
 A, B. X 1.0. Santa Cruz Miocene. 
 
 superiorly, with a tongue-shaped lobe extending inferiorly. In the ectocuneiform of 
 Hapalops elongatus (No. 15160, Prin. Univ. Coll.) the lower lobe is much more broadly 
 connected with the superior portion of the surface. The outer end of the upper portion 
 is deflected decidedly to the proximal side and articulates with the fourth metatarsal. 
 In other words, there is indicated a slight overlapping of the ectocuneiform on the fourth 
 metatarsal, a character not seen in the pes of Hapalops. 
 
 Superiorly the ectocuneiform is twice as thick antero-posteriorly as it is inferiorly. 
 The proximal surface articulating with the navicular varies in depth. The depth of the 
 upper portion of the facet depends in part on the extent to which the outer margin has 
 been turned back. The facet along the outer margin of its upper portion is continuous 
 with a small articulating surface for the cuboid. In this respect Nothrotherium resembles 
 Hapalops and Megalonyx. The facet for the cuboid faces outward as well as to the 
 proximal side. 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 91 
 
 ENTOCUNEIFORM AND METATARSAL I. 
 
 Two specimens are available from Rancho La Brea, representing the first metatarsal 
 with which the entocuneiform is evidently fused. Winge (1915, plate 25) has shown this 
 element in the pes of Nothrotherium maquinense, and the material from the asphalt resem¬ 
 bles this closely. The element (fig. 35, a, b, c) is thickened somewhat in the proximal 
 region and thins toward the distal end. Along the external side are two facets: (1) a 
 narrow surface, flanking the side of the terminal face, articulating with the mesocunei- 
 form, and (2) a more distally situated surface articulating with the inner proximal facet 
 of metatarsal II. The slightly concave facet at the proximal end, truncated along the 
 outer side, articulates with the navicular. At the distal end no articulating surface is 
 present, indicating an entire loss of the inner digit. This is a notable specialization 
 beyond the stage seen in Hapalops, in which form the inner digit of the pes is still func¬ 
 tional. 
 
 Fig. 35 .—Nothrotherium shastense Sinclair. Right entocuneiform and metatarsal I, No. 1880-R-2. 
 A, ventral view; B, external view; C, dorsal view; D, proximal view showing these elements in 
 articulation with mesocuneiform and metatarsal II; X 0.50. Rancho La Brea Pleistocene. 
 
 It is evident from the above that the proximal part of the innermost element of the 
 pes of Nothrotherium, bearing facets for the mesocuneiform and navicular, must be homol¬ 
 ogous with the entocuneiform. In both specimens from Rancho La Brea the ento¬ 
 cuneiform has fused with the first metatarsal, whereas in the pes of Hapalops these 
 elements were probably separate. 
 
 Table 43. — Measw'ements (in millimeters) of ectocuneiform. 
 
 
 N. shastense, 
 No. 1943-L-2 
 
 N. shastense, 
 No. 1943-L-3 
 
 Hapalops 
 
 elongatus, 
 
 No. 15160 P. U.C. 
 
 Megalonyx 
 jeffersonii 
 (after Leidy). 
 
 Greatest depth. 
 
 51.4 
 
 50.5 
 
 18.4 
 
 76.2 
 
 Greatest width. 
 
 43 3 
 
 39.7 
 
 12.6 
 
 50.8 
 
 Greatest antero-posterior diameter superiorly. 
 
 25.2 
 
 18.5 
 
 6.2 
 
 21.8 
 
 METATARSAL II. 
 
 In general shape the second metatarsal (fig. 36, a to f) of Nothrotherium is more like 
 that of Megalonyx than like that of Hapalops. The largest of the second metatarsals 
 from Rancho La Brea is still decidedly smaller than 
 in Megalonyx jejjersonii. It is more slender than in 
 Hapalops elongatus, in which respect it approaches 
 Megalonyx. In the Miocene genus and in Megalonyx 
 the depth of the proximal end exceeds the width, 
 while in Nothrotherium the width in most specimens 
 decidedly exceeds the depth. In the latter form the 
 proximal end, when viewed from above, is seen in 
 some specimens to extend beyond the end of the facet 
 
 Table 44. —Measurements (in mil¬ 
 limeters) of metatarsal I and ento¬ 
 cuneiform of N. shastense. 
 
 1 
 
 No. 
 
 No. 
 
 
 1880-R-l 
 
 1880-R-2 
 
 Length. 
 
 56.8 
 
 53.6 
 
 Width. 
 
 30.3 
 
 33 5 
 
92 
 
 CENOZOIC GRAYIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 for metatarsal III and even farther beyond the facet for the first metatarsal. In the 
 metapodial of Megalonyx the outer and inner facets for the adjoining metatarsals are 
 closer to the borders of the proximal articulation and do not stand off as distinctly as 
 in that of Nothrotherium. 
 
 The mesocuneiform facet in Nothrotherium differs from that in Hapalops in its more 
 distinct triangular shape, thus resembling that in Megalonyx. It also lacks the median 
 dorsal notch present in the metapodial of the Miocene genus. The depth of the tri¬ 
 angular facet is usually less than the length of the base (the dorsal border of the articu¬ 
 lating surface). 
 
 A single element (No. 1881-R-2) of the three or more second metatarsals available 
 from Rancho La Brea shows fusion with the mesocuneiform. The proximal surface in 
 this specimen is concave, broad above and narrowing below, and articulates with the 
 navicular. Continuous with this surface along the inner side, but at right angles to it, 
 is a narrow facet (fig. 36 c) that articulates with a similar surface situated on the outer 
 side of the entocuneiform. 
 
 Fig. 36. —Nothrotherium shastetise Sinclair. Right mesocuneiform and metatarsal II, No. 1881-R-2. A, outer view; B, 
 dorsal view; C, inner view; D, proximal end; E, distal end; F, proximal end of metatarsal II, No. 1881-R-l; 
 X 0.50. Rancho La Brea Pleistocene. 
 
 Hapalops elongatus Ameghino. Right metatarsal II, No. 15545, P. U. C. A 1 , B l , C 1 , D 1 , E l , same view as in A, B, 
 C, D, E. X 1.0. Santa Cruz Miocene. 
 
 The facet for metatarsal I is directed slightly downward and may be plane or slightly 
 concave. The outer facet for metatarsal III may have its greatest diameter vertical or 
 transverse with reference to the outer side. The surface is convex. In the correspond¬ 
 ing metatarsal of H. elongatus (No. 15545, Prin. Univ. Coll.) this facet is concave and 
 is not placed upon so distinct a pedicle as in the species from the asphalt beds. It articu¬ 
 lated presumably with the ectocuneiform. The second metatarsal never overlapped 
 the third at the proximal end, as in the Miocene Eucholoeops or Analcimorphus. „ 
 
 The shaft is broad, with the ventral surface rounded. The distal articulation, as 
 in Hapalops, is placed somewhat oblique to a vertical plane of the shaft. It differs from 
 that in the Miocene form in lacking the broad offset on the inner side of the carina. In 
 the species from the asphalt there is an articulating offset along the outer side of the 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 93 
 
 oarina, but along the inner side its place is taken by a narrow and more or less defined 
 groove. Leidy (1855, p. 42) states that at the distal end of metatarsal II in Megalonyx 
 there is “a long vertical, ellipsoidal articulating process, with a short vertical offset upon 
 each side.” 
 
 PHALANX I, DIGIT II, PES. 
 
 The first phalanx (fig. 37, a, b, c) is shortened antero-posteriorly, this distance 
 being less than the depth of the proximal end. The greatest length of the phalanx is 
 along the inner side. A deep vertical groove at the proximal end received the carina of 
 the second metatarsal. The groove notches the ventral surface. It is bordered along 
 nearly the entire outer side by an articulating surface. Below the latter face and on 
 the ventral side is a facet for a sesamoid. The distal articulating surface is slightly 
 
 Fig. 37 .—rNothrotherium shastense Sinclair. Right Phalanx II, digit II, pes, No. 1887-R-l. 
 
 phalanx I, digit II, pes, No. 1886-R-l. A, proxi- Proximal, lateral, and distal views. X 0.50. 
 
 mal view; B, lateral view; C, distal view. X 0.50. Rancho La Brea Pleistocene. 
 
 oblique to’a vertical plane passing through the proximal groove. It consists of two con¬ 
 dylar areas separated by a median V-shaped groove. The outer condyle has a slightly 
 greater dorso-ventral extent than the inner condyle. 
 
 The measurements (in millimeters) of phalanx I, digit II, pes, of N. shastense, No. 
 1886-R-l, are: Length along inner side, 28.6; depth, 41.3; greatest width, 32.8. 
 
 PHALANX II, DIGIT II, PES. 
 
 The second phalanx (fig. 38) is not so long and slender as the corresponding segment 
 in the second digit of the manus. It is presumably more comparable in size to the second 
 phalanx of digit III, manus. The proximal articulation consists of two concave surfaces 
 separated by a median acute ridge which widens below and above. At the distal end 
 the two condyles diverge downward, with the outer of only slightly greater depth than 
 the inner condyle. The surfaces of the condyles widen toward the inferior side. 
 
 The measurements (in millimeters) of phalanx II, digit II, pes, of N. shastense, 
 No. 1887-R-l, are as follows: Length through middle, 50.9; width of proximal end, 
 27; depth of proximal end, 36.5; depth of distal end, 23.7. 
 
 PHALANX III, DIGIT II, PES. 
 
 The terminal phalanx of the second digit of the pes, as exemplified by specimen 
 1888-R-l, fig. 415, may have a length equal to that of the corresponding phalanx in 
 the manus, but the hooded region is deeper and the claw-process is narrower than in 
 the latter. Compared with other ungual phalanges of the manus, No. 1888-R-l is 
 larger than that of the fourth digit but smaller than that carried by the median finger. 
 No. 1888-R-l is actually wider at the proximal end than phalanx III, digit III, manus 
 (No. 1890-L-l). Here the articulating surfaces for the second phalanx tend to diverge 
 ventrally and increase in width in this direction. In No. 1890-L-l, however, the dorso- 
 ventral extent of the proximal articulating surfaces is greater, but they do not tend to 
 diverge ventrally. The posterior overhanging process is relatively slender in contrast 
 to that in No. 1890-L-l. The tip of the claw-process in No. 1888-R-l does not extend 
 so far below the horizontal plane of the subungual base as in No. 1890-L-l. In the ter- 
 
94 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 minal phalanx of digit IV, manus, the anterior end of the claw-process does not extend 
 below this horizontal plane. The dorsal surface of the projecting portion of the claw- 
 process in No. 1888-R-l is rounded. In No. 1890-L-l and in the ungual phalanx of 
 
 Fig. 39 .—Nothrotherium shastense Sinclair. Right metatarsal III, No. 1882-R-l. A, distal view; 
 
 B, inner view; C, outer view; D, proximal view. X 0.50. Rancho La Brea Pleistocene. 
 
 Hapalops elongatus Ameghino. Right metatarsal III, No. 15545 P. U. C. A 1 , B l , C 1 , D 1 , 
 same views as in A, B, C, D. X 1.0. Santa Cruz Miocene. 
 
 digit IV, manus, this portion of the dorsal surface forms a sharp edge. On the ventral 
 surface of the base of the phalanx the discoid area for tendon attachment is situated for¬ 
 ward. Lying on each side posterior to this area are the foramina for blood-vessels to the 
 base of the claw. The inner foramen lies slightly in advance of the outer. 
 
 Table 45. — Measurements {in millimeters) of second metatarsal. 
 
 
 Nothrotherium 
 
 shastense, 
 
 No. 1881-R-2 1 
 
 N. shastense, 
 No. 1881-R-l 
 
 N. shastense, 
 No. 1881-R-4 
 
 Megalonyx 
 jeffersonii, 
 (after Leidy). 
 
 Greatest length. 
 
 76 6 
 
 71.4 
 
 72.4 
 
 88.9 
 
 Greatest width of proximal end, from 
 third metatarsal facet to first meta¬ 
 tarsal facet. 
 
 43 5 
 
 47.6 
 
 44.7 
 
 Greatest depth of proximal end. 
 
 39 
 
 36.3 
 
 34.5 
 
 57.2 
 
 Least depth of shaft. 
 
 21.3 
 
 20.6 
 
 20.8 
 
 Greatest depth of distal end. 
 
 53.5 
 
 48.4 
 
 47 
 
 63.5 
 
 Greatest width of distal end. 
 
 35.5 
 
 38 
 
 33.5 
 
 
 
 1 Fused with mesocuneiform. 
 
 The measurements (in millimeters) of phalanx III, digit II, pes, of N. shastense 
 (No. 1888-R-l) are as follows: 
 
 Greatest length from proximal end of overhanging process to tip of claw-process. 106.3 
 
 Greatest depth from dorsal surface of hood to discoid area on ventral surface (approximate). 39 
 
 Width of proximal end. 31 
 
 Length of subungual base. 46 
 
 Width of ventral surface of claw-process just anterior to hooded region. 13.3 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 95 
 
 METATARSAL III. 
 
 In Nothrotherium, as in Megalonyx, the width of metatarsal III is greater than the 
 length, the Pleistocene genera differing in this respect from Hapalops. The element 
 (fig. 39, a to d) is more compressed antero-posteriorly in Nothrotherium than in Mega¬ 
 lonyx. The ectocuneiform surface is more deeply concave than in Megalonyx or in 
 Hapalops, while the ventral, tongue-like process of the posterior end is not so broad as 
 in Megalonyx or in H. elongatus. The inner facet for metatarsal II is concave, while 
 the outer facet for metatarsal IV is less like Megalonyx and more like Hapalops in its 
 large size. The latter articulating surface is directed more backward and less outward 
 than in Megalonyx and in H. elongatus. At the distal end the trochlea resembles that 
 in the Miocene genus. The carina is not so large and heavy as in Megalonyx. 
 
 Fig. 40 .—Nothrotherium shastense Sinclair. Co-ossified phalanges I and II, digit III, pes. A, B, C, No. 1889-L-2 
 with inner sesamoid bone. Proximal lateral and distal views. X 0.50. D, No. 1889-L-l. Dorsal view. X 0.50. 
 Rancho La Brea Pleistocene. 
 
 CO-OSSIFIED PHALANGES I AND II, DIGIT HI. PES. 
 
 In N othr other ium, as in Megalonyx, the first and second phalanges of the median digit 
 of the pes have fused to form a single element (fig. 40, a to d). Both Pleistocene genera 
 are more advanced in this character than Hapalops, in which the phalanges are separate. 
 The single element in Nothrotherium greatly resembles that in Megalonyx, but differs in its 
 inferior size (compare measurements). 
 
 Table 46. —Measurements {in millimeters ) of metatarsal III, of 
 
 N. shastense. 
 
 
 No. 
 
 1882-R-l 
 
 No. 
 
 1882-L-l 
 
 Antero-posterior diameter through middle.... 
 
 x 43.4 
 
 *45.2 
 
 Greatest width. 
 
 63.6 
 
 62.4 
 
 Greatest depth of proximal end. 
 
 59 
 
 59.7 
 
 Depth of distal carina. 
 
 67.6 
 
 62.5 
 
 1 Approximate. 
 
 The proximal end is characterized by a deep median vertical groove with articu¬ 
 lating offsets on each side as in Megalonyx. The furrow does not widen dorsally as in 
 the latter genus. As in Megalonyx, the inner offset is the wider, but in Nothrotherium 
 this articulating surface does not extend dorsally so far as the outer offset. In a single 
 specimen from Rancho La Brea the inner sesamoid has fused with the lower side of the 
 proximal end. The outer end of the sesamoid curves anteriorly. A facet on the inner 
 side continues ventrally the articulating surface of the median furrow, while a posterior 
 facet likewise continues downward the inner offset. As in Megalonyx, the inner sesamoid 
 joins the co-ossified phalanges by a larger facet than does the outer sesamoid. 
 
9G 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 The distal condyles have relatively greater depth in the Rancho La Brea specimen 
 than in Leidy’s specimen of Megalonyx jeffersonii, and in dorsal view are seen to be 
 separated by a wider groove. In the co-ossified first and second phalanges (No. 20004 
 U. C. C.), pes, of Megalonyx from Rancho La Brea, the groove of the distal trochlea is 
 narrow above and widens below; in N othrotherium it is practically of same width through¬ 
 out. The outer condyle possesses greater depth than the inner. It is semicircular in 
 outline, while the inner condyle presents a somewhat more extensive surface for articu¬ 
 lation with the ungual phalanx. 
 
 Table 47. —Measurements {in millimeters) of co-ossified 'phalanges 
 I and II, digit III, pes, of N. shastense. 
 
 
 No. 
 
 1889-L-l 
 
 No. 
 
 1889-L-2 
 
 Length along middle of outer side. 
 
 55.9 
 
 55.4 
 
 Greatest depth of proximal end. 
 
 54.6 
 
 58.7 
 
 Depth of lateral condyle of distal end. 
 
 41 
 
 43.8 
 
 Greatest width of proximal end. 
 
 49.5 
 
 54 
 
 Depth of proximal end from superior border to 
 
 
 
 median inferior border of sesamoid. 
 
 • • • • 
 
 69.7 
 
 Width of sesamoid. 
 
 • • • • 
 
 38.9 
 
 Thickness of sesamoid along inner side. 
 
 .... 
 
 20.6 
 
 PHALANX III, DIGIT III, PES. 
 
 The ungual phalanx carried by the median digit of the pes (fig. 41a) is the largest 
 of the ungual series in Nothrotherium, being much larger than the corresponding phalanx 
 of the median digit of the manus. The bony sheath is well preserved in specimens from 
 the asphalt and encases more than 
 one-half of the claw-process. The 
 sheath is best developed along the 
 sides. At the lower margin it may 
 be separated from the side of the 
 claw-process by a space measuring 
 approximately 8 mm. The subun¬ 
 gual base has a well-marked discoid 
 area on its anterior half for tendon 
 attachment. This area may become 
 quite tuberous. Just posterior to it 
 and on either side is a foramen trans¬ 
 mitting blood-vessels to the base of 
 the claw. The inner foramen is dis¬ 
 tinctly the larger. The dorsal sur¬ 
 face of the claw-process within the 
 bony hood is broadly convex. An¬ 
 teriorly, however, this convex sur¬ 
 face is transformed into a sharp dorsal edge which extends to the tip of the phalanx. 
 The sides of the claw-process at the anterior margin of the hood are concave dorso- 
 ventrally. The ventral surface of the process may be flat or slightly convex. It may 
 vary in width. 
 
 The proximal articulation consists of two concavities extending parallel in a vertical 
 direction and separated by a thick median ridge. The concavities are of equal width, 
 but the outer has a slightly greater dorso-ventral extent. The articulations and median 
 ridge extend upward and backward in an arch forming the lower surface of the over¬ 
 hanging process. The posterior end of this process is indented. 
 
 Fig. 41 .—Nothrotherium shastense Sinclair. Ungual phalanges, right pes. 
 Lateral view. X 0.50. a, digit III, No. 1890-R-l; b, digit II, 
 No. 1888-R-l. Rancho La Brea Pleistocene. 
 
PLEISTOCENE MEGALONYCIIINAE AND MYLODONTIDAE OF RANCHO LA BIIEA. 97 
 
 METATARSAL IV. 
 
 The fourth metatarsal of Nothrotherium (fig. 42, a to e) may be actually longer than 
 that of Megalomyx, but the depth of the proximal end is not quite so great as in that of 
 the latter genus. The proximal end (fig. 42d) is nearly quadrate in shape, lacking the 
 pronounced downward prolongation which is present in the metatarsal of Hapalops. A 
 
 Fig. 42 .—Nothrotherium shastense Sinclair. Right metatarsal IV, No. 1883-R-l. A, outer view; B, dorsal view; C, inner 
 view; D, proximal end; E, distal end. X 0.50. Rancho La Brea Pleistocene. 
 
 Hapalops elongatus Ameghino. Right metatarsal IV, No. 15545 P. U. C. A 1 , B\ C l , D l , E l , same views as 
 in A, B, C, D, E. X 1.0. Santa Cruz Miocene. 
 
 small facet at the dorso-internal angle of the proximal face, not represented in Megalonyx 
 or in H. elongatus , forms a lateral continuation of the ectocuneiform surface of metatarsal 
 III. This ectocuneiform facet is separated by a broad interarticular channel from the 
 large cuboid facet. On the inner side the ectocuneiform surface is continuous with an 
 
 Table 48. —Measurements {in millimeters ) of phalanx III, digit III, pes, of N. shastense. 
 
 
 No. 
 
 1890-R-l 
 
 No. 
 
 1890-R-2 
 
 No. 
 
 1890-R-3 
 
 No. 
 
 1890-R-4 
 
 Greatest length from posterior end of posterior process to tip of 
 claw-process. 
 
 161.8 
 
 *158.4 
 
 *161.3 
 
 165 8 
 
 Greatest depth from dorsal surface of hood to discoid area on 
 subungual base. 
 
 75.8 
 
 72.7 
 
 75.4 
 
 Width of proximal end. 
 
 52.5 
 
 47.7 
 
 50.6 
 
 48.4 
 
 Greatest length of subungual base. 
 
 69 
 
 67 
 
 68.7 
 
 72.9 
 
 Width of ventral surface of claw-process just anterior to hood.. 
 
 19.8 
 
 21.8 
 
 25.4 
 
 21.5 
 
 Approximate. 
 
98 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 articulating face for metatarsal III. The latter facet is convex antero-posteriorly and 
 is situated more distally on the shaft than the facet for metatarsal V. The nearly flat 
 cuboid surface is large and widens ventrally, differing in the latter respect from Meg- 
 alonyx. It is less extensive than in Hapalops. The cuboid facet is separated from the 
 
 Fig. 43 .—Nothrotherium shastense Sinclair. Right metatarsal V, No. 1884-R-l. A, dorsal view; B, inner view; C, ventral 
 view. X 0.50. Rancho La Brea Pleistocene. 
 
 Hapalops elongatus Ameghino. Right metatarsal V, No. 15545 P. U. C. A 1 , B 1 , C 1 , same views as in A, B, C, 
 X 1.0. Santa Cruz Miocene. 
 
 large lateral facet for metatarsal V by a sharp angle, the degree of which is greater than 
 that in Hapalops. 
 
 The shaft is well rounded internally, but the outer side is flattened more than in 
 Hapalops. The distal extremity is expanded dorso-ventrally, but the articulating sui¬ 
 
 table 49 .—Measurements (in millimeters ) of metatarsal IV. 
 
 
 N othrotherium 
 shastense 
 
 No. 1883-R-l 
 
 N. shastense 
 No. 1883-R-2 
 
 N. shastense 
 No. 1883-R-3 
 
 N. s. hawveri 
 No.19875 U.C.C. 
 
 Greatest length. 
 
 120 
 
 123 
 
 121.9 
 
 106.7 
 
 Greatest depth of proximal end. 
 
 56.7 
 
 57.5 
 
 54 
 
 Greatest width of proximal end. 
 
 51.2 
 
 58.2 
 
 50 
 
 49 
 
 Greatest depth of distal end. 
 
 04.1 
 
 61 
 
 68.5 
 
 *67.7 
 
 Greatest width of distal end. 
 
 43.3 
 
 42.4 
 
 40.3 
 
 40 
 
 Least depth of shaft 
 
 26.3 
 
 26.3 
 
 29 
 
 25 
 
 
 1 Approximate. 
 
PLEISTOCENE MEGALONYCIIINAE AND MYLODONTIDAE OF RANCIIO LA BREA. 99 
 
 face is much less developed than in either Megalonyx or in the Miocene genus. The 
 dorso-external border of the metatarsal ends distally in a large tuberosity which sur¬ 
 mounts the carina. The latter is decidedly less prominent than in either Hapalops or 
 Megalonyx. It may be bordered along the inner, ventral side by a small offset. A 
 small sesamoid facet is situated on the ventral and external side of the carina. 
 
 METATARSAL V. 
 
 The outermost metatarsal in the pes of Nothrotherium (fig. 43 a, b, c), in contrast 
 to the corresponding metapodial of Megalonyx from Rancho La Brea, is a larger element 
 and resembles in some respects the fifth metatarsal of Hapalops more closely than does 
 the latter. The lateral process extends considerably beyond the facet for the cuboid, 
 and the distal end of the lateral margin of this process extends downward, forming a 
 decided ridge. In the deflection of the outer region of the lateral process, the meta¬ 
 podial differs noticeably from that of Hapalops (compare in fig. 43, c with c 1 )- The 
 shaft is rather broad, deepest along the inner side, and narrowing gradually to the outer 
 side. The articulating surface for the cuboid is slightly concave. The surface for meta¬ 
 tarsal IV is also slightly concave and approaches in size the cuboid facet. The facets 
 resemble in shape those in metatarsal V of Megalonyx and differ from those in the cor¬ 
 responding metapodial of Hapalops. At the distal end of the shaft a rounded ridge is 
 formed, which joins with a rudimentary digit. 
 
 Table 50. —Measurements {in millimeters) of metatarsal V, N . shastense, 
 
 No. 1884-R-l. 
 
 Length from union of proximal facets to distal end of shaft. 107.7 
 
 Greatest width from union of proximal facets to outer border of lateral process. 92.8 
 
 Greatest dorso-ventral extent of lateral process. 100.3 
 
 "Width of shaft distal to lateral process.. 43.4 
 
 Depth of proximal articulation. 43.9 
 
 Depth of distal end. 38 7 
 
 COMPARISON OF PES. 
 
 With the exception of the early representation of the metatarsal series of Megalonyx 
 jeffersonii by Leidy (1860) and the reconstruction of the pes of N othr other ium maquinense 
 by Winge (1915), little information is available concerning the structure of the hind foot 
 in Pleistocene Megalonychinae. More recently the incomplete information relating to 
 the structure of the pes in Nothrotherium and in Megalonyx has been reviewed by Betty 
 Schulthess (1920, pp. 102-105). The skeletal material from Rancho La Brea is sufficient 
 to permit an almost complete reconstruction of the pes (plate 15) and offers opportunity 
 to contrast adequately this part of the skeleton of the Pleistocene genus with that of the 
 L T pper Miocene Hapalops. 
 
 In general appearance, the pes of Nothrotherium is strongly suggestive of that of 
 Hapalops. A number of specializations are, however, exhibited. The modifications 
 have arisen as a result of change in position of the foot from a type resting with sole 
 on the ground to one resting mainly on the outer side. Such a foot has deviated perhaps 
 no less decidedly from that possessed by a primitive plantigrade mammal than has the 
 hind foot of a digitigrade form. In one the original horizontal plane of the foot inclines 
 forward, in the other it slopes principally to the outer side. Nothrotherium is not the only 
 ground-sloth so characterized, for in Mylodon the hind foot exhibits similar deformation. 
 
 These ground-sloths and their related forms may be spoken of as latero-pedal types, 
 employing the descriptive term to emphasize a character of the hind foot which under¬ 
 goes progressive or rather retrogressive, change in the phylogenetic history of the group. 
 Moreover, it should be recognized that in such a form as Mylodon , a change has also 
 taken place in position of front foot, suggested by structural modifications in the digits 
 similar to those seen in the digits of the hind foot. 
 
100 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 As in Mylodon harlani, the lateral metapodials have been strengthened and the 
 inner metatarsal has been reduced. The astragalus has been curiously modified in its 
 articulation with the foreleg. The pes of Hapalops, as reconstructed by Scott, already 
 shows a tendency to rest on the outer side, but the rotation has by no means progressed 
 to the extent seen in Nothrotherium. The calcaneum in the Pleistocene genus has a fan¬ 
 shaped posterior process similar to that in Hapalops, but lacking the great expansion 
 seen in Megalonyx. The astragalus becomes deeper in the Pleistocene form, and the 
 two divisions of the articulating surface for the tibia are bent at a sharp angle. In 
 this respect the astragalus of Megalonyx is less specialized than that of Nothrotherium 
 and is not greatly in advance of the type seen in Hapalops. In Nothrotherium the head 
 of the astragalus, articulating with the navicular, has also greater individuality than that 
 in Hapalops. In the distal row of tarsal elements there should be noted a more extensive 
 articulation between cuboid and metatarsal V than exists in Hapalops longiceps. 
 
 In the series of metatarsals, No. I is rudimentary and has fused with the ento- 
 cuneiform. In Mylodon harlani no element has been recognized as lying to the inner 
 side of the mesocuneiform and representing perhaps the combined entocuneiform and 
 metatarsal I. As in the pes of H. longiceps, metatarsal II is longer than metatarsal 
 III, but distinctly shorter than either the fourth or fifth metapodials. With the excep¬ 
 tion of the inner metapodial, metatarsal III is the shortest of the series. There is close 
 agreement between Miocene and Pleistocene forms in degree of divergence of metatarsal 
 IV from metatarsal III. This divergence is greater than that between the correspond¬ 
 ing elements of Megalonyx as shown by Leidy (1860, plate 6, fig. 1) and by the material 
 referred to Megalonyx from Rancho La Brea. Metatarsal V resembles the corresponding 
 metapodial of Hapalops more closely than does that of Megalonyx, particularly in the 
 shape of the lateral process. The downward and inward deflection of the external border 
 of the lateral process in Nothrotherium furnishes a broad support for the foot resting on 
 its side and accompanies the change in position of the pes from that seen in Hapalops. 
 
 The hallux has disappeared entirely in Nothrotherium, while in Hapalops the toe is 
 still functional. Digit II consists of three phalanges, of which the second is relatively 
 longer and more slender than that in the pes of the Miocene genus. A more decided 
 difference in size between the ungual phalanges borne by digits II and III exists in 
 Nothrotherium than in Hapalops. In the latter, phalanges I and II of digit III are sep¬ 
 arate, while in Nothrotherium, as in Megalonyx, they have co-ossified to form a single 
 element. Such co-ossification occurs occasionally in Mylodon. The co-ossification may 
 be associated, perhaps, with the large claw-phalanx in the Pleistocene megalonychids. 
 Metatarsal IV supported apparently such rudimentary elements as are present in the 
 corresponding digit of Mylodon. In Hapalops the digit was apparently fully formed 
 and functional. In specimens of metatarsal V from Rancho La Brea, a facet at the 
 distal end indicates the presence of a nodule. In the reconstruction of H. longiceps, 
 Scott shows a complete fifth digit in the pes. 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 101 
 
 CHARACTERS OF GENUS MEGALONYX. 
 
 Species of Megalonyx averaging larger in bulk than those of the genus Nothrotherium. 
 Skull short, broad, and deep, with anterior end truncated. Mandible deep without 
 spout-like predental region. Dentition f, f. First tooth separated by diastema from 
 cheek-tooth series and cross-section is meniscoid with inner bulge. Last superior 
 tooth tending to possess a triangular cross-section. Appendicular skeleton heavier than 
 in Notlirotherium. Wing-like, posterior process of calcaneum broad; astragalus retains 
 more the structure seen in Hapalops and is distinctly less modified than in Nothrotherium. 
 Metatarsal IV short; metatarsal V with relatively slender lateral process. 
 
 Genus MEGALONYX Jefferson. 
 
 Megalonyx. T. Jefferson, Trans. Amer. Philos. Soc. vol. 4, p. 248, 1799. 
 
 Megalonyx Jeffersonii Californicus Stock. 
 
 Megalonyx californicus Stock, Univ. Calif. Publ., Bull. Dept. Geol., vol. 7, p. 352, 1913. 
 
 DESCRIPTION OF MATERIAL. 
 
 It is an unusual character of the Rancho La Brea Pleistocene that relatively few 
 individuals of the genus Megalonyx have been preserved. Megalonyx is decidedly less 
 abundant in the asphalt beds than the genus Mylodon and occurs even more infrequently 
 than the genus Notlirotherium. Not more than four or five individuals have been 
 recorded. Unfortunately, the skull of Megalonyx is not available. The collection 
 includes, however, the lower jaw and representative parts of the skeleton. 
 
 MANDIBLE. 
 
 The mandible (No. 21429, U. C. C., plate 17) is well preserved, lacking only por¬ 
 tions of the coronoid processes. The specimen apparently belongs to an adult indivi¬ 
 dual and is smaller than the lower jaw of Megalonyx jeffersonii described by Leidy (1855, 
 pp. 13-15, plates 1 and 5). The horizontal ramus is deep and thick in the region of the 
 cheek-teeth, while the vertical ramus is plate-like and resembles somewhat that portion 
 in the lower jaw of Mylodon. At the anterior end a keel is developed along the upper 
 half of the symphyseal region and on each side of the median ridge is situated a large 
 mental foramen. Smaller foramina are also present. Behind the caniniform tooth the 
 outer face of the horizontal ramus is deeply excavated along the upper half. The notch 
 between the posterior border of the coronoid process and the condyle is relatively nar¬ 
 row, not wide as in Megalonyx jeffersonii. The condyle is large and is somewhat flat¬ 
 tened dorsally. The heavy mandible of Megalonyx from Rancho La Brea is decidedly 
 unlike the slender jaw of Nothrotherium. The latter difference is similar to that which 
 exists between the Miocene genera Eucholoeops and Hapalops. In comparing the man¬ 
 dible of Megalonyx with that of Eucholoeops it should be noted that the anterior end does 
 not extend in front of the caniniform teeth to the extent seen in the lower jaw of the 
 Miocene genus. Between the caniniform tooth and the cheek-tooth series the dorsal 
 portion of the ramus is narrower transversely in Megalonyx than in Eucholoeops. 
 
 DENTITION. 
 
 Dental formula: l f 
 4> 4* 
 
 The anterior tooth in the lower tooth-row is separated from the cheek-tooth series 
 by a diastema. This tooth is comparable to that present in Megalonyx jeffersonii and 
 possesses a convex external surface and an internal face which is developed into a con¬ 
 vexity along the median line. The posterior face of the first cheek-tooth is wider trans¬ 
 versely than the anterior face. Antero-posteriorly the tooth is widest toward the outer 
 side and only faint grooves are indicated on the lateral surface. The second cheek¬ 
 tooth is also widest antero-posteriorly toward the outer side. In contrast to the first 
 cheek-tooth, the second narrows decidedly toward the inner side. The principal trans¬ 
 verse axis of the third or last cheek-tooth is distinctly oblique to the fore-and-aft axis 
 
102 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 of the tooth-row. The obliquity of this tooth with reference to the long axis of the 
 tooth-row occurs not only in Megalonyx, but also in Nothrotherium, and seems to be 
 characteristic of the Megalonychinae. The tooth corresponds in shape to that in 
 Nolhr other ium and the triturating surface wears in a similar way. The inner surface is 
 round, while the outer face is practically flat. 
 
 Measurements {in millimeters) of mandible and lower dentition, No. 21429 U. C. C. 
 
 Length, anterior end of symphysis to posterior end of 
 
 angle. 287 
 
 Length, anterior end of symphysis to posterior end of 
 
 condyle. 264.8 
 
 Depth of ramus below middle of diastema. 80.5 
 
 Depth of ramus between first and second cheek-teeth 86 
 
 Depth of ramus at condyle. 122 
 
 Greatest width across outer sides of anterior teeth. . 89.6 
 
 Width across second cheek-teeth. 120 
 
 Width across outer ends of condyles. 161.5 
 
 Length of symphysis along inner side. 91 
 
 Thickness of ramus below second cheek-tooth. 49.5 
 
 Length of diastema. 42 
 
 Length of lower dental series, anterior end of ante¬ 
 rior tooth to posterior side of last cheek-tooth.. 137.3 
 Length of lower cheek-tooth series. 65 
 
 Least distance between inner sides of second cheek¬ 
 teeth. 19.8 
 
 Anterior tooth, greatest diameter. 35.5 
 
 Anterior tooth, width normal to greatest diameter. 16.5 
 
 First inferior cheek-tooth, greatest transverse diameter 26.5 
 
 First inferior cheek-tooth, greatest antero-posterior di¬ 
 ameter. 18 
 
 Second inferior cheek-tooth, greatest transverse diam¬ 
 eter. 28.3 
 
 Second inferior cheek-tooth, greatest antero-posterior 
 
 diameter. 17.8 
 
 Third inferior cheek-tooth, greatest transverse diam¬ 
 eter. 26.2 
 
 Third inferior cheek-tooth, greatest antero-posterior 
 
 diameter. 20.7 
 
 In the lower dentition of Eucholoeops, the first or caniniform tooth is also separated 
 from the cheek-teeth by a diastema. The tooth is either round or oval in cross-section 
 and never assumes the shape seen in anterior teeth of Megalonyx. The two anterior 
 cheek-teeth in Eucholoeops are oval, but with corners well rounded, in which respect 
 they differ from the semi-quadrate teeth of Megalonyx. The third or last tooth in Eu¬ 
 choloeops is either round or oval, but the transverse axis is already oblique to the long 
 axis of the tooth-row. 
 
 A single tooth (No. 6000-1, fig. 44, a, b) in the Los 
 Angeles Museum, collected from Pit 77, apparently repre¬ 
 sents the superior fifth of the right side. Specimen 6000-1 
 differs from the last superior tooth of Nothrotherium in shape 
 and in the large transverse diameter. The tooth approaches 
 that of Megalonyx in size, but it is somewhat smaller and 
 possesses a slightly different shape than the last superior 
 tooth usually present in Megalonyx jeffersonii. 
 
 A striking resemblance exists between the Rancho La 
 Brea specimen and a tooth described by Leidy (1855, pp. 
 
 45-46, plate 14, figs. 7, 8) from the Pleistocene near Natchez, 
 
 Mississippi, as Megalonyx dissimilis. The two teeth agree 
 very closely in size and are almost if not identical in shape. 
 
 Both the Natchez specimen and the Rancho La Brea tooth 
 differ in cross-section from the last superior tooth of M. 
 jeffersonii in lacking a distinctly triangular outline. No. 
 
 6000-1 is of greatest antero-posterior diameter toward the 
 lingual side, and for approximately two-thirds of the trans¬ 
 verse distance there is no appreciable difference in this diameter. In the outer one-third 
 the tooth narrows considerably, and in the corresponding region of the anterior face 
 there is developed a shallow longitudinal groove. A broad and shallow longitudinal 
 groove is present in median line on the posterior face. Anteriorly and internally the 
 tooth is longitudinally concave. 
 
 The close similarity between No. 6000-1 from Rancho La Brea and the fifth superior 
 tooth from near Natchez, Mississippi, indicates that perhaps the same specific form is 
 represented. Whether these teeth are to be referred to a species distinct from Mega¬ 
 lonyx jeffersonii is, however, another question. Cope (1899, p. 213) concluded from a 
 
 Fig. 44. —Megalonyx sp. A and B, 
 fifth superior tooth, No. 6000-1. 
 X 1.0. Rancho La Brea Pleistocene. 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 103 
 
 study of the superior cheek-teeth of Megalonyx wheatleyi that the last superior tooth 
 varies somewhat in shape. Cope, therefore, does not regard the characters given by 
 Leidy for M. dissimilis as of specific importance and states that the species is synony¬ 
 mous with M. jeffersonii. 
 
 VERTEBRAE. 
 
 Only a few vertebrae of Megalonyx (plate 19) are available from Rancho La Brea. 
 
 A single specimen of the first dorsal vertebra (No. 22111, U. C. C.) is available 
 and lacks only the posterior epiphysis. The vertebra closely resembles the corresponding 
 segment in Nothrotherium (plate 18, figs, 1, la, 2). In both forms the spine of the first 
 dorsal is elongated, rising straight above the neural canal, and is thickened terminally, 
 thus differing from the spine in the vertebra of Mylodon. The centrum is relatively not 
 so long as in Hapalops elongalus. Remnants of the haemapophyses are to be seen on the 
 ventral surface of the centrum. The transverse process as in Nothrotherium is more 
 slender than in the Miocene species. There is no small canal perforating this process 
 at the base of the metapophysis as in Nothrotherium. The concave articulating facet 
 for the tubercle of the rib is quite large, while the surface for the capitulum is even 
 smaller than that in Nothrotherium. The position of the facets is more like in the latter 
 genus than in H. elongatus. The anterior zygapophyses are relatively much smaller 
 than in H. elongatus. 
 
 Table 51. —Measurements (in millimeters ) of dorsal vertebrae. 
 
 
 No. 22111 
 
 U. C. C. 
 
 No. 24243 
 
 U. C. C. 
 
 No. 24244 
 
 U. C. C. 
 
 Length of centrum. 
 
 4o 
 
 *50 
 
 57.7 
 
 Width of centrum measured over anterior face. 
 
 40.1 
 
 59.8 
 
 76 
 
 Depth of centrum measured over anterior face. 
 
 29.4 
 
 47.3 
 
 61.1 
 
 Width across metapophyses. 
 
 90.4 
 
 Width across anterior zygapophyses. 
 
 82.2 
 
 84.2 
 
 Greatest width across transverse processes. 
 
 145.6 
 
 130.9 
 
 143.2 
 
 Length of neural arch along middle and at base of dorsal spine. 
 
 50.8 
 
 67.7 
 
 72 
 
 Height, measured from middle of ventral border, posterior face of centrum to 
 end of dorsal spine. 
 
 *192 
 
 1 185 
 
 214 
 
 Dorso-ventral diameter of neural canal, measured at anterior end. 
 
 30 
 
 39.5 
 
 47 
 
 Transverse thickness of dorsal spine at middle. 
 
 17.5 
 
 21.9 
 
 17.3 
 
 Transverse thickness of end of dorsal spine. 
 
 32.8 
 
 36.8 
 
 43.5 
 
 Antero-posterior extent of end of dorsal spine . 
 
 55.0 
 
 49 
 
 54.6 
 
 
 
 1 Approximate. 
 
 Two additional dorsal vertebrae are available in the collections from Rancho La 
 Brea now at the University of California. No. 24243 comes presumably from the 
 anterior thoracic region, while No. 24244 belongs to the posterior part of the series. 
 Both vertebrae are decidedly larger than corresponding segments in Nothrotherium, but 
 resemble the latter fairly closely in shape. In each specimen the neural spine thickens 
 considerably at the dorsal end. The anterior and posterior zygapophyses form rather 
 large articulating surfaces. 
 
 Table 52.—Measurements (in millimeters) of caudal vertebrae. 
 
 
 No. 6003-1 
 
 No. 6003-2 
 
 Length of centrum measured along ventral surface.. 
 
 38.2 
 
 34.6 
 
 Width of centrum measured over anterior face. 
 
 52 
 
 48.2 
 
 Depth of centrum measured over anterior face.... 
 
 42.8 
 
 39.6 
 
 Width across transverse processes. 
 
 116.3 
 
 ' 105 
 
 Two caudal vertebrae, Nos. 6003-1, 6003-2, in the collections of the Los Angeles 
 Museum (plate 18, figs. 7, 7a, 8, 8a) presumably belong to Megalonyx. Both specimens 
 
104 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 lack a complete neural arch and apparently come from the posterior region of the caudal 
 series. The specimens are much larger than posterior caudal vertebrae of Nothrotherium 
 and differ from them considerably in shape. The centra are compressed antero-pos- 
 teriorly, much more so than in caudal vertebrae of the latter genus. The anterior and 
 posterior faces of the centrum in each vertebra are closer together dorsally than ven- 
 trally. The wing-like transverse processes are far better developed than in caudal 
 vertebrae of Nothrotherium, in which the neural arch has been greatly reduced. The 
 transverse processes in the specimens from the asphalt beds are thick posteriorly and 
 have a decided slope forward and downward. 
 
 HUMERUS. 
 
 The humerus (No. 21003, U. C. C., plate 19) is close in structure to that of Mega¬ 
 lonyx jeffersonii figured by Leidy. It is somewhat smaller, but this discrepancy in size 
 may be due in part to a difference in age of the animals represented. In the specimen 
 from the asphalt deposits the head has not fused entirely with the shaft, a separation 
 which is not indicated in the humerus shown by Leidy. In anterior view (plate 19, 
 fig. 1), the head is more prominently placed than in the humerus of M. jeffersonii. It 
 is also more prominent than in the upper arm-bone of Nothrotherium (compare with 
 plate 9, fig. 1). The head is also like that in Hapalops in its proportions—-the antero¬ 
 posterior diameter being decidedly greater than the transverse. In Nothrotherium the 
 antero-posterior diameter is equal to the transverse diameter. As stated in the descrip¬ 
 tion of the humerus of Nothrotherium, the greater and lesser tuberosities are connected 
 by a shelf of bone in front of the head, which is distinctly less developed in Megalonyx. 
 In the latter genus, also, there is no deep gutter separating head from lesser tuberosity, 
 as in Nothrotherium, in which respect Megalonyx is like Hapalops. The bicipital groove 
 is narrower and deeper than in the humerus of Nothrotherium, but no deep channel tra¬ 
 verses the anterior surface of the lesser tuberosity, as in the latter genus. 
 
 Fig. 45 .—Megalonyx jeffersonii californicus Stock. Right unciform, No. 22775, U. C‘ C. A, proximal view; 
 B, dorsal view; C, ventral view; D, distal view. X 0.50. Rancho La Brea Pleistocene. 
 
 The pectoral and deltoid crests are both prominent features, and between them is 
 situated a third ridge which is also distinct. This short middle ridge is present also in 
 Hapalops. The deltoid crest swings well to the outer side of the shaft, much more so 
 than in the humerus of Nothrotherium. An entepicondylar foramen is present, inclosed 
 by a wide bar. The internal tuberosity is massive and supports on the superior surface 
 a shallow but broad groove extending from the superior opening of the entepicondylar 
 foramen to the outer extremity, and which near the outer end is completely arched by 
 bone to form a closed canal (plate 19, fig. 2). The distal contour of the trochlea resembles 
 that in Megalonyx jeffersonii and differs from that in Hapalops and Nothrotherium. The 
 ulnar surface is wider transversely than in the humerus of the species described by 
 Leidy. The surface is relatively wider antero-posteriorly than that in the humerus of 
 N othrotherium. 
 
 Table 53. —Measurements {in millimeters) of humerus, No. 21003. 
 
 Greatest length. 447 
 
 Width at proximal tuberosities. 128 
 
 Width of shaft at middle. 81.3 
 
 Least width of shaft. 60.5 
 
 Thickness of shaft at end of deltoid crest. 59 
 
 Greatest width of distal expansion. 221 
 
 Width of distal trochlea. 127.5 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCIIO LA BREA. 105 
 
 ULNA. 
 
 A single specimen of an ulna (No. 23192, U. C. C., plate 20) of Megalonyx probably 
 represents the same individual to which the humerus belongs. In length it is some 43 
 mm. shorter than the ulna of M. jeffersonii described by Leidy and approximately 16 
 mm. longer than the ulna of Nothrotherium, also from Rancho La Brea. 
 
 The ulna possesses a much greater antero-posterior diameter than does the corre¬ 
 sponding forearm element of Nothrotherium] but in transverse diameter it is, rela¬ 
 tively and absolutely narrower. In greater antero-posterior diameter of shaft the ulna 
 of Megalonyx is less like that of Hapalops than is that of Nothrotherium. Below the sig¬ 
 moid notch the inner surface is deeply depressed, much more so than in the ulna of 
 Nothrotherium. A large facet articulates with the radius. The olecranon process is rela¬ 
 tively longer and narrower than that in the ulna of Nothrotherium. At the distal end the 
 surface articulating with the carpus is more extensive than in Nothrotherimn. 
 
 Table 54 . — Measurements (in millimeters ) of ulna, No. 23192, U. C. C. 
 
 Greatest length. 464.6 
 
 Greatest width from summit of coronoid process. .. . 125.4 
 
 Width of shaft at middle. 69.2 
 
 Thickness of shaft at middle. 33 
 
 Greatest width of shaft at distal end. 
 
 Thickness of distal articulating surface. 
 
 Distance from summit of coronoid process to end of 
 olecranon. 
 
 74.3 
 46.6 
 
 152.3 
 
 UNCIFORM. 
 
 The unciform (No. 22775, U. C. C., fig. 45, a to d) resembles very closely that 
 described and figured by Leidy (1855, p. 32, plate 8, figs. 10, 12, 13 b) for Megalonyx 
 jeffersonii. It is also quite similar in shape to the corresponding element in Hapalops. 
 In dorso-palmar extent the unciform of Hapalops is deeper along the external side, but 
 the dorso-internal angle is less prominent than in Megalonyx. With the inner ventral 
 portion of the cuneiform surface forming a boss or knob, this surface is sigmoid in both 
 the Miocene and Pleistocene forms. 
 
 Fig. 46 .—Megalonyx jeffersonii californicus Stock. Patella, No. 22776 U. C. C. .4, anterior 
 view; B, femoral view. X 0.50. Rancho La Brea Pleistocene. 
 
 The internal articulating face for the magnum resembles that in Hapalops, but is 
 relatively somewhat less extensive. It is continuous along its proximal border with the 
 cuneiform surface. In the Miocene genus the facet for metacarpal III is relatively 
 wider than in Megalonyx and is grooved in its dorsal half, the furrow deeply notching the 
 dorsal border of the distal end. The facet for metacarpal III in No. 22775 extends 
 the whole dorso-palmar depth of the unciform, as in M. jeffersonii, but dorsally is not as 
 wide as in the latter. Its dorsal border lacks entirely the notching noted in Hapalops, 
 The facet is concave transversely in the lower half. The facet for metacarpal IV is 
 larger than in M. jeffersonii. It is concave, but relatively not so deep as in the Miocene 
 genus, and the facet is less extensive. The lateral facet for the fifth metacarpal occu¬ 
 pies a position much as in Hapalops. 
 
106 
 
 CENOZOIC GR AVI GRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 The measurements (in millimeters) of unciform (No. 22775) are: length, proximo- 
 distal diameter, measured from plane of dorsal half of facet for metacarpal III to cunei¬ 
 form surface, 35; width, 52.4; depth, dorso-palmar diameter, 46.9. 
 
 PATELLA. 
 
 Two specimens of the patella (fig. 46, a, b) are available, one in the collections of 
 the Los Angeles Museum and another (No. 22776) in the collections of the University 
 of California. 
 
 The knee-cap of the California form resembles closely that in Megalonyx jeffersonii, 
 differing apparently only in minor details. The specimens from the asphalt beds are 
 slightly smaller and approach more nearly a triangular shape than does the patella of 
 the latter species as described by Leidy. 
 
 The patella of Megalonyx differs distinctly in shape from that of Nothrotherium (see 
 plate 12, figs. 2 and 3). In Megalonyx the knee-cap is wide transversely and articulates 
 with the femur by a broad surface. In Nothr other ium this facet is deep. The patella of 
 Megalonyx resembles that of Hapalops closely in shape. In the Miocene genus the ele¬ 
 ment is triangular, broad above where it articulates with the femur and narrowing 
 below the femoral facet. 
 
 Table 55. —Measurements (in millimeters) of patella. 
 
 
 No. 22776, 
 
 u. c. c. 
 
 No. 6001-1 
 L. A. M. 
 
 Dorso-ventral diameter. 
 
 106 
 
 109 
 
 Width. 
 
 99 
 
 90.8 
 
 Dorso-ventral diameter of femoral surface. 
 
 52 
 
 45.4 
 
 Transverse diameter of femoral surface. 
 
 97.6 
 
 90 
 
 Thickness through femoral surface. 
 
 42.2 
 
 41 
 
 CALCANEUM. 
 
 A well-preserved calcaneum of Megalonyx , No. 20095, U. C. C., from Rancho La 
 Brea is shown in figure 47, a, b, c. No. 20095 resembles rather closely the correspond¬ 
 ing element in Megalonyx jeffersonii. The specimen from the asphalt beds differs from 
 
 Fi<;. 47 .—Megalonyx jeffersonii californicus Stock. Left calcaneum, No. 20095 U. C. C. A, outer view; 
 B, view of articulating end; C, inner view. X 0.33. Rancho La Brea Pleistocene. 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 107 
 
 that of Nothrotherium (see p. 85, fig. 30, a, b, c) in having a short and broad neck and a 
 great posterior expansion. 
 
 The articulating end is, on the whole, similar to that of M. jeffersonii and differs in 
 shape from that of the calcaneum of Nothrotherium (compare fig. 30 b and fig. 47 b). The 
 grooves separating the three articular facets are well defined. The groove between the 
 large astragalar facet and that for the cuboid is particularly broad. The large astragalar 
 facet is flat externally and becomes slightly concave internally and curves down to 
 the inferior border. The small astragalar facet is concave in its longest diameter. 
 The cuboidal facet is concave vertically and curves toward the external margin. The 
 inner lip is flattened almost at right angles to the rest of the surface and articulates 
 with the astragalus. 
 
 The measurements (in millimeters) of calcaneum, No. 20095, of Megalonyx jeffersonii 
 californicus are as follows: width of neck, 67.8; greatest antero-posterior diameter, 206; 
 greatest diameter of tuber calcis, 202; greatest thickness at posterior border, 38.2. 
 
 METATARSAL III AND DIGIT III. 
 
 The third metatarsal (No. 20001, U. C. C., fig. 48, a to d) differs from that in Nothro- 
 therium in possessing greater antero-posterior length. The metapodial is, however, much 
 less extended in this direction than is the corresponding element of Hapalops. The ecto- 
 cuneiform surface is more constricted at the middle than in Nothrotherium, and the ven¬ 
 tral process of the proximal end is broader. The facet for metatarsal II is flattened or 
 slightly convex, with greatest diameter dorso-ventral. On the lateral side the facet for 
 metatarsal IV has a smaller area, but is longer antero-posteriorly than in Nothrotherium. 
 At the distal end the carina is much thicker than in Nothrotherium and resembles Hapa¬ 
 lops. It has a large articulating offset along the inner side, which is especially broad 
 dorsally. The offset along the outer side is narrower than in Nothrotherium. 
 
 Fig. 48 .—Megalonyx jeffersonii calif ornicus Stock. Right metatarsal III, No. 20001 U. C. C. A, proximal view; 
 B, outer view; C, inner view; D, distal view. X 0.50. Rancho La Brea Pleistocene. 
 
 The element formed by the co-ossification of the first and second phalanges (No. 
 20004) of Megalonyx approaches that of Nothrotherium more closely in size than does that 
 of M. jeffersonii. Some of the characters of this element have already been noted in a 
 description of the corresponding segment of Nothrotherium. The ventral portion of the 
 proximal extremity is much thicker antero-posteriorly than in Nothrotherium. At the 
 distal end the median groove of the trochlea is narrower dorsally than in M. jeffersonii, 
 and the dorsal depression is smaller. The lateral condyle is practically of same depth 
 as in one of the specimens from Rancho La Brea belonging to Nothrotherium. 
 
 Table 56 .—Measurements (in millimeters ) of metatarsal IV, No. 6002-1. 
 
 Greatest length. 
 
 Greatest depth of proximal end 
 Greatest width of proximal end 
 
 93.4 
 65 
 
 49.4 
 
 Greatest depth of distal end 
 Greatest width of distal end 
 Least depth of shaft. 
 
 63.3 
 41.6 
 
 33.4 
 
108 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 The ungual phalanx (No. 20004) is intermediate in size between the corresponding 
 ungual phalanges of M. jeffersonii and Nothrotherium. In the Rancho La Brea specimen 
 (C. Stock, 1913, p. 357, figures 17 and 18) the anterior portion of the claw-process, as 
 well as part of the hood, is not preserved. The claw-process appears to be relatively 
 more slender than in N othr other ium. In both genera the claw is of the compressed type, 
 differing thus from Mylodon. The bony sheath is thickest on the inner side. The basal 
 discoid area is large. On each side posteriorly is a foramen with the larger on the inner 
 side. The median ridge of the proximal articulating surface is thin above and widens 
 below. It is much less robust than in Nothr otherium. The overhanging process is not 
 so deeply depressed posteriorly as in the latter form. 
 
 METATARSAL IV. 
 
 The fourth metatarsal (No. 6002-1, fig. 49, a to d) is shorter than that of Megalonyx 
 jeffersonii. It is decidedly shorter and heavier than the corresponding metatarsal of 
 Nothr other ium. In this character the Rancho La Brea Megalonyx is less like Hapalops 
 than is N othr otherium. The proximal extremity has greater depth than in Nothr other ium 
 and is more as in Hapalops. It is, however, more nearly rectangular in shape, due to a 
 greater width ventrally. The cuboid facet has greater dorso-ventral extent and is 
 narrower ventrally than in metatarsal IV of N othr other ium. It is to a slight extent sigmoid 
 dorso-vent.rally, with the dorsal half well inclined toward the inner side. The proximal 
 surface differs further from N othr other ium in lacking a broad ectocuneiform facet. 
 
 Fig. 49. —Megalonyx jeffersonii californicus Stock. Right metatarsal IV, No. 6002-1. A, inner view; B, dorsal view; 
 C, outer view; D, proximal view. X 0.50. Rancho La Brea Pleistocene. 
 
 The facet for metatarsal III extends farther along the shaft than in Nothr otherium. 
 The facet for metatarsal V, in contrast to that of Nothrotherium, is situated oblique to 
 the longitudinal axis of the bone, whereas in the latter genus it parallels this axis. 
 
 The position of the facet for metatarsal V, together with the inwardly directed facet 
 for the cuboid, may denote a greater divergence of the two outer metatarsals in Mega¬ 
 lonyx than occurs in N othrotherium. The shaft in the Rancho La Brea specimen is quad¬ 
 rate in section, with no development of a dorso-external crest. On the distal extremity 
 the most prominent feature is the very thick carina, resembling that in Hapalops and 
 differing from that in Nothrotherium. There is no offset along the outer side, while that 
 along the inner side is much more reduced than in the Miocene form. The carina is not 
 surmounted by such a heavy tuberosity as in Nothrotherium. A small facet for a sesa¬ 
 moid is present on the ventro-external side of the carina. 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF IIANCHO LA BREA. 109 
 
 METATARSAL V. 
 
 A right and left metatarsal V (Nos. 22773, 22774) are in the collections of the Uni¬ 
 versity of California and probably belong to the same individual. The specimens from 
 the asphalt beds are similar in shape, but smaller than the fifth metatarsal of Megalonyx 
 jeffersonii. Metatarsal V (fig. 50, a, b) differs decidedly in shape from the corresponding 
 metapodial of Nothrotherium. In general appearance this element in Megalonyx is less 
 like that of Hapalops than is that of Nothrotherium. The lateral process in the Rancho 
 La Brea specimen extends almost at right angles to the shaft, the outer end not reaching 
 far, if at all, proximally beyond the articulating surface for the cuboid. In Nothrotherium 
 the lateral process extends proximally considerably beyond this surface and the outer 
 border forms a pronounced ridge beneath the element. 
 
 Fig. 50 .—Megalonyx jeffersonii californicus Stock. Right 
 metatarsal V, No. 22773 U. C. C. A, dorsal view; 
 
 B, inner view. X 0.50. Rancho La Brea Pleistocene. 
 
 The shaft in Megalonyx is shorter and less plate-like than in Nothrotherium. The 
 articulating surfaces for cuboid and metatarsal IV are subequal and are somewhat com¬ 
 parable in shape to the corresponding surfaces in Nothrotherium. Both Megalonyx and 
 Nothrotherium differ in shape of these facets from Hapalops. In metatarsal V of the Mio¬ 
 cene genus the articulating surfaces for cuboid and metatarsal IV are elongated dorso- 
 ventrally, the former being wide below and the latter being wide above. In Megalonyx 
 the articulating surface for the rudimentary digit extends the full depth of the distal 
 end of the shaft and is slightly convex. 
 
 Table 57 .—Measurements {in millimeters) of metatarsal V. 
 
 
 No. 22773, 
 
 U. C. C., right. 
 
 No. 22774, 
 
 U. C. C., left. 
 
 Length from angle formed by proximal facets to distal end. 
 
 85.1 
 
 85 
 
 Greatest width, from angle formed by proximal facets to outer border of lateral 
 process. 
 
 85.3 
 
 SO. 7 
 
 Width of lateral process. 
 
 39.6 
 
 46.9 
 
 Width of shaft anterior to lateral process. 
 
 32.3 
 
 32.5 
 
 Depth of proximal articulating end. 
 
 40.9 
 
 45.9 
 
 Depth of distal end. 
 
 40.2 
 
 39.7 
 
 
OCCURRENCE AND DESCRIPTION OF MEGALONYCHINAE IN THE 
 PLEISTOCENE OF WESTERN NORTH AMERICA. 
 
 GREAT BASIN PROVINCE. 
 
 White Bluffs, near Hanford, south central Washington. —The White Bluffs consist of 
 a series of sedimentary deposits, probably of flood-plain origin, exposed in section sev¬ 
 eral hundred feet in height along the eastern bank of the Columbia River, 12 miles above 
 Pasco, Washington. The beds were examined by J. C. Merriam and J. P. Buwalda' 1 
 and were designated the Ringold formation. The Ringold is regarded as Pleistocene in 
 age. The following collection is listed from a locality in the Ringold formation near 
 Hanford: 
 
 Megalonyx, small species, probably new. 
 
 Equus or Pliohippus, calcaneum. 
 
 Camelid, small, near Pliauchenia. 
 
 Camelid, possibly larger than first-mentioned 
 form. 
 
 Cervid, antler fragments. 
 
 Leporid, new genus. 
 
 Fish vertebrae and other elements of skeleton. 
 
 Testudo fragments. 
 
 It would be interesting to know 
 the relationship of the White Bluffs 
 fauna to that of the Washtuckna Lake 
 Pleistocene, Franklin County, Washing¬ 
 ton, for the two localities are not very 
 distantly separated. Horses, camels, 
 and deer occur at both localities, but 
 Mylodon is the ground-sloth recorded 
 from Washtuckna Lake. 
 
 A fragmentary portion of the skull 
 with right upper cheek-tooth series (No. 
 22779, fig. 52) and a single vertebra 
 (No. 24118) from localities 3031 and 
 3031 b, respectively, in the White Bluffs 
 near Hanford, Washington, represent 
 the specimens available. 
 
 DESCRIPTION OF GROUND-SLOTH MATE¬ 
 RIAL FROM WHITE BLUFFS 
 NEAR HANFORD. 
 
 In the skull-fragment (No. 22779 
 U. C. C.) some of the suture lines are 
 still well defined and the specimen may 
 not belong to a fully mature individual. 
 Judging from specimen, the ground- 
 sloth from the White Bluffs was of 
 slightly smaller size than Megalonyx 
 wheatleyi and approximately two-thirds 
 the size of an adult individual of the 
 species M. jeffersonii. 
 
 Fig. 51. —Map showing occurrences of Pleistocene Megalonychinae 
 in western North America. 1, Rancho La Brea; 2, Alameda 
 County; 3, Hawver Cave, Eldorado County; 4, Potter Creek Cave 
 and Samwel Cave, Shasta County; 5, Bautista Creek Badlands, 
 Riverside County; 6, White Bluffs near Hanford, south central 
 Washington; 7, Murphys, Calaveras County; 8, Douglas City, 
 Trinity County; 9, San Pedro, Los Angeles County. 
 
 “J. C. Merriam and J. P. Buwalda. Age of strata referred to the Ellensburg formation in the White Bluffs of the 
 Columbia River. Univ. Calif. Publ., Bull. Dept. Geol., vol. 10, pp. 255-266, plate 13, 1917. 
 
 110 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. Ill 
 
 The cheek-teeth of No. 22779 (fig. 52) resemble in their general characters those of 
 other Pleistocene species of Megalonyx. They are smaller than corresponding teeth of 
 M. wheatleyi and of M. leidyi, but are slightly larger than those of Nothrotherium. The 
 first cheek-tooth is wider externally than that shown by Leidy (1855, plate 3) in ventral 
 view of the skull of M. jeffersonii. In Leidy’s specimen the anterior face of the tooth 
 seems to round into the postero-external face, while in No. 22779 a distinct outer flat 
 
 Fig. 52 .—Megalonyx sp. Skull-fragment with superior cheek-teeth of right side. X 1.0. White 
 
 Bluffs, near Hanford, Washington. 
 
 face is developed. The tooth resembles more that in the specimen of M. jeffersonii 
 shown as figure 1 on plate 6 of Leidy’s memoir. The last cheek-tooth of No. 22779 
 resembles more the corresponding tooth of other species of Megalonyx than it does that 
 of Nothrotherium. The tooth in the species from the White Bluffs differs, however, from 
 that of M. jeffersonii, M. wheatleyi, and M. leidyi in greater width of internal or medial 
 side. In Nothrotherium this tooth is subject to some variation in shape. 
 
 A single vertebra (No. 24118, from Univ. Calif. Coll., Loc. 3031b) is the only addi¬ 
 tional specimen of the ground-sloth available from the White Bluffs Pleistocene. The 
 vertebra belongs to the caudal series, but unfortunately lacks the neural arch and the 
 
112 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 greater portions of the transverse processes. The anterior and posterior epiphyses are 
 thoroughly fused with the body of the vertebra. The antero-posterior diameter of the 
 centrum is slightly less than the transverse diameter. Two nutritive foramina, one on 
 each side of the median line, are present on the dorsal surface of the centrum, while the 
 ventral surface at about the middle is perforated by a single foramen. In caudal verte¬ 
 brae of Nothrotherium several foramina are present ventrally. A well-defined canal on 
 the ventral surface of the centrum leads backward and outward from each side of the 
 middle foramen. This canal passes around the posterior side of the base of the trans¬ 
 verse process and extends upward, forward, and inward to the nutritive foramina of the 
 dorsal surface. 
 
 Two pairs of chevron facets were presumably present in No. 24118, but the 
 articulating surfaces have been badly weathered. The two posterior surfaces are much 
 larger than the two anterior and are situated close together. In the caudal series of 
 Nothrotherium a decided difference in size between posterior and anterior chevron facets 
 becomes noticeable in the sixth vertebra from the anterior end. The transverse process 
 in No. 24118 is quite stout posteriorly, lessening in thickness toward the anterior side. 
 The posterior side is situated much more closely to the posterior face of the centrum 
 than in caudal vertebrae of Nothrotherium. 
 
 Characters possessed by the specimens from the White Bluffs Pleistocene indicate 
 the presence of Megalonyx, but the type may be specifically separable from other known 
 forms of the genus. As stated above, it is possible, however, that the individual repre¬ 
 sented by the skull-fragment may not be an adult. There is reason for believing, also, 
 that the shape of the superior cheek-teeth, particularly that of the last upper tooth, 
 may vary somewhat in megalonychid sloths, and in the present instance may not 
 furnish a character of truly diagnostic value. Definite specific determination is there¬ 
 fore withheld and awaits fuller information. 
 
 Table 58 .—Measurements {in millimeters ). 
 
 Dentition, No. 22779: 
 
 Length of cheek-tooth series, from anterior face 
 
 of 2 to posterior face of 5. 58.8 
 
 2, greatest antero-posterior diameter. 12.7 
 
 2, greatest transverse diameter. 14.8 
 
 3, greatest antero-posterior diameter. 13.3 
 
 3, greatest transverse diameter. 18.2 
 
 4, greatest antero-posterior diameter. 12.6 
 
 Dentition, No. 22779— Continued. 
 
 4, greatest transverse diameter. 18.4 
 
 5, greatest antero-posterior diameter. 9.1 
 
 5, greatest transverse diameter. 16.4 
 
 Caudal vertebra. No. 24118: 
 
 Antero-posterior diameter of centrum. 37 
 
 Transverse width across posterior face. 41.8 
 
 Depth measured across middle of posterior face. 34.3 
 
 PACIFIC COAST PROVINCE. 
 
 Douglas City, Trinity County. —A last inferior tooth of the right side (No. 6739, 
 U. S. Nat. Mus.) was collected in 1910 by J. S. Diller. The label accompanying the 
 specimen states that the localitj 7, is the Union Hill Mine. 
 
 Specimen 6739 is smaller than corresponding teeth of Megalonyx jeffersonii, but 
 larger than comparable teeth of Nothr other ium. It approaches in dimensions the last 
 inferior tooth of Megalonyx sierrensis. The top of the pulp-cavity is situated approx¬ 
 imately 35 mm. below the triturating surface. No. 6739 possesses the characteristic 
 shape seen in the last inferior tooth (4) of Megalonyx and of Nothr other ium. The anterior 
 and posterior faces show slight curvature in dorso-ventral extent. A well-marked groove 
 traverses the length of the posterior face. The greatest transverse diameter is 23 mm.; 
 greatest antero-posterior diameter, 18 mm. 
 
 Shasta and Eldorado Counties. —Limestone caves situated in the mountain belt of 
 California, east and northeast of the Sacramento Valley, have yielded important Pleisto¬ 
 cene mammalian faunas in which ground-sloths occur. The three principal caverns fur- 
 
PLEISTOCENE MEGALONYCIIINAE AND MYLODONTIDAE OF RANCHO LA BREA. 113 
 
 nishing vertebrate remains are Potter Creek Cave,° near Baird, on the McCloud River, 
 in Shasta County; Samwel Cave, * * 6 on the McCloud, approximately 15 miles north of 
 Baird; and Hawver Cave/ on the Middle Fork of the American River, 5 miles east of 
 Auburn, Eldorado County, and approximately 150 miles southeast of the Shasta caves. 
 
 Table 59.— Pleistocene mammalian faunas from three cave deposits of California. 
 
 POTTER CREEK CAVE. 
 
 Scapanus latimanus (Bachman). 
 
 Antrozous pallidus pacificus C. H. Merriam. 
 JArctotherium simum Cope. 
 fUrsus, n. sp. 
 
 JAenocyon dims (Leidy). 
 
 Vulpes cascadensis C. H. Merriam. 
 
 Urocyon cinereoargenteus townsendi C. H. Merriam. 
 Bassariscus astutus raptor (Baird). 
 
 Mustela arizonensis (Mearns). 
 fSpilogale, n. sp. 
 
 Mephitis occidentalis Baird. 
 
 Taxidea, n. sp. (?). 
 
 tFelis, probably n. sp. „ 
 
 fFelis, n. sp. 
 
 Lynx fasciatus Rafinesque. 
 
 Lynx fasciatus, n. subsp. (?). 
 
 Neotoma cinerea occidentalis Baird. 
 
 Microtus californicus (Peale). 
 fThomomys microdon Sinclair. 
 
 Thomomys leucodon C. H. Merriam. 
 
 *Aplodontia, californica fossilis Sinclair. 
 
 Marmota flaviventer (Audubon and Bachman). 
 
 Citellus beecheyi douglasi (Richardson). 
 
 Eutamias sp. 
 
 Callospermophilus chrysodeirus C. H. Merriam. 
 Sciurus douglasi albolimbatus Allen. 
 
 Sciuropterus alpinus klamathensis C. H. Merriam. 
 Sylvilagus auduboni (Baird). 
 
 JNothrotherium shastense Sinclair. 
 
 JMegalonyx wheatleyi (?) Cope. 
 
 JMegalonyx jeffersonii (?) (Desmarest). 
 JMegalonyx, n. sp. 
 
 JMegalonyx sp. 
 
 JCamelid. 
 f Bison sp. 
 
 Odocoileus, sp. a. 
 
 Odocoileus, sp. b. 
 
 JEuceratherium collinum Sinclair and Furlong. 
 OreamnOs americanus (Ord). 
 
 JPlatygonus (?), sp. 
 fEquus occidentalis Leidy. 
 fEquus pacificus Leidy. 
 
 JMammut americanum (Kerr). 
 fElephas primigenius Blumenbach. 
 
 SAMWEL CAVE. 
 
 Ursus americanus Pallas. 
 fUrsus, n. sp. 
 
 Ursus sp. 
 
 Vulpes sp. 
 
 Urocyon cinereoargenteus townsendi C. H. Merriam. 
 Procyon, near lotor (Linnaeus). 
 
 Mustela arizonensis (Mearns). 
 
 Mustela sp. 
 
 Mephitis occidentalis Baird. 
 fFelis, probably n. sp. 
 
 Peromyscus maniculatus gambeli (Baird). 
 
 Neotoma cinerea occidentalis Baird. 
 
 Microtus calif ornicus (Peale). 
 fThomomys microdon Sinclair. 
 
 Thomomys leucodon C. H. Merriam. 
 
 Erethizon epixanthum Brandt. 
 
 *Aplodontia californica fossilis Sinclair. 
 
 Citellus beecheyi douglasi (Richardson). 
 
 Callospermophilus chrysodeirus C. H. Merriam. 
 Sciurus griseus fossilis Kellogg. 
 
 Sciurus douglasi albolimbatus Allen. 
 
 Sciuropterus alpinus klamathensis C. H. Merriam. 
 Castor subauratus Taylor. 
 
 Lepus washingtoni klamathensis C. H. Merriam. 
 Sylvilagus auduboni (Baird). 
 
 JNothrotherium shastense Sinclair. 
 
 JNothrotherium sp. 
 
 JMegalonyx sp. 
 
 Odocoileus, sp. a. 
 
 Odocoileus, sp. b. 
 
 JEuceratherium collinum Sinclair and Furlong. 
 JPreptoceras sinclairi Furlong. 
 
 Oreamnos, sp. 
 fEquus occidentalis Leidy. 
 fElephas sp. 
 
 Scapanus, possibly n. sp. 
 Ursus sp. 
 
 JAenocyon, near dims (Leidy). 
 Canis ochropus Eschscholtz. 
 Procyon psora Gray. 
 
 Mephitis occidentalis Baird. 
 JSmilodon (?) sp. 
 fFelis hawveri Stock. 
 fFelis, probably n. sp. 
 Peromyscus boylei (Baird). 
 Neotoma fuscipes Baird. 
 Microtus sp. 
 
 HAWVER CAVE. 
 
 Thomomys sp. 
 
 Aplodontia (?) sp. 
 
 Citellus beecheyi (Richardson). 
 fLepus, probably n. sp. 
 
 JMylodon harlani Owen. 
 
 JMegalonyx (?) sp. 
 
 JNothrotherium shastense hawveri Stock. 
 Odocoileus sp. 
 
 JEuceratherium collinum (?) Sinclair and Furlong. 
 tBison sp. 
 fEquus sp. 
 
 JMammut sp. 
 
 J Extinct genus. f Extinct species. * Extinct subspecies. 
 
 a W. J. Sinclair. The exploration of Potter Creek Cave. Univ. Calif. Publ., Dept. Amer. Arch, and Ethnol., vol. 2 
 
 pp. 1-27, plates 1—14, 1904. 
 
 6 E. L. Furlong. The exploration of Samwel Cave. Amer. Jour. Sci, ser. 4, vol. 22, pp. 235-247, 1906. 
 c C. Stock. The Pleistocene fauna of Hawver Cave. Univ. Calif. Publ., Bull. Dept. Geol., vol. 10, pp. 461-515, 1918. 
 
114 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 We note the presence of the genus Nothrotherturn in all three Pleistocene cave faunas. 
 The type locality of the species N. shastense is Potter Creek Cave. Megalonyx is also 
 present at Potter Creek Cave and at Samwel Cave and possibly at Hawver Cave. 
 Mylodon is found only at the latter locality. The megalonychid ground-sloths are 
 therefore characteristic of the cave faunas of California and are regarded as types 
 commonly occurring in the foot-hill and more elevated regions of California during 
 the Pleistocene. Associated with the ground-sloths in this environment, and perhaps 
 even more characteristic of it, are the ungulates Euceratherium and Preptoceras. 
 
 Fra. 53 .—Nothrotherium shastense Sinclair. Right metacarpal IV, No. 9443, U. C. C. A, inner view; 
 
 B, dorsal view; C, outer view; D, proximal end; E, distal end. X 0.50. Samwel Cave 
 Pleistocene, California. 
 
 Hapalops sp. Left metacarpal IV, No. 15171. P. U. C. A 1 , B l , C 1 , D l , E l , same views as 
 in A, B, C, D, E. X 1.0. Santa Cmz Miocene. 
 
 The mammalian assemblages from the cave deposits are probably not all of the 
 same stage or division of the Pleistocene, but are to be distributed over a considerable 
 time period. One of the important problems in the palaeontology of western North 
 America awaiting solution is the succession of Pleistocene vertebrate life. It is safe 
 to say that the large assemblages known from the California cave deposits form an 
 important link in the chain of evidence partly constructed, and detailed comparisons 
 between the cave forms and those occurring at Rancho La Brea should cast light on 
 the age enigma of the southern asphalt beds. 
 
 The materials from the cave deposits consist mainly of fragments of skulls and 
 skeletons and separate teeth. The type specimen of Nothrotherium shastense from Potter 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 115 
 
 Creek Cave is an incomplete right ramus of the mandible (No. 8422 U. C. C.) (Sinclair, 
 1905, pp. 153-155, plate 23). A number of teeth of this species are also available. In 
 addition, teeth have been found that apparently represent several species of Megalonyx. 
 
 Teeth, a well-preserved scaphoid (No. 9437 U. C. C.), a fourth metacarpal (No. 
 9443 U. C. C.), and a navicular (No. 9371 U. C. C.) comprise the best material of 
 Nothrotherium from Samwel Cave. 
 
 The fourth metacarpal (No. 9443, fig. 53, a to e) is similar in shape and in slender¬ 
 ness to the corresponding element in the Brazilian species of iV othr other ium shown by 
 Winge (1915, plate 24). It is relatively more slender than that of Megalonyx. No, 
 9443 evidently belongs to an individual of rather advanced age. 
 
 In proportion to length of shaft, the proximal and distal ends are much deeper in 
 Megalotiyx jeffersonii than in No. 9443. The Samwel Cave specimen in this respect is 
 more like the corresponding metacarpal of Hapalops (No. 15171, Prin. Univ. Coll.). 
 
 Fig. 54 .—N othr other ium shastense hawveri Stock. Left unciform, No. 22899 U. C. C. 
 A, proximal view; B, dorsal view; C, ventral view; D, distal view. X 0.50. Hawver 
 Cave Pleistocene, California. 
 
 The dorsal surface of the metacarpal near the proximal end is nearly flat and the 
 carpal border rounds toward the inner side. In carpal view (fig. 53 d) the superior 
 border of the proximal end in Nothrotherium is seen to be straight. In Megalonyx jeffer¬ 
 sonii this border is somewhat convex. The palmar border in Megalonyx is drawn down¬ 
 ward more than in Nothrotherium. The articulating surface for the unciform is relatively 
 wider than in Megalonyx. It is continuous with the facet for the median metacarpal 
 along the lower half of the carpal surface. Continuous with the unciform surface, but 
 separated from it by a sharp angle, is the small, reniform facet for metacarpal V (fig. 
 53 c). This facet in the Samwel Cave specimen is more lateral in position than in M. 
 jeffersonii. In Nothrotherium the surface for the unciform is sigmoid in dorso-palmar 
 direction, but the articulating face for metacarpal III is flattened or somewhat concave. 
 Both surfaces in the Miocene form are sigmoid in dorso-palmar direction. The facet 
 for the median metacarpal in Hapalops is situated lower on metacarpal IV, so that the 
 proximal end slopes as a whole obliquely inward from the outer side. 
 
 The shaft of the metacarpal is subquadrate, palmar and outer surfaces meeting in 
 a sharp angle. The dorsal surface forms a sharp median ridge along approximately the 
 distal two-thirds of the shaft, No. 9443 thus agreeing with metacarpal IV in the Brazil¬ 
 ian species and differing from that in Hapalops and from that in Megalonyx. The distal 
 articulation (fig. 53 e) is very close to that in Hapalops and in Megalonyx. The rather 
 heavy carina is of nearly same thickness throughout. Along its entire inner side is an 
 offset that widens below. The carina and the inner offset are deflected below for articu¬ 
 lation with the sesamoid. The former is here transformed into a rather sharp ridge. 
 
 The measurements (in millimeters) of No. 9443 are: Greatest length, 129.8; greatest 
 width of proximal end, 33; greatest depth of proximal end, 36; greatest width of distal 
 end, 33; greatest depth of distal end, 51. 
 
 Some skull material, several teeth, and many parts of the skeleton of Nothrotherium 
 are known from the Hawver Cave deposit (Stock, C., 1918, pp. 492-513, figs. 11 to 32c). 
 
116 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 In the anterior extremity two elements are known that have not been found at Rancho 
 La Brea. 
 
 An unciform (No. 22899) is nearly complete, lacking only the dorso-external portion, 
 which has been broken away. This has destroyed the greater part of the facet for meta¬ 
 carpal V as well as some of the cuneiform surface, but the remaining facets are well 
 preserved. 
 
 When viewed from the proximal end (fig. 54, a to d), the unciform of Nothrotherium 
 approaches that of Hapalops rather more closely than does the unciform of Megalonyx 
 in the expanse of bone at the ventro-external angle. The cuneiform surface does not 
 form internally such a distinct boss or knob as in Megalonyx. This is due in part to the 
 fact that the articulating surface slopes away more gradually from the top of the knob 
 to the dorsal margin of the unciform. In Megalonyx a shelf is formed at the base of the 
 knob and the end of this shelf is the dorsal margin. The proximal knob differs further 
 from Megalonyx and Hapalops in being curiously flattened along the inner side. On 
 the inner side the facet for the magnum resembles that in Megalonyx and is relatively 
 less extensive than that in Hapalops. Dorsally a small facet for the magnum, contin¬ 
 uous distally with the facet for metacarpal III, is not connected with the proximal 
 articulating surface for the magnum as in the unciform (No. 22775 U. C. C.) of Mega¬ 
 lonyx from Rancho La Brea. The facet for metacarpal III, is wide dorsally and narrows 
 decidedly to the ventral side, thus resembling the corresponding feature in Megalonyx 
 jeffersonii as figured by Leidy (1855, plate 8, fig. 136) and differing from that in Mega¬ 
 lonyx from Rancho La Brea. The widest portion of the facet differs from that in the 
 latter genus in being deeply excavated, especially in an oblique line extending from the 
 inner to the dorsal margin. In the unciform of Hapalops, the oblique groove traversing 
 the facet dorsally is very much deeper than in Nothrotherium and notches the distal border 
 of the carpal element. This deep groove is to be associated with a more extensive over¬ 
 lapping of metacarpal III upon the fourth metacarpal. In Hapalops the unciform joins 
 with the external and internal sides of the overlapping processes, while in Nothrotherium 
 articulation is with the external or true proximal face of the metacarpal. 
 
 In Nothrotherium the facet for metacarpal IV differs from the corresponding surface 
 in Hapalops in its much smaller size, but resembles the latter in being deeply excavated. 
 The facet in Nothrotherium is comparable in size to that in M. jeffersonii, but is decidedly 
 smaller than in No. 22775 U. C. C., from Rancho La Brea. A small articulating facet, 
 also for metacarpal IV, truncates the proximal portion of the boundary between facets 
 for the third and fourth metacarpals. 
 
 The dorsal surface of the unciform (fig. 54 b) differs from that in Megalonyx in 
 being wider proximo-distally along the inner side, and the dorso-internal corner is not 
 as prominent. 
 
 A distal portion of a metapodial (No. 19861 U. C. C.) appears to belong to meta¬ 
 carpal III. In the specimen the shaft at the middle is nearly square in section, and is 
 more slender than that of the corresponding metapodial of Megalonyx or of Hapalops. 
 Its distal end is also not so large as that in the latter genus. The carina is nearly straight, 
 thus differing somewhat from that in either the Miocene form or in Megalonyx. Articu¬ 
 lating offsets are present along the lower half of each side of the distal carina. 
 
 The measurements (in millimeters) of metapodial, No. 19861 U. C. C., are: Greatest 
 depth of distal end, 45.8; width of distal end, 31. 
 
 Calaveras County. —Mercer’s Cave, near the town of Murphys, is located in the mid- 
 Sierran region of California. Although of only minor faunal importance, the cavern is 
 of interest as the type locality of the ground-sloth Megalonyx sierrensis, described by 
 Sinclair (1905, pp. 155-160, plate 20, figs. 5 to 8; plate 21, figs. 1 and 2; plate 22, figs. 
 1 to 3). 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 117 
 
 The species is based on a lower jaw (No. 14986) in the Harvard Museum of Com¬ 
 parative Zoology, Cambridge, Massachusetts, and skeletal elements (No. 8130) in the 
 collections of the University of California. 
 
 Middle California. —The genus Morotherium was established by Marsh (1874) on a 
 fragmentary humerus and a nearly complete femur, said to have come from Pliocene 
 strata (Pleistocene) of middle California. The type description follows: 
 
 Morotherium gigas, gen. et sp. nov. 
 
 The present genus appears to be most nearly related to Megalonyx and Mylodon. From the former, as 
 well as from Scelidotherium, it may readily be distinguished by the humerus, which has no supra-condylar 
 foramen; while from the latter genus it differs in the femur, which is without a depression for the round ligament. 
 The more slender femur and the concave ulnar articulation of the humerus separate it likewise from Megathe¬ 
 rium. The skull of the present genus is not known. 
 
 In this species the femur is stout, and in its general proportions resembles that of Megalonyx jeffersonii 
 Harlan. The head of the femur is hemispherical, but slightly expanded in an antero-posterior direction. The 
 great trochanter is massive and rugose, and is raised slightly above the neck. It is somewhat recurved, and 
 encloses a deep pit, as in the elephant. The third trochanter is represented only by a rugosity, mainly on the 
 lower half of the shaft. On the posterior surface there is a prominent ridge extending from the great trochanter 
 obliquely across the shaft toward the inner condyle. The trochlear surface for the patella is broad and shallow, 
 and is separated by a deep groove from each of the condyles, the inner groove being about twice the width 
 of the other. The humerus is expanded at its distal end. The olecranon cavity has little depth, and there is 
 no perforation above the inner condyle. The latter has its articular face concave transversely, as in Mylodon. 
 The outer and inner deltoid ridges unite below, terminating in a double tuberosity. 
 
 Measurements (in millimeters). 
 
 Length of femur. 450 
 
 Transverse diameter at proximal end. 217 
 
 Antero-posterior diameter of head of femur. 106 
 
 Transverse diameter. 98 
 
 Least transverse diameter of shaft. 130 
 
 Least antero-posterior diameter. 55 
 
 Width of trochlear surface. 82 
 
 Vertical extent! . 50 
 
 Transverse diameter of outer condyle. 78 
 
 Least transverse diameter of shaft of humerus. 93 
 
 Antero-posterior diameter of ulnar condyle of humerus.. 70 
 
 Transverse diameter. 63 
 
 In 1899, J. C. Merriam described a humerus from Pleistocene strata near Tomales 
 Bay, California. It resembled closely the known portion of Marsh’s specimen and was 
 referred to Morotherium , although the validity of the genus was questioned by that 
 writer. In 1914, Stock concluded that the humerus described by Merriam belongs to 
 Mylodon. It is inferred, therefore, that Marsh’s fragmentary specimen belongs also to 
 this genus. 
 
 The femur of Morotherium was later stated by Marsh (1897) to be the type of the 
 genus. In 1919 the writer examined this specimen (No. 11898) in the collections of the 
 Peabody Museum, Yale University, and was impressed by its resemblance to the thigh 
 bone of Nothrotherium from Rancho La Brea. Unfortunately, the only known femur, No. 
 1871-L-l, of Nothrotherium from the asphalt was not then available for direct compari¬ 
 son. Later a cast of the latter specimen was sent to Professor R. S. Lull, who, with his 
 assistants, kindly made comparisons with Marsh’s type. To quote from Professor Lull 
 (letter dated June 28, 1922): 
 
 “We have reached the conclusion that generic distinctions are not apparent, and that such differences 
 as exist may be specific, and it is perhaps safer to assume that they are. On the other hand, they may be due 
 to age or sex. 
 
 “The differences are that the Marsh specimen is slightly larger, and has strong rugosities, some of which 
 do not appear at all on your femur. In the latter, the head of the femur is not sharply separated from the neck, 
 while in Morotherium it is larger and overhangs. The exterior distal face is larger, has greater curvature, and 
 extends around the end as though to straighten the leg for greater weight in Morotherium, and the surface 
 for the patella is wider transversely and not so curved. The loss of the inner condyle of the Marsh type leaves 
 a certain element of doubt as to whether it w^as the same in both specimens.” 
 
 As indicated by Professor Lull, Marsh’s type is larger than the femur of Nothrothe- 
 rium from Rancho La Brea. Judging from this character, Marsh’s type might belong to 
 
118 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 a species of Megalonyx smaller than M. jeffersonii. Small species of this genus are 
 known to occur in the Pleistocene of western North America. There is reason for believ¬ 
 ing, however, that No. 1871-L-l belonged to a younger individual than that repre¬ 
 sented by No. 11898. This becomes evident when the original femur from Rancho La 
 Brea is examined, but can not be seen in the cast. In determining the relationship of 
 Nothrotherium to Morotherium the following should be emphasized: (1) the type of Moro- 
 therium is a fragmentary specimen, (2) the geologic occurrence of No. 11898 is not 
 definitely known, (3) certain structural characters in No. 11898, as pointed out by Lull, 
 distinguish this specimen from the Rancho La Brea femur, (4) the femur of Nothrotherium 
 shares in the general resemblance that exists between the skeletal elements of this genus 
 and those of Megalonyx , and (5) only the femur of the large species Megalonyx jeffersonii 
 is known; the comparable element in the smaller species of Megalonyx being unknown. 
 
 In view of the above the writer hesitates to assert definitely that Marsh’s type, 
 described under the name Morotherium gigas, represents the genus known as Nothrothe¬ 
 rium and which occurs at Rancho La Brea. It seems more desirable to await further 
 information concerning the femur of a fully matured individual of Nothrotherium and 
 of femora belonging to the small species of Megalonyx known to occur in the Pleistocene of 
 North America. 
 
 The name Nothrotherium was proposed by Richard Lydekker" to replace Coelodon 
 Lund, a name preoccupied for a genus of long-horned, wood-boring beetles (Fam. Cer- 
 ambycidae) from Senegal. * 6 If the type specimen of Morotherium gigas is ultimately 
 established as generically the same as Nothrotherium, then Marsh’s name Morotherium 
 has precedence over Nothrotherium Lydekker. Dr. T. S. Palmar (in a letter dated 
 August 12, 1922) and Professor R. S. Lull concur with the writer in this conclusion. 
 
 San Pedro, Los Angeles County, California. —The San Pedro Pleistocene deposits 
 are divisible, according to Ralph Arnold/ into an upper and a lower series, the division 
 being based on the presence of an unconformity as well as differences in lithology, faunas, 
 and inferred climatic conditions. Both the lower and upper San Pedro beds are very 
 fossiliferous, the horizons yielding splendid marine invertebrate faunas. The Upper San 
 Pedro deposits consist of sands and gravels and are presumably indicative of strand-line 
 conditions. The beds have been involved in earth movements of the late Pleistocene 
 or Recent, and are generally regarded as not representing the latest stage of the Pleis¬ 
 tocene as recorded in this region of southern California. Fragmentary vertebrate re¬ 
 mains occur in the Upper San Pedro in association with marine invertebrates. Interest 
 attaches to the vertebrate collections because of light which may be thrown by them 
 on the relationship of the upper San Pedro Pleistocene to the period of the Rancho La 
 Brea accumulation. The following is a provisional list rf of mammals comprising the 
 Upper San Pedro fauna: 
 
 Canid (Aenocyon) (?) sp. 
 
 Felis, sp. 
 
 Sylvilagus sp. 
 
 Neotoma sp. 
 
 Citellus cf. beecheyi (Richardson). 
 
 Microtus, cf. californicus (Peale). 
 Thomomys bottae (Eydoux and 
 Gervais). 
 
 Megalonyx sp. 
 
 Megalonychid (?) indet. 
 
 Equus, near occidentalis Leidy. 
 Odocoileus (?) sp. 
 
 Capromeryx, near minor Taylor. 
 Camelops hesternus (?) (Leidy). 
 Bison sp. 
 
 The ground-sloth remains consist of two specimens representing second phalanges. 
 A specimen in the collections of the Los Angeles Museum exhibits the transversely 
 compressed trochlea at the distal end, with deep median groove characteristic of Mega¬ 
 lonyx. This phalanx is larger than second phalanges of Megalonyx known from Rancho 
 
 “ Nicholson and Lydekker. Manual of palaeontology, ed. 3, vol. 2, p. 1299, 1889. 
 
 6 Audinet-Serville. Ann. Soc. Entom. France, 1832, p. 164. 
 
 c Ralph Arnold. The palaeontology and stratigraphy of the marine Pliocene and Pleistocene of San Pedro, Cali¬ 
 fornia. Mem. Calif. Acad. Sci., vol. hi, 420 pp., 37 plates, 1903. 
 
 d Determination of the Pleistocene mammalian remains from San Pedro has been made with the assistance of Mr. 
 Harold F. Blum. Material collected in Upper San Pedro beds exposed near yards of San Pedro Lumber Company. 
 
PLEISTOCENE MEGALONYCIIINAE AND MYLODONT1DAE OF RANCHO LA BREA. 119 
 
 La Brea. Its measurements (in millimeters) are as follows: Greatest length through 
 middle of shaft 69.5; depth of proximal end 41.5; width of distal end 45.3. 
 
 The second specimen (No. 19720, U. C. C.) is long, narrow, and much less robust 
 than second phalanges of Mylodon or Megalonyx. The shaft in No. 19720 is particu¬ 
 larly long, while the distal articulation is small. The median groove of the distal trochlea 
 is broad and rather shallow, differing in this respect from second phalanges of Mega¬ 
 lonyx from Rancho La Brea and from Pleistocene cave deposits of California. 
 
 The specimen from the Upper San Pedro resembles somewhat the second phalanx, 
 digit 2, manus of N othrotherium, in being very long and slender. No. 19720 is, however, 
 much larger than the phalanx in Nothrotherium. In Megalonyx jeffersonii, according to 
 the figures given (Leidy, 1855, plate 10), phalanx II, digit II, manus, is stouter than 
 the corresponding element in Nothrotherium. 
 
 The measurements (in millimeters) of No. 19729 U. C. C. are: Greatest length through 
 middle of shaft, 73; depth of proximal end, 34.6; width of distal end, 23.2. 
 
 The horse, camel, and antelope remains from the Upper San Pedro may be com¬ 
 parable to those from Rancho La Brea and may indicate contemporaneity of the two 
 faunas. While mylodont ground-sloths may be present in the Upper San Pedro mam¬ 
 malian fauna, the peculiarity remains that in the small and fragmentary collection from 
 this horizon megalonychid ground-sloths are recognized. At Rancho La Brea, a locality 
 not far removed from San Pedro, where Pleistocene mammalian remains are abundant 
 and are well preserved, members of the Megalonychinae, and particularly individuals 
 of Megalonyx, form a very small percentage of the total fauna. It is possible that 
 phalanx No. 19720, U. C. C., may belong to a ground-sloth species or genus not here¬ 
 tofore recognized in Pleistocene deposits of the West. The presence of this specimen 
 may be regarded as suggesting a faunal difference between the Upper San Pedro and 
 Rancho La Brea. 
 
 Bautista Creek Badlands, Riverside County. —The Bautista Pleistocene deposits are 
 typically exposed in the badlands along Bautista Creek, a tributary of the San Jacinto 
 River, several miles southeast of Hemet, in southern California. The beds are stated 
 by Childs Frick" to consist of fine and coarse sands and black clays that yield the follow¬ 
 ing fauna: 
 
 Megalonyx (?) sp. Antilocapra (?) one or more sp. 
 
 Lepus sp. Equus bautistensis Frick. 
 
 Odocoileus (?), two or more sp. Tapirus merriami Frick. 
 
 Capromeryx(?) sp. 
 
 An interesting member of this assemblage of mammals is the tapir. The horse, 
 E. bautistensis, according to Frick, possesses apparently a tooth-pattern more specialized 
 than that of E. occidentalis. Unfortunately the ground-sloth remains consist of a single 
 specimen, a second phalanx belonging presumably to a megalonychid. 
 
 a C. Frick. Extinct vertebrate faunas of the badlands of Bautista Creek and San Timoteo Canon, southern California. 
 Univ. Calif. Publ., Bull. Dept. Geol., vol. 12, pp. 277-424, plates 43-50, 1921. 
 
MYLODONTiDAE OF RANCHO LA BREA. 
 
 CHARACTERS OF FAMILY MYLODONTIDAE. 
 
 Hairy ground-sloths with dermal ossicles. Skull elongate with pre-orbital reigon 
 short or long. Premaxillaries reduced, sometimes joined with nasals. Dentition 
 
 Teeth round, oval, lozenge-shaped, or lobate. First tooth when present is canini- 
 form and is not separated from cheek-teeth by long diastema. Last lower tooth bilo- 
 bate, sometimes with median or third lobe present. Humerus with or without entepi- 
 condylar foramen; femur without third trochanter; manus with five digits, but digits 
 I, IV, and V may be rudimentary; hallux absent and digits II, IV, and V may be rudi¬ 
 mentary in the pes. Claw-processes of ungual phalanges broad and with convex dorsal 
 surfaces. 
 
 Table 60 .—Measurements (■in millimeters) of paired premaxillaries of Mylondon harlani. 
 
 
 No. 1711-3 
 
 No. 1711-1 
 
 No. 1711-2 
 
 No. 1711-4 
 
 Greatest transverse width. 
 
 126.5 
 
 109.8 
 
 J 120 
 
 Ll4 
 
 Greatest length of median arm. 
 
 Greatest length of diverging arm measured from antero-internal 
 
 93.8 
 
 87.8 
 
 83.6 
 
 89 
 
 angle to end.. 
 
 92 
 
 76.8 
 
 80.4 
 
 84.3 
 
 1 Approximate. 
 
 CHARACTERS OF GENUS MYLODON. 
 
 Skull elongate but with short pre-orbital region, truncated anteriorly; dorsal surface 
 of cranium flattened; premaxillaries small, loosely joined with maxillaries; malar verti¬ 
 cally expanded and palmate; pterygoid plate-like; cranial elements around brain-chamber 
 with many sinuses. Mandible deep. Teeth of the mylodont type. Lumbar vertebrae 
 ankylosed with sacral vertebrae; haemapophyses are Y-shaped; humerus without entepi- 
 condylar foramen; manus with digits IV and V rudimentary. Pelvis with greatly 
 expanded ilia; pubic symphysis short. Femur long and broad and without third tro¬ 
 chanter; foreleg much shortened; pes with hallux completely reduced and with digits 
 IV and V rudimentary. Calcaneum and metatarsal V enlarged. Largest ungual pha¬ 
 lanx in digit III, manus; second largest in digit II, manus. 
 
 Genus MYLODON Owen. 
 
 Megalonyx. R. Harlan, Medical and physical researches, p. 334, 1835. 
 
 Mylodon. R. Owen, The zoology of the voyage of H. M. S. Beagle, etc. Part 1, Fossil Mammalia, p. 68, 1840. 
 
 Mylodon. R. Owen, Description of the skeleton of an extinct gigantic sloth, Mylodon robustus, Owen, etc., p. 169, 1842. 
 
 Orycterotherium. R. Harlan, Amer. Jour. Sci. and Arts, ser. 1, vol. 44, p. 69, 1843. 
 
 Paramylodon. B. Brown, Bull. Amer. Mus. Nat. Hist., vol. 19, p. 569, 1903. 
 
 Mylodon. C. Stock, Science, n. s., vol. 39, pp. 761-763, 1914. 
 
 Mylodon. R. S. Lull, Amer. Jour. Sci. and Arts, ser. 4, vol. 39, p. 380, 1915. 
 
 Species Mylodon harlani Owen. 
 
 Megalonyx laqueatus Harlan. Medical and physical researches, p. 334, 1835. 
 
 Mylodon harlani Owen. The zoology of the voyage of H. M. S. Beagle, etc., part 1, Fossil mammalia, p. 68, 1840. 
 Orycterotherium missouriense Harlan. Amer. Jour. Sci. and Arts, ser. 1, vol. 44, p. 69, 1843. 
 
 Orycterotherium oregonensis Perkins. Amer. Jour. Sci. and Arts, ser. 1, vol. 44, p. 80, footnote. 
 
 Mylodon sodalis Cope. Bull. U. S. Geol. and Geog. Surv. Terr., vol. 4, p. 385, 1878. 
 
 Mylodon renidens Cope. Proc. Amer. Philos. Soc., vol. 34, p. 460, 1895. 
 
 Mylodon sulcidens Cope. Proc. Amer. Philos. Soc., vol. 34, p. 463, 1895. 
 
 Paramylodon nebrascensis Brown. Bull. Amer. Mus. Nat. Hist., vol. 29, p. 569, 1903. 
 
 DESCRIPTION OF MATERIAL. 
 
 DERMAL OSSICLES. 
 
 In the excavation of skeletal remains of Mylodon harlani at Rancho La Brea, numer¬ 
 ous dermal ossicles (plate 21) were encountered, lying occasionally in a definite layer of 
 asphalt that adhered to the larger bones. Professor J. C. Merriam (1906) directed 
 attention to this occurrence, as follows: 
 
 “During the first examination of the beds several small, pebble-like bones were obtained which resembled 
 the dermal ossicles of the ground-sloth, Grypotherium, recently described by Dr. A. Smith Woodward from skin 
 
 120 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 121 
 
 fragments obtained in a cave at Last Hope Inlet, Patagonia. The ossicles were in association with remains of 
 a large ground-sloth somewhat similar to Mylodon in foot structure. Realizing that the peculiar conditions of 
 accumulation offered an especially favorable opportunity for preservation of the dermal armor of a ground- 
 sloth, during the second study of the deposits an attempt was made to find a specimen in which the armor might 
 be recognized. Several hundred yards from the location of the first specimen, a large scapula resembling that 
 of a mylodont was found partly exposed, with a row of small ossicles immediately over the outer side. The 
 section of the bed containing these bones has recently been worked out, and the row of small bones proves to be 
 the edge of a distinct layer including between 250 and 300 individuals. They mantle over the outer surface 
 of the scapula, being removed from it by about an inch of asphalt. 
 
 “The layer of bones as we find it has probably been disturbed somewhat and does not occupy its original 
 position exactly, but the fact that it remains as a distinct layer with a tendency toward similar orientation of the 
 individual ossicles indicates that the disturbance has not been great. As the position of the layer in the asphalt 
 was nearly vertical, the presence of the large number of ossicles together may not be attributed to the washing 
 together of scattered elements on the floor of a small basin of deposition. 
 
 “The ossicles are not closely pressed together and are not superimposed. The individuals range in size 
 from a cross-section of 6.5 x 4.5 mm. to 21 x 16 mm. Excepting a few of the largest ones, which are nearly 
 square, the greater number are rounded and rather irregular in form. The outer side is in some cases more 
 regularly modeled than the inner. The surface of the bones is somewhat roughened or pitted in some instances, 
 but no markings are present which would be considered as definite sculpturing. The microscopic structure has 
 not yet been examined. 
 
 “In general the form, size, and arrangement of the ossicles are much as in the bones in the Grypotherium 
 skin from Patagonia. The skin fragment first described by Woodward was thought to represent mainly the 
 region of the neck and shoulder. The Californian specimen mantles over the outer side of the scapula, and is 
 presumably not far removed from its original position with relation to this bone.” 
 
 In a statement concerning the ossicles from the asphalt, Dr. W. J. Sinclair (1910), 
 after quoting the above from Professor Merriam, continues: 
 
 “Subsequent investigation has added little to this, so far as the localization of particular types of ossicles 
 is concerned. Some idea of the diversity of forms assumed by them may be gathered from the accompanying 
 figure, in which several of the types, mentioned in the citation may be recognized (especially in fig. 1 , b, c, d, g, 
 and h), nor do these differ essentially from the great mass of scattered and unlocated ossicles. Pitted and 
 smooth forms occur in the same individual. In some of the bones, except for minor undulations and the pores 
 for the entrance of blood-vessels, the outer surfaces are smooth and polished (fig. 1, e, /). Others show a highly 
 irregular pattern of small anastomosing ridges (fig. 1 , b), but, as previously noted, there is no constancy or 
 regularity in the pattern. Grooves cut across some of the ossicles, as shown in the figure. Some of these may 
 be due to the fusion of two adjacent elements, as suggested by Woodward for the origin of a similar structure 
 in Grypotherium.” 
 
 “Thin sections of the ossicles of Paramylodon were submitted 
 to Professor E. G. Conklin, who has kindly examined them and 
 furnished the following note: 
 
 “ ‘Histological character of dermal bone, horizontal and vertical 
 sections. —Penetrated by many canals for blood-vessels, which, in 
 general, begin perpendicular to the surface, but soon branch 
 repeatedly and frequently anastomose. Occasionally Haversian 
 systems of lamellae and lacunae may be seen around these canals, 
 but generally these lamellae and lacunae are very irregularly 
 disposed between the canals. One of the most striking features is the presence of bundles of fibres which run in 
 all directions through the bone, making the latter look almost like a woven fabric. These fibres are most 
 abundant in the spaces between Haversian systems.’ ” 
 
 Fig. 55 .—Mylodon har- 
 lani Owen. Section 
 through middle of a 
 dermal ossicle show¬ 
 ing entrance of a 
 vascular canal and 
 inner honeycombed 
 cavity. X 2.0. 
 
 CRANIUM. 
 
 The greater number of skulls of Mylodon harlani from Rancho La Brea range approxi¬ 
 mately between 460 and 525 mm. in length (from the anterior end of the maxillary to 
 the posterior end of the occipital condyles). None of the larger skulls reaches quite the 
 length of the Colorado specimen described by Cockerell (1909). In several specimens 
 the region behind the postorbital process is narrowed transversely, and an approach is 
 made to the type of skull seen in Mylodon garmani. This compression above the brain- 
 case is, however, never so great in the material from the asphalt as in M. garmani. 
 Contrasted with the skull of Mylodon robustus, that of M. harlani averages larger and 
 is relatively more slender. 
 
122 
 
 CENOZOIC GRAYIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 Viewed from above (plate 23), the skull is seen to be essentially similar to the speci¬ 
 men from Nebraska described by Brown (1903) and to that figured by Cockerell. The 
 muzzle is somewhat inflated at the middle and narrows to a greater or less extent ante¬ 
 riorly. In M. robustus the muzzle widens from the region immediately in front of the 
 malar process to the anterior end of the skull. The nasals are more or less constricted 
 at the middle. Normally the fronto-nasal suture is V-shaped. 
 
 Fig. 56 .—Mylodon harlani Owen. Skull, dorsal view. X 0.33. n, nasal; mx, maxillary; l, lachrymal ;/r, frontal; sq, squamosal; 
 
 pa, parietal; so, supraoccipital. Rancho La Brea Pleistocene. 
 
 Fig. 57 .—Mylodon harlani Owen. Skull, ventral view. X 0.33. mx, maxillary; pi, palatine; pi, pterygoid; sq, squamosal; 
 per, periotic; bo, basioccipital; bs, basisphenoid; flm, foramen lacerum medium; ftp, foramen lacerum posterius; fc, 
 condylar foramen: 2, second superior tooth; 5, fifth superior tooth. Rancho La Brea Pleistocene. 
 
 The postorbital processes are not conspicuous. The supraorbital region does not 
 widen backward so much as in M. robustus, and in some skulls it may narrow posteriorly. 
 The flattened dorsal surface of the frontal and parietal, outlined laterally by the margins 
 of the temporal muscles, varies in width, irrespective of the degree of slenderness of the 
 skull. In slender skulls, for example, the width of the surface may be equal to or greater 
 than that in some of the broader types. In skull 1717-19 the dorsal surface is very 
 wide, thus resembling closely that in M. robustus as figured by Owen (1842, plate 3). 
 In young individuals the lateral borders of the dorsal surface are not so distinctly 
 defined. 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 123 
 
 The premaxillary (plate 23, fig. 2) is not firmly united with the maxillary in most 
 skulls from the asphalt. A number of separate elements have been recovered from the 
 deposit. They reflect in their size the broad type of skull possessed by Mylodon harlani, 
 and resemble in their shape the corresponding bones of M. robustus. When paired they 
 have a curious dart-like appearance and resemble the predentary structures in the dino- 
 saurian genus Triceratops. The median arm is deep dorso-ventrally, and in transverse 
 width tapers gradually to the posterior end. The lateral or diverging arm is compara¬ 
 tively thin, but is broad transversely. It is bent into undulating curves, and the poste¬ 
 rior end, which may fuse with that portion of the maxillary lying in front of the tooth 
 row, is often thickened. The anterior border of the pair forms a broad convex curve. 
 In the skull of Scelidotherium magnum (Winge, plate 36, fig. 1), which is much narrower 
 than that of Mylodon, the premaxillaries form anteriorly an acute angle and the arms 
 are slender. Even in the narrow skull of M. garmani the premaxillaries still present 
 an anterior convex border. At the posterior angle formed by the junction of the two 
 arms a small canal is partially inclosed by bone. On the dorsal side a ridge may be 
 developed anteriorly which swings from the posterior angle in a convex curve to the 
 inner border, reaching forward about one-half the distance between the posterior angle 
 and the anterior end. Here it meets a similar ridge developed on the opposite element. 
 
 tal; If, lachrymal foramen; fla, fo, combined optic foramen and foramen lacerum anterius; fro, combined 
 foramen rotundum and foramen ovale; remaining letters as in other figures. Rancho La Brea Pleistocene. 
 
 The combined width of the premaxillaries of Mylodon harlani is always greater than 
 that in M. garmani, and the median arm is also distinctly longer in the Rancho La Brea 
 species. 
 
 The average measurements (in millimeters) of 20 unpaired elements are as follows: 
 Transverse width from medial face to end of diverging arm, 55.3; greatest length of 
 median arm, 84.3; greatest length of diverging arm measured from antero-internal angle 
 to end, 74.3. 
 
 The palate (plate 23, fig. 1) is noticeably convex antero-posteriorly opposite the 
 third and fourth superior teeth. Anteriorly the palate may be either slightly or deeply 
 concave along the median line. Between the last superior teeth and posterior to them 
 the surface of the palatine is concave and slopes to the posterior nasal opening. The 
 surface of the palate is perforated by a number of small foramina. The palatine posterior 
 to the fifth superior teeth is very narrow transversely in the slender type of skull from 
 Rancho La Brea. This character apparently does not depend on the slenderness of the 
 cranium in the forms represented in the asphalt deposits. It should be noted also that 
 in M. garmani the least width across the palatal portion of the palatine is actually greater 
 
124 
 
 CENOZOIC GRAVIC4RADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 than in a broad skull of M. harlani (No. 21158 U. C. C.) of approximately the same 
 length as that of the former. 
 
 The pterygoids are plate-like structures which may diverge slightly posteriorly. 
 The inferior border is rounded. The base of the pterygoid at the inner posterior end 
 forms a tuberosity. Between this and the wall of the pterygoid is the groove for the 
 Eustachian canal, which widens in its forward extent. Between the pterygoid plates 
 and on each side of the basicranial region may be seen frequently the two open channels 
 that were traversed by the nerves making their exit through the anterior lacerate foramen 
 and the combined foramina rotundum and ovale. 
 
 The base of the pterygoid is indicated along the inner side in some specimens by 
 the suture line that extends back from the palatine and separates the element from the 
 basisphenoid. The latter is the principal element in the basicranial region between the 
 pterygoids. In front of the basisphenoid is the presphenoid, with the vomer probably 
 also represented. A narrow process of the basisphenoid extends outward and backward 
 and lies between the pterygoid and the basioccipital. Between this portion of the basi¬ 
 sphenoid with the adjacent pterygoid and the periotic, lies the carotid foramen ( flm ). 
 As seen in plate 23, the periotic and the structures immediately adjacent to it are com¬ 
 parable to those in Mylodon robustus and have been fully described by Owen (1842, p. 
 24). The tympanic is not always preserved, but when 
 present it resembles that in other mylodont ground- 
 sloths. Behind the large jugular foramen ( flp ) is the 
 foramen for the twelfth nerve (/c). The condylar 
 foramina are occasionally of unequal size in the same 
 skull. 
 
 Auditory ossicles have been obtained from several 
 of the skulls. The malleus and incus are available, 
 but the stapes has not been recognized. Unfortu¬ 
 nately, the manubrium is not preserved in the malleus 
 (plate 23, fig. 3). The remainder of the specimen 
 resembles that of the malleus of Grypotherium shown 
 by A. Smith Woodward (1900, plate 6, fig. 4, m), but 
 possesses only a single surface of union with the incus. 
 
 The latter ossicle (plate 23, fig. 4) also resembles in shape that of Grypotherium (A. S. 
 Woodward, 1900, plate 6, fig. 4, i). The diverging arms may be of nearly equal length, 
 but in one specimen the arms are quite unequal. 
 
 The suture between basisphenoid and basioccipital extends between the anterior 
 inner end of the periotic of each side. At each side of the anterior end the basioccipital 
 is tuberous. Toward the margin of the foramen magnum the surface of the element 
 flattens, and on each side of the median line at the middle it may be perforated by a 
 vascular foramen. The ventral surface of the occipital condyles has the transverse 
 diameter greater than the longitudinal, as in M. garmani, but in some skulls from the 
 asphalt the surface may be quite wide longitudinally, as in the Colorado skull described 
 by Cockerell. 
 
 In lateral view (plate 24) the dorsal surface is seen to be slightly depressed at the 
 anterior end of the frontal. In this view it becomes apparent also that the occiput 
 slopes downward at varying degrees. The lachrymal extends distinctly into the orbit 
 and its foramen is conspicuous. The malar, which is generally found separate, has 
 three forks or prongs. The upper and middle forks are usually of unequal length. The 
 lower and middle forks are separated by a wide notch, as in M. garmani. 
 
 A rather large depression in the orbit is bounded above by the frontal, in back by 
 the alisphenoid, and below by the orbital portion of the palatine. The orbitosphenoid 
 is not clearly defined by suture. The optic nerve leaves the skull presumably through 
 the anterior lacerate foramen ( fla ). Lying in front of the foramen is a rather large 
 
 Fig. 59. —Mylodon harlani Owen, Skull, pos¬ 
 terior view. X 0.33. Letters as in other 
 figures. Rancho La Brea Pleistocene. 
 
PLEISTOCENE MEGALONYCIIINAE AND MYLODONTIDAE OF RANCHO LA BREA. 125 
 
 opening through the palatine. At the anterior and lower end of the depression is situated 
 the entrance to the canal that traverses the palate. 
 
 The channel through which the optic, first branch of the trigeminal, and other 
 nerves reach the orbit exhibits further details. A small opening on the inner wall and 
 a short distance behind the exit leads into a channel that has been traced back to be¬ 
 tween the third and fourth endoturbinals. A small channel, apparently for the fourth 
 nerve, leaves the brain-chamber and extends forward a short distance and merges with 
 the large one already described. 
 
 Fig. 60 .—Mylodon harlani Owen. Vertical longitudinal section of skull, No. 1717-2. X 0.33. mxt, maxillo-turbinal; et, 
 ethmo-turbinals; s, sinus in frontal* parietal, and basisphenoid; ofV 1 , canal for first branch of trigeminal nerve and 
 possibly the optic nerve; F 2 , F 3 , canal for second and third branches of the trigeminal nerve; IX, X, XI, exit for ninth, 
 tenth, eleventh cranial nerves; XII, exit for twelfth nerve; remaining letters as in other figures. Rancho La Brea 
 Pleistocene. 
 
 The exit (fro) for the remaining branches of the trigeminal nerve is situated 
 below the front end of the base of the zygomatic process, where squamosal, alisphenoid, 
 and pterygoid meet. This exit represents the combined foramina rotundum and ovale. 
 
 Skull 1717-2 has been bisected longitudinally. In viewing the internal structures 
 (fig. 60), striking features are the relatively small size of the brain-chamber and the 
 large number of air-cavities or sinuses (s) present in the bones that surround it. Anterior 
 to the brain-chamber the vomer and mesethmoid have been removed and underlying 
 parts come to view. 
 
 The maxilloturbinals are very much complicated in structure and occupy to a con¬ 
 siderable extent the anterior portion of the nose on either side of the median septum. 
 They are covered medially by a thin plate of bone, similar to that described in the 
 
 Table 61. 
 
 
 No. of 
 endo¬ 
 turbinals. 
 
 No. of 
 olfactory 
 scrolls. 
 
 Remarks. 
 
 Mylodon harlani. 
 
 7 
 
 8 
 
 Second from top endoturbinal divided. 
 
 Choloepus didactylus. 
 
 7 
 
 7 
 
 
 Tamandua tetradactylus. 
 
 5 
 
 6 
 
 Second from top endoturbinal divided. 
 
 Myrmecophaga jubata. 
 
 5 
 
 7 
 
 Second and fifth from top endoturbinals divided. 
 
 Dasypus villosus. \ 
 
 Dasypus sexcinctus./ 
 
 6 
 
 8 
 
 Second and sixth from top endoturbinals divided. 
 
 Xenvru. duodecimcinctus. \ 
 
 Dasypus novemcinctus./ 
 
 8 
 
 9 
 
 Second endoturbinal divided. 
 
CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA 
 
 126 
 
 skull of Choloepus didactylus by Paulli,“ and which is a forward continuation of the 
 first endoturbinal of the ethmoid. 
 
 The ethmoturbinals also present a complicated form and indicate that the sensory 
 epithelium of the nose was well developed in Mylodon harlani. We are principally in- 
 
 Table 62. —Measurements {in millimeters) of skulls of Mylodon harlani. 
 
 Catal. No. 1717, L. A. M. 
 Coll. 
 
 Length from anterior margin 
 of maxillaries to posterior 
 end of occipital condyles. 
 
 Length of palate from ante¬ 
 rior end of maxillaries to 
 postpalatine notch. 
 
 Greatest width measured 
 across ventral surface an¬ 
 terior to first teeth. 
 
 Width of palate measured be¬ 
 tween middle of inner sides 
 of second superior teeth. 
 
 Width of palate measured be¬ 
 tween inner sides of anterior 
 lobes of fifth superior teeth. 
 
 Least width posterior to 
 
 fifth superior teeth. 
 
 Least distance from fifth 
 
 superior tooth to middle of 
 
 postpalatine notch. 
 
 Least distance across ventral 
 
 margins of pterygoid plates. 
 
 ' 
 
 Mastoid width above stylo- 
 
 hyal processes. 
 
 Greatest width across occi¬ 
 
 pital condyles. 
 
 Transverse diameter of fora¬ 
 
 men magnum. 
 
 Dorso-ventral diameter of 
 
 foramen magnum. 
 
 Greatest width of muzzle. 
 
 Least width behind postor¬ 
 
 bital processes. 
 
 Height measured from plane 
 
 of basioccipital to dorsal 
 plane. 
 
 1 
 
 492.8 
 
 216.8 
 
 124.8 
 
 72.8 
 
 49 
 
 66.7 
 
 53.4 
 
 
 184.9 
 
 127.8 
 
 43.5 
 
 36.9 
 
 130.9 
 
 107.5 
 
 139.5 
 
 2 
 
 497.9 
 
 217.5 
 
 151.6 
 
 79.6 
 
 49.3 
 
 68.3 
 
 50.1 
 
 120.6 
 
 191.5 
 
 131.2 
 
 40.4 
 
 39.5 
 
 147.6 
 
 114.2 
 
 123.6 
 
 3 
 
 466.2 
 
 212.5 
 
 
 74 
 
 48.5 
 
 60.4 
 
 44.7 
 
 
 188.8 
 
 133.9 
 
 47.6 
 
 40 
 
 124.1 
 
 115.6 
 
 143.5 
 
 4 
 
 493.7 
 
 221.8 
 
 131.1 
 
 66.9 
 
 54.4 
 
 71 
 
 *51 
 
 119.1 
 
 185.5 
 
 129.9 
 
 41.7 
 
 38.8 
 
 132.2 
 
 114.6 
 
 
 5 
 
 505.8 
 
 223 
 
 147 
 
 90.5 
 
 59.3 
 
 72.4 
 
 53.8 
 
 114 
 
 185.9 
 
 134.6 
 
 39.5 
 
 40 
 
 *146 
 
 105.3 
 
 138.5 
 
 6 
 
 520.6 
 
 226.3 
 
 142.4 
 
 77.6 
 
 55.3 
 
 68.5 
 
 41.3 
 
 122.4 
 
 *201.3 
 
 125.5 
 
 40 
 
 37.8 
 
 146.2 
 
 121.3 
 
 134.8 
 
 
 
 
 128.8 
 
 
 
 65.5 
 
 
 
 194.6 
 
 127.4 
 
 
 
 134.8 
 
 107.9 
 
 129.5 
 
 8 
 
 480.8 
 
 213.5 
 
 127.3 
 
 67 
 
 52 
 
 69.1 
 
 43 
 
 
 186.3 
 
 133.3 
 
 43.9 
 
 42 
 
 128.6 
 
 115.3 
 
 142.2 
 
 9 
 
 516.5 
 
 
 150.9 
 
 74.7 
 
 55.5 
 
 69.7 
 
 
 
 199.6 
 
 131.9 
 
 43.9 
 
 36.8 
 
 147.1 
 
 116.7 
 
 143.2 
 
 10 
 
 481.6 
 
 
 135 
 
 74.5 
 
 58.9 
 
 69.1 
 
 
 120.6 
 
 196.6 
 
 131.1 
 
 44 
 
 34.8 
 
 141.1 
 
 113 
 
 136.9 
 
 11 
 
 12 
 
 483.1 
 
 210.6 
 
 136.8 
 
 68.7 
 
 51.5 
 
 64.4 
 
 54 
 
 
 188.2 
 
 134.5 
 
 42.5 
 
 36 
 
 *138 
 
 112.8 
 
 135.7 
 
 13 
 
 *445.5 
 
 205.9 
 
 131.4 
 
 64.7 
 
 50.1 
 
 65.1 
 
 43.3 
 
 108.7 
 
 178.6 
 
 125.9 
 
 35.8 
 
 33.4 
 
 125.5 
 
 111.9 
 
 
 14 
 
 500.8 
 
 217 
 
 134 5 
 
 74.3 
 
 51.8 
 
 67.4 
 
 43.5 
 
 109.8 
 
 
 136.4 
 
 46.4 
 
 38.5 
 
 135.8 
 
 116.1 
 
 
 15 
 
 519 6 
 
 *233.1 
 
 *149 
 
 79.1 
 
 62.5 
 
 77.3 
 
 *47.3 
 
 
 
 
 51 
 
 36.2 
 
 147.2 
 
 
 146.3 
 
 16 
 
 491.8 
 
 219.6 
 
 141.4 
 
 70.3 
 
 49.9 
 
 69.1 
 
 45.8 
 
 105 
 
 200.3 
 
 124.7 
 
 52.9 
 
 36.8 
 
 136.9 
 
 114.6 
 
 130.6 
 
 17 
 
 462.1 
 
 203.6 
 
 118.4 
 
 75.1 
 
 50.9 
 
 73.2 
 
 50.9 
 
 
 179.6 
 
 131 
 
 45.3 
 
 38.2 
 
 127.1 
 
 112.5 
 
 142.3 
 
 18 
 
 502.9 
 
 
 128.2 
 
 69.8 
 
 55.6 
 
 71.4 
 
 
 
 196.3 
 
 
 42 
 
 36.3 
 
 131.3 
 
 111.2 
 
 129 
 
 19 
 
 492.2 
 
 
 146.2 
 
 66 
 
 46.3 
 
 63.7 
 
 
 
 
 133.2 
 
 44.6 
 
 32 9 
 
 
 119.8 
 
 137.5 
 
 20 
 
 499.4 
 
 
 
 73.6 
 
 57.4 
 
 67.7 
 
 
 
 192.8 
 
 137.5 
 
 44.4 
 
 37.1 
 
 
 120.2 
 
 148.2 
 
 21 
 
 
 176.5 
 
 hie 
 
 85.4 
 
 42.2 
 
 57.8 
 
 36.8 
 
 87 
 
 158.3 
 
 *131.5 
 
 41.7 
 
 33.6 
 
 *122 
 
 96.8 
 
 123 
 
 22 
 
 492.6 
 
 218.4 
 
 141 
 
 76.7 
 
 45.6 
 
 64.2 
 
 43.5 
 
 
 188.5 
 
 132.1 
 
 50.8 
 
 42 
 
 137.3 
 
 120.6 
 
 136.1 
 
 23 
 
 461.8 
 
 199.8 
 
 139.8 
 
 69 
 
 45.2 
 
 63.3 
 
 42.2 
 
 106.4 
 
 182.5 
 
 
 48.4 
 
 37.6 
 
 132.6 
 
 113.3 
 
 138.6 
 
 24 
 
 507.8 
 
 
 139.5 
 
 78.6 
 
 58 
 
 
 
 
 191.6 
 
 135.6 
 
 49.5 
 
 36.2 
 
 *145.5 
 
 112.3 
 
 137.9 
 
 25 
 
 
 
 
 78.2 
 
 57.9 
 
 
 
 
 
 136.9 
 
 43 
 
 40.8 
 
 *139.5 
 
 
 152.6 
 
 26 
 
 *424.5 
 
 183.3 
 
 *115.5 
 
 61.8 
 
 41.7 
 
 59.4 
 
 *43 
 
 
 162.8 
 
 *129 
 
 41.3 
 
 38.7 
 
 117.9 
 
 *94.5 
 
 119.7 
 
 27 
 
 499.7 
 
 223.9 
 
 145.5 
 
 78.8 
 
 53.7 
 
 *66.2 
 
 53.9 
 
 
 193.8 
 
 
 43.5 
 
 37.8 
 
 142.8 
 
 *108 
 
 
 28 
 
 
 
 135 
 
 65.8 
 
 42.6 
 
 60.7 
 
 
 
 
 
 
 
 135.7 
 
 106.5 
 
 
 29 
 
 
 
 132.8 
 
 77.5 
 
 51.5 
 
 
 
 
 
 
 
 
 131.8 
 
 96.4 
 
 
 30 
 
 
 
 
 *61.9 
 
 *43.5 
 
 
 
 
 
 
 
 
 
 106.4 
 
 
 31 
 
 
 
 
 
 
 
 
 
 179.3 
 
 120.9 
 
 43.5 
 
 33.5 
 
 
 
 140.1 
 
 32 
 
 527 6 
 
 237.8 
 
 151.3 
 
 
 
 72.4 
 
 56.6 
 
 
 198.2 
 
 138.7 
 
 
 
 143.8 
 
 120.7 
 
 141.6 
 
 No. 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 21170 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 u.c.c. 
 
 513 
 
 237 
 
 156 
 
 
 
 72.2 
 
 55.8 
 
 124.4 
 
 210 
 
 143.3 
 
 
 
 157 
 
 134 
 
 157 
 
 No. 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 21158 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 U.C.C. 
 
 498 
 
 222 
 
 135.4 
 
 
 
 69.9 
 
 54.3 
 
 113.9 
 
 196 
 
 120 
 
 
 
 137 
 
 113.7 
 
 135 
 
 No. 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 1 21156 
 
 
 
 
 
 
 
 
 
 
 
 
 . 
 
 
 
 
 lu.C.C. 
 
 1 * 
 
 512 
 
 218 
 
 138.8 
 
 
 
 61.4 
 
 46.5 
 
 118 
 
 193 
 
 143.9 
 
 
 
 1-34 
 
 106.6 
 
 151 
 
 1 Approximate. 
 
 debted to Paulli for a very complete comparative study of the structure of the mam¬ 
 malian ethmoid, and it is interesting to compare the structures found in the Pleistocene 
 form from Rancho La Brea with those noted by Paulli in Recent edentates. Seven 
 endoturbinals are present in M. harlani, a number corresponding to that in Choloepus 
 
 ° S. Paulli Uber die Pneumaticitiit des Schadels bei den Siiugethieren. Morphol. Jahrb., vol. 28, pp. 147-251, 483- 
 564, plates 7, 8-14, 27-29, 1899-1900. 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 127 
 
 didaclylus. The first endoturbinal in the Rancho La Brea genus forms a relatively 
 large scroll. A forward continuation of this endoturbinal forms a cover to the maxillo- 
 turbinal, as already mentioned. The second endoturbinal, in contrast to that of Cho- 
 loepus, gives rise to two scrolls, of which the ventral one is the larger. 
 
 The endoturbinals of M. harlani are given in table 61 with those of recent edentates 
 recorded by Paulli. 
 
 It is seen from table 61 that Mylodon, in agreement with other American 
 edentates, possesses a number of endoturbinals in excess of that usually found in mam¬ 
 mals (5). The number of endoturbinals in this Pleistocene ground-sloth is duplicated 
 only in Choloepus among the forms investigated by Paulli. 
 
 Fig. 61. —Mylodon harlani Owen. Skull-fragment with right superior teeth of young individual, No. 
 1717-35; inferior and lateral views. X 0.50. Note small size of first superior tooth. Rancho La 
 Brea Pleistocene. 
 
 MANDIBLE. 
 
 In the mandible (plate 24 and plate 27, fig. 1) of Mylodon harlani the horizontal 
 ramus is thick. In many specimens from Rancho La Brea the ramus decreases dis¬ 
 tinctly in depth from the base of the coronoid process to the anterior side of the first 
 lower tooth, in which character they closely resemble the material from Nebraska 
 described by Brown. In other mandibles, as, for example in No. 1717-10 L. A. M., 
 and in No. 21576 U. C. C., this decrease in depth is not so noticeable and the speci¬ 
 mens resemble more the type of lower jaw in M. garmani and in M. robustus. The 
 predental region possesses usually concave lateral borders, so that the width across the 
 anterior end of the mandible is slightly greater than that across the middle. In some 
 the lateral border may be fairly straight. The width across the front end is always less 
 than the distance across the front teeth measured from their outer surfaces. The ante¬ 
 rior upper end of the predental region may extend distinctly above the upper surface of 
 the ramus just in front of the first tooth. 
 
 The symphyseal keel varies somewhat in size and in outline when viewed from the 
 side. The anterior margin of the keel slopes gently downward and backward from the 
 anterior border of the mandible. Below a level with the lower end of the anterior mental 
 foramen the anterior margin curves more decidedly backward and then turns downward 
 again toward the ventral border. Usually two mental foramina are present, and of 
 these the lower anterior one is the larger. The bridge of bone separating the two fora¬ 
 mina may be narrow or wide. Often additional foramina are present in this region, 
 but the total number usually does not exceed three. Sometimes three foramina are 
 present in one ramus and two in the opposite ramus. 
 
 The postero-external opening of the dental canal is situated just below the front 
 end of the base of the coronoid process or opposite the posterior lobe of the last lower 
 tooth. Occasionally it varies slightly in position. The coronoid process is broad and 
 rather erect. It is sometimes narrow. When erect, the sigmoid notch may open widely. 
 
128 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 In some specimens the end of the process extends backward to a point vertically above 
 the anterior portion of the condyle and the sigmoid notch is fairly narrow. The condyle 
 projects inwardly and has a rugose surface. The obliquity of the long axis of the con¬ 
 dyle with reference to the fore-and-aft direction of the ramus varies somewhat. The 
 angle may be broad, or it may be relatively narrow. The angle may resemble that in 
 the mandible of M. garmani. In some specimens it is acute. The postero-internal 
 opening of the dental canal always appears to be larger than that of M. robustus. 
 
 DENTITION. 
 
 Dental formula: • 
 
 4 4 
 
 In the possession of lobate teeth the mylodont ground-sloths exhibit considerable 
 difference from other gravigrade edentates in which the teeth have a simpler form. The 
 individual teeth in Mylodon harlani, as in other mylodont ground-sloths, consist of a 
 relatively soft inner material, the vascular dentine of Owen, surrounded by a harder 
 layer of dentine. The external surface of the hard layer is marked by transverse undu¬ 
 lating lines. Covering this surface is a thin layer of cement, with external surface 
 showing longitudinal lines. In Rancho La Brea specimens the outer layer peels away 
 readily from the hard dentine. 
 
 Fig. 62. —Mylodon harlani Owen. 
 Occlusal views of second supe¬ 
 rior tooth, showing variation 
 in shape of crowns. X 0.50. 
 Rancho La Brea Pleistocene. 
 
 Fig. 63 .—Mylodon harlani Owen. 
 Occlusal views of left fourth 
 superior tooth, showing varia¬ 
 tion in shape of crowns. X 
 0.50. Rancho La Brea Pleis¬ 
 tocene. 
 
 Fig. 64. —Mylodon harlani 
 Owen. Occlusal views of 
 right fifth superior tooth, 
 showing variation in shape 
 of crowns. X 0.50. Ran¬ 
 cho La Brea Pleistocene. 
 
 The superior tooth-rows in most specimens from Rancho La Brea do not diverge 
 anteriorly as much as in Mylodon robustus. In No. 21159 U. C. C. the divergence of 
 the tooth-rows approximates closely that in a skull of M. robustus considered by Lydek- 
 ker (1894, p. 79, plate 49, fig. 2) as belonging to a female. 
 
 Thirty-one skulls in the Los Angeles Museum and 19 in the Museum of Paleon¬ 
 tology, University of California, furnish the following data on the presence and absence 
 of the first tooth in the superior dentition: 
 
 First superior tooth present on right and left sides. 21 
 
 First superior tooth absent on right and left sides. 14 
 
 First superior tooth present on right side and absent on left side. 5 
 
 First superior tooth present on left side and absent on right side. 7 
 
 Doubtful. 3 
 
 50 
 
 It should be noted in this connection that when the first tooth is present it is not 
 always of normal size. This is well shown, for example, in No. 21158 U. C. C., a skull 
 of average size, in which the alveolus for the first tooth measures approximately 8.5 by 
 8.2 mm. 
 
 The series of specimens from the asphalt demonstrates that the tooth may have 
 actually disappeared during the life of the individual. In skull 1717-9 the alveolus for 
 the first tooth of the left side was originally as large as that of the right side. Around 
 the border, however, a growth of bone has taken place which shows a tendency to reach 
 toward the middle of the tooth-socket. That this actually takes place is clearly seen 
 in 1716-1, in which the closing of the first superior alveolus has progressed much further. 
 In this skull the left alveolus is almost completely closed, while that of the right side has 
 a shallow, crescentic opening along the lingual border. 
 
PLEISTOCENE MEGALONYCHJNAE AND MYLODONTIDAE OF RANCHO LA BREA. 129 
 
 No. 1717-12 illustrates how the entire alveolus may be filled by a growth of bone. 
 The final stage is reached in No. 21156 U. C. C. or in the Nebraska skull described by 
 Brown (1903), where the first tooth has disappeared entirely, its former presence being 
 indicated by a scar in front of the second superior tooth. Burmeister (1886, p. 1132) 
 directed attention to a similar closing of the first superior alveolus in the South Amer¬ 
 ican Grypotherium. 
 
 Table 63. —Measurements {in millimeters) of mandibles of Mylodon harlani. 
 
 Catalogue 
 No. 1718. 
 
 Length from 
 anterior end 
 of symphysis 
 to posterior 
 end of condyle. 
 
 Greatest 
 length 
 of sym¬ 
 physis. 
 
 Greatest 
 
 pre¬ 
 
 dental 
 
 width. 
 
 Depth of 
 ramus between 
 third and 
 fourth inferior 
 teeth, meas¬ 
 ured normal to 
 inferior margin. 
 
 Catalogue 
 No. 1718. 
 
 Length from 
 anterior end 
 of symphysis 
 to posterior 
 end of condyle. 
 
 Greatest 
 length 
 of sym¬ 
 physis. 
 
 Greatest 
 
 pre- 
 
 dental 
 
 width. 
 
 Depth of 
 ramus between 
 third and 
 fourth inferior 
 teeth, meas¬ 
 ured normal to 
 inferior margin. 
 
 1 
 
 359.5 
 
 113.7 
 
 143.5 
 
 93.5 
 
 21 
 
 381.1 
 
 96.2 
 
 135.3 
 
 83.2 
 
 2 
 
 409.8 
 
 108.9 
 
 x 154 
 
 95.6 
 
 22 
 
 382.1 
 
 101.5 
 
 128.7 
 
 83.2 
 
 3 
 
 
 97.8 
 
 123.6 
 
 80.4 
 
 23 
 
 355.4 
 
 94.6 
 
 122.2 
 
 83.5 
 
 L5 
 
 363.8 
 
 96.9 
 
 126.6 
 
 79.4 
 
 24 
 
 *382 
 
 
 145.1 
 
 92.1 
 
 L6 
 
 386.8 
 
 
 
 J 93.3 
 
 25 
 
 372.6 
 
 90.9 
 
 128.2 
 
 85.1 
 
 L7 
 
 *330 
 
 
 
 73.3 
 
 26 
 
 377.1 
 
 108.2 
 
 137.4 
 
 79.7 
 
 L8 
 
 
 
 
 73.8 
 
 27 
 
 383.6 
 
 100.6 
 
 142.3 
 
 85 
 
 L9 
 
 374.1 
 
 
 
 87.9 
 
 28 
 
 388.5 
 
 97.8 
 
 142.1 
 
 88 
 
 L10 
 
 387.8 
 
 
 
 97.1 
 
 29 
 
 386.9 
 
 113.5 
 
 *151 
 
 96 
 
 Lll 
 
 *295.5 
 
 
 
 69 
 
 30 
 
 382.5 
 
 92.1 
 
 130.1 
 
 83.8 
 
 R12 
 
 
 
 
 87.8 
 
 31 
 
 
 
 
 91.9 
 
 R13 
 
 375.1 
 
 
 
 
 32 
 
 394.4 
 
 111.4 
 
 145.4 
 
 84.6 
 
 R14 
 
 353.1 
 
 
 
 78.3 
 
 33 
 
 367.3 
 
 96.9 
 
 130.3 
 
 77.3 
 
 R15 
 
 379.1 
 
 
 
 85.2 
 
 No. 21170 
 
 
 
 
 
 R16 
 
 385.4 
 
 
 ..... 
 
 94.6 
 
 u. c. c. 
 
 386 
 
 115.6 
 
 *109.5 
 
 92 
 
 17 
 
 x 375 
 
 93 
 
 124.9 
 
 79 
 
 No. 21158 
 
 
 
 
 
 18 
 
 
 99.6 
 
 144.1 
 
 *95 
 
 U. C. C. 
 
 386 
 
 116.4 
 
 * 95.9 
 
 83 
 
 19 
 
 
 
 130.9 
 
 85 
 
 No. 21156 
 
 
 
 
 
 20 
 
 393.1 
 
 104.6 
 
 141.8 
 
 88.5 
 
 U. C. C. 
 
 385 
 
 109.5 
 
 102.7 
 
 94.7 
 
 1 Approximate. L, left ramus; R, right ramus. 
 
 The crown of the first superior tooth (1) is distinctly curved and is round or oval 
 in cross-section. The surface occluding with the first inferior tooth is worn obliquely 
 to the crown. The layer of hard dentine is relatively thicker in this tooth than in other 
 superior teeth. The second superior tooth (2) has the greatest antero-posterior diameter 
 of any of the upper teeth. The tooth varies somewhat in the long diameter (see fig. 
 62), but seems never to be shortened in this direction so much as in Mylodon robustus. 
 The sides of the tooth may be flat or slightly convex. In some specimens a median 
 longitudinal sulcus extends the length of the crown on the inner and outer sides. The 
 occlusal surface is beveled anteriorly and posteriorly, the former surface wearing against 
 1, the latter against the anterior half of the occlusal surface of 2. 
 
 The third superior tooth (3) remains quite constant in shape and resembles closely 
 the corresponding tooth in M. garmani. The description which Allen (1913, p. 324) 
 gives of the tooth can be applied also to 3 in the specimens from Rancho La Brea. The 
 antero-internal lobe of the tooth occludes with 2, while the postero-internal lobe and 
 the antero-external lobe wear against 3. 
 
 The fourth superior tooth (4) may resemble somewhat 3 in shape, but as Allen has 
 indicated, the former tooth is more compressed in the long axis of the tooth-row. A 
 review of the series from the asphalt reveals some variation in the shape of 4 as shown 
 in figure 63. Occasionally, 4 may be so compressed as to resemble closely the tooth in 
 M. garmani; 4 occludes with the anterior lobe of 4. 
 
Table 64 . — Measurements {in millimeters ) of the dentition of Mylodon harlani. 
 
 130 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA 
 
 
 • 0 qoj jouojsod 
 
 GO 
 
 X 
 
 
 
 
 CO 
 
 
 
 
 
 to 
 
 • ^ 
 
 
 
 
 rH 
 
 
 to 
 
 
 CO 
 
 Ol 
 
 
 rH 
 
 X 
 
 
 to CM 
 
 to 
 
 
 1 > 
 
 
 
 
 
 
 
 jo jojeramp JS 0 JU 0 J 3 ‘f. 
 
 rH 
 
 Ol 
 
 rH 
 
 CM 
 
 
 o 
 
 01 
 
 
 GO 05 
 
 rH rH 
 
 05 
 
 rH 
 
 
 
 05 
 
 rH 
 
 • 05 
 
 • rH 
 
 rH 
 
 Ol 
 
 
 
 CO 
 
 O l> 
 CM »-h 
 
 
 o 
 
 Ol 
 
 rH 
 
 Ol 
 
 05 
 
 rH 
 
 X 
 
 Ol 
 
 Ol 
 
 Ol 
 
 Ol 
 
 05 rH 
 rH Ol 
 
 o 
 
 CM 
 
 
 X 
 
 CM 
 
 
 
 
 
 Ol 
 
 
 
 
 
 
 
 rH rH 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 •0CJOJ JOU0JUB 
 
 x 
 
 CO 
 
 
 
 
 to 
 
 
 to 
 
 
 
 CO 
 
 • to 
 
 
 rH 
 
 
 05 
 
 I> 
 
 
 
 »o 1 > 
 
 05 
 
 
 
 
 X X 
 
 
 
 
 
 
 
 X 
 
 
 JO J0J0UIUtp JS0JU0J3 ‘f 
 
 CM 
 
 »0 
 
 Ol 
 
 
 o 
 
 01 
 
 
 CM 
 
 CO 
 
 Ol 
 
 rH 
 
 Ol 
 
 
 
 Ol 
 
 Ol 
 
 • Ol 
 
 • Ol 
 
 to 
 
 Ol 
 
 CO 
 
 CM 
 
 
 X 
 
 rH 
 
 CO 
 
 Ol 
 
 Ol 
 
 Ol 
 
 co 
 
 Ol 
 
 Tt 4 
 
 CM 
 
 co 
 
 Ol 
 
 rH 
 
 CM 
 
 CO 
 
 Ol 
 
 ‘O 
 
 Ol 
 
 to 
 
 CM 
 
 Ol T* 
 Ol Ol 
 
 rH 
 
 Cl 
 
 
 05 
 
 Ol 
 
 
 
 
 
 Ol 
 
 X 
 
 
 
 
 
 
 
 —« 
 
 
 
 
 rH rH 
 
 
 
 
 rH rH 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 •j d^eun? 
 
 
 05 
 
 rH 
 
 CO 
 
 
 
 CO 
 
 
 
 CO 
 
 CO 
 
 
 05 
 
 CO 
 
 
 05 
 
 
 
 
 to 
 
 
 1 > 
 
 CM 
 
 X 
 
 
 ^ CM 
 
 to 
 
 
 
 
 
 
 
 
 
 -ip jou 0 jsod-oa 0 juu l f 
 
 CO 05 
 tO to 
 
 CM 
 
 to 
 
 05 
 
 Tf 
 
 
 X 
 
 rH 
 
 to 
 
 TfH Ol 
 lO 
 
 to 
 
 rH 
 
 to 
 
 • o 
 
 • to 
 
 rH 
 
 to 
 
 X 
 
 to 
 
 
 to 
 
 -r 
 
 05 
 
 05 
 
 
 CO 
 
 tO 
 
 to 
 
 O 
 
 to 
 
 Tt 4 
 
 to 
 
 rH 
 
 to 
 
 to 
 
 to 
 
 X X 
 
 to to 
 
 to 
 
 to 
 
 
 o 
 
 JP 
 
 
 
 
 
 CO 
 
 to 
 
 
 •jajoiunip JS 0 JB 0 J 8 
 
 tO 
 
 rH 
 
 oo 
 
 
 
 
 
 
 
 
 
 
 
 
 05 
 
 to 
 
 H 
 
 
 
 CO 
 
 
 
 . 
 
 
 
 H • 
 
 X 
 
 
 
 
 
 T}H 
 
 
 
 
 oj feuLiou oo^jjns 
 
 Q 
 
 Q 
 
 
 
 
 
 
 
 
 
 
 
 05 
 
 X 
 
 CO 
 
 H 
 
 CO 
 
 . 
 
 
 05 
 
 
 
 
 
 
 t> • 
 
 
 
 o 
 
 
 
 
 
 X 
 
 
 jusnjooo jo J 0 j 0 iuuip ‘g 
 
 Ol 
 
 01 
 
 
 rH 
 
 
 
 
 
 
 
 
 
 rH 
 
 rH 
 
 rH 
 
 rH 
 
 rH 
 
 *' 
 
 
 rH 
 
 
 
 
 
 
 rH • 
 
 rH 
 
 
 CM 
 
 
 
 rH 
 
 
 CM 
 
 H 
 
 
 *0D13J 
 
 
 l- 
 
 CM 
 
 
 
 
 
 
 
 
 
 
 
 
 Ol 
 
 X 
 
 to 
 
 
 
 X 
 
 rH 
 
 CO 
 
 
 
 
 co • 
 
 Ol 
 
 
 X 
 
 
 
 
 
 X 
 
 r; 
 
 -jns j^snpDo ssojd'b 
 
 CM 
 
 Ol 
 
 O 
 
 rH 
 
 
 
 
 
 
 
 
 • H 
 
 rH 
 
 
 X 
 
 to !> 
 
 
 
 05 
 
 to 1 > 
 
 
 
 
 X • 
 
 X 
 
 
 X 
 
 
 
 
 
 CO 
 
 4) 
 
 J0j0uinip JS0JU0J3 ‘g 
 
 CO 
 
 CO 
 
 CO 
 
 CO 
 
 
 
 
 
 
 
 
 • Ol 
 
 co 
 
 
 Ol 
 
 CM 
 
 Ol 
 
 
 
 CM 
 
 Ol 
 
 CM 
 
 
 
 
 Ol • 
 
 CM 
 
 
 X 
 
 
 
 Ol 
 
 
 Ol 
 
 O 
 
 
 
 to 
 
 
 CO 
 
 GO • 
 
 rH 
 
 
 
 
 
 rH 05 
 
 
 
 
 CM 
 
 . 
 
 . 
 
 
 
 
 
 CO 
 
 rH 
 
 . 
 
 • X 
 
 
 
 
 
 X 
 
 
 . 
 
 o 
 
 •J0J0lUBtp 0S.I0ASUBJJ 'Z 
 
 CO 
 
 CO 
 
 rH 
 
 Ol 
 
 01 • 
 
 CM 
 
 
 
 . 
 
 
 CO 05 
 
 CO 
 
 
 
 05 
 
 
 
 CM 
 
 
 
 
 X 
 
 X 
 
 
 • X 
 
 o 
 
 
 X 
 
 
 
 o 
 
 
 ! 
 
 
 
 Ol 
 
 01 
 
 01 
 
 Ol 
 
 Ol • 
 
 Ol 
 
 
 
 
 
 Ol rH 
 
 Ol 
 
 
 
 rH 
 
 
 
 Ol 
 
 
 
 CM 
 
 Ol 
 
 CM 
 
 
 • Ol 
 
 Ol 
 
 
 Ol 
 
 
 
 Ol 
 
 
 
 <D 
 
 
 r- 1 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 *3 
 
 t-H 
 
 •J0J0UI12 
 
 
 05 
 
 GO 
 
 05 
 
 co • 
 
 co 
 
 
 . 
 
 
 
 CO 
 
 to 
 
 
 
 
 
 
 
 
 
 CO 
 
 
 CM 
 
 
 to to 
 
 X 
 
 
 H 
 
 
 
 CO 
 
 
 • 
 
 
 -ip joi.i 0 jsod-oj 0 juu ‘z 
 
 rH 
 
 CO 
 
 05 
 
 CM 
 
 to b- 
 Ol Ol 
 
 00 • 
 Ol • 
 
 cc 
 
 CM 
 
 
 
 
 
 rH 
 
 CO Ol 
 
 X 
 
 Ol 
 
 
 
 »o 
 
 CM 
 
 
 
 05 
 
 Ol 
 
 
 
 01 
 
 rH 
 
 X 
 
 05 
 
 Ol 
 
 
 XC5N 
 Ol CM CM 
 
 
 X 
 
 Ol 
 
 
 
 CO 
 
 Ol 
 
 
 • 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 lO 
 
 co 
 
 
 
 
 
 tO 
 
 
 
 
 
 • to CO 
 
 to 
 
 
 05 
 
 CO 
 
 
 
 
 to 05 
 
 X 
 
 
 
 rH 05 
 
 rH 
 
 
 Ol 
 
 
 
 
 
 Ol 
 
 
 •J0J0lUUip 0SJ0ASUBJJ ‘ J 
 
 r- 
 
 
 to 
 
 
 CO • 
 
 co 
 
 . 
 
 
 
 
 • CM 
 
 CO 
 
 to 
 
 . 
 
 
 to 
 
 
 . 
 
 
 
 co 
 
 CO 
 
 
 
 co X 
 
 X 
 
 
 o 
 
 
 
 ! 
 
 
 X 
 
 
 
 r—i 
 
 rH 
 
 rH 
 
 
 
 • 
 
 rH 
 
 
 
 
 
 • rH 
 
 rH 
 
 rH 
 
 
 rH 
 
 rH 
 
 
 
 
 rH 
 
 rH 
 
 rH 
 
 
 rH 
 
 rH rH 
 
 7-1 
 
 
 Ol 
 
 
 
 
 
 rH 
 
 
 "Jd^dure 
 
 CM 
 
 
 Ol 
 
 
 
 
 
 • 
 
 
 
 • 
 
 
 t> CO 
 
 
 CM 
 
 to 
 
 
 
 
 
 t> 
 
 05 
 
 • 
 
 
 05 
 
 to 
 
 
 r- 
 
 
 
 
 
 rH 
 
 
 -ip Jtou 0 jsod-oj 0 juB *x 
 
 CM 
 
 to 
 
 CM 
 
 CO 
 
 rH 
 
 
 05 • 
 
 rH • 
 
 o 
 
 Ol 
 
 
 
 
 
 • o 
 
 • CM 
 
 co 
 
 Ol 
 
 CO 
 
 CM 
 
 
 05 
 
 rH 
 
 Ol 
 
 CM 
 
 
 
 
 Ol 
 
 Ol 
 
 o 
 
 CM 
 
 rH 
 
 CM 
 
 
 o 
 
 Ol 
 
 CM X CO 
 
 Ol CM rH 
 
 
 X 
 
 CM 
 
 
 
 
 
 CO 
 
 Ol 
 
 
 "juomojnsuora 
 
 tO 
 
 
 
 
 
 
 
 
 
 
 
 
 rH 
 
 Ol 
 
 
 
 CO 
 
 
 
 
 05 
 
 rH 
 
 X 
 
 
 Tt< 
 
 05 
 
 
 
 X 
 
 
 
 to 
 
 
 to 
 
 
 JBJO 0 AJB ‘S 0 IJ 0 S qjooj 
 
 tO 
 
 
 
 
 
 
 
 
 
 
 
 
 05 
 
 X 
 
 
 CO 
 
 to 
 
 
 
 
 o 
 
 05 
 
 X 
 
 
 X 
 
 Ol CO 
 
 
 
 to 
 
 
 
 rH 
 
 
 05 
 
 
 CO 
 
 to 
 
 
 
 
 
 co 
 
 
 
 
 
 • Ol 
 
 co 
 
 CO 
 
 
 CM 
 
 CO 
 
 
 
 
 ro 
 
 CO 
 
 X 
 
 
 -V 
 
 
 X 
 
 
 
 
 
 
 
 
 
 JOIJ 0 JUI jo qjSuo'q 
 
 r—1 
 
 rH 
 
 
 
 
 
 
 
 
 
 
 
 rH 
 
 rH 
 
 
 rH 
 
 rH 
 
 
 
 
 rH 
 
 rH 
 
 rH 
 
 
 rH 
 
 rH rH 
 
 rH 
 
 
 rH 
 
 
 
 rH 
 
 
 rH 
 
 
 •ll°0 TV V '1 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 p d 
 
 
 GO d 
 
 
 co d 
 
 
 rH 
 
 Ol 
 
 CO to 
 
 co r> 
 
 X 
 
 05 
 
 O 
 
 rH 
 
 Ol 
 
 CO ^ 
 
 to 
 
 X 
 
 05 
 
 o 
 
 rH 
 
 Ol 
 
 co 
 
 H 
 
 to 
 
 X 
 
 05 
 
 o 
 
 rH Ol 
 
 X 
 
 c 
 
 
 o 
 
 to 
 
 d 
 
 6 
 
 tO . 
 
 dq 
 
 
 ‘Sill N 
 
 
 
 
 
 
 
 
 
 rH 
 
 
 rH 
 
 rH rH 
 
 rH 
 
 rH 
 
 rH 
 
 CM 
 
 CM 
 
 Ol 
 
 Ol 
 
 Ol 
 
 Ol 
 
 CM 
 
 Ol 
 
 CM 
 
 X 
 
 X X 
 
 X 
 
 £ 
 
 s? 
 
 rH 
 
 % 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 iM ^3 
 
 
 Cl 
 
 d 
 
 
 
 
 
 
 CO 
 
 
 CO 
 
 05 
 
 
 
 05 
 
 
 co 
 
 • rH 
 
 
 
 
 CM 
 
 tO 
 
 05 
 
 
 
 
 
 X 
 
 
 
 X 
 
 
 . 
 
 
 
 
 
 
 
 
 •J0J0UI13ip 0SJ0ASUBJJ ‘e 
 
 CM 
 
 05 
 
 
 CO 
 
 05 rH 
 
 
 
 05 
 
 O 05 
 
 • 05 
 
 X 05 
 
 05 
 
 X 
 
 CO 
 
 co 
 
 X 
 
 
 
 X 
 
 Ol 
 
 
 
 Ol 
 
 
 
 
 o 
 
 
 
 
 
 
 
 CM 
 
 rH 
 
 
 rH 
 
 r— 
 
 1 Ol 
 
 
 
 rH 
 
 Ol 
 
 rH 
 
 • rH 
 
 rH 
 
 rH 
 
 rH 
 
 rH 
 
 rH 
 
 rH 
 
 rH 
 
 
 
 r—H 
 
 Ol 
 
 
 
 CM 
 
 
 • 
 
 
 Ol 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 rH 
 
 
 
 
 
 
 
 
 
 
 to 
 
 CO t> 
 
 
 
 CO 05 
 
 CM l> ^ 
 
 
 
 05 
 
 CM 
 
 05 
 
 X 
 
 
 to 
 
 
 
 
 
 
 X 
 
 
 
 
 CM 
 
 
 
 
 
 
 -ip roij 0 jsod-oj 0 juB ‘cj 
 
 to 
 
 CM 
 
 CO 
 
 CM 
 
 
 CM 
 
 Ol 
 
 >0 T* 
 
 CM CM 
 
 
 
 CM 
 
 CM 
 
 CO T- O 
 CM CM CO 
 
 CM 
 
 Ol 
 
 Th 
 
 CM 
 
 to 
 
 Ol 
 
 CM 
 
 CM 
 
 O CO CO 
 CM CM Ol 
 
 CM 
 
 
 rH 
 
 CM 
 
 TP 
 
 CM 
 
 
 
 CM 
 
 
 
 
 O 
 
 Ol 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 rH 
 
 
 . 
 
 ^ CO 05 
 
 X 
 
 
 
 05 
 
 • CO 
 
 CM 
 
 rH 
 
 
 
 X 
 
 to 
 
 X 
 
 
 tO 
 
 X 
 
 rH 
 
 
 
 
 
 . 
 
 
 
 
 
 
 
 
 •J0J0lUUip 0SJ0ASUBJJ ‘f 
 
 . 
 
 to 
 
 
 
 CO Rr 
 
 Ol 
 
 Ol 
 
 
 »o 
 
 co 
 
 ■ ^ 
 
 Ol 
 
 05 
 
 
 
 
 05 
 
 CO 
 
 
 o 
 
 X 
 
 X 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 Ol 
 
 
 
 Ol Ol 
 
 Ol 
 
 Ol 
 
 
 Ol 
 
 Ol 
 
 • Ol 
 
 Ol 
 
 rH 
 
 
 
 Ol 
 
 rH 
 
 Ol 
 
 
 
 CM 
 
 Ol 
 
 
 
 Ol 
 
 
 
 
 CM 
 
 
 
 
 
 
 •jojouns 
 
 
 rH 
 
 
 
 oo »o 
 
 X 05 
 
 
 to 
 
 
 • 05 
 
 
 
 CM 
 
 
 CO 
 
 rH 
 
 
 
 rH 
 
 
 05 
 
 
 
 i> 
 
 
 . 
 
 
 
 
 
 
 
 
 -JP JOU0JSOd-OJ0JUB ‘f 
 
 
 CO 
 
 Ol 
 
 
 
 b- to 
 
 Ol Ol 
 
 o 
 
 Ol 
 
 rH 
 
 CM 
 
 
 CM 
 
 Ol 
 
 o 
 
 CM 
 
 • o 
 
 • Ol 
 
 05 
 
 rH 
 
 CO 
 
 Ol 
 
 Ol 
 
 Ol 
 
 
 Ol 
 
 Tt 4 
 
 CM 
 
 o 
 
 CM 
 
 
 o 
 
 CM 
 
 CM 
 
 Ol 
 
 X 
 
 Ol 
 
 
 
 Ol 
 
 Ol 
 
 
 
 
 Ol 
 
 CM 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 rH 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 X 
 
 . 
 
 • 05 
 
 
 
 X 
 
 • ^ 
 
 
 • TJH 
 
 
 X 
 
 rH 
 
 
 CO 
 
 CO 
 
 rH 
 
 
 X 
 
 to 
 
 rH 
 
 
 
 X 
 
 
 ’J0J0lUBip 0SJ0ASUUJJ 
 
 ‘e 
 
 X 
 
 CM 
 
 X 
 
 CM 
 
 • 
 
 • rH 
 
 • CM 
 
 to X 
 
 Ol Ol 
 
 rH 
 
 CM 
 
 • Tt< 
 
 • CM 
 
 CM 
 
 CM 
 
 • Ol 
 
 • CM 
 
 
 CM 
 
 Ol 
 
 Ol 
 
 CO 
 
 Ol 
 
 Ol 
 
 Ol 
 
 Ol 
 
 01 
 
 Ol 
 
 
 t> 
 
 rH 
 
 Ol 
 
 CO 
 
 Ol 
 
 CO 
 
 CM 
 
 CM 
 
 CM 
 
 rH 
 
 CM 
 
 
 •ja^ouiii 
 
 
 rH 
 
 CM 
 
 . 
 
 ♦ rH 
 
 05 
 
 X co 
 
 • X 
 
 05 
 
 • 1> 
 
 
 
 X 
 
 to CM 
 
 to 
 
 Ol 
 
 
 CM 
 
 to 
 
 Ol 
 
 
 CM 
 
 X 
 
 H— > 
 
 <D 
 
 O 
 
 HH 
 
 -ip JOIJ0JSOd-OJ0JUB 
 
 ‘g 
 
 X 
 
 CM 
 
 X 
 
 CM 
 
 
 • Ol 
 
 • X 
 
 O l> 
 X Ol 
 
 CM 
 
 • T- 
 
 • CM 
 
 CO 
 
 CM 
 
 • CO 
 
 • CM 
 
 
 co 
 
 Ol 
 
 di 
 
 CO 
 
 Ol 
 
 o 
 
 X 
 
 >o 
 
 CM 
 
 to 
 
 Ol 
 
 
 Ol 
 
 Ol 
 
 1^ 
 
 CM 
 
 X 
 
 CM 
 
 CO 
 
 CM 
 
 CO 
 
 Ol 
 
 CO 
 
 Ol 
 
 Sh 
 
 O 
 
 
 
 05 
 
 
 
 • rH 
 
 l> 
 
 05 
 
 X 
 
 • 05 
 
 
 • to 
 
 to 
 
 05 
 
 
 X 
 
 H 1 
 
 CM 
 
 05 
 
 
 
 rH 
 
 X 
 
 tr 
 
 
 
 •G 
 
 o 
 
 a 
 
 3 
 
 M0;ocu^ip 0SJ0ASin3J^ 
 
 1 z 
 
 05 
 
 rH 
 
 05 
 
 rH 
 
 rH 
 
 • l> 
 
 • rH 
 
 rH 
 
 to 
 
 rH 
 
 rH 
 
 • CO 
 
 ■ rH 
 
 CO 
 
 rH 
 
 • CO 
 
 • rH 
 
 CO 
 
 rH 
 
 rH 
 
 rH 
 
 rH 
 
 o 
 
 CM 
 
 CO 
 
 rH 
 
 rH 
 
 X 
 
 rH 
 
 
 CO 
 
 rH 
 
 X 
 
 rH 
 
 o 
 
 01 
 
 to 
 
 rH 
 
 rH 
 
 02 
 
 *J0)0UIt? 
 
 
 CC 
 
 to 
 
 
 • T- 
 
 X 
 
 to 
 
 to 
 
 • 
 
 
 • to 
 
 X 
 
 05 
 
 CM 
 
 05 
 
 
 
 CM 
 
 05 
 
 
 to 
 
 
 X 
 
 05 
 
 'X 
 
 
 -ip .iouojsod-oj 0 jut: 
 
 ‘z 
 
 rH 
 
 cp 
 
 Ol 
 
 X 
 
 X 
 
 X 
 
 • X 
 
 • X 
 
 CO 
 
 X 
 
 05 
 
 Ol 
 
 rH 
 
 X 
 
 • o 
 
 • X 
 
 rH 
 
 X 
 
 • 05 
 
 • X 
 
 X 
 
 X 
 
 rH 
 
 TP 
 
 o 
 
 X 
 
 H 
 
 X 
 
 X 
 
 X 
 
 X 
 
 X 
 
 Ol 
 
 JX 
 
 
 — 
 
 X 
 
 Ol 
 
 X 
 
 X 
 
 CM 
 
 o 
 
 X 
 
 rH 
 
 •jejoranip ©sjoasubjj ‘j 
 
 13.9 
 
 .... 
 
 15.4 
 
 13.2 
 
 13.6 
 
 4 21 
 
 13.5 
 
 17.1 
 
 20 
 
 *J0;01U'G 
 -Tp JOIJ0^SOCl-OJ0^UT? 
 
 15.2 
 
 16.7 
 
 14.9 
 
 17 
 
 22 
 
 16 
 
 20.7 
 
 20.5 
 
 *iu0ui0jnsi30ui 
 J r BJO0A|'l3 ‘S0TJ0S IJ^OO^ 
 
 jojjoclns jo 
 
 119.8 
 
 141.9 
 
 111.3 
 
 142.8 
 
 136.1 
 
 138.2 
 
 137.8 
 
 114.8 
 
 129.3 
 
 127.1 
 155.5 
 
 118.9 
 
 153.1 
 
 117.4 
 122.8 
 
 117.8 
 
 149.1 
 
 153 
 
 136.2 
 
 126.3 
 
 •n°o tv y i 
 ‘L I ll 'ON 
 
 od xo i-d 
 
 HNX^t0C0NX05OHXt0NX©ONX^t0C0 05N aN • n ^ • 
 
 rn rH cc co -r CC Cl CJ hhhhhhhOIOICMCMCMCMOIX Jd H r - ) l_" rH f) Jd H ^ > 
 
 w 't ci rr tj th fi n m t ci ^ cc ci i w ^,— 
 
 w D M & 
 
 Approximate. 2 First superior tooth present on both sides. 3 First superior tooth present on one side only. 4 First superior tooth absent on both sides. 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 131 
 
 Iii 5 (fig. 64) the anterior lobe is always more expanded than the posterior lobe. 
 Usually the sulcus on the outer side of the tooth is shallow, but in some specimens the 
 groove may deepen; 5 occludes with the posterior portion of 4. 
 
 The form of the triturating surface of the first inferior tooth (T) is dependent upon 
 the presence or absence of the first superior tooth. When the latter is present, the 
 wearing surface of 1 consists of an anterior and a posterior beveled face meeting'in a 
 transverse ridge across the middle of the crown. When the first superior tooth is absent 
 a single wearing surface occurs in 1 and is directed to the inner posterior side. This 
 type of tooth resembles that in Brown’s specimen from the Pleistocene of Nebraska. In 
 passing, it should be noted that 1 in the Nebraska ramus has moved from the normal 
 position in the alveolus. 
 
 Fig. 65. —Mylodon harlani Owen. 
 Occlusal views of left third 
 inferior tooth, showing variation 
 in shape of crowns. X 0.50. 
 Rancho La Brea Pleistocene 
 
 Fig. 66. —Mylodon harlani Owen. Occlusal views of right fourth inferior 
 tooth, showing variation in shape of crowns. X 0.50. Rancho 
 La Brea Pleistocene. 
 
 2 is not unlike the tooth in M. garmani. A deep longitudinal sulcus extends the 
 length of the inner face at approximately the middle and assists in defining the antero- 
 internal and postero-internal lobes. A shallow and broad sulcus situated on the outer 
 posterior side defines also an antero-external lobe. 
 
 The third inferior tooth (3) may occasionally be quite broad from the antero- 
 internal to the postero-external side (fig. 65). Teeth from the asphalt may resemble 
 closely in shape 3 of the types of M. harlani and M. garmani. 
 
 Table 65 .—Measurements (in millimeters ) of endocranial cast of Mylodon harlani, 
 
 skull No. 1717-33. 
 
 Greatest length from anterior end of cerebrum to posterior end of cerebellum. 1 117 
 
 Greatest transverse diameter of anterior end of cerebrum. 81 
 
 Greatest transverse diameter of posterior end of cerebrum. *96 
 
 Transverse diameter of cerebellum. 1 90.4 
 
 Height from infundibulum to dorsal surface of cerebrum. 75 
 
 1 Approximate. 
 
 As shown in figure 66, the shape of the fourth inferior tooth (4) varies considerably. 
 The anterior lobe is always more or less oblique to the long axis of the tooth-row, while 
 the posterior lobe may sometimes be at right angles to it. The external surface of the 
 isthmus connecting the two lobes may form a convex ridge extending longitudinally, 
 giving the tooth an appearance similar to that in Brown’s specimen of Mylodon from 
 Nebraska. 
 
 ENDOCRANIAL CAST. 
 
 The dorsal, lateral, and ventral views of the endocranial cast of Mylodon harlani are 
 shown in plate 26. A greater difference in width between the anterior and posterior 
 portions of the cerebrum in Mylodon harlani is seen when a comparison is made of the 
 dorsal views of the endocranial casts of the North American genus and of Grypotherium 
 (Woodward, 1900, plate 7). In lateral view the cerebral part of the cast is seen to 
 descend more gradually to the olfactory lobes than in Grypotherium. Judging from the 
 surface of the cast, the cerebrum was rather richly convoluted. The cerebellum is rela¬ 
 tively less prominent in M. harlani than in Grypotherium. It is decidedly smaller than 
 in Glyptodon. In the form from Rancho La Brea the olfactory lobes extend noticeably 
 in front of the cerebrum and are divergent when viewed from above. 
 
132 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 ELEMENTS OF HYOID ARCH. 
 
 The elements of the hyoid apparatus (plate 27, fig. 2) in Mylodon harlani resemble 
 closely in their general features those of M. garmani, which have been adequately 
 described by Allen (1913, p. 327, pi. 3, figs. 5, 6), as follows: 
 
 “The stylohyal is the longest, 130 mm. in extreme length, with an irregularly rounded stem, expanding 
 dorsally into a squarish plate at whose anteroventral corner is a rounded projection for articulation with the 
 skull. Distally it bears a transverse articular surface. What corresponds to an epihyal articulates by a flat 
 surface with the ventral end of this bone. It is compressed laterally and its distal portion is transversely ex¬ 
 panded with an elliptical articular facet, 14 X 10 mm., on its posterior face. Its extreme length is 62 mm., or 
 about half that of the stylohyal. A third smaller bone of trapezoidal outline, with a small elliptical facet 
 on each of its convergent sides, serves for the articulation of the epihyal with the body of the hyoid. This series 
 of three bones is present on both sides, and the two smallest, or ceratohyals, articulate each with the facet of a 
 protuberance marking either end of the body or basihyal. The last is thoroughly ankylosed with the thyrohyals 
 to form a V-shaped bone, whose sides are a little expanded dorsally and whose point is widely emarginate at its 
 posterior border. A small facet at the posterior tip of each cornu probably marks the attachment of cartilage.” 
 
 Specimens of the stylohyal from Rancho La Brea average 129.3 mm. in extreme 
 length (shortest, 117.7 mm.; longest, 139.4 mm.), and thus approximate closely the 
 corresponding element of M. garmani. The medial surface of the proximal expanded 
 portion is often formed into several concavities separated by low ridges. The lateral 
 surface, on the other hand, is comparatively smooth and even with exception of that 
 portion near the dorsal border where rugosities are developed which aid in the attach¬ 
 ment of the stylohyal to the skull. The epihyal of M. harlani is somewhat longer than 
 that of M. garmani; specimens from the asphalt beds average 70.8 mm. in total length 
 (shortest, 63 mm.; longest 79.1 mm.). A single example of fusion of stylohyal and epi¬ 
 hyal has been noted in the museum collections. 
 
 The chain of elements constituting the hyoid apparatus in M. harlani resembles 
 that of the South American form figured by Burmeister. 0 
 
 On the posterior side and at the end of each cornu of the basihyal a small facet is 
 present, as already noted by Allen. These facets indicate points of contact with the 
 thyroid cartilage. In the collections of the Los Angeles Museum occur specimens, 
 indicative of at least four individuals of Mylodon harlani, in which the thyroid cartilage 
 is ossified. The element thus formed, when viewed in posterior aspect, has somewhat the 
 appearance of a large avian wishbone. Unfortunately, all the specimens from the 
 asphalt beds are more or less fragmentary, and this condition is to be attributed alike 
 perhaps to their somewhat delicate structure and to the incomplete ossification of the 
 cartilage. Among the living South American edentates, an ossification of the thyroid 
 cartilage is known to occur in the giant ant-eater ( Myrmecophaga jubata). It has not 
 been determined whether in Mylodon harlani the ossification of the cartilage is a con¬ 
 stant characteristic of the animal or whether such ossification occurred only in old 
 individuals. The fact that only four individuals are certainly known to possess an 
 ossified thyroid may suggest the latter, but the chances of preservation in a sufficiently 
 complete state to permit recognition may likewise be a potent factor in accounting for 
 the shortage in the representation of this element. It is entirely probable that a number 
 of additional specimens will become available when the entire Rancho La Brea collection 
 of vertebrate material is examined in detail. It is also to be noted that, in the small 
 series at hand, considerable variation is shown in the size of the specimens, and that the 
 slender element, No. 1705-3, does not represent an old individual. 
 
 The thyroid ossification is broad in antero-posterior extent, with the ventral surface 
 sharply keeled in median line and toward the anterior end. Posteriorly the keel ends in 
 a gentle and broad convexity which on the dorsal surface is expressed by a concavity 
 
 ° H. Burmeister. Lista de los mamiferos fosiles del terreno diluviano. Ann. Mus. Buenos Aires, vol. 1, plate 5, fig. 
 
 9, 1864-1869. 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA.BREA. 133 
 
 lying adjacent to the posterior border and at the median junction of the sides of the 
 thyroid. This concavity is divided by a low median ridge which supports a tubercle. 
 The dorsal or free extremities of the sides of the element are modified to secure attach¬ 
 ment with adjacent structures. At the antero-dorsal angle of the side a process is formed 
 which bears an elliptical facet directed forward and outward and which secures the at¬ 
 tachment of the thyroid with the lateral arms of the basihyal. Between the anterior 
 and posterior processes the wall of the thyroid, in specimen No. 1705-2, is pierced by a 
 short canal which extends from the dorsal surface to the inner side. This foramen is 
 absent in the second specimen (No. 1705-4), in which this region of the thyroid is pre¬ 
 served. 
 
 The posterior process is much more prominent than the anterior; it may be quite 
 thick at the base, and the external surface is strongly ridged and bears at the tip an 
 obliquely placed, elongated elliptical facet which is directed mainly inward and slightly 
 upward. This facet joined with the cricoid element. From the nature of the surface of 
 this facet, which, in contrast to that of the anterior facet articulating with the cornu of 
 the basihyal, is not smooth, but roughened, seems to indicate that the cricoid cartilage 
 was not ossified. 
 
 Table 66 . —Measurements {in millimeters). 
 
 Stylohyal: 
 
 Greatest length, average 20 specimens, right 
 
 series. 129.3 
 
 Width of proximal end, average 21 specimens, 
 
 right series. 47.6 
 
 Epihyal: 
 
 Greatest length, average 25 specimens, right 
 
 series. 70.8 
 
 Greatest width, average 24 specimens, right 
 
 series. * . 32.6 
 
 Ceratohyal; left series: 
 
 Greatest length. 30.2 27.8 
 
 Width. 19.9 22.5 
 
 Basihyal: 
 
 Greatest extent across extremities of diverging 
 
 arms, average 5 specimens. 108.8 
 
 Greatest length of arm (average of 11 specimens) 80.2 
 Greatest width over facets for ceratohyals (aver¬ 
 age of 16 specimens). 37.9 
 
 Antero-posterior diameter, measured between fa¬ 
 cets for ceratohyals (average of 18 specimens) 16.8 
 
 VERTEBRAE. 
 
 The vertebral formula for Mylodon harlani is as follows: Cervicals 7, thoracics 16, 
 lumbar-sacrals 9 or 8, caudals 21. The lumbar vertebrae have ankylosed with the sacral 
 series in the North American species, as in M. robustus. 
 
 Cervical vertebrae. —In Mylodon harlani the length of the series of seven cervical 
 vertebrae (plate 27, figs. 3 and 4) in close articulation is 273 mm. This measurement 
 is actually smaller than the comparable measurement in Nothrotheriurn, and is indicative 
 of the relatively short neck which Mylodon possessed. The individual vertebrae, in 
 contrast to those in the megalonychid forms from Rancho La Brea, are decidedly com¬ 
 pressed. The antero-posterior compression is shown particularly in cervicals following 
 the axis. 
 
 The atlas of M. harlani resembles closely the corresponding vertebral segment in 
 M. garmani and differs from that in M. robustus in the same way as does the species 
 described by Allen. For example, the lateral borders of the wings may be straight, 
 though in some specimens from the asphalt the curvature corresponds to that in the 
 atlas of the South American species. On the dorsal surface this border is distinctly 
 raised. The posterior border of the lateral process does not extend in back of the artic¬ 
 ulating surface for the axis as far as in the atlas of Owen’s specimen, M. harlani and 
 M. garmani thus being similar. This border reaches farther back than in the atlas of 
 N othr other ium. 
 
 Another resemblance between the two North American mylodonts and a character 
 which separates them from M. robustus is the position of the two dorsal foramina on 
 the lateral process, as already pointed out by Allen for M. garmani. In M. harlani the 
 posterior foramen is situated considerably to the rear of the anterior depression. In M. 
 
134 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 robustus these structures are very close together. In the atlas of Scelidotherium magnum 
 shown by Winge (1915, plate 37), the two foramina are widely separated. An interest¬ 
 ing difference between Mylodon and Nothrotherium is seen in the absence of the posterior 
 superior foramen in the atlas of the latter genus. In Mylodon harlani the posterior border 
 of the arch over the neural canal is carried farther back and allows a bridge of bone to 
 arch over the dorsal furrow leading from the anterior dorsal depression to the posterior 
 opening of the vertebrarterial canal. The anterior margin of the neural arch may be 
 more deeply indented in median line in the atlas of M. harlani than in that of M. robus¬ 
 tus, but this feature is variable. 
 
 In posterior view the dorsal arch of the neural canal is seen to be thick and the 
 ventral arch apparently sags more posteriorly than in M. robustus. The facets for the 
 axis face a little more directly to the posterior side than they do in the atlas of Nothro- 
 therium. Also, external to these facets in Mylodon harlani there is a lateral expanse of 
 bone, representing a downward deflection of the lateral wing, which is more extensive 
 than in M. robustus. Such a pronounced deflection is entirely lacking in the atlas of 
 Nothr other ium. 
 
 The depression on the ventral surface 
 of the lateral wing, leading to the verte¬ 
 brarterial canal, appears as a more promi¬ 
 nent feature in the atlas of M. harlani than 
 in that of M. robustus. It is relatively less 
 extensive than in the atlas of Nothrotherium, 
 and consequently the dorsal and posterior 
 openings of the canal can not be seen in 
 the floor of the depression. The anterior 
 margin of the floor of the neural canal is 
 not deeply concave, as in Nothr otherium. 
 
 This bridge is relatively narrow and situ¬ 
 ated well posteriorly, in contrast to that 
 of Nothr otherium. Its antero-lateral mar- 
 gins do not overhang the articulating sur- if 
 
 faces for the occipital condyles as they do J J verse process; az, anterior zygapo- 
 
 in Nothrotherium. ' physi3 - Rancho La Brea Pleistocene. 
 
 The axis again resembles more closely the corresponding element in M. garmani 
 than in M. robustus. Allen states (Allen, G. M., 1913, p. 328) that the axis of M. gar¬ 
 mani “differs from that of M. robustus mainly in that the superior margin of the spinous 
 process is at a much less angle to the long axis, due to greater elevation of its anterior 
 projection .” This is true also for the axis of M. harlani. Viewed from the top, the 
 superior margin in most specimens is represented by a flattened surface anteriorly 
 which changes posteriorly to a sharp ridge. In specimens of the axis belonging to excep¬ 
 tionally large individuals the superior margin of the spinous process is formed by a very 
 broad surface extending the length of the process. The spinous process may somteimes 
 be exceptionally long antero-posteriorly, in which case it overhangs the postzyga- 
 pophyses to a considerable extent. In all the axes of M. harlani from Rancho La Brea 
 the posterior border of the spinous process, extending from the postero-dorsal angle to 
 the postzygapophyses, is concave, not convex, as in the axis of M. robustus. 
 
 With one exception, the facet on the lower side of the odontoid process is directed 
 more to the ventral side than in the axis of M. robustus. The facet itself is usually 
 convex, but may be concave in some specimens. The upper surface of the odontoid is 
 smooth, and, according to Leidy, comes in contact with the transverse ligament. In 
 M. harlani this surface is bounded posteriorly by a transverse ridge and assumes the 
 
 ter 
 
 Fig. 67 .—Mylodon harlani Owen. Cervi¬ 
 cal vertebra VII, No. 1573-6, with 
 cervical rib, No. 1737-1; anterior 
 view. X 0.33. cr, cervical rib; ccr, 
 facet for capitulum of cervical rib; ter, 
 
 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 135 
 
 character of a broad transverse groove. The lateral articulating surfaces for the atlas 
 are convex and ovate in shape. The principal diameter of the facet extends from the 
 postero-dorsal side to the antero-ventral side. In a few specimens this diameter extends 
 from the antero-dorsal to the postero-ventral side, with other axes showing degrees of 
 variation between these two extremes. In one specimen the left lateral facet for the 
 atlas is in direct connection with the facet on the ventral side of the odontoid. 
 
 The transverse process may be narrow or quite broad dorso-ventrally. It does not 
 extend well below the lower margin of the lateral articulating surface, as in Megalonyx 
 jeffersonii. The process is pierced by a large vertebrarterial canal. In Mylodon harlani 
 this canal does not open also on the ventral surface at the base of the transverse process, 
 as in Nothrotherium. The median keel of the ventral surface is prominent and broadens 
 at either end. 
 
 Behind the axis the cervical vertebrae carry slender neural spines which increase in 
 length and in antero-posterior diameter from the third to the seventh vertebrae, inclu¬ 
 sive. In the seventh vertebra the neural spine is inclined and is wide posteriorly, with 
 right and left margins of the posterior side distinct and crenulated. The neural spine 
 in the seventh vertebra of Mylodon differs, therefore, in several characters from that in 
 the corresponding segment of Nothr otherium. 
 
 Table 67. —Measurements (in millimeters) of eight specimens of atlas of Mylodon harlani. 
 
 
 No. 
 
 1567-2 
 
 No. 
 
 1567-4 
 
 No. 
 
 1567-5 
 
 No. 
 
 1567-6 
 
 No. 
 
 1567-7 
 
 No. 
 
 1567-8 
 
 No. 
 
 1567-9 
 
 No. 
 
 1567-10 
 
 Greatest transverse width across lateral wings. 
 
 211 
 
 212.8 
 
 232 
 
 225 
 
 223 
 
 237 
 
 214.7 
 
 218 
 
 Greatest antero-posterior diameter. 
 
 83 
 
 77.7 
 
 91 
 
 90.8 
 
 87.3 
 
 96.8 
 
 87.8 
 
 86.5 
 
 Antero-posterior diameter of dorsal wall of neural 
 
 
 
 
 
 
 
 
 
 canal alor g median line. 
 
 46.4 
 
 47.9 
 
 51.8 
 
 57.4 
 
 47.8 
 
 53.5 
 
 51 
 
 46.4 
 
 Antero-posterior diameter of ventral wall of neural 
 
 
 
 *38.3 
 
 
 
 
 
 
 canal along median line. 
 
 33.2 
 
 33.3 
 
 43.9 
 
 41.4 
 
 40.7 
 
 37.4 
 
 38.4 
 
 Distance from antero-external border to lateral 
 
 
 
 
 
 
 
 
 
 facet for axis. 
 
 83.6 
 
 90.8 
 
 92 
 
 84.7 
 
 90 
 
 94.8 
 
 85.9 
 
 85.5 
 
 Greatest transverse distance between posterior 
 
 fill 
 
 
 
 
 
 
 
 
 borders of facets for axis. 
 
 93 
 
 98.8 
 
 89.4 
 
 98.2 
 
 102.3 
 
 95 
 
 98 
 
 Least transverse distance between anterior borders 
 
 
 
 
 
 
 
 
 
 of facets for axis. 
 
 55.5 
 
 39 
 
 52.3 
 
 41.3 
 
 43.8 
 
 46.3 
 
 43.3 
 
 46.2 
 
 Greatest antero-posterior diameter of lateral process. 
 
 77.7 
 
 80 
 
 84.6 
 
 S3.6 
 
 97.8 
 
 87 
 
 81 
 
 78 
 
 1 Approximate. 
 
 The lateral wing of the third cervical is broad dorso-ventrally, with the upper outer 
 end projecting backward as a distinct process. In the fourth vertebra the anterior 
 (inner) and posterior (outer) ends of the lower portion of the wing-like process tend 
 to thicken and to separate. This separation becomes more pronounced in the fifth 
 cervical, where the anterior (inner) end projects as a flattened process, with long axis 
 parallel to the fore-and-aft diameter of the vertebra, and representing apparently the 
 inferior lamella. When the fifth vertebra is viewed from the anterior side, the inferior 
 lamella is seen to extend well below the ventral surface of the centrum. In the fourth 
 vertebra, on the contrary, the inferior lamella, as seen in front view, extends only slightly 
 beneath that surface. In the sixth cervical the inferior lamella extends farther forward 
 beyond the anterior face of the centrum than in preceding vertebrae. In the sixth 
 cervical, also, the connection between inferior lamella and the outer lower portion of 
 the lateral process that incloses the vertebrarterial canal on the lower outer side is 
 quite narrow. In the preceding vertebrae this connection is broad. 
 
 The seventh cervical vertebra is relatively simple in structure of lateral process, but 
 may show some interesting characters. As a rule, only a heavy lateral process is present, 
 which projects slightly downward and may fuse with a rudimentary cervical rib. When 
 
136 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 such fusion takes place the capitulum disappears entirely and the vertebrarterial canal 
 remains open below, or the capitulum remains and a vertebrarterial foramen is formed 
 comparable to that in preceding vertebrae. Specimen 1573-6 (fig. 67) is interesting in 
 this connection, for a cervical rib actually articulates with this vertebra on the right 
 side. Facets on the left side demonstrate the presence of the opposite rib, also with 
 distinct capitulum and tuberculum. In the seventh cervical the inferior lamella is far 
 more reduced than in the corresponding vertebra of Nothrotherium. As noted by Allen 
 (1913, p. 328) for M. garmani, this vertebra possesses a large facet on each side of the 
 lateral border of the posterior face of the centrum. In specimens from Rancho La 
 Brea this facet may be concave or flat and articulates with the capitulum of the first 
 costal rib. One specimen has been noted in which the facet was absent on one side. 
 
 In the series of cervical vertebrae from the third to the seventh inclusive the centra 
 gradually broaden transversely and flatten dorso-ventrally. The median keel on the 
 ventral surface of the centrum, most pronounced on the axis, tends to widen posteriorly 
 in these vertebrae. It practically disappears in the fifth cervical. 
 
 Table 68 .— Measurements (in millimeters ) of six specimens of axis of Mylodon harlani. 
 
 
 No. 
 
 1568-1 
 
 No. 
 
 1568-2 
 
 No. 
 
 1568-4 
 
 No. 
 
 1568-5 
 
 No. 
 
 1568-6 
 
 No. 
 
 1568-7 
 
 Greatest length along median line of ventral surface. 
 
 88 
 
 86.8 
 
 84.4 
 
 92.3 
 
 82 
 
 87.6 
 
 Greatest height. 
 
 134 
 
 141.3 
 
 139 
 
 143 
 
 138.9 
 
 141.2 
 
 Least width behind articulating surfaces for atlas. 
 
 83.3 
 
 87.9 
 
 76 
 
 88 
 
 84 
 
 85.6 
 
 Dorso-ventral diameter of centrum across posterior surface. 
 
 45.8 
 
 49.9 
 
 45 
 
 48.9 
 
 49.9 
 
 46.5 
 
 Greatest width of centrum. 
 
 58.4 
 
 56.7 
 
 53 
 
 62.7 
 
 61.5 
 
 59.6 
 
 Greatest transverse diameter of neural canal at anterior end. 
 
 61.4 
 
 49 
 
 46 
 
 49.8 
 
 45 
 
 48.6 
 
 Greatest width across posterior ends of transverse processes. 
 
 Distance from anterior end of dorsal spine to posterior end of neural 
 
 163.4 
 
 153.2 
 
 141.4 
 
 146.8 
 
 163.2 
 
 155.9 
 
 arch. 
 
 78.2 
 
 70 
 
 67.9 
 
 81.7 
 
 69 
 
 71.7 
 
 Width across posterior zygapophyses. 
 
 Least distance from anterior border of neural canal to border of notch 
 
 81.1 
 
 80.3 
 
 78.7 
 
 95 
 
 89.3 
 
 96.1 
 
 below posterior zygapophyses. 
 
 25.6 
 
 27.3 
 
 24.3 
 
 29.2 
 
 25.2 
 
 26.3 
 
 Greatest width across outer ends of lateral facets for atlas. 
 
 Distance from anterior border of lateral facet for atlas to posterior end 
 
 100.6 
 
 100.9 
 
 85.8 
 
 102 
 
 97.3 
 
 102 
 
 of transverse process. 
 
 107.7 
 
 97.4 
 
 83.3 
 
 99.4 
 
 101.8 
 
 104.5 
 
 Thoracic vertebrae. —The first thoracic vertebra differs noticeably from that in Nothro¬ 
 therium in possessing a slender spine without terminal enlargement. The spine is inclined 
 backward and is approximately twice the length of that in the last cervical vertebra. 
 The anterior zygapophyses are concave upward (plate 28, fig. 8), while the posterior 
 zygapophyses are flat or slightly concave downward. The centrum is relatively shorter 
 than in the first thoracic of Nothrotherium. The facet for the capitulum of the first rib is 
 situated lower in the vertebra of Mylodon , extending nearly to the ventral border. The 
 facet for the capitulum of the second rib is situated well along the lateral border of the 
 posterior face. Nutrient foramina pierce both the dorsal and ventral surfaces of the 
 centrum. 
 
 In the second thoracic the spine is slightly longer and wider than that in the first 
 vertebra. The dorsal border of the spine meets the anterior border to form a distinct 
 angle, in which respect this segment differs from the first. The centrum is still com¬ 
 pressed dorso-ventrally; the flattened ventral surface is, however, not as wide as in the 
 preceding vertebra. There is a noticeable decrease in the width across the transverse 
 processes. A curious feature is seen on the right side of all the specimens of this vertebra 
 examined from Rancho La Brea. The pedicle of the neural arch on this side of the ver¬ 
 tebra is slender and the intervertebral space behind it, forming the exit for the spinal 
 nerve, is very large. On the left side the intervertebral space is of normal size. A 
 slender right pedicle and an enlarged intervertebral space may also occur in the third 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 137 
 
 thoracic vertebra. There appear to be some specimens, however, in which these struc¬ 
 tures are comparable in their development to those of the left side and are of the normal 
 type. 
 
 A nearly complete series of thoracic vertebrae, presumably belonging to a single 
 individual, is shown on plate 28, figure 1. The vertebrae have been tentatively deter¬ 
 mined as representing the third to fourteenth inclusive. In this series is clearly shown 
 the change in the size and shape of the dorsal spines and the change in the size, shape, 
 and position of the facets articulating with the ribs. The gradual development of the 
 metapophyses should be also noted. The centra have concave sides and gradually 
 increase in depth posteriorly. In each vertebra at least two nutrient foramina perforate 
 the dorsal surface, while the ventral surface is pierced by similar foramina. The change 
 in the shape of the anterior face of the centrum and in the size and shape of the neural 
 canal can be seen in plate 28, figures 2, 3, 4, and 8. With the twelfth vertebra the ventral 
 side of the centrum begins to flatten, and this character becomes more noticeable in the 
 following thoracic segments. In the fourteenth the ventral surface is bounded on each 
 side by a ridge. Several vertebrae in the middle portion of the thoracic series are 
 uniquely characterized by a facet on the anterior and on the posterior side of the dorsal 
 spine. These facets vary in size and shape and permit an articulation of the spines of 
 the vertebrae. Such surfaces have been noted by Allen in thoracic vertebrae of Mylo¬ 
 don garmani. The anterior facet is situated at the flattened base of the spine, while the 
 posterior facet lies frequently between the posterior zygapophyses and the middle of the 
 posterior side. 
 
 Table 69. —Measurements (in millimeters ) of cervical vertebrae of Mylodon harlani , series 1569-8 
 
 to 1573-3 inclusive. 
 
 Number in cervical series. 
 
 3 
 
 4 
 
 5 
 
 6 
 
 7 
 
 Length of centrum measured over ventral surface. 
 
 38 
 
 37.5 
 
 37 
 
 38.6 
 
 42.2 
 
 Width across centrum measured over anterior face and between inner borders 
 of vertebrarterial canals. 
 
 75.3 
 
 76.7 
 
 77.2 
 
 81.3 
 
 94.2 
 
 Depth of centrum measured over anterior face and normal to dorsal surface. . . 
 
 50.3 
 
 51.1 
 
 49.5 
 
 47 
 
 46.2 
 
 Height from ventral border of posterior face of centrum to tip of neural spine. . 
 Width across outer sides of anterior zygapophyses. 
 
 136.5 
 
 114 
 
 138 
 
 *131 
 
 133.7 
 
 146.8 
 
 138.9 
 
 Greatest width across posterior zygapophyses. 
 
 91.7 
 
 93 
 
 92 
 
 94 
 
 *95 
 
 Greatest width across posterior ends of ventral wings. 
 
 166 
 
 175.6 
 
 183 
 
 194.5 
 
 202 
 
 Length of ventral wing. 
 
 45 
 
 64.4 
 
 63.8 
 
 59.5 
 
 43.3 
 
 1 Approximate. 
 
 In the last four vertebrae of the complete thoracic series of Mylodon harlani the 
 zygapophyses undergo an interesting change from the simple type of the anterior region 
 of the trunk. In the first of these four vertebrae the anterior zygapophyses are broad, 
 flat, articulating surfaces, situated close to the median line and at the base of the dorsal 
 spine. The posterior zygapophyses may be represented also by two broad surfaces. In 
 some individuals, however, the posterior zygapophyses may occur as two pairs of facets, 
 in which case two large surfaces having a posterior position form the medial pair, while 
 two small articulating surfaces form the lateral and anterior pair. Moreover, individuals 
 are known in which a large and a small facet may be present on one side of the median 
 line, while on the opposite side only a single large facet occurs. 
 
 In the second of the four vertebrae the anterior zygapophyses form articulating 
 surfaces to correspond with those forming the posterior zygapophyses of the preceding 
 vertebra. The outer facet, when present, is not on the same plane with the inner or 
 medial facet, but is deflected downward. The posterior zygapophyses form two distinct 
 pairs of articulating surfaces. On each side the inner or medial facet is situated at the 
 base of the dorsal spine and a small extent of its surface is directed downward, while 
 
138 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 the major portion is directed outward and downward. In some specimens this facet is 
 directed downward for the most part. The outer or lateral facet is situated on the 
 upper posterior side of the pedicle of the neural arch and is directed downward and 
 inward. 
 
 In the third of the four vertebrae (see fig. 68) the medial facet ( az 2 ) is usually 
 separated from the lateral facet (az 1 ) by a strong ridge extending forward from the base 
 of the metapophysis (m). The medial facet is concave upward, with the greater portion 
 of the surface directed inward. The lateral facet is flat and is directed upward and 
 outward. The posterior zygapophyses (see fig. 68) somewhat resemble those in the 
 preceding vertebra. The medial facet ( pz 2 ) is bent sharply upward along the side of 
 the dorsal spine. This facet is separated by a deep notch from the lateral facet (pz 1 ). 
 
 In the last vertebra of the series, presumably the sixteenth of the entire thoracic 
 series, the arrangement of the zygapophyses is similar to that in the preceding vertebra. 
 
 The posterior articulating surfaces are 
 rugose and frequently ankylose the verte¬ 
 bra to the sacrum. 
 
 Lumbar and sacral vertebrae. —The 
 sacrum of Mylodon is a complex structure 
 formed by the fusion of vertebrae in the 
 posterior region of the trunk with the true 
 sacral segments. In this respect Mylodon 
 differs considerably from Nothrotherium. 
 In Mylodon robustus, as described by 
 Owen, the sacrum consists of 7 sacral 
 vertebrae ankylosed with 3 lumbar verte¬ 
 brae. In Owen’s specimen the last tho¬ 
 racic is also attached to the lumbar series. 
 
 Fig. 68 .—Mylodon harlani Owen. Thoracic vertebra XV, 
 No. 1588-4, anterior and posterior views of right half. 
 X 0.33. az 1 , lateral anterior zygapophysis; az" 1 , medial 
 anterior zygapophysis; pz 1 , lateral posterior zygapophy¬ 
 sis; pz~, medial posterior zygapophysis; m, metapophy¬ 
 sis; tp, transverse process; cr, facet for capitulum 
 of rib; tr, facet for tuberculum of rib. Rancho La 
 Brea Pleistocene. 
 
 In many of the specimens from Rancho La Brea, 9 vertebrae form the sacrum. 
 Thus the lumbar-sacral series in M. harlani lacks 1 vertebra present in M. robustus. 
 Several specimens from the asphalt have been noted in which only 8 vertebrae occur 
 in the sacrum. This is characteristic of No. 1719-25. In many sacra with 9 vertebrae 
 the segments are completely fused and the neural spines form a continuous ridge over 
 the series. 
 
 The 3 anterior vertebrae in the sacrum apparently represent the lumbar series. The 
 centra are deep and have their ventral surfaces perforated by large nutrient foramina. 
 In the first vertebra the transverse process may not be completely fused with the ante¬ 
 rior inner margin of the ilium and large openings may persist in this region. The ante¬ 
 rior zygapophyses are grouped in two pairs, as in the last thoracic vertebra. The neural 
 canal increases in size throughout the series. The exits for the lumbar nerves are large. 
 Extending obliquely downward and forward across the lateral face of the centrum from 
 each of these openings is a well-marked groove. The last pair of nervous foramina 
 through the neurapophysial plate of the lumbar region has the characters of the fourth 
 pair in M. robustus, as described by Owen (1842, p. 65). The 5 pairs of foramina for 
 the sacral nerves are also similar to those in M. robustus. In sacra having 8 vertebrae, 
 4 pairs of foramina for the sacral nerves are present. The neural canal continues to 
 
Table 70 .—Measurements {in millimeters) of thoracic vertebrae of Mylodon harlani. 
 
 PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA 
 
 139 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 21 
 
 
 Cl 
 
 
 05 
 
 • co 
 
 
 CO 
 
 CO 
 
 
 CO 
 
 
 to 
 
 to 
 
 
 
 
 I- 
 
 
 X 
 
 • X 
 
 
 to 
 
 1-H 
 
 
 CO 
 
 
 
 to 
 
 
 
 
 
 
 
 • l-H 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 *o 
 
 
 CO 
 
 
 CO 
 
 
 
 CO 
 
 
 
 CO 
 
 »o 00 
 
 
 CO 
 
 05 O 
 
 
 o 
 
 Cl 
 
 o 
 
 X 
 
 
 Tt< 
 
 o 
 
 
 
 CO 
 
 CO 
 
 
 X 
 
 rH 05 
 d »-H 
 
 
 CO 
 
 rH 
 
 co 
 
 Cl 
 
 
 TjH 
 
 to 
 
 
 
 
 Cl 
 
 
 Cl 
 
 
 
 
 Cl 
 
 
 Cl 
 
 
 
 to 
 
 • 
 
 Cl 
 
 CO 
 
 CO 
 
 
 »o 
 
 r-H (X 
 
 
 05 
 
 CO 
 
 05 
 
 to 
 
 
 X 
 
 
 
 
 CO 
 
 CO 
 
 
 X 
 
 r-H 05 
 Cl ^H 
 
 
 to 
 
 1-H 
 
 Cl 
 
 Cl 
 
 
 CO 
 
 to 
 
 
 
 
 Cl 
 
 
 
 
 
 
 rH 
 
 Cl 
 
 to 
 
 
 
 
 
 r-H 
 
 
 Cl 
 
 
 r-H 
 
 to o 
 
 
 X 
 
 to 
 
 CO 
 
 co 
 
 
 o 
 
 Cl 
 
 
 rH 
 
 
 CO 
 
 
 X 
 
 o o 
 
 Cl Cl 
 
 
 to 
 
 H 
 
 Cl 
 
 Cl 
 
 
 
 to 
 
 
 
 
 
 
 
 eo 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 to 
 
 
 
 
 
 
 
 
 o 
 
 
 
 
 
 Hff iO 
 
 
 X 
 
 rH 
 
 Cl 
 
 H 
 
 
 
 C5 
 
 
 
 
 
 
 
 05 O 
 1-H Cl 
 
 
 to 
 
 Cl 
 
 Cl 
 
 Cl 
 
 
 rf 
 
 Tti 
 
 CO 
 
 
 
 
 
 
 
 
 
 CO 
 
 
 
 
 
 to 
 
 
 05 
 
 . 
 
 
 
 
 • 05 
 
 
 X 
 
 
 
 o 
 
 
 
 F- 
 
 00 
 
 
 • 
 
 
 
 
 • 05 
 
 
 to 
 
 i-H 
 
 
 Cl 
 
 
 
 -V 
 
 »o 
 
 
 
 
 
 
 * 1-H 
 
 
 
 
 
 
 
 
 
 H 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 o 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 H3 
 
 
 
 
 
 CO 
 
 • 
 
 
 to 
 
 
 
 co 
 
 
 
 to 
 
 CO 
 
 00 
 
 CM 
 
 05 
 
 
 to 
 
 • o 
 
 
 
 
 
 H 
 
 
 
 r-H 
 
 CO 
 
 I- 
 
 CO 
 
 CO 
 
 to 
 
 
 L- 
 
 • 05 
 
 • r-H 
 
 
 to 
 
 
 
 Cl 
 
 
 
 to 
 
 to 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 cu 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 05 
 
 
 to 
 
 Cl Cl 
 
 
 CO 
 
 CO 1> 
 
 co 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 r-H 
 
 
 1^ 
 
 o 
 
 00 
 
 
 Cl 
 
 »o r - 
 
 
 CO I - 
 
 o 
 
 co 
 
 
 
 to 
 
 
 
 to 
 
 to 
 
 
 
 X X 
 
 r-H i-H 
 
 
 to 
 
 rH 
 
 Cl 
 
 Cl 
 
 
 
 
 
 
 
 CO 
 
 
 o 
 
 co 
 
 
 to 
 
 to 
 
 CO 
 
 
 
 
 CO 
 
 
 CO 
 
 
 00 
 
 
 o 
 
 rH CO 
 
 
 
 CO 
 
 Cl 
 
 Cl 
 
 
 H 
 
 CO 
 
 
 to 
 
 to 
 
 
 l> 
 
 05 X 
 
 
 to 
 
 rH 
 
 Cl 
 
 Cl 
 
 
 
 
 
 
 
 
 
 
 rH rH 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 CO 
 
 
 
 rt< 
 
 CO 
 
 to 
 
 
 
 
 Cl 
 
 
 iO 
 
 to 
 
 C5 
 
 
 t> 
 
 Cl co 
 
 
 CO 
 
 X 
 
 to 
 
 CO 
 
 
 o 
 
 o 
 
 
 
 to 
 
 to 
 
 
 CO 
 
 05 
 
 i-H »-H 
 
 
 Tf 
 
 i-H 
 
 Cl 
 
 Cl 
 
 
 
 CO 
 
 
 
 to 
 
 Cl 
 
 
 
 
 
 X 
 
 CO 
 
 
 00 
 
 
 
 
 
 
 
 CO 
 
 
 CO 
 
 CO TjH 
 
 
 CO 
 
 Cl 
 
 to 
 
 CO 
 
 
 CO 
 
 co 
 
 
 to 
 
 to 
 
 
 CO 
 
 05 L- 
 
 
 
 Cl 
 
 Cl 
 
 Cl 
 
 
 co 
 
 Cl 
 
 
 
 
 
 
 
 rH rH 
 
 
 
 
 
 
 
 
 
 
 
 
 CO 
 
 
 1> 
 
 Cl 
 
 
 CO 
 
 Cl 
 
 Cl 
 
 CO 
 
 
 
 
 
 co 
 
 to 
 
 Cl 
 
 
 CO 
 
 o 
 
 
 Cl 
 
 05 
 
 CO 
 
 TfH 
 
 
 Cl 
 
 C5 
 
 
 to 
 
 to 
 
 
 CO 
 
 r-H 
 
 
 
 i-H 
 
 Cl 
 
 Cl 
 
 
 
 rH 
 
 
 
 co 
 
 
 
 
 Cl H 
 
 co 
 
 
 
 
 
 
 
 
 
 CO 
 
 1 
 
 
 Cl 
 
 05 
 
 
 Cl 
 
 Cl 
 
 
 
 Cl 
 
 
 
 
 
 
 tO 
 
 <N 
 
 Cl 
 
 00 
 
 
 o 
 
 CO C l 
 
 
 o 
 
 05 
 
 i-H 
 
 
 
 
 
 »o 
 
 
 
 o 
 
 Cl 1^ 
 
 
 
 i-H 
 
 CO 
 
 
 
 
 
 
 
 CO 
 
 co 
 
 
 
 Cl H 
 
 
 
 
 
 
 
 
 
 r-H 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 CO 
 
 1 
 
 
 t'- 
 
 05 
 
 
 X 
 
 Cl 
 
 
 l> 
 
 co 
 
 Cl 
 
 
 
 
 
 -T 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 I- 
 
 rH 
 
 CO 
 
 X 
 
 
 Cl 
 
 X to 
 
 
 on 
 
 
 •-H 
 
 
 
 
 
 tO 
 
 
 
 
 
 CO 
 
 -H C5 
 
 
 CO 
 
 rH 
 
 Cl 
 
 
 
 
 
 rH 
 
 
 
 
 
 
 Cl 1-H 
 
 
 
 
 
 
 
 
 
 
 
 
 © 
 
 Fh 
 
 • rH 
 
 
 43 
 
 
 
 
 
 
 • T3 
 
 
 
 
 
 © 
 
 . © 
 
 
 d 
 
 
 
 • © 
 
 
 o 
 
 • d 
 
 
 
 
 
 
 > 
 
 . d 
 
 • ?! 
 
 
 
 © 
 
 © © 
 
 
 ^H 
 
 
 
 
 
 
 
 n 
 
 . Fh 
 
 . © 
 
 
 
 
 a ca 
 
 
 
 
 
 
 
 
 
 
 . o 
 
 co 
 
 d 
 
 
 'd 
 
 •rH «HH 
 
 
 © 
 
 © 
 
 ' © 
 
 
 
 
 
 o 
 
 Tj 
 
 . Xt 
 
 . o 
 
 d 
 
 
 
 a f . 
 
 
 • © 
 
 
 
 
 
 
 
 . —. 
 
 . © 
 
 d 
 
 
 
 o 
 
 
 d 
 
 . H 
 
 
 
 
 
 
 Fh 
 
 . d 
 
 • o 
 
 © 
 
 
 s 
 
 rt 
 
 £ « 
 o a 
 
 
 
 . © 
 
 
 
 
 
 43 
 
 d 
 
 d 
 
 M 
 
 © 
 
 d 
 
 CO 
 
 d 
 
 © 
 
 s 
 
 • 3 
 
 • c 
 
 ' aj 
 : > 
 
 . Fh 
 
 . a 
 
 • © 
 
 • CO 
 
 • Fh 
 
 3 
 
 Fh 
 
 
 -H 
 
 d 
 
 © 
 
 d 
 
 
 tin 
 
 o 
 
 tH 
 
 © 
 
 . rQ 
 
 ; -+-> 
 
 . t— 
 
 . o 
 
 
 o 
 
 CO 
 
 'V 
 
 © 
 
 CO 
 
 o 
 
 a 
 
 > 
 
 o 
 
 d 
 
 © 
 
 a 
 
 g 
 
 . d 
 • ^d 
 
 'd 
 
 . © 
 
 • > 
 
 • CO 
 
 • d 
 
 • d 
 
 — * Fh 
 © 4-> 
 
 .2 w 
 a g 
 ” 2 
 
 — a 
 
 d 
 
 *4H 
 
 o 
 
 
 *E 
 
 CO 
 
 09 
 
 0.2 
 •o ^ 
 
 C u 
 
 
 © 
 
 2 
 
 _d 
 
 ; co 
 
 -H 
 • © 
 
 • © 
 
 
 £ 
 
 .2 
 
 'V 
 
 © 
 
 Vl 
 
 0 
 
 u 
 
 43 
 
 d 
 
 © 
 
 © 
 
 «*H 
 
 Fh 
 
 d 
 
 CO 
 
 d 
 
 © 
 
 2 
 
 43 
 
 d 
 
 © 
 
 © 
 
 *«H 
 
 o 
 
 : a 
 : s 
 
 • o 
 
 • Fh 
 
 . «H 
 
 F- 
 
 © 
 
 -H> 
 
 © 
 
 6 
 
 d 
 
 
 o 
 
 H 
 
 O 
 
 T3 
 
 O 
 
 © o 
 
 o s 
 
 S o 
 a ft 
 
 rd 
 
 ’V 
 
 "d 
 
 'd 
 
 (H 
 
 -H 
 
 d 
 
 © 
 
 • d 
 
 • %H 
 
 • c 
 
 • © 
 
 • © 
 
 rQ ^ 
 
 rd -+-> 
 
 -Q 
 
 
 d 
 
 u 
 
 o 
 
 o 
 
 rd 
 
 +3 
 
 a 
 
 2 
 
 rd 
 
 43 
 
 S 5 
 
 $ 
 
 o ^ 
 
 d 
 
 T3 
 
 rd 
 
 ^d 
 
 • rH 
 
 S o 
 .2 « 
 
 *4H 
 
 o 
 
 © 
 
 > 
 
 1 
 
 6 
 
 ^H © 
 «HH r^ 
 
 o © 
 
 V 
 
 <H 
 
 o 
 
 
 rd 
 
 43 
 
 fl 
 
 tc 
 
 C d 
 
 o © 
 rd O 
 
 d 
 
 d 03 
 
 ^ d 
 
 Fh © 
 
 T3 -a 
 
 ■g'S 
 
 d 
 
 *H 
 
 43 
 
 d 
 
 d 
 
 © 
 
 s- 
 
 © 
 
 ■Ofl 
 S 03 
 
 ~© 
 
 © 
 
 o 
 
 TJ 
 
 a 1 s 
 
 3 if 3 
 
 
 © 
 
 -O 
 
 43 
 
 c o 
 © 
 
 o 
 
 rd 
 
 43 
 
 to 
 
 © 
 
 43 
 
 .y a 
 © ^ 
 
 ft-S 
 
 s s 
 
 © 
 
 > 
 
 i 
 
 o 
 
 .2 
 
 © 
 
 (H 
 
 © 
 
 r* 
 
 CO 
 
 Sh ^ 
 
 £ ® 
 £ 5 
 
 JJ 
 
 d 
 
 75 
 
 © 
 
 3 .2 
 
 ."d'd 
 
 3 
 
 *a 
 
 
 s 
 
 d 
 
 d 
 
 © 
 
 Fh 
 
 a 
 
 © 
 
 d 
 
 © 
 
 n 
 
 d .£P 
 
 d q 
 
 o d 
 ° 2 
 
 ou 
 
 Fh 
 
 o 
 
 d 
 
 d 
 
 d 
 
 3 
 
 Sh 
 
 d d 
 d © 
 
 *H *H 
 
 o 
 
 d 
 
 © 
 
 f-1 
 
 c3 2 
 
 d 
 
 © 
 
 
 
 OQO 
 
 w 
 
 0 
 
 Q 
 
 
 H 
 
 H O 
 
 
 o 
 
 vJ 
 
 
 rH 
 
 I 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 05 
 
 X 
 
 ‘O 
 
 CO 
 
 
 Cl 
 
 X 
 
 
 X 
 
 
 
 
 
 
 
 
 
 
 
 CO 
 
 X 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 eo 
 
 
 
 
 
 
 
 
 
 
 rH 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 X 
 
 to 
 
 CO 
 
 Cl 
 
 / 
 
 
 
 
 tO 
 
 O 
 
 
 
 
 
 
 
 X 
 
 to 
 
 rH 
 
 r- 
 
 co 
 
 Cl 
 
 o 
 
 
 X 
 
 
 X 
 
 CO 
 
 o 
 
 
 
 
 
 
 
 CO 
 
 X t- 
 
 
 (£ 
 
 Cl 
 
 
 CO 
 
 Cl 
 
 X 
 
 
 
 
 
 
 
 
 t> 
 
 
 
 
 X 
 
 
 X 
 
 
 CO 
 
 
 to 
 
 
 
 
 
 to 
 
 Cl 
 
 CO 
 
 X 
 
 
 to 
 
 
 
 X 
 
 X 
 
 
 1 _ 
 
 
 CO 
 
 05 
 
 
 r- 
 
 X I- 
 
 
 — 
 
 
 to 
 
 Cl 
 
 X 
 
 
 Cl 
 
 
 -4 
 
 
 X 
 
 
 
 
 
 
 Cl 
 
 
 
 
 
 
 
 
 
 
 X 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ‘0 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 i-H 
 
 
 CO 
 
 Cl 
 
 rH 
 
 
 CO 
 
 
 05 
 
 X 
 
 
 
 
 
 CO 
 
 X 
 
 o 
 
 rt< 
 
 CO 
 
 CO 
 
 TfH 
 
 
 X 
 
 Tf 
 
 05 
 
 pH 
 
 05 
 
 
 o 
 
 
 CO 
 
 to 
 
 
 
 CO 
 
 
 t>» 
 
 
 X 
 
 X 
 
 to 
 
 Cl 
 
 X 
 
 
 X 
 
 
 tO 
 
 to 
 
 
 
 
 
 
 
 Cl 
 
 rH 
 
 
 
 
 
 
 
 
 
 1 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 X 
 
 to 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 05 
 
 
 X 
 
 
 
 X 
 
 Cl 
 
 l> 
 
 05 
 
 
 05 
 
 
 to 
 
 
 IT* 
 
 CO 
 
 to 
 
 Cl 
 
 Cl 
 
 
 
 O b- 
 
 i-H 
 
 
 
 T+1 
 
 
 -r 
 
 Cl 
 
 
 
 CO 
 
 I- 
 
 l- 
 
 
 X 
 
 X 
 
 to 
 
 Cl 
 
 X 
 
 
 Cl 
 
 
 -r 
 
 to 
 
 
 
 
 
 
 
 Cl 
 
 rH 
 
 
 
 
 
 
 
 
 
 
 
 CO 
 
 X 
 
 CO 
 
 
 
 
 X 
 
 to 
 
 
 
 05 
 
 
 rp 
 
 
 
 o 
 
 05 
 
 X 
 
 05 
 
 
 
 1- 
 
 X 
 
 X 
 
 
 
 Cl 
 
 
 CO 
 
 CO 
 
 
 
 to 
 
 co 
 
 o 
 
 
 
 X 
 
 to 
 
 Cl 
 
 
 
 Cl 
 
 
 
 to 
 
 
 
 
 
 
 
 
 iH 
 
 
 
 
 
 
 
 
 
 
 
 CO 
 
 to 
 
 to 
 
 
 
 Cl 
 
 
 X 
 
 
 
 05 
 
 
 05 
 
 05 
 
 
 05 
 
 co 
 
 
 o 
 
 
 Cl 
 
 CO 
 
 CO 
 
 X 
 
 
 
 X 
 
 
 
 o 
 
 1 
 
 CO 
 
 X 
 
 
 CO 
 
 l- 
 
 CO 
 
 
 Cl 
 
 X 
 
 to 
 
 Cl 
 
 
 
 rH 
 
 
 X 
 
 CO 
 
 
 
 
 
 
 Cl 
 
 rH 
 
 
 
 
 
 
 
 
 
 to 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 TP 
 
 X 
 
 
 
 ‘O 
 
 Cl 
 
 to 
 
 to 
 
 
 
 
 
 ■7 1 
 
 
 
 X 
 
 I- 
 
 X 
 
 to 
 
 
 
 CO 
 
 
 rH 
 
 05 
 
 
 CO 
 
 
 ^H 
 
 o 
 
 
 to 
 
 CO 
 
 co 
 
 
 c 
 
 X 
 
 to 
 
 Cl 
 
 Cl 
 
 
 Cl 
 
 
 
 CO 
 
 1 
 
 
 C*5 
 
 C*5 
 
 
 
 Cl 
 
 rH 
 
 
 
 
 
 
 
 
 
 05 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 I- 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 to 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 i-H 
 
 o 
 
 
 
 
 
 
 X 
 
 X 
 
 »o 
 
 X 
 
 
 
 
 
 
 Th 
 
 
 
 CO 
 
 X 
 
 Cl 
 
 
 to 
 
 
 X 
 
 o 
 
 X 
 
 
 -V 
 
 
 Cl 
 
 Cl 
 
 
 
 CO 
 
 l> 
 
 CO 
 
 
 H 
 
 X 
 
 to 
 
 Cl 
 
 Cl 
 
 
 Cl 
 
 
 
 to 
 
 
 
 
 
 
 
 Cl 
 
 FH 
 
 
 
 
 
 
 
 
 
 
 
 
 H* 
 
 X 
 
 
 
 Cl 
 
 
 to 
 
 
 
 i> 
 
 
 
 w*5 
 
 
 CO 
 
 CO 
 
 X 
 
 Cl 
 
 
 CO 
 
 CO 
 
 to 
 
 o 
 
 
 
 X 
 
 
 CO 
 
 
 
 
 to 
 
 CO 
 
 CO 
 
 
 o 
 
 X 
 
 to 
 
 Cl 
 
 X 
 
 
 Cl 
 
 
 Cl 
 
 
 
 
 
 
 
 
 Cl 
 
 H 
 
 
 
 
 
 
 
 
 
 
 
 TP 
 
 
 X 
 
 
 to 
 
 to 
 
 Cl 
 
 
 
 
 05 
 
 
 X 
 
 X 
 
 TfH 
 
 to 
 
 05 
 
 05 
 
 H 
 
 
 Cl 
 
 l> 
 
 o 
 
 to 
 
 Cl 
 
 
 r- 
 
 
 X 
 
 X 
 
 
 to 
 
 CO 
 
 CO 
 
 
 Cl 
 
 
 to 
 
 Cl 
 
 X 
 
 
 Cl 
 
 
 X 
 
 1-H 
 
 1 
 
 X 
 
 
 
 
 
 
 Cl 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 rH 
 
 
 
 X 
 
 
 
 
 X 
 
 
 Cl 
 
 to 
 
 
 X 
 
 
 rH 
 
 Cl 
 
 o 
 
 
 Cl 
 
 Cl 
 
 o 
 
 
 X 
 
 Cl 
 
 to 
 
 X 
 
 o 
 
 
 Cl 
 
 
 CO 
 
 05 
 
 
 CO 
 
 CO 
 
 CO 
 
 
 Cl 
 
 
 
 Cl 
 
 X 
 
 
 Cl 
 
 
 X 
 
 H 
 
 1 
 
 
 
 
 
 
 Cl 
 
 rH 
 
 
 
 
 
 
 
 
 
 CO 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 to 
 
 i-H 
 
 
 
 05 
 
 X 
 
 
 to 
 
 
 
 to 
 
 l> 
 
 
 
 
 Cl 
 
 CO 
 
 CO 
 
 Cl 
 
 to 
 
 
 
 Cl 
 
 to 
 
 Cl 
 
 rH 
 
 Cl 
 
 
 o 
 
 
 
 X 
 
 -r 
 
 •o 
 
 to 
 
 to 
 
 
 X 
 
 1^ 
 
 TtH 
 
 Cl 
 
 X 
 
 
 X 
 
 
 Cl 
 
 i-H 
 
 
 
 n 
 
 CO 
 
 
 
 Cl 
 
 i-H 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 CO 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 H 
 
 
 «-TH 
 
 
 
 
 
 
 
 © 
 
 o 
 
 
 
 
 
 
 o 
 
 
 O 
 
 . d 
 
 
 
 
 
 
 0 
 
 d 
 
 a 
 
 o 
 
 
 
 © 
 
 
 
 JLs 
 
 3 
 
 
 2 
 
 . o 
 
 
 
 
 
 
 
 H 
 
 
 
 
 
 F- 
 
 
 ~ 
 
 ■ 
 
 
 
 
 
 
 o 
 
 o 
 
 
 
 Fh 
 
 o 
 
 
 
 © 
 
 
 d 
 
 • a 
 
 
 
 
 
 
 •— 
 
 © 
 
 H—> 
 
 
 
 d 
 
 
 
 d 
 
 H 
 
 
 "a 
 
 . © 
 
 
 
 
 
 © 
 
 o 
 
 d 
 
 d 
 
 d 
 
 © 
 
 — 
 
 Fh 
 
 
 
 -4—' 
 
 d 
 
 
 
 Fh 
 
 o 
 
 «HH 
 
 
 © 
 
 Fh 
 
 • 3 
 
 
 
 
 
 u 
 
 F-i 
 
 © 
 
 > 
 
 o 
 
 rO 
 
 • 
 
 to 
 
 © 
 
 m 
 
 CO 
 
 © 
 
 -H 
 
 d 
 
 d 
 
 d 
 
 © 
 
 0 
 
 *a 
 
 © 
 
 © 
 
 d 
 
 
 «HH 
 
 © 
 
 • jj 
 
 ’ 3 
 
 • Fh 
 
 
 
 
 
 
 
 d 
 
 
 d 
 
 
 CO 
 
 
 
 © 
 
 • © 
 
 
 
 
 
 d 
 
 © 
 
 Fh 
 
 -H 
 
 • 
 
 o 
 
 Fh 
 
 d 
 
 d 
 
 d 
 
 3 
 
 H 
 
 o 
 
 
 d 
 
 tin 
 
 * rd 
 
 • d 
 
 
 o 
 
 3 
 
 
 
 d 
 
 Fh 
 
 © 
 
 > 
 
 o 
 
 d 
 
 Fh 
 
 d 
 
 CO 
 
 d 
 
 © 
 
 a 
 
 © 
 
 > 
 
 d 
 
 .2 
 
 d 
 
 • 
 
 a 
 
 © 
 
 r 
 
 Fh 
 
 © 
 
 > 
 
 © 
 
 3 
 
 Fh 
 
 d 
 
 © 
 
 H 
 
 d 
 
 © 
 
 ^d 
 
 Q 
 
 CO 
 
 to 
 
 Fh 
 
 O 
 
 d 
 
 «*H 
 
 o 
 
 Fh 
 
 © 
 
 4—> 
 
 © 
 
 a 
 
 
 3 
 
 Fh 
 
 © 
 
 43 
 
 © 
 
 s 
 
 * -*-> 
 
 • F- 
 
 • o 
 
 • H 
 
 • CO 
 
 • 43 
 
 . o 
 
 
 © 
 
 X 
 
 CO 
 
 .2 
 
 *c 
 
 © 
 
 CO 
 
 © 
 
 *© 
 
 d 
 
 Fh 
 
 o 
 
 Jd 
 
 H-> 
 
 p 
 
 Fh 
 
 H-J 
 
 d 
 
 © 
 
 © 
 
 o 
 
 rd 
 
 bfi 
 
 d 
 
 r2 
 
 © 
 
 Fh 
 
 3 
 
 CO 
 
 d 
 
 © 
 
 2 
 
 s 
 
 2 
 
 H 
 
 d 
 
 © 
 
 © 
 
 2 
 
 p 
 
 Fh 
 
 -H 
 
 d 
 
 © 
 
 © 
 
 V*_» 
 
 o 
 
 rd 
 
 H 
 
 d 
 
 5 
 
 © 
 
 2 
 
 o 
 
 Fh 
 
 t+H 
 
 d 
 
 © 
 
 Fh 
 
 d 
 
 73 
 
 d 
 
 © 
 
 © 
 
 ’a 
 
 CD 
 
 d 
 
 d 
 
 Fh 
 
 H> 
 
 CO 
 
 to 
 
 o 
 
 Fh 
 
 © 
 
 d 
 
 -d 
 
 H. 
 
 d 
 
 5 
 
 »H-I 
 
 o 
 
 Fh 
 
 © 
 
 -H 
 
 © 
 
 a 
 
 .2 
 
 d 
 
 3 
 
 F- 
 
 -H 
 
 d 
 
 d 
 
 to 
 
 Fh 
 
 O 
 
 d 
 
 H 
 
 o 
 
 rd 
 
 -H 
 
 d 
 
 5 
 
 © 
 
 CO 
 
 d 
 
 d 
 
 © 
 
 **H 
 
 o 
 
 Fh 
 
 © 
 
 -H 
 
 © 
 
 a 
 
 .2 
 
 © 
 
 to 
 
 d 
 
 Fh 
 
 o 
 
 * - 
 © 
 — 
 
 CO 
 
 O 
 
 a 
 
 i 
 
 o 
 
 Fh 
 
 © 
 
 43 
 
 d 
 
 d 
 
 
 .2 
 
 d 
 
 3 
 
 Fh 
 
 d 
 
 © 
 
 > 
 
 i 
 
 o 
 
 CO 
 
 Fh 
 
 o 
 
 d 
 
 . d 
 . **“• 
 
 • d 
 
 • © 
 
 . o 
 
 * ^ 
 
 * 43 
 
 * © 
 
 : 
 
 * © 
 
 . © 
 
 • d 
 
 • d 
 
 • 43 
 
 
 
 Fh 
 
 © 
 
 rO 
 
 H-3 
 
 CO 
 
 © 
 
 H-> 
 
 o 
 
 rd 
 
 H 
 
 CO 
 
 © 
 
 -H 
 
 a 
 
 *5 
 
 XI 
 
 3 
 
 CO 
 
 •H 
 
 CO 
 
 © 
 
 H> 
 
 © 
 
 > 
 
 i 
 
 o 
 
 Fh 
 
 © 
 
 > 
 
 to 
 
 Fh 
 
 © 
 
 CO 
 
 CO 
 
 © 
 
 H> 
 
 
 43 
 
 CO 
 
 © 
 
 * r 00 
 
 i d 
 
 n 
 
 *C 
 
 
 3 
 
 d 
 
 
 d 
 
 bC 
 
 o 
 
 d 
 
 CO 
 
 d 
 
 d 
 
 d 
 
 rd 
 
 d 
 
 -d c /3 
 
 
 
 3 
 
 © 
 
 Fh 
 
 •— 
 
 © 
 
 © 
 
 M 
 
 “© 
 
 d 
 
 © 
 
 Fh 
 
 o 
 
 d 
 
 Fh 
 
 d 
 
 Fh 
 
 © 
 
 Fh 
 
 3 
 
 © 
 
 Fh 
 
 ‘V, d 
 © 
 
 
 
 S5 
 
 OQO 
 
 E 
 
 
 o 
 
 M 
 
 H 
 
 H 
 
 O 
 
 
 o 
 
 l-J 
 
 
 • • 
 
 CO I- 
 
 1 Figured on plate 29, figs. 8 and 9. 3 Approximate. 6 Figured on plate 29, fig. 10. These vertebrae 6 Figured on plate 29, fig. 
 
 2 Figured on plate 29, figs. 1 to 4. 4 Figured on plate 29, fig. 5. tentatively determined as sixth to tenth inclusive. 7 Figured on plate 29, fig. 
 
140 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 increase in size to about the middle of the sacral series and then decreases. The last 
 sacral vertebra is wide transversely. A small facet for a chevron-bone may be present 
 at each end of the ventral border of the posterior surface. Occasionally a small process 
 may occur in this region. 
 
 Caudal vertebrae. —A representative series of caudal vertebrae of Mylodon harlani is 
 shown in plate 29, figures 1 and 2. In this series the centrum reaches its greatest length 
 at the end of the proximal third of the tail. Behind this region there is a gradual decrease 
 in antero-posterior diameter to the end of the tail. The centrum reaches its greatest 
 depth, as measured over the anterior surface, in the third or fourth vertebra. In the 
 following segments a gradual decrease is noted in this measurement. The centra are 
 similar in shape to those in M. robustus. The neural arch remains complete to approx¬ 
 imately the end of the middle third of the tail. With the reduction of the arch the 
 anterior and posterior zygapophyses become small and disappear. In the four anterior 
 vertebrae of the series the transverse processes are stout structures, but are not as wide 
 in fore-and-aft direction as in vertebrae of the middle portion of the tail. 
 
 In the first caudal a facet for the haemapophysis is present at each end of the ventral 
 border of the posterior surface. In vertebrae behind the first, two pairs of facets are 
 present, the two facets on the ventral border of the posterior surface being usually larger 
 than those on the ventral border of the anterior surface. Between the haemapophysial 
 facets of each side the lateral border of the ventral surface of the centrum may be notched, 
 and from this notch extends upward and backward a groove to the posterior side of the 
 transverse process. As indicated by Owen for M. robustus, this groove undoubtedly 
 carried a vertebral artery from the haemal vessel. Occasionally the lower end of the 
 groove may be inclosed by a bridge of bone. Specimens have also been observed in 
 which the opening of the groove into the haemal canal is closed. In the region of the 
 seventeenth vertebra the haemapophyses become quite small, and the facets for their 
 articulation tend to disappear. The haemapophysial facets on the ventral border of the 
 anterior surface of the centrum are usually the first to vanish. With the disappearance 
 of the haemapophyses, the haemal canal extends backward as an open groove on the 
 ventral side of the centra. 
 
 HAEMAPOPHYSES. 
 
 All the caudal vertebrae in Mylodon harlani, with the exception of those at the end 
 of the tail, carry chevron-bones (plate 29). Where typically developed these elements 
 are Y-shaped, in which respect Mylodon differs from Nothrotherium. In the first vertebra 
 of the caudal series, where the facets for the haemapophysis are widely separated trans¬ 
 versely, the right and left elements do not unite ventrally to inclose the haemal canal 
 (plate 29, fig. 3). Each element articulates mainly with the posterior facet of the first 
 caudal and tapers downward. The second haemapophysis is completely formed, the 
 right and left elements joining below to form a wide V-shaped canal (plate 29, fig. 4). 
 Below this union extends a short spine. Each arm of the chevron is broad antero- 
 posteriorly and carries at the upper anterior and posterior end a facet that articulates 
 with second and third vertebra respectively. 
 
 In the following four haemapophyses there is to be noted a gradual lengthening of 
 the elements ventrally and a broadening antero-posteriorly of the haemal spines. With 
 the narrowing of the caudal vertebrae and the close proximity of right and left facets 
 for the chevron-bones, the two arms of the haemapophysis approach each other more 
 closely at the upper ends. In the middle region of the tail the haemapophyses are 
 broad antero-posteriorly and the haemal spine forms a distinct keel (plate 29, figs. 5 and 6). 
 Posterior to about the ninth chevron-bone the elements show gradual reduction in 
 size, but the spine may remain broad to the thirteenth. Behind this haemopophysis 
 the fourteenth and fifteenth elements still articulate by anterior and posterior facets 
 with the caudal vertebrae. 
 
PLEISTOCENE MEGALONYCIIINAE AND MYLODONTIDAE OF RANCHO LA BREA. 141 
 
 COSTAL RIBS. 
 
 There are presumably sixteen pairs of costal ribs in Mylodon harlani, a number 
 similar to that in M. robustus. The anterior ribs are joined below with the ossified 
 sternal ribs. The first costal rib is short and broad and is firmly united with the sternal 
 rib in adult individuals. Occasionally the ribs immediately behind the first also co¬ 
 ossify with the sternal ribs. The costal ribs over the middle portion of the body are 
 strong, broad elements, increasing in thickness at the lower ends. In the series the 
 capitulum and tuberculum gradually move apart and are widely spaced in the posterior 
 region. At the end of the series the head and tubercle tend to fuse with the thoracic 
 vertebrae with which they articulate. In the last thoracic vertebra the rib is greatly 
 reduced in length, is firmly united with this segment, and has the appearance of a later¬ 
 ally extended transverse process. 
 
 Table 71.— Measurements (in millimeters) of caudal vertebrae of Mylodon harlani, series 1590-1 to 
 
 1609-1 inclusive. 
 
 
 
 
 
 No. in caudal series. 
 
 
 
 
 
 1 
 
 2 
 
 3 
 
 4 
 
 5 
 
 6 
 
 7 
 
 8 
 
 9 
 
 10 
 
 Height from ventral border of pos¬ 
 terior face. 
 
 Length of centrum. 
 
 149.6 
 
 54.2 
 
 139 
 
 56.5 
 
 134.2 
 
 59 
 
 131 
 
 60.3 
 
 125.7 
 
 59.8 
 
 115.6 
 
 60.3 
 
 104.8 
 
 61 
 
 61 
 
 82.3 
 
 60.2 
 
 75.5 
 
 59.1 
 
 Width of centrum across anterior 
 face. 
 
 77.2 
 
 70.7 
 
 69.6 
 
 67.3 
 
 65.1 
 
 65.3 
 
 65.3 
 
 62.3 
 
 59.2 
 
 53 
 
 Depth of centrum across anterior 
 face. 
 
 64.2 
 
 65.6 
 
 66 
 
 66 
 
 63.3 
 
 60.3 
 
 57.6 
 
 56.3 
 
 53.1 
 
 51.7 
 
 Greatest width across transverse 
 processes. 
 
 *255 
 
 251 
 
 246 
 
 236.2 
 
 213 
 
 204.5 
 
 178 
 
 160 
 
 134.9 
 
 112 
 
 Greatest width across metapophy- 
 ses. 
 
 101.6 
 
 109.8 
 
 108.6 
 
 104.2 
 
 97.2 
 
 91.7 
 
 76.2 
 
 67.2 
 
 53 
 
 50.3 
 
 Greatest width across posterior 
 zygapophyses. 
 
 70.5 
 
 72.2 
 
 64.4 
 
 61 
 
 51 
 
 38.8 
 
 32.7 
 
 • 
 
 28.7 
 
 21.4 
 
 
 
 
 
 
 No. in 
 
 caudal 
 
 series. 
 
 
 
 
 
 
 11 
 
 12 
 
 13 
 
 14 
 
 15? 
 
 16 
 
 17 
 
 18 
 
 19 
 
 20 
 
 21 
 
 Height from ventral border of poste¬ 
 rior face. 
 
 64 
 
 49.7 
 
 43.4 
 
 
 
 
 
 
 
 
 
 Length of centrum. 
 
 55 
 
 51.5 
 
 49.5 
 
 47 
 
 46.9 
 
 43.3 
 
 42.6 
 
 38.8 
 
 37.6 
 
 29.3 
 
 
 Width of centrum across anterior face. 
 
 52.6 
 
 50.5 
 
 47.4 
 
 46.8 
 
 44.5 
 
 42.5 
 
 37.7 
 
 35.5 
 
 33.4 
 
 28.4 
 
 
 Depth of centrum across anterior face. 
 
 52.8 
 
 46.6 
 
 42.4 
 
 43 
 
 39 
 
 38.6 
 
 34.2 
 
 31 
 
 29.4 
 
 26.3 
 
 
 Greatest width across transverse pro¬ 
 cesses . 
 
 95 
 
 92 
 
 79.7 
 
 59.4 
 
 69.6 
 
 56 
 
 43.4 
 
 42.6 
 
 39.9 
 
 33.5 
 
 
 Greatest width across metapophyses. . 
 
 41 
 
 *45 
 
 
 39.2 
 
 41.5 
 
 30.3 
 
 30.5 
 
 .... 
 
 27.7 
 
 21.3 
 
 
 Greatest width across posterior zyga¬ 
 pophyses. 
 
 14.1 
 
 
 .... 
 
 .... 
 
 .... 
 
 .... 
 
 .... 
 
 .... 
 
 .... 
 
 .... 
 
 
 1 Approximate. 
 
 STERNAL RIBS. 
 
 The sternal ribs (plate 30) inclose the thoracic cavity ventrally and afford with the 
 sternum a framework of considerable strength. The series in M. harlani consists of at 
 least 10 elements resembling those in M. robustus. Usually 7 of the ribs join with sternal 
 segments, while the eighth, ninth, and tenth do not reach the sternum, but overlap each 
 other. The first sternal rib is very often fused with the vertebral rib, as i nNothr other ium. 
 The sternal ribs gradually increase in length from the first back to the seventh. Con¬ 
 nection between adjacent ribs begins with and continues behind the fifth. 
 
142 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 In the third and fourth sternal ribs the ventral portion of the medial articulation 
 for the sternum is usually symmetrical. In following ribs this articulation with the 
 sternum becomes asymmetrical with reference to a transverse axis. 
 
 The fifth rib establishes a connection with the sixth by an elongate surface directed 
 downward and situated near the middle of the posterior border. In the sixth rib a 
 corresponding surface is situated on the anterior border and faces upward. The postero¬ 
 lateral end of the rib is truncated obliquely by a large roughened surface that was con¬ 
 nected by cartilage with the seventh rib. 
 
 The seventh sternal rib is the longest and stoutest of the series. The medial artic¬ 
 ulation varies somewhat in shape, depending on the structure of the sixth and seventh 
 sternal segments. The lateral half of the rib broadens antero-posteriorly, is thick in 
 front, and thins to the posterior edge. At approximately the middle of the anterior 
 border is situated the roughened surface for union with the sixth sternal rib. In some 
 specimens of the seventh sternal rib a small surface may be elevated above the ventral 
 side and serves for attachment of the medial end of the eighth rib. Occasionally this 
 surface migrates to the posterior border. In such specimens the postero-lateral end of 
 the rib is truncated as in the sixth and serves also for cartilaginous attachment of the 
 eighth rib. In other words, the eighth rib may in some individuals join at two places 
 with the seventh rib, in others only the medial end extends over the ventral surface of 
 the seventh. 
 
 The sternal ribs posterior to the seventh resemble those seen in species of Mylodon 
 from South America. 
 
 STERNUM. 
 
 The segments comprising the sternum in Mylodon harlani (plate 30, figs. 1 to 6) 
 number seven in all, and are comparable in their general characteristics to those of 
 South American ground-sloths of the Pleistocene. The manubrium is the largest and 
 most conspicuous of these elements, forming a broad plate. Sternal segments II to V 
 inclusive possess individually the peculiar pawn-like structure, commonly found in 
 ground-sloths and permitting a double articulation with the sternal ribs. Segments 
 II, III, and IV have the ventral process elongated antero-posteriorly, while in No. V 
 the process is usually much shorter. The sixth segment may have also a ventral pro¬ 
 cess, but this element is sometimes represented only by a dorsal plate. 
 
 Occasionally other variations ma 3 ^ be noted in this region of the sternum. The 
 dorsal plate sometimes consists of a right and left piece not fused in median line, while 
 the ventral head also remains separate in some specimens (see plate 30, fig. 3). The 
 teratological characters exhibited by the sixth sternal segment in some individuals of 
 mylodont sloths from Rancho La Brea are a result of ossification of this element from 
 several centers, which have not united in later development. Such variations are known 
 to occur in other mammals, 0 but in Mylodon they seem to be restricted to the sixth seg¬ 
 ment of the sternum. With what degree of frequency a median cleft occurs in the 
 metasternum can not be adequately ascertained, because of the very limited number of 
 specimens of that element that are available. The xiphisternum possesses a thick median 
 ridge ventrally, but the development of a distinct ventral process does not occur in M. 
 harlani. The last segment, thinning to the posterior border, has more the appearance 
 of that of M. robustus. 
 
 There is a gradual increase in transverse width of mesosternum from the second to 
 the fourth segment, the latter having the greatest width of the mesosternal elements. 
 Posterior to the fourth, a gradual decrease in this measurement occurs to the xiphister¬ 
 num. The dorsal surface, particularly in the middle region of the sternum, may become 
 
 “A. M. Paterson. The human sternum. Three lectures delivered at the Royal College of Surgeons, England, No¬ 
 vember, 1903. London, 89 pp., 10 plates, 1904. 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 143 
 
 quite concave. This is especially noticeable in the fourth segment, which is broad. 
 Union of elements sometimes occurs in the sternum of Mylodon harlani. Specimens 
 are present in the collections from Rancho La Brea, in which the first and second seg¬ 
 ments are united, several in which the second and third are fused, and one in which the 
 third and fourth segments are fused. 
 
 A downward deflection of the sixth segment may be occasionally noted in side- 
 view of the sternum. The xiphisternum is apparently always somewhat deflected 
 downward from the plane of the sternum. 
 
 The individual elements of the sternum of M. harlani are discussed more in detail 
 below. 
 
 The manubrium is a broad element (plate 30, fig. 1), with ventral surface concave 
 in antero-posterior axis and convex in transverse direction. The dorsal surface may be 
 divided into three principal areas, of which the median concave one is by far the most 
 extensive. Anterior to this concavity and on either side of the median line lie the com¬ 
 paratively large excavations that lodge the medial ends of the clavicles. Along the sides 
 of the median concave area are situated the facets for the first sternal rib. The ante¬ 
 rior portion of each facet faces more toward the dorsal side than in Nothrotherium. 
 The manubrium of Mylodon harlani is thickest at the posterior end, where it articulates 
 with the second sternal segment and with the second sternal ribs. The surface for the 
 second sternal segment is directed posteriorly, not decidedly to the dorsal side as in 
 Nothrotherium, thus indicating that the manubrium was not bent so sharply from the 
 rest of the sternum. 
 
 All specimens of the first sternal segment of M. harlani differ from the corresponding 
 element in M. rohustus in shortness of that portion of the segment anterior to the dorsal 
 concavity, and in the greater transverse width across the anterior border of the con¬ 
 cavity. The transverse diameter anteriorly is also much greater in the manubrium of 
 the Rancho La Brea form than in that of M. gracilis. 
 
 The measurements (in millimeters) of manubrium (average of 25 specimens) are: 
 greatest antero-posterior diameter, 134.5; greatest transverse diameter, 128.5; greatest 
 thickness at posterior end, 42.7. 
 
 The second sternal segment has the plate-like dorsal piece tapering considerably 
 toward the anterior end. The anterior border is usually convex. The dorsal or internal 
 articulating surface at the anterior end is continuous ventrally along the sides, with 
 facets for the second sternal ribs. These in turn join below with another articulating 
 surface for the manubrium. It is to be noted that the articulation for the second sternal 
 rib forms a single surface, not divided into an upper and a lower part, as at the posterior 
 end or in the sternal elements that follow. The conformation and extent of the facets 
 at the anterior end of the second sternal element vary considerably. In a number of 
 specimens all four facets merge to form an extensive anterior surface, for the most part 
 convex in transverse direction, concave in dorso-ventral direction. The ventral knob or 
 process is elongated in antero-posterior diameter. There are more specimens preserved 
 in which this knob bears posteriorly a facet for the following sternal element than those 
 in which it is lacking. 
 
 The measurements (in millimeters) of sternal segment II are: Greatest length over 
 dorsal face (average of 21 specimens), 62; greatest width at posterior end (average of 
 21 specimens), 56.6; depth through middle (average of 21 specimens), 44.9; length of 
 ventral keel (average of 19 specimens), 64.5; width of ventral keel at posterior end 
 (average of 19 specimens), 44.9. 
 
 The third sternal segment is similar to sternal segment II in being longer than 
 wide. At each end is present the full complement of facets for articulation with the 
 sternal ribs, namely, two above, situated at the angle formed by the anterior and lateral 
 
J44 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 borders, and two below, which are larger than those above and situated on each side of 
 the end of the ventral process. The articulation of the ventral knob with that of the 
 segment in front and behind is variable. This is indicated by a variable presence of 
 the median facet in front and behind. 
 
 The measurements (in millimeters) of sternal segment III are: Greatest length 
 over dorsal face (average of 22 specimens), 60.4; greatest width at posterior end (aver¬ 
 age of 23 specimens), 71.7; depth through middle (average of 24 specimens), 45.7; 
 length of ventral keel (average of 18 specimens), 57.3; width of ventral keel at pos¬ 
 terior end (average of 20 specimens), 43.6. 
 
 In the fourth sternal segment the mesosternum reaches its greatest width, the 
 transverse measurement of this element being always decidedly greater than the antero¬ 
 posterior diameter. The ventral keel is usually large and massive. It often reaches its 
 greatest depth toward the posterior end. 
 
 The measurements (in millimeters) of sternal segment IV are: Length along median 
 line of dorsal side (average of 27 specimens), 59.6; greatest width of dorsal piece (average 
 of 27 specimens), 79.1; greatest depth (average of 25 specimens), 50.7; antero-posterior 
 diameter of ventral keel (average of 24 specimens), 54.7; width of keel (average of 24 
 specimens), 45.8. 
 
 Table 72. —Measurements (in millimeters ) of xiphisternum of 
 
 Mylodon harlani. 
 
 
 No. 1726-2. 
 
 No. 1726-6. 
 
 No. 1726-10. 
 
 Greatest length. 
 
 67 
 
 J 89 
 
 96.5 
 
 Greatest width at anterior end. 
 
 45.2 
 
 57.7 
 
 70.2 
 
 Depth. 
 
 21.5 
 
 28 
 
 28.6 
 
 1 Approximate. 
 
 The fifth sternal segment is not so wide as the fourth, indicating that the sternum 
 narrows gradually behind the latter and toward the xiphisternum. The ventral keel is 
 also distinctly shorter in antero-posterior diameter as compared to that of the fourth. 
 This causes the two lower facets for sternal ribs on each side and end of the keel to 
 approach each other more closely and makes the interspace of less extent. The ventral 
 keel is deepest toward the posterior end. The articulating surface of the dorsal plate 
 with that of the sixth segment is usually convex in transverse direction. 
 
 The measurements (in millimeters) of sternal segment V are: Length across dorsal 
 plate (average of 16 specimens), 62.5; greatest width at posterior end (average of 16 
 specimens), 72.1; greatest depth (average of 13 specimens), 54.3; total length of ventral 
 keel (average of 15 specimens), 43.7; greatest width of ventral keel (average of 14 speci¬ 
 mens), 37.7. 
 
 The sixth sternal segment is remarkable for its variable shape and structure. When 
 completely formed it resembles the sternal segments anterior to it in having a dorsal 
 plate which subtends a peduncle with an articulating head. This head or keel is note¬ 
 worthy because of its shortness. The four facets which it bears approach each other 
 very closely and often become confluent, remaining demarcated only by the distinct 
 angle between them. The keel seldom reaches so far posteriorly as to lie well between 
 the postero-lateral facets of the dorsal plate, thus differing from sternal segment V. 
 The posterior articulation for the xiphisternum may be concave or sharply indented. 
 A peculiar feature noted in a number of specimens is the separateness of the ventral 
 head. Sometimes it has apparently disappeared entirely, for there is no sign of an artic¬ 
 ulating surface on the ventral side of the dorsal plate. In other specimens a distinct 
 but small facet occurs in the median line of the plate. Corroborative evidence is sup- 
 
FLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 145 
 
 plied by the presence in the collections of specimens which undoubtedly represent the 
 ventral part of this sternal segment, and which possess a small facet that articulates 
 with a corresponding facet on the dorsal plate. In 16 specimens of sternal segment VI, 
 5 have the ventral keel fused with the plate, 4 specimens indicate the presence of the 
 ventral keel as a separate element, and the remaining 7 lack the keel and do not exhibit 
 its presence by the small facet on the ventral surface. It is possible that in the latter 
 group the ventral keel was actually present but very loosely connected with the dorsal 
 plate. Finally, an added peculiarity to be noted in this segment of the sternum is the 
 division of the dorsal plate into a right and left piece, a separation that has occurred in 
 5 of the 16 specimens under observation. 
 
 The xiphisternum resembles that in Mylodon robustus. The anterior border may be 
 straight or may slope away on each side of a median apex. The dorsal surface may be 
 flat or distinctly concave anteriorly. On the ventral side and at the two antero-external 
 corners are the small facets for the last sternal rib to articulate with the sternum. In 
 median line a tumid keel is developed which does not possess facets. Posteriorly this 
 keel gradually sinks into the dorsal plate. The posterior margin is rounding and was 
 probably fringed with cartilage. 
 
 Two specimens are noted in which xiphisternum has fused with the sixth sternal 
 segment. 
 
 Table 73. — Average measurements {in millimeters ) of scapulae of Mylodon harlani. 
 
 Length, measured from outer border of glenoid cavity to supra-scapular border and along the base 
 
 of the spine. 327.7 
 
 Greatest length, measured from end of clavicular facet to supra-scapular border and parallel to 
 
 length of spine. 435 
 
 Greatest width of blade, measured between ends of supra-scapular border. 422.1 
 
 Width measured below base of spine. 227.3 
 
 Height, measured from inner border of glenoid cavity to point directly above on surface of acro¬ 
 mial process. 162.5 
 
 Greatest antero-posterior extent of glenoid cavity. 117.5 
 
 Greatest transverse width of glenoid cavity. 72.3 
 
 Greatest width of coraco-acromial arch. 68.2 
 
 Greatest diameter of supra-scapular aperture inclosed by coraco-acromial arch. 168.4 
 
 Greatest diameter of coraco-scapular foramen. 29.9 
 
 Greatest extent of clavicular facet. 41.6 
 
 SCAPULA. 
 
 The lateral aspect of the scapula (plate 31, fig. 1) differs from that of Mylodon 
 robustus in the unequal areas separated by the spine. In M. harlani the prescapular 
 fossa is broader than the postscapular fossa, while in the South American species, and 
 also in M. garmani , the spine divides the lateral surface into two nearly equal parts. 
 The inequality in size of these fossae is also a feature noted in the scapulae of Notliro- 
 therium and Hapalops. The spine is sharply defined in the Rancho La Brea species, 
 with the anterior and posterior borders usually diverging dorsally toward the supra¬ 
 scapular border. In Nothrotherium the spine is not broad dorsally, nor are its borders 
 so well defined as those of Mylodon. It sinks gradually into the dorsal surface of the 
 scapula. The size and shape of the coraco-scapular foramen varies considerably in the 
 scapula of Mylodon harlani. It may be relatively smaller than the similar foramen in 
 the scapula of Nothr other ium, or it may be very large, in which specimen it greatly ex¬ 
 ceeds the size in M. robustus and in M. garmani. 
 
 In specimens belonging to young individuals (No. 1715-L-12, for example) the 
 coracoid element is still distinctly defined by suture. It becomes evident that the 
 greater portion of the border of the coraco-scapular foramen is formed by the cora¬ 
 coid. In specimen 1715-L-12, where the anterior border or epiphysis of the coracoid 
 is lacking, due to immaturity, this bone possesses a striking similarity to the reptilian 
 “coracoid” in shape. The coracoid is excluded from the glenoid cavity, as shown by 
 Ameghino. It is, however, intimately connected with the epiphysial element forming 
 
146 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 the anterior part of the glenoid articulating surface and designated metacoracoid by 
 Ameghino. 
 
 The acromian is evidently still separate in young individuals, but in the mature 
 animal it is firmly consolidated with adjacent structures to form a complete span of 
 bone from the scapular spine to the coracoid region. 
 
 CLAVICLE. 
 
 The clavicle (plate 31, fig. 2) is a strongly developed element of approximately twice 
 the size of that in Nothrotherium. The shaft is somewhat compressed and has a double 
 curvature. At the outer end of the shaft an oval, convex surface articulates with a 
 facet on the acromial arch of the scapula as in M. robustus. The opposite end of the 
 clavicle is tuberous, giving a broad surface for attachment to the first sternal segment. 
 
 Table 74. —Average measurements (in millimeters ) of clavicle of Mylodon 
 
 harlani. 
 
 G reatest length. 220.9 
 
 Least width of shaft at middle. 19.9 
 
 Greatest width at outer end below scapular facet. 21.6 
 
 Long diameter of scapular facet. 33.6 
 
 Greatest width of sternal end. 62.3 
 
 Greatest thickness of sternal end. 27.9 
 
 HUMERUS. 
 
 As suggested by Owen (1842, p. 81) the humerus of Mylodon differs from that of 
 the megalonychid ground-sloths in being heavily constructed. This is clearly shown by 
 the size of the shaft and of the extremities (see plate 32). Mylodon differs also notice¬ 
 ably from Megalonyx and Nothrotherium in the absence of the entepicondylar foramen. 
 
 In anterior view of the humerus (plate 32, fig. 2) the head is seen to be not very 
 prominent, rising but slightly above the level of the greater tuberosity. Occasionally 
 the head may be even less prominent, the articulating surface lying below this level. 
 The long axis of the proximal articulating surface has an antero-posterior direction. 
 The greater tuberosity is particularly large. In the groove between the tuberosities 
 and the head are lodged a number of large nutrient foramina. The pectoral ridge 
 varies in distinctness, as does also the eminence which terminates the ridge distally. 
 The pectoral ridge is never quite so well defined as it is in the humerus of M. robustus 
 figured by Owen. The deltoid ridge, overhanging the musculo-spiral course, is a noticeable 
 feature in the humerus of Mylodon , showing greater prominence than in the megalonychid 
 ground-sloths. As in the latter, this ridge becomes better defined in old individuals. 
 
 Distally the humerus is broad, with expanded pronator and supinator plates above 
 the distal trochlea. The articulating surfaces for radius and ulna are broad antero- 
 posteriorly. The surface for the ulna is flattened considerably in this direction, more so 
 than in the humerus of the megalonychid forms. The posterior surface of the humerus 
 above the distal trochlea is broadly concave, but as a rule the olecranon fossa is not 
 so distinct as in Nothrotherium. 
 
 The medullary foramen is frequently absent. When present it has a position at 
 about the middle of the posterior surface and is relatively small. 
 
 Table 75. —Average measurements (in millimeters) of humeri of Mylodon harlani. 
 
 Length through middle of shaft, measured from proximal articulating surface to middle of distal 
 
 articulating surface. 
 
 Length from proximal end of greater tuberosity to distal end of radial surface. 
 
 Greatest antero-posterior diameter of head. 
 
 Greatest width across tuberosities. 
 
 Distance from greater tuberosity to distal end of deltoid ridge. 
 
 Greatest width of shaft measured at outer end of deltoid ridge. 
 
 Antero-posterior diameter of shaft at distal end of deltoid ridge. 
 
 Greatest distance between inner and outer tuberosities of distal end, measured obliquely across 
 
 distal expansion. 
 
 Width of distal trochlea. 
 
 429.6 
 456.9 
 
 123.8 
 
 172.2 
 
 296.8 
 
 122.6 
 79.3 
 
 261.4 
 
 133.2 
 
PLEISTOCENE MEGALONYCIIINAE AND MYLODONTIDAE OF RANCHO LA BREA. 147 
 
 ULNA. 
 
 The ulna of Mylodon harlani (plate 33, figs. 1, 2, and 3) resembles in many char¬ 
 acters that of M. robustus (Owen, R., pp. 83 to 86). This element of the forearm is 
 shorter and much heavier than that of the megalonychid ground-sloths. The shaft is 
 very broad and is compressed. The two sides of the shaft below the articulation for 
 the humerus are concave. As indicated by Owen, the olecranon process is characterized 
 by great size and is bent obliquely inward. The coronoid process is also prominent. 
 The outer articulation for the humerus presents a broad, flattened or slightly concave 
 surface, while the inner articulation is deeply concave. The surface for the radius is 
 relatively small. The shaft thickens distally and terminates by an oblique truncation. 
 The surface for the cuneiform is very broad and flat. 
 
 Table 76 —Average measurements [in millimeters ) of ulnae of Mylodon harlani. 
 
 Greatest legenth. 395.9 
 
 Distance from end of olecranon process to summit of coronoid process. 199.6 
 
 Width from posterior border to summit of coronoid process. 161.2 
 
 Greatest thickness at distal end of shaft. 69.1 
 
 Greatest width at distal end of shaft. 90 
 
 Greatest width of distal articulating surface. 55 
 
 RADIUS. 
 
 This element of the forearm, (plate 33, figs. 4, 4 a, 4 b, and 5) also conforms closely in 
 its characters to that of Mylodon robustus as described by Richard Owen (1842, pp. 86, 
 87; plate 14). The radius is much shorter and heavier than that of the megalonychid 
 ground-sloths Megalonyx and Nothrotherium. The proximal articulation is distinctly 
 concave and elongated in the axis that extends from the ulnar to the outer border. The 
 articulating surface for the ulna is convex. Below this surface the middle portion of the 
 shaft may be deeply scarred where 
 the interosseous membrane attaches. 
 
 The ulnar border of the distal articu¬ 
 lating surface may be broadly con¬ 
 cave or notched in the middle. The 
 outer border and the posterior or outer 
 surface of the radius possess the de¬ 
 tailed structural characters described 
 by Owen for Mylodon. The anterior 
 or inner surface is relatively smooth. 
 
 The average measurements (in 
 millimeters) of 21 specimens of the radius of Mylodon harlani are as follows: Greatest 
 length measured from proximal end to end of styloid process, 296.1; long diameter of 
 proximal end, 77.5; short diameter of proximal end, 60.4; greatest length of distal end, 
 120.3; greatest width of distal end, 81.7. 
 
 FALCIFORM BONE OF MANUS. 
 
 Fig. 69 .—Mylodon harlani Owen. A, B, and C, views of falciform bone 
 of manus, No. 1700-19. X 0.50. Rancho La Brea Pleistocene. 
 
 Winge a has demonstrated the presence of the os falciforme manus in Catonyx gigan- 
 teus and in Scelidotherium magnum, and states that the element was designated by Lund 
 the os proprium flexorum manus. Winge notes that the falciform bone is a sesamoid 
 situated in the palm of the hand and lying in the tendons of the flexor digitorum pro¬ 
 fundus muscle. 
 
 In the collections from Rancho La Brea, 30 specimens are available of an element 
 (fig. 69, a, b, c) which is very similar in structure to the os falciforme manus of Sceli¬ 
 dotherium magnum as shown by Winge, and undoubtedly represents this sesamoid bone 
 
 ° Jordfundne og nulevende Gumlere (Edentata) fra Lagoa Santa, Minas Geraes, Brasilien. Med Udsigt over Gum- 
 lerne 3 indbyrdes Slaegtskab, E. Museo Lundii, vol. 3, part 2, p. 162, plate 31, and pp. 201, 202, plate 30, 1915. 
 
148 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 in the manus of Mylodon harlani. The series seems divisible into two parts, represent¬ 
 ing the right and left manus, this division being based upon the shape of the element 
 and the surface appearance of the bony texture. One division includes 20 specimens, 
 the other 10 specimens. 
 
 The falciform bone is a compressed element, somewhat elongated, with one end 
 broader than the other. It is entirely devoid of articulating facets. One of the principal 
 faces is generally somewhat concave, while the opposite face is more or less flattened. 
 The borders are smooth and rounding. Considering the 20 specimens which appear to 
 belong to one side, we note the following average measurements: Greatest length, 61.5 
 mm.; greatest width, 46.2 mm.; greatest thickness, 15 mm. 
 
 Fig. 70. —Mylodon harla?ii Owen. Right scaphoid, No. 24261 U. C. C. A, distal surface with facet for 
 trapezoid at right and facet for magnum at left; B, proximal surface with articulation for radius; 
 
 C, dorsal view with facet for trapezium at right; D, dorso-external view. X 0.50. Rancho La Brea 
 Pleistocene. 
 
 SCAPHOID. 
 
 This carpal element (fig. 70, a to d) has been described by Lull (1915, p. 368), 
 from whose description the following may be quoted: 
 
 "The bone is roughly triangular when viewed from above, with a rounded antero-internal angle and a 
 deeply concave postero-external side. Three-quarters of the superior surface is articular and is convex in both 
 dimensions in its outer part, becoming concave first along the transverse diameter and then in all directions in 
 (he broadly expanded inner portion. The anterior aspect of the bone is continuous with the outer face and the 
 latter bears a rather long, narrow lunar facet which is continuous with that of the upper aspect though demar¬ 
 cated by a pronounced angle. The lower margin bears two deeply rounded bays, the limits respectively of the 
 magnum and trapezoid facets. Except for the facets the entire surface of the bone is highly rugose. Pos¬ 
 teriorly the bone is prolonged into a compressed angle bearing on its external face a sub-circular facet with 
 which articulated the metacarpal of the first digit. When the bone is held with the superior face horizontal, 
 this facet nearly coincides with the vertical antero-posterior plane. In its proper orientation, however, as 
 indicated by the distal end of the radius, the upper surface of the scaphoid is inclined sharply downward and 
 outward, which inclines the metacarpal facet obliquely downward in such a manner as to make the articulation 
 feasible.” 
 
 In Rancho La Brea specimens the articulating facet for the lunar may be continuous 
 below by a narrow strip with the surface for the magnum, but distinguished from the 
 latter by a sharp angle. The facets for trapezoid and magnum, where they approach 
 each other most closely and are sometimes confluent, are situated on either side of a 
 prominent projection (fig. 70, a and d). In Megalonyx the articulation with the mag¬ 
 num consists of two distinct facets instead of the single surface seen in Mylodon. In 
 the manus of Hapalops the articulation between scaphoid and lunar is much more 
 extensive than that of Mylodon. There is, however, but a single facet for the magnum 
 as in the latter form, while the articulation between scaphoid and trapezoid is much more 
 extensive in the Miocene genus and includes an additional facet close to the palmar border. 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 149 
 
 Owen (1842, pp. 87, 88) considers, in the description of the scaphoid, that the 
 prominently developed process in this element represents the trapezium. There seem 
 good reasons for believing, however, that the trapezium has not united with the scaphoid 
 to form the scapho-trapezial element in Mylodon, but that it has in reality fused with 
 the first metacarpal. This has already been indicated by Anthony,® who states: 
 
 La main pentadactyle du Mylodon pleistocene a pu sembler a certains auteurs avoir poss^de un 
 veritable scapho-trapeze en connexion avec le premier doigt. Mais cette interpretation du scaphoide du 
 Mylodon comme un scapho-trapeze est tres vrai-semblablement erronee, etant donut; qu’il est probable qu’h 
 1 etat jeune le ler segment du ler doigt de cet animal presentait un element pouvant etre interprets comme un 
 trapeze. 
 
 Fig. 71 .—Mylodon harlani Owen. Right lunar, No. 24262 U. C. C. A, proximal view of articu¬ 
 lating surface for radius; B, distal view with surface for unciform in middle, a small portion of 
 facet for magnum shown on left, and facet for cuneiform on right; C, dorsal view; D, inner 
 view with surface for scaphoid above and that for magnum below; E, outer view. X 0.50. 
 
 Rancho La Brea Pleistocene. 
 
 The primitive relation of the elements in question is well shown in the Mioceno 
 genus Hapalops, a manus of which (No. 15171, Prin. Univ. Coll.) is available for 
 comparison with that of Mylodon. In Hapalops a process of the scaphoid is well 
 developed and bears upon its distal side an elongate facet for the trapezium. This 
 process is undoubtedly homologous with the corresponding structure of the scaphoid of 
 Mylodon. In the Miocene genus the trapezium is a separate element of relatively large 
 size. It articulates proximally with the scaphoid, laterally with the second metacarpal, 
 and distallv with the first metacarpal. There is no articulation between trapezium and 
 trapezoid. This position of the trapezium causes the first metacarpal to be well sepa¬ 
 rated from metacarpal II, a feature which characterizes also the manus of Mylodon. 
 In Hapalops the trapezium and metacarpal I when articulated approximate closely in 
 shape the corresponding element in Mylodon. In the latter genus, however, the artic¬ 
 ulating surface for metacarpal II is much less extensive, and this is true also for the 
 scaphoid articulation. Thus the weight of evidence seems to indicate that the trapezium 
 has fused with the first metacarpal and not with the scaphoid. 
 
 The average measurements (in millimeters) of 39 specimens of Mylodon harlani are 
 as follows: Greatest distance across articulating surface, 75.3; greatest distance from 
 border of facet for metacarpal I to border of radial surface, 82.8; distance from radial 
 surface to tip of distal process measured normal to radial surface, 45.3. 
 
 ° R. Anthony. Recherches anatomiques sur les Bradypes arboricoles le squelette du Paresseux a collier; ses rapports 
 morphologiques avec celui des autres bradypes, Ann. Sci. Nat., Ser. 9 (Zool.), vol. 9, p. 238. 1909. 
 
150 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 LUNAR. 
 
 The proximal surface of the lunar (fig. 71 a) articulating with the radius is convex 
 in dorso-palmar extent and narrows toward the palmar border to less than one-half the 
 transverse width of the surface dorsally (figs, a to e). This decrease in width is brought 
 about by the obliqueness of its inner border, indicating the scaphoid surface. When the 
 lunar is placed in normal relation to the remaining elements of the carpus, the radial 
 surface slopes steeply inward. The radial surface encroaches well upon the dorsal face 
 of the lunar, from which it is separated by a decided offset. Along the inner side (fig. 
 71 d) the radial surface is continuous with an articulating area for the scaphoid. The 
 latter surface is widest toward the dorsal side. In distal aspect of lunar (fig. 71 b) three 
 articulating surfaces are seen. An inner concave facet for the magnum is continuous at 
 a pronounced angle with the narrow surface for the unciform. The latter is the prin¬ 
 cipal surface of the distal end, conforms in shape to that on the unciform with which 
 it articulates, and is continuous externally, at an obtuse angle, with the third articu¬ 
 lating area, a small concave facet for the cuneiform. 
 
 Fig. 72 .—Mylodon harlani Owen. Right cuneiform, No. 24263 U. C. C. A, proximal view of ulnar 
 surface; B, dorsal view; C, distal view of surface for unciform; D, palmar view showing facet for 
 pisiform; E, inner view with facet for lunar. X 0.50. Rancho La Brea Pleistocene. 
 
 The average measurements (in millimeters) of 39 specimens of Mylodon harlani are 
 as follows: Greatest distance from radial surface to face for unciform, measured over 
 dorsal surface, 54.3; greatest width across radial surface, 66.1; dorso-palmar diameter 
 across unciform facet, 51.2. 
 
 CUNEIFORM. 
 
 The cuneiform (fig. 72, a to e) is a very compact bone with extensive, flattened 
 proximal face devoted to articulation with the ulna. The dorsal surface (fig. 72 b) is 
 much more extensive than the palmar (fig. 72 d) and is rugose, projecting well above 
 the general dorsal plane of the carpus. Distally the cuneiform (fig. 72 c) decreases 
 considerably in dorso-palmar diameter, and there is developed a prominent convexity 
 articulating with the unciform. This surface, when continued toward the inner and 
 palmar sides, becomes concave, and is confluent over a rounded ridge with a small 
 convex facet on the inner side of the cuneiform (fig. 72 e) for the lunar. At the union 
 of posterior and palmar borders the ulnar surface is continuous at a right angle, with a 
 small convex facet for the pisiform (see fig. 72, a and d). 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 151 
 
 The average measurements (in millimeters) of 42 specimens of Mylodon harlani are 
 as follows: Greatest distance across dorsal surface from inner to outer side, 68.4; greatest 
 proximo-distal distance, measured normal to ulnar surface, 50.5; greatest dorso-palmar 
 distance across ulnar articulating surface, 61.3. 
 
 PISIFORM. 
 
 The pisiform (fig. 73, a, b, c) of Mylodon is a heavy nodular bone. When viewed 
 from the outer distal aspect (fig. 73 a) it is seen to be usually wide above and narrowing 
 below. The surface thus viewed is convex, except along the border of the facet for the 
 cuneiform, where a groove is developed, and in which are located a number of nutrient 
 foramina. The surface on the opposite side of the pisiform (fig. 73 c) tends to be concave. 
 
 The concave articulating surface is usually subtriangular in shape, but the latter 
 varies somewhat and the facet may be oval. That portion of the surface forming the 
 base of the triangle is deflected slightly from the cuneiform articulation and joins with 
 the ulna. This small deflection of surface is not always present in specimens from 
 Rancho La Brea. Below the apex of the subtriangular facet and close to its margin 
 the rugose surface of the pisiform frequently shows a small but deep and characteristic 
 depression. The long axis of the element is oblique to the principal axis of the facet 
 for the cuneiform. 
 
 Fig. 73 .—Mylodon harlani Owen. Right pisiform. 
 No. 242^4 U. C. C. A, outer distal view ; B, view 
 of surface for cuneiform; C, proximo-internal 
 view. X 0.50. Rancho La Brea Pleistocene. 
 
 The average measurements (in millimeters) of 22 specimens of Mylodon harlani are 
 as follows: Greatest length (a axis), 55.6; width, measured at right angles to a axis and 
 below articulating facet (b axis), 34.7; depth measured from plane of cuneiform facet 
 and normal to the a and b axes, 36.4. 
 
 UNCIFORM. 
 
 This element in Mylodon (fig. 74, a to e) differs mainly from the unciform of Hapa- 
 lops in its method of articulation with the metapodial bones. This is expressed in dorsal 
 aspect (fig. 74 a) by entire absence of a proximo-internal development of bone articu¬ 
 lating distally with the lateral process of metacarpal III. In Mylodon the articulation 
 between unciform and third metacarpal consists only of two facets, seen on the inner 
 side (fig. 74 b), situated along the distal border and separated by an interarticular 
 gutter. These facets are also usually separated from the surface for the magnum, lying 
 along the proximal border of the inner side, by a rugose channel. The interarticular 
 areas are perforated by a number of foramina. The dorsal facet for metacarpal HI is 
 always the smaller (not well shown in fig. 74 b, surface to the left), and may be often 
 continuous proximally w T ith the magnum facet, from which it is to be distinguished by 
 a ridge. The palmar facet is but rarely connected with the magnum surface, such a 
 confluence having been noted only in 2 of the 38 specimens of the right series. From 
 the above it may be readily inferred that the articulating surface for the magnum varies 
 considerably in shape. In Nothrotherium and Megalonyx the surface for metacarpal III is 
 not interrupted in its extent from the dorsal to the palmar border. In Nothrotherium, 
 and presumably in Megalonyx, there is no sharp indentation of the dorsal portion of 
 the metacarpal surface with an articulating area lying on each arm of a right angle, 
 
152 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 as in the unciform of Hapalops. Thus the articulation between the unciform and meta¬ 
 carpal III in Mylodon is much more simple than that in the megalonychids, and among 
 the latter that in Hapalops is decidedly the most complex (see discussion under Nothro- 
 therium and Megalonyx). 
 
 The articulating surface for metacarpal IV (fig. 74 e) is narrow and band-like, ex¬ 
 tending from the border of the dorsal facet for metacarpal III along the distal border 
 of the dorsal surface for a short distance and thence to the palmar side, where it termi¬ 
 nates at the border of the palmar facet for the third metacarpal. This surface is de¬ 
 marcated from the facets for metacarpal III by approximate right angles, in which 
 respect it resembles that in the unciform of Hapalops, but the surface differs from the 
 latter in being relatively less extensive. It preserves the sigmoid curvature in dorso- 
 palmar direction which this surface possesses in Hapalops. Along its outer border it is 
 continuous over a ridge, with the relatively broad surface for metacarpal V. 
 
 Fig. 74 .—Mylodon harlani Owen. Right unciform, No. 24265 IT. C. C. A, dorsal view; B, 
 inner view; C, outer view showing surface for cuneiform; D, proximal view; E, distal view 
 showing surfaces for metacarpals IV and V. X 0.50. Rancho La Brea Pleistocene. 
 
 The articulating surface for lunar and cuneiform forms a boss or knob (fig. 74 d) 
 which is more prominent than in the unciform of Megalonyx or Nothrotherium. Along 
 the inner border of the former surface is that for the magnum. The facet for the lunar is 
 narrow and of sigmoid curvature. It differs from that in the unciform of megalonychids 
 in being distinctly demarcated along the outer side from the surface for the cuneiform. 
 
 The average measurements (in millimeters) of 38 specimens of Mylodon harlani are 
 as follows: Greatest transverse diameter across dorsal face, 69.4; proximo-distal diameter 
 from tip of proximal boss to palmar half of surface for metacarpal IV, 58.8; dorso-palmar 
 distance across surface for metacarpal IV, 57.4. 
 
 MAGNUM. 
 
 As remarked by Lull (1915, p. 369), the shape of this element (fig. 75, a to e) is 
 very irregular, and there are a number of surfaces which articulate with those of the 
 surrounding carpal bones. The facets do not always preserve their individuality, but 
 their boundaries are often merged or confluent with those of neighboring facets. 
 
 The variation which occurs most frequently is the union of the dorsal and palmar 
 facets situated on the distal side of the magnum and articulating with metacarpal III. 
 The surface thus formed is expanded at the dorsal border and somewhat more so at the 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 153 
 
 palmar border, these areas being connected by a narrow band. It possesses a sigmoid 
 curvature in dorso-palmar extent and the outline is sinuous. The dorsal portion of the 
 surface is separated by a rugose interarticular area from a small facet (fig. 75 b), artic¬ 
 ulating also with the median metacarpal. The ventral expansion is continuous with, 
 although demarcated by a right angle from, a small facet directed upward and articu¬ 
 lating with metacarpal II (fig. 75 d, upper right facet). Along the dorsal border for 
 about half its distance and on the inner side of the magnum (fig. 75 d, lower right facet) 
 is a facet for the trapezoid, which lies confluent with, but demarcated by, a sharp angle 
 from the usually smaller facet for the third metacarpal. The former facet may barely 
 touch the proximal convex surface for the scaphoid (fig. 75 d, upper left facet) which 
 extends to the palmar side. 
 
 E 
 
 Fig. 75. — Mylodon harlani Owen. Right magnum, No. 24266 U. C. C. A, proximal view; B, distal view; 
 
 C, dorsal view; D, inner view; E, outer view. X 0.50. Rancho La Brea Pleistocene. 
 
 All three facets—for scaphoid, trapezoid, and metacarpal III—are separated from 
 the small facet for metacarpal II by an interarticular gutter which joins on the distal 
 surface of the magnum with the space lying between the articulating facets for the 
 median metacarpal. In Megalonyx, according to Leidy (1855, p. 31): 
 
 “On the outer® side it [the magnum] presents at the dorsal border a tripartite articular facet, separated 
 by a deep interarticular groove from an oval facet at the palmar border. The tripartite facet articulates with 
 the scaphotrapezial, the trapezoid, and median metacarpal bones; and the oval facet .... articulates with 
 the trapezial portion of the scaphotrapezium.” 
 
 Table 78. — Variation in confluence of facets in 32 specimens (of right magnum ). 
 
 
 Specimens. 
 
 Percentage. 
 
 Scaphoid facet with trapezoid facet. 
 
 10 
 
 31.25 
 
 Unciform with dorsal facet on distal face for metacarpal III. . . 
 
 12 
 
 37.5 
 
 Dorsal and palmar facets on distal face for metacarpal III. 
 
 14 
 
 43.75 
 
 Palmar facet on distal face with dorso-internal facet, both for 
 
 
 
 metacarpal III. 
 
 2 
 
 6.25 
 
 In Mylodon there is but a single articulating surface for the scaphoid, while the 
 small facet for metacarpal II is absent in the magnum of Megalonyx. In the latter the 
 principal surface of the distal end (Leidy, fig. 13 a, plate 8) articulating with metacarpal 
 III extends from the dorsal to the palmar border as in Hapalops. In Mylodon the cor¬ 
 responding surface is represented by two facets which are disconnected in over half the 
 number of specimens examined of the right series. The articulating surface for the lunar 
 (fig. 75 a, left facet), as indicated by Lull (1915, p. 369), is separated from that for the 
 
 ° The designation outer and inner side as used by Leidy is not quite clear to the writer. The articulating surfaces 
 for scaphoid and trapezoid lie certainly on the inner rather than on the outer side of the magnum of Mylodon. It is possible 
 that Leidy considered the terms outer and inner with reference to a different orientation of the specimen than that held in 
 the present text. 
 
154 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 scaphoid by crest or faint groove. It is a large facet with sigmoid curvature and is bounded 
 on the outer side by a narrow but long articulating surface (fig. 75 e) for the unciform. 
 The latter varies considerably in shape. 
 
 The average measurements (in millimeters) of 32 specimens of Mylodon harlani are 
 as follows: Greatest dorso-palmar distance, measured across distal surface, 64.5; width 
 of dorsal surface, 44.5; proximo-distal diameter, 44.4; width across palmar surface, 
 measured normal to unciform surface, 43.1. 
 
 TRAPEZOID. 
 
 The trapezoid (fig. 76, a to e) is the smallest element of the carpus. Viewed from 
 the dorsal side (fig. 76 c), this bone is seen to be roughly triangular in shape. The largest 
 facet possessed by the trapezoid is situated on the proximal side (fig. 76 a) where it artic¬ 
 ulates with the scaphoid. It is ovoid in shape and for the most part flattened, with 
 the outer portion sharply deflected distally (fig. 76 d). On the dorsal surface this is 
 expressed by an obtuse angle between the proximal and proximo-external borders. 
 
 Fig. 76 .—Mylodon harlani Owen. Right trapezoid, No. 24267 U. C. C. A, proximal view; B, inner 
 view showing below the facet for metacarpal II; C, dorsal view; D, outer view showing above 
 deflected surface for scaphoid and below facet for magnum; E, distal view showing from left to 
 right the facets for magnum, metacarpal III, metacarpal II. X 0.50. Rancho La Brea Pleistocene. 
 
 Distally the trapezoid joins the metacarpals by two facets and with the magnum 
 by a single facet. A small facet articulating with the metacarpal III truncates the 
 distal prominence, on the inner side of which is situated the surface for metacarpal II 
 and on the outer side the facet for the magnum. The surface for the second metacarpal 
 is relatively large, with principal axis in dorso-palmar direction, while that for the mag¬ 
 num is smaller, with principal axis more in a proximo-distal direction. The great dorso- 
 palmar extent of the facet for the second metacarpal carries the lower surface of the 
 trapezoid well below the inferior margin of the facet for the scaphoid (fig. 76 b). The 
 extent of articulation between trapezoid and metacarpal III varies considerably in 
 specimens from the asphalt beds. 
 
 The average measurements (in millimeters) of 21 specimens of Mylodon harlani are 
 as follows: Greatest dorso-palmar diameter (a axis), 30.9; greatest distance from sur¬ 
 face for scaphoid to surface for metacarpal III, measured across dorsal face (b axis), 
 34.6; greatest distance across dorsal face, measured approximately normal to b axis (c 
 axis), 40.5. 
 
 From the average measurements, given above, it is seen that the length of the c 
 axis is the greatest of the three measurements, while the a axis is the shortest and the 
 b axis lies between these two in length. There are, however, a few specimens which do 
 not conform to the average measurements. In No. 1489-R-7, for example, the a axis 
 is greater than the b axis and slightly greater than the c axis, and No. 1489-L-l has the 
 a axis greater than the 6 axis, but distinctly less than the c axis. In No. 1489-R-7 the 
 excessive length of the a axis is due to the great palmar extent of the facet for meta¬ 
 carpal II. 
 
 TRAPEZIUM AND METACARPAL I. 
 
 It has already been noted in the discussion of the scaphoid that the trapezium has 
 not fused with this element, as Owen supposed, but that the trapezium has co-ossified 
 with the first metacarpal. In Mylodon the right-angled element thus produced (fig. 77, 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 155 
 
 A to d) is very similar in shape to the articulated trapezium and metacarpal I in Hapa- 
 lops. Winge (1915, p. 193) has noted the fusion of metacarpal I and trapezium in Sceli- 
 dotherium magnum , but in the compound element thus formed (H. Winge, plate 39) 
 the lateral process is not so prominent as in that of Mylodon. In the manus of S. magnum 
 as shown by Winge, the trapezium-metacarpal I element articulates distinctly with the 
 trapezoid; an articulation which is totally lacking in M. harlani and M. robustus. 
 
 Of interest in this discussion is the occurrence in the Rancho La Brea collection of 
 several specimens in which the lateral process has ossified separately and has not fused 
 with the principal shaft representing metacarpal I. This is well shown by the two speci¬ 
 mens in figure 77, e to i, both of the left side, but not of the same individual. They 
 have been placed in juxtaposition to indicate the relation of parts. The lateral piece 
 possesses the usual articulating surface for metacarpal II. Internally and upon the 
 proximal face is a small facet for the scaphoid, which is a continuation of the surface 
 for the same element situated on the inner piece. Since in earlier forms the trapezium 
 
 Fig. 77. —Mylodon harlani Owen. A to D, right metacarpal I and trapezium, No. 24268 U. C. C. 
 
 A, dorsal view; B, proximal view; C, outer view showing on left a small facet for metacarpal II; 
 
 D, distal view showing on right surface for digit I and on left the distal portion of facet for meta¬ 
 carpal II. X 0.50. E, left metacarpal I, No. 1482-L-21, dorsal vew; F, trapezial moiety, dorsal 
 view; G, trapezial moiety showing contact surface for metacarpal II; H, metacarpal I, No. 
 
 1482-R-ll, dorsal view; I, No. 1482-R-ll, inner view showing facet for trapezial moiety. 
 
 X 0.50. Rancho La Brea Pleistocene. 
 
 and metacarpal I arose from separate centers of ossification, we may expect to find 
 these centers occasionally remaining distinct in Mylodon and giving rise to two inde¬ 
 pendent elements in the mature individual. Perhaps in such instances the lateral piece 
 is the homologue of the trapezium and the medial piece the homologue of the first meta¬ 
 carpal. It is to be noted, however, that in Hapalops the articulation exists between 
 scaphoid and trapezium, whereas in the present case a portion of the scaphoid surface 
 is borne by the medial piece and the remaining portion by the lateral piece. 
 
 The separation of the metacarpal I trapezial element in Mylodon harlani into two 
 distinct parts is known to occur 3 times in the left series of 27 specimens and once in 
 the right series of 21 specimens. Since each of these occurrences is known from a dif¬ 
 ferent pit at Rancho La Brea, it is safe to assume that at least 4 individuals were 
 characterized by the division of the compound element. 
 
 When the element formed by the union of trapezium and metacarpal I is viewed in 
 dorsal aspect (fig. 77 a) it is seen to consist of two arms which meet in a right angle. 
 The principal shaft is well developed, notwithstanding its shortness, and articulates at 
 the proximal end (fig. 77 b) with the process of the scaphoid by an ovoid facet whose 
 long axis is transverse. At the distal end articulation with the digit is formed by an 
 
156 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 oval convexity with flattened base (fig. 77 d). The main axis of this surface is oblique. 
 A small portion of the articulating surface continues upon the palmar side and occa¬ 
 sionally the middle of the palmar border is notched. 
 
 As remarked, the lateral arm meets the main shaft in a right angle, but when the 
 metapodial is viewed from the proximal end (fig. 77 b) the former is seen to rise obliquely 
 upward to its articulation with the second metacarpal. This articulation consists of a 
 small facet (fig. 77 c) directed laterally, with a portion of the surface curving well to the 
 distal side (fig. 77 d). With the exception of the articulating facets, the surface of the 
 element is highly rugose and pierced by a number of nutrient foramina. 
 
 The average measurements (in millimeters) of 21 specimens are as follows: Proximo- 
 distal diameter of shaft measured normal to plane of facet for scaphoid, 39.8; greatest 
 distance between inner side of shaft and end of lateral process, 45.8. 
 
 CO-OSSIFIED PHALANGES I AND II, DIGIT I, MANUS. 
 
 The first and second phalanges of digit I are fused in the manus of Mylodon, forming 
 an element (fig. 78, a to d) somewhat similar to that of the second digit of the pes. In 
 the manus, the element is distinctly smaller and more slender than that in the pes. The 
 proximal end (fig. 78 a) is oval in shape, with flattened base, and conforms with the sur¬ 
 face on metacarpal I, with which it joins. A principal axis extending through the middle 
 
 Fig. 78. —Mylodon harlani Owen. Co-ossified pha¬ 
 langes I and II, digit I, right manus, No. 24269. 
 U. C. C. A, proximal view; B, outer view; C, 
 dorsal view; D, distal view. X 0.50. Rancho 
 La Brea Pleistocene. 
 
 of the proximal articulating face in dorso-palmar direction is oblique to the median 
 vertical plane of the phalanx. This suggests a convenient method of distinguishing right 
 from left element, as follows: Orient the phalangeal segment in proximal view with 
 palmar border horizontal. The principal oblique axis of the proximal surface lies to 
 the left or to the right of an axis normal to the base, according to whether the element 
 belongs to the left or right manus, respectively. 
 
 At the distal extremity (fig. 78 d) the articulating surface is directed mainly forward 
 and downward. It is unevenly divided by a furrow into inner and outer portions, of 
 which the former is distinctly the larger and extends farther upon the palmar surface. 
 The inner portion forms a complete convex curve, whereas the outer forms but half of 
 such a curve. In the phalanx of the pes such a complete convex surface is never 
 developed. 
 
 The average measurements (in millimeters) of 19 specimens of Mylodon harlani are 
 as follows: Total length, 34; depth of proximal end measured normal to base, 28.7; 
 width of proximal end, 21.5; greatest transverse width behind distal articulation, 21.2. 
 
 PHALANX III, DIGIT I, MANUS. 
 
 The terminal phalanx of the first digit, manus (fig. 79 a, b, c), is the smallest of 
 the ungual series, being smaller and not so heavy as the ungual phalanx of digit II, 
 pes. Considerable variation is shown in the form of the proximal articulating surface. 
 The usual type consists of an inner large and an outer small area continuous over a low 
 vertical ridge, with the posteriorly projecting proximal end situated mainly over the 
 inner area. In some specimens the two articulating surfaces may be of approximately 
 equal size, with the ridge median in position. In other specimens (No. 1471-L-6, for 
 example) the ridge tends to disappear, and a single concave surface articulates with the 
 proximal element. 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCIIO LA REBA. 157 
 
 The base of the phalanx is flattened near the proximal articulation, but becomes 
 tuberous toward the distal end. On each side of the subungual tubercle are two fora¬ 
 mina, which lie in the same general depression. The lowermost one on each side pene¬ 
 trates the base, while the foramen above and behind the latter permits the entrance of 
 vessels which supply the base of the nail. On the outer side of the phalanx the foramen 
 piercing the sheath lies farther above the foramen in the base than it does on the inner 
 side, and causes this side of the sheath to be more extensively scarred. The sides of 
 the claw-process indicate the path taken by the vessels which penetrate the sheath. 
 This path on the inner side extends forward and upward to the rugose area at the free 
 end of the process, where it bifurcates. The claw-process is asymmetric, for the greater 
 portion of the dorsal convex surface lies to the inner side of a vertical plane passing 
 through the process, and the separation between the palmar convex surface of the 
 process and the lateral wall is lower on the inner side than on the outer side. 
 
 Allen describes the corresponding element in Mylodon garmani as follows (p. 336): 
 
 “The first ungual phalanx .... measures 90 mm. in extreme length and differs from the others in 
 having a stem some 25 mm. in length, the long axis of which is bent at an angle with that of the claw-bearing 
 portion. This angle seems much greater than in M. robustus. The proximal face is slightly concave. The bony 
 core of the claw is some 55 mm. long, laterally compressed, and regularly tapering in side view to a blunt point. 
 The basal half is surrounded by a thin bony sheath which becomes much thickened and rugose at the base 
 ventrally.” 
 
 Fig. 79. —Mylodon harlani Owen. Ungual phalanx, digit I, right manus, No. 1471-R-l. A, dorsal view; B, 
 outer view; C, ventral view. X 0.50. Rancho La Brea Pleistocene. 
 
 A comparison between the Rancho La Brea specimens and the figure of Allen’s 
 specimen (Allen, 1913, plate 3, fig. 10) discloses the fact that the “stem” of the first ungual 
 phalanx is in reality the co-ossified first and second phalanges, and this compound structure 
 has in turn been fused with the ungual phalanx, so that the entire digit is here represented. 
 It is worthy of note that among all the material of Mylodon harlani from the asphalt 
 beds a comparable fusion has not been found. 
 
 The average measurements (in millimeters) of 23 specimens of Mylodon harlani are 
 as follows: Greatest length, measured from most projecting part of proximal end to 
 distal end of claw-process, 74.8; depth at proximal end, 28.7; distance from subungual 
 tuberosity to dorsal surface of claw-process, measured normal to dorsal border, 32.1; 
 greatest width at proximal end, 28.9; width of claw-process at distal end of subungual 
 tubercle, 15.8. 
 
 METACARPAL II. 
 
 This metapodial (fig. 80, a to e) is, next to metacarpal I, the smallest of the meta¬ 
 carpal series. The proximal extremity (fig. 80 d) is of triangular shape and bears an 
 articulating surface for the trapezoid. The latter is usually narrow and lies along the 
 lateral border, where it is confluent with, but demarcated by, a sharp angle from the 
 articulating concavity for metacarpal III on the outer side. In No. 1488-R-29 the 
 surface for the trapezoid is quite extensive and reaches across the proximal extremity 
 to its inner border. The apex of the triangular-shaped proximal end is truncated 
 obliquely by a small facet which articulates with the magnum. The distal extremity 
 (fig. 80 e) is modified to form a strong articulation for a well-developed digit, and this 
 surface encroaches well upon the dorsal side of the metapodial (fig. 80 b), permitting a 
 strong retraction of the digital elements. The carina is prominent, rounded above, but 
 
158 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 becoming sharper below, and is oblique with reference to a median plane passing through 
 the metapodial, the lower end lying on the inner side of such a plane. On each side of 
 the carina is an offset articulating also with the first phalanx. The offset on the inner 
 or radial side of the carina is broad above and continues to the palmar side with about 
 the same width, while that on the outer or ulnar side increases from little or nothing to 
 a considerable width below. Both inner and outer offsets articulated below with sesa- 
 moids. 
 
 Fig. 80. —Mylodon harlani Owen. Right metacarpal II, No. 24270 U. C. C. A, outer view; B, dorsal view; C, 
 inner view; D, proximal view; E, distal view. X 0.50. Rancho La Brea Pleistocene. 
 
 At the proximo-dorsal angle of the inner side of the metapodial (fig. 80 c) is an 
 elongate, concave articulating surface for metacarpal I. On the opposite side (fig. 80 a) is 
 a larger concavity articulating with the third metacarpal. 
 
 The average measurements (in millimeters) of 27 specimens of Mylodon harlani are 
 as follows: Greatest length, 71.6; greatest diameter of distal carina, 51.4; depth of 
 shaft, 35.1; width of distal articulating surface, 33.7. 
 
 PHALANX I, DIGIT II, MANUS. 
 
 This element (fig. 81 a, b, c) is nearly as broad at the base as it is high. It is much 
 shortened proximo-distally, and in this direction the inner side (fig. 81 b) is longer than 
 the outer. The proximal extremity (fig. 81 a) is devoted entirely to articulation with 
 metacarpal II, and consists principally of a median groove which narrows below. Along 
 the inner border the furrow is flanked for the greater part of its extent by an articu¬ 
 lating offset for the metapodial which tends to incline away from the groove. Along 
 approximately the lower half of the opposite border of the groove is also an offset, but 
 this one, in contrast to the inner offset, lies horizontal. In palmar view it is to be noted 
 that the angle between the inner offset and the inner face of the median groove approaches 
 more nearly a right angle than does the angle between the outer offset and the outer face 
 of the groove. This is due in part to a more steeply inclined inner surface of the furrow 
 
 Fig 81. — Mylodon harlani Owen. Phalanx I, digit 
 II, right manus, No. 24271 U. C. C. A, proximal 
 view; B, inner view; C, distal view. X 0.50. Rancho 
 La Brea Pleistocene. 
 
 when contrast is made with the outer surface. Both offsets are continued below by 
 small facets which articulate with sesamoid bones. On the distal extremity (fig. 81 c) 
 articulation with phalanx II is accomplished by an inner and outer convexity separated 
 by a median groove. The outer convexity or condyle is of greater dorso-palmar extent 
 than the inner condyle. These distal convex surfaces readily distinguish the phalanx 
 from the corresponding element of digit III, pes, in which the condyles are flatter. The 
 palmar border of the phalanx below the distal articulation is notched. 
 
PLEISTOCENE MEGALONYCIIINAE AND MYLODONTIDAE OF RANCHO LA BREA. 159 
 
 The average measurements (in millimeters) of 34 specimens of Mylodon harlani are 
 as follows: Greatest depth, 49.1; greatest width, 41.5; proximo-distal diameter across 
 inner side, 27.4. 
 
 PHALANX II, DIGIT II, MANUS. 
 
 This is a powerful phalangeal segment (fig. 82, a to d) suitable to support a well- 
 developed ungual phalanx. The proximal extremity (fig. 82 c) is elongated in dorso- 
 palmar axis, narrowing above and with flattened base below. It supports two concave 
 facets separated in middle line by a tumid ridge which widens below, separating the 
 facets farthest in this region. The distal extremity (fig. 82 b) is modified to form a 
 trochlea, the outer condyle of which is the larger and is separated from the inner con¬ 
 dyle by a broad groove. This groove is broader and in most instances more shallow 
 than in second phalanges of Megalonyx. It terminates dorsally in a deep pit for recep¬ 
 tion of the posteriorly projecting process of the ungual phalanx. The circumference of 
 the inner condyle describes two-thirds of a circle. At the postero-dorsal angle of the 
 outer surfaces of the condyles heavy overhanging ridges are developed, indicating strong 
 tendinous attachments. The shaft of the phalanx is deepest as measured over the outer 
 surface. 
 
 Fig. 82. —Mylodon har¬ 
 lani Owen. Phalanx 
 II, digit II, right 
 manus, No. 24272 
 U. C. C. A, outer 
 view; B, distal view; 
 
 C, proximal view; 
 
 D, dorsal view. X 
 0.50. Rancho La 
 Brea Pleistocene. 
 
 The average measurements (in millimeters) of 30 specimens of Mylodon harlani are 
 as follows: Greatest length measured through middle, 54.1; depth of proximal end, 47; 
 depth of outer condyle, 34.1; width of proximal end, 36.8; greatest width across pos¬ 
 terior region of distal trochlea, 36. 
 
 PHALANX III, DIGIT II, MANUS. 
 
 Next to the ungual phalanx of the median digit of the manus, this element (fig. S3, 
 a, b, c) is the largest of the terminal series. The proximal articulation consists of two 
 concave surfaces, continuous over a median low and broad ridge, of which the outer is, 
 of course, the larger one. Both facets are continued upward along the under surface of 
 the posterior projection. About one-half of the basal length of the phalanx is devoted 
 to the base of the claw-process, and from which arises the bony sheath that encases the 
 posterior portion of the claw. The base is flattened, with discoid area roughly outlined 
 distally, behind which and on each side of the base are located the foramina for the 
 subungual vessels. The inner foramen is larger than the outer one and is often divided 
 into two by a bony partition. The bony sheath, although well developed along the 
 sides of the claw-process, becomes thin above, and hence is rarely preserved entire in 
 specimens from Kancho La Brea. It is doubtful, too, whether the sheath was always 
 developed over the dorsal surface of the claw. The claw-process is convex dorsally, 
 but the curvature is not symmetric with reference to a median vertical plane. Toward 
 the tip the upper surface is marked by a sharp angle from the sides, lhe border mark¬ 
 ing the separation between palmar and lateral surfaces is lower on the inner side than 
 on the outer side. 
 
 No. 1470-R-l and No. 1470-R-25 represent the extremes in size of the ungual 
 series of the second digit. Another interesting comparison can be made between speci¬ 
 mens 1470-R-l9 and 1470-R-3, which indicate the extremes in thickness of claw- 
 
160 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 process. The former is one of the most slender of ungual phalanges for digit II which 
 has been examined, while the latter is one of the heaviest. In the left series of terminal 
 phalanges, No. 1470-L-24 exceeds even No. 1470-R-3 in thickness of claw-process, the 
 / axis (see below) in this specimen measuring 35.7 mm. 
 
 Table 79. —Average measurements (in millimeters) of 21 specimens of Mylodon harlani. 
 
 Greatest length, measured from most projecting portion of proximal end to end of claw-process 
 
 (a axis). 153.6 
 
 Distance from proximal end of subungual base to end of claw-process (b axis). 125.6 
 
 Greatest proximo-distal diameter of subungual base (c axis). 61.8 
 
 Depth of proximal end (d axis). 43.9 
 
 Width of proximal end (e axis). 46.2 
 
 Greatest transverse width of claw-process at distal end of subungual base (/ axis). 28.1 
 
 Fig. 83 .—Mylodon har¬ 
 lani Owen. Ungual 
 phalanx, digit II, 
 right manus, No. 
 24273 U. C. C. A, 
 outer view with 
 c r o s s-s e c t i o n of 
 claw-process; B, 
 dorsal view; C, ven¬ 
 tral view. X 0.50. 
 Rancho La Brea 
 Pleistocene. 
 
 METACARPAL III. 
 
 Metacarpal III (fig. 84, a to f) is the stoutest of the metacarpal series, preserving 
 the heaviness which characterizes this metapodial in the earlier ground-sloths. The 
 shaft is nearly quadrate in section, with both the proximal and distal extremities greatly 
 modified to articulate with adjoining elements. The proximal border of the dorsal sur¬ 
 face is indented and marks the separation between inner and outer processes of the 
 proximal end. This extremity (fig. 84, d and f) may be divided into three parts, of 
 which the lateral processes form two and the ventral or palmar projection the third 
 division. 
 
 Table 80. —Measurements (in millimeters) of extremes of ungual 
 
 series in digit II, manus. 
 
 
 a axis. 
 
 h axis. 
 
 c axis. 
 
 d axis. 
 
 e axis. 
 
 / axis. 
 
 1470-R-l. 
 
 170.4 
 
 142.9 
 
 67.5 
 
 45 
 
 48.5 
 
 29.3 
 
 1470-R-25. 
 
 139 
 
 116 
 
 52.1 
 
 39.4 
 
 41.8 
 
 25.6 
 
 1470-R-l 9. 
 
 148.9 
 
 123.5 
 
 60.3 
 
 41.6 
 
 42.8 
 
 25.2 
 
 1470-R-3. 
 
 157.7 
 
 127.8 
 
 63.2 
 
 48.8 
 
 52.6 
 
 33.6 
 
 Close to the middle of the dorsal border a pit is developed on the proximal surface 
 which separates the two dorsal facets for the magnum. The outer of the two facets 
 forms the inner proximal surface of the lateral process. It forms a right angle with an 
 articulating surface on the outer proximal side of the process for the unciform. This 
 facet, in turn, is separated by a right angle from an articulating surface on the outer 
 distal side of the process for metacarpal IV. In other words, the lateral process is 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 1G1 
 
 sharply defined on three sides by articulating facets. The inner process supports mainly 
 an articulating convexity for metacarpal II (fig. 84 c). Between this surface and the 
 inner dorsal facet for the magnum lies the facet for the trapezoid. The palmar division 
 of the proximal extremity supports along its inner lower border a large facet, also for 
 the magnum. Its outer side supports two facets meeting in a sharp angle, one directed 
 proximally, the other, which is small, directed distally. The former facet is for the 
 unciform, the latter for the fourth metacarpal. The palmar portion of the proximal 
 end supporting the two facets is separated from the lateral process by a deep groove. 
 
 Fig. 84 .—Mylodon harlani Owen. Right metacarpal III, No. 24274 U. C. C. A, outer view; B, dorsal 
 view; C, inner view; D, proximal view; E, distal view; F, proximal view of another specimen, No. 
 23148 U. C. C., showing confluence of facets. X 0.50. Rancho La Brea Pleistocene. 
 
 As may be inferred from the description of the magnum, the outer dorsal facet and 
 the palmar facet, both of which articulate with this element, are frequently confluent in 
 metacarpal III. This is shown in figure 84 f. 
 
 Table 81 .—Average measurements {in millimeters ) of 30 specimens of metacarpal 
 
 III of Mylodon harlani. 
 
 Greatest length. 103.1 
 
 Greatest depth of shaft across outer surface. 38.3 
 
 Greatest extent of carina. 69.8 
 
 Greatest width of proximal end. 75.5 
 
 Least width of shaft, measured across dorsal surface. 42.6 
 
 Greatest width of distal end. 50.9 
 
 Depth of proximal end, measured across proximal surface. 66 
 
 The distal extremity (fig. 84 e) resembles that of metacarpal II and allows a very 
 free and easy rotation of the median digit. The articulation encroaches well upon the 
 dorsal surface. The great carina is relatively not as broad above as that of the second 
 
1G2 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 metacarpal, and extends obliquely downward and inward from the upper surface to 
 form the pronounced ridge between the surfaces for the sesamoids. This obliqueness 
 does not characterize the carina of metacarpal III in earlier types of ground-sloths. On 
 its inner side is a very broad articulating offset for the first phalanx, while along the 
 outer side this offset becomes perceptible only near the sesamoid surface. 
 
 SESAMOID BONES FOR METACARPAL III. 
 
 These elements are shown on plate 35, figure 2. The inner sesamoid is a broad 
 element, being compressed in dorso-palmar direction, with development of a prominent 
 flange extending downward along the entire inner side. This causes the palmar surface 
 to be broadly and deeply concave. The opposite face of the bone bears an articulating 
 facet for the inner offset of metacarpal III. This surface is longer than wide and is 
 concave. It is continuous with but demarcated by a right angle from an elongated 
 lateral facet articulating with the side of the carina of the metapodial. Dorsally the 
 facet for the offset is continued at a right angle with a narrow, transversely elongated 
 facet which articulates with the inner base of the first phalanx. In some specimens the 
 extent of the latter facet may be quite small. 
 
 Fig. 85. —Mylodon harlani Owen. 
 Phalanx I, digit III, right 
 manus, No. 24275 U. C. C.; 
 A, proximal view; B, outer 
 view; C, distal view. X 0.50. 
 Rancho La Brea Pleistocene. 
 
 The outer sesamoid differs entirely in shape from the inner sesamoid. It is a heavier 
 element than the latter; what it loses in transverse width it compensates in dorso-palmar 
 diameter. The inner surface of the sesamoid is concave between the facet for the outer 
 wall of the carina and the lower border of the bone. The inner articulating surface is 
 slightly concave in its long axis and rounds dorsally into an articulating surface for the 
 outer offset of the metapodial. In contrast to the inner sesamoid, this surface in the 
 sesamoid of the outer side is of smaller extent than the articulating surface for the side 
 of the carina. Above the inner articulation and separated from it by a sharp angle is 
 a broad, flat facet which joins with the outer base of the first phalanx. The two facets 
 may often be continuous, and the latter shows considerable variation in shape. It is 
 much broader in proximo-distal direction than the corresponding facet of the inner 
 sesamoid. 
 
 The average measurements (in millimeters) of 24 specimens of inner sesamoid of 
 
 Mylodon harlani are as follows: Greatest length, 46; greatest width, 29.9. 
 
 The average measurements (in millimeters) of 19 specimens of outer sesamoid of 
 
 Mylodon harlani are as follows: Greatest length, measured parallel to dorsal surface, 
 
 47.4; width through middle, 20.2; greatest depth, 32.8. 
 
 PHALANX I, DIGIT III, MANUS. 
 
 The depth of this phalanx (fig. 85, a, b, c) exceeds slightly the greatest width. It 
 is foreshortened in proximo-distal direction, as is characteristic of all the first phalangeal 
 elements. The proximal end (fig. 85 a) is traversed by a deep median groove which 
 extends the entire dorso-palmar depth and narrows below, permitting the phalanx to 
 articulate snugly with the metapodial. Along the inner side of the groove is a flattened 
 or concave surface conforming in outline to the surface on the inner side of the distal 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF IIANCIIO LA BREA. 163 
 
 carina of metacarpal III, with which it articulates. Below this surface and continued 
 with it at a sharp angle is a small facet for a sesamoid bone. On the outer side of the 
 median groove an extensive articulating surface is lacking. On the palmar side, how¬ 
 ever, there is a facet for a sesamoid bone which is larger than that of the opposite side. 
 
 1 he distal articulation (fig. 85 c) consists of two flattened condyles separated by a 
 broad groove. The inner one is convex transversely, and is of greater dorso-palmar 
 extent than the outer condyle. The median groove may be rugose, and this roughened 
 area may encroach upon the surface of the outer condyle. The distal extremity appar¬ 
 ently indicates but little movement between the first and second phalanges. The phalanx 
 is of greatest proximo-distal extent as measured over the outer surface (fig. 85 b). 
 
 The average measurements (in millimeters) of 40 specimens of Mylodon harlani are 
 as follows: Greatest depth, 62.7; greatest width, 58; proximo-distal diameter across 
 outer surface, 39.6. 
 
 Fig. 86 .—Mylodon harlani Owen. Phalanx II, digit III, right manus, No. 24276 U. C. C. A, proximal view; 
 B, dorsal view; C, outer view; D, distal view. X 0.50. Rancho La Brea Pleistocene. 
 
 PHALANX II, DIGIT III, MANUS. 
 
 Although this element (fig. 86, a to d) is directly comparable in length to the cor¬ 
 responding phalanx in the second digit of the manus, it greatly exceeds the latter in 
 stoutness. The proximal extremity (fig. 86 a) is greatly elongated in dorso-palmar 
 direction, with dorsal and palmar sides flattened and lateral borders convex. The two 
 articulating facets are separated by a thick and low median ridge, and the inner facet 
 is concave. At the distal extremity (fig. 86 d) the outer condyle of the trochlea is the 
 larger, possessing a greater dorso-palmar diameter than the inner condyle. The median 
 groove separating the two condyles is deep. The circumference of the outer condyle 
 describes a half circle, while that of the inner condyle describes approximately two- 
 thirds of a circle. The articulating surface of the latter condyle extends upon the inner 
 side of the distal trochlea. Although the outer side of the phalanx (fig. 86 c) is deeper 
 than the inner side, it is not so long in a proximo-distal direction as the latter. 
 
 The average measurements (in millimeters) of 42 specimens of Mylodon harlani are 
 as follows: Length measured across middle of inner side, 48.9; greatest depth of inner 
 condyle, 39.9; depth of proximal end, 57.8; greatest width of proximal end, 47.9. 
 
 PHALANX III, DIGIT III, MANUS. 
 
 Viewed from the outer side (fig. 87 a), the proximal articulation of the phalanx is 
 seen to describe a semicircle, with the dorsal portion of the element projecting far back¬ 
 ward. The proximal surface of this projecting portion is deeply indented. The inner of 
 the two articulating surfaces is deeply concave, of smaller dorso-palmar extent than the 
 outer, and is well inclosed along the inner side. In this phalanx the principal axis of 
 the proximal articulation is oblique to the median vertical plane passing through the 
 element, and the upper end of the axis lies to the outside of the plane, whereas in the 
 terminal segment of the second digit, manus, this axis is vertical. 
 
164 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 The base of the claw-process is similar to that of the ungual phalanx, digit II, 
 manus, with the distal discoid area more prominently developed than in the latter. 
 This area (fig. 87 c) is situated closer to the inner than to the outer border of the base. 
 Immediately behind the area and on each side of the base are the two subungual fora¬ 
 mina. The inner of these is much the larger and is often divided into two by a bony 
 partition. Probing the openings thus formed indicates that the lateral one leads into 
 the space between bony sheath and claw-process, while the medial one permits an 
 entrance into the base of the claw-process. The bony sheath which incases the claw- 
 process, but is well separated from its sides, arises from the base. In specimens from 
 Rancho La Brea the sheath is preserved in various stages of completeness, and a num¬ 
 ber of ungual phalanges are present in which it is still intact over the top, reaching 
 forward some 60 mm. from the proximal end. The sheath is often incompletely fused 
 with the proximal end along the dorsal margin. 
 
 Fig. 87 .—Mylodon harlani Owen. Ungual phalanx, digit III, right manus, No. 24277 U. C. C. A, outer view with 
 cross-section of claw-process; B, dorsal view; C, ventral view. X 0.50. Rancho La Brea Pleistocene. 
 
 The claw-process is higher and relatively more narrow transversely than that in the 
 terminal phalanx of digit II, manus. The palmar portion of the process not incased is 
 more prominently convex than that of the latter phalanx. For a distance of approxi¬ 
 mately 70 mm. from the free end of the phalanx the dorsal surface of the claw-process 
 is grooved (fig. 87 b) and the surface of this groove is directed toward the inner side. 
 The free end of the process reaches a level indicated by the horizontal plane of the sub¬ 
 ungual base. Sometimes it extends slightly below this level. 
 
 Table 82. —Average measurements (in millimeters) of 16 specimens of ungual phalanx, 
 
 digit III, manus, of Mylodon harlani. 
 
 Greatest length, measured from most projecting portion of proximal end to end of claw-process. 174.1 
 
 Distance from proximal end of subungual base to end of claw-process. 141.2 
 
 Greatest proximo-distal diameter of subungual base. 73.5 
 
 Vertical dorso-palmar distance between discoid area to dorsal surface of claw-process. 57.2 
 
 Width of proximal end. 55 
 
 Transverse width of claw-process at distal end of subungual base. 31.3 
 
 METACARPAL IV. 
 
 This strongly developed metacarpal (fig. 88, a to e) is the longest of the anterior 
 metapodial series, although not as heavy as metacarpal III. The proximal and distal 
 extremities are quite heavy and the middle portion of the shaft may become relatively 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 165 
 
 slender. The outer or ulnar side of the metapodial is longer than the inner or radial 
 side, due to the oblique truncation of the proximal end. The latter extremity is trian¬ 
 gular in shape (fig. 88 d), and along the outer side is devoted to articulation with the 
 unciform. This surface extends practically the entire dorso-palmar depth of the bone, 
 and in this direction has a sigmoid curvature. It extends internally along the dorsal 
 border and touches the usually concave facet for the third metacarpal. In such speci¬ 
 mens the two articulating surfaces are distinguished by an intervening ridge. In other 
 metapodials, however, the two are merged at this point. Along the palmar border a 
 small facet is demarcated by an obtuse angle from the surface for the unciform and 
 articulates also with metacarpal III. A narrow facet on the outer side of metacarpal 
 IV (fig. 88 c) and articulating with the fifth metapodial is continuous with but demar¬ 
 cated by a sharp angle from the surface for the unciform. 
 
 Fig. 88 .—Mylodon harlani Owen. Right metacarpal IV, No. 24278 U. C. C. A, inner view; B, dorsal view; 
 C, outer view; D, proximal view; E, distal view. X 0.50. Rancho La Brea Pleistocene. 
 
 At the distal end (fig. 88 e) a prominent carina is not developed and the articulating 
 surface indicates the reduced and functionless fourth digit. A narrow, convex articu¬ 
 lating surface is present, which is elongated in dorso-palmar axis and bears a small olf- 
 set, also for the first digital element, along its inner side. Below the distal convexity 
 a prominent ridge is present and separates the two surfaces for the sesamoids. Hie 
 outer of these surfaces is the larger one. The metapodial develops a large tuberosity 
 to the outer dorsal side of the distal articulation. The latter surface is directed well 
 to the inner side, which causes the rudimentary toe to assume a position iar to the 
 inner side of a median vertical plane through the metapodial. 
 
 Table 83. —Average measurements {in millimeters) of 30 specimens of metacarpal 
 
 IV of Mylodon harlani. 
 
 Length. 
 
 Dorso-palmar depth of proximal extremity, measured along outer side 
 
 Least depth of shaft. 
 
 Greatest extent of distal articulation. 
 
 Width of proximal end, measured normal to lateral surface. 
 
 Least width of shaft. 
 
 Greatest width of distal extremity. 
 
 121.4 
 
 62.3 
 33.8 
 
 59.3 
 47.2 
 30 
 
 49.4 
 
 SESAMOID BONES FOR METACARPAL IV. 
 
 Because of the resemblance which exists between the distal extremities of meta¬ 
 carpal IV and metatarsal IV, there is also some similarity between the sesamoid bones 
 which articulate with the metapodials. The outer element is broader than the innei, 
 
166 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 as is also true of the sesamoids articulating with metatarsal IV. The outer sesamoid 
 bone (fig. 89, a, b, c) differs from the corresponding element of the latter metatarsal in 
 the distinct deflection of a portion of the articulating surface for the metapodial upon 
 the outer side of the sesamoid. This is due to the presence in the anterior metapodial 
 of a more defined or ridge-like outer border to the surface for the sesamoid, as con¬ 
 trasted with the fourth metatarsal. The outer sesamoid and the inner lack entirely 
 the small dorsal facet which in the posterior sesamoids indicate an articulation with 
 the first phalanx of the fourth digit. This absence is to be associated with the lack 
 of any specimens in the collections of phalanx I, digit IV, manus, that have fused with 
 sesamoid bones—a fusion which is known to occur in the pes. 
 
 Fig. 89. —Mylodon harlani Owen. A, B, and C, outer sesamoid, meta¬ 
 carpal IV, left manus, No. 1462-L-4. X 1.0. Rancho La Brea 
 Pleistocene. 
 
 The inner sesamoid bone (fig. 90, a, b, c), although approximating in size the cor¬ 
 responding element articulating with metatarsal IV, is noticeably thicker. It differs 
 from the latter also in the more extensive articulating surface, not restricted to the 
 dorsal half of the element. The portion of the surface articulating with the side of the 
 ridge in the metapodial separating the two sesamoid bones is more extensive in dorso- 
 palmar direction than that touching the surface at the side of the ridge. 
 
 Fig. 90 .—Mylodon harlani Owen. A, B, and C, inner sesamoid, 
 metacarpal IV, left manus, No. 1468-L-5. X 1.0. Rancho 
 La Brea Pleistocene. 
 
 PHALANX I, DIGIT IV, MANUS. 
 
 The first phalanx of the fourth digit (fig. 91, a, b, c) is peculiarly twisted out of 
 the regular shape, as is shown by the fact that the median vertical plane is not normal 
 to the plane of the proximal end, but forms an acute angle with that plane. When the 
 phalanx is viewed from the proximal end, the principal axis of the latter face is not 
 normal to the base, but joins it to form an acute angle. The dorsal end of the axis is 
 to the outer side of the normal. These distortions assist in the pronounced inward 
 direction of the rudimentary digit. 
 
 The proximal articulation (fig. 91 a) consists of a broad but shallow concavity, 
 continuous along the inner lower side, with an offset. The latter surface is continued 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF IIANCIIO LA BREA. 167 
 
 upon the palmar side at a right angle, with a small facet for a sesamoid. On the distal 
 surface (fig. 91 c) the articulating facet is much wider transversely than deep. It is 
 concave in the former direction, convex to flattened in the latter, and the inner half 
 is the more extensive area. The outer side (fig. 91 b) is decidedly longer in proximo- 
 distal extent than the inner side. On the upper half of the outer face there is almost 
 invariably present a fairly large nutritive foramen. Leading from the foramen is a 
 groove which extends to the dorsal surface. It is plainly indicated in some specimens 
 and more faintly so in others. These characters are not present, or very rarely so, on 
 the outer face of phalanx I, digit IV, pes. The palmar border of the phalanx is deeply 
 notched at the middle. 
 
 The average measurements (in millimeters) of 31 specimens of Mylodon harlani are 
 as follows: Dorso-palmar diameter 45.2; greatest width, 39.1; greatest proximo-distal 
 extent of outer side, 29.8. 
 
 Fig. 91 .—Mylodon harlani Owen. Phalanx I, digit IV, right 
 manus, No. 24279 U. C. C. A, proximal view; B, outer 
 view; C, distal view. X 0.50. Rancho La Brea Pleisto¬ 
 cene. 
 
 Fig. 92. —Mylodon harlani Owen. Terminal phalanx, 
 digit IV, right manus, No. 1484-R-2. A, proxi¬ 
 mal view; B, lateral view; C, dorsal view. X 0.50 
 Rancho La Brea Pleistocene. 
 
 RUDIMENT REPRESENTING PHALANGES II AND III, DIGIT IV, MANUS. 
 
 This element (fig. 92, a, b, c) is very similar to the corresponding rudiment of the 
 fourth digit of the pes, but appears to be, on the average, somewhat smaller than the 
 latter. At the proximal end the articulating surface varies considerably in extent. This 
 segment has also undergone distortion, as is shown by the fact that if a perpendicular 
 be taken to the base of the proximal extremity, the principal axis of this face will be 
 oblique and the dorsal end of the axis will lie to the right or left of the perpendicular, 
 according as the element pertains to the right or left manus, respectively. 
 
 The average measurements (in millimeters) of 9 specimens of Mylodon harlani are 
 as follows: Greatest length, 36.5; depth of proximal end, measured normal to base, 30; 
 width of proximal end, 26.5. 
 
 Table 84. —Average measurements (in millimeters) of 84 specimens of metacarpal 
 
 V of Mylodon harlani . 
 
 Length. 
 
 Width of proximal extremity. 
 
 Least width of shaft. 
 
 Width of distal extremity. 
 
 Greatest depth of proximal extremity, measured across unciform facet 
 
 Least depth of shaft. 
 
 Depth of distal extremity. 
 
 METACARPAL V. 
 
 Metacarpal V (fig. 93, a to e) is the second in length of the anterior series of meta- 
 podials. The shaft is of small extent, and the metapodial is chiefly constituted by the 
 proximal and distal extremities, which are large and heavily developed. 1 he proximal 
 end is truncated obliquely by the articulating surface for the unciform. This surface 
 is frequently continuous with, although demarcated by, a ridge from a small concave 
 facet situated at the tip, which articulates with the cuneiform. Along the inner side 
 of the proximal extremity is the facet for the adjoining metacarpal, continuous with 
 that for the unciform over a rounded ridge. 
 
 107.8 
 48.8 
 32.5 
 40. fi 
 51 
 
 26.7 
 
 54.4 
 
168 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 The massive distal extremity (fig. 93 c) bears along the inner margin a flattened 
 facet, elongated in dorso-palmar axis, which articulates with the rudimentary fifth digit. 
 Occasionally a very small facet for a sesamoid is present below this surface. 
 
 PROXIMAL RUDIMENTARY ELEMENT, DIGIT V, MANUS. 
 
 This is a much reduced segment (fig. 94, a to d) bearing on the proximal end a 
 slightly concave facet which is elongated in dorso-palmar extent and articulates with 
 the metapodial. A very small facet for a distal nodule is situated at the dorso-internal 
 angle of the distal face and is directed upward as well as distally. This facet is directed 
 more upward and less to the distal side than is the corresponding surface on the prox¬ 
 imal rudimentary element of digit V, pes. Several specimens in the collections of the 
 Los Angeles Museum show a fusion with the distal nodule (fig. 94, e and f) and indicate 
 that the latter was a tiny element. 
 
 Fig. 93. —Mylodon harlani Owen. Right metacarpal V, No. 24280 U. C. C. A, outer view; B, dorsal view; C, distal 
 view; D, inner view; E, ventral view. X 0.50. Rancho La Brea Pleistocene. 
 
 The average measurements (in millimeters) of 9 specimens of Mylodon harlani are 
 as follows: Greatest depth of proximal end, 34.6; greatest width of proximal end, 24.9; 
 greatest proximo-distal diameter, 21.5. 
 
 COMPARISON OF MANUS. 
 
 Mylodon harlani agrees closely with the South American M . robustus in the structure 
 of the manus. While the individual elements differ somewhat in shape, no change has 
 taken place in the major relationships of the bones or in the reduction of segments. 
 
 In dorsal view of manus (plate 35, fig. 1) the pisiform does not always come to 
 view. As in M. robustus, the scaphoid sends downward a process, and the trapezium 
 is rarely represented by a separate element. As discussed above (see p. 149) the trape¬ 
 zium has not fused with the scaphoid, but has joined with metacarpal I to form an 
 element in M. harlani similar in form to that in M. robustus. In both North American 
 and South American mylodonts the claw-bearing phalanges are found in the three inner 
 digits, while the two lateral digits are rudimentary. In digit I of both forms the first 
 and second phalanges have fused, presumably to form a single segment. The distinct 
 inequality in size between ungual phalanges of the second and third digits can not be 
 regarded as a valid difference between M. robustus and M. harlani, for Stock (1917c) 
 has shown that in Owen’s figures (R. Owen, 1842, plates 15 and 21) of the manus and 
 pes of the former species the second and third phalanges of digit III, pes, have been 
 interchanged with the corresponding phalanges of digit II, manus. 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF HANCIIO LA BREA. 169 
 
 The manus of Scelidotherium leptocephalum differs from that of M. liarlani in the 
 greater reduction of the innermost digit. In the hand of S. magnum, as figured by 
 Winge (1915, plate 39), the inner digit and metacarpal I are comparable in develop¬ 
 ment to those in the North American mylodont. 
 
 Catonyx, a near relative of Scelidotherium , resembles M. liarlani in the retention of 
 ungual phalanges in digits I, II, and III and in the rudimentary structure of digits IV 
 and V. According to Lund, and to Schulthess (1920, p. 77), the trapezoid and magnum 
 may unite in the manus of Catonyx. A union of these two elements has never been 
 observed in the North American form. In C. giganteus, as shown by Winge (1915, 
 plate 31, fig. 1), metacarpals IV and V are more slender than the corresponding meta- 
 podials of Mylodon. 
 
 INNOMINATE BONE. 
 
 This element in Mylodon liarlani, shown in plates 35, 36, and 37, is similar in many 
 of its structural features to that of the South American species M. robustus. The great 
 wing-like ilium in Mylodon is even more prominent than in Nothrotherium, and the upper 
 anterior border is carried much farther forward than in the latter genus. When the os 
 innominatum is viewed from the dorsal or from the posterior side (plates 36 and 37) it 
 is seen that the ridge extending backward from the inner anterior border of the ilium 
 does not approach the spinal ridge as closely as in M. robustus. The ridge formed by 
 
 Fig. 94 .—Mylodon harlani Owen. A to D, 
 phalanx I, digit V, right manus, 
 No. 24281 U. C. C.; E and F, pha¬ 
 lanx I, fused with rudimentary pha¬ 
 lanx II, No. 1485—R-5. A, proximal 
 view; B, outer view; C, inner view; 
 D, distal view; E, proximal view; 
 F, outer view. X 0.50. Rancho La 
 Brea Pleistocene. 
 
 the fusion of the outer ends of the transverse processes of the sacral vertebrae does not 
 appear to diverge posteriorly as much as in the latter form. Along the inner base of 
 this ridge may be seen the posterior sacral foramina. These foramina increase in size 
 posteriorly. At the base of the anterior portion of the spinal crest is a series of three 
 or more deep depressions into which enter canals that extend through the neurapophy- 
 sial plate and open inward above the large foramina for the lumbar nerves. The inner 
 surface of the neurapophysial plate between each of these foramina and the large open¬ 
 ings for the lumbar nerves is usually marked by a distinct groove. In a specimen of a 
 young individual of M. harlani, No. 1719-17, the transverse processes of the two ante¬ 
 rior sacral vertebrae are seen to establish the principal connection with the ilium, while 
 those of the last two vertebrae connect with the ischium. The ends of the transverse 
 processes of the intermediate vertebrae form the inner boundary of the sacro-ischiadic 
 foramen. 
 
 The inner or anterior and the outer or posterior surfaces of the ilia are shown in 
 plates 36 and 37. In anterior view a well-marked groove is seen to extend from the 
 inner end of the inferior border of the iliac wing to the border of the pelvic opening. 
 The acetabulum is round or oval in shape. The sulcus for the round ligament incises 
 the articulating surface as in M. robustus. It is interesting to note that the depression 
 varies considerably in size and in many specimens is actually smaller than in the os 
 innominatum of Nothrotherium. In No. 1719—1 the obturator border of the acetabulum 
 is notched, but no depression extends toward the middle of the femoral articulating 
 surface. The sacro-ischiadic foramen is ovate in form. The obturator foramen is rela¬ 
 tively smaller than in Nothrotherium. The pubis and ischium resemble the corresponding 
 
Table 85. —Mylodon harlani, measurements of pelvis (in millimeters). 
 
 170 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 CD 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 rH 
 
 X 
 
 CM • 
 
 
 
 
 
 
 
 CO 
 
 
 
 
 
 
 05 
 
 
 
 
 . 
 
 rH 
 
 
 X 
 
 
 05 • 
 
 
 IO 
 
 
 
 
 CO 
 
 tH 
 
 
 
 
 
 
 
 
 
 
 X 
 
 
 o 
 
 01 
 
 o • 
 
 
 05 
 
 
 
 
 oo 
 
 iO 
 
 
 
 
 
 
 10 
 
 
 
 
 
 rH 
 
 
 rH 
 
 rH 
 
 rH 
 
 
 
 
 
 
 
 
 
 
 
 
 
 CO 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 (M 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 rH 
 
 
 Tt< 
 
 
 
 
 
 
 
 X 
 
 CO 
 
 t- 
 
 
 rH 
 
 
 
 
 
 
 
 X 
 
 Ttt 
 
 >2 a> 
 
 
 o 
 
 o 
 
 
 
 
 
 
 
 X 
 
 ‘0 
 
 rH 
 
 ^ • 
 
 
 
 
 
 
 05 
 
 CM 
 
 CO 
 
 Z *-c 
 
 
 t- 
 
 CO 
 
 
 
 
 
 
 X 
 
 X 
 
 05 
 
 rH 
 
 CM • 
 
 
 
 
 
 X 
 
 X 
 
 IO 
 
 IO 
 
 r- 
 
 
 
 
 
 
 
 
 
 rH 
 
 rH 
 
 
 rH 
 
 rH • 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 rH 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 rH 
 
 
 
 
 
 
 
 
 
 
 
 CO 
 
 X 
 
 X • 
 
 
 X 
 
 
 
 
 CO 
 
 rH 
 
 CM 
 
 fS cL 
 
 
 
 
 05 
 
 
 
 
 
 CM 
 
 
 rH 
 
 rH 
 
 CM • 
 
 
 X 
 
 
 
 
 05 
 
 05 
 
 X 
 
 /h *-l 
 
 
 iO 
 
 
 05 
 
 
 
 
 
 X 
 
 
 05 
 
 o 
 
 rH • 
 
 
 05 
 
 
 
 
 X 
 
 IO 
 
 IO 
 
 
 
 TtH 
 
 
 -t 
 
 
 
 
 
 rH 
 
 
 
 rH 
 
 rH 
 
 
 
 
 
 
 
 
 
 t-H 
 
 
 
 
 CO 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 03 
 
 
 
 
 
 
 
 
 
 CM 
 
 
 rH 
 
 CO 
 
 
 
 !> 
 
 
 
 CO 
 
 IO 
 
 X 
 
 
 O 05 
 
 
 co 
 
 
 
 
 
 
 
 iO 
 
 to 
 
 O 
 
 rH 
 
 IO • 
 
 
 rH 
 
 
 
 CM 
 
 rH 
 
 o 
 
 rH 
 
 
 
 CO 
 
 
 OI 
 
 
 
 
 
 X 
 
 CM 
 
 o 
 
 X 
 
 rH 
 
 
 05 
 
 
 
 
 05 
 
 CO 
 
 l> 
 
 rH 
 
 
 Tji 
 
 
 iO 
 
 cc 
 
 
 
 
 
 rH 
 
 rH 
 
 rH 
 
 rH 
 
 rH 
 
 rH 
 
 
 
 
 
 CM 
 
 
 
 
 CO 
 . 1 
 
 
 
 
 
 
 
 l- 
 
 H 
 
 rH 
 
 
 
 
 CO 
 
 X 
 
 H 
 
 CM X 
 
 
 
 
 
 CO 
 
 O 05 
 
 
 
 
 iO 
 
 o 
 
 co 
 
 IO 
 
 rH 
 
 CM 
 
 rH 
 
 
 
 X X 
 
 05 
 
 CO 
 
 05 rH 
 
 
 
 
 CM 
 
 4 
 
 
 
 
 
 X 
 
 CO 
 
 
 X 
 
 05 
 
 X 
 
 CM 
 
 
 
 O CM 
 
 IO 
 
 05 
 
 X CO 
 
 
 
 
 iO 
 
 CO 
 
 
 
 
 
 IO 
 
 
 co 
 
 rH 
 
 IO 
 
 rH 
 
 rH 
 
 
 
 rH rH 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 rH 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 . J 
 
 
 
 X 
 
 
 
 X 
 
 t- 
 
 IO 
 
 
 
 05 
 
 
 
 X 
 
 
 H H 
 
 
 
 T)t 
 
 IO 
 
 
 O 05 
 
 
 co 
 
 CM 
 
 IO 
 
 X 
 
 X 
 
 05 
 
 rH 
 
 o 
 
 CM 
 
 CO 
 
 X 
 
 rH O 
 
 X 
 
 X 
 
 X co 
 
 
 
 CM 
 
 o 
 
 
 
 
 IO 
 
 CM 
 
 TfH 
 
 CO 
 
 X 
 
 Ol 
 
 rH 
 
 X 
 
 CM 
 
 o 
 
 CM 
 
 O CM 
 
 IO 
 
 05 
 
 i- »o 
 
 
 
 X 
 
 CO 
 
 
 rH 
 
 
 
 
 »o 
 
 CO 
 
 Tt< 
 
 X 
 
 rH 
 
 CO 
 
 rH 
 
 rH 
 
 rH 
 
 rH 
 
 rH rH 
 
 
 
 
 
 
 
 
 
 CO 
 . 1 
 
 
 
 
 
 Ht 
 
 IO 
 
 
 IO 
 
 rH 
 
 
 
 CM 
 
 X 
 
 
 
 IO 05 
 
 05 
 
 CM 
 
 
 iO 
 
 CM 
 
 O 05 
 
 
 rH 
 
 05 
 
 Ol 
 
 o 
 
 CO 
 
 »o 
 
 IO 
 
 10 
 
 X 
 
 IO 
 
 05 
 
 IO IO 
 
 H 
 
 X 
 
 CM tO 
 
 iO 
 
 iO 
 
 br 
 
 i 
 
 CO 
 
 
 
 X 
 
 co 
 
 
 rH 
 
 rH 
 
 05 
 
 
 X 
 
 X 
 
 05 
 
 rH 
 
 CM X 
 
 
 X 
 
 05 X 
 
 05 
 
 H) 
 
 X 
 
 b* 
 
 
 Tf< 
 
 H 
 
 IO 
 
 
 X 
 
 X 
 
 »o 
 
 rH 
 
 rH 
 
 
 rH 
 
 rH rH 
 
 
 
 
 rH 
 
 X 
 
 
 
 
 
 
 
 
 co 
 
 
 
 
 
 
 
 
 
 
 
 
 
 co 
 
 
 
 
 
 iO 
 . 1 
 
 
 
 
 CO 
 
 
 
 
 
 
 t> 
 
 X 
 
 
 X • 
 
 
 CO 
 
 
 
 
 X 
 
 t> 
 
 
 O 05 
 
 
 <N 
 
 
 iO 
 
 
 
 
 
 to 
 
 X 
 
 CO 
 
 IO 
 
 I-- . 
 
 
 rH 
 
 
 
 
 rH 
 
 X 
 
 
 
 
 05 
 
 
 r- 
 
 
 
 
 
 rH 
 
 X 
 
 o 
 
 CM 
 
 CM • 
 
 
 05 
 
 
 
 
 05 
 
 CO 
 
 
 rH 
 
 
 
 
 iO 
 
 CO 
 
 
 
 
 
 rH 
 
 rH 
 
 rH 
 
 rH 
 
 rH 
 
 
 
 
 
 
 
 
 
 
 
 CO 
 
 
 
 
 
 
 
 05 
 
 
 »o 
 
 X 
 
 t'T * 
 
 
 CO 
 
 
 
 
 
 
 
 • 1 
 
 O 05 
 
 IO 
 
 Ol 
 
 
 
 
 
 
 rH 
 
 05 
 
 o 
 
 rH 
 
 05 
 
 o • 
 
 
 CM 
 
 
 
 o 
 
 
 
 
 
 I- 
 
 
 
 
 
 
 05 
 
 CM 
 
 rH 
 
 o 
 
 rH 
 
 rH 
 
 
 05 
 
 
 
 CM 
 
 
 
 
 £ 4 
 
 o 
 
 
 
 
 
 
 
 IO 
 
 rH 
 
 rH 
 
 rH 
 
 rH 
 
 rH • 
 
 
 
 
 
 X 
 
 
 
 
 
 •» 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 rH 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 CO 
 
 6 d® 
 
 
 
 
 X 
 
 
 
 
 
 t> 
 
 IO 
 
 
 
 
 
 iO 
 
 rH 
 
 
 
 
 
 rH 
 
 
 
 
 05 
 
 
 
 rH 
 
 
 rH 
 
 rH 
 
 
 
 H • 
 
 
 rH 
 
 X rH 
 
 O 
 
 H 
 
 I> 
 
 X 
 
 CM 
 
 
 
 
 
 rH 
 
 
 
 X 
 
 
 rH 
 
 X 
 
 
 
 rH • 
 
 
 05 
 
 X CO 
 
 CO 
 
 rH 
 
 X 
 
 IO 
 
 CO 
 
 
 
 
 
 CO 
 
 
 
 rH 
 
 
 rH 
 
 rH 
 
 
 
 rH 
 
 
 
 
 IO 
 
 X 
 
 
 
 
 
 
 
 H 1 
 
 
 
 rH 
 
 
 
 
 
 
 
 
 
 
 CO 
 
 
 
 
 
 <N 
 
 
 
 CO 
 
 
 
 
 X 
 
 
 IO 
 
 
 IO 
 
 
 CO l> 
 
 CO 
 
 X 
 
 CO 
 
 
 
 X 
 
 X 
 
 CM 
 
 . 1 
 
 O 05 
 
 Ol 
 
 X 
 
 X 
 
 Ht 
 
 10 
 
 rH 
 
 CM 
 
 I> 
 
 X 
 
 
 rH 
 
 
 rH IO 
 
 CO 
 
 4 
 
 co 
 
 
 
 rH 
 
 05 
 
 rH 
 
 GO 
 
 IO 
 
 
 CO 
 
 rH 
 
 rH 
 
 05 
 
 rH 
 
 CM 
 
 CM 
 
 CM 
 
 
 CM X 
 
 CO 
 
 05 
 
 r- th 
 
 
 o 
 
 X 
 
 IO 
 
 CO 
 
 
 o 
 
 -ft 
 
 b 1 
 
 IO 
 
 CO 
 
 H 
 
 X 
 
 X 
 
 CO 
 
 rH 
 
 rH 
 
 rH 
 
 
 rH rH 
 
 
 
 
 X 
 
 
 
 rH 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 rH 
 
 IO 
 
 
 o 
 
 CO 
 
 CO 
 
 
 
 rH 
 
 X 
 
 
 CM 
 
 IO 
 
 IO X 
 
 CM 
 
 05 
 
 X rH 
 
 
 rH 
 
 
 
 H 
 
 
 CO 
 
 X 
 
 05 
 
 iO 
 
 05 
 
 X 
 
 o 
 
 CM 
 
 
 t- 
 
 4 
 
 CM 
 
 CM rH 
 
 rH 
 
 X 
 
 rH ^H 
 
 
 CO 
 
 
 
 X 
 
 
 »Q 
 
 CO 
 
 CM 
 
 
 CM 
 
 rH 
 
 05 
 
 CO 
 
 CM 
 
 CM 
 
 O 
 
 CM 
 
 rH 
 
 X 
 
 05 
 
 05 
 
 
 X 
 
 
 
 CO 
 
 
 O 
 
 H 
 
 H 
 
 OJ 
 
 
 X 
 
 X 
 
 IO 
 
 rH 
 
 rH 
 
 rH 
 
 rH 
 
 rH rH 
 
 
 
 
 
 CM 
 
 
 
 
 
 rH 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 to § 
 n -> 
 C 43 
 
 F-. • — 
 
 o 
 
 73 
 
 73 
 
 -H 
 
 GO 
 
 O 
 
 h-> 
 
 d 
 
 <D 
 
 Fh 
 
 o 
 
 O 
 
 Fh 
 0> 
 73 
 
 Fh 
 
 O 
 
 X 
 
 : s 
 
 . o 
 
 S T3 
 
 £ £ 
 o 3 
 o? 2 
 
 u a 
 
 ■3 .2 
 
 !h —i 
 
 O 
 
 ^ o 
 
 .2 ’S 
 
 o 
 
 +=> > 
 p > 
 73 ® 
 
 o 
 
 bC 
 
 .s 
 
 '3 
 
 <d 
 
 a 
 
 o 
 
 O bfl 
 
 x -3 
 g,* 
 
 Jr © 
 
 o.2 
 
 A — 
 
 4 ° 
 
 o © 
 
 -g 
 
 2 _g 
 
 CD X 
 
 rJl 73 
 
 t3 
 
 _ d 
 
 O Fh 
 
 CO d) 
 
 3 O 
 
 c3 *C 
 r* <D 
 
 b d 
 
 o S 
 
 F. 75 
 ©f-H 
 
 e I 
 g £ 
 
 <53 
 
 2° 
 
 ► Fh 
 
 a o 
 
 c -c 
 
 1 o 
 S'® 
 
 CO 
 
 2 4 
 
 Fh 
 
 GO -4-J 
 
 d 
 
 ° > 
 43 A 
 ■P 0 
 
 be 
 
 d ^ 
 
 CD 
 
 ►-3 
 
 Fh 
 
 O 
 
 T3 
 
 Fh 
 
 c 
 
 42 
 
 bC o 
 
 £ 
 
 *b ^ 
 d o 
 
 8.-H 
 
 o ^ 
 
 o 
 
 a 
 
 4 '■ ■£ 4 
 
 u . ? fc. 
 
 ® So ? 
 
 r x> u 7 
 
 O © <D 73 
 
 *H *H 
 
 s S S S 
 ► a _. 
 u .2 S 
 
 oo o 
 
 o .2 £ 
 
 ® 0 «4 
 
 Ilf I 
 
 F 2 £ 2 
 
 or 1 Fh 
 
 73 r^ vr 0 
 
 72 CO bC 
 
 S o .S ^ 
 
 >>•* 3 ° 
 
 GO X ~ 
 
 X . O 
 
 © . ^ 
 ~ CO 
 GO CO t-i 
 CG © Q 
 
 o ’-3 -o 
 
 Fh • .H 
 
 © 
 
 c3 
 
 o X 
 
 Fh 
 
 r 
 
 b 42 
 " P 
 
 42 
 
 3 o 
 
 A Fh 
 
 CO 
 
 O 
 
 C, 
 
 P 
 
 a 
 
 £ ~2 C2 
 
 O 73 
 
 +— Fh —* 
 J Cfl -P 73 
 
 d Q d © 
 
 ^ a ® 
 
 ® p 7 
 CO 
 O 
 
 © 
 
 u 
 4 
 
 CO 
 
 O d 
 w 2 
 
 CD ^ 
 
 ft O 
 
 ® GO 
 
 g o.H © 
 
 4 o-§"H 
 
 -24 g. o 
 
 t. to 32 
 
 (3J '-t-H 
 
 ^ 2 43 2 
 
 o 
 
 73 
 CD 
 
 3 
 
 1 O » 
 b t» 73 
 O O 
 o 2 h 
 n ci P 
 
 -*• 4h (« 
 
 ft 33 
 
 CO 
 
 O o3 
 O P 
 
 - •— 
 
 ^*-i o 
 
 73 t? 
 <D O 
 
 *— 
 
 33 
 
 11 
 
 . a 
 
 2 ° 
 
 *7 
 
 4 
 
 c3 
 
 -p> 
 
 <D 
 
 D 
 
 c3 ■ 
 
 O 
 
 Q 
 
 73 • 
 
 73 
 
 ° d . 
 43 43 
 
 -P> Fh '♦P 
 
 ^ O 73 
 
 i2 O 
 5 F-< 
 
 X <D 
 o3 ^ 
 
 a "Si 
 
 % So 
 
 P S CO 
 4P Fh i 
 
 c3 b o 
 
 ^ Cfi 4 
 
 1 s § 
 * £.: 
 b o 
 
 © 
 o 
 
 be o 
 
 “ 2 
 
 73 
 
 b 
 
 3-8 
 
 d) *-< 
 
 ^ o 
 
 <D 
 O 
 
 «4-H -f-1 
 
 o -V 
 ** {-> 
 4.2 
 2 S 
 
 i- -+e 
 o g 
 
 O M 
 
 4 .2 
 
 c3 o 
 
 '3 M 
 
 •S £ 
 
 .2 o 
 
 I 
 
 2 o 
 
 CO 
 
 O 
 
 £3 
 
 P P 
 
 73 O 
 
 03 .2 43 
 
 Q 
 
 p 
 
 CD 
 
 _ a 
 
 -«a 2 
 
 to d 1 o 
 
 <D O o 7-. 
 
 ■’•g S.S 
 $ -e -3 
 
 X 
 
 73 
 
 75 
 
 £ ■£ 
 
 O 7 
 
 T3 o 
 
 3 S 
 A § 
 m a 
 » o 
 S ^ 
 A 
 
 C 
 
 O O 
 
 £ S 
 £ -2 
 3.2 
 2 "O 
 24 
 
 4 7 
 CD 
 X\ 
 
 t-t 
 
 <D 
 
 73 
 
 f-t 
 
 O 
 
 Fh 
 
 # o 
 
 *Fh 
 
 ft 
 
 P 
 
 CO 
 
 O o 
 
 fe S 
 
 d d 
 
 <D S 
 
 a a 
 
 03 <D 
 
 Fh Fh 
 
 o b 
 
 GO 
 
 c3 
 
 o 
 
 3 
 
 bC 
 ‘o 
 
 bC 
 <D 
 
 Fh 
 
 o 
 
 d*' 73 
 
 S -2 
 
 1 
 
 : O 
 
 bfi 
 
 d 
 
 D 
 
 a «*- 2 
 2 d « 
 a d 
 
 52 co „ 
 
 d „ 4P 
 
 ^Ss 
 
 a 
 a 
 3 
 
 a 
 0) 
 a 
 
 ? to 
 
 o a> 
 
 to 
 
 3 > 
 to «1 
 ^ © 
 
 o o 
 
 O 2 B 
 - 3 ei 
 
 ;£ 3 "£ 
 4 ^ 
 
 .2 e3 ® 
 b -p* © 
 -d m 
 © S 03 
 Art© 
 
 0 
 
 co b 
 CD o 
 
 <D ^ 
 
 Fh 
 
 £ a 4 
 £ a a 
 a © a 
 
 2 £ 
 
 >• £ *« 
 1 - ■£ 0 
 
 0 c 10 
 
 T3 
 
 <D 
 
 Fh 
 
 3 
 
 CO 
 
 c3 
 
 <D 
 
 2 
 
 a 
 
 .2 § 
 gt; 
 
 u ft 
 *b co 
 4^ 
 
 ft “ P 73 
 
 d ^ ft a 
 r s - ^ 
 c° 
 
 CO Fh 
 
 o 
 
 O 73 
 
 c3 
 
 X . 
 
 2 03 
 ft d 
 
 Fh 
 
 O 
 
 o 73 
 
 b 4)4 b 
 +2 X X 
 73 c3 t 3 
 .b -*-> c3 +-> 
 
 o 
 
 o b 
 
 CO £ 
 
 co 
 
 2 73 
 d <D 
 
 X g 
 
 a b 
 
 .£ bJD 
 to d 
 O (D 
 
 Ah 1-1 
 
 2 rt 
 
 2 » 
 ra to 
 
 M 
 
 e3 
 
 to t. 
 
 C ® 
 
 ® 4 
 
 «H « 
 
 A t ' 
 
 £ § 
 to — 
 to ^ 
 
 2 « 
 c3 a> 
 
 s 4 
 
 H ^ 
 
 cj 4 
 
 to © 
 
 c t) 
 
 >* S 
 ® £ 
 
 0 
 
 
 
 u-< 
 
 o 
 
 
 0 
 
 c3 
 
 
 c3 
 
 Fh 
 
 
 Fh 
 
 rQ 
 
 
 X 
 
 CJ 
 
 
 a> 
 
 
 
 
 Fh 
 
 
 Fh 
 
 <D 
 
 
 <D 
 
 > 
 
 
 > 
 
 -H 
 
 
 
 GQ 
 
 
 GQ 
 
 73 
 
 
 73 
 
 r-H 
 
 
 rH 
 
 U-t 
 
 
 <H-< 
 
 o 
 
 
 O 
 
 <v 
 
 
 <D 
 
 o 
 
 
 O 
 
 c3 
 
 
 c3 
 
 «4-l 
 
 
 «H-t 
 
 Fh 
 
 
 Fh 
 
 _o 
 
 
 o 
 
 '£ 
 
 
 'h 
 
 o 
 
 
 <D 
 
 -H 
 
 
 -H 
 
 m 
 
 b 
 
 GQ 
 
 O 
 
 p 
 
 O 
 
 a 
 2 21 
 
 s 1 
 
 o *33 
 
 73 7b 
 
 co . 4 ; 
 
 a 
 
 o "b 
 c3 0 
 
 >1 Q, 
 
 to a? 
 
 Q 
 
 Approximate. 2 Eight vertebrae. d Nine vertebrae. 4 Ten vertebrae. 
 
FLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 171 
 
 elements in M. robustus. In M. harlani the flattened external face of the plate of bone 
 below the border of the obturator foramen extends downward between the converging 
 and thickened anterior and posterior borders. The pubic symphysis is short, with the 
 lower or outer surface forming a distinct tubercle. The pelvic opening appears to be 
 relatively narrower in M. harlani than in M. robustus. 
 
 FEMUR. 
 
 The femur of Mylodon harlani (plate 38, figs. 1, la, lb, and plate 39) resembles 
 closely that of M. robustus, although the specimens from the asphalt beds average 
 considerably larger (average, 546.4 mm.; shortest, 509.2 mm.; longest, 579.8 mm.). 
 The proportion of thigh-bone to tibia is, however, practically the same for the North 
 American and South American species. The femur of M. harlani is decidedly longer 
 than that of Scelidotherium, and the proportion of femoral length to tibial length is 
 distinctly less in the latter than that in the Rancho La Brea species. 
 
 The head of the femur is directed more to the inner side and less upward than in 
 Nothrotherium and more as in the femur of Hapalops, but the neck is not distinct. In 
 the femora of Scelidotherium magnum and Catonyx giganteus, as shown by Winge (plates 
 40 and 31), the head is directed more distinctly upward than in Mylodon. The border 
 of the head on the posterior side is notched to form the depression for the ligamentum 
 teres. In femora from Rancho La Brea this notch varies considerably in extent, and in 
 one specimen it is entirely absent. The notch may be of triangular outline, thus re¬ 
 sembling that in M. garmani, or it may be similar in shape to that in M. robustus. 
 Occasionally a scar extends from the apex of the notch inward toward the center of 
 the articulating surface. On the antero-internal side of the head the border of the 
 articulating surface is usually indented, but the indentation is broader and not so deep 
 as the notch for the ligamentum teres. In one specimen (No. 1704-1^15), however, 
 the antero-internal indentation is very deep, much more so than the posterior notch. 
 The digital fossa is deep and in one specimen (No. 1704-R-14) it is partially divided 
 within by a median vertical partition. The lesser trochanter is not so prominent a 
 feature as it is in the femur of Scelidotherium magnum shown by Winge (plate 41). 
 
 Table 86. —Average measurements (in millimeters) of 26 specimens of femur 
 
 of Mylodon harlani. 
 
 Total length, measured from head to inferior surface of inner condyle. 546.4 
 
 Length, measured from great trochanter to inferior surface of outer condyle. 525.2 
 
 Transverse diameter of head, measured at base. 127.8 
 
 Width, measured from inner surface of head to outer surface of great trochanter. 282.9 
 
 Thickness of shaft at middle of inner border. 67.5 
 
 Least width of shaft. 164.6 
 
 Greatest width across distal tuberosities (above condyles). 234.8 
 
 Width across condyles. 188.5 
 
 Width of intercondyloid space. 47 
 
 Greatest width of inner condyle. 88.7 
 
 Vertical extent of inner condyle. 120 
 
 As in the femur of Nothrotherium, the lower half of the thigh-bone in Mylodon lies 
 in a plane which has been rotated inward somewhat and is oblique to a vertical plane 
 passing transversely through the head and great trochanter. This is to be correlated 
 with the inwardly directed tibia. The smallest transverse diameter of shaft in M. 
 harlani occurs distinctly below the middle of the femur. In other words, the least 
 width of shaft is closer to the distal extremity than in the femur of Nothrotherium. The 
 femora of Scelidotherium and Catonyx do not show any marked constriction of shaft, 
 as in Mylodon. The anterior surface of the upper portion of the shaft is marked by 
 three longitudinal ridges for muscle attachments. The outermost ridge is situated 
 closer to the lateral border of the great trochanter than that of the femur of Scelido¬ 
 therium magnum. A third trochanter, present in Nothrotherium and Megalonyx, is lack¬ 
 ing in Mylodon. 
 
172 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 The distal articulating condyles are entirely obscured in anterior view (plate 38, 
 fig. 1). In the femur of M. harlani there is a less expanse of bone present external to the 
 rotular groove than in that of SceUdotherium magnum. This is much better developed 
 in the femur of Nothrotherium. The rotular surface is continuous with the articulating 
 surfaces of the condyles (plate 38, fig. 16), Mylodon thus differing from Nothrotherium, 
 in which it is separate from both. 
 
 PATELLA. 
 
 The patella of Mylodon harlarii (plate 38, figs. 2 and 3) resembles that of the Miocene 
 genus Hapalops more closely than does the knee-cap of either Nothrotherium or Megalonyx. 
 It is much thicker in antero-posterior diameter than the patella of Megalonyx, the greatest 
 thickness occurring just external to the median vertical line. On the dorsal side this is 
 produced by a tuberous development which is distinctly elevated above the surrounding 
 surface. In the latter respect M. harlani agrees with M. garmani and differs from M. 
 robustus, Megalonyx, and Nothrotherium. Occasionally, however, this dorsal tuberosity is 
 not well developed, and the patella appears more like that of M. robustus. The femoral 
 articulating surface is broadly convex in transverse direction, with the outer portion 
 slightly concave and the inner portion flattened in dorso-ventral extent. This surface 
 is of greatest dorso-ventral diameter toward the inner side. The ventral process below 
 the surface terminates usually rather sharply. The end is recurved posteriorly, giving 
 the process a decidedly concave posterior surface. Contrasted with patellae of Catonyx 
 and SceUdotherium, that of Mylodon harlani possesses a smaller ventral process which 
 terminates less broadly. 
 
 Table 87. —Average measurements {in millimeters ) of 34 specimens of patella 
 
 of Mylodon harlani. 
 
 Greatest depth over all. 124 
 
 Depth measured from dorsal border of femoral surface to end of ventral process. 106.3 
 
 Transverse width of femoral surface. 108 
 
 Greatest thickness through middle of femoral surface. 72 
 
 TIBIA. 
 
 The average measurements for the tibia of Mylodon harlani from Rancho La Brea 
 indicate a size slightly larger than that of M. robustus. In SceUdotherium the tibia is 
 longer and the extremities are relatively narrower than in the lower leg element of 
 Mylodon. The tibia of Catonyx giganteus reached a much larger size than that of Mylo¬ 
 don, according to the measurements given by Winge (1915, p. 163), and is likewise 
 relatively longer with extremities narrower than in the latter. The tibia of Lestodon 
 armatus is also considerably longer than that of Mylodon. 
 
 Viewed in proximal aspect (plate 40, fig. la), the specimens from the asphalt beds 
 are seen to resemble in general shape the tibia of M. robustus. Brown (1903, p. 580), 
 in a comparison of the Nebraska form with M. robustus, has noted the difference in 
 shape of outer femoral articulating surface, and has described it as being pyriform in 
 the former. Tibiae from Rancho La Brea frequently exhibit a shape of outer femoral 
 surface similar to that in Brown’s specimen, but materials are not lacking in which the 
 inner anterior portion of the facet is fuller in its development and gives the surface an 
 approximation in shape to that seen in M. robustus. 
 
 Principal support for the femur is furnished by the large inner articulating surface 
 which is borne by the shaft of the tibia. The outer femoral articulation, as described 
 by Lull, is not supported directly by the shaft, but by a strong buttress arising from it. 
 Anterior to the outer femoral articulation is a large extent of rugose surface which 
 descends upon the antero-lateral side of the tibia to form the place of attachment of 
 the ligamentum patellae. The position of this insertion is curiously out of place with 
 reference to the principal face of the tibia and is one of the results of torsion undergone 
 by this element, causing the anterior face to be directed more to the inner side. With 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 173 
 
 the inward turning the area for attachment of the patella ligament comes to rest adja¬ 
 cent to the upper outer border of the tibia. A small lateral process from the upper 
 portion of the cnemial crest partly, and occasionally completely, incloses a canal for 
 the transmission of vessels lying along the outer base of the crest. In specimen 1702- 
 L^33, in which the fibula has fused with the tibia, this process touches an inwardly 
 directed extension from the fibula, with which it tends to fuse. In M. harlani this 
 channel, as developed in front of the outer femoral articulation, is situated farther from 
 the outer border of the surface than in M. robustus. 
 
 A small facet on the posterior side of the outer femoral surface indicates the presence 
 of a fabella. Along the inner margin of the inner femoral surface a narrow, elongate 
 facet is occasionally present which also articulates with a fabella. The proximal fibular 
 surface is similar in shape and position to that of M. robustus. A single broad channel 
 traverses the posterior side of the inner malleolus. In Nothrotherium two grooves are 
 present in this region of the tibia. On the distal extremity (plate 40, fig. 16) the fibular 
 facet is sometimes narrower and longer than that in the tibia of the South American 
 species. The shape of this facet varies, however, in M. harlani. 
 
 The concavity for the reception of the inner process of the astragalus is broad or 
 narrow, depending upon the size of the process. The rugose distal surface of the tibia 
 lying adjacent to the inner posterior border of this surface is pierced by several large 
 nutrient foramina. 
 
 Proportion of length of tibia to length of femur is as 0.452 is to 1. 
 
 In Scelidotherium leptocephalum the lower leg, in proportion to the thigh, is longer 
 than in Mylodon. This is indicated by the ratio, length of tibia : length of femur as 
 0.642:1. A slight amount of variation is shown by the individuals whose measurements 
 are given by Winge (1915, pp. 201-202), and the ratio may range from 0.631: 1 to 0.652:1. 
 Gregory® gives the tibio-femoral ratio for Lestodon armatus as 0.51:1. 
 
 Table 88.— Average measurements {in millimeters) of 32 specimens of tibia 
 
 of Mylodon harlani. 
 
 Length, measured from intercondyloid eminence of proximal extremity to distal spine. 247.3 
 
 Greatest length along inner side... 246.3 
 
 Greatest distance across proximal extremity. 185.2 
 
 Greatest thickness across inner femoral surface measured along principal axis. 123.2 
 
 Least thickness of proximal extremity measured between femoral surfaces. 87.3 
 
 Least width of shaft. 93.8 
 
 Greatest width of distal extremity. 142.5 
 
 Greatest thickness of distal extremity. 101.9 
 
 FIBULA. 
 
 In Mylodon harlani the fibula (plate 40, figs. 3 and 4) is always longer than the 
 corresponding element figured by Owen for M. robustus. In the series of specimens from 
 Rancho La Brea the average measures 263 mm. in total length, the shortest giving a 
 measurement of 248.2 mm. and the longest a measurement of 248.7 mm. The fibula 
 described by Brown from Nebraska falls readily within the range in size exhibited by 
 M. harlani , and this is true not only for the length of the element but for other meas¬ 
 urements as well. The fibula of Mylodon described by Williston from Seneca, Kansas, 
 exceeds in total length the largest specimen available from Rancho La Brea by approx¬ 
 imately 5 mm. The corresponding measurements stated by Allen for specimens of this 
 element in M. garmani approach in the one case (265 mm.) near the average for the 
 Rancho La Brea series and in the other (234 mm.) indicate a size distinctly smaller 
 than the shortest of the latter series. 
 
 The proximal extremity of the fibula is truncated mainly by an elongate oval artic¬ 
 ulating surface for the tibia. The small facet for a sesamoid is a variable feature of 
 the outer anterior surface of this extremity. The hook-like process noted by Brown as 
 
 ° W. K. Gregory. Notes on the principles of quadrupedal locomotion and on the mechanism of the limbs of hoofed 
 animals. Ann. N. Y. Acad. Sci., vol. 22, pp. 267-294, pi. 34, 1912. 
 
174 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 extending backward from the end of the tibial surface is present also in tibiae from the 
 asphalt beds, but there are a number of specimens available in which the process is not 
 prominently developed. The tibial surface of the shaft is flattened, while the opposite 
 face is convex. Above the articulation of the distal extremity the tibial surface of the 
 shaft is very rugose and depressed and represents a portion of the end of the shaft which 
 has not fused with the distal epiphysis. The distal extremity is roughly trihedral in 
 shape, the three principal surfaces being the tibial, anterior, and outer. The tibial sur¬ 
 face is composed of two parts represented principally by the articulating facets for the 
 tibia and the astragalus. These facets are confluent but demarcated by a right angle. 
 The tibial facet is broad in fore-and-aft diameter, the astragalar facet is of greatest 
 extent in vertical direction and lies close to the anterior border of the extremity. The 
 two articulating surfaces are always confluent over a sharp angle in specimens from 
 Rancho La Brea, and thus the fibula of M. harlani agrees with Brown’s specimen from 
 the Pleistocene of Nebraska, with the fibula of M. garmani, and differs from that in M. 
 robustus, in which the two facets are separate. Propor¬ 
 tion of length of fibula to length of femur is as 0.481 
 is to 1. 
 
 The average measurements (in millimeters) of 25 
 specimens of Mylodon liarlani are as follows: Total 
 length, 263; greatest width of proximal extremity at 
 right angles to vertical axis, 103.6; least width of shaft, 
 
 44.3; least thickness of shaft, 24.7; greatest width of 
 distal extremity across anterior face, 73.5; greatest width 
 of distal extremity across outer face, 66.9. 
 
 Fig. 95. —Mylodon harlani Owen. A and B, 
 falciform bone of pes, No. 1701. X 0.50. 
 Rancho La Brea Pleistocene. 
 
 FALCIFORM BONE OF PES. 
 
 Five specimens in the Rancho La Brea collection of the Los Angeles Museum may 
 possibly represent the falciform bone (fig. 95, a, b) of the pes. Such an element is known 
 to occur in the pes of Scelidotherium. The specimens from the asphalt beds are rudely 
 quadrangular, with surfaces very much roughened. One of the principal faces is con¬ 
 cave and is pierced by several large foramina. . 
 
 The average measurements (in millimeters) of 5 specimens of Mylodon harlani are 
 as follows: Long diameter, measured between opposite apices, 52.7; short diameter, 
 measured between opposite apices, 45.6. 
 
 CALCANEUM. 
 
 As remarked by Owen, the most noteworthy feature of the calcaneum of Mylodon 
 (fig. 96, a to d) and one possessed alike by the South American M. robustus and the 
 Rancho La Brea species, is the great development of the posterior portion. There is 
 some variation in the shape of this part of the calcaneum, but in the specimens from 
 Rancho La Brea it is always more or less triangular with the inferior surface, highly 
 rugose, and pierced by a number of nutrient foramina, thus resembling M. robustus. In 
 the calcaneum of Lestodon a posterior apex is not defined and the border is broadly 
 rounded, and this apparently is true also for Scelidotherium. In specimens belonging 
 to immature individuals the greater part of this posterior portion of the calcaneum is 
 seen to be formed by the epiphysis of the bone. 
 
 The inferior surface of the posterior region is connected with the anterior extremity 
 bearing the surfaces of articulation by a constricted part which in some specimens may 
 be quite narrow. The tendinal groove, which occasionally may be nearly completely 
 arched, lies along the outer border of this region. The anterior extremity is devoted 
 mainly to articulation with the astragalus. This surface is narrow above, where it is 
 transversely concave, and wide below. Along the outer side the astragalar facet is 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 17f) 
 
 convex in dorso-ventral extent. In Mylodon the articulauon for the astragalus is not 
 divisible into two parts, as it is in Scelidotherium and Lestodon. Below the astragalar 
 facet, but confluent with it, is the concave facet for the cuboid. 
 
 Table 89. — Average measurements (in millimeters ) of 21 specimens of calcaneum 
 
 of Mylodon harlani. 
 
 Length. 224.8 
 
 Greatest width across inferior side of posterior expansion.. 123.9 
 
 Least width across inferior side of neck. 60.2 
 
 Greatest width at anterior end, measured across astragalar surface. 91.8 
 
 Greatest depth, measured across outer side. 125.4 
 
 ASTRAGALUS. 
 
 The astragalus of Mylodon (fig. 97, a to d), representing the connecting element 
 between the foot and epipodial segment of the limb, expresses in its peculiar shape the 
 result of a torsion which has caused the entire foot to rest upon the outer side. The 
 dorsal or tibial surface (fig. 97 a) may be considered as of two parts because of their 
 
 Fig. 96 .—Mylodon harlani Owen. Left calcaneum, No. 24282 U. C. C. A, ventral view; B, outer view; C, dorsal view; 
 
 D, view of articular end. X 0.33. Rancho La Brea Pleistocene. 
 
 characteristics, although both are directly connected. The inner part forms the prom¬ 
 inently convex surface of a process, likened to an odontoid process, which extends 
 inward and somewhat upward. The principal borders of the articulating surface are 
 parallel, or they may converge somewhat to the inner end of the process. The articu¬ 
 lating surface is continuous externally, with a broad flange-like or shelf-like area con¬ 
 stituting the lateral portion of the tibial articulation. The angle which separates the 
 two parts is pronounced and is as well defined posteriorly as it is anteriorly. In the 
 astragalus of Nothrotherium the inner process carrying the tibial surface is shorter, but 
 the latter is broader in antero-posterior extent. The angle between inner and outer 
 portions of the tibial articulation is greater, and the separation is consequently not so 
 well defined, particularly in the posterior region, as in the astragalus of Mylodon. In 
 Scelidotherium the inner tibial surface is also broader antero-posteriorly than that in 
 Mylodon. 
 
 The distinction between tibial surface and fibular side is sharp. The former artic¬ 
 ulation is continuous with a fibular facet along this side (fig. 97 b) which is broad above, 
 but extends to the calcaneal border as a narrow surface. In Nothrotherium the fibular 
 
170 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 surface is much more extensive in antero-posterior direction, flanking the tibial surface 
 for nearly its entire extent. On the lower side of the astragalus the articulation for 
 calcaneum (fig. 97 c) forms a continuous surface from the posterior end to the border of 
 the cuboid-navicular surface. This is a constant characteristic in specimens from Rancho 
 La Brea and distinguishes Mylodon harlani from Scelidotherium (R. Owen, 1840, plate 26, 
 fig. 6). On the lower surface of the process bearing the inner tibial articulation above, 
 and seen best in inferior aspect, a knob may be developed which is quite prominent in 
 some specimens from the asphalt beds. The inferior surface for navicular and cuboid is 
 flattened in antero-posterior direction, but possesses a convex curve in transverse extent. 
 It thus resembles the articulation in the astragalus of Lestodon armatus, but differs from 
 that in Scelidotherium (Owen, figs. 2 and 6), in which a portion of this surface is concave. 
 The anterior surface for the navicular (fig. 97 d), which rounds gently into that portion 
 of the inferior surface articulating also with this element, may show considerable varia¬ 
 tion in the development of a central depression. The surface may be practically flat or 
 deeply excavated, and there are numerous specimens showing gradations. Lull has noted 
 a similar variation in astragali from the Rock Creek Pleistocene of Texas. The inter- 
 articular area adjacent to the inner border of the calcaneal surface is pierced by a number 
 of large nutrient foramina. 
 
 Fia. 97. —Mylodon harlani Owen. Left astragalus, No. 23133 U. C. C. A, tibial view; B, lateral view; 
 C, calcaneal view; D, distal view. X 0.33. Rancho La Brea Pleistocene. 
 
 The average measurements (in millimeters) of 41 specimens of Mylodon harlani are 
 as follows: Antero-posterior diameter, 140.2; greatest distance from fibular facet to 
 cuboid-navicular surface across front depression, 104.4; greatest distance from end of 
 fibular facet to end of ascending process with inner tibial surface, 134.3; distance from 
 fibular border of lateral tibial surface to navicular surface, 115.3; antero-posterior extent 
 of lateral tibial surface, 121.4; antero-posterior extent of inner tibial surface, 49.1. 
 
 CUBOID. 
 
 The cuboid (fig. 98, a to e) is a compact element with dorso-palmar and proximo- 
 distal diameters nearly equal, and with the greatest width across the dorsal surface near 
 the metapodial border exceeding either of these. The greatest extent of articulating 
 surface is that on the distal side (fig. 98 c), which joins with the two elements, metatar¬ 
 sals IV and V. The surface for the latter metapodial is broad, flattened, or slightly 
 convex, being wider toward the dorsal side and narrowing to the palmar side. Contin¬ 
 uous along the inner border of this surface is the articulation for metatarsal IY. It is 
 also widest along the dorsal border and the palmar half is a strip deflected along the 
 inner side of the cuboid, thus giving the surface a peculiar curvature. Along the inner 
 side of the dorsal portion the surface is also continuous, but at a right angle, with a small 
 facet for metatarsal III. 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 177 
 
 On the proximal side of the cuboid (fig. 98 a) is situated the large articulating 
 surface for the astragalus, which is concave in dorso-palmar extent. Along the inner 
 side it is continuous at an obtuse angle, with a surface for the navicular. Occasionally 
 a very small portion of the articulating surface near the dorsal border, and opposite 
 that for metatarsal III, may be deflected still more upon the inner side, and touches 
 the external cuneiform. 
 
 Along the entire outer border the astragalar surface is continued at a sharp angle 
 with an articulating surface for the calcaneum. The latter (fig. 98 d) possesses a sig¬ 
 moid curvature, and is wide above and narrows to the palmar side. 
 
 The average measurements (in millimeters) of 32 specimens of Mylodon harlani 
 are as follows: Proximo-distal diameter through middle, 54.3; dorso-palmar diameter 
 through middle, 55.3; greatest transverse diameter along metapodial border of dorsal 
 surface, 71.3. 
 
 Fig. 98. —Mylodon harlani Owen. Left cuboid, No. 23134 U. C. C. A, proximal view; B, dorsal 
 view; C, distal view; D, outer or calcaneal view; E, inner view. X 0.50. Rancho La Brea 
 Pleistocene. 
 
 NAVICULAR. 
 
 The navicular (fig. 99, a, b, c) is compressed in proximo-distal direction and de¬ 
 cidedly concave in its long axis to accommodate the element in articulation with the 
 astragalus. The proximal surface is entirely devoted to this articulation, and the knob 
 or boss (fig. 99 a), which is a peculiar feature of the surface, is variable in its develop¬ 
 ment. It may be very prominent in some specimens and entirely lacking in others. 
 This variability in the development of the proximal boss, which articulates with a 
 depression in the head of the astragalus, has been noted also by Lull (1915, pp. 377-378) 
 in specimens secured from the Rock Creek Pleistocene of Texas. Along the outer side 
 the astragalar surface of the navicular is continued by a sharp angle, with a narrow but 
 elongated facet for the cuboid (fig. 99 b). For a small extent of its distal border the 
 cuboidal area is continuous with the convex surface on the distal side of the navicular 
 (fig. 99 c), which articulates with the cuneiforms. The latter surface constitutes approx¬ 
 imately one-half the width of the distal side and is distinctly shorter than the principal 
 axis of that side. The major portion of the area is devoted to the ectocuneiform and is 
 demarcated only by a faint groove from the adjacent part, which articulates with meso- 
 cuneiform. A small inner lobe, distinguishable from the remaining portion of the cunei- 
 
178 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 form surface also by a faint groove, may indicate an articulation with a rudimentary 
 representative of the entocuneiform, metatarsal I, and digit I. 
 
 The average measurements (in millimeters) of 34 specimens of Mylodon harlani are 
 as follows: Greatest length, 85.1; greatest width, 60.5; proximo-distal thickness, 30. 
 
 Fig. 99 .—Mylodon harlani Owen. Left navicular, No. 23135 U. C. C. A, astragalar or proximal view; B, 
 lateral view showing astragalar and cuboidal surfaces; C, distal view. X 0.50. Rancho La Brea Pleisto¬ 
 cene. 
 
 ECTOCUNEIFORM. 
 
 The ectocuneiform (fig. 100, a to d) is a trigonal-shaped bone, with the outer side 
 the longest of the three borders. The proximal and distal surfaces are devoted almost 
 entirely to the support of articulating facets, the former (fig. 100 a) bearing a broad and 
 concave facet for the navicular, and the latter (fig. 100 c), an extensive surface for 
 metatarsal III. The surface for the navicular extends from the angle formed by the dor¬ 
 sal and outer borders diagonally across the ectocuneiform to the inner side. Here it is 
 continued at a right angle with a small articulating surface on the inner side (fig. 100 b) 
 for the mesocuneiform. The latter facet varies much in size; it may sometimes touch the 
 surface for the third metatarsal and occasionally may disappear entirely. In some 
 specimens there is observed a small continuation of the surface for the navicular upon the 
 outer side of the ectocuneiform near the dorsal border. In one specimen (No. 1446-L- 
 17) in which this has occurred a similar slight continuation from the surface for the 
 metapodial is present on the outer side and directly opposite the former surface. It would 
 appear that these small facets articulated with the cuboid. In another specimen (No. 
 1446-L-12) there is a much more extensive surface deflected, at, however, an obtuse 
 angle from the surface for the navicular, and may have articulated with the cuboid. 
 
 Fig. 100 .—Mylodon harlani Owen. 
 Left ectocuneiform, No. 23136 
 U. C. C. A, proximal view; 
 B, inner view; C, distal view; 
 D, dorsal view. X 0.50. 
 Rancho La Brea Pleistocene. 
 
 D 
 
 The surface for the metatarsal is defined by a raised border, except where this 
 facet is continuous with an articulating surface on the outer side, as noted above. 
 
 The average measurements (in millimeters) of 30 specimens of Mylodon harlani are 
 as follows: Greatest dorso-palmar diameter, 58.9; greatest width, 44.8; greatest depth 
 of dorsal face, 29.8. 
 
 MESOCUNEIFORM. 
 
 In Mylodon harlani the mesocuneiform (fig. 101, a to d) is compressed laterally and 
 roughly wedge-shaped, with the proximo-distal diameter of the rounded and narrow dor¬ 
 sal surface nearly twice as great as the corresponding diameter of the palmar surface. 
 
PLEISTOCENE MEGALONYCIIINAE AND MYLODONTIDAE OF RANCHO LA BREA. 179 
 
 The depth of the mesocuneiform may be greater or less than the greatest proximo-distal 
 diameter. The palmar surface is flattened and is wider transversely than the dorsal 
 surface. The outer side is flatter than the inner, and in all specimens from Rancho La 
 Brea bears along the proximal border a facet for the ectocuneiform. There is no indi¬ 
 cation of a facet along the proximal border of the inner side for an entocuneiform. The 
 proximal articulating surface for the navicular is rectangular in shape and is slightly 
 concave. At the distal end the articulating surface for the metatarsal is placed oblique 
 to the vertical axis of the cuneiform. It is of greater dorso-palmar extent than the prox¬ 
 imal articulation and is more deeply concave in that direction. The surface narrows 
 dorsally and in most specimens is constricted at about the middle of the inner side. 
 The mesocuneiform is noticeably larger than in M. robustus. 
 
 Fig. 101 .—Mylodon harlani Owen. 
 Left mesocuneiform, No. 23137 
 U. C. C. A, outer view showing 
 at right facet for ectocuneiform; 
 B, proximal view; C, inner view; 
 D, distal view. X 0.50. Rancho 
 La Brea Pleistocene. 
 
 The average measurements (in millimeters) of 16 specimens of Mylodon harlani are 
 as follows: Greatest distance between dorsal borders of proximal and distal facets, 38.2; 
 distance between palmar borders of proximal and distal facets, 21.1; greatest width of 
 palmar surface, 27.2; dorso-palmar diameter through middle of mesocuneiform, 37.1. 
 
 METATARSAL II. 
 
 In Mylodon harlani the second metatarsal (fig. 102, a to d) differs decidedly from 
 that of M. robustus in its relative length. This element, as compared with that of M. 
 robustus, is exceedingly short in proximo-distal diameter, often shorter, in fact, than the 
 mesocuneiform in M. harlani. In the marked shortening of metatarsal II, the pes of 
 the latter species may be construed as being more advanced than that of M. robustus. 
 It is to be noted, however, that the mesocuneiform is relatively longer in M. harlani than 
 in M. robustus. 
 
 The proximal and distal articulating surfaces of the second metatarsal may almost 
 meet dorsally, giving the specimen a triangular shape in side view. In some specimens 
 the proximal face is almost vertical, and the metatarsal therefore approximates a quad- 
 
 Fig. 102 .—Mylodon harlani Owen. Left 
 metatarsal II, No. 23138 U. C. C. 
 A, proximal view; B, outer view; 
 C, inner view; D , distal view. X 0.50. 
 Rancho La Brea Pleistocene. 
 
 rilateral form. The proximal articulating surfaces (fig. 102 a) conforms in outline to the 
 distal articulating face of the cuneiform and may be convex below and slightly concave 
 above. The distal articulation (fig. 102 d) is elongate in dorso-palmar direction, convex, 
 and faces upward and slightly inward. The lower border of the surface is straight. 
 The inner side of the metatarsal (fig. 102 c) is convex dorso-palmarly, while the outer 
 side (fig. 102 b), as in M. robustus, bears a narrow vertical groove, which joins below 
 an excavation between the proximal and distal extremities. 
 
 In the series of 20 specimens in the collections of the Los Angeles Museum, repre¬ 
 senting the right second metatarsal, two (10 per cent) have fused with the mesocuneiform. 
 In such specimens all indications of an original separation are obliterated, although the 
 two elements are easily recognized by their otherwise characteristic shape (see fig. 103, 
 
180 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 c, d). These specimens are of especial interest in their great similarity to Owen’s figure 
 and description of metatarsal II of M. robustus. It suggests, at least, that in the latter 
 species the second metatarsal as determined by Owen may possibly be the homologue 
 of the compound element from Rancho La Brea. In M. garmani the cuneiform has fused 
 with metatarsal II only in the right pes. 
 
 D 
 
 Fig. 103 .—Mylodon harlani Owen. 
 A and B, inner and dorsal views, 
 right mesocuneiform and meta¬ 
 tarsal II, No. 22771 U. C. C. 
 C and D, dorsal and inner views 
 of co-ossified mesocuneiform and 
 metatarsal II, No. 22772. X 0.50. 
 Rancho La Brea Pleistocene. 
 
 The average measurements (in millimeters) of 18 specimens of Mylodon harlani are 
 as follows: Proximo-distal diameter through middle of metatarsal, 36.1; greatest width 
 of distal end measured parallel to base, 33; greatest depth of distal end, 25.8. 
 
 CO-OSSIFIED PHALANGES I AND II, DIGIT II, PES. 
 
 The compound element representing the fused first and second phalanges of digit 
 II, pes, (fig. 104, a to e) is broader and heavier than its homologue in the first digit of 
 the manus (compare average measurements). The proximal articulating surface (fig. 
 104 a) is broader than that of the anterior phalanx, and possesses a flattened palmar 
 border and well-rounded dorsal side. The palmar border is notched at the middle for 
 the passage of a tendon to the ungual phalanx. Occasionally this notch is absent. The 
 dorsal end of the principal axis of this surface lies to the left or to the right of the per¬ 
 pendicular to the base, according to whether the phalanx is of the left or right pes 
 respectively, but this deviation from the normal is not as great as in the corresponding 
 element of digit I, manus. The shaft (fig. 104, b, d, and e), although broader, is relatively 
 not so deep as in the element of the manus. At the distal extremity (fig. 104 c) the artic¬ 
 ulating surface is broader and more evenly divided by a broad median groove into an 
 inner and outer portion. The outer surface extends considerably below the level of the 
 inner surface (seen when element is oriented with base of proximal extremity horizon- 
 
 Fig. 104 .—Mylodon harlani 
 Owen. Co-ossified pha¬ 
 langes I and II, digit II, 
 left pes, No. 23139 U. C. 
 
 C. A, proximal view; B, 
 inner view; C, distal view; 
 
 D, dorsal view; E, ventral 
 view. X 0.50. Rancho 
 La Brea Pleistocene. 
 
 tal). In the homologue of the manus the distal extremity is unevenly divided into an 
 outer and inner division, of which the former is the smaller. The dividing groove tends 
 to flatten out toward the dorsal surface, whereas in the element of the pes it remains 
 distinct throughout its course. The greater expanse of the outer division causes this 
 surface to be carried farther from the outer side of the phalanx of the pes than that of 
 the manus. Posterior to the inner surface and along the side of the distal end is a strong 
 ridge (fig. 104 b) for tendinous attachment, which is not so prominent in the homologue 
 of the manus. 
 
 The average measurements (in millimeters) of 22 specimens of Mylodon harlani are 
 as follows: Total length 37.1; depth of proximal end measured normal to base, 31; 
 width of proximal end, 25.9; greatest tranverse width behind distal articulation 24.2. 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 181 
 
 PHALANX III, DIGIT II, PES. 
 
 This ungual phalanx (fig. 105, a, b, c) is larger and heavier than the corresponding 
 element of digit I, manus. At the proximal end the articulating surface is more evenly 
 divided into two parts by a low vertical and well-rounded ridge. The inner division of 
 this surface is, however, slightly larger than the outer. The subungual base (fig. 105 b) 
 is broader and flattened more than in digit I, manus. This permits the phalanx to rest 
 securely on a level surface, whereas that of the manus can not remain in such a position. 
 On each side of the flattened base and near its proximal end is the foramen for the sub¬ 
 ungual vessels. These openings do not extend upward along the side of the sheath as 
 they do in phalanx III, digit I, manus, for in the former the vessels supplying the base of 
 the nail and those entering the subungual base pierce the element through openings more 
 closely situated or very often through a single opening. 
 
 The claw-process is noticeably thicker than that of phalanx III, digit I, manus, but 
 possesses the same general shape. In both specimens, when viewed in distal aspect, the 
 principal transverse plane of the process is seen to be oblique to the horizontal plane and 
 slopes decidedly downward to the inner side. In several phalanges the bony sheath sur¬ 
 rounding the claw-process is well preserved. It forms a strong roof over the dorsal 
 surface and is completely fused with the proximal end. 
 
 Fio. 105 .—Mylodon harlani Owen. 
 Ungual phalanx, digit II, left pes, 
 No. 23140 U. C. C. A, dorsal 
 view; B, ventral view; C, inner 
 view with cross-section of claw- 
 process. X 0.50. Rancho La 
 Brea Pleistocene. 
 
 Specimens of this phalanx from Rancho La Brea range in total length from 74.3 
 mm. to 93.7 mm., with an average of 85.3 mm. This length is considerably greater 
 than that of phalanx III, digit I, manus (average 74.8 mm.). In the phalanx of the 
 pes the claw-process ranges in thickness from 17.4 mm. to 22.5 mm., with an average 
 of 19.4 mm. In the corresponding phalanx of digit I, manus, this average is 15.8 mm. 
 
 The average measurements (in millimeters) of 22 specimens of Mylodon harlani are 
 as follows: Greatest length measured from most projecting part ol proximal end to 
 distal end of claw-process, 85.3; depth at proximal end, 33.3; distance from subungual 
 tuberosity to dorsal surface of claw-process, measured normal to dorsal surface, 35.2; 
 greatest width at proximal end, 35.3; width of claw-process at distal end of subungual 
 base, 19.4. 
 
 METATARSAL III. 
 
 This metapodial (fig. 106, a to d) is short in proximo-distal direction, with t he 
 proximal end peculiarly modified to accommodate the metatarsal in its articulation 
 with the surrounding elements. The latter extremity is compressed in the dorsal region 
 to form a wing or process, which apparently overlaps or articulates with the fourth 
 metatarsal to a greater extent in Mylodon harlani than in M. rohustus. lhe palmar 
 portion of the proximal end is extended downward in a decided process. I he dorso¬ 
 lateral wing bears on the inner proximal side (fig. 106 c) a very large facet lor the ecto- 
 
182 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 cuneiform. Continuous with this facet, but situated on the proximo-palmar surface (fig. 
 106 d) of the projection, is a small and narrow rectangular facet for the cuboid, while 
 the second largest facet, for metatarsal IV, is confluent, but at a sharp angle with the 
 cuboidal facet, and is directed outward and forward (fig. 106 b). In the left series of 32 
 specimens, representing metatarsal III, only 1 was noted (No. 1463-Lr-24) which pos¬ 
 sessed a very small facet on the palmar projection of the proximal end directed approx¬ 
 imately like the dorsal cuboidal facet and undoubtedly also articulating with the cuboid. 
 In the figure of the palmar side of the pes in Mylodon robustus (Owen, plate 22) the 
 palmar process of the third metatarsal is shown also to articulate with the cuboid. 
 
 Fig. 106. —Mylodon harlani Owen. Left metatarsal III. No. 23141 U. C. C. A, dorsal view; B, outer view with large 
 articulating surface for metatarsal IV; C, inner view with large surface for ectocuneiform; D, proximal view with large 
 surface on right for ectocuneiform and small surface of articulation on left for cuboid. X 0.50. Rancho La Brea 
 Pleistocene. 
 
 The inner side of the shaft in some specimens from the asphalt beds may become 
 quite tuberous near the border of the facet for the cuneiform. At the distal extremity 
 the broadly convex carina is flanked principally along the inner side by an articulating 
 surface, while along the outer side only in the lower region near the sesamoid surface is 
 an offset developed. The sesamoid surfaces are of small extent and are separated by a 
 short and sharp ridge. These bones have not been recognized in the collections. 
 
 The average measurements (in millimeters) of 30 specimens of Mylodon harlani are 
 as follows: Length measured along outer side of metapodial below surface for meta¬ 
 tarsal IV, 64.6; greatest depth of proximal end, 73.8; width of dorso-lateral process at 
 proximal end, 26.1; greatest width of shaft measured across dorsal surface, 41.1; greatest 
 width of distal extremity, 39; greatest depth of distal extremity, 53.9; depth of shaft 
 measured across outer surface, 37.9. 
 
 PHALANX I, DIGIT III, PES. 
 
 The height of the first phalanx of the third digit (fig. 107, a, b, c) is slightly greater 
 than the basal width. The phalanx narrows but little dorsally from the base. It differs 
 from the corresponding element in the second 
 digit, manus, in possessing a distinctly shorter 
 proximo-distal diameter, with more shallow proxi¬ 
 mal furrow and flatter condylar surfaces at the 
 distal end. On the proximal extremity (fig. 107 a) 
 the median furrow is flanked along nearly the 
 entire inner side by an articulating offset which 
 widens below. This surface, particularly the 
 lower part, is deflected forward. At the lower 
 end of the outer side of the groove a small offset 
 is also present. These marginal articulating surfaces are continuous below, with facets 
 for sesamoid bones, the outer of which is usually the larger. The palmar border below 
 and between the facets for the sesamoids is concave. At the distal end (fig. 107 c) the 
 
 Fig. 107. —Mylodon harlani Owen. Phalanx I, digit 
 III, left pes, No. 22769 U. C. C. A, proximal 
 view; B, inner view; C, distal view. X 0.50. 
 Rancho La Brea Pleistocene. 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 183 
 
 articulating surface for phalanx II is flatter, the median groove being of greater width 
 and the condyles distinctly less prominent than in phalanx I, digit II, manus. The outer 
 portion of the distal articulation is of greater dorso-palmar extent, while the inner 
 approaches more nearly a convexity in transverse profile. 
 
 The average measurements (in millimeters) of 29 specimens of M ylodon harlani are 
 as follows: Greatest depth, 48.1; greatest width, 42.5; proximo-distal diameter measured 
 across middle of inner surface, 23.2. 
 
 PHALANX II, DIGIT III, PES. 
 
 The second phalanx (fig. 108, a to h) is frequently only half so large as the corre¬ 
 sponding element in the second digit of the manus. The proximal extremity (fig. 108 a) 
 is broad at the base and narrows dorsally to a posteriorly projecting process. The 
 inner of the two articulating surfaces, continuous over a rounded median ridge, is more 
 deeply excavated than the outer. In the second phalanx of digit II, manus, these sur¬ 
 faces are much more deeply excavated. The outer condyle of the distal extremity (fig. 
 108 d) has a greater depth than the inner, and its articulating surface is flatter. 
 
 Fig. 108. —Mylodon harlani Owen. Phalanx 
 II, digit III, left pes. A to D, No. 
 22769 U. C. C.; E to H, No. 23147 
 U. C. C. A and E, proximal views; 
 B and F, dorsal views; C and G, inner 
 views; D and H, distal views. X 0.50. 
 Rancho La Brea Pleistocene. 
 
 Several specimens from Rancho La Brea present an interesting variation in the loss 
 of the dorsal process of the proximal end, with a noticeable flattening of the proximal 
 facets (fig. 108 e). On the dorsal surface of such specimens there is often a small facet 
 situated close to the proximal border, and which evidently articulated with an element 
 corresponding to the process in the completely ossified phalanx. 
 
 In the right series of 32 specimens, 4 (12.5 per cent) are co-ossified with the first 
 phalanx. The length of the compound element, as measured over the middle of the 
 inner side, is 48.8 mm. in the largest and 42.8 mm. in the smallest specimen. A similar 
 co-ossification occurs in the third digit, pes, of Megatherium, Megalonyx, and Nothrothe- 
 rium, though it is worthy of note that in the Miocene genus Hapalops the two phalanges 
 are separate. Among the Mylodontidae the same structure obtains in Catonyx and 
 Scelidotherium, but the co-ossified element is broader and more foreshortened than the 
 Rancho La Brea specimens. The two phalanges are separate in Lestodon, Pseud olestod on, 
 and Mylodon robustus. 
 
 The lack of any great movement between the separate phalanges, together with the 
 location of principal vertical movement between the first phalanx and the metatarsal in 
 Mylodon, may be in part conducive to such co-ossification. It is more probable, how¬ 
 ever, that the co-ossification of the two phalanges in the genera cited above is to be 
 correlated with the large ungual phalanx of the median digit, which prevails in these 
 forms and requires a rigid support, while in Mylodon the relatively much smaller ungual 
 element is accompanied by a more loosely articulating digit. Fusion of the two pha¬ 
 langes in Mylodon harlani may depend upon the age of the individual. In both the 
 left and the right pes of M. garmani the first and second phalanges of digit III are 
 fused. 
 
184 
 
 CENOZOIC GRAVIGRADE EDENTATES OF 'WESTERN NORTH AMERICA. 
 
 The average measurements (in millimeters) of 27 specimens of Mylodon harlani are 
 as follows: Length through middle, 35.4; greatest depth of proximal end, 38; greatest 
 width of proximal end, 35.1; depth of inner condyle of distal trochlea, 21.8. 
 
 PHALANX III, DIGIT III, PES. 
 
 The average measurements of phalanx III (fig. 109, a, b, c) indicate a size distinctly 
 larger than the ungual phalanx of digit II of the pes. There is a range in length from 
 120.8 mm. at one extreme of the series to 94.7 mm. at the other extreme. It will be 
 seen that although in total length the averages for the terminal phalanges of digits II 
 and III, pes, are quite distinct, the minimum length of phalanx (94.7 mm.) in the latter 
 digit approximates closely the maximum length of phalanx (93.7 mm.) in the former 
 digit. There is not a very great difference in average thickness of claw-process in each 
 type of phalanx. 
 
 The two terminal elements are easily distinguishable, however, by their structural 
 peculiarities. At the proximal articulation the posterior projection of the dorsal surface 
 ends in a rather broad surface which is often deeply indented for tendinous attachment. 
 
 Fig. 109 .—Mylodon harlani Owen. Ungual phalanx, digit III, left pes, No. 23142 U. C. C. A, dorsal 
 view; B, ventral view; C, inner view with cross-section of claw-process. X 0.50. Rancho La 
 Brea Pleistocene. 
 
 In the phalanx of digit II, pes, this process ends more sharply and tends to overhang 
 more the inner articulating surface. The distal discoid area of the subungual base 
 (fig. 109 b) is flattened and in some specimens is quite prominent. The subungual fora¬ 
 mina are of unequal size, the inner being larger than the outer foramen. In the cor¬ 
 responding phalgnx of digit II, pes, these foramina are subequal. The difference in size 
 of these foramina in the former phalanx appears due to the fact that on the inner side 
 vessels supply both the base of the nail and the subungual base, while through the outer 
 foramen passed vessels which supplied only the base of the nail. 
 
 When the phalanx is viewed in distal aspect the claw-process is seen to differ from 
 that in the third element, digit II, pes, in that the principal transverse plane is nearly 
 horizontal and not decidedly oblique, as in the latter. When viewed in dorsal aspect 
 (fig. 109 a), the claw-process in some specimens appears to bend outward to a slight 
 degree. Occasionally the bony sheath is well preserved and is seen to extend over the 
 claw-process, being fused completely with the proximal border. 
 
 The terminal element is considerably shorter and smaller in every respect than the 
 corresponding phalanx in digit II, manus (compare average measurements). It ranks 
 next to the latter in size. 
 
 The average measurements (in millimeters) of 26 specimens of Mylodon harlani are 
 as follows: Greatest length measured from most projecting part of proximal end to 
 distal end of claw-process, 111.1; depth at proximal end, 39.8; distance from discoid 
 
PLEISTOCENE MEGALONYCIIINAE AND MYLODONTIDAE OF RANCHO LA BREA. 185 
 
 area of sublingual base to dorsal surface of claw-process, measured normal to dorsal 
 border, 39.9; greatest width at proximal end, 42.6; width of claw-process at distal end 
 of subungual base, 21.8. 
 
 METATARSAL IV. 
 
 In comparison with the fourth metacarpal, the corresponding metapodial of the pes 
 is only slightly shorter (compare average measurements), but differs noticeably in the 
 greater oblique truncation of the proximal end (fig. 110 b). The truncating surface, 
 which along the inner side reaches almost one-half the length of the bone, is composed 
 principally of two articulating facets, namely, for the cuboid and metatarsal III. Both 
 facets are usually confluent (fig. 110 a) and demarcated by a faint ridge. In the left 
 series of 31 metapodials from Rancho La Brea, only 2 were noted in which the two 
 facets are actually separated by a rugose interarticular area. The dorsal portion of the 
 cuboidal facet lies in the general oblique plane with the surface for metatarsal III. Its 
 lower portion, which reaches along the palmar extension of the proximal end, is turned 
 proximally to face more truly in that direction. The proximo-palmar process of the 
 third metatarsal may sometimes touch the lower end of this surface. In the fourth 
 metatarsal of Mylodon robustus (Owen, plate 21) the facets for cuboid and metatarsal 
 
 Fig. 110 .—Mylodon harlani Owen. Left metatarsal IV, No. 23143, U. C. C. .4, inner view; B, dorsal 
 view; C, outer view; D, distal view. X 0.50. Rancho La Brea Pleistocene. 
 
 Ill are apparently demarcated by a distinct angle and the former surface is of greater 
 extent, as inferred from the dorsal view, both of which characters are not true for M. 
 harlani. 
 
 Along the outer border of the proximal extremity (fig. 110 c) is the articulating facet 
 for the fifth metatarsal. This surface is plane, wide dorsally, and narrows considerably 
 toward the palmar end. Its surface dorsally is separated from the plane of the cuboidal 
 facet by an acute angle. The shaft is of less depth than that of metacarpal IV. There 
 is some similarity in distal extremities between the fourth metapodials of the manus and 
 pes. Both show the effect of torsion, and the articulating surface indicates in each case 
 a reduced digit. The carina is somewhat more prominent in metatarsal IV (fig. 110 d), 
 but there is no large tuberous development of bone at the dorso-lateral side of the carina, 
 as in the fourth metacarpal. Along the lower inner side of the carina in metatarsal IV 
 a relatively broad offset is present, while along the outer side in this region a marginal 
 surface may show incipient development. The sesamoid surfaces are separated by a 
 shorter and thicker ridge than in metacarpal IV. Extending backward from this ridge 
 in the posterior metapodial is often a pronounced process of bone. 
 
186 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 The average measurements (in millimeters) of 29 specimens of Mylodon harlani are 
 as follows: Greatest length measured along outer side close to dorsal border, 119.3; 
 greatest depth of proximal end measured along proximal border, 60.7; greatest width 
 of proximal end, 45.5; least width of shaft, 36.3; least depth of shaft, 27.; depth of 
 distal end, 50.1; width of distal end, 43.5. 
 
 SESAMOID BONES FOR METATARSAL IV. 
 
 These structures may be described with equal propriety when the first phalanx of 
 the fourth digit is considered, since they are intimately related to the lower extremity of 
 the latter element. Their close association with the first phalanx is evidenced by the 
 fact that they are occasionally fused with the base of the latter. In the left series of 27 
 specimens of phalanx I, digit IV, pes, 1 is fused with an inner sesamoid, 2 indicate fusion 
 with an outer sesamoid, and 1 possesses both inner and outer sesamoid bones which have 
 united with the base. In the right series of 25 specimens, 4 phalanges are fused with an 
 outer sesamoid, 1 with an inner sesamoid, and 2 have united with both sesamoid bones. 
 In specimens where both sesamoids are present, the latter do not unite across the mid¬ 
 dle, but excrescences of bone are developed on the inner side of each sesamoid, indicating 
 an attempt at such a union. In addition to the sesamoid bones fused with the first 
 phalanx, there are available in the collections of the Los Angeles Museum two separate 
 outer sesamoids of the right side and one right and one left specimen tentatively deter¬ 
 mined as outer sesamoids, and two inner sesamoids of the left side. 
 
 Fig. 111 .—Myloclon harlani Owen. A, B, and C, outer sesamoid bone, 
 digit IV, left pes, No. 1493-L-l. X 1.0. Rancho La Brea 
 Pleistocene. 
 
 Fig. 112 .—Mylodon harlani Owen. A 
 and B, inner sesamoid bone, digit 
 IV, pes, No. 1444-L-l. X 1.0. 
 Rancho La Brea Pleistocene. 
 
 Since the outer side of phalanx I is longer in proximo-distal extent than the inner 
 side, the sesamoid bone (fig. Ill, a, b, c) which lies below it is likewise larger than the 
 inner. The outer palmar surface of the sesamoid is smooth and convex and forms a 
 continuation downward of the outer surface of the phalanx. On the side articulating 
 with metatarsal IV there is a large, broad facet which continues downward the outer 
 portion of the median groove of the phalanx. The outer part of this facet may be 
 deflected upon the outer side, where it then forms a palmar continuation of the small 
 articulating surface or offset along the lower outer margin of the groove in phalanx I, 
 when such an offset is present. Continuous with the larger articulating face at a right 
 angle is a small facet situated on the inner side of the dorsal surface of the sesamoid, 
 which joins with the base of the phalanx. 
 
 As already stated, the inner sesamoid bone (fig. 112, a, b) is smaller than the outer. 
 A prominent ridge extends the length of the palmar surface in specimens examined. 
 This sesamoid bears three facets in its dorsal half, all of them continuous but situated 
 on different surfaces. The sesamoid articulates by a small facet on the dorsal surface 
 with the base of the phalanx. Another facet, the largest of the three, which articulates 
 with the metapodial, forms a palmar continuation of the broad articulating offset along 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 187 
 
 the lower inner border of the median groove of phalanx I. It rounds into a much smaller 
 surface on the outer side of the sesamoid, which also articulates with the metapodial and 
 forms a continuation of the inner palmar portion of the groove of phalanx I. 
 
 Table 90.— Measurements {in millimeters ) of inner and outer sesamoid 
 
 bones for metatarsal IV. 
 
 
 Inner, 
 
 No. 1444-L—1. 
 
 Inner, 
 
 No. 1444-L-2. 
 
 Outer, 
 
 No. 1493-R-l. 
 
 Outer, 
 
 No. 1493-R-2. 
 
 Length. 
 
 33.6 
 
 36.2 
 
 35.9 
 
 30.7 
 
 Width. 
 
 19 
 
 19.5 
 
 22.9 
 
 24.3 
 
 Thickness. 
 
 14.4 
 
 18 
 
 15.9 
 
 15.7 
 
 PHALANX I, DIGIT IV, PES. 
 
 In general form this phalanx (fig. 113, a, b; fig. 114, a, b, c) resembles closely the 
 corresponding element of digit IV, manus. It approximates the latter in size (compare 
 average measurements), but it is absolutely and relatively wider. In this element the 
 distal articulating surface has shifted somewhat to one side of a median vertical plane 
 and the outer side is of greater length than the inner. The presence of foramina on the 
 outer side is more variable in this phalanx than in that of the manus, and when these 
 are present their position is usually closer to the dorsal edge, they are not so large, and 
 there is usually no groove present extending from the foramina to the dorsal surface. 
 The nutritive foramina immediately below the distal articulating surface are smaller and 
 less prominent features in the element of the pes than in that of the manus. 
 
 Fig. 113 . —Mylodon harlani Owen. Phalanx I 
 with outer sesamoid, digit IV, left pes, No. 
 1453-L-15. A, proximal view; B, distal view. 
 X 0.50. Rancho La Brea Pleistocene. 
 
 Fig. 114. — Mylodon harlani Owen. Phalanx I, digit. 
 IV, left pes, No. 23144 U. C. C. A, proximal view; 
 B, outer view; C, distal view. X 0.50. Rancho La 
 Brea Pleistocene. 
 
 The proximal articulation (fig. 114 a) consists principally of a broad median groove 
 extending from the dorsal to the palmar border, and which is much deeper than that of 
 the corresponding phalanx of the manus. Occasionally the groove may be shallow, but 
 as a general rule the depth of the median furrow readily distinguishes this element from 
 the corresponding phalanx of the manus. Along the lower part of the inner border of 
 the groove is an articulating surface continuous at an obtuse angle with that of the 
 groove. The surface resembles that in the phalanx of the manus, but is smaller, being 
 wider transversely, but not extending along the margin of the groove as far as it often 
 does in the latter. An articulating surface may be present also along the lower outer 
 border of the groove, but it is of smaller extent than that on the inner side. This sur¬ 
 face is not represented in phalanx I, digit IV, manus. Both the inner and outer surfaces 
 are continuous below, with small facets on the palmar surface lor the sesamoid bones. 
 That below the outer side is the larger. In the phalanx of the manus a iacet for a sesa- 
 
188 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 moid is never present below the outer side, and only occasionally is a slight indication of 
 one seen below the inner side. 
 
 The distal surface of articulation (fig. 114 c) resembles that in the corresponding 
 phalanx of the manus. Its inner portion may be quite flattened or convex. The phalanx 
 is notched on the palmar side, and this indentation is broader than in the phalanx of the 
 manus. 
 
 The average measurements (in millimeters) of 25 specimens of Mylodon harlani are 
 as follows: Dorso-palmar diameter, 45.2; greatest width, 44.8; greatest proximo-distal 
 extent of outer side, 28.7. 
 
 RUDIMENT REPRESENTING PHALANGES II AND III, DIGIT IV, PES. 
 
 The reduction of the second and third phalanges in the fourth digit of the pes 
 (fig. 115, a, b, c) results in an element which resembles somewhat in general shape the 
 corresponding element in the manus. It is, however, on the average slightly larger than 
 the latter, with the distal extremity heavier or more swollen transversely. The proximal 
 end (fig. 115 a) bears an articulating surface which is usually of greater area in the 
 rudiment of digit IV, pes. The palmar border of this extremity is generally without a 
 median notch, although one or two apparent exceptions to the rule have been noted. 
 In the distal rudiment, digit IV, manus, a small groove on the palmar side, which forms 
 a notch in the palmar border of the proximal end, is almost invariably present. When 
 the element of the pes is viewed from the side (fig. 115 b) the dorsal border extends 
 forward and suddenly descends steeply to the distal end. This gives the element a 
 “roman-nose” appearance in lateral aspect. In the corresponding rudiment of the 
 manus the dorsal border descends, as a rule, with a more gradual slope from the proximal 
 to the distal end and the “roman-nose” effect is not so pronouncedly developed. An 
 exception to the rule is indicated by specimen 1484-R-5, which undoubtedly articulates 
 with a first phalanx (No. 1480-R-20) of the fourth digit, manus. This specimen is much 
 shortened in proximo-distal axis and the “roman-nose” character is well developed. 
 
 Fig. 115. —Mylodon harlani Owen. Rudiment represent¬ 
 ing phalanges II and III, digit IV, left pes, No. 
 1450-L-2. A, proximal view; B, inner view; C, dorsal 
 view. X 0.50. Rancho La Brea Pleistocene. 
 
 The average measurements (in millimeters) of 11 specimens of Mylodon harlani are 
 as follows: Greatest length, 41.3; depth of proximal end, 32.9; width of proximal end, 
 30.3. 
 
 METATARSAL V. 
 
 In Mylodon harlani the fifth metatarsal (fig. 116, a, b, c) is a heavy element, with 
 well-developed and broadened proximal heel suited to assist in the support of the pes. 
 This heel or heavy lateral process of the proximal end is apparently better formed in 
 the metapodial of the North American species than in that of Mylodon robustus, and 
 resembles that in Scelidotherium. A surface at the proximal end indicating an articula¬ 
 tion between metatarsal V and calcaneum does not occur in any of the specimens from 
 Rancho La Brea. Such an articulation occurs in M. robustus, but not in species of 
 Scelidotherium. The cuboidal facet (fig. 116 b) is very extensive and is concave in dorso- 
 palmar direction. It is continuous with, but demarcated by an obtuse angle from, the 
 facet for metatarsal IV, which lies on the inner side. In conforming to the surface on the 
 latter metapodial with which it articulated, it is broad dorsally and narrows to the 
 palmar side. 
 
 The distal extremity is tuberous, with the outer palmar surface flattened. The facet 
 of articulation for the rudimentary digit exhibits the effect of the torsion undergone by 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 189 
 
 the metapodial, for when this element rests upon the outer side the principal axis of the 
 facet is close to the vertical. Normally, the position of the axis under such change 
 of position would lie close to the horizontal. The facet is elongated in dorso-palmar 
 extent, often convex above and flattened below. It is often associated on the palmar 
 surface with one or two sesamoid facets. 
 
 The average measurements (in millimeters) of 26 specimens of Mylodon harlani are 
 as follows: Length from proximal border of cuboidal facet on distal extremity, 113.9; 
 dorso-palmar distance measured along ridge separating cuboidal surface from facet for 
 metatarsal IV, 70.7; distance from surface for metatarsal IV to end of lateral tuberosity, 
 76.4. 
 
 Fig. 116 .—Mylodon harlani Owen. Left metatarsal V, No. 23145 U. C. C. A, dorsal view; B, 
 inner view; C, ventral view. X 0.50. Rancho La Brea Pleistocene. 
 
 PROXIMAL RUDIMENTARY ELEMENT, DIGIT V, PES. 
 
 This element (fig. 117, a to e) is very similar to that in the corresponding digit of 
 the manus. It differs from the latter in not being so deep and in being wider trans¬ 
 versely and longer in a proximo-distal direction (compare average measurements). 
 At the proximal end (fig. 117 a) the surface articulating with the metapodial is broad, 
 with usually flat sides and palmar margin, and the dorsal margin cqnvex. The palmar 
 surface of the element immediately adjacent to the proximal end is grooved, and this 
 groove may notch the palmar border of the facet. The proximal facet may be concave, 
 particularly in its upper portion, and this concavity may become quite deep, as in speci¬ 
 men 1458-R-4. At the distal end (fig. 117 d) is a small facet for a nodule, this facet 
 being directed inward as well as forward. It is directed more forward and not as much 
 upward as is the distal facet in the corresponding element of digit V, manus. If a median 
 vertical plane be passed through the proximal face, normal to the base, and extended 
 distally, the distal facet is seen to lie well upon the inner side of such a plane. This is 
 true also for the corresponding element in the manus and indicates that in both the 
 front and hind extremities this rudimentary digit has assumed an oblique position. Due 
 to this obliquity, the outer face (fig. 117 b) is longer than the inner (fig. 117 c). The 
 former is relatively smooth, while the latter is more rugose and is marked by a rather 
 broad groove extending from the dorsal to the palmar surface. 
 
 Fortunately, there is present in the collections of the Los Angeles Museum one speci¬ 
 men (No. 1458-L-15) of the proximal rudiment, with which is fused the sesamoid element. 
 The latter supports two distinct facets, which articulate with the fifth metatarsal. It is 
 possible that each facet was borne by a separate element, the two having been fused in 
 middle line in the specimen under observation. 
 
190 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 The size and shape of the distal nodule is readily inferred from specimen 1458-L-3 
 (fig. 117 e), in which it is fused with the proximal element. 
 
 The average measurements (in millimeters) of 10 specimens of Mylodon harlani are 
 as follows: Greatest depth of proximal end, 31.2; greatest width of proximal end, 29.7; 
 greatest proximo-distal diameter, 26.8. 
 
 Fig. 117 .—Mylodon harlani Owen. A to D, proximal phalanx, digit V, left pes, No. 
 
 23146 U. C. C.; E, co-ossified proximal and terminal rudiments, digit V, left pes, 
 
 No. 1458-L-3. A, proximal view; B and E, outer views; C, inner view; D, distal view. 
 
 X 0.50. Rancho La Brea Pleistocene. 
 
 COMPARISON OF PES. 
 
 Comparison with Mylodon robustus. —In number of elements present the pes of 
 Mylodon harlani (plate 41) is similar to that of M. robustus. In both species digit I has 
 disappeared. The entocuneiform may be also absent, but in some specimens from 
 Rancho La Brea the extent of the articulating surface on the navicular for the cuneiform 
 bones is sufficiently large to support a reduced entocuneiform. The latter element has 
 not been recognized in the collection from the asphalt. Presence of ungual phalanges 
 in digits II and III and the amount of reduction in the fourth and fifth digits are as 
 characteristic of the North American as of the South American species. 
 
 Two noticeable differences exist between M. haxlani and M. robustus, namely, (1) 
 the size of the second and third phalanges of digit III, and (2) the size of the mesocunei- 
 form and metatarsal II. In a former discussion of the structure of the pes in Mylodon, 
 the conclusion was reached by Stock (1917c, p. 277) that in Owen’s reconstruction, as 
 figured on plates 21 and 22 of the memoir on M. robustus, the second and third phalanges 
 of digit III have been interchanged with the corresponding phalanges of digit II, manus. 
 Thus one of the differences between the two forms to which specific significance may be 
 attached loses its importance. 
 
 In Mylodon harlani the mesocuneiform is compressed laterally and is roughly wedge- 
 shaped, with the antero-posterior diameter of the rounded and narrow dorsal surface 
 twice as great as the corresponding diameter of the ventral side. This element is larger 
 and appears to differ in shape from that figured in M. robustus (R. Owen, 1842, plate 
 21/). The second metatarsal in M. harlani is very short, often shorter, in fact, than 
 the mesocuneiform. In contrast, metatarsal II of M. robustus (R. Owen, 1842, plates 
 21 and 22, m2) is fully twice as long as the cuneiform with which it articulates. 
 
 In the pes of M. harlani, as represented by materials from Rancho La Brea, the 
 mesocuneiform rarely unites with the second metatarsal. Such union has been observed 
 in specimen 22772 U. C. C. All indications of an original separation are obliterated, 
 although the characters of mesocuneiform and metatarsal II can still be recognized. 
 Along the proximal margin of the outer side in No. 22772 there is no distinct facet for 
 the ectocuneiform. The specimen is of special interest because of its similarity to meta¬ 
 tarsal II of M. robustus, as figured and described by Owen (1842, plates 21 and 22, p. 
 121). There is a suggestion, at least, that in the South American species the second 
 metatarsal as determined by Owen may possibly be comparable to No. 22772 from the 
 asphalt deposits. 
 
 Comparison with Lestodon. —The calcaneum in M. harlani differs from that in Les- 
 todon in its greater constriction below the astragalus, and in the less broadly rounded 
 posterior end. The tendinal groove on the outer border of the calcaneum is larger in 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 191 
 
 M. harlani. An important difference between the two genera, to which attention has 
 been directed by Gervais (1873), is the division of the astragalar facet. In this respect 
 Lestodon is more like the Megalonychinae. Gervais also points out that in general 
 shape the calcaneum of Lestodon is more like that element in Megatherium. 
 
 The navicular in the Rancho La Brea species is less compressed than in Lestodon, 
 and the ectocuneiform is thicker in the latter. In the latter also the mesocuneiform 
 and second metatarsal are united. Gervais states that the third metatarsal is relatively 
 much longer than in Mylodon. 
 
 The second and third digits resemble those in Lestodon. In both forms the proximal 
 element in digit II is a composite structure. The ungual of this digit is comparable in 
 size to that of Lestodon , and thus both agree in differing from the Pleistocene Megalony¬ 
 chinae. The fourth and fifth digits in M. harlani have suffered a reduction similar to 
 that in Lestodon. 
 
 Table 91. —Measurements {in millimeters ) of skulls and mandible of Mylodon harlani tenuiceps. 
 
 
 No. 
 
 1716-1. 
 
 No. 
 
 1716-2. 
 
 No. 
 
 1716-3. 1 
 
 No. 
 
 1716-4. 
 
 Length from anterior margin of maxillaries to posterior end of occipital con- 
 
 
 
 
 
 dyles. 
 
 495 
 
 486.4 
 
 2 459.4 
 
 471.5 
 
 Length of palate from anterior end of maxillaries to postpalatine notch. 
 
 219 
 
 217.6 
 
 
 213.2 
 
 Greatest width measured across ventral surface anterior to first teeth. 
 
 136.4 
 
 2 138.6 
 
 
 
 Width of palate, measured between middle of inner sides of second superior 
 
 
 
 
 
 teeth. 
 
 65.2 
 
 70.5 
 
 72.6 
 
 62.4 
 
 Width of palate, measured between inner sides of anterior lobes of fifth supe- 
 
 
 
 
 
 rior teeth. 
 
 38.4 
 
 46.2 
 
 42.8 
 
 41.4 
 
 Least width posterior to fifth teeth. 
 
 52.7 
 
 57.4 
 
 
 2 55.5 
 
 Least distance from fifth superior tooth to middle of postpalatine notch. 
 
 45.6 
 
 47.3 
 
 
 50.8 
 
 Least distance across ventral margins of pterygoid plates. 
 
 
 111.7 
 
 
 
 Mastoid width above stylohyal processes. 
 
 184.5 
 
 2 184 
 
 2 167.7 
 
 178 
 
 Greatest width across occipital condyles. 
 
 131 
 
 132 
 
 2 122.8 
 
 118.5 
 
 Transverse diameter of foramen magnum. 
 
 41.4 
 
 43.5 
 
 42.3 
 
 41.4 
 
 Dorso-ventral diameter of foramen magnum. 
 
 38.9 
 
 35.5 
 
 37.9 
 
 38.5 
 
 Greatest width of muzzle. 
 
 133.5 
 
 2 135 
 
 2 131 
 
 
 Least width behind postorbital processes. 
 
 94.6 
 
 97.7 
 
 2 100 
 
 91.7 
 
 Height measured from plane of basioccipital to dorsal plane. 
 
 134.8 
 
 2 131.8 
 
 128 
 
 129.6 
 
 Mandible: 
 
 
 
 
 
 Greatest length of symphysis. 
 
 
 
 2 92.7 
 
 
 Greatest predental width. 
 
 
 
 98 
 
 
 Depth of ramus between third and fourth inferior teeth, measured normal 
 
 
 
 
 
 to inferior border. 
 
 
 
 2 83 
 
 
 
 
 
 
 
 1 Skull and mandible tentatively referred to M. h. tenuiceps. 2 Approximate. 
 
 Comparison with Scelidotherium. —In the pes of Scelidotherium the calcaneum is 
 heavy and, as Owen remarks, the posterior end “is broader, and terminated by a less 
 angular convexity than in the Mylodon.” According to Schulthess (1920, p. 101), the cal¬ 
 caneal surface of the astragalus is divided, thus differing from that in Mylodon harlani. 
 
 In S. leptocephalum (see H. Burmeister, 1881a, plate 1) and in S. magnum as shown 
 by Winge (1915, plate 41, fig. 1) the first and second digits of the pes have been reduced, 
 the genus thus exhibiting greater specialization than M. harlani. 
 
 In the marked widening and shortening of the third metatarsal and the presence 
 of a heavy proximal segment in the digit which supports distally a very large ungual, 
 Scelidotherium is decidedly different from M. harlani and is strongly suggestive of the 
 Pleistocene Megalonychinae. The large ungual in Scelidotherium is supported by the 
 co-ossified first and second phalanges, while in Mylodon the corresponding ungual is 
 much smaller and is supported by first and second phalanges, which are usually separate. 
 In the slender fourth metatarsal of the former genus the articulating faces for cuboid 
 and metatarsal III are separated by a distinct angle, while in M. harlani they are con- 
 
192 
 
 CEXOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 tinuous in the same plane. In Scelidotherium the three phalanges of digit IV are repre¬ 
 sented. The proximal and middle elements have co-ossified, while the distal or terminal 
 phalanx is rudimentary. In M. harlani only two segments of this digit are present, but 
 the middle or second phalanx has not become fused with the proximal element. The 
 reduction of the fourth digit in Scelidotherium is, therefore, not so advanced as in Mylo- 
 don. Burmeister believes that a closer agreement of Scelidotherium with Megatherium 
 than with Mylodon is shown in that “die beiden Metatarsusknochen der zwei ausseren 
 Zehen nur am cuboideum gelenken, und nicht, wie bei Mylodon, der ausserste zugleich 
 noch den calcaneus beriihrt.” In M. harlani the fifth metatarsal agrees closely in shape 
 with Scelidotherium, and does not touch the calcaneum. 
 
 MYLODON HARLANI TENUICEPS STOCK. 
 
 Mylodon harlani tenuiceps Stock. Univ. Calif. Publ., Bull. Dept. Geol., vol. 10, pp. 171-173, pis. 3 and 4, 1917. 
 
 Type specimen 1716-1, a skull (plate 42, fig. 1; plate 43, fig. 5) contained in the 
 Rancho La Brea collection of the Los Angeles Museum. 
 
 Table 92. —Measurements (in millimeters ) of dentition of Mylodon harlani tenuiceps. 
 
 
 No. 
 
 1716-1. 
 
 No. 
 
 1716-3. 1 
 
 No. 
 
 1716-4. 
 
 Length from anterior end of first alveolus to posterior end of 
 fifth alveolus. 
 
 146.5 
 
 
 *137 
 
 Length from anterior end of second alveolus to posterior end of 
 fifth alveolus. 
 
 119.6 
 
 
 110.8 
 
 3, antero-posterior diameter. 
 
 *28.6 
 
 
 *23.2 
 
 3, transverse diameter. 
 
 *24.3 
 
 
 *20.3 
 
 4, antero-posterior diameter. 
 
 22 
 
 
 *22.4 
 
 4, transverse diameter. 
 
 23.3 
 
 
 23.7 
 
 5, antero-posterior diameter. 
 
 *27.3 
 
 
 22.3 
 
 5, transverse diameter. 
 
 *20 
 
 
 18 
 
 Length of inferior series, alveolar measurement. 
 
 130.4 
 
 1, antero-posterior diameter. 
 
 
 *14.5 
 
 
 i 1, transverse diameter. 
 
 
 *15.4 
 
 
 2, antero-posterior diameter. 
 
 
 25.7 
 
 
 2, transverse diameter. . 
 
 
 20.2 
 
 
 3, diameter of occlusal surface normal to greatest diameter. 
 
 
 17 
 
 
 4, antero-posterior diameter. 
 
 
 52 
 
 
 4, greatest diameter of anterior lobe. 
 
 
 *28.7 
 
 
 4, greatest diameter of posterior lobe. 
 
 
 *19.7 
 
 
 
 
 
 1 Mandible tentatively referred to M. h. tenuiceps. * Approximate. 
 
 Subspecific characters. —Skull more constricted behind postorbital processes and 
 cranium more slender than in M. harlani from Rancho La Brea. Associated with these 
 characters is a very narrow palate posterior to the fifth tooth. In possessing a slender 
 cranium, the subspecies approaches M. garmani more than does the typical M. harlani 
 from the asphalt deposits, differing, however, from both in distinctly narrower palate. 
 It resembles the Nebraska skull (No. 2780, Amer. Mus. Coll.) described by Brown in 
 degree of constriction behind the postorbital processes, but differs in having a narrower 
 palate. 
 
 Discussion .—The extensive range of variation exhibited by the skull and dentition 
 on Mylodon harlani makes it difficult to establish a type absolutely distinct from the 
 parent species. It does not seem possible to establish a series showing gradations between 
 one extreme, as represented by No. 1716-1, and the other extreme as exemplified by 
 No. 21158, Univ. Calif. Coll. Palae., because of difficulties encountered in endeavoring 
 to bridge the gap between the narrow and broad types of skulls. 
 
 No. 1717-1 was considered at first to supply a necessary link in establishing a 
 series of nearly perfect gradation, but after further comparison its affinities appeared 
 to be rather more with the broader skulls. In least width of palate and least width of 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 193 
 
 skull behind postorbital processes (see table of measurements), Xo. 1717-1 is undoubt¬ 
 edly more like the broader skulls than are the slender specimens referred to M. h. tenui - 
 ceps. Xo. 1717-1 belongs apparently to a younger individual than those having the 
 slender skulls, judging from the closure of the coronal and parietosquamosal sutures. 
 These sutures are hardly, if at all, discernible in the latter specimens. It may be possible 
 that the sutures have closed for other reasons than age alone. 
 
 An objection to the separation of a new subspecies may be raised by the apparent 
 inconstancy in Mylodon harlani of the variations given as characteristic of M. h. tenui - 
 ceps. Xo 21160 Univ. Calif. Coll. Palae. can be cited as an example in which a small 
 palatal index is associated with a broad cranium, while in Brown’s specimen from 
 Xebraska a small cranial index is associated with a comparatively broad palate posterior 
 to the fifth tooth. It is also worth noting that in M. garmani, although the cranial 
 index is very small, the palate is very broad. In establishing the subspecies the writer 
 has been impressed by the lack of continuity in the gradations exhibited by the Rancho 
 La Brea series when the two characters under consideration are taken in combination, 
 rather than by the variability of the characters taken separately. 
 
OCCURRENCE AND DESCRIPTION OF MYLODONTIDAE IN THE 
 PLEISTOCENE OF WESTERN NORTH AMERICA. 
 
 GREAT BASIN PROVINCE. 
 
 Fig. 
 
 118.—Map showing occurrences of Pleistocene 
 Mylodontidae in western North America. 1, Ran¬ 
 cho La Brea; 2, McKittrick, Kern County; 3, San 
 Francisco peninsula; 4, Petaluma, Sonoma County; 
 5, Contra Costa County; 6, Marysville, Yuba 
 County; 7, Hawver Cave, Eldorado County; 8, 
 Carson City, Nevada; 9, Fossil Lake region, Oregon; 
 10, Dayton, Oregon; 11, Willamette Valley, Oregon; 
 12, Washtuckna Lake, Washington; 13, Klamath 
 River; 14, Point Conception, California. 
 
 ° G. A. Waring. Geology and water resources of a portion of south-central Oregon. U. S. Geol. Sur., Water Supply 
 
 Paper 220, plate 6, 1908. 
 
 6 E. D. Cope. The Silver Lake of Oregon and its region. Amer. Nat., vol. 23, pp. 970-982, 1889. 
 e For a recent discussion of this subject, see J. C. Merriam, Univ. Calif. Publ., Bull. Dept. Geol., vol. 10, pp. 517- 
 521, 1918. 
 
 Fossil Lake region, Oregon .—The largest assemblage of Pleistocene mammalian re¬ 
 mains known at present from the Great Basin province is that recorded by Cope from 
 sandy beds at Fossil Lake, in south-central Oregon. Waring’s 0 geologic reconnaissance 
 map shows Fossil Lake to be situated 7 miles northeast of Christmas Lake, in an area of 
 lake and stream deposits of Pleistocene age. Christmas Lake is approximately 25 miles 
 northeast of Silver Lake. It is to be presumed that all the Pleistocene vertebrate mate¬ 
 rial listed by Cope * * 6 and others from the region of Fossil Lake was collected in the area 
 
 lying north and northeast of Christmas 
 Lake. The matter of location will 
 be again referred to in mentioning the 
 Pleistocene vertebrate remains collect¬ 
 ed by I. C. Russell in the Fossil Lake 
 region. 
 
 Unfortunately, the position of the 
 Fossil Lake fauna in the Pleistocene 
 succession of mammalian life in North 
 America has not been definitely deter¬ 
 mined, for there is still conflict of opinion 
 between geologist and palaeontologist. 
 A consideration of the geological aspects 
 of the problem led G. K. Gilbert to the 
 belief that the accumulation of the de¬ 
 posit at Fossil Lake approximated in 
 time that which occurred in Lake La- 
 hontan and in Lake Bonneville during 
 the last stages of their history. The 
 beds, according to Gilbert, should be 
 referred, perhaps, to the Wisconsin 
 epoch. On the other hand, palaeonto¬ 
 logical evidence has been regarded as 
 favoring early Pleistocene age of the 
 horizon. 0 
 
 194 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 195 
 
 The list of mammals known from Fossil Lake has been subjected to several revi¬ 
 sions 0 and is given below: 
 
 Cams, cf. occidentalis Richardson. 
 Canis? latrans Say. 
 
 Canis ochropus Eschscholtz. 
 
 Felis, near oregonensis Rafinesque. 
 Felis sp. 
 
 Felis sp., large. 
 
 Ursus sp. 
 
 Vulpes, cf. pennsylvanicus Rhoads. 
 Lutra canadensis Schreber. 
 
 Fiber zibethicus Linnaeus. 
 
 Microtus sp. 
 
 Thomomys sp. 
 
 Geomys sp. 
 
 Castor sp. 
 
 Castoroides sp. 
 
 Lepus sp (cf. campestris Bachman). 
 
 Mylodon sodalis Cope (? = M. harlani Owen). 
 Platygonus cf. vetus Leidy. 
 
 Platygonus sp. 
 
 Eschatius conidens Cope. 
 
 Camelops kansanus Leidy. 
 
 Camelops vitakerianus (Cope). 
 
 ?Camelops sp. Max. 
 
 Antilocapra sp. 
 
 Equus pacificus Leidy. 
 
 Elephas columbi Falconer. 
 
 We are concerned here chiefly with the species Mylodon sodalis, described by Cope 
 (1878, pp. 385, 386) from very fragmentary specimens obtained at Fossil Lake. Since 
 Cope’s statement of the characters of M. sodalis, considerable doubt has arisen as to 
 the validity of this species. Osborn * * * * * 6 made the suggestion that the ground-sloth occur¬ 
 ring at Fossil Lake may possibly be referred to Megalonyx. As indicated in the above 
 list, Matthew regards M. sodalis as possibly the same as M. harlani. 
 
 The ungual phalanx, which Cope (1889, p. 661, fig. 1) considered most distinctive 
 of M. sodalis, has been compared with Mylodon garmani by Allen (1913), who states 
 that the “description applies well enough to the large ungual phalanx of the forefoot 
 in Mylodon garmani but does not indicate that the two are identical.” He adds further 
 that “until more material from the same locality and formation is available, it does 
 not seem best to attempt the definite identification of Cope’s Mylodon sodalis .” In a 
 description of M. harlani from Texas, Lull (1915) states, with reference to Cope’s type 
 material: 
 
 “We have therefore in the M. sodalis type a larger claw of much the same proportions as the Texas 
 specimens, standing between two in the degree of preservation of the basal sheath. I do not believe, therefore, 
 that Cope’s description brings out characters other than those of individual variation or degree of preservation 
 and hence it is entirely inadequate as the characterization of a valid species.” 
 
 Lull, therefore, considers M. sodalis synonymous with M. harlani. 
 
 Compared with the series of terminal phalanges of Mylodon from Rancho La Brea, 
 Cope’s specimen appears to resemble closest in shape the ungual of the second digit of 
 the manus. As restored by Cope, the tip of the ungual from Fossil Lake reaches farther 
 below the level of the base of the bony sheath than in any of the third phalanges from 
 Rancho La Brea. The writer agrees with Allen and Lull that the incomplete appearance 
 of the bony sheath in Cope’s specimen is due to breakage. 
 
 Further materials of the mylodont species from the region of Fossil Lake are 
 available. Portions of the axial and appendicular skeleton of Mylodon are in the collec¬ 
 tions of the U. S. National Museum. Field labels accompanying the specimens state 
 that they were collected by I. C. Russell in 1882, near Button Ranch, Christmas Lake, 
 Oregon. 
 
 The writer has not been able to secure information as to the exact location of the 
 Button Ranch. On the sketch map in Russell’s report 0 on the geology and water 
 resources of Central Oregon, the location of Button Spring is shown approximately 
 10 miles north of Christmas Lake. 
 
 ° W. D. Matthew. List of Pleistocene fauna from Hay Springs, Nebraska. Bull. Amer. Mus. Nat. Hist., vol. 16, 
 
 p. 320, 1902. 
 
 W. J. Sinclair. The exploration of Potter Creek Cave. Univ. Calif. Publ., Amer. Arch, and Ethnol., vol. 2, pp. 
 
 17-18, 1904. 
 
 J. C. Merriam. Personal communication. 
 
 6 H. F. Osborn. The age of mammals in Europe, Asia, and North America, p. 459, 1910. 
 c I. C. Russell. U. S. Geol. Surv. Bull. 252, plate 1, 1905. 
 
196 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 While Button Spring does not necessarily indicate the position of the ranch of that 
 name, such designation presumably implies that the Button Ranch was not distantly 
 located, particularly since Christmas Lake was also given on the labels with Russell's 
 specimens. The following specimens are determined: 
 
 No. 3859 U. S. National Museum: Left ramus with third tooth and fragment of fourth tooth, fig. 119; caudal 
 vertebra; right humerus, proximal half; left radius, proximal two-thirds; left manus, cuneiform, 
 unciform, metacarpals II, III, and IV, phalanges II and III of digit II and phalanges I, II, III of 
 digit III, fig. 120; left pes, cuboid, navicular. 
 
 Several specimens belonging to a mylodont sloth and collected by Miss Annie M. 
 Alexander at Fossil Lake, in 1901, are in the Museum of Paleontology, University of 
 California. The ground-sloth specimens obtained by Miss Alexander have been deter¬ 
 mined as a first phalanx, No. 22523, of the second digit of the manus, a third metatarsal, 
 No. 22524, and two caudal vertebrae. 
 
 All the remains of Mylodon , collected by Russell and bearing the number 3859 
 (U. S. Nat. Mus. Coll.), were secured from one locality. The collection may represent 
 a single individual. All the specimens indicate a mature animal of rather large size, 
 exceeding in this character Mylodon harlani from Rancho La Brea. 
 
 Fig. 119. — Mylodon, near harlani 
 Owen. Left ramus of mandible, 
 No. 3859 U. S. National Mu¬ 
 seum. X 0.33. fmr, fin 2 , mental 
 foramina: 1, 2, 3, 4, inferior 
 alveoli and teeth. Button Ranch, 
 near Christmas Lake, Oregon. 
 
 The right ramus of Mylodon (fig. 119) in the U. S. National Museum collection is 
 large and heavy. The upper portion of the symphyseal region is not preserved, but 
 two large openings of the dental canal occur below the superior border. The posterior 
 foramen lies 27 mm. in advance of the alveolus for the first lower tooth. A small open¬ 
 ing is present below and between the two larger foramina. The depth of the ramus 
 between the third and fourth inferior teeth exceeds that in M . harlani from Rancho La 
 Brea and is greater also than_in M. garmani. The third inferior tooth is present while 
 only the posterior portion of 4 remains. 3 resembles closely the corresponding tooth in 
 the species from the asphalt deposits. In 4 the isthmus connecting anterior and posterior 
 lobes does not show a lateral bulging, in which respect the Oregon form resembles 
 Mylodon harlani from Rancho La Brea. 
 
 Table 93 .—Measurements {in millimeters) of ramus of mandible. 
 
 Depth of ramus, measured below anterior border of alveolus of first lower tooth. 94.2 
 
 Depth of ramus, measured between third and fourth lower teeth and normal to lower border. 105.7 
 
 Thickness of ramus at 3. 50.2 
 
 Length of lower tooth series, alveolar measurement.•. 158.6 
 
 3, greatest diameter across occlusal surface. 30.6 
 
 3, diameter of occlusal surface normal to greatest diameter. 21.4 
 
 4, greatest length, alveolar measurement. 66 
 
 4, transverse width of posterior lobe. 21.7 
 
 A review of the ground-sloth material from Fossil Lake and its vicinity indicates 
 clearly that Mylodon occurred in the Pleistocene fauna of this region. The megalony- 
 chid sloths are not known to have been present. While exhibiting no noteworthy skull 
 
PLEISTOCENE MEGALONYCIIINAE AND MYLODONTIDAE OF RANCIIO LA BREA. 197 
 
 and skeletal differences from the Rancho La Brea forms, the ground-sloths occurring in 
 and near Fossil Lake during the Pleistocene were apparently larger than the mylodonts 
 from the asphalt. A further recognition of this fact may warrant the grouping of the 
 mylodont sloths of south-central Oregon under a distinct subspecies, for which Cope’s 
 name sodalis might be retained. 
 
 Carson City, Nevada .—The earliest statement known to the writer of the occurrence 
 of Pleistocene vertebrates near Carson City, Nevada, is that by A. R. Conkling, 0 in an 
 appendix to the annual report of the Wheeler Sur¬ 
 vey for 1877. In this report Conkling notes an 
 outcrop of granular yellow sandstone, underlain by 
 clay, a mile east of Carson, and observes that 
 invertebrate fossils, particularly fresh-water shells, 
 are commonly found in the deposits. Conkling 
 mentions also the finding of vertebrate remains. 
 
 The Pleistocene beds near Carson City form a 
 part of a series of such deposits exposed in Washoe, 
 
 Eagle, and Carson Valleys, along the eastern base 
 of the Sierra Nevada. The geology of the region 
 has been discussed in recent years by John A. 
 
 Reid. * * * * * 6 The Pleistocene beds exposed east of Car- 
 son City, where the Nevada State Penitentiary 
 is now located, have received considerable atten¬ 
 tion. In the construction of the prison buildings 
 and the excavation of the prison yard a rock section 
 was exposed, consisting of sandstone with occa¬ 
 sional clay layers. The sedimentary accumulation 
 has been regarded as of lacustrine, fluviatile, or 
 aeolian origin. In recent years it has been sug¬ 
 gested that the sediments were originally alluvial 
 and aeolian sands and were strongly cemented by 
 hot-spring deposits. The size of the area, character 
 of rocks, the similarity of the sediments to other 
 sandstones in the adjoining region, and the present 
 and past existence of hot springs in the immediate 
 vicinity are considered by W. S. T. Smith* as 
 comprising the chief evidence in support of this 
 view. 
 
 In the course of the quarrying operations for 
 sandstone at the Nevada State Penitentiary, not 
 only were twigs, invertebrates, and fragmentary 
 remains of vertebrates unearthed, but at a certain 
 horizon there were also uncovered tracks and other 
 impressions of mammals and birds. The Carson 
 footprints were discussed by Le Conte (1882, 1883), 
 
 Fig. 120.— Mylodon, near haiiani, Owen. Incom¬ 
 plete left manus, No. ."1850, IT. S. Nat. Mus.; 
 dorsal view. X 0.33. Button Ranch, near 
 Christmas Lake, Oregon. 
 
 ° A. R. Conkling. Geological report on the portions of western Nevada and eastern California between the parallels 
 
 of 39°30' and 38°30' explored in the field season of 1876. Appendix H, pp. 1285-1205. See G. M. Wheeler, Annual Report 
 
 upon the geological surveys west of the one hundredth meridian in the States and Territories of California, Oregon, Nevada, 
 
 Texas, Arizona, Colorado, Idaho, Montana, New Mexico, Utah, and Wyoming. Appendix NN of the Annual Report of 
 
 the Chief of Engineers for 1877. 
 
 6 J. A. Reid. The geomorphogeny of the Sierra Nevada northeast of Lake Tahoe. Univ. Calif. Publ., Bull. Dept. 
 Geol., vol. 6, pp. 89-161, plates 19-28, 1911. 
 
 c W. S. T. Smith. Origin of the sandstone at the State prison near Carson City, Nevada. (Abstract) Proc. 12th 
 Ann. Meet. Cordill. Sec., Geol. Soc. Amer., Berkeley, Mar. 31 and Apr. 1, 1911, Bull. Geol. Soc. Amer., vol. 23, p. 73, 1912. 
 
198 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 Marsh (1883), and others,® and from these studies it was concluded that of the verte¬ 
 brates the ground-sloth, mammoth, horse, possibly deer and wolf, and birds were living in 
 the region during the Pleistocene. 
 
 Many tracks are still visible on the floor of the prison yard, where they occur both 
 in the coarse sandstone and in the fine clay layers. The most striking impressions are 
 those of the ground-sloth and mammoth in a stratum exposed on the floor of the yard 
 along the east wall, and particularly where they have been exposed by tunnels into the 
 east wall. Here most of the tracks are shallow and the outlines are clearly defined. 
 Near the west wall and in the sandstone occur footprints, apparently of the ground- 
 sloth, but the impressions are deep and the outlines are not so distinct. 
 
 Through the kindness of Mr. R. B. Henrichs, warden of the Nevada State Peni¬ 
 tentiary, the writer was permitted to study the footprints in detail and also to make 
 casts of the best-preserved tracks in the prison yard and in the tunnels. The latter 
 task has been accomplished by Mr. J. W. Lytle, of the Los Angeles Museum of History, 
 Science, and Art, and exact reproductions of these footprints have been installed as an 
 exhibit at that museum (plate 46). 
 
 Considerable interest is attached to the ground-sloth tracks, for they were regarded 
 at one time as the impressions of the sandaled foot of primeval man. Le Conte (1882, 
 1883) pointed out, however, certain obvious objections to this view, namely, the large 
 size and the shape of the individual imprint and the span of the straddle as measured 
 by the distance between the right and left imprints. Le Conte considered the possi¬ 
 bility that the tracks had been made by a bear or by a ground-sloth, and mentions 
 the suggestion of Marsh that the supposed human footprints were perhaps formed by 
 either Mylodon or Morotherium. b At that time remains of Mylodon were known to occur 
 in Great Basin deposits of the same age as the Carson strata. Marsh (1883) expressed 
 an opinion in favor of the view that the tracks were formed by Mylodon. 
 
 The objections to the quadrupedal origin of the footprints were stated by Le Conte 
 to rest in “the apparent singleness of the tracks, and the absence of claw-marks.” The 
 first of these he explained by the observed fact that the imprint of the posterior foot is 
 often superimposed upon that of the anterior foot, while he attributed the second pecu¬ 
 liarity to the “clogging of the feet with mud.” 
 
 Four of the footprints, forming part of the series exposed on the floor of the tunnel 
 in the east wall of the prison yard, are shown in plate 46. Measurements of the tracks 
 are given in the accompanying diagram (see plate 46). The individual footprints in the 
 clay stratum are quite shallow. As indicated in a former paper (Stock, 1917c, pp. 284, 
 285, figs. 5, 6), the imprints correspond in size to the hindfoot of Mylodon harlani from 
 Rancho La Brea. They represent the impressions of the hind feet that have been super¬ 
 imposed on and have largely obliterated the imprints of the front feet. The best pre¬ 
 served tracks show a raised border where the soft mud bulged upward along the side of 
 the foot as pressure was exerted downward when the animal walked across the surface 
 of the ground. The lower surface of the track shows sometimes a slightly raised middle 
 region. This elevation corresponds probably to the arch in the foot of Mylodon existing 
 between calcaneum and metatarsal V. One end of the footprint is wider than the other 
 and possibly represents the anterior or digital portion of the foot in Mylodon. The 
 inner border is always distinctly concave. 
 
 a H. W. Harkness. Footprints found at the Carson State Prison. Proc. Calif. Acad. Sci., 7 pp., 5 diag., 1882. 
 
 C. D. Gibbes. Prehistoric footprints in the sandstone quarry of the Nevada State Prison. Proc. Calif. Acad. Sci., 
 8 pp.. 4 photos., 3 diag., 1882. 
 
 W. P. Blake. The Carson City ichnolites. Science, vol. 4, pp. 273-276, 1884. 
 
 G. D. Louderback. Where mammoths roved. Recent discoveries in Carson, Nevada, State prison—footprints that 
 make geologists wonder—theory of their human origin. Sunset Mag., vol. 19, pp. 205-216, frontispiece, 1907. 
 
 b Morotherium is based on a fragmentary humerus belonging to Mylodon and a femur belonging to mfegalonychid 
 ground-sloth. (See pp. 117, 118.) 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 199 
 
 Mention has already been made of the apparent absence of claw-marks. As shown 
 by the manus and pes of Mylodon harlani from the asphalt deposits, the two largest 
 claws possessed by this animal are borne by the second and third digits of the front 
 foot. A close examination of the tracks in the tunnel seems to suggest that an impres¬ 
 sion was occasionally left of an ungual phalanx (probably the third) of the manus. 
 Such an impression is possibly indicated at a in the diagram (plate 4(j; also plate 47, 
 fig- 1). 
 
 Further substantiation of the belief that the footprints which have been discussed 
 were made by Mylodon rather than by some other quadruped comes from the presence 
 of osseous remains belonging to this animal, found in the Pleistocene deposits at the 
 prison. The specimens consist of skull fragments and a cervical vertebra embedded in 
 a thoroughly cemented, coarse quartzitic sandstone. The skull remains belonged to an 
 individual possessing 4 upper teeth on the right side and apparently 4 teeth also on 
 the left side. 
 
 Washluckna Lake, Franklin County, Washington .—The Washtuckna Lake locality 
 lies not far to the east of the Ringold Pleistocene 0 of the White Bluffs along the Columbia 
 River. The Washtuckna Pleistocene fauna may have some elements in common with 
 the assemblage from the latter formation, but much more material will have to be col¬ 
 lected before adequate comparisons can be made. Matthew * 6 cites the following forms 
 from Washtuckna: 
 
 Taxidea sulcata (americana), parts of skulls, jaws, 
 limbs, and foot-bones. 
 
 Felis, cf. imperialis, parts of limb-bones. 
 
 Felis, cf. concolor, parts of limb-bones. 
 
 Felis, cf. canadensis, parts of limb-bones. 
 
 Mylodon sp., astragali and foot-bones. 
 
 Equus sp., bones of feet and some teeth. 
 
 ? Camelops, cf. kansauus, foot-bones. 
 
 ? Camelops, cf. vitakerianus, foot-bones. 
 
 ? Camelops sp., max. foot-bones. 
 
 Alces brevitrabalis, parts of antlers, foot-bones, etc. 
 Alces, semipalmatus, parts of antlers, foot-bones, etc. 
 Cariacus ensifer, parts of antlers, foot-bones, etc. 
 Oreamnos, parts of horn. 
 
 PACIFIC COAST PROVINCE. 
 
 Willamette Valley, Oregon .—The first remains of mylodont sloths from the Will¬ 
 amette Valley were recorded by H. C. Perkins (1842). The specimens were a tooth and 
 a humerus, and for the animal represented Perkins c proposed the name Orycterotherium 
 oregonensis. G. M. Allen (1913 T p. 343) has remarked that the humerus is now in the 
 collections of the Boston Society of Natural History, but the tooth is no longer available. 
 Richard Owen (1843, p. 344) and Joseph Leidy (1855, p. 48) regarded the material as 
 belonging to Mylodon harlani. 
 
 Allen (1913, pp. 343, 344, plate 4, fig. 16) further describes and figures a fragmen¬ 
 tary specimen from the Willamette region. Allen states: 
 
 “In the collection of the Museum of comparative zoology there is a fragment of the tip of a Mylodon 
 mandible (Plate 4, fig. 16) labeled ‘Walhaumet River, Oregon/ which, though it has no further history was 
 evidently received many years ago, before the spelling ‘Willamette’ for this river was adopted. The bone is 
 stained a dark brown like the humerus from the same locality, and it is probable that it is from the same place 
 or even from the same specimen. Assuming that it represents M. harlani, it supplies a. portion hitherto unde¬ 
 scribed, namely the predental part of the jaw. It includes the tip of the left ramus broken slightly to the right 
 of the middle line of the symphysis, with at its posterior edge the basal part of the socket for the first tooth. 
 It is clearly not referable to Paramylodon nebrascensis, in which the symphysis is considerably longer and 
 narrower with a decided keel. From the Mylodon whose skeleton I have just described, it differs equally in the 
 breadth of the truncate tip of the jaw, in the nature of the symphysis, and in the arrangement of the openings 
 of the dental canal. These last in the Oregon fragment consist of two large subequal openings, the anterior 
 25 mm. in advance of the posterior and some 15 mm. in long diameter, with a very small third opening (4 mm. in 
 diameter) below and between these two. In the other specimen on the contrary, the three openings are in a row, 
 the anteriormost large, the two others small. Further, the symphysis of the Oregon specimen is about 109 mm. 
 
 ° See p. 110. 
 
 6 W. D. Matthew. List of the Pleistocene fauna from Hay Springs, Nebraska. Bull. Amer. Mus. Nat. Hist., vol. 
 16, p. 321, 1902. 
 
 c Amer. Jour. Sci., ser. 1, vol. 44, p. 80, footnote, 1843. 
 
200 
 
 CENOZOIC GRAVIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 long or some 15 mm. shorter than in the Nebraska species, and there is at its base a flattened triangular area 
 whose apex extends for nearly 70 mm. toward the tip of the ramus. The terminal part of the ramus was much 
 broader transversely than in either Paramylodon or Mylodon garmani; it was more abruptly truncate, and 
 about 120 mm. across as against about 85 in the latter species. It therefore more nearly resembles M. robustus 
 in having a broad truncate lower lip, and was thus adapted more for a grazing habit, while Paramylodon and 
 M. garmani with their elongate and compressed rami, were probably browsing animals.” 
 
 From a Pleistocene locality near Dayton, northwestern Oregon, on the Yamhill 
 River, a tributary of the Willamette, Ellen Condon McCornack“ records the finding of 
 ground-sloth, mammoth, and horse. The material consists of two ungual phalanges of 
 the genus Mylodon. The specimens are the property of the University of Oregon. The 
 larger of the two phalanges is of digit III, manus. The second specimen probably 
 belongs to digit II, manus. The latter, when compared with a series of ungual pha¬ 
 langes of digit II, manus, of Mylodon harlani from Rancho La Brea, is characterized 
 by (1) greater length and (2) relatively greater width of subungual base and of the 
 proximal articulation. No. 1470-1^24 and No. 1470-R-3 in the Los Angeles Museum 
 collections approach the Oregon specimen closely in width of proximal end, but in each 
 of the phalanges from Rancho La Brea the claw-process is distinctly wider. The dorsal 
 surface in the second specimen from the Yamhill locality terminates with a straight 
 posterior edge over the proximal articulation. 
 
 Table 94 .—Measurements {in millimeters) of phalanges from near Dayton, Oregon. 
 
 
 Manus, 
 phal. Ill, 
 dig. HI. 
 
 Manus, 
 phal. Ill, 
 dig. II. 
 
 Greatest length, measured from most projecting portion of proximal 
 end to end of claw-process. 
 
 *187 
 
 1 147 
 
 72.8 
 *56 
 
 29.9 
 
 *173 
 
 h50 
 
 80.5 
 54.2 
 
 29.6 
 
 Distance from proximal end of subungual base to end of claw-process. . 
 Greatest proximo-distal diameter of subungual base. 
 
 Width of proximal end. 
 
 Greatest transverse width of claw-process at distal end of subungual 
 base. 
 
 
 1 Approximate. 
 
 San Francisco Bay region , California. —Marsh (1874) first tentatively recorded the 
 presence of ground-sloths in California in the description of Morotherium gig as. Mate¬ 
 rials consisting of a fragmentary humerus and a femur were said to come from Alameda 
 County, California. A second species, Morotherium leptonyx, was based on an ungual 
 phalanx from Pliocene beds of Idaho. Later, Marsh * * 6 described the genus Moropus, two 
 species of which were based on material from the John Day region, eastern Oregon, and 
 a third, a specimen obtained in Nebraska. In the same year Marsh, 0 following the 
 custom of the times, referred both Moropus and Morotherium to the Edentata. The 
 determination of the relationship of Moropus to the chalicotheres led some authors to 
 refer Morotherium also to this group. 
 
 In 1897, Marsh (1897, figs. 8, 9, p. 144) figured the femur of Morotherium gigas and 
 commented as follows: 
 
 “To this group [i. e. the Chalicotheria] the genera Moropus and Morotherium , described by the writer as 
 Edentates, have been recently referred by some authors not familiar with the specimens on which these genera 
 were based. While it is possible that the nature of some of the remains attributed to the former genus may be 
 fairly in doubt, there can be no question that the two known species of Morotherium are both true Edentates.” 
 
 ° E. C. McCornack. Contributions to the Pleistocene history of Oregon. Univ. Oreg. Leaflet Ser., Geol. Bull., 
 
 vol. 6, No. 3, pt. 2, p. 18, 1920. 
 
 6 O. C. Marsh. Note of some new vertebrate fossils. Amer. Jour. Sci., ser. 3, vol. 14, pp. 249-256, 1877. 
 
 c Introduction and succession of vertebrate life in America. An address delivered before the A. A. A. S. at Nashville, 
 Tenn., Aug. 30, 1877. 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 201 
 
 J. C. Merriam (1899) described a humerus from Pleistocene strata near Tomales 
 Bay, Marin County, California. It resembled closely the known portion of Marsh's 
 specimen and was referred to Morotherium by Professor Merriam, although the validity 
 of the genus was questioned by the writer. Stock® compared the humerus from near 
 Tomales Bay with specimens of Mylodon harlani from Rancho La Brea and concluded 
 that the former belonged to Mylodon. This leaves only the femur of Morotherium, to 
 account for. The latter belongs undoubtedly to a megalonychid ground-sloth (see pp. 
 117, 118). 
 
 Ground-sloth remains from Pleistocene beds in the San Francisco Bay region have 
 been noted by Professor Merriam (1899, 1900) at Bull’s Head Point near Martinez and 
 near the station of Rodeo. At Rodeo the Pleistocene deposits rest upon the San Pablo 
 Miocene and the Pinole Tuff Pliocene and are exposed along the shore of the bay. The 
 thickness of the Pleistocene beds near Rodeo is not over 40 feet. In the basal layers 
 occur many oysters and mussels, while the overlying deposits, either of estuarine or 
 alluvial origin, yield fragmentary vertebrate remains. The following mammals have 
 been identified from this horizon: Felis, near atrox Leidy; Smilodon ? sp.; Mylodon sp.; 
 Equus sp. (possibly new); Bison , near antiquus Leidy; camelid (possibly new); Elephas 
 columbi Falconer. 
 
 Remains of the ground-sloth from the Rodeo beds consist of parts of skeletal 
 structures. The genus Mylodon is undoubtedly represented, but the materials are 
 too incomplete for further determination. 
 
 Professor A. C. Lawson * * 6 has noted that, during the cutting of a tidewater canal 
 at the east end of the city of Alameda, 13 feet of sand were exposed overlying 5 feet 
 of the San Antonio formation (Pleistocene), in which occurred a femur and a pelvis of 
 Mylodon (Morotherium gigas.). 
 
 On the San Francisco peninsula at least one occurrence of the genus Mylodon is 
 now definitely known from available material. A humerus (No. 24242, U. C. C.) of 
 this form was found at a depth of 60 feet, about 300 feet west of the intersection of 
 Eighteenth Street and the Twin Peaks Tunnel, San Francisco. The specimen was 
 obtained during the construction of the Twin Peaks Tunnel, and was presented to the 
 Department of Palaeontology by Dr. L. W. Ely, of Palo Alto. 
 
 A second occurrence of ground-sloth remains on the peninsula is indicated in the 
 following letter referred to Professor A. C. Lawson by Dr. C. D. Walcott: 
 
 United States Department of Agriculture, 
 
 Bureau of Plant Industry, 
 Washington, D. C., June 15, 1906. 
 
 Professor Charles D. Walcott, 
 
 U. S. National Museum, Washington, D. C. 
 
 Dear Sir: In looking up some references to old volumes of the Proceedings of the California Academy 
 of Science, I ran across an item which perhaps may be of interest from a palaeontological point of view. As the 
 item is not indexed and might be easily overlooked, I venture to call your attention to it. It announces the 
 discovery of Dr. C. F. Winslow in 1853 of a portion of a tibia which was pronounced by Professor Leidy to 
 “belong to a gigantic sloth of an extinct and undetermined form.” Both Professor Leidy and Professor Baird 
 of the Smithsonian Institution took plaster casts of this tibia, the original of which was preserved by Dr. 
 Winslow in Boston at the time of this report in 1876. Now this bone was found in digging a well, where Dr. 
 Frederick Zeile’s baths were afterwards located, in the rear of 524 to 528 Pacific Street, San Francisco, about 
 the year 1852. The bone was found at a depth of 23 feet, and one of the workmen informed Dr. Winslow that 
 the whole skeleton was still imbedded in the yellow silt through which the well was dug until water was reached. 
 The skeleton was estimated to lie at a depth of 23 feet below the surface. “When Dr. Zeile’s brick building was 
 put up, Dr. Winslow observed that the rear wall just embraced the well within its area; and he has also con¬ 
 sidered it possible to reach the skeleton without injury to the edifice, by careful excavation. This gigantic fossil 
 is probably entirely new to science, and would be of great value to the collections of the Academy.” 
 
 ° C. Stock. The systematic position of the mylodont sloths from Rancho La Brea. Science, n. s., vol. 39, pp. 7G1- 
 
 7G3, 1914. 
 
 6 Geological Atlas of the United States, San Francisco Folio, No. 193, p. 15, 1914. 
 
202 
 
 CENOZOIC GRAYIGRADE EDENTATES OF WESTERN NORTH AMERICA. 
 
 A considerable part of Pacific Street was burned down during the recent catastrophe in San Francisco, and 
 if the building in question is among those that have been razed preparatory to reconstruction, there might be 
 an excellent opportunity to secure the rest of the skeleton. Perhaps the effort was made to secure the skeleton 
 sometime after 1875. All the information I have is from the Proceedings of the California Academy of Science, 
 volume 6, for 1875, page 141. 
 
 I have written this thinking that perhaps the Museum presumably has a plaster cast of the tibia; it 
 might be worth while to make some effort to secure the rest of the skeleton in the event that it has not yet been 
 excavated. 
 
 Very truly yours, 
 
 Walter F. Swingle, 
 Physiologist in Charge of Plant Life History. 
 
 The tibia discovered by Dr. Winslow and determined by Joseph Leidy may have 
 belonged to Mylodon. No further information is known to the writer regarding the 
 Pacific Street locality. 
 
 Sonoma County, California. —Mr. William McGrew, of Petaluma, California, dis¬ 
 covered at a locality approximately a mile west and slightly south of the station of 
 Stoney Point, a pelvis and several vertebrae of Mylodon. Stoney Point is 7 miles north¬ 
 west of Petaluma. The ground-sloth remains occurred in a bluish clay with gravels of 
 Pleistocene age, overlying unconformably a sandstone belonging probably to the Francis¬ 
 can series (Jurassic?). 
 
 Marysville, California. —A fragmentary distal end of a right femur (No. 24245), pre¬ 
 sumably belonging to Mylodon, was presented to the Department of Palaeontology, 
 University of California, by Dr. J. H. Barr, through Mr. I. C. Schumacher. The label 
 attached to the specimen states that it was found on the ranch of N. S. Wilson, in an ex¬ 
 cavation caused by the flood of 1890. The locality is near Marysville, California. No. 
 24245 is comparable in size to the corresponding part in femora belonging to Mylodon 
 liarlani from Rancho La Brea. The articulating surfaces of the condyles are continuous 
 with that for the patella, as in the femur of Mylodon. The width of the intercondyloid 
 space is greater than in average specimens of Mylodon liarlani from Rancho La Brea. 
 
 Point Conception, California. —A cervical vertebra, a rib fragment, and a complete 
 fibula belonging to Mylodon were collected on a raised beach, 120 feet high, a mile north 
 of Conception Station on the Southern Pacific Railroad. The specimens are now in the 
 collections of the Museum of Palaeontology, University of California. The cervical 
 vertebra (No. 11261) and the fibula (No. 11262) are comparable in size and shape to 
 similar elements from the Rancho La Brea Pleistocene. 
 
 Eldorado County, California. —The presence of megalonychid ground-sloths in the 
 Hawver Cave deposit has been already discussed (seepp. 115,116). A first lower tooth and 
 two phalanges belonging to Mylodon harlani were also found (Stock, 1918, pp. 489-491, 
 figs. 7 to 105). The mylodont ground-sloths probably encountered little difficulty in 
 extending their range over the foothill belt, and their presence in deposits as at Hawver 
 Cave should be expected where such accumulations were taking place during the Pleisto¬ 
 cene, in fairly close proximity to the region of the Great Valley of California. 
 
 McKittrick, California. —Remains of mylodont ground-sloths have been found asso¬ 
 ciated with other Pleistocene mammals in an asphalt deposit near McKittrick, on the 
 southwest side of the San Joaquin Valley. 0 The forms so far recognized in the assem¬ 
 blage are listed below: 
 
 Aenocyon dirus (Leidy). 
 Canis, near ochropus Esch. 
 Felis atrox Leidy. 
 
 Felis, near daggetti Merriam. 
 
 Arctotherium, near simum Cope. 
 Taxidea sp. 
 
 Mylodon? sp. 
 
 Equus occidentalis Leidy. 
 
 Antilocapra? sp. 
 
 Bison sp. 
 
 Camel, slender-limbed form. 
 Mammut, near americanum (Kerr). 
 
 ° J. C. Merriam and C. Stock. Occurrence of Pleistocene vertebrates in an asphalt deposit near McKittrick, Cali¬ 
 fornia. Science, n. s., vol. 54, pp. 566-567, 1921. 
 
PLEISTOCENE MEGALONYCHINAE AND MYLODONTIDAE OF RANCHO LA BREA. 203 
 
 A review of the fauna indicates several types, for example Aenocyon dims, Felis 
 atrox, Equus occidentalis, that are found also at Rancho La Brea. Striking differences 
 between the two faunas are seen in the absence of the saber-tooth tiger at McKittrick 
 and the presence of a slender-limbed camel generically distinct from the Rancho La Brea 
 Camelops and closely related to Lama. The environment of the Great Valley of Cali¬ 
 fornia seems to be suggested by the presence of the antelope. That conditions of life 
 during the Pleistocene, in the region of McKittrick, were quite unlike those in the vicin¬ 
 ity of Rancho La Brea seems entirely probable. This becomes evident also from a 
 study of the Pleistocene birds of McKittrick, made by Dr. L. H. Miller. There appears 
 to be some reason for interpreting the differences between the McKittrick and Rancho 
 La Brea faunas on the basis of a time separation in the Pleistocene. 
 
LITERATURE. 
 
 In the following list an attempt is made to include only the more specific references to the literature on the gravi- 
 gradc edentates bearing particularly on the Megalonychinae and the Mylodontidae. References of more general 
 nature are included in the text as footnotes. 
 
 Allen, G. M., 1913. A new Mylodon. Mem. Mus. Comp. 
 
 Zool. Harv. Coll., vol. 40, pp. 319-346, 4 pis. 
 Ameghino, F., 1889. Contribucion al eonocimiento de los 
 Mammiferos fosiles de la Republico Argentina. 
 Act. Acad. nac. de cienc. Cordoba, pp. 661-757, 
 pis. 39-49. 
 
 -, 1899. Further remarks on Neomylodon listai. Proc. 
 
 Zool. Soc. Lond., p. 830. 
 
 -, 1902. Notas sobre algunos mamiferos fosiles nuevos 
 
 o poco conocidos del valle de Tarija. An. Mus. 
 Nac. Buenos Aires, ser. 3, vol. 1, pp. 225-261, 7 pis. 
 
 -, 1907. Notas sobre una pequena coleccion de huesos 
 
 mamiferos procendentes de las grutas calcareas de 
 Iporanga en el Estado de Sao Paulo-Brazil. Rev. 
 Mus. Paulista, vol. 7, pp. 59-124. 
 
 Anthony, H. E., 1918. The indigenous land mammals of 
 Porto Rico, living and extinct. Mem. Amer. Mus. 
 Nat. Hist., vol. 2, pp. 331-435, pis. 55-75. 
 
 Boule, M., 1920. Mammiferes fosiles de Tarija. With col¬ 
 laboration of A. Thevenin. (Mission scientifique 
 
 G. de Crequi-Monfort et E. Senechal de la Grange.) 
 
 H. le Soudier (Paris), Roy. 4to., pp. vii + 255. 
 Brown, B., 1903. A new genus of ground sloth from the 
 
 Pleistocene of Nebraska. Bull. Amer. Mus. Nat. 
 Hist., vol. 19, pp. 569—583, pis. 50, 51. 
 Burmeister, H., 1864-1869. Lista de los mamiferos fosiles 
 del terreno diluviano. An. Mus. Publ. Buenos 
 Aires, vol. 1, p. 173, pi. 5, figs. 8a, 85. 
 
 -, 1865. Hautpanzer bei Mylodon. Arch. f. Anat. 
 
 Physiol, u. wissensch. Med., pp. 334-336. 
 
 --, 1879. Descriptive physique de la Republique Argen¬ 
 tine d’apres des observations personnelles et etran- 
 geres. Vol. 3, Animaux vertebres, pt. 1, mammi¬ 
 feres vivants et eteints. Edentata, pp. 274-449. 
 
 — --, 1881a. Bericht iiber ein Skelet von Scelidotherium 
 
 leptocephalum. Monatsber. Akad. Wiss. Berlin, 
 pp. 374-381, 1 pi. 
 
 -, 18816. Atlas de la description physique de la Re¬ 
 publique Argentine. Deuxieme section, mammi¬ 
 feres. Dritte Abhandlung, Osteologie der Gravi- 
 graden oder Riesenfaulthiere, pp. 65-125, pis. 12-16. 
 
 -, 1882. Nothropus priscus, ein bisher unbekanntes 
 
 fossiles Faulthier. Sitzber. k. preuss. Akad. Wiss. 
 Berlin, pp. 613-620, pi. 11. 
 
 ■-, 1885. Berichtigung zu Coelodon. Sitzber. k. preuss. 
 
 Akad. Wiss. Berlin, pp. 567-573, pi. 5. 
 
 -, 1886a. Weitere bemerkungen iiber Coelodon. 
 
 Sitzber. k. preuss. Akad. Wiss. Berlin, pp. 357, 358. 
 
 -— -, 18866. Nochmalige berichtigung zu Coelodon. 
 
 Ibid., pp. 1127-1132. 
 
 -, 1887. Neue beobachtungen von Coelodon. Ibid., 
 
 pp. 857-862. 
 
 Claypole, E. W., 1891. Megalonyx in Holmes County, 
 Ohio, 1890. Amer. Geol., vol. 7, pp. 122-132; 149- 
 153. 
 
 Cockerell, T. D. A., 1909. A fossil ground-sloth in Colo¬ 
 rado. Univ. Colo. Studies, vol. 6, pp. 309-312, 2 
 pis. 
 
 Cope, E. D., 1871. Preliminary report on the vertebrata 
 discovered in the Port Kennedy bone cave. Proc. 
 Amer. Philos. Soc., vol. 12, pp. 73-102. 
 
 -, 1878. Descriptions of new extinct vertebrata from 
 
 the upper Tertiary and Dakota formations. Bull. 
 U. S. Geol. and Geog. Surv. Terr., vol. 4, pp. 379- 
 396. 
 
 — -, 1880. Mylodon from Klamath River, near Yreka 
 
 California. Amer. Nat., vol. 14, p. 62. 
 
 Cope. E. D., 1889. The Edentata of North America. 
 Amer. Nat., vol. 23, pp. 657—664, pis. 31-32. 
 
 -, 1893. A preliminary report on the vertebrate pale¬ 
 ontology of the Llano Estacado. 4th Ann. Rept. 
 Geol. Surv. Texas, 1892, 136 pp., 23 pis. 
 
 -, 1895. On some Pleistocene mammalia from Petite 
 
 Anse, La. Proc. Amer. Philos. Soc., vol. 34, pp. 
 458-468, pis. 10-12. 
 
 -, 1899. Vertebrate remains from Port Kennedy bone 
 
 deposit. Jour. Acad. Nat. Sci. Phila., ser. 2, vol. 
 11, pp. 193-267, pis. 18-21. 
 
 Cordovez, M., 1903. Los restos del Mylodon i la gruta de 
 su nombre en la Patagonia occidental. Actas Soc. 
 cient. Chile, vol. 12, pp. 1-19, 2 pis. 
 
 Eaton, G. F., 1914. Vertebrate fossils from Ayusbamba, 
 Peru. Amer. Jour. Sci., ser. 4, vol. 37, pp. 141-154, 
 pis. 5, 6, and 7. 
 
 Flower, W. H., 1882 On the mutual affinities of the ani¬ 
 mals composing the order Edentata. Proc. Zool. 
 Soc. Lond., pp. 358-367. 
 
 Gervais, H., and F. Ameghino, 1880. Les mammiferes 
 fossiles de L’Amerique du Sud. Paris and Buenos 
 Aires, 225 pp. 
 
 Gervais, P., 1855. Recherches sur les mammiferes fossiles 
 de l’Amerique meridionale. Extrait de la Zoologie 
 de l’Expedition dans les parties centrales de l’Ame- 
 rique du Sud, publiee sous la direction de M. le 
 comte Francois de Castelnau. Paris, 64 pp., 13 pis. 
 
 -, 1869. Memoire sur les formes cerebrales propres 
 
 aux fidentes vivants et fossiles. Nouv. Arch. Mus. 
 d’Hist. Nat. Paris, vol. 5, pp. 1-56, pis. 1-5. 
 
 -, 1873. Memoire sur plusieurs especes de mammi¬ 
 feres fossiles propres a l’Amerique meridionale. 
 Mem. Soc. Geol. France, ser. 2, vol. 9, 44 pp., pis. 
 21-29. 
 
 Harlan, R., 1825. Fauna Americana: being a description 
 of the mammiferous animals inhabiting North 
 America, pp. 199-203. 
 
 -, 1835. Medical and physical researches, pp. 271- 
 
 275; 314-318; 319-333; 334-336; 336; pis. 12-15, 16. 
 
 -, 1843. Description of the bones of a new fossil ani¬ 
 mal of the order Edentata. Amer. Jour. Sci., ser. 
 1, vol. 44, pp. 69—80, 3 pis. 
 
 Hatjthal, R., 1900a. Die haustiereigenschaft des Grypo¬ 
 therium domesticum Roth, die glacialverhiiltnisse 
 bei Ultima Esperanza und die berechtigung des 
 namens Grypotherium domesticum. Globus, vol. 
 78, pp. 333-338; 357-360. 
 
 -, 19006. Quelques rectifications relatives au Grypo¬ 
 therium de la caverne Eberhardt. Comunicaciones 
 del Mus. Nac. Buenos Aires, vol. 1, pp. 241-252. 
 
 -, 1904. Die bedeutung der funde in der Grypother- 
 
 iumhohle bei Ultima Esperanza (siidwestpata- 
 gonien). Ber. Senkenb. Naturf. Ges. Frankfurt a. 
 M., pp. 89-91, Dec. 12, 1903. 
 
 Hauthal, R., S. Roth, and R. Lehmann-Nitsche. 1899. 
 
 El mamifero misterioso de la Patagonia “Grypo¬ 
 therium domesticum.” Rev. Mus. de la Plata, vol. 
 9, pp. 409-473, 5 pis. 
 
 Hay, O. P., 1916. Descriptions of two extinct mammals 
 of the order Xenarthra from the Pleistocene of 
 of Texas. Proc. U. S. Nat. Mus., vol. 51, pp. 
 107-123, pis. 3-7. 
 
 -, 1919. Descriptions of some mammalian and fish 
 
 remains from Florida of probably Pleistocene age. 
 Proc. U. S. Nat. Mus., vol. 56, pp. 103-112, pis, 
 26-28. 
 
 204 
 
205 
 
 LITERATURE. 
 
 Holmes, F. S., 1858. Remains of domestic animals among 
 fossils in South Carolina. Charleston, S. C., 16 
 pp. (Reviewed in Amer. Jour. Sci., ser. 2, vol. 
 25, pp. 442-443.) 
 
 Jacob, C., 1902. Examen microscopico de in pieza cutanea 
 del mamifero misterioso de la Patagonia “ Grypo - 
 therium domesticurn Rev. Mus. de la Plata, vol. 
 
 10, pp. 61-62, 1 pi. 
 
 - , 1911. Vom tierhirn zum mensclienhirn. Verglei- 
 
 chend morphologische, histologische u. biologische 
 studien zur entwicklung der grosshirnhemispharen 
 und ihre rinde. 
 
 Jefferson, T., 1799. A memoir on the discovery of certain 
 bones of a quadruped of the clawed kind in the 
 western parts of Virginia. Trans. Amer. Philos. 
 Soc., vol. 4, pp. 246-260. 
 
 Kraglievich, L., 1921a. Descripcion comparativa del 
 genero Pleurolestodon. Rev. Mus. Nac. Buenos 
 Aires, vol. 31, pp. 95-118. 
 
 - , 1921. Estudios sobre los Mylodontinae. Descrip¬ 
 cion del cranes y mandibula del Pseudolestodon 
 myloides gallenii, n. sbsp. An. Mus. Nac. Buenos 
 Aires, vol. 31, pp. 119-134. 
 
 -, 1922. Estudios sobre los Mylodontinae. An&lisis 
 
 comparado de los valores cr&neometri<;os de los 
 Milodontinos de Norte y Sud America. An. Mus. 
 Nac. Hist. Nat. Buenos Aires, vol. 31, pp. 457-464. 
 
 Le Conte, J., 1882. On certain remarkable tracks, found 
 in the rocks of Carson quarry. Proc. Calif. Acad. 
 Sci., Aug. 27, 10 pp. 
 
 -, 1883. Carson footprints, a letter to Nature. 
 
 Nature, vol. 28, pp. 101-102. 
 
 Lehmann-Nitsche, R., 1902. Die gleichzeitigkeit der 
 siidpatagonischen hohlenbewohner mit dem Grypo- 
 therium und anderen ausgestorbenen thieren der 
 argentinischen hohlenfauna. Arch. Anthrop. 
 Braunschweig, vol. 27. 
 
 Leidy, J., 1855. A memoir on the extinct sloth tribe of 
 North America. Smithson. Contrib. Knowl., vol. 
 7, 68 pp., 16 pis. 
 
 -, 1860. Remarks on the structure of the feet of 
 
 Megalonyx. Trans. Amer. Philos. Soc., n. s., vol. 
 
 11, p. 107, pi. 6. 
 
 -, 1868. Notice of some vertebrate remains from the 
 
 West Indian Islands. Proc. Acad. Nat. Sci. Phila., 
 vol. 20, pp. 178-180. 
 
 -, 1870. Remarks on Mylodon? robustus, from Central 
 
 America, and on Dromatherium sylvestre. Proc. 
 Acad. Nat. Sci. Phila., vol. 22, pp. 8-9. 
 
 -, 1885. Remarks on Mylodon. Proc. Acad. Nat. 
 
 Sci. Phila., vol. 37, pp. 49-51, 
 
 -, 1889. Notice of some mammalian remains from the 
 
 salt mine of Petite Anse, La. Trans. Wag. Free 
 Inst. Sci., vol. 2, pp. 33-40. 
 
 Lindahl, J, 1892. Description of a skull of Megalonyx 
 leidyi, n. sp. Trans. Amer. Philos. Soc., vol. 17, 
 pp. 1-10, pis. 1-5. 
 
 Loomis, F. B., 1914. The Deseado formation of Patagonia. 
 Concord, 232 pp. 
 
 Lonnberg, E., 1899. On some remains of “Neomylodon 
 listai” Ameghino brought home by the Swedish 
 expedition to Tierra del Fuego 1896. Wissensch. 
 Ergebn. Schwedish. Exped. Magellansland. 1895— 
 1897 unter leitung von Dr. Otto Nordenskjold, 
 vol. 2, pp. 149-170, pis. 12-14. 
 
 Lull, R. S., 1915. A Pleistocene ground sloth, Mylodon 
 harlani, from Rock Creek, Texas. Amer. Jour. 
 Sci., ser. 4, vol. 39, pp. 327-385. 
 
 Lund, P. AY., 1836. Om huler i kalksteen i det cudre af 
 Brazilien, der tildeels ondeholde fossile lcnokler. 
 Kgl. Danske Vidensk. Selsk. Naturwidenskabelig. 
 og math. 1 Afd., 6 deel, pp. 209-249, pis. 1 and 2. 
 
 - , 1839. Recherches sur les mammiferes fossiles du Bre- 
 
 sil. Ann. Sci. Natur., ser. 2 (Zool), vol. 11, p. 220. 
 
 - , 1842. Blik paa Braziliens dyreverden for sidste 
 
 jordomvaeltning. Kgl. Dansk. Vid. Selsk, Skr., 4 
 Raekke, 9 deel, pp. 137-208, pis. 28-38. 
 
 Lutken, C., 1886. Antikritiske bemaerkninger i anledning 
 af kaempledovendyrslaegten Coelodon. Over- 
 siggt K. Danske Selsk. Forhandl., pp. 78-84. 
 
 Lydekker, R., 1886. Description of three species of Scelido- 
 therium. Proc. Zool. Soc. Lond., pp. 491-498, 
 pis. 46-49. 
 
 -, 1894. Contributions to a knowledge of the fossil 
 
 vertebrates of Argentina. Pt. 2: Extinct edentates 
 of Argentina. An. Mus. de la Plata, Paleontologia 
 Argentina, pt. 3, 103 pp., 69 pis. 
 
 Marsh, O. C., 1874. Notice of new Tertiary mammals. 
 Amer. Jour. Sci., ser. 3, vol. 7, pp. 531-532. 
 
 -, 1883. On the supposed human foot-prints recently 
 
 found in Nevada. Amer. Jour. Sci., ser. 3, vol. 26, 
 pp. 139-140. 
 
 -, 1897. The Stylinodontia, a suborder of Eocene 
 
 edentates. Amer. Jour. Sci., ser. 4, vol. 3, pp. 
 137-146. 
 
 Matthew, W. D., 1911. The ground-sloth group. Amer. 
 Mus. Jour., vol. 11, pp. 113-119. 
 
 -, 1912. Ancestry of the edentates. Amer. Mus. 
 
 Jour., vol. 12, pp. 300-303. 
 
 -, 1915. Climate and evolution. Ann. N. Y. Acad. 
 
 Sci., vol. 24, pp. 259-262. 
 
 -, 1919. Recent discoveries of fossil vertebrates in the 
 
 West Indies and their bearing on the origin of the 
 Antillean fauna. Proc. Amer. Philos. Soc., vol. 58, 
 
 pp. 161-181. 
 
 Merriam, J. C., 1899. Ground-sloths in the California 
 Quaternary. Bull. Geol. Soc. Amer., vol. 11, pp. 
 612-614, pi. 58. 
 
 -, 1900. On the occurrence of ground-sloths in the 
 
 Quaternary of middle California. Science, n. s., 
 vol. 11, p. 219. 
 
 -, 1906. Recent discoveries of Quaternary mammals 
 
 in southern California. Science, n. s., vol. 24, pp. 
 248-250. 
 
 Moreno, F. P., 1899. On a portion of mammalian skin, 
 named Neomylodon listai, from a cavern near 
 Consuelo Cove, Last Hope Inlet, Patagonia. 
 Proc. Zool. Soc. Lond., pp. 144-148. 
 
 Nordenskjold, E., 1899. Neue untersuchungen iiber Neo¬ 
 mylodon listai (vorlaufige mittheilung). Zool. 
 Anz., vol. 22, pp. 335-336. 
 
 -, 1900. Iakttagelser och fynd grottor vid Ultima 
 
 Esperanza i sydvestra Patagonien. Kongl. Svenska 
 Vetensk.-Akad. Handl., Stockholm., vol. 33, 24 
 pp., 7 pis. 
 
 -, 1901. Tiber die saugethierfossilien im Tarijathal, 
 
 Siidamerika. Bull. Geol. Instit. Univ. Upsala, 
 vol. 5, pp. 261-266. 
 
 -, 1908. Ein neuer fundort fur saugetierfossilien in 
 
 Peru. Arkiv for Zoologi, vol. 4, 22 pp., 2 pis. 
 
 Osborn, H. F., 1910. The age of mammals in Europe, Asia, 
 and North America. N. Y., xvn + 635 pp. 
 
 Owen, R., 1840. The zoology of the voyage of H. M. S. 
 
 Beagle, 1832-36. Pt. 1, Fossil Mammalia, Lond., 
 pp. 73-99, pis. 20-28. 
 
 -, 1842. Description of the skeleton of an extinct 
 
 gigantic sloth, Mylodon robustus Owen, with 
 observations on the osteology, natural affinities, 
 and probable habits of the megatherioid quadrupeds 
 in general. London, 176 pp., 24 pis. 
 
 —--, 1843. Letter from Richard Owen, esq., F. R. S., 
 
 F. G. S., etc., etc., on Dr. Harlan’s notice of new 
 fossil mammalia. (Letter dated 1842). Amer. 
 Jour. Sci., ser. 1, vol. 44, pp. 341-345. 
 
 -, 1857. On the scelidothere (Scelidotherium leptoce- 
 
 phalum Owen). Phil. Trans. Roy. Soc. Lond., vol. 
 147, pp. 101-110, pis. 8-9. 
 
 Perkins, H. C., 1842. Notice of fossil bones from Oregon 
 Territory, in a letter to Dr. C. T. Jackson. Amer. 
 Jour. Sci., ser. 1, vol. 42, pp. 136-140. 
 
 Philippi, R. A., 1893. Vorlaufige nachricht iiber fossile 
 saugethierknochen von Ulloma, Bolivia. Zeitschrift 
 deutsch. Geol. Ges.. vol. 45, pp. 87-96. 
 
206 
 
 LITERATURE. 
 
 Philippi, R. A., 1900. Contribucion & la ost^olojia del 
 Grypotherium domesticum Roth i un neuvo delfin. 
 Anal, de la Univ. de Chile, vol. 107, pp. 105-114, 
 4 pis. 
 
 -, 1901. Beitrage zur kenntniss der knochen des 
 
 Grypotherium domesticum Roth. Arch. f. natur- 
 gesch., 67, vol. 1, pp. 271-275, pi. 15. 
 
 Pouchet, G., 1868-1869. Memoire sur l’enc^phale des 
 fldentes. Jour, de l’anatomie et de la physiologie, 
 vol. 5, 1868, pp. 658-675; vol. 6, 1869, pp. 1-18, 
 147-162, 302-316, 349-363, pis. 1-6. 
 
 Rautenberg, M., 1906. Uber Pseudolestodon hexaspon- 
 dylus. Palaeontographica, vol. 53, pp. 1-50, pis. 
 1-6. 
 
 Reinhardt, J., 1866. Ueber den hautpanzer der megathe- 
 roiden thiere. Arch. Anat. Physiol, u. wissensch. 
 Med., pp. 414-415. 
 
 -, 1875. Bidrag til kundskab om kjaempodovendyret 
 
 Lestodon armatus. Vidensk. Selsk. Skr., 5 Raekke, 
 naturvidenskabelig og mathematisk Afd., vol. 11, 
 pp. 1-38, 3 pis. 
 
 -, 1878. Kaempedovendyr-Slaegten Coelodon. 
 
 Vidensk, Selsk, Skr., 5 Raekke, naturvidenskabelig 
 og mathematisk Afd., vol. 12, pp. 257-335, 5 pis. 
 
 -, 1879. Beskrivelse af Hovedskallen af et Kaempe- 
 
 dovendyr, Grypotherium darwinii, fra La Plata- 
 Landenes pleistocene Dannelser. Vidensk. Selsk. 
 Skr., 5 Raekke, naturvidenskabelig og mathe¬ 
 matisk Afd., vol. 12, pp. 351-380, 2 pis. 
 
 Richters, F., 1904. Mikr. untersuchung von Grypo¬ 
 therium - Dung. Ber. Senckenb. Naturf. Ges. 
 Frankfurt a. M., p. 46. 
 
 Ridewood, W. C., 1901. On the structure of the hairs of 
 Mylodon listai and other South American Edentata. 
 Quart. Jour. Micros. Sci., vol. 44, n. s., pp. 393-411, 
 pi. 26. 
 
 Rovereto, C., 1914. Los Estratas Araucanos y sus fosiles. 
 
 Anal. Mus. Nac. Hist. Nat. Buenos Aires, vol. 25, 
 249 pp., 31 pis. 
 
 Safford, J. M., 1892. The pelvis of a Megalonyx and other 
 bones from Big Bone Cave, Tennessee. Bull. 
 Geol. Soc. Amer., vol. 3, pp. 121-123. 
 
 Scott, W. B., 1903-04. Reports of the Princeton Uni¬ 
 versity expeditions to Patagonia, 1896-1899, 
 J. B. Hatcher in charge. Vol. 5, Palaeontology, 
 Mammalia of the Santa Cruz beds. pp. 161-364, 
 pis. 30-63. 
 
 -, 1913. A history of land mammals in the western 
 
 hemisphere, xiv + 693 pp., 32 pis. 
 
 Sefve, I., 1915a. Uber einen Scelidotherium-Schadel aus 
 Tarija, Bolivia. Kongl. Svenska Vetensk.-Akad. 
 Handl., Stockholm, vol. 53, 12 pp., 2 pis. 
 
 -, 19155. Scelidotherium-Reste aus Ulloma, Bolivia. 
 
 Bull. Geol. Instit. Univ. Upsala, vol. 13, pp. 61-92, 
 pis. 10 and 14. 
 
 Schulthess, B., 1920. Beitrage zur kenntnis der Xenarthra 
 auf grund der “Santiago Roth’schen sammlung” 
 des zoologischen museums der Universitat Zurich. 
 Mem. Soc. Pale. Suisse, vol. 44, 119 pp., 6 pis. 
 
 Sinclair, W. J., 1905. New mammalia from the Quaternary 
 caves of California. Univ. Calif. Publ., Bull. Dept. 
 Geol., vol. 4, pp. 145-161, pis. 19-23. 
 
 -, 1906. Some edentate-like remains from the Mascall 
 
 beds of Oregon. Ibid., vol. 5, pp. 65-66. 
 
 -, 1910. Dermal bones of Paramylodon from the 
 
 asphaltum deposits of Rancho La Brea, near Los 
 Angeles, California. Proc. Amer. Philos. Soc., 
 vol. 49, pp. 191-195. 
 
 Stock, C., 1913. Nothrotherium and Megalonyx from the 
 Pleistocene of southern California. Univ. Calif 
 Publ. Bull. Dept. Geol., vol. 7, pp. 341-358. 
 
 -, 1914. Skull and dentition of the mylodont sloths of 
 
 Rancho La Brea. Ibid., vol. 8, pp. 319-334. 
 
 -, 1917a. Recent studies on the skull and dentition 
 
 of Nothrotherium from Rancho La Brea. Ibid., 
 vol. 10, pp. 137-164. 
 
 -, 19175. Further observations on the skull structure 
 
 of mylodont sloths from Rancho La Brea. Ibid., 
 vol. 10, pp. 165-178, pis. 3-4. 
 
 -, 1917c. Structure of the pes in Mylodon harlani. 
 
 Ibid., vol. 10, pp. 267-286. 
 
 -, 1918. The Pleistocene fauna of Hawver Cave. 
 
 Ibid., vol. 10, pp. 461-515. 
 
 -, 1920a. Origin of the supposed human footprints 
 
 of Carson City, Nevada. Science, n. s., vol. 51, p. 
 514. 
 
 -, 19205. A mounted skeleton of Mylodon harlani. 
 
 Univ. Calif. Publ. Bull. Dept. Geol., vol. 12, pp. 
 425-430, pi. 51. 
 
 Studer, T., 1905. Ueber neue funde von Grypotherium 
 listai Amegh, in der Eberhards hohle von Ultima 
 Experanza. Neue Denkschr. allg. schweiz. Ges. 
 Nat., vol. 40, pp. 1-18, pis. 1-3. 
 
 Wing e, H., 1915. Jordfundne og nulevende Gumlere 
 ( Edentata ) fra Lagoa Santa, Minas Geraes, Brasi- 
 lien. Med Udsigt over Gumlernes indbyrdes 
 Slaegtskab. E Museo Lundii, vol. 3, pt. 2, 321 
 pp. 42 pis. 
 
 Woodward, A. S., 1899. On a portion of mammalian skin, 
 named Neomylodon listai, from a cavern near 
 Consuelo Cove, Last Hope Inlet, Patagonia. With 
 a description of the specimen. Proc. Zool. Soc. 
 Lond., pp. 148-156, pis. 13-15. 
 
 -, 1900. On some remains of Grypotherium ( Neo¬ 
 mylodon) listai and associated mammals from a 
 cavern near Consuelo Cove, Last Hope Inlet, Pata¬ 
 gonia. Proc. Zool. Soc. Lond., pp. 64-79, pis. 5-9. 
 
 Wyman, L. E., 1922. Notes on the Pleistocene fossils 
 obtained from Rancho La Brea asphalt pits. Los 
 Angeles Museum, Misc. Publ. No. 2, 35 pp., illust. 
 

STOCK 
 
 PLATE 1 
 
 View looking north, showing Los Angeles Museum Pits 61 and 67, Rancho La Brea. Photo¬ 
 graph by L. E. Wyman. Courtesy of Los Angeles Museum. 
 
 View looking northeast, showing final stage in removal of wall separating Los Angeles 
 Museum Pits 61 and 67, Rancho La Brea. Photograph by L. E. Wyman. Courtesy 
 of Los Angeles Museum. 
 
5jv;V--PP. v l -'V 
 
STOCK 
 
 PLATE 2 
 
 Nothrotherium shastense Sinclair. 
 
 Figs. 1 and 2. Skull, No. 1800-11, dorsal and ventral views. X 0.44. 
 
 1 igs. .1 and 4. Skull and mandible, No. 1800—11 and No. 1801—7, anterior and posterior views. X 0.44. 
 Los Angeles Museum Collection. Rancho La Brea Pleistocene. 
 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 Wfc tmiRY 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 «w»snnr sjfF mm 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
STOCK 
 
 PLATE 3 
 
 3 6 
 
 Nothrotherium shastense Sinclair. 
 
 Figs. 1, 2, and 3. Endocranial cast. Fig. 1, dorsal view; Fig. 2, lateral view; Fig. 3, ventral view. X 0.75. 
 
 Hapcilops sp. 
 
 Figs. 4, 5,"and 6. Endocranial cast. Fig. 4, dorsal view; Fig. 5, lateral view; Fig. 6, ventral view, x 0.75. 
 
 Los Angeles Museum Collection. 
 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 flUfVffc ‘iV 1 )l ' 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
STOCK 
 
 PLATE 4 
 
 Nothrotherium shastense Sinclair. 
 
 Fig. 1. Skull and mandible, No. 1800-11 and No. 1801-7, lateral view. X 0.44. 
 
 Fig. 2. Mandible, No. 1801-7, dorsal view. X 0.44. 
 
 Figs. 3 and 4. Cervical vertebrae I-VII inclusive, lateral and dorsal views. X 0.25. 
 Los Angeles Museum Collection. Rancho La Brea Pleistocene. 
 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
STOCK 
 
 PLATE 5 
 
 Nothrotherium shastense Sinclair. 
 
 Figs. 1 and 2. Thoracic and lumbar vertebrae, dorsal and lateral views. X 0.25. 
 
 Fig. 3. First thoracic vertebra, No. 1809-1, anterior and ventral views. X 0.25. az, anterior zygapophysis; 
 
 facet for capitulum of rib; tf., facet for tuberculum of rib. 
 
 Fig. 4. Third thoracic vertebra, No. 1811—1, anterior view. X 0.25. 
 
 Fig. 5. Last thoracic vertebra, No. 1825-1, anterior view. X 0.25. 
 
 Fig. (5. First lumbar vertebra, No. 1826-1, anterior view. X 0.25. 
 
 Fig. 7. Third lumbar vertebra, No. 1828-1, anterior view. X 0.25. 
 
THfc (iSRIRY 
 of m 
 
 iwwfRsn'? (j mm 
 
STOCK 
 
 PLATE 6 
 
 Nothrotherium shastense Sinclair. 
 
 Figs. 1 and 2. Caudal vertebrae, dorsal and lateral views. X 0.25. 
 Fig. 3. First caudal vertebra, No. 1830-1, anterior view. X 0.25. 
 
 Fig. 4. Sixth caudal vertebra, No. 1835-1, anterior view. X 0.25. 
 Fig. 5. Eleventh caudal vertebra, No. 1840-1, anterior view. X 0.25. 
 Los Angeles Museum Collection. Rancho La Brea Pleistocene. 
 

 
STOCK 
 
 PLATE 7 
 
 Nothrotherium shastense Sinclair. 
 
 Fig. 1. Sternal segments and sternal ribs, ventral view. X 0.33. I and VI, first and sixth 
 sternal elements; II (?) and VIII, sternal ribs. 
 
 Fig. 2. Costal ribs, posterior view. X 0.33. I, VI, X, XIII, XVII, first costal rib fused with 
 first sternal rib, sixth, tenth, thirteenth, seventeenth costal ribs. 
 
 Los Angeles Museum Collection. Rancho La Brea Pleistocene. 
 

 
 
 
 TKt tiHRiPK 
 
 mm a- w n mm 
 
 
 
 
 
 
 
 
 
 
STOCK 
 
 PLATE 8 
 
 Nothrotherium shastense Sinclair. 
 
 Figs. 1 , la. Right humerus, No. 1874-R-l. Fig. 1 , anterior view; Fig. la, view of distal end. 
 
 Figs. 2, 2a. Right humerus, No. 1874-R-l. Fig. 2, inner view; Fig. 2a, view of proximal end. 
 
 Figs. 3, 3a. Right scapula, No. 1868-R-2. Fig. 3, outer view; Fig. 3a, view of proximal end. 
 
 Fig. 4. Right clavicle, No. 1878-R-2, anterior view. X 0.50. 
 
 Los Angeles Museum Collection. Rancho La Brea Pleistocene. 
 
 X 0.50. 
 X 0.50. 
 X 0.25. 
 
mt imm 
 or- ist 
 
 uw¥s»»it m nmm 
 
 
STOCK 
 
 PLATE 9 
 
 Nothrotherium shastense Sinclair. 
 
 I'i<;s. 1, 2 and 3. Left ulna, No. 1873 I/—1. I'i^. 1, inner view; Fig 
 1' nis. 1 and 5. Left radius without distal epiphysis, No. 1872-L-l. 
 
 Los Angeles Museum Collection. 
 
 2, outer view; Fig. 3, radial view. X 0.50. 
 
 Fig. 4, anterior view; Fig. 5, inner view, x 0.50. 
 ltuncho La Brea Pleistocene. 
 
rut i inuiir 
 
 or m 
 
 W<u«®|s 
 
 
 
STOCK 
 
 PLATE 10 
 
 Nothrotherium shastense Sinclair. 
 
 I'igs. 1 and 2. Pelvis, No. 1892-1. P'ig. 1 , posterior view; Fig. 2, dorsal view. X 0.22. 
 Los Angeles Museum Collection. Rancho La Brea Pleistocene. 
 
(Hi i 
 
 Of TNfc 
 W'VR it 
 
STOCK 
 
 PLATE 1 1 
 
 Nothrotherium shastense Sinclair. 
 
 Figs. 1 and 2. Pelvis, No. 1892-1. Fig. 1 , lateral view; Fig. 2, anterior view. X 0.22. 
 Los Angeles Museum Collection. Rancho La Brea Pleistocene. 
 
rat iiBS^Ry 
 
 Of THfc 
 
 um?vse55TT wf (ij ws 
 
STOCK 
 
 PLATE 12 
 
 Nothrotherium shastense Sinclair. 
 
 Figs. 1, la, and 16. Left femur, No. 1871-L-l. Fig. 1, anterior view; Fig. la, view of proximal end; Fig. 16, view of distal end. X 0.50. 
 Figs. 2 and 3. Left patella, No. 1895-L-l. Fig. 2, femoral view; Fig. 3, anterior view. X 0.50. 
 
 Los Angeles Museum Collection. Rancho La Brea Pleistocene. 
 

 m imm 
 or THt 
 
 mMm yr ii mm 
 
STOCK 
 
 PLATE 13 
 
 Nothrotherium shastense Sinclair. 
 
 Figs. 1 and 2. Left femur, No. 1871-1-1. Fig. 1, posterior view; Fig. 2, inner view. X 0.50. 
 Los Angeles Museum Collection. Rancho La Brea Pleistocene. 
 

 
 
 
 
 
 THt U»B4B¥ 
 
 o? m 
 
 p,W£$OT U4WM||$ 
 
 
 
 
 
 
 
 
 
 
 
 
 
STOCK 
 
 PLATE 14 
 
 Nothrotherium shastense Sinclair. 
 
 Figs. 1, la, and 16. Right tibia, No. 1870-R-l. Fig. 1, anterior view; Fig. la, view of proximal end; Fig. 16, view of distal 
 end. X 0.50. 
 
 Fig. 2. Right tibia, No. 1870-R-l, inner view. X 0.50. 
 
 Fig. 3. Right fibula, No. 1869-R-l, tibial view. X 0.50. 
 
 Los Angeles Museum Collection. Rancho La Brea Pleistocene. 
 

 
 
 
 fHt i.afufy 
 
 OF f»fc 
 
 WiVij§$JfY W MNQf$ 
 
 
 
 * 
 
 
 
 
 
STOCK 
 
 PLATE 15 
 
 Nothrotherium shastense Sinclair. 
 
 Figs. 1 and 2. Right pes. Fig. 1, dorsal view. Fig. 2 , inner view. X 0.50. 
 Los Angeles Museum Collection. Rancho La Brea Pleistocene. 
 

 V 
 
 - •*» 
 
 
 
 
 
 
Nothrotherium shcistense Sinclair. 
 
 STOCK 
 
 PLATE 16 
 
iHt iMm 
 Or THt 
 
 umvfiflsiTV w auiiqis 
 
 
 
 
 
 
 
STOCK 
 
 PLATE I? 
 
 Megalonyx jeffersonii californicus Stock. 
 
 Figs. 1 and 2. Mandible, No. 21429. Fig. 1, dorsal view; Fig. 2, view of left side. X 0.50. 
 
 University of California Collection. Rancho La Brea Pleistocene. 
 
STOCK 
 
 PLATE 18 
 
 Megalonyx jeffersonii californicus Stock. 
 
 Figs. 1, 1 a, and 2. First thoracic vertebra, No. 22111 U.C. C. Fig. 1, anterior view; Fig. la, ventral view; Fig. 2, 
 lateral view; X 0.33; az, anterior zygapophysis; c.f., facet for capitulum of rib; /./., facet for tubercultfm of rib. 
 Figs. 3 and 4. Anterior thoracic vertebra, No. 24243 U. C. C. Fig. 3, anterior view; Fig. 4, lateral view, x 0.33. 
 Figs. 5 and 0. Posterior thoracic vertebra, No. 24244 U. C. C. Fig. 5, anterior view; Fig. 6, lateral view, x 0.33. 
 Figs. 7, 7a and 8, 8a. Two caudal vertebrae, Nos. 6003-1, 6003-2 L. A. M.; anterior and dorsal views. X 0.33. 
 Specimens referred tentatively to M. j. californicus. 
 
 Los Angeles Mus. and T'niv. California Collections. Rancho La Brea Pleistocene. 
 

 
 of m 
 
 
 
 
 
 
 
 
STOCK 
 
 PLATE 19 
 
 Megalonyx jeffersonii californicus Stock. 
 
 Figs. 1, la, and 16. Loft humerus, No. 21003. Fig. 1, anterior view; Fig. la, view of proximal end; Fig. 16, view of distal articula 
 tion. X 0.50. 
 
 Fig. 2. Left humerus, No. 21003, inner view. X 0.50. 
 
 Univ. California Collection. Rancho La Brea Pleistocene. 
 

 IHfc t mU 
 
 sjf mmm 
 
 
STOCK 
 
 PLATE 20 
 
 Megalonyx jeffersonii califamicus Stock. 
 
 Figs. 1, 2, and 3. 
 
 Loft ulna, No. 23192. Fig. 1, outer view; Fig. 2, inner view; Fig. 3, radial view. 
 Univ. California Collection. Rancho La Brea Pleistocene. 
 
 X 0.50. 
 

 
 m 4;HK4«1 
 
 Or IHt 
 
 w uxm 
 
 
STOCK 
 
 PLATE 21 
 
 Dermal ossicles. 
 
 Mylodon harlani Owen. 
 
 X 1. Los Angeles Museum Collection. Rancho La Brea Pleistocene. 
 

 
 THt t i»S4BT 
 Of TKh 
 
 umflW w tnm\% 
 
 
 
 
STOCK 
 
 PLATE 22 
 
 Mylodon harlani Owen. 
 
 Skull, No. 1717-32, dorsal view, x 0.40. Los Angeles Museum Collection. Rancho 
 
 La Brea Pleistocene. 
 

 
 
 
 
 
 
 IHfe ‘ 
 
 Of Wk 
 
 U-Hivaisiu iif 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
STOCK 
 
 PLATE 23 
 
 Mylodon harlani Owen. 
 
 Fig. 1. Skull, No. 1717-32, ventral view. X 0.40. 
 
 Fig. 2. Premaxillaries, ventral view. X 0.40. 
 
 FiG^{_llight incus. X 2. Fig^ Left malleus. X 2. 
 Los Angeles Museum Collection. Rancho La Brea Pleistocene. 
 

 
 
 
 
 m uttiufti 
 ^.vp-ar? m lumais 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Mylodon harlani Owen. 
 
 Skull and mandible, No. 1717-32 and No. 1718-1, lateral view. X 0.40. 
 Los Angeles Museum Collection. Rancho La Brea Pleistocene. 
 
 STOCK 
 
 PLATE 24 
 

 
 c 
 
 ftft 
 
 Gr m 
 
 UH1YMTY ftp fcMROIS 
 
STOCK 
 
 PLATE 25 
 

 Of THfc 
 
 
 
 
 
 
 v 
 
 
 
 
 
STOCK 
 
 PLATE 26 
 
 Mylodon harlani Owen. 
 
 Figs. 1, 2, and 3. Endocranial cast. Fig. 1, dorsal view; Fig. 2, lateral view; Fig. 3, 
 ventral view. X 0.75. 
 
 Los Angeles Museum Collection. Rancho La Brea Pleistocene. 
 

 !Mfc 
 
 or- Tut 
 
 U^VEHSSfY & IHum 
 
 
 
 
 
STOCK 
 
 PLATE 27 
 
 Mylodon harlani Owen. 
 
 Fig. 1 . Mandible, No. 1718-1, dorsal view, x 0.40. 
 
 Fig. 2. Hyoid arch, lateral view, x 0.40. Sh, stylohyal; eh, epihyal; ch, ceratohyal; bh, 
 tc, ossified thyroid cartilage. 
 
 Figs. 3 and 4. Cervical vertebrae I-VII. Fig. 3, dorsal view; Fig. 4, lateral view, x 0.25. 
 
 Los Angeles Museum Collection. Rancho La Brea Pleistocene. 
 
 basihyal; th, thvrohyal: 
 

 
 
 
 
STOCK 
 
 PLATE 28 
 

 
 
 
 
 
STOCK 
 
 PLATE 29 
 
 Mylodon harlani Owen. Caudal vertebrae and Haemapophyses. 
 
 Fk;s. 1 and 2, X 0.20; Fic;s. 3 to 7, X 0.50. 
 
 Los Angeles Museum Collection. Rancho La Brea Pleistocene. 
 
Of frit 
 
 Uv'-.:T«'.rr ( -v ^ ^ 
 
STOCK 
 
 PLATE 30 
 
 Mylodon harlani Owen. 
 
 Sternal segments and sternal ribs. X 0.25. 
 
 Los Angeles Museum Collection. Rancho La Brea Pleistocene. 
 

 m I* 
 
 Or 
 
 issssn « 
 
 
 
 
STOCK 
 
 PLATE 31 
 
 Mylodon harlani Owen. 
 
 I’ igs. 1 and 1 a. Right scapula, No. 1715-R-15. Fig. 1, outer view; Fig. la, view of proximal end. X 0.33. 
 Fig. 2. Right clavicle, No. 1738-R-l, anterior view, x 0.33. 
 
 Los Angeles Museum Collection. Rancho La Brea Pleistocene. 
 

 
 
 ‘iumis 
 
 
 
STOCK 
 
 PLATE 32 
 
 Mylodon harlani Owen. 
 
 Figs. 1 and la. Right humerus, No. 1712-R-l. Fig. 1, posterior view; Fig. la, view of proximal end. X 0.33. 
 Figs. 2 and 2a. Right humerus, No. 1712-R-l. Fig. 2, anterior view; Fig. 2a, view of distal end. x 0.33. 
 
 Los Angeles Museum Collection. Rancho La Brea Pleistocene. 
 
Ittt 
 
 Of THt 
 
STOCK 
 
 PLATE 33 
 
 Mylodon harlani Owen. 
 
 l' Kis. 1, 2, and 3. Right ulna, No. 1714-R-27. Fig. 1, outer view; Fig. 2, inner view; Fig. 3, radial view, x 0.33. 
 
 I-i<;s. 4, 4a, and 4/; Right radius, No. 1713-R-3; Fig. 4, anterior view; Fig. 4a, view of proximal articulation; Fig. 46, view of 
 distal end. x 0.33. 
 
 Fig. 5. Right radius, No. 1713-R-3, ulnar view, x 0.33. 
 
 Los Angeles Museum Collection. Rancho La Brea Pleistocene. 
 

 
 
 
 tot tmw)t 
 
 Of IHt 
 
 |i jjn.'rrn 
 
 w umn 
 
 
STOCK 
 
 PLATE 34 
 
 w rc 
 
 <» 1 <g. § n 
 
 c~*- c-*~ r- ^ | 
 
 3 3^3 
 
 c c p = 
 J® J® TT 
 
 g-o P -, '* CL 
 3 EL o 
 » 3- h » 
 2 & *-* SL 
 
 < ^ HH 
 
 <J 
 
 § 5 
 
 < Q 
 
 Co 
 
 <T> 
 
 CO 
 
 x 
 
 §fp 
 
 >o §3 
 
 3 r 
 jq 
 
 2- *, ^ 
 
 cd cd S. 
 
 08 3 L 
 
 !> P 
 I—-I 3* co 
 
 e g.s; 
 
 “32 
 
 3 S*-. 
 
 p 
 
 O Li P 
 
 - X 
 &o 
 °S. oo 
 
 co 
 
 Co 
 
 X CO 
 
 3 — o 
 • p CO 
 
 5 g 
 
 ►h 3 3 
 
 rD q 
 
 p *-i P. 
 3 P CL 
 
 2- ^ — 
 |_j ^ 
 
 o P o 
 ce 3 
 fI H . » 
 
 m a ® 
 
 w 2. 
 
 3 3* 
 
 3 CD 
 
 cr 
 
 a cd 
 
 P 3h 
 
 <r 35 
 
 CD — 
 co 3 
 
 " 3 
 
 <1 
 
 K 
 
 3 
 
 CD ^ 
 3i Cs£ 
 O 
 
 o 
 
 3 
 
 «S 
 
 3 
 
 p 
 
 crq 
 
 a 
 
 S* 
 
 o 
 
 a- 
 
 a 
 
 ~i 
 
 c-w 
 
 a 
 
 o> 
 <S> . 
 
 <! 
 
 CD 
 
 x £> 
 
 P 3 
 
 W 3- 
 
 m cr 
 
 g 
 
 cd 
 
 CD 
 
 N 
 
 C 
 
 N 
 
 5 
 
 31 
 
 / 
 
IHt t 
 
 \SMWEOH Sr XS&W 
 
 
 * 
 
 Li ' * 
 
 <•- r. ^ 
 
 v V 
 
 *':«a 
 
 :■ 
 
 V'} 
 
 
 
Mylodon harlani Owen. 
 
 Pelvis, No. 1719-1, anterior view, x 0.20. 
 
 Los Angeles Museum Collection. Rancho La Brea Pleistocene. 
 
 STOCK 
 
 PLATE 35 
 

 
 
 mi 
 
 
 
 
 l 
 
 * 
 
 
 
 
 
Mylodon harlani Owen. 
 
 Pelvis, No. 1719-1, posterior view. X 0.20. 
 
 Los Angeles Museum Collection. Rancho La Brea Pleistocene. 
 
 STOCK 
 
 PLATE 36 
 
 i 
 
 \ 
 
 

 
 
 \ty' fP V' v.f rpiMIS 
 
 
STOCK 
 
 PLATE 37 
 
 Mylodon harlani Owen. 
 
 I KjS. 1 and 2. Pelvis, X’o. 1719-1; hig. 1, lateral view; Fig. 2, dorsal view, x 0.20. 
 Los Angeles Museum Collection. Rancho La Brea Pleistocene. 
 
 

 
 
 
 
 
 
 
 
 
 ttSSISMHt ' 
 
 
 
 
STOCK 
 
 PLATE 38 
 
 Mylodon harlani Owen. 
 
 Figs. 1, la, and lb. Right femur, No. 1704-R-15. Fig. 1, anterior view; Fig. la, view of proximal end; Fig. 16, view of distal 
 end. X 0.33. 
 
 Figs. 2 and 3. Right patella, No. 1710-R-15. Fig. 2, femoral view; Fig. 3, anterior view. X 0.33. 
 
 Los Angeles Museum Collection. Rancho La Brea Pleistocene. 
 
THt t mm 
 Mr umoc? 
 
STOCK 
 
 PLATE 39 
 
 Mylodon harlani Owen. 
 
 Figs. 1 and 2. Right femur, No. 1704-R-15. Fig. 1, posterior view; Fig. 2, inner view, x 0.33. 
 Los Angeles Museum Collection. Rancho La Brea Pleistocene. 
 

STOCK 
 
 PLATE 40 
 
 Mylodon harlani Owen 
 
 Figs. 1, la, and lb. Right tibia, No. 1702-R-5. Fig. 1, anterior view; Fig. la, view of proximal end; 
 
 Fig. lb, view of distal end. X 0.33. 
 
 Fig. 2. Right tibia, No. 1702-R-5, inner view. X 0.33. 
 
 Figs. 3 and 4. Right fibula, No. 1703-R-7; Fig. 3, tibial view; Fig. 4, outer view. X 0.33. 
 
 Los Angeles Museum Collection. Rancho La Brea Pleistocene. 
 

 
STOCK 
 
 PLATE 41 
 
 Mylodon harlani Owen. 
 
 Kios. 1, 2 and 3. Right pes. Fig. 1, dorsal (outer) view; Fig. 2, inner (dorsal) view; Fig. 3, plantar (inner) 
 
 Storm ™'/ u ca ) lc “ n< '“ ra: T' «• ““bold: navicular; ec, eetocuneiform; ,nc. ,„es„. 
 
 cuneiform, m(. II, mt. V, second and fifth metatarsals; II, III, IV, V, digits. 
 
 Los Angeles Museum Collection. Rancho La Brea Pleistocene. 
 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 - 
 
 
 
 
 
STOCK 
 
 PLATE 42 
 
 Mylodon harlani tenuiceps Stock. 
 
 Fig. 1 . Ventral, posterior and dorsal views of skull, No. 1716-1. 
 
 Mylodon harlani Owen. 
 
 Figs. 2, 3, and 4. Same views as in Fig. 1. Fig. 2, No. 1717-1; Fig. 3, No. 1717-8; Fig. 4, No. 
 Los Angeles Museum Collection. Rancho La Brea Pleistocene. 
 
 Photographs by L. E. Wyman. Courtesy of Los Angeles Museum. 
 
 1717-22. 
 
STOCK 
 
 PLATE 43 
 
 1 
 
 2 
 
 5 6 7 8 
 
 Mylodon harlani Owen. Figs. 1 to 4. Mandibles, dorsal and lateral views. Fig. 1 , No. 1718-20; Fig. 2, No. 1718- 
 21; Fig. 3, No. 1718-29; Fig. 4, No. 1718-3. 
 
 Mylodon harlani tenuiceps Stock. Fig. 5. Skull, No. 1716-1, lateral view. 
 
 Mylodon harlani Owen. Figs. 6, 7, and 8. Skulls, Nos. 1717-1, 8, 22, lateral views. 
 
 Los Angeles Museum Collection. Rancho La Brea Pleistocene. Photographs by L. E. W yman. 
 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 SF 3 UK!)IJ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Mylodon harlani Owen. 
 
 Mounted skeleton, view from right side. X 0.077. 
 
 STOCK 
 
 PLATE 44 
 
Mylodon harlani Owen. 
 
 Restoration by ( liarles R. Knight under the direction of Chester Stock. 
 
 STOCK 
 
 PLATE 45 
 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ■ 
 
 
 
 
 >■ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Mylodon harlani Owen. 
 
 Replica of footprints in Pleistocene deposits at the Nevada State Penitentiary, Carson City, Nevada. Diagram indicates size of imprints, length of stride, and width of 
 straddle. Cast prepared by J. W. Lytle. 
 
 a, possible impression of ungual phalanx, digit III, manus. Los Angeles Museum Collection. 
 
 STOCK 
 
 PLATE 46 
 

 
 * 
 
 
 * 
 
 
 * 
 
 * 
 
 
 
 
 0* 
 
 
 
 
 • 
 
 
 
 
 
 
 
 
 * 
 
 
 
 
 
 
 
 
 
 
 
 * 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ' 
 
 
 
 
 
 
 
 
 
 
 - 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
STOCK 
 
 PLATE 47 
 
 Mylodon liarlani Owen. 
 
 Figs. 1 to G. Footprints on floor of tunnel in east wall of prison yard, Nevada State Penitentiary, Carson City, 
 
 Nevada.