(navigation image)
Home American Libraries | Canadian Libraries | Universal Library | Community Texts | Project Gutenberg | Children's Library | Biodiversity Heritage Library | Additional Collections
Search: Advanced Search
Anonymous User (login or join us)
Upload
See other formats

Full text of "The titanotheres of ancient Wyoming, Dakota, and Nebraska"

X /<^. /^.^:S^ 



Digitized by the Internet Archive 

in 2009 with funding from 

Boston Library Consortium IVIember Libraries 



http://www.archive.org/details/titanotheresofan01osbo 



Department of the Interior 

Ray Lyman Wilbur, Secretary 






S Oi 



U. S. GEOLOGICAL SURVEY 
George Otis Smith, Director 



Monograph 55 



THE TITANOTHERES OF ANCIENT WYOMING, 
DAKOTA, AND NEBRASKA 



BY 



HENRY FAIRFIELD OSBORN 



VOLUME 1 



\ ) i 




DocuMmrs department 

RECEIVED 

JAM 2 m? 
Wilbui Cross Library 
Univeisity ci Connecticut 



UNITED STATES 

GOVERNMENT PRINTING OFFICE 

WASHINGTON; 1929 



Note. — Monograph 55 is issued in two volumes. Volume 1 contains Chapters 
I- VII and Plates I-XLII; volume 2 contains Chapters VIII-XI, Plates XLIII- 
CCXXXVI, an appendix, and the index to both volumes. 



xi^ii 



CONTENTS 



Page 

Letter of transmittal xix 

Preface i xxi 

Vertebrate paleontology in the national surveys xxi 

Preparation of the present monograph xxi 

Work by the author, 1878-1919 xxii 

Research and collaboration xxii 

Cooperation of museums xxiii 

Work on text and illustrations xxiii 

Summary of geologic and anatomic principles xxiii 

Chapter I. Introduction to mammalian paleontology 1 

Section 1. Exploration and research made in the preparation of this monograph 1 

Section 2. Preliminary survey of the monograph and the conclusions presented 2 

Range of the titan otheres in geologic time 2 

Hay den's subdivisions of the Eocene and the Oligocene 5 

Discovery of the titanotheres of the plains 6 

Discovery of the mountain-basin environment of the titanotheres 6 

Discovery and delimitation of periods of sedimentation and of life zones 8 

Principle of local and continental adaptive radiation 10 

Comparison of the four life phases of Europe and North America during Eocene and early Oligocene time 12 

Old and new systems of classification 13 

Old terminology retained 13 

Linnaean methods of defining species, genera, and phyla of titanotheres 14 

Recognition of many lines of descent; polyphyly the key to interpretation of the family 14 

Relation of the phylogenetic classification to the Linnaean classification 15 

Comparison between zoologic and paleontologic species 18 

Proportions of the skull in bears and in titanotheres 19 

Features distinguishing phyla of titanotheres - 19 

Mutations of Waagen 19 

Zoologic and paleontologic nomenclature 20 

Summary of differences between old and new systems 22 

Study of the evolution of single characters 22 

Phylogeny of the nine typical families of the Perissodactyla 23 

Wide geographic distribution of the Perissodactyla 24 

Causes of evolution 27 

Adaptive evolution and overs volution of the form of skull, tooth, and foot 27 

Phyletic divergence in the evolution of new proportions in horses and in titanotheres 28 

Evolution of the limbs and feet of the titanotheres 33 

Origin of new characters as distinguished from changes in proportion 34 

Velocity in the development of characters and in phylogeny 39 

Summary of the evolution of the titanotheres 41 

Section 3. Bibliography of literature cited or consulted in the preparation of Chapter I 42 

Chapter II. Environment of the titanotheres and effect of adaptive radiation on their variation 43 

Section 1 . Geology and geography 43 

Correlation of early Tertiary events in the Rocky Mountain region with those in western Europe 43 

Late Cretaceous and early Tertiary climates 45 

Eocene geography of western North America and its relation to f aunal migrations 47 

Geographic divisions and their bearing on migration 47 

Character of the mountain-basin, plateau, and plains regions 51 

Eocene topography in the Rocky Mountain region 51 

Contrast in phj-siographic conditions east and west of the Rocky Mountain Front Range 53 

Lateral and main river systems in the mountain-basin region 54 

Section 2. Eocene and lower Oligocene formations and f aunal zones 56 

First f aunal phase (basal Eocene) 56 

Seventeen life zones 56 

Basal Eocene time in Montana and New Mexico 60 

Summary of faunal events of basal Eocene time 60 

Basal Eocene faunal zones 63 

Zones 1 and 2: Ectoconus and Polymastodon zones (Puerco fauna; part of Thanetian of Europe) 63 

Zones 3 and 4: Deltatherium and Pantolamhda zones (Torrejon and Fort Union faunas; part of Thanetian 

of Europe) 64 



IV CONTENTS 

Chaptek II — Continued. 

Section 2 — Continued. Page 

Second f aunal phase (lower Eocene) 64 

Transitional basal Eocene faunas 64 

Zone 5: Phenacodus-Nothodectes-Coryphodon zone (base of Wasatch formation of Big Horn Basin, first 

Wasatch Ufe zone, Big Horn A; Cernaysian of Europe) 64 

Early Eocene time 65 

Lower Eocene faunal zones 68 

Zone 6: Eohippus-Coryphodon zone (second Wasatch life zone, Big Horn B; lower Sparnacian of 

Europe) 68 

Zone 7: Systemodon-Coryphodon-Eohippus zone (third Wasatch life zone, Big Horn C; upper 

Sparnacian of Europe) 69 

Zone 8: Heptodon-Coryphodon-Eohippics zone (fourth Wasatch life zpne, Big Horn D and Wind 

River A; lower Ypresian of Europe) 69 

Zone 9: Lambdotherium-Eotitanops-Coryphodon zc*ne (fifth Wasatch life zone. Big Horn E, Wind 

River B, and Huerfano A; upper Ypresian of Europe) 69 

Transitional lower to middle Eocene deposits 74 

Huerfano formation of Colorado 74 

Wind River beds and their fauna 74 

Third faunal phase (middle and upper Eocene) 77 

Correlation of American zones with European stages 77 

Typical Bridger formation 78 

Zone 10: Eometarhinus-Trogosus-Palaeosyops fontinalis zone (Bridger A and Huerfano B; lower 

Lutetian of Europe) ^ 82 

Zone 11: Palaeosyops paludosus-Orohippus zone (Bridger B; upper Lutetian of Europe) 84 

Zone 12: Uintatherium-Manteoceras-Mesatirhinus zone (Bridger C and D, Washakie A, and Uinta 

A; part of Bartonian of Europe) 84 

Washakie Basin, Wyo 85 

Stratigraphy of the basin 85 

Zones 13 and 14: Metarlmius zone and Eobasileus-Dolichorhinus zone (Uinta B 1 and Washakie B 1; 

Uinta B 2) 89 

Uinta Basin, Utah 91 

Physiographic, climatic, and volcanic conditions in the Uinta Basin during middle (?) and later Eocene 

time 91 

Geologic horizons in the Uinta Basin 91 

Uinta B 1 (Metar/iinus zone = zone 13) 94 

Uinta B 2 {Eobasileus-Dolichorhinus zone = zone 14) 94 

Zone 15: Diplacodon-Protiianotherium-Epihippus zone (Uinta C 1; Ludian of Europe) 94 

Summary of faunas of Uinta B and C 97 

Adaptive radiation of the titanotheres in the Uinta Basin 97 

Genera and species represented 97 

Adaptive radiation of phyla 98 

Fauna unrepresented 99 

Zone 16: Theoretic Uinta C 2 99 

Composite Eocene and lower Oligocene section at Beaver Divide 99 

Fourth faunal phase (lower Oligocene) 101 

Lower Oligocene mammals 101 

Correlation of European and American forms 101 

Zone 17: Titanotheriuni-Mesohippus zone (Chadron A, B, and C; Sannoisian of Europe) 101 

Oligocene flood-plain sedimentation in the western Great Plains region 103 

Conditions of deposition 103 

South Dakota in Titanotherium time : 106 

Rapid fluviatile sedimentation in the Cypress Hills, Saskatchewan 109 

Slow sedimentation in South Dakota 109 

Geographic distribution of the Chadron formation 110 

Comparison of basins in western United States with the flood plain of the Nile 112 

Faunal divisions in the Chadron formation 113 

Three f aunistic levels determined 113 

Stratigraphic distribution of species of Oligocene titanotheres 113 

Hatcher's coUections, 1886-1888 115 

Sources of error in determining stratigraphic levels 116 

Mammalian life of the lower Oligocene Titanotherium zone 117 

Notes on the habitat of the fauna of the clay and sandstone as a whole 120 

Section 3. Adaptive radiation, primary and secondary, through change of environment a cause of diversification of the 

titanotheres 121 

Habitat of the ungulates 121 

Polyphyly among hoofed mammals ' ; 121 

The titanotheres and other extinct forms 121 

The existing African antelopes 124 



CONTENTS V 

Chapter II — Continued. Page 
Section 3 — Continued. 

Continental adaptive radiation of the African antelopes 125 

Adaptive radiation in the feeding habits of antelopes 126 

Causes of variation and polyphyly among quadrupeds 127 

Habits of the rhinoceroses parallel to those of the Oligocene titanotheres 128 

Habits of the existing tapirs parallel to those of the Eocene titanotheres 128 

Vertical geographic range of quadrupeds -- 129 

Vertical geographic range of the titanotheres 129 

Ten chief habitat zones of mammals _ 129 

Conclusions as to habitats of the titanotheres 132 

Section 4. Bibliography for Chapter II 132 

Chapter III. Discovery of the titanotheres and original descriptions of the types 141 

Section 1. History of discovery 141 

The Oligocene titanotheres 141 

The pioneer period: Prout, Owen, Evans, Leidy (1846-1873).-^ 141 

Taxonomic arrangement and comparison 144 

Work of Marsh and Cope (1870-1887) 144 

Summary of Marsh's contributions 145 

Summary of Cope's contributions 146 

Reinterpretation and phylogenetio study (Osborn, 1887-1919) 146 

Study of certain features 146 

Geologic levels and succession of t3'pes (Hatcher, 1886-1893) 147 

First European notice (Toula, 1892) 148 

Distinctions of sex (Osborn and Wortman, 1895) 148 

Monoph3'Ietic interpretation (Osborn, 1896) 148 

Polyphyletic interpretation (Osborn, 1902-1919) 148 

Recent discoveries by Lull, Lambe, and others 149 

The Eocene titanotheres 149 

Pioneer discoveries 149 

Work in the Bridger, Washakie, and Uinta Basins by Leidy, Marsh, Cope, Scott, Osborn, and others 

(1870-1889) 149 

Discovery in Hungary 150 

Princeton and Cope-Wortman expeditions 150 

First systematic and evolutionary revision (Earle, 1889-1891) 150 

American Museum and other explorations of the Eocene basins (1891-1895) 151 

Investigations and explorations made in preparation for the present monograph (1900-1919) 152 

Section 2. Original descriptions of types of Eocene titanotheres 153 

Five rules for determining the names of titanotheres 153 

The genera and species of Eocene titanotheres 155 

Descriptions of the species- 157 

Section 3. Original descriptions of types of Oligocene titanotheres 201 

List of genera and species 201 

Prout's descriptions of a fragmentary lower jaw, the first titano there made known to science 202 

Pomel's genus Menodus, based on Prout's description and figure 204 

Early notices by Leidy and others, 1850-1870 205 

Species described by Marsh and Cope in 1873-1876 209 

First notice of Canadian titanotheres by Cope, 1886 219 

Species described by Scott and Osborn in 1887 219 

Species described by Marsh in 1887 222 

Canadian species described by Cope in 1889 225 

Last five species described by Marsh, 1890-91 227 

Last species described by Cope, 1891 229 

First European Oligocene species, described by Toula, 1892 230 

Species described by Osborn in 1896 and 1902 231 

Species described by Lull in 1905 234 

Species described by Osborn in 1908 235 

Canadian species described by Lambe in 1908 235 

Second European Oligocene species, described by Kiernik, 1913 240 

Final Oligocene species described by Osborn in 1916-1919 241 

Chapter IV. Systematic classification of the titanotheres 243 

Section 1. Phyletic versus Linnaean S3'stem of classification 243 

Neo-Linnaean systematic divisions (zoologic) and evolutionary phyla (paleontologic) 243 

Superfamily names proposed by Osborn (1898) and Hay (1902) 243 

Family names proposed or adopted by Marsh (1873), Flower (1875), Cope (1879-1889), and Osborn (1889) 243 

Subfamily names and phsda proposed by Steinmann and Doderlein (1890), Earle (1892), and Riggs (1912) 245 

Division of the Oligocene titanotheres into four contemporary phyla, Osborn (1902) 245 

Reclassification of the Eocene and Oligocene subfamilies by Osborn (1914) 246 

Species wrongly referred to the titanotheres 246 

Section 2. Classification of the titanotheres adopted in this monograph 247 



VI CONTENTS 

Page 

Chapter V. Evolution of the skull and teeth of Eocene titanotheres 251 

Section 1. General principles of the study of the characters of the skuU and teeth 251 

Proportion characters and tendencies of evolution distinguished by analysis and synthesis 251 

Distinctions between proportion characters and new rectigradation characters 251 

Steps in transformation of characters 252 

Proportion and flexures of the skull 254 

Summary as to craniometrj' 255 

Changing proportions of the cranium and face 256 

Cyptocephaly , or f aciocranial flexure 256 

Dolichocephaly, brachycephaly, and correlation 257 

Zygomatic cephalic indices in the titanotheres and other perissodactyls 259 

Relative values of indices 259 

Indices of skulls of Eocene and Oligocene titanotheres 259 

Differences in terminology of skull proportions in titanotheres and in man 260 

Contrast in features of brachy cephaUc and dolichocephalic skulls and teeth 261 

List of abbreviations used in illustrations of skulls -262 

Terminology of the upper molar teeth 263 

Section 2. Introduction to the anatomy of the skuU and teeth of the Eocene titanotheres ._- 264 

Tj-pes of skull of Eocene titanotheres 264 

Feeding habits of broad-headed and long-headed titanotheres 264 

Origin and structure of the "horns" in titanotheres 266 

Proportion and rectigradation in the grinding teeth of Eocene titanotheres 267 

Mechanism of the titanothere grinding teeth 269 

Molarization of the premolars 270 

Correlation of dimensions of upper and lower teeth 272 

Geologic succession and geographic distribution of the Eocene titanotheres 272 

Section 3. The lower Eocene titanotheres 273 

Ancestral titanotheres of the Lambdolherium zone of Wyoming at the end of lower Eocene time 273 

Physiographic environment at the end of lower Eocene time 273 

Contrasts and resemblances between Lambdotherium and Eotitanops 276 

Explorations and discoveries 279 

Systematic descriptions of the lower Eocene titanotheres 279 

Section 4. The middle and upper Eocene titanotheres 297 

Phyla distinguished 297 

Species of Palaeosj'opinae and Dolichorhininae from the upper Huerfano {Trogosus zone) 297 

S3'stematic descriptions of the middle and upper Eocene titanotheres 297 

The palaeosy opine group 297 

The Manteoceras-DolichorJiinus group 357 

Successors to the Manteoceras-Dolichorhinus group 434 

Chapter VI. Evolution of the skull and dentition of OHgocene titanotheres 443 

Section 1. Review of the environment, geologic succession, and geographic distribution of the lower Oligocene titano- 
theres 443 

Section 2. Introduction to the anatomy of the skull and the dentition of the Oligocene titanotheres 444 

Horns : transformation, elongation 444 

Nasals : expansion, abbreviation 446 

Zygomatic arches: expansion, buccal plates 446 

Occipital pillars : auditory meatus 446 

Sexual characters common to all phyla 448 

Teeth : distinctive features and evolution 448 

Development of the skull and dentition 451 

Summary of the replacement of the teeth in OUgocene titanotheres 455 

Stages of wear of the adult grinding teeth 456 

Age and other characters common to both sexes of titanotheres of all stratigraphic levels 456 

Section 3. Division of the Oligocene titanotheres into groups and subfamilies 457 

Characters of the skuU and teeth of the menodontine and brontotheriine groups 457 

Characters and relations of the subfamilies 465 

Possible Eocene ancestors of the brontotheriine group — 468 

Section 4. Oligocene genera accepted as vahd in this monograph 469 

Section 5. The menodontine group 470 

Subfamily Brontopinae, including the phyla Manteoceras, Protilanotherium, Teleodus, Brontops, and Diploclonus... 470 

Stratigraphic level and distinguishing features 470 

Subfamily characters of Teleodus, Brontops, and Diploclonus 471 

Comparisons and contrasts 471 

Conspectus of characters of the subfamily ' 477 

Conspectus of characters of species 478 

Measurements of the Brontops series 479 

Systematic descriptions of genera and species in the Brontops-Diploclonus phylum ; 481 



CONTENTS VII 

Chaptjsr VI — Continued. 

Section 5 — Continued. Page 

Subfamily Menodontinae 505 

Systematic descriptions of genera and species in the Alloys phylum 506 

The Menodus monophjdum 518 

Systematic descriptions of genera and species in the Menodus phylum 522 

Section 6. The brontotheriine group 538 

Group characters 538 

Sexual characters 540 

Subfamily Megaceropinae 540 

Systematic descriptions of genera and species in the Megacerops phjdum 541 

Subfamily Brontotheriinae 550 

Systematic descriptions of genera and species in the Broniotherium phylum 555 

Chapter VII. Evolution of the skeleton of Eocene and Oligocene titanotheres 583 

Section 1. Methods by which the titanothere skeleton has been studied 583 

Principles of the evolution of the limbs of hoofed animals 583 

Size and proportions of Eocene titanotheres 584 

Divergence and convergence in the skeleton of polyphyletic series 586 

Diverse adaptive types of limb structure ^ 586 

Terms used in describing the skeleton of the titanotheres 588 

Section 2. The postcranial skeleton of lower Eocene titanotheres 590 

Subfamily Lambdotheriinae 590 

Subfamily Eotitanopinae 59 1 

Section 3. Middle Eocene groups and phyla J : 598 

Double parallelism in the palaeosyopine and Manteoceras-DoKchorhinus groups 598 

Family and subfamily characters of skeletal parts in middle Eocene titanotheres 599 

Systematic descriptions of middle Eocene titanotheres 612 

Subfamily Palaeosyopinae 612 

Subfamily Manteoceratinae 631 

Section 4. The postcranial skeleton of upper Eocene titanotheres 636 

Subfamily Dolichorhininae 636 

Subfamilies Telmatheriinae, Brontopinae?, and Diplacodontinae 652 

Section 5. The postcranial skeleton of Oligocene titanotheres 662 

Subfamily Brontopinae 664 

Subfamily Menodontinae 678 

Subfamily Brontotheriinae 689 

Bibliography for Chapters III- VII 698 



ILLUSTRATIONS 



Plate 
Frontispiece. Herd of Brontotherium platyceras. Page 

I. .4, Eruption of the crater of Taal, Philippine Islands; B, Flooded area 140 

II. A, Qjo Alamo, San Juan County, N. M.ex., looking north; B, Base of Puerco formation resting on eroded 

surface of Ojo Alamo sandstone 140 

III. A, Upper Torrejon beds, Torrejon Arroyo, Sandoval County, N. Mex.; B, Exposures of Puerco formation 

east of Ojo Alamo, N. Mex 140 

IV. A, Eohippus-Coryphodon zone. Little Sand Coulee, Clark Fork Basin, Wyo.; B, Phenacodus-Nothodecles- 

Coryphodon zone, Clark Fork Basin, north of Ralston, Wyo 140 

V. A, Typical "Lysite" locality, at Cottonwood Draw, north of Lost Cabin, Wyo.; B, Typical "Gray Bull" 

locality, south of Otto, Big Horn Basin, Wyo 140 

VI. A, A typical Huerfano locality, west of Gardner, Huerfano Basin, Colo.; B, A typical "Lost Cabin" locality, 

east of Lost Cabin, Wind River Basin, Wyo 140 

VII. A, Henrys Fork Table, looking northward across Henrys Fork, Bridger Basin, Wyo.; B, Grizzly Buttes, south 

of Mountain View, Uinta County, Wyo 140 

VIII. A, Northwest point of Haystack Mountain, head of Bitter Creek, Sweetwater County, Wyo.; B, View 

southeastward from Laclede station, Sweetwater County, Wyo 140 

IX. A, Columnar sandstones, LTinta A, White River Canyon, Uinta Basin, Utah; B, Panoramic view. White 

River Canyon below Wagonhound Bend, Uinta Basin, Utah 140 

X. A, Northern boundary of Coyote Basin, Uinta Basin, Utah; B, Divide between White River Canyon and 

Coyote Basin, Uinta Basin, Utah 140 

XL A, North face of Beaver Divide, Wind River Basin, Wyo.; B, Exposures at Wagonbed Spring, Beaver 

Divide, Fremont County, Wyo 140 

XII. A, Contact between Titanotherium zone and Pierre shale. Cedar Creek, Big Badlands, S. Dak.; B, Badlands 

south of White River, Utah, showing the Diplacodon zone 140 

XIII. " Mauvaises Terres" or Big Badlands of South Dakota 140 

XIV. Exposures at Quinn Draw, Big Badlands, S. Dak., showing summit of Chadron formation 140 

XV. A, South end of Sheep Mountain, near head of Corral Draw, Big Badlands, S. Dak.; B, Cedar Creek, Big 

Badlands, S. Dak., showing the Oreodon zone overlying the Titanotherium zone 140 



VIII CONTENTS 

Plate Page 

XVI. The region of the horn swelling in Palaeosyops, Manteoceras, and Telmatherium 266 

XVII. The region of the horn swelling in Manteoceras, Mesatirhinus, and Dolichorhinus 267 

XVIII. Restorations of the heads of four genera of Oligocene titanotheres 582 

XIX. Incisors and canines of Brontotherium and Teleodus 582 

XX. Upper and lower canines of Oligocene titanotheres 582 

XXI. Left upper premolars of Oligocene titanotheres 582 

XXII. Third left lower molar in Menodus and Brontotherium 582 

XXIII. Juvenile jaw referred by Marsh to Brontops 582 

XXIV. Juvenile jaws and teeth of Oligocene titanotheres 582 

XXV. Superior deciduous and permanent grinding teeth of Menodus giganteus 582 

XXVI. Type skeleton of Eotitanops princeps 702 

XXVII. Mounted skeleton of Palaeosyops leidyi 702 

XXVIII. Restoration of Palaeosyops of the Bridger Basin, Wyo 702 

XXIX. Restoration of Manteoceras and Dolichorhinus of the Uinta Basin, Utah 702 

XXX. Restoration of the skeleton of Dolichorhinus longiceps 702 

XXXI. Manus and pes of Dolichorhinus longiceps 702 

XXXII. Skeleton of Dolichorhinus longiceps 702 

XXXIII. Mounted skeletons of Brontops dispar and Brontops robustus (type) 702 

XXXIV. Mounted skeleton of Brontops robustus (type) , oblique front and side views 702 

XXXV. Mounted skeleton referred to Brontops robustus? 702 

XXXVI. Vertebral column of Brontop srobustus • 702 

XXXVII. Manus and hind limb of Diploclonus tyleri . 702 

XXXVIII. Mounted skeleton of Allops marshi L 702 

XXXIX. Mounted skeleton of Brontotherium hatcheri, left side view ___ 702 

XL. Mounted skeleton of Brontotherium hatcheri, right side view 702 

XLI. Mounted skeleton of Brontotherium hatcheri, front view 702 

XLII. Mounted skeleton of Brontotherium hatcheri, back view 702 

Figure 

1. "Fragment of the inferior maxillary of the left side" of Front's "gigantic Palaeotherium" 1 

2. Type of Palaeotherium? proutii 1 

3. Geologic ages and orogenic periods in North America 2 

4. Successive and overlapping Oligocene and early Eocene formations of the Rocky Mountains 3 

5. Map showing areas throughout the world in which remains of titanotheres have been found and areas in which titano- 

theres were probably in migration during Eocene and Oligocene time 4 

6. The Meek and Hayden Tertiary section of 1862 5 

7. Panoramic section of the Big Badlands of South Dakota, looking southeastward across Cheyenne and White Rivers 

to Porcupine Butte 6 

8. Map showing the type locality of the Titanotherium zone on Bear Creek, S. Dak 7 

9. Map showing cluster of typical lower, middle, and upper Eocene sedimentary basins in the Rocky Mountain region. _ 8 

10. Restorations of Eotitanops borealis and Brontotherium platyceras 10 

1 1 . Ambly poda : Skeletons and restorations of an ancestral and a specialized form 11 

12. Diagram showing the gradual extinction of archaic mammals and their replacement by modernized mammals 14 

13. Phenacodus and Coryphodon drawn to the same scale 15 

14. Contrast between the Linnaean and phylogenetic systems of classification 16 

15. The family tree of the titanotheres 17 

16. Theoretic descent of existing members of the dog family from a common ancestor 19 

17. Successive invasion of nine families of perissodactyls in North America and western Europe 23 

18. Outlines of the body form of the perissodactyls, drawn to the same scale 25 

19. The family tree of the Perissodactyla 26 

20. Periods of expansion and extinction of the perissodactyls and contemporary forms 27 

21. Phyletie divergence in the evolution of new proportions in horses and in titanotheres 29 

22. Contours of the head and of parts of the mouth in browsing and grazing perissodactyls 30 

23. Heads of lower Eocene and modern perissodactyls, showing changes of proportion and of the lip structure 31 

24. Restorations of the heads of some of the principal types of titanotheres 32 

25. Lower jaws of the first and the last of the titanotheres 33 

26. Structure of the feet in extinct and living odd-toed ungulates 34 

27. Restorations of nine species of titanotheres 35 

28. Evolution of the skeleton of the titanotheres 36 

29. Evolution of the skull and molar teeth in the titanotheres 37 

30. Adaptive radiation in the evolution of the upper molar teeth in the perissodactyls 38 

31. Three types of teeth of members of nine typical famihes of perissodactyls 39 

32. The family tree of the perissodactyls, showing adaptive radiation of the nine families and thirty-five subfamilies 40 

33. Outlines of the bodies of titanotheres at different stages of evolution 44 

34. Map showing the known areas and the hypothetical areas of titanothere migration and habitat 45 

35. General geologic sketch map of the Rocky Mountain region, showing existing topography and drainage areas and their 

relation to areas of Eocene and lower Oligocene sedimentation 46 

36. Map of western North America showing supposed routes of migration of animals 49 

37. Map showing the orogeny of the western mountain and plateau region 50 



CONTENTS IX 

Figure Page 

38. Geologic map of the Uinta Range, showing the Tertiary sediments of the Bridger Basin, Wye, at the north, and of the 

Uinta Basin, Utah, at the south 52 

39. Chronologic relations of formations in the mountain-basin region 54 

40. Section of deposits near Barrel Springs, Washakie Basin, Wyo 55 

41. Eocene and lower Oligocene mammalian life zones in eleven typical correlated areas in New Mexico, Colorado, Utah, 

Wyoming, South Dakota, and Montana 59 

42. Section of Upper Cretaceous and basal Eocene (Fort Union) deposits in Sweet Grass County, Mont 61 

43. Section of Eocene deposits in the San Juan Basin, N. Mex 62 

44. Columnar section of Cretaceous and Eocene sediments exposed along Bear River, Wyo., showing the typical Wasatch 

group of Hayden 66 

45. Generalized section through Upper Cretaceous and basal and lower Eocene deposits near Pumpkin Buttes, Powder 

River Valley, Wyo 68 

46. Composite section of the Eocene deposits of the Big Horn and Clark Fork Basins, Wyo 70 

47. A typical "Lost Cabin" locality. Alkali Creek, Wind River Basin, Wyo 71 

48. Section through the Wind River formation (lower Eocene) near Lost Cabin, Wyo 72 

49. Map showing cluster of lower, middle, and upper Eocene sedimentary basins in southwestern Wyoming and northern 

Utah, exhibiting parts of areas of the Wasatch, Wind River, Bridger, and Uinta formations 73 

50. Sketch map of the region of the Huerfano and Cuchara formations in southern Colorado 74 

51. Section of the Huerfano formation in southeastern Colorado 75 

52. Section of exposures from lower Eocene to lower Oligocene at Green Cove, on Beaver Divide, Wind River Basin, Wyo_- 76 

53. Section across Wind River Basin, Wyo., from Hudson to top of Beaver Divide 77 

54. Map showing the Eocene sediments encircling the Uinta Mountains of southwestern Wyoming and northern Utah 78 

55. Geologic section of the Bridger formation in the Bridger Basin, Wyo 80 

56. Map of the Bridger Basin, Wyo., and section of the Bridger formation 82 

57. Section of the lower part of the Bridger formation in the Bridger Basin, Wyo 83 

58. Section of the upper part of the Bridger formation in the Bridger Basin, Wyo 86 

59. Section of deposits near Barrel Springs, Washakie Basin, southern Wyoming 87 

60. Section of the Washakie Basin, Wyo., from north to south 88 

61. Sketch map of the Washakie Basin region, Wyo 88 

62. Columnar section of Washakie Basin, Wyo., showing life zones 90 

63. Section of the Uinta formation exposed in the north wall of White River Canyon, Utah 91 

64. Section of the Uinta formation from Kennedy's Basin to White River Canyon, Utah 1 92 

65. Section of the Eobasileus-Dolichorhinus and Metarhinus zones in the Uinta Basin, Utah 93 

66. Badlands near mouth of White River, Uinta Basin, Utah 95 

67. Section of deposits at Green Cove, Beaver Divide, Wyo 100 

68. Section across the Wind River Basin, Wyo., from Hudson to top of Beaver Divide 101 

69. Map showing exposures originally described as the "White River group" by Meek and Hayden 102 

70. Facsimile of the Meek and Hayden Tertiary section of 1862 103 

71. Map showing tributaries of Chej'enne River, S. Dak., and the type locality of the " Titanotherium beds" of Hayden. _ 104 

72. Type locality of the " Titanotherium beds" of Hayden, on Bear Creek, S. Dak 105 

73. Panoramic section of the Big Badlands of South Dakota 106 

74. Section of the Big Badlands of South Dakota, showing the chief faunal zones of the OUgocene (White River group) 

and the Miocene 107 

75. Map showing principal exposures of the Chadron formation 108 

76. Section showing the results of stratigraphic leveling in the Chadron formation (Titanotherium zone) in the badlands 

of White River, S. Dak ■ 115 

77. The family tree of the Perissodactyla 116 

78. Geographic cross section showing the nature of the habitats of the larger existing ungulates and of the titanotheres as 

illustrating adaptive radiation 122 

79. Original radiation of the unguligrade Herbivora, Carnivora, and Insectivora, with adaptations to environment 123 

80. Adaptations in the structure of the skull and teeth of Herbivora to diverse habits of feeding 125 

81. Convergent adaptations in the structure of the limbs and feet of ungulates 125 

82. Adaptive radiation in the feeding habits of antelopes 126 

83. Mauvaises Terres, Nebraska 142 

84. "Vertical view of the posterior tooth belonging to the lower jaw of Mr. Front's Palaeotherium" 143 

85. Original figures of Front's "gigantic Palaeotherium" 143 

86. Osborn's first restoration of Palaeosyops paludosus Leidy 151 

87. Four stages in the origin and evolution of the horns in titanotheres 152 

88. Leidy's cotypes of Palaeosyops paludosus 157 

89. Leidy's type (holotype) of Palaeosyops major 158 

90. Leidy's type of Palaeosyops humilis 159 

91. Leidy's cotypes of Palaeosyops Junius 159 

92. Marsh's type of Palaeosyops laticeps 160 

93. Marsh's type of Telmatherium validus 161 

94. Marsh's type of Limnohyus robustus 161 

95. Cope's cotypes of Palaeosyops vallidens 162 

96. Cope's cotypes of Limnohyops laevidens 163 

97. Cope's type (holotype) of Limnohyus fontinalis ■ — 164 



X CONTENTS 

Figure Page 

98. Cope's type (holotype) o{ Palaeosyops diaconus 1(55 

99. Marsh's type of Diylacndon elatus 166 

100. Type (holotype) lower jaw of Brachydiastematherium transilvanicum 167 

101. T3'pe (holot^'pe) oi Leurocephalus cultridens 168 

102. Type (holotype) of Palaeosyops borealis 168 

103. Type (holotype) of Lambdotherium popoagicum 169 

104. Cope's type of Lambdotherium brownianum 1 170 

105. Type (holotype) of Palaeosyops hyognathus 170 

106. Type (holotype) of skull of Palaeosyops megarhinus 171 

107. Earle's cotypes of Palaeosyops minor 172 

108. Earle's type of Palaeosyops longirostris 173 

109. Type (holotype) of Telmatotherium diploconum 173 

110. Type (holotype) of Telmatotherium cornutum 174 

111. Type (holotype) of Sphenocoelus uintensis 175 

112. T\'pe (holotype) of Diplacodon emarginatus 176 

113. Cotypes of Manleoceras manieoceras {Telmatotherium vallidens) 179' 

114. T^'pe (holotype) of Lambdotherium primaevum 180 

115. Type (holotype) of Limnohyops prisons 180 

116. Type (holotype) skull oi Limnohyops matthewi 180 

117. Type (holotype) skull oi Limnohyops 7nonoconus _' 180 

118. Type (holotype) skull of Palaeosyops leidyi 181 

119. Type (holotype) of Palaeosyops grangeri 181 

120. Type (holotype) of Palaeosyops copei 182 

121. Type (holotype) skull of Manteoceras washakiensis 182 

122. Type (holotype) skull of Mesatirhinus petersoni 183 

123. Type (holotype) skull of Metarhinus fltwiatilis 183 

124. Type (holotype) skull of Metarhinus earlei 183 

125. Type (holotype) skull oi Dolichorhinus intermedius 184 

126. Type (holotype) skull of Telmatherium ullimum 184 

127. Type (holotype) of Telmatherium? altidens 185 

128. Type (holotype) of Protitanotherium superbum 185 

129. Type (holotype) skull of Telmatherium? incisivum 186 

130. Type (holotype) of Telmatherium? incisivum -- 187 

131. Type (holotype) skull of Manieoceras uintensis -- 187 

132. Type (holotype) of Manteoceras uintensis 187 

133. Type (holotype) skull of Dolichorhinus heterodon 188 

134. Type (holotype) of Dolichorhinus heterodon 188 

135. Type (holot.ype) skull of Dolichorhinus longiceps 188 

136. Type (holotype) of Dolichorhinus longiceps 189 

137. Tj'pe (holotype) skull of Mesatirhinus superior 190 

138. Type (holotype) skull of Metarhinus riparius 191 

139. Type (holotype) skull of Metarhinus cristatus 191 

140. Type (holot3'pe) skull oi Dolichorhinus fluminalis . 192 

141. Type (holotype) skull oi Rhadinorhinus abbolti 193 

142. Type (holotype) teeth of Eotitanops gregoryi 193 

143. Lower jaws of Lambdotherium and Eotitanops 194 

144. Type (holotype) of Eotitanops princeps 195 

145. Type (holotype) of Eotitanops major 195 

146. Type (holotype) of Lambdotherium priscum 195 

147. Type (holotype) of Lambdotherium progressum 196 

148. Type of Diploceras oshorni 196 

149. Type of Diploceras osborni 197 

1.50. Type (holotype) skeleton of H eterotitanops parvus 198 

151. Type (holotj'pe) skull of H eterotitanops parvus 198 

152. Type (holotype) of H eterotitanops parvus 198 

153. Cotypes of Telmatherium? birmanicum : 198 

154. Type (holotype) of Lambdotherium magnum 199 

155. Type (holotype) of Eotitanops minimus 199 

156. Type (holotype) skull of Eometarhinus huerfanensis 200 

157. "Vertical view of the posterior tooth belonging to the lower jaw of Mr. Prout's Palaeotherium" 203 

158. Original figures of Prout's "gigantic Palaeotherium" 203 

159. Type of Menodus giganteus 204 

160. Owen's specimens of Palaeotherium? proutii ,-- 205 

161. Type (holotype) of Palaeotherium maximum 206 

162. Cotypes of Rhinoceros americanus 206 

163. Cotypes of Palaeotherium giganteum 207 

164. Type (holotype) of Megacerops coloradensis. 208 

165. Type (lectotype) of Brontotherium gigas 210 



CONTENTS XI 

Figure Page 

166. Type (lectotype) jaw of Symborodon ioruus ^ 211 

167. Type (holotype) skull of Megaceratops acer 212 

168. Type (holotype) skull of Megaceratops heloceras 213 

169. Type (lectotj'pe) skull of Symborodon bucco 214 

170. Type skulls of Symborodon altirostris, S. bucco, and Megaceratops acer 215 

171. Type (holotype) skull of Symborodon altirostris 216 

172. Type (holotype) skull of Symborodon trigonoceras 217 

173. Type (holotype) skull of Brontolherium ingens 218 

174. Type (lectotype) of Symborodon hypoceras 218 

175. Type (holotype) of Anisacodon montanus 219 

176. Cope's cotypes of Menodus angustigenis 220 

177. Anterior part of skulls of " Megacerops coloradensis," Menodus iichoceras, and Menodus dolichoceras 221 

178. Type (holotype) horns of Menodus platyceras 222 

179. Type (holotype) skeleton of Brontops robustus 222 

180. Type (holotype) lower jaw of Brontops dispar 223 

181. Type (holotype) skull of Menops varians 223 

182. Type (holotype) skull of Titanops curtus 224 

183. Type (holotype) skull of Titanops elatus 224 

184. Type (holotype) skull of Allops serotinus 225 

185. Type of Menodus selwynianus 225 

186. Type of Menodus syceras 226 

187. Type skull of Diploclonus amplus 227 

188. Type of Teleodus avus , 228 

189. Type skull of Allops crassicornis 229 

190. Type (holotype) skull of Brontops validus 230 

191. Type (holotype) skull of Titanops medius ■ 231 

192. Type (holotype) nasofrontal shield of Menodus peltoceras 232 

193. Cotypes of Menodus? rumelicus 232 

194. Type (holotype) skull of Titanotherium ramosum 232 

195. Type skull of Megacerops hrachycephalus 233 

196. Type (holotype) skull and lower jaw of Megacerops bicornutus 234 

197. Type skull of Megacerops marshi 234 

198. Type (holotype) skull of Brontolherium leidyi 235 

199. Upper premolars of type skull of Brontolherium leidyi 235 

200. Type (holotype) skull of Megacerops lyleri 236 

201. Right manus and right hind limb of the type of Megacerops iyleri 237 

202. Type (holotype) skull of Brontolherium haicheri 238 

203. Type (holotype) skull of Symborodon copei 238 

204. Type (holotype) jaw of Megacerops primilivus 239 

205. Type (holotype) jaw of Megacerops assiniboiensis 239 

206. Type of Titanotherium bohemicum 240 

207. Type (holotype) skull of Allops walcotli 241 

208. Type (holotype) jaw of Megacerops riggsi 242 

209. Characteristic basal sections of horns of Oligocene titanotheres 245 

210. Skulls showing different numerical and proportional characters in five separate phyla of titanotheres 253 

211. Standard measurements of Eocene titanothere skulls 255 

212. Unequal elongation of face and cranium in titanotheres and horses 256 

213. Faciocranial flexure, or cyptocephaly 256 

214. Faciocranial flexure in Patoeosyops and Dolichorhinus 256 

215. Cranial proportions of Eocene titanotheres — Palaeosyops, Manteoceras, and Dolichorhinus 257 

216. Cranial proportions in man and in the titanotheres 258 

217. Natural and artificial brachycephaly and dolichocephaly ■ 258 

218. Contra.sting forms of upper teeth in Eocene titanotheres 264 

219. Skulls of Eocene titanotheres of the principal genera 265 

220. Heads of Eocene titanotheres of four phyla 266 

221. Upper and lower molars of bunoselenodont pattern 268 

222. Upper and lower molar patterns of Hyracotherium 268 

223. Bunoselenodont patterns of upper and lower molars in Tertiary perissodactyls 268 

224. Relations of upper and lower molars in Telmatherium cultridens 269 

225. Dental mechanism of titanotheres 269 

226. Grinding teeth of a titanothere and an insectivore 270 

227. Contrast of braohyodont and semihypsodont molars in titanotheres 270 

228. Cross sections through second upper and lower molars of Lambdotherium and Menodus 270 

229. Upper premolar-molar teeth of the earliest and latest known titanotheres 271 

230. Reconstructed skeletons and restorations of Lambdotherium popoagicum and Eotitanops borealis 277 

231. Lower jaws of Lambdotherium, Eotitanops, and Tapirus 278 

232. Restored contours of skulls of Lambdotherium and Eotitanops . 278 

233. Skull of Lambdotherium popoagicum, reconstructed , 281 



XII CONTENTS 

Figure Page 

234. Lower premolars of three "species" or mutations of Lambdotherium 282 

235. Upper and lower grinding teeth of Lambdotherium 283 

236. Lower jaws and teeth of Lambdotherium popoagicum 284 

237. Lower jaws and teeth of Lambdotherium popoagicum, side view 285 

238. Front part of type lower jaw of Lambdotherium priscum 286 

239. Incomplete lower jaw of Lambdotherium priscum 286 

240. Jaws and teeth of Lambdotherium priscum and L. magnum 287 

241. Lower jaw and teeth of Lambdotherium progressum • 288 

242. Upper teeth of Lambdotherium progressum 288 

243. Restoration of Eotitanops borealis 289 

244. Skulls of the oldest known titanotheres, Lambdotherium popoagicum and Eotitanops borealis 290 

245. Model of skull of Eotitanops gregoryi 291 

246. Lower premolars and molars of Eotitanops 291 

247. Lower jaws of Eotitanops gregoryi and E. brownianus 292 

248. Lower jaw of Eotitanops borealis 293 

249. Lower teeth of Eotitanops borealis 294 

250. Skull of Eotitanops borealis, palatal and side views 294 

251. Skull of Eotitanops borealis, top and occipital views 294 

252. Lower jaw of Eotitanops princeps 296 

253. Lower grinding teeth of three species of Eotitanops from the upper Huerfano formation 296 

254. Skull sections of brachycephalic and dolichocephalic Eocene titanotheres 299 

255. Cross sections of the skull m middle Eocene titanotheres 300 

256. Three skulls typical of the palaeosyopine group 301 

257. Distribution of Palaeosyops and associated fauna in the Bridger formation 301 

258. Anterior part of skull of Limnohyops laevidens 306 

259. Skull of Limnohyops priscus 307 

260. Back part of skull of Limnohyops priscus 308 

261. SkuUs of three species of Limnohyops 309 

262. SkuU of Limnohyops matthemi 309 

263. Skull of Limnohyops monoconus 310 

264. Skull of Limnohyops laticeps 311 

265. Third right upper molar of Limnohyops laticeps 311 

266. Lower jaws of Limnohyops and Palaeosyops 314 

267. Lower jaws of Palaeosyops 314 

268. Lower jaws of three species of Palaeosyops 316 

269. Young skull of Palaeosyops fontinalis 317 

270. Upper molars of Palaeosyops fontinalis 318 

271. Teeth of Palaeosyops fontinalis 1 318 

272. Skull of Palaeosyops major 319 

273. SkuU and head of Palaeosyops leidyi 324 

274. Incisors and canines ot Limnohyops a.nd Palaeosyops 325 

275. Skull of Palaeosyops leidyi 326 

276. Type skull of Palaeosyops leidyi 327 

277. Type skull of Palaeosyops leidyi, top view 328 

278. Type skull of Palaeosyops leidyi, palatal view 328 

279. Skulls of Palaeosyops major and P. leidyi 329 

280. Lower jaws of Palaeosyops leidyi : 330 

281. Skulls of Palaeosyops leidyi and P. copei? (aff. P. robustus) 331 

282. Jaws and deciduous teeth of Palaeosyops leidyi? 332 

283. Deciduous cheek teeth of Palaeosyops leidyi? 332 

284. Fragments of jaws of Palaeosyops 333 

285. Skull of Palaeosyops robustus 333 

286. Hyperbrachycephalic old male skull of Palaeosyops robustus 334 

287. Basicranial region of Palaeosyops robustus 334 

288. Nasals of Palaeosyops robustus 335 

289. Progressive hypsodonty of the molars in Telmatherium 341 

290. Upper jaw of Telmatherium cuUridens 342 

291. Upper and lower teeth of Telmatherium cultridens 343 

292. Upper and lower teeth of Telmatherium cultridens, interlocked 343 

293. Lower jaw of Telmatherium cultridens 344 

294. Type skull and lower jaw of Telmatherium ultimum 346 

295. Type skull of Telmatherium ultimum, side, front, and occipital views 347 

296. Type skull of Telmatherium ultimum, palatal and top views 348 

297. Paratype skull of Telmatherium ultimum 349 

298. Lower jaw of Telmatherium ultimum 350 

299. Hypothetical reconstruction of the skull of Telmatherium altidens 352 

300. Lower jaws of Telmatherium ultimum and T. altidens 353 

301. Type skull of Sthenodectes incisivus . — 356 



CONTENTS XIII 

Figure Page 

302. Skulls of titanotheres of the Manteoceras-DoKchorhinus group 359 

303. Skulls of Manteoceras manieoceras 363 

304. Type skull of Manieoceras manteoceras 366 

305. Skulls of Manteoceras manteoceras and Palaeosyops leidyi 367 

306. Skulls of Manteoceras manteoceras and M. washakiensis 367 

307. Skull of Manteoceras manteoceras, side view 368 

308. Skull of Manteoceras manteoceras, anterior half 368 

309. Incisors and canines of Manteoceras manteoceras 369 

310. Lower jaw of Manteoceras 370 

311. Skulls of Manteoceras manteoceras and M. washakiensis 371 

312. Type skull of Manteoceras uintensis •- 373 

313. Upper canines and incisors of Manteoceras uintensis 374 

314. Restoration of Protitanotherium emarginatum 374 

315. Lower jaws of Protitanotherium and Brachydiastematherium 375 

316. Type skull of Protitanotherium emarginatum; reconstruction, side view 376 

317. Type skull of Protitanotherium emarginatum, front and side views 376 

318. Nasal region in three specimens of Protitanotherium 377 

319. Sections of the nasals and horns of Protitanotherium emarginatum 377 

320. Lower jaw of Protitanotherium emarginatum 378 

321. Lower jaw of Protitanotherium superbum 381 

322. Phylogenetic relations of the species of Metarhinus, Mesatirhinus, Dolichorhinus, and Rhadinorhinus 383 

323. Top view of the skull in the Manteoceras-Dolichorhinus group 385 

324. Palatal view of the skull in the Manteoceras-Dolichorhinus group 385 

325. Leidy's cotypes of Palaeosyops Junius 386 

326. Type skull of Mesatirhinus megarhinus 389 

327. Type skull of Mesatirhinus petersoni 390 

328. Skull of Mesatirhinus petersoni, side, top, and palatal views 391 

329. Skulls of Mesatirhinus petersoni, front and occipital views 392 

330. Incisors, canines, and premaxillae of Mesatirhinus 392 

331 . Lower jaws of Mesatirhinus 394 

332. Lower jaw of Mesatirhinus sp. with deciduous dentition 395 

333. Imperfect cranium of Mesatirhinus petersoni? 396 

334. Geologic section of the Bridger formation in the Washakie Basin 397 

335. Restoration of Dolichorhinus longiceps ■ 398 

336. Skull and lower jaw of Dolichorhinus hyognathus 398 

337. Skulls of Dolichorhinus hyognathus and modern horse 399 

338. Geologic section of the Eobasileus-Dolichorhinus and Metarhinus zones in the Uinta Basin 400 

339. Skulls showing progressive dolichoceplialy in the Mesatirhinus-Dolichorhinus phylum, side view 401 

340. Skulls showing progressive dolichocephaly in the Mesatirhinus-Dolichorhinus phylum, top and palatal views 402 

341. Upper premolars of Mesatirhinus, Dolichorhinus, and Metarhinus 403 

342. Skull of Dolichorhinus intermedins 406 

343. Skulls of Dolichorhinus intermedins, D. heterodon, and D. longiceps 408 

344. Skull referred to Dolichorhinus longiceps? 409 

345. Hyoid apparatus of Dolichorhinus longiceps? compared with that of a modern tapir 410 

346. Skulls of Dolichorhinus 411 

347. Skull of Dolichorhinus hyognathus, palatal view 412 

348. Skulls of Dolichorhinus hyognathus, front and occipital views 413 

349. Skull of Dolichorhinus hyognathus, side view 413 

350. Upper incisors and canines of Dolichorhinus hyognathus : 414 

351. Lower incisors and canines of Dolichorhinus hyognathus 414 

352. Left upper canine of Dolichorhinus hyognathus 414 

353. Lower jaws of Dolichorhinus 415 

354. Skull of Sphenocoelus uintensis 418 

355. Type skull of Eometarhinus huerfanensis 419 

356. Skull of Metarhinus fluviatilis 423 

357. Right lower premolars of Metarhinus fluviatilis 424 

358. Lower jaws of Metarhinus 425 

359. liOwer jaw o{ Metarhinus? (Rhadinorhinus?) sp 426 

360. Skull and deciduous teeth of type of lieterotitanops parvus 426 

361. Type skull of Metarhinus earlei ; 427 

362. Type skull of Rhadinorhinus diploconus, side and top views 432 

363. Type skull of Rhadinorhinus diploconus, top and palatal views 433 

364. Type skull of Rhadinorhinus diploconus, side, front, and occipital views 434 

365. Skulls of Eotitanotherium osborni , 436 

366. Nasals and horn swellings of Eotitanotherium osborni 437 

367. Two upper raolara of Eotitanotherium (" Diploceras") osborni 438 

368. Type skull of Diplacodon elatus, partial reconstruction, palatal view 439 

369. Type skull of Diplacodon elatus, upper jaw and zygoma 440 



XIV CONTENTS 

Figure Page 

370. Third and fourth upper premolars of Diplacodon elatus 440 

371. Upper molars of Diplacodon and Proiitanotherium compared 441 

372. Facial region of Eotitanoiherium osborni and Bronlotherium leidyi 441 

373. Map showing areas in which remains of titanotheres have been found 443 

374. Comparison of upper Eocene and lower OUgocene titanotheres 444 

375. Sections at base of horn in the six chief generic types of Oligocene titanotheres 445 

376. Position of the standard sections and contours of Oligocene titanotheres skulls 445 

377. Male and female skulls of Bronlotherium gigas 446 

378. Occipital view of skulls in different phyla of OHgocene titanotheres 447 

379. Influence of progressive brachycephaly on the auditory region of perissodactyls 447 

380. Inferior aspect of chin in Manleoceras • 449 

381. Upper molars of Menodus giganteus and Allops marshi 450 

382. Extreme dolichocephaUc and brachycephahc types of upper premolar-molar series in Oligocene titanotheres 450 

383. Third left lower molar of Bronlotherium leidyi 451 

384. Development of jaws and teeth, stage 4 452 

385. Development of jaws and teeth, stage 6 453 

386. Occiput of young skull of Brontops? brachycephalus 454 

387. Stages of wear in the adult upper grinding teeth of Ohgooene titanotheres 455 

388. Skull contours showing extreme divergence between Menodus giganteus and Bronlotherium platyceras 456 

389. Skulls of the menodontine group, side view 459 

390. Skulls of the bronototheriine group, side view 460 

391. Skulls of the menodontine group, top view 461 

392. Skulls of the brontotheriine group, top view 462 

393. Skulls of the menodontine and brontotheriine groups, palatal view 463 

394. Skulls of the menodontine and brontotheriine groups, front view 464 

395. Lower jaws of the Bronlotherium phylum 465 

396. Lower jaws of the Brontops and Menodus phyla 466 

397. Lower jaws of the Diplodonus and Allops phyla 466 

398. Heads of Oligocene titanotheres 466 

399. Sections at base of horn in five principal lower Ohgocene phyla of titanotheres 468 

400. Restorations of lower OUgocene titanotheres of the four principal genera 469 

401. Skulls of Rhadinorhinus and Bronlotherium, palatal view 470 

402. Skulls of Rhadinorhinus and Bronlotherium, side view 471 

403. Skulls of Rhadinorhinus and Bronlotherium, top view 472 

404. Lower jaws of Metarhinus fluvialilis and Bronlotherium halcheri 473 

405. Progressive evolution of the upper premolars in Bronlotherium. and its predecessors 474 

406. Progressive evolution of the upper premolars in Menodus and Brontops and their predecessors 474 

407. Progressive evolution of the lower premolars in Bronlotherium and its predecessors 475 

408. Progressive evolution of the lower premolars in Brontops and its predecessors 475 

409. Phyla of the Brontopinae and Menodontinae 477 

410. Evolution of the horns in the Brontops phylum 477 

411. Basal section of the horns in the Brontops phylum 477 

412. Progressive broadening of the nasals in the Brontops phylum 478 

413. Lower jaws of Teleodus primitivus, Brontops brachycephalus, and Allops walcotti? _ 478 

414. Sections and contours of skulls of Brontops brachycephalus 483 

415. Upper canines and incisors of Brontops brachycephalus 484 

416. Reconstruction of crushed skull of Brontops brachycephalus 485 

417. Left upper grinding teeth of Brontops brachycephalus 486 

418. Skull and horn region of Brontops brachycephalus? 486 

419. Sections and contours of skulls of Brontops brachycephalus and B. dispar 487 

420. Restoration of Brontops robuslus 492 

421. Sections and contours of skull of Brontops robuslus 493 

422. Skull of Brontops robuslus 495 

423. Lovi'er jaws of Brontops dispar and B. robuslus 496 

424. Sections and contours of skull of Brontops sp 497 

425. Hyoid bones of Brontops compared with those of the tapir, black rhinoceros, and horse 497 

426. Sections and contours of skulls of Diplodonus hicornutus and D. tyleri 498 

427. Sections and contours of skull of Diplodonus ampins 499 

428. Lower jaws of Diplodonus bicornutus and D. tyleri 503 

429. Sections and contours of skulls of Allops walcotti and A. marshi 610 

430. Upper teeth of Allops walcotti 511 

431. Skull of Allops marshi 513 

432. Lower jaws of Allops marshi and Allops? sp 514 

433. Sections and contours of skulls of Allops serotinus and A. crassicornis 516 

434. Coossified nasals and proximal part of horns of Allops? serotinus? 517 

435. Sections and contours of skulls of Menodus heloceras and M. trigonoceras 519 

436. Skull of Menodus heloceras 526 

437. Lower jaws of Menodus (Symhorodon) lorvus and M. trigonoceras 527 



CONTENTS XV 

Figure Page 

438. Upper teeth of Menodus ■proutii 528 

439. Skull of Menodus trigonoceras 629 

440. Skull of Menodus trigonoceras belonging with the mounted skeleton in the Colorado Museum, Denver 530 

441. Restoration of Menodus giganteus 531 

442. Skull of Menodus giganteus, front view 532 

443. Skull of Menodus giganteus, palatal view 532 

444. Sections and contours of skulls of Menodus giganteus and M. various 534 

445. Sections and contours of skull of Menodus giganteus 535 

446. Lower jaws of Menodus giganteus 536 

447. Teeth and nasals of Menodus montanus i 538 

448. Sections and contours of nasals and horns of Megacerops coloradensis 544 

449. Sections and contours of skull of Megacerops bucco 545 

450. Sections and contours of skulls of Megacerops copei and M. acer 546 

451. Upper part of occiput of Megacerops acer 547 

452. Skull of Megacerops acer, side and top views 548 

453. Skull of Megacerops acer, palatal view 549 

454. Restoration of Megacerops copei 549 

455. Lower jaws of Megacerops assiniboiensis and M. riggsi 550 

456. Sections and contours of skull of Megacerops? syceras 550 

457. Composite sections showing the evolution of the horns and reduction of the free nasals in the Brontotherium phylum. _ 551 

458. Basal sections of the horns in the Brontotherium phylum 552 

459. Skulls of male and female brontotheres 552 

460. Contrast in contour of horns and nasals between male and female brontotheres 552 

461. Sections and contours of skulls of Brontotherium leidyi and B. hypoceras 558 

462. Lower jaws of Brontotherium leidyi ' 559 

463. Two lower molars and symphyseal region of Brontotherium? rumelicum 560 

464. Lower jaws of Brontotherium rumelicum? and B. leidyi 561 

465. Sections and contours of skull of Brontotherium? hatcheri 563 

466. Skull of Brontotherium hatcheri, side view j 564 

467. Skull of Brontotherium hatcheri, front view 564 

468. Lower jaws of Brontotherium hatcheri and B. gigas 566 

469. Sections and contours of skull of Brontotherium? tichoceras 567 

470. Sections and contours of skulls of Brontotherium hatcheri and B. gigas 568 

471. Lower jaws of Brontotherium gigas and B. medium 569 

472. Sections and contours of skull of Brontotherium gigas? 572 

473. Sections and contours of skull of Brontotherium dolichoceras 572 

474. Skull of Brontotherium dolichoceras 572 

475. Sections and contours of skulls of Brontotherium medium and B. curtum 674 

476. Horns of Brontotherium curtum 576 

477. Sections and contours of skull of Brontotherium curtum 576 

478. Left horn and nasals of Brontotherium curtum? 577 

479. Sections and contours of skull of Brontotherium ramosum 578 

480. Restoration of Brontotherium platyceras 579 

481. Sections and contours of skull of Brontotherium platyceras 580 

482. Evolution of the skeleton in titanotheres 584 

483. Estimated height at shoulder of Eocene and Oligocene titanotheres and tapir 585 

484. The phyla of Eocene titanotheres, as represented by the manus 587 

485. Progressive broadening of the magnum in Eocene titanotheres 587 

486. Reconstructed skeleton and restoration of Lambdotherium popoagicum 591 

487. Atlas and scapula of Lambdotherium popoagicum 591 

488. Fore limb of Lambdotherium popoagicum 592 

489. Forearm and manus of Lambdotherium popoagicum 592 

490. Left manus of Lambdotherium and Eotitanops 592 

491. Astragalus of Lambdotherium popoagicum 593 

492. Restorations of Lambdotherium popoagicum, Eotitanops princeps, and E. gregoryi 593 

493. Metatarsals of Eotitanops 593 

494. Reconstructed skeleton and restoration of Eotitanops borealis 594 

495. Atlas of Eotitanops borealis 595 

496. Vertebrae of Eotitanops princeps 595 

497. Radius of Eotitanops borealis 595 

498. Lunars of Eotitanops 595 

499. Manus of Eotitanops princeps 595 

500. Humerus and femur of Eotitanops princeps 596 

501. Pelvis of Eotitanops borealis -• 596 

502. Left pes of cursorial and subeursorial Eocene Perissodactyla 597 

503. Astragalus and calcaneum of cursorial and submediportal Eocene Perissodactyla 598 

504. Astragalocalcaneal facets in lower Eocene Perissodactyla , 598 

505. Left astragalus and calcaneum of Eotitanops sp 599 



XVI CONTENTS 

Figure Page 

506. Metatarsal and tibia of Eotitanops major 599 

507. Restoration of Eoiitmiops horealis 600 

608. Atlas of Eocene titanotheres 601 

509. Types of scapula in middle Eocene titanotheres 602 

510. Tj'pes of fore limb in Eocene and Oligocene titanotheres 603 

511. Characteristic details of radius and ulna in middle and upper Eocene titanotheres 604 

512. Manus of lower and middle Eocene titanotheres 605 

513. Comparison of the riglit scaphoid in middle Eocene titanotheres 605 

514. Terminal phalanges of the manus in middle Eocene titanotheres and amyuodonts 605 

515. Progressive graviportal adaptation in the pelvis of Eocene and Oligocene titanotheres ■ 606 

516. Femora and tibiae of middle Eocene titanotheres 609 

517. Distal end of the femur in Manteoceras and Amynodon 610 

518. Angulation of the knee joint: relation of patellar facet to long axis of femur * 611 

519. Inner side view of left fibula of Palaeosyops, Limnohyops, and Brontotherium 611 

520. Comparison of pes in four species of middle Eocene titanotheres 613 

521. Astragali of Eocene titanotheres 614 

622. Calcanea of Eocene titanotheres 615 

523. Left ectocuneiform tarsi of lower and middle Eocene titanotheres j 615 

524. Principal measurements of the carpus and tarsus 615 

525. Humerus, radius, and ulna of Limnohyops monoconus? 615 

526. Left manus, radius, and ulna of Mesatirhinus petersoni 616 

527. Manus, radius, and ulna of Limnohyops monoconus 616 

528. Right scaphoid of Palaeosyops sp. and Limnohyops monoconus 617 

529. Left hind limb of Limnohyops monoconus 618 

530. Right pes of Limnohyops monoconus? 618 

631. Ventral surface of sacrum of Limnohyops laticeps 618 

632. Right OS innominatum of Limnohyops laticeps 619 

533. Pelvis of Palaeosyops major 619 

534. Right femur and tibia of Palaeosyops major 620 

635. Astragalus and calcaneum of Palaeosyops major 620 

536. Composite mounted skeleton of Palaeosyops leidyi 621 

537. Manus of Palaeosyops leidyi 622 

538. Pelvis of Limnohyops 624 

539. Pelvis of Palaeosyops cf. P. leidyi 624 

640. Left pes of Palaeosyops leidyi 626 

641. Relations of facets of the astragalus and calcaneum in Palaeosyops 626 

542. Atlas of Palaeosyops robustus 627 

543. Atlas and axis of Palaeosyops leidyi? 627 

544. Cervicals and dorsals of Palaeosyops robustus 627 

645. Left scapula of Palaeosyops robustus 627 

546. Bones of forearm of Palaeosyops 628 

647. Left astragalus of Palaeosyops copei? 629 

648. Fore limb of Palaeosyops copei? 629 

649. Left manus of Palaeosyops copei? 629 

550. Right hind limbs of Palaeosyops major and P. copei? 630 

551. Atlas of Manteoceras manteoceras 632 

562. Seventh cervical vertebra of Manteoceras manteoceras compared with that of Palaeosyops leidyi 633 

553. Left humerus of Manteoceras manteoceras 633 

554. Right manus of Manteoceras manteoceras 633 

655. Pelvis of Manteoceras? 1 634 

556. Femora and tibiae of Manteoceras manteoceras 635 

557. Left astragalus of Manteoceras manteoceras 635 

558. Restoration of the skeleton of Mesatirhinus petersoni 637 

559. Restorations of Mesatirhinus petersoni and Palaeosyops leidyi 637 

660. Atlas of Mesatirhinus megarhinus * 638 

561. Humerus of Mesatirhinus megarhinus 638 

562. Radius and ulna of Mesatirhinus petersoni 638 

563. Left forearm and manus of Mesatirhinus petersoni? 639 

564. Right manus and fragments of radius and ulna of Mesatirhinus petersoni 639 

565. Left manus, radius, and ulna of Mesatirhinus petersoni? 639 

566. Right scaphoid of Mesatirhinus and Manteoceras 639 

567. Right manus of Mesatirhinus petersoni? 641 

568. Left femur and tibia of Mesatirhinus petersoni? 642 

569. Left pes of Mesatirhinus petersoni? 642 

570. Left astragali of Mesatirhinus petersoni? 642 

571. Left entocuneiform tarsi of Palaeosyops and Mesatirhinus 642 

572. Pes referred to Mesatirhinus 644 

573 Pes of Meiarhinus cf. M. earlei 644 



CONTENTS XVII 

Figure i'ags 

574. Astragalus, calcaneum, and navicular of Metarhinus cf. M. earlei 644 

575. Astragalus of Metarhinus cf. M. earlei 644 

576. Left scapula of Metarhinus? sp 645 

577. Left radius and ulna of Metarhinus earlei'? 645 

578. Skeleton of a newly born animal, provisionally identified as Metarhinus sp 646 

579. Provisional restoration of the skeleton of Dolichorhinus hyognathus 646 

580. Vertebral column of Dolichorhinus hyognathus 647 

581. Atlas referred to Dolichorhinus sp 647 

582. Left scapula of Dolichorhinus? hyognathus 649 

583. Humerus of Dolichorhinus hyognathus 649 

584. Radius and ulna of Dolichorhinus hyognathus 649 

585. Metatarsals of Dolichorhinus hyognathus 649 

586. Manus of AmynodoJi and Mesatirhinus compared 650 

587. Left fore limb of the amphibious rhinoceros Amynodon intermediusf 650 

588. Left astragali of Dolichorhinus and allied types 651 

589. Cervical vertebrae of Dolichorhinus longiceps? 651 

590. Right fore limb of Dolichorhinus longiceps? 652 

591. Manus of Dolichorhinus longiceps? 652 

592. Hind limb referred to Telmatherium ultimum 653 

593. Pes of Tehnatheriumf ultimum? 653 

594. Vertebrae and fore limb of Diplacodon or Protitanotherium 654 

595. Astragalus and calcaneum of Diplacodon or Protitanotherium 655 

596. Left astragalus of Protitanotherium superbu7n 655 

597. Incomplete ilium and ischium of Diplacodon elatus 656 

598. Atlas and axis of Eotitanotherium osborni : 656 

599. Vertebrae of Eotitanotherium osborni 657 

600. Scapula of Eotitanotherium osborni 657 

601. Humerus, radius, and ulna of Eotitanotherium osborni , 657 

602. Manus of Eotitanotherium osborni 658 

603. Femur, tibia, and pelvis of Eotitanotherium osborni 658 

604. Pes of Eotitanotherium osborni 658 

605. Restoration of skeleton of Eotitanotherium osborni 659 

606. Mounted skeleton of Brontops ^ 670 

607. Three views of mounted skeleton of Brontops -. 671 

608. Scapulae of Oligocene titanotheres 673 

609. Manus of Brontops? sp. and B. dispar? 674 

610. Mounted skeleton of Brontops brachyeephalus? 676 

611. Mounted skeleton of Brontops brachyeephalus?, oblique front view 677 

612. Parts of skeleton of Allops crassicornis? 680 

613. Pes of Menodus trigonoceras, referred, and M. heloceras 681 

614. Manus of Menodus trigonoceras? 682 

615. Restorations of Menodus trigonoceras and Allops marshi 683 

616. Mounted skeletons of Brontops dispar? and Menodus trigonoceras. 684 

617. Left astragalus of Menodus giganteus - 685 

618. Cervical and first four dorsal vertebrae of Brontops robustus and Menodus giganteus 686 

619. Manus referred to Menodus giganteus 687 

620. Restorations of Brontotherium leidyi and B. platyceras 688 

621. Atlas and axis of Brontotherium leidyi 689 

622. Vertebrae of Brontops robustus and Brontotherium gigas 689 

623. Scapulae of Oligocene titanotheres 690 

624. Humeri of Brontops robustus and Brontotherium leidyi 691 

625. Humeri of Megacerops? acer? and Brontotherium gigas? 691 

626. Radii of Brontops robustus, Brontotherium leidyi, and Brontotherium gigas 691 

627. Radius and ulna of Brontotherium 692 

628. Ulnae of Brontops robustus, Brontotherium leidyi, and Brontotherium gigas 692 

629. Olecrana of Brontotherium and Megacerops? 692 

630. Manus of Oligocene titanotheres . 693 

631. Manus and pes referred to Brontotherium gigas hatcheri 694 

632. Manus and pes referred to Brontotherium hatcheri? 695 

633. Manus referred to Brontotherium gigas, as restored 695 

634. Pelvis of Brontotherium gigas hatcheri 696 

635. Femora of Brontops robustus and Brontotherium leidyi 696 

636. Tibiae of Brontops robustus and Brontotherium leidyi 696 

637. Tibia and fibula of Brontotherium leidyi 696 

638. Femora of Megacerops? and Brontotherium? 696 

639. Pes of Oligocene titanotheres 697 

101959— 29— VOL 1 2 



LETTER OF TRANSMITTAL 



Dr. George Otis Smith, 

Director United States Geological Survey, 

Washington, D. C. 

Dear Sir: I have the honor to transmit herewith 
a monograph on the evolution of a pecuharly American 
family of quadrupeds known as the titanotheres. 
This designation was given to them in 1852 by Joseph 
Leidy while he was employed as vertebrate paleon- 
tologist in David Dale Owen's survey of a part of 
the ancient territory of Nebraska. This family is 
one of a group of vertebrate animals whose fossil 
remains, found in the western United States, were 
long studied by Othniel Charles Marsh, my distin- 
guished predecessor in this work in the United States 
Geological Survey. Early in the eighties Professor 
Marsh projected a monograph on the Brontotheridae 
(here called the titanotheres), and subsequently he 
made the largest and most valuable contributions to 
our knowledge of this family and of its evolution. He 
planned the monumental field work of John Bell 
Hatcher, by which the great collection for the United 
States National Museum was made, and he super- 
vised the preparation of sixty lithographic plates, 
which are here reproduced. Unfortunately he died 
before he had even begun to prepare the manuscript. 
The duty of continuing his work was intrusted to me 
June 30, 1900, by your predecessor, Charles D. Wal- 
cott. During this period of nearly 20 years I have 
supervised the preparation of the monograph on the 
Ceratopsia by Hatcher and Lull and have half com- 
pleted the monograph on the Sauropoda. The mono- 
graph on the Stegosauria has not yet been prepared. 

The task of preparing the present monograph has 
been long and difficult. First, it proved necessary 
to reexplore the entire Eocene and lower Oligocene 
series of rocks in Wyoming, Colorado, and South 
Dakota, where the fossilized remains of titanotheres 
are found, both to determine precisely their geologic 
succession and to close up gaps in the stages of evolu- 
tion; second, it proved necessary to examine and com- 
pare the titanotheres of these geologic epochs in all 
the museums of this country and in several museums 
abroad; third, it pi-oved necessary, in order thoroughly 
to understand the titanotheres, to discover and to 
follow many side lines of investigation that have not 
hitherto been followed in vertebrate paleontology. 

This work has been done with the aid of many 
specialists, foremost among whom is my junior col- 
league Prof. William K. Gregory, without whose in- 



telligent and unremitting cooperation the monograph 
could never have been completed. 

It is perhaps not too much to say that this work 
has transformed our knowledge of the early Tertiary 
geology of the Rocky Mountain basin region. First, 
the six life periods recognized by Marsh and his no 
less distinguished contemporary Edward Drinker Cope 
may now be replaced by sixteen life periods, which may 
be clearly defined and separated and certain of which 
may be more or less precisely correlated with life 
periods established for western Europe. Second, a 
much clearer notion has been gained of the changing 
geographic, physiographic, climatic, and volcanic con- 
ditions in Wyoming and Dakota and of their influence 
on the migration and succession of forms of life. 
Third, the whole method of attack on problems of 
vertebrate paleontology has been developed; we seek 
to know the entire living animal, its musculature, its 
mode of locomotion, and its feeding habits, in order 
to insure the complete restoration of the body. Fourth, 
the study of the many branches of this group has given 
the most convincing demonstration that evolution, 
even in any one geographic region, seldom moves along 
a single line of descent ; more frequently it moves along 
many lines — it is polyphyletic; in other words, it 
radiates, following the principles of local adaptive 
radiation. Finally, the history of the titanothere 
family in its evolution from very small and relatively 
weak forms into titanic quadrupeds, second in size 
only to the elephants, has afforded us a unique oppor- 
tunity to enlarge our previous knowledge of the actual 
modes of evolution as well as to revise our theories as 
to the causes of evolution and of extinction. 

I desire to express my appreciation of the support 
given by the Geological Survey under your direction 
in the completion and publication of this work. 

With the aid of many coworkers I have endeavored 
to set a new standard of broad, thorough, and ex- 
haustive research in vertebrate paleontology which 
shall be worthy of the great geologic traditions of our 
national Geological Survey. I trust that this mono- 
graph, like Leidy's classic memoir of 1869, may ex- 
ercise a permanent influence upon future studies of 
the geologic history of the great West. 

Henry Fairfield Osborn, 

Vertebrate Paleontologist. 

American Museum of Natural History, 
December 19, 1919. 



PREFACE 



VERTEBRATE PALEONTOLOGY IN THE NATIONAL 
SURVEYS 

Joseph Leidy, Edward Drinker Cope, and Othniel 
Charles Marsh, who successively served as members of 
United States Government surveys of the West, were 
the founders of American vertebrate paleontology. 
Leidy's memoir of 1869, entitled "The extinct mam- 
malian fauna of Dakota and Nebraska, including an 
account of some allied forms from other localities, 
together with a synopsis of the mammalian remains of 
North America," marked the end of the first period of 
exploration. Cope's great memoir of 1885, entitled 
"The Vertebrata of the Tertiary formations of the 
West," marked the end of the second period of explor- 
ation. 

Meanwhile the subject had become too broad to be 
comprehended in a single work. Accordingly Marsh, 
as vertebrate paleontologist, planned a series of ex- 
haustive monographs on special groups of extinct 
birds, reptiles, and mammals, which should treat in 
great detail the anatomical structure and form the 
basis of a systematic classification. For these mono- 
graphs he carried out the most intensive field explora- 
tions known to science and published a large number of 
preliminary papers, which fairly revolutionized our 
knowledge of these and many other groups. In 1880 
the Fortieth Parallel Survey published his monograph 
on the Odontornithes, an extinct group of birds of North 
America. In 1883 the United States Geological Sur- 
vey published his paper entitled "Birds with teeth," 
and in 1886 his monograph on the Dinocerata, an 
extinct order of gigantic mammals. This was the 
first of the series of five monographs projected for pub- 
lication by the United States Geological Survey on 
the Dinocerata, the Stegosauria, the Sauropoda, the 
Ceratopsia, the Brontotheridae. The monograph last 
indicated has developed into the present monograph 
on the titanotheres, which covers a much broader field 
than that contemplated by Marsh for the monograph 
on the Bronototheridae. 

For the monographs on the Ceratopsia and on the 
Brontotheridae exploration on an unprecedented scale 
was begun by the United States Geological Survey 
under the direction of Marsh. For the four mono- 
graphs on the Stegosauria, Sauropoda, Ceratopsia, and 
Brontotheridae 204 superb lithographic plates were 
completed under Marsh's direction. Altogether he 
had been engaged on this work nearly 17 years when 
death interrupted his monumental labors on March 
18, 1899. 



PREPARATION OF THE PRESENT MONOGRAPH 

The first important step taken by Marsh in his series 
of contributions to our knowledge of this extinct fam- 
ily was the publication of his paper on "The structure 
and affinities of the Brontotheridae," published in 
1874, based on the collections at Yale University. 
The second was his paper entitled "Principal charac- 
ters of the Brontotheridae," published in 1876. In 
the meantime he had made a geologic excursion to 
White River in South Dakota, in the vicinity of the 
Red Cloud Agency. This visit marks an interesting 
epoch in the history of paleontologic exploration for 
the titanotheres. 

Late in the autumn of 1875 Marsh, accompanied by 
an escort from Fort Laramie to the Red Cloud Agency, 
went to the Badlands of Nebraska and Dakota. The 
consent of the Indians was deemed necessary to permit 
safe search for fossil bones in their country. This con- 
sent was obtained with difficulty, and after it had been 
obtained the Indians withheld their assistance. An 
account of Marsh's visit is given in a manuscript en- 
titled "Sketches of the life of Red Cloud," by Capt. 
James H. Cook, of Agate, Nebr., at that time serving as 
a scout for the United States Army. Captain Cook 
writes: 

It was in the autumn of 1875 that I visited the Red Cloud 
Agency, which was at that time located on the White River, in 
the northwestern part of Nebraska, the agency buildings stand- 
ing about 2 miles up the river from the place where the city of 
Crawford is now situated. The chief of the Sioux, Red Cloud, 
made me welcome to his lodge. 

It was on this visit that I first learned of the petrified bones 
of strange creatures that had once occupied the lands to the 
eastward of the agency. Two of Red Cloud's subchiefs, 
American Horse and Little Wound, took me to the lodge of 
Afraid of Horses, where I was shown a piece of bone, perfectly 
petrified, containing a molar tooth .3 inches or more in diameter. 
American Horse explained that the tooth had belonged to a 
"Thunder Horse" that had lived "away back" and that then 
this creature would sometimes come down to earth in thunder- 
storms and chase and kill buffalo. 

His old people told stories of how on one occasion man_v, 
many j^ears back, this big Thunder Horse had driven a herd 
of buffalo right into a camp of Lacota people during a bad 
thunderstorm, when these people were about to starve, and 
that they had killed many of these buffalo with their lances 
and arrows. The "Great Spirit" had sent the Thunder Horse 
to help them get food when it was needed most badly. This 
story was handed down from the time when the Indians had 
no horses; 

While I was the guest of Red Cloud on this occasion, Prof. 
O. C. Marsh, of the Smithsonian Institution and Yale Uni- 
versity, came over from Fort Laramie to Camp Robinson and 
the Red Cloud Agency to get permission to collect fossils in 

XXI 



XXII 



the Sioux country. The Sioux, however, did not take kindly 
to this proposition, thinking it was yellow lead (gold) that the 
white chief wanted, not stone bones. 

I met Professor Marsh at that time and talked with him. 
I showed him the tooth the Indians had shown me. When I 
returned to Red Cloud's lodge I told Red Cloud that Professor 
Marsh was a friend of the "Great Father" (the President) at 
Washington, and that I thought if he were allowed to hunt for 
stone bones in the Sioux country he would be a good friend 
to the Sioux people. Red Cloud said that if Professor Marsh 
was a good man he would help the Sioux people to get rid of 
the agent that was then in charge of the agency, whom the.y 
disliked very much. This being brought to the attention of 
Professor Marsh, he took the matter in hand, and an investi- 
gation of affairs took place at the Red Cloud Agency, the re- 
sult of which was at least pleasing to the Indians concerned, 
as the agent was removed. 

Professor Marsh was allowed to collect with his field parties 
unmolested from that time on. He was named by Red Cloud 
"Wicasa Pahi Hohu" (pronounced we-ch5-shJl pa-he ho-hii), 
Man-that-Pioks-Up-Bones. The professor and Red Cloud 
became friends to the extent that Red Cloud was entertained 
at the home of the professor in New Haven, Conn., and the 
two were photographed there with clasped hands and the 
"peace pipe" between them. 

The first collections made for this monograph were 
those brought together from Colorado and South 
Dakota, part of them under the direction of Marsh, 
for the Peabody Museum of Yale University. By far 
the greatest collection was that brought together by 
John Bell Hatcher for the Geological Survey, now 
preserved in the United States National Museum. 
Between 1870 and 1891 Marsh published 14 papers 
on these collections. These papers relate more or 
less directly to the Brontotheridae; the last appeared 
in 1891 and contained descriptions of three new types 
from South Dakota — AUops crassicornis, Brontops 
dispar, and Brontotherium medium. 

WORK BY THE AUTHOR, 1878-1919 

In the meantime the present author made his first 
contribution to the history of this family in 1878 in 
a paper on the results of the Princeton collections of 
1877 and 1878 in the Bridger Basin. His second 
contribution was made in 1887 in a paper entitled 
"Preliminary report on the vertebrate fossils of the 
Uinta formation collected by the Princeton expedition 
of 1886." His third and fourth contributions were 
made in 1890, in the two papers entitled, respectively, 
"Preliminary account of the fossil mammals from the 
White River and Loup Fork formations," which 
related to a collection made in South Dakota by Dr. 
S. Garman for the Harvard University Museum, and 
"The MammaUa of the Uinta formation," Parts III 
and IV, on the Perissodactyla. These have been 
followed by 38 papers by the author, based chiefly 
on his paleontologic and geologic expeditions in the 
field for the American Museum of Natural History, 
planned by the author and ably directed by Dr. J. L. 
Wortman, Mr. O. A. Peterson, and Mr. Walter 
Granger. To these indefatigable field explorers science 
is indebted for the wonderful series of Eocene titano- 



theres which have enabled us to trace the ancestry 
of the Oligocene titanotheres and to establish all the 
early phases in the history of this family. To Peter- 
son, Earl Douglass, and Elmer S. Riggs in the Uinta, 
and especially to Granger in the entire series from the 
basal Eocene to the base of the Uinta, is due the 
remarkable precision of the geologic records by which 
the faunal life zones of the Eocene have been deter- 
mined. 

The stratigraphic succession of the Eocene and of 
the lower Oligocene mammal life has been determined 
chiefly by the field observations and collections of 
Granger in the Eocene and of John Bell Hatcher in 
the lower Oligocene and by the systematic examina- 
tions of species by Dr. William Diller Matthew and 
by the author. 

RESEARCH AND COLLABORATION 

Prof. William K. Gregory has been in the closest 
cooperation with the author in all the details of the 
preparation of the monograph since the beginning of 
the work in the year 1900. Words are inadequate 
to express the author's sense of indebtedness to his 
former student and present colleague in the American 
Museum and in Columbia University. 

The author desires also to acknowledge his special 
indebtedness to Mr. Granger for his valuable notes 
and his cooperation in the preparation of the text 
and the geologic sections, as presented in Chapter II, 
on the Eocene and Oligocene formations of the Rocky 
Mountains, as well as to Prof. William J. Sinclair 
for his work on the volcanic nature of the middle 
Eocene deposits and to Mr. Albert Johannsen of the 
Geological Survey for his analyses of the material 
of these deposits. It is hoped that that chapter will 
furnish a key to future exploration of this mountain- 
basin region as well as to the Oligocene sections of the 
Great Plains. Matthew, by means of the rich col- 
lections in the American Museum, has furnished 
critical determinations for the discrimination of mam- 
malian species in the sixteen life zones and has cooper- 
ated with the author in the preparation of "Cenozoic 
mammal horizons of western North America," pub- 
lished by the Geological Survey in 1909 as its Bulletin 
361, which forms the foundation of the more de- 
tailed life-zone work whose results are presented in 
Chapter II. 

Details of the history of the collections at home and 
abroad are presented in Chapter III under the head- 
ing "History of explorations and discoveries and 
original descriptions of the Eocene and Oligocene 
titanotheres." Every known significant specimen 
is referred to, its species and its sex are determined, 
and its principal characters are described. This 
monograph will furnish a much desired key to the 
present and future collections and surveys in Wyo- 
ming, Nebraska, Colorado, the Dakotas, and Assin- 
iboia. 



XXIII 



COOPERATION OF MUSEUMS 

To the museums of the United States, Great 
Britain, and Bavaria, where titanothere remains are 
preserved, the author is indebted for cordial coopera- 
tion in furnishing materials for study and in affording 
every possible facility for measurements and illustra- 
tions. The author would mention especially Prof. 
Charles Schuchert and Prof. Richard S. Lull, of the Yale 
University Museum, present custodians of the great 
Marsh collections, as well as their assistant, Mr. 
Thomas A. Bostwick, who is in charge of all the field 
records of Marsh. In connection with the superb 
Hatcher collection in the United States National 
Museum, which far surpasses any other in existence. 
Dr. Charles W. Gilmore and Dr. James W. Gidley 
have rendered every possible assistance. The author 
is especially indebted to the director of the Carnegie 
Museum at Pittsburgh, Dr. W. J. Holland, and to 
Mr. O. A. Peterson of that museum for the liberal 
use of collections of the Uinta titanotheres; also to 
Mr. Earl Douglass of the same institution for his 
invaluable field notes and observations on the Uinta 
section. The systematic location of species in the 
great Uinta section is due to the precise field work of 
Mr. Elmer S. Riggs of the Field Museum of Natural 
History, Chicago, an institution that is especially 
rich in remains from the horizon known as Uinta B. 
To his former colleague Prof. William B. Scott of 
Princeton University, as well as to his colleague 
Prof. William J. Sinclair, the author is indebted for 
the liberal use of valuable collections, including many 
types from several levels of the Bridger and from the 
uppermost Eocene horizon, known as Uinta C. 

From 1846, when the earliest remains of titanotheres 
were found, until 1918 almost every year has added 
one or more stages or types to the history of this 
great family. The stages still to be discovered are 
in the unknown interval between the uppermost 
Eocene horizon, known as Uinta C, and the lowermost 
Oligocene horizon, known as Chadron A. 

WORK ON TEXT AND ILLUSTRATIONS 

The great task of preparing the bulk of the manu- 
script — a task performed between 1902 and 1918 — 
fell upon Miss H. Ernestine Ripley, the work being 
done chiefly from the dictation and notes of thp 
author. The preparation of the bibliography and 
the first revision of the entire manuscript were also 
undertaken by Miss Ripley with interest and per- 
formed with precision. The author warmly appreci- 
ates this invaluable service to paleontology. The 
final arrangement of the illustrations together with 
the preparation of the accompanying legends, was 
undertaken, under the author's general direction, by 
Doctor Gregory with the cooperation of Miss Chris- 
tine D. Matthew, Mr. Erwin S. Christman, and Mrs. 
Lindsey Morris Sterling. The preliminary editorial 



work has been performed with celerity and skill by 
Miss Mabel Rice Percy, of the American Museum. 
The final arrangement and verification of illustra- 
tions and captions were the work of Miss Christine 
D. Matthew. 

The final editorial work and preparation of the text 
for the printers were accomplished by Messrs. George 
M. Wood and Bernard H. Lane, Mr. Wood continuing 
the work as a member of the American Museum staff 
after his retirement from the Geological Survey. 

The illustrations, which are taken from many 
sources, date back to the early lithographic figures of 
Leidy. They include the unpublished lithographic 
plates prepared under the direction of Cope, and 
especially the superb lithographic drawings made 
for the United States Geological Survey by Mr. F. 
Berger under Marsh's direction. These lithographic 
plates are supplemented by numerous plates based 
upon photographs taken chiefly by Mr. A. E. Ander- 
son of the American Museum staff. 

The text and plates are adorned with reproductions 
of the fine series of drawings from the pen and brush 
of Mr. Christman and from the numerous pen draw- 
ings of Mrs. Sterling. The geologic sections in 
Chapter II are the work of Mr. William E. Belanske. 
To Mrs. Sterling, Mr. Christman, and Mr. C. A. Weck- 
erly of the Geological Survey were assigned the final 
preparation for the photoengraver of all the illustra- 
tions for the monograph, which, including those in the 
Appendix, consist of 797 figures and 236 plates. 

This review affords a partial explanation of the 
great length of the period of time occupied by the 
author in the preparation of this monograph. The 
work has involved repeated explorations in the West 
in search of the remains of all the ancestors of the 
family and in establishing the full chronology. It has 
necessitated repeated journeys to all the museums of 
the country and long and painstaking research. The 
greatest effort, however, has been expended on the 
solution of the series of related problems in stratig- 
raphy, in adaptation, in the origin of new characters, 
in the mechanics of locomotion, in the modes of 
evolution, and in the causes of evolution and of 
extinction that presented themselves as essential to 
the exposition of the life history of a long extinct 
family. To restore the living and the lifeless environ- 
ment of the Rocky Mountain region and to present 
the titanotheres as living forms has been the persistent 
purpose of this monograph. 

SUMMARY OF GEOLOGIC AND ANATOMIC 
PRINCIPLES 

The following is a brief statement of the principles 
developed and discriminated in this monograph : 

1. The principle of the division and correlation of geologic 
formations in Eocene and lower Oligocene time by mamma- 
lian life zones and bv the subdivision of these zones. 



XXIV 



2. The principle of the correlation of local physiographic 
diversity with the adaptive radiation, local and continental, 
of titanotheres and other ungulates. 

3. The principle of adaptive radiation as expressed in adap- 
tations to aquatic, forest, savanna, and plains life at different 
altitudes. 

4. The principle of multiple lines of descent in the same 
regions, of polyphyly and of polyphyletic evolution as more 
common among ungulates than monophyletic evolution. 

5. The principle of distinguishing phyla by contrasting pro- 
portions of the head (dolichocephaly and brachycephaly) , of 
the Umbs (dohchomely and brachj'mely) , of the feet (dolicho- 
pody and brachypody), and of the teeth (hypsodonty and 
brachyodonty) . 



6. The principles of the lengthening and shortening of the 
limb segments in harmony, respectively, with adaptation to 
speed and to weight. 

7. The principles of evolution by rectigradation (origination 
of new characters) and by allometry (changes of proportion) 
as effecting the chief changes in the hard parts. 

8. The principles of continuity and of orthogenesis — the 
direct continuation of animal form estabUshed in adaptation 
to environment and of the evolution of new types irrespective 
of external influences. ' 

The theoretic causes underlying these principles of 
evolution are briefly stated in Chapter I, and the con- 
clusions reached are summarized in Chapter XI. 



THE TITANOTHERES OF ANCIENT WYOMING, DAKOTA, 

AND NEBRASKA 



By Henry Fairfield Osborn 



CHAPTER I 

INTRODUCTION TO MAMMALIAN PALEONTOLOGY 



SECTION 1. EXPLORATION AND RESEARCH MADE IN 
THE PREPARATION OF THIS MONOGRAPH 

The preparation of this monograph was actually 
begun in 1846, when a part of a jawbone of a titano- 
there was found in the region now known as South 
Dakota and sent first to Dr. Hiram A. Prout of St. 
Louis and then to Dr. 
Joseph Leidy of Phila- 
delphia for description. 
This bit of bone gave the 
first hint to science of 
the wonderful deposits 
of vertebrate fossils in 
the Rocky Mountain 
region that have revo- 
lutionized vertebrate 
paleontology. The de- 
tails of this epoch-mak- 
ing discovery are given 
in Chapter III. The 
original fragment bears 
the generic name Meno- 
dus, which was assigned 
to it by the keen system- 
atic paleontologist of 

France, Nicolas Auguste Pomel,who gave it the specific 
name giganteus. Menodus giganteus is thus the first 
titanothere known to science, and it is a representative 
of the most imposing family of mammals that was 
evolved in ancient North America. 




Figure 1. — "Fragment of the inferior maxillary of the left side' 
Front's "gigantic Palaeotherium," the first titanothere discovered 
After Prout (1847). One-fourth natural size. 



of America — Joseph Leidy, Edward Drinker Cope, 
Othniel Charles Marsh, John Bell Hatcher — up to 
the time when the whole long and difficult study of 
family history, of geologic succession, and of environ- 
ment was intrusted to the present author. 

From the first it seemed desirable that this study 
should encompass more 
than a dry, systematic 
description — that these 
animals and their envi- 
ronment should, so far 
as possible through pale- 
ontology, be made to 
live again as the domi- 
nant animals of a long 
and very interesting 
epoch in the history of 
North America — the 
first third of the Terti- 
ary period. The field 
explorations made in 
the prosecution of this 
research should, more- 
over, sustain the guiding 
principles of the union 
of paleontology and geology established by the pioneers 
of our national surveys, as seen especially in the com- 
bined work of the geologist, Frederick V. Hayden, and 
the paleontologists, Charles A.White and Joseph Leidy,^ 
whose reports are still fundamental standards of Terti- 




FiGURE 2. — Tj'pe of Palaeotherium ? proutii 
Owen's specimen, Nat. Mus. 113. After Leidy (1852). One-third natural size. This was one of the specimens referred to by 
Leidy (1852.1)1 in proposing the name Titanolherium. 



This family, from its earliest known beginnings in the 
Wind River Mountains of the present State of Wyoming 
to the height of its development on the plains of the 
ancient Dakota- Nebraska -Colorado region, attracted 
the attention of the leading vertebrate paleontologists 



1 The figures in parentheses refer to entries i 
chapter. 



, the bibliography at the end of this 



ary geologic and paleontologic history. Subsequent 
works have surpassed these in specialization and in 
number and variety of animal forms, and the geologic 
areas and life zones have been greatly increased by 
subsequent discovery, but none have surpassed them 

' See reports of Hayden and White (1867-73.1, 1868.1), based on surveys begin- 
ning in 1802, and Leidy's great memoh (1869.1). 



TITANOTHERES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



in scientific method — in the constant union of paleon- 
tologic with geologic evidence in the reconstruction of 
the slow succession of events in the wonderful history 
of this western resion. 




ROCKY 
MOUNTAIN 



LARAMIDE 



SIERRA NEVADA 



APPALACHIAN 



7/ , , , , ■ ,, 

/ r^y /// /// ,,. 

/PENNSYLVANjAN 

/ ' MISSfeSIPPlKN^ / 
.''//•> / ^, / /'/// 




? PRE-CAMBRIAN 



? ARCHEAN 



FiGtTRE 3. — Geologic ages and orogenic 
periods in North America 

Age of mammals, stipple; age of reptiles, vertical lines; 
age of amphibians and fishes, oblique lines. The peri- 
ods of the birth and elevation of the chief American 
mountain systems, notably the Rocky Mountains 
(including the Laramide revolution), are indicated 
approximately by incisions on the right. Modified 
from diagram by Henry Shaler Williams. 

The present monograph is made up of this introduc- 
tory chapter and of ten other chapters, covering the 
following six main lines of exploration and research 



that have been followed in order to restore, at least 
in part, the life and times of the titanotheres : 

1. Geologic, physiographic, climatic, and faunal 
environmental conditions of the titanothere epoch — • 
the Eocene and lower Oligocene divisions of the 
Tertiary. Principles of adaptive radiation in animals 
as explaining the variation of the titanotheres. 
(Chap. II.) 

2. History of the discoveries of the remains of 
titanotheres, the original published descriptions, and 
the previous and present classification of genera and 
species. (Chaps. Ill and IV.) 

3. Systematic study of the titanotheres: Eocene 
and lower Oligocene subfamilies, genera, and species. 
Characters of the skull, dentition, and postcranial 
skeleton. (Chaps. V, VI, and VII.) 

4. Muscular anatomy of the titanotheres: Princi- 
ples of locomotion and evolution of limb structure in 
the hoofed mammals (Ungulata) in relation to habits. 
(Chaps. VIII and IX.) 

5. Origin, ancestry, and adaptive radiations of the 
titanotheres and other odd-toed ungulates. (Chap. X.) 

6. Evolution and extinction of the titanotheres: 
Evidence regarding modes and causes of evolutionary 
development and decline in mammals. (Chap. XI.) 

SECTION 2. PRELIMINARY SURVEY OF THE MONO- 
GRAPH AND OF THE CONCLUSIONS PRESENTED 

RANGE OF THE TITANOTHERES IN GEOLOGIC TIME 

Geographic distribution. — The earliest known titano- 
theres lived near the end of early Eocene time, after 
the appearance in the Rocky Mountain region of 
three kinds of quadrupeds — the horses, the related 
forest-living tapirs, and the more remotely related 
rhinoceroses, which still exist elsewhere. 

The successive immigrations of related odd-toed 
ungulates are recorded in the Eocene deposits of the 
region now included in the State of Wyoming, which 
during Eocene time was a fertile land inhabited by 
an abundant fauna. The Eocene titanothere epoch 
in northern Utah, south of the great Uinta Mountain 
range, which, according to Powell, rose to majestic 
heights, ended in late Eocene time. 

In lower Oligocene time the titanotheres had 
seemingly become the largest mammals in North 
America. They were second in size to the existing 
elephants only, but recent paleontologic evidence 
indicates the existence in Oligocene time in India of 
mammals that exceeded in size both the titanotheres 
and the elephants. In 1913 Mr. C. Forster-Cooper 
(1913. 1) described a new genus of perissodactyls from 
the upper Oligocene deposits of the Bugti Hills of 
Baluchistan, BalucTiiiherium (Thaumastotherium) os- 
borni, an animal of proportions so gigantic that it 
dwarfs the largest known titanothere. 

Sedimentary divisions and faunal life zones. — The 
lower Eocene to lower Oligocene sediments in which 
titanothere remains have been found occur here and 



INTRODUCTION TO MAMMALIAN PALEONTOLOGY 



there in several of the ancient river drainage basins 
of Wyoming. While the remains of the animals and 
plants of the period were accumulating in these sedi- 
ments the titanotheres and other herbivorous quadru- 
peds and the carnivores that preyed upon them, as 
well as the other mammals and invertebrates of the 
land, of the water, and of the air, were constantly 
evolving, appearing and disappearing through mi- 
gration and extinction. Thus where the sediments 



of Front's "gigantic PalaeotJierium" (Menodus gigan- 
teus) in 1846 to the present time, it has been found 
that the lower division of this zone is distinguished by 
the presence of 85 species of vertebrates. The names 
of the dominant form or forms of each zone are used 
to designate the several life zones. For the designa- 
tion of the Titanotherium zone the name of this single 
genus Titanotherium (Menodus) is used, for it is the 
most distinctive form in that zone. 




swEETGRAss co.jNTRODUCTION OF ANCESTORS 

(MONT.) 

2 ^I'aunal Period 



g::=a ARCHAIC MAMMALS 
ONLY 




1-1/ ^^Faunal Period 



CorypTiodon - Ayriblypods 

and. 
Eohippjis -^vrstSorses 



W$m wm MAMMALS 




Pan.toZcmibda -^-mhlypods 
Polynhastodan. 



Figure 4. — Successive and overlapping Oligocene and early Eocene formations of the Rocky Mountains 
The duration of the titanothere epoch is indicated by the arrow. 



are very rich in fossils of all kinds — mammals, reptiles, 
iishes, and rarely birds — we are able to restore the 
life that was distinctive of certain more or less con- 
tinuous phases of geologic sedimentation. These 
time divisions are designated life zones, as distin- 
guished from the sedimentary divisions of groups and 
formations. 

After an exploration of the Titanotherium zone that 
covered a period of over 70 years, from the discovery 



Many genera persist through several successive life 
zones. Two genera, the large-hoofed Coryphodon and 
the small primitive horse EoMppus, persist through 
four lower Eocene geologic phases or life zones, during 
which a succession of other species, as well as migra- 
tions, extinctions, etc., may be clearly observed. It 
may therefore be necessary to select more than one 
genus, perhaps as many as three genera, in order to 
define clearly a certain life zone. For example, the 



TITANOTHEHES OP ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



amblypod Coryphodon, the horse EoMppus, the tapir 
Systemodon unite to define the Systemodon-Coryphodon- 
EoMppus life zone of the lower Eocene. 

It is through these zonal resemblances in the 
mammalian life, and more rarely in the plant life, 
that relatively sure estimates are made of the time 
during which the sediments containing certain fossils 
were deposited, irrespective of such geologic data as 
whether the sediments are thick or thin, whether they 
are products of erosion or of volcanic eruption, whether 
they were deposited in still water or in rapidly moving 
water, or whether they are composed of clay, sand, 
gravel, conglomerate, or tuff. The life zone, when 
adequately defined, is an absolutely reliable means of 
time correlation as distinguished from other means — 
physiographic, geologic, or lithologic. 

Similar sediments. — It is true that in the Rocky 
Mountain region there prevailed at times over wide 



mentation in one region (for example, the Cypress 
Hills, Saskatchewan) and with excessively slow sedi- 
mentation on river flood plains in another region (Chey- 
enne and White Rivers, S. Dak.), or with a fall of 
volcanic ash in still another region (Beaver Divide, 
Wyo.). 

Evolution of mammals a stable process. — ^By com- 
paring all the events in the history of the American 
continent for which the records afforded by geology 
and paleontology harmonize with others afforded by 
paleontology alone we reach the conclusion that one of 
the most uniform, the most stable geographically, and 
the most widespread is the evolution of mammalian life. 
This evolution proceeds more or less uniformly in 
Europe, in Asia, and in North and South America. The 
apparently sensitive protoplasm (body substance) and 
germ plasm (hereditary substance) are far more stable 
and far more uniform in their progressive evolution 




Former land orea^ Former migration areas Known fossil areas 

Figure 5. — Map showing areas throughout the world in which remains of titanotheres have been 
found (solid black) and areas in which titanotheres were probably in migration during Eocene 
and Oligocene time (oblique lines) 

Titanotheres have been found in the northwestern United States, the Gobi Desert (Mongolia), Burma, and southeastern 

Europe. 



areas similar physiographic, climatic, and eruptive 
volcanic conditions, as, for example, during what we 
designate Fort Union time, Wasatch time, upper 
Bridger time. During such periods of uniform con- 
ditions the geologic evidence is concordant or harmoni- 
ous with the paleontologic evidence afforded by life 
zones, and doubtless any paleobotanic evidence that 
may be found must also be concordant. In basal 
Eocene (Fort Union) time, for instance, the forests, 
the mammals, the reptiles, the climate, the physiogra- 
phy of the chief areas of sedimentation of the whole 
Rocky Mountain region were all more or less similar, 
and in this particular epoch these several means of time 
correlation afford more or less harmonious evidence. 

Unlilce sediments. — Such similar sediments, however, 
become increasingly rare in the continental deposits 
of Eocene and Oligocene time. A single life zone, 
such as the Titanotherium zone, may be contempo- 
raneous with violent fluviatile action and heavy sedi- 



than the surface of the earth. For this reason they 
form superior data for time correlation. This is one 
of the chief generalizations that have grown out of the 
long series of observations and studies of the correla- 
tion of Tertiary geologic events in America and Europe 
that were specially made in the preparation of this 
monograph. 

Life zones of the titanothere epoch. — By the method 
of determining geologic time by discriminating life 
zones the whole epoch of the evolution of the titano- 
theres has been subdivided into titanothere zones, 
distinguished not only by successive genera and species 
of titanotheres but by corresponding changes in all the 
environmental forms of life. Each of these life zones 
probably represents a very long period of time, for in 
each there was a very considerable evolution of the 
titanotheres as well as of other forms. These zones 
(17-9; see table, p. 9), named in descending order, 
are as follows: 



INTRODUCTION TO MAMMALIAN PALEONTOLOGY 



17. Titanotherium-Mesohippus zone (Brontops robustus zone, 
fauna; Chadron C fauna; Brontops dispar zone, Chadron B 
and Brontops brachycephalus zone, Chadron A fauna) . 

16. Theoretic zone (Uinta C 2). 

15. Diplacodon-Protitanotherium-Epihippus zone (Uinta C 1 
fauna) . 

14. Eobasileus-DoKchorhinus zone (Uinta B 2 and Washakie B 2 
faunas) . 

1.3. Metarhinus zone (Uinta B 1 and Washakie B 1 faunas). 

12. Uintatherium-Manteoceras-Mesatirhinus zone (Bridger C 
and D and Washakie A faunas). 

11. Paleosyops paludosus-Orohippus zone (Bridger B 
fauna) . 

10. Eornetarhinus-Trogosus-Palaeosyops fontinalis zone 
(Bridger A and Huerfano B faunas). 
9. Lambdotherium-E otitanops-C oryphodon zone 
(Wind River B and Huerfano A faunas) . 

Estimated duration of tJie titanothere epoch. — 
The duration of the titanothere epoch, from 
the time of the earhest known member of 
the family {Larnbdotherium) to that of the 
last product of titanothere evolution {Bron- 
totherium) is estimated as 600,000 years. 
This estimate is based on the calculation of 
Walcott, made from measurements of the 
rate of geologic sedimentation, that the 
entire Tertiary period covered not more than 
3,000,000 years. If estimates made by Bar- 
rell (1917.1, p. 892), based on radioactivity, 
can be verified the duration of Tertiary time 
should be extended to 54,000,000 years. If 
this estimate is accepted the duration of 
the titanothere epoch alone would extend to 
11,000,000 years. Though the geologic esti- 
mate of 600,000 years for titanothere evolu- 
tion seems to be too small, the physical esti- 
mate seems to be too great, and for the 
present we may regard the estimate based on 
geologic data as ranging between 600,000 and 
1,000,000 years. 

HAYDEN'S SUBDIVISIONS OF THE EOCENE AND THE 
OLIGOCENE 

The geologic formations in which titano- 
there remains occur and the life zones into 
which these formations are subdivided have 
been discovered and described during the last 
56 years, the first report on them being that 
of Meek and Hayden (1862.1), in which the 
entire Tertiary geologic column is represented 
in a "General section of the Tertiary rocks 
of Nebraska," reproduced here in facsimile. 

There is little doubt that when Hayden described 
the White River group as "1,000 feet or more" in 
thickness, as including the "Bad Lands of White 
River; under the Loup River beds, on the Niobrara, 
and across the country to the Platte," and as com- 
posed of "white and light-drab clays, with some beds 
sandstone, and local layers limestone," he had in mind 
the area extending from Cheyenne River of South 
Dakota to the region south of North Platte River, 



displayed in the accompanying map and panoramic 
section. This section includes at its base the Titano- 
therium and Oreodon zones (Chadron and Brule for- 
mations), from which Hayden listed certain char- 
acteristic forms of animal life, such as TitanotJierium 
{=Menodus), Choeropotamus {=Ancodus, Hyopota- 
mus), "Rhinoceros" {=Caenopus), AncJdtherium 
{= Mesohippus)' , Hyaenonodon {= Hyaenodon) , Ma- 
chair odus { = Dinictis). 

Gemral Section of the Tertiary rocks of Nebraska. 



Names. 


SUBDIVISIONS. 


Thick- 


LOCALITIES. 


Foreign 
Equiva- 
lents. 


.a 
> 

2 
3 


Fine loose sand, with some 
layers of limestope, — contains 
bones of Canis, Felis, Castor, 
Eguiis, Mastodon, Tesiudo, &c., 
some of which are scarcely dis- 
tinguishable from living spe- 
cies. Also Helix, Pliysa succinea, 
probably of recent species. All 
fresh water and land types. 


o 
o 

o 


On Loup fork of 
Platte River ; extend- 
ing north to Niobrara 
River, and south to 
an unknown distance 
beyond the Platte. 


a 


u 
> 

5 


White and light drab clays, 
with some beds sandstone, and 
local laj-ers limestone. Fossils, 
Oreodon, Titanotherium, Ch(?ro- 
potamus, Rhinoceras, Anchitke- 
rium,Bycenonodon,A/achairodus, 
Trionyx, Testudo, Helix, Plan- 
orbis, Limnma, Petrified wood, 
&c. &c. All extinct. No 
brackish water or marine re- 
mains. 


i 
a 

o 

o 
o 

r-t 


Bad Lands of White 
River ; under the 
Loup River beds, on 
Niobrara, and across 
the country to the 
Platte. 


o 


13 P. 


Light gray and ash colored 
sandstones, with moro or less 
argillaceous layers. Fossils, — 
fragments of Trionyx, Testudo, 
with large Helix, Vivipara, 
Petrified wood, &c. No marine 
or brackish water types. 


O 
O 
O 
IM . 


Wind River valley. 
Also west of Wind 
River Mountains. 


- 


'3 

'a 
1 


Beds of clay and sand, with 
round ferruginous concretions, 
and numerous beds, seams and 
local deposits of Lignite ; great 
numbers of dicotyledonous 
leaves, stems, &c.of the genera 
Platanus, Acer, Ulmus, Populus, 
&e., with very large leaves of 
true fan Palms. Also, Helix, 
Melania, Vivipara, Corbicula, 
Unio, Ostrea, Potamomya, and 
scales Lepidotus, with bones of 
Trionyx, Emys, Compsemys, 
Orocodilus, he. 


U 

o 

a 

u 

o 


Occupies the whole 
country around Fort 
Union, — extending 
north into the British 
possessions, to un- 
known distances ; 
also southward to 
Fort Clark. Seen un- 
der the White River 
Group on North Plat- 
te River above Fort 
Laramie. Also on 
west side Wind River 
Mountains. 


1 



Figure 6. — The Meek and Hayden Tertiary section of 1862 

The deposits named are now known to include the following: 

"Loup River beds." The lower Pleistocene fauna listed is found in an area that includes 
deposits of the Pliocene and upper Miocene (Ogalalla formation of Darton) . 

"White River group," including lower Miocene (Arikaree formation of Darton) and Oligo- 
cene (Brule and Chadron formations of Darton). The " Choeropotamus" is Ancodus 
ameTicanus, the ancodont of the Chadron formation (Titanotlierium zone). 

"Wind River deposits" (summit of the lower Eocene). 

"Fort Union or Great Lignite group" (basal Eocene). 



These Titanotherium and Oreodon zones are now 
regarded as lower and middle Oligocene, respectively, 
and above them have been discovered the Protoceras 
and Leptauchenia zones, which embrace the highest 
sediments assigned to the Oligocene. The combined 
thickness of the Oligocene at this point is 600 to 650 
feet. Above it, to the east, are "light-drab clays," 
having a total thickness of 500 feet, and these, when 
combined (1,150 feet), correspond to the "1,000 feet 



6 



TITANOTHEEES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



or more" of Hayden's section. It therefore appears 
that Hayden's description of the White River group 
conforms with the accompanying panoramic section 
of the Oligocene and lower Miocene exposed on the 
south side of White River, South Dakota, shown in 




of his White River group apparently came from 
beds now classified as Oligocene. The name White 
River group has therefore for years been restricted 
to the beds of Oligocene age (Brule and Chadron 
formations). 

DISCOVERY OF THE TITANOTHEEES OF 
THE PLAINS 

At the base of this great section lies 
the Titanotherium zone, or " Titano- 
therium beds" of the Hayden-Leidy 
memoirs, fully described in Chapter II, 
composed in part of clays, in part of 
river-channel sandstones, in which 
titanothere remains are extraordi- 
narily abundant. 

The northern borders of this wonder- 
ful region appear to have been first 
explored around Bear Creek, a dry 
tributary on the south side of Cheyenne 
River, from which Thaddeus A. Cul- 
bertson brought back the first collection 
of fossils in 1850. From these expo- 
sures of the Titanotherium. and Oreodon 
life zones were obtained the greater 
part of Leidy's types, which are de-c 
scribed in Chapter III. The Brule and 
Arikaree formations, which overlie the 
Chadron, belong to a period succeed- 
ing the titanothere epoch, with which 
this monograph closes. 

The physiography of this ancient 
flood-plain region — its broad level 
stretches, its meandering rivers, its 
fringing river-border forests, its distant 
mountains and active volcanic peaks — 
as restored from our present knowledge 
of its fauna and flora, is described in 
Chapter II. It forms a wide contrast 
to the mountain-basin region, in the 
heart of which lie the Wind River de- 
posits, described by Hayden in 1862. 

DISCOVERY OF THE MOUNTAIN - BASIN 
ENVIRONMENT OF THE TITANOTHEEES 



As the entire lower Oligocene history 
Figure 7 —Panoramic section of the Big Badlands of South Dakota, looking of the titanotheres is recorded chiefly 
southeastward across Cheyenne and White Rivers to Porcupine Butte 



This section of the ancient flood-plain sediments now expoi^ed cuts through five great life zones — the 
Titanotherium, Oreodon, Leptauchenia, Promenjcochoerus, and Merycochoerus zones. It includes also four 
ancient river-channel sandstones and conglomerates— the •'Titanotherium sandstones," " Metamynodon 
sandstones," "Protoceras sandstones," and " Fromcrycochoerus sandstones" — each of which includes a 
more or less distinct river-border and forest fauna. (See map, fig. 69, vicinity of section B.) 

Figure 7, as sketched under the direction of Osborn 
for the United States Geological Survey in 1909. 
(Osborn and Matthew, 1909.321.) Hayden did not, 
however, specifically define the upper limit of his 
group, and all the fossils listed by him as characteristic 



in the Great Plains region east of the 
Front Range of the Rocky Mountains, 
so their entne Eocene history is 
recorded almost exclusively in the 
mountain -basin region west of the 
Front Range, in western Wyoming, northwestern 
Colorado, and northeastern Utah. The interpreta- 
tion of these remnants of the great Eocene sediments 
(given in Chapter II) involves far more difficult 
problems and has required more prolonged and in- 



INTRODUCTION TO MAMM.tLIAN PALEONTOLOGY 



D A K O T A 




Chadron Formation 
(Tit<xjvo i^herTZi^n/ZoThe) 



Figure 8. — Map showing the type geologic locaUty (X) of the Titanotherium zone on Bear Creek, branch of 

Cheyenne River, S. Dak. 

The map shows the present exposures of the Cbadron formation in South Dakota, Nebraska, Montana, eastern Wyoming, and Colorado. These exposures 
of the Titanotherium zone form the northern and western fringes of the overlying sediments, composing the Brule and Arikaree formations {the great 
"White River group" of Hayden). Map after Darton, U. S. Geological Survey, 1905, modified from observations of Matthew and Thomson, 1906, 1907. 



TITANOTHEEES OP ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



tensive geologic researcli than tlie interpretation of 
the Oligocene sediments. The program for this 
exploration was proposed by the author to the Director 
of the United States Geological Survey in 1900. 

Each of the typical lower, middle, and upper Eocene 
basins shown in the accompanying map has had its 




Wind River Valley. Also west of Wind River Moun- 
tains." It is possible that Hayden here refers to the 
Wasatch or the Bridger formation, which lie southwest 
of the Wind River Range. 

Subsequent exploration by Hayden revealed the 
typical Bridger, Wasatch, and "Washakie"^ forma- 
tions, each affording portions of separate 
chapters in the history of the ancient 
mammalian life of the mountain-basin 
region, which has proved to be no less 
remarkable than that of the Great Plains. 
Hayden was aided by the early paleon- 
tologic observations of Leidy on the 
Bridger fauna. 

The survey along the fortieth parallel 
by Clarence King was supplemented by 
the paleontologic observations of Marsh, 
who described the life areas south of 
the Uinta Mountains and defined the 
Diplacodon zone of the Uinta. Cope 
hastened to describe the life of the Wind 
River, of the Wasatch, and of the 
"Washakie" formations and made known 
a very rich fauna contemporaneous with 
the Wasatch of the Big Horn Basin, to 
the north, and of the San Juan Basin of 
northern New Mexico, to the south, where 
he also discovered the basal Eocene fauna 
(Puerco). Five of these six geologic for- 
mations were long regarded also as fau- 
nistic units and were described as single 
life zones, namely, the "Diplacodon beds" 
(Uinta formation), the "Dinoceras beds" 
(Bridger and "Washakie" formations), 
the " LambdotJierium beds" (Wind River 
formation), the " Coryphodon beds" 
(Wasatch formation), and the Puerco 
formation. 

The intensive observation of these 
six formations and the analysis of their 
fauna has enabled us to divide them 
into sLxteen known life zones, which in 
turn afford the key to the time of origin 
and of cessation of sedimentation in each 
basin. 



Figure 9. — Map showing cluster of typical lower, middle, and upper Eocene 
sedimentary basins in the heart of the Rocky Mountain region 



DISCOVERY AND DELIMITATION OF PERIODS 
OF SEDIMENTATION AND OF LIFE ZONES 



Mapped chiefly aftst the explorations of Hayden, King, and Powell of successive Government 
surveys. The arrows indicate the lines along which were taken the chief geologic sections de- 
scribed and illtistrated in Chapter II. Modified from Osbom and Matthew, 1909.321. 



antecedent historic and its recent analytic treatment, 
beginning with the Wind River deposits of Hayden 
(Meek and Hayden, 1862.1, p. 433), who described 
these deposits as "light-gray and ash-colored sand- 
stones, with more or less argillaceous layers. Fossils — 
fragments of Trionyx, Testudo, with large Helix, 
Vivipara, petrified wood, etc. No marine or braclcish- 
water types * * * 1,500 to 2,000 feet * * * 



The fact that these sediments ac- 
cumulated very slowly, during long 
periods of geologic time and in the course of profound 
changes in climatic and physiographic environment, 
with consequent variations in the fauna and flora, 
has gradually become recognized, and the explorations 
and researches that have led to this recognition have 

s The Washakie was contemporaneous with the upper two-thirds of the Bridger 
formation, and the name Washakie is now abandoned by the United States Geolog- 
ical Survey for the name Bridger. 



INTEODUCTION TO MAMMALIAN PALEONTOLOGY 



9 



formed a considerable part of the work done for this 
monograph. At first the periods of sedimentation 
were regarded as broadly equivalent to a similar 
number of life zones. For example, up to the year 
1900 the two chief formations, the Wasatch and the 
Bridger, were treated as containing one fauna each. 
It was not known that the Wasatch represents five 
distinct life zones, that the Bridger represents four 
and perhaps five life zone^, and that the partly con- 
temporaneous Washakie represents three distinct life 
zones. The correlation of different areas of sedimen- 
tation by means of fossils was similarly loose and in- 
exact. The evidence discovered since 1900 by parties 
sent out from the American Museum of Natural 
History proves that there was considerable change of 



environment as well as a great faunal change during 
Bridger time. The careful recording of the precise 
geologic level at which every specimen, especially 
every type specimen, was collected, together with 
close analysis of lithologic evidence that the rocks 
afford as to modes of deposition, has worked a com- 
plete revolution in our knowledge of the history of 
these mountain basins in Eocene time and of the 
flood plains in early Oligocene time and has afforded 
the relatively precise and far more interesting sequence 
of events that is described in Chapter II. 

Our geologic studies show that from basal Eocene to 
early Oligocene time there were six great physiographic 
and climatic epochs of sedimentation, shown in the 
accompanying table. 



Epochs of sedimentation and life zones from iasal Eocene to early Oligocene time in hasins in the Rocky Mountain 

region 



Physiographic epochs 



6. Lower Oligocene, represented by Chadron formation. Flood plains east of 
the Rocky Mountains. Sedimentation extremely slow. Moderate rain- 
fall. Warm temperate climate. 

5. Latest upper Eocene, represented by Uinta formation (Uinta C). Flood- 
plain basins south of the Uinta Mountains. Sedimentation relatively 
rapid; fine material. Heavy rainfall, diminishing. 

4. Upper Eocene, represented by contemporaneous deposits in Washakie and 
Uinta Basins (horizons Washakie B and Uinta B) and probably by upper- 
most part of Bridger formation, or Bridger E. Violent river and stream 
action from the north and south sides of the Uinta Mountains. Erup- 
tions of volcanic dust; coarse material. Heavy rainfall. 

3. Middle Eocene, represented by Bridger formation (horizons Bridger A, B, 
C, andD). More quiescent flood-plain conditions in the Bridger Basin; 
eruptions of volcanic dust; intervals of evaporation. Sediments com- 
posed in part of eroded material, generally laid down on lacustrine 
deposits. 

2. Lower Eocene, represented by Wasatch, Wind River, and Green River 
formations. Warm temperate climate of the Green River lake period, 
and evidently arid conditions in the contemporaneous Wind River sedi- 
ments. Alternation of arid and fluviatile conditions characteristic of 
Wind River and Wasatch time. Evidence of open country, favorable to 
cursorial mammals. 

1. Basal Eocene, represented by the Puerco, Torrejon, and Fort Union forma- 
tions. Forests, base-leveled areas, flood plains, and swamps widespread. 
Evidence of somewhat cooler climate. 



17. Titanotherium-Mesohippus. 
16. Theoretic zone. No fauna discovered. 
1 5 . Diplacodon-Protitanotherium-Epihippus. 
14. Eobasileus-Doliohorhinus. 



13. Metarhinus. 

12. Uintatherium-Manteoceras-Mesatirhinus. 

11. Palaeosyops paludosus-Orohippus. 
10. Eornetarhinus-Trogosus-Palaeosyops fonti- 
nalis. 

9. Lambdotherium-Eotitanops-Coryphodon. 

8. Heptodon-Coryphodon-Eohippus. 

7. Systemodon-Coryphodon-Eohippus. 

6. Eohippus-Coryphodon. 

5. Phenacodus-Nothodectes-Coryphodon. 

4. Pantolambda. 

3. Deltatherium. 

2. Polymastodon. 

1. Ectoconus. 



The evidence of the existence of these successive 
climatic, physiographic, and biologic conditions is 
derived from studies by Berry of the flora; by Hay of 
the reptiles; by Osborn, Scott, Wortman, Granger, 
Matthew, Peterson, Douglass, and Riggs of the mam- 
mals; and by Sinclair and Johannsen of the lithology. 
These studies, the results of which are in part set forth 
in Chapter II, show a great advance upon the pioneer 
studies by Leidy, Marsh, and Cope, which were based 
chiefly on characters of the skeleton and teeth. 
101959^29— VOL 1 3 



Our paleontologic division of the strata of the Eocene 
and lower Oligocene epochs into sixteen known life zones 
and one theoretic life zone enables us to fix the date of 
the immigrations of animals into this region, as well as 
the emigrations and extinctions, with much greater 
precision than formerly. Remains of titanotheres 
have been found in the upper eight of the known life 
zones. 

Extremely important is the realization that the zonal 
fossil fauna reflects local conditions of sedimentation. 



10 



TITANOTHEKES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



which have a significant bearing on the kinds of animals 
preserved. For example, violent fluviatile action 
may preserve for us chiefly the river-border and aquatic 
fauna; but remains of the animals of the surrounding 
plains and of the distant forests may not have entered 
the river-channel sandstones. Forest-living animals, 
like the chalicotheres [Moroyus], are relatively rare; 
and arboreal animals, like the lemurs (NotJiardus) , are 
seldom preserved in channel sandstones. Certain 
mammals apparently arriving as new immigrants, like 
the giant uintatheres, which suddenly appear in 
Bridger C, doubtless came from the surrounding plains 
or mountain regions, where the conditions were un- 
favorable for their entombment and fossilization. 

The threefold division of the Wasatch and Bridger 
mammals by Matthew (1909.1) and Loomis (1907.1) 
according to their habitats, into meadow, forest, and 




Figure 10.- 



- Restorations of Eotilanops borealis (A) and Brontotherium plaly- 
ceras (B) , drawn to the same scale 



titanotheres has extended we invariably find more than 
one of the branches of the titanotheres, as in Wind 
River and early Bridger time, and in some areas as many 
as five or six contemporaneous branches. Altogether 
twenty branches of the great titanothere family tree 
have thus far been discovered in Eocene and lower Oligo- 
cene strata. This multiple branching, known as poly- 
phyletic evolution, has made the study of the titano- 
theres more difficult and at the same time more 
fascinating than if these mammals presented only 
a single line of descent, as in monophyletic evolution. 
Some of the phyla of the titanotheres can be traced 
through a long series of successive evolutionary stages, 
such as Palaeosyops, Manteoceras, and Dolichorhinus 
in the Eocene, Brontops, Menodus, and Brontotherium 
in the Oligocene. Other phyla, such as the supposed 
river-dwelling Eometarhinus and Metarhinus, appear 
in two life zones only, in the middle Eocene, 
Huerfano B, and the upper Eocene, Uinta 
B 1, under fluviatile conditions of sedi- 
mentation favorable to fossilization. 

Extremes of evolution. — Members of these 
twenty branches wandered in and out of 
the regions favorable to fossilization, and 
consequently no single branch (phylum) can 
be traced over the whole period of time. 
Even if this period covered 600,000 years 
(minimum estimate), or 11,000,000 years 
(maxunum estimate), the descent of a gi- 
gantic horned quadruped, such as Bronto- 
therium platyceras, from a small and de- 
fenseless animal akin to Eotitanops borealis 
would appear almost incredible were it not 
that unremitting exploration during the last 
half century has unearthed many phyla of 



One of the earliest members of the titanothere family (£. borealis of the Wind River formation, lower 
Eocene) and one of the latest and most formidable ( B. platyceras of the White River group, lower 
Oligocene). Frommodelsinthe American Museum of Natural History made by Erwin S. Christman spCcicS that are mOre Or IcSS intermediate 
imder the direction of the author and of William K. Gregory. 



river living groups, not only has important bearing on 
the gaps in the fossil record and on the interpretation 
of the evidence relating to immigration and emigration 
but is in accord with the principle of local adaptive 
radiation developed by Osborn, as fully set forth at the 
end of Chapter II. 

PRINCIPIE OF LOCAL AND CONTINENTAL ADAPTIVE 
RADIATION 

The changes in the climatic and physiographic 
conditions during the Eocene epoch, which favored not 
only the evolution but the fossilization of this or that 
type of animal, supply the key to the divergence in 
anatomical structure and to the presence in the 
diversified Eocky Mountain region and adjacent 
plains of a great variety of titanotheres, in a measure 
comparable to the great variety of ruminants found 
to-day in the plain and plateau regions of the continent 
of Africa. 

Twenty branches of titanotheres. — In the eight life 
zones through which the observed evolution of the 



between these two extremes. Although the 
whole period of life of the titanotheres was relatively brief 
as compared with that of the surviving horses, tapirs, 
and rhinoceroses, yet within this period the titano- 
theres became much more specialized than the modern 
tapirs; in fact, although in lower Eocene time they 
resembled superficially the existing tapir {Tapirus 
terrestris), by middle Eocene time they had reached and 
passed the tapir-like stage of evolution. As compared 
also with the contemporary horses they were more 
rapidly progressive; the difference between the lower 
Oligocene Brontotherium and the lower Eocene Eoti- 
tanops is vastly greater than that between the lower 
Oligocene horse Mesohippus and the lower Eocene 
Eohippus. The titanotheres evolved rapidly, partly 
because the environment was peculiarly favorable 
to their rapid evolution; partly because their internal 
germinal hereditary conditions favored their rapid 
evolution and differentiation. 

Competition oj the titanotheres with other ungulates. — 
In the course of their evolution the titanotheres came 
into competition as herbivorous quadrupeds with 
members of four orders of hoofed mammals. They 



INTEODUCTION TO MAMMALIAN PALEONTOLOGY 



11 



competed with members of two archaic orders, the 
Amblypoda, typified by Coryphodon, and the Condy- 
larthra, typified by Phenacodus. The titanotheres 
survived both these archaic orders. They came 
into competition with members of several other 
families of the Perissodactyla and rapidly outstripped 
them in evolution. The period of the extinction of 
the titanotheres, at the end of lower Oligocene time, 
marked also the decline of several other of the great 



rhinoceroses are the only odd-toed ungulates that 
outlived the titanotheres and survived to the present 
time. The fourth order of quadrupeds that competed 
with the titanotheres were the Artiodactyla, the dimin- 
utive ancestors of the even-toed ungulates, including 
the ruminants, which entered a great era of expansion 
soon after the titanotheres became extinct. 

The earliest known types of titanothere evolution, 
Lambdotherium and Eotitanops, which were contem- 




FiGUEE 11. — Amblypoda: Skeletons and restorations of an ancestral form (A) and a specialized form (B) 

A, Pantolamida of the basal Eocene Torrejon formation; B, Coryphodon of the Wasatch formation, persisting throughout five life zones of 
lower Eocene time, contemporaneous in its later stages of development with Eotitanops and Lambdothtriiim, ancestral titanotheres. 



families of perissodactyls, especially the aquatic 
rhinoceroses (amynodonts), the cursorial rhinoceroses 
(hyracodonts), and the fleet lophiodonts (Colodon), 
all of which became extinct soon after the titanotheres 
disappeared. The aberrant perissodactyl chalico- 
theres, which are in many respects similar to the titan- 
otheres, survived, perhaps because they retreated, 
like the okapi of the Congo region of Africa, into the 
recesses of the forests. The tapirs, horses, and true 



poraneous, appear in the fourth Coryphodon life zone. 
Coryphodon is a clumsy but powerful mammal of very 
archaic type, heavily armed with great canine tusks. 
It is descended from Pantolamhda of the basal Eocene. 
As Coryphodon appears in the far distant region of 
the Sparnacian of France as the companion of a giant 
bird (Gastornis) and of a primitive horse {Hyracothe- 
rium) similar to the American Eohippus, France and 
western America are brought close together in their 



12 



TITANOTHEEES OF ANCIENT WYOMING, DAKOTA, AND NEBKASKA 



mammalian life during lower Eocene time, so that we 
shall probably discover a similar Coryphodon fauna 
in the intermediate regions of eastern Europe, northern 
Asia, and British Columbia. 

COMPARISON OF THE FOUE LIFE PHASES IN EUROPE 
AND IN NORTH AMERICA DURING EOCENE AND EARIY 
OLIGOCENE TIME 

Length of Eocene time. — It is the comparison of the 
ancient life of the Old and the New World, especially 
by means of the results of the successive studies of 
Cope, Filhol, Deperet, Osborn, and Matthew, that 
has led to the demonstration by Osborn of four great 
continental faunal phases in Eocene and lower Oligo- 
cene time — phases that probably extended over the 
entire Northern Hemisphere and that were separated 
by the rise and fall of the archaic forms of life, by the 
union or separation of western Europe and western 
America into one single or two distinct centers of mam- 
malian life, and by the severance of all connection be- 



tween North and South America. Together these 
three series of events form a sequence that affords evi- 
dence of the great length of Eocene time. In other 
words, the biologic evidences of very marked evolution 
in single families like the titanotheres, of the zoogeo- 
graphic events of migration, and of the succession and 
extinction of faunas together indicate that the Eocene 
epoch alone may have been longer than the 600,000 to 
1,000,000 years allotted to the titanothere epoch in 
accordance with Walcott's estimates of Tertiary time 
based upon purely geologic data. 

The archaic succeeded hy the modernized mammals. — 
The long duration of Eocene time is further indicated 
by the subdivision of the Wasatch {Coryphodon) epoch 
(the "Coryphodon beds" of Marsh and Cope) into 
five lesser time divisions. Thus the term Coryphodon 
alone no longer serves as the designation of a life zone, 
because Coryphodon is now known to have survived 
through at least five life zones, Nos. 5-9 in the 
zonal series (p. 57), as follows: 



"Coryphodon beds" oj Marsh and Cope 
9. Lambdotherium-Eotitanops-Coryphodon zone o( Osborn 



8. Heptodon-Coryphodon-Eohippus zone 

7. Systemodon-Coryphodon-Eohippus zone__ 
6. Eohippus-Coryphodon zone 

5. Phenacodus-Nothodedes-Coryphodon zone. 



The modernized mammals in the series tabulated 
above are the titanotheres, lophiodonts, tapirs, horses; 
the archaic mammals are the condylarths (Phenacodus) 
and amblypods (Coryphodon). 

As remarked above, no single biologic phenomenon 
affords stronger evidence of the long duration of 
Eocene time than the complete replacement of the 
archaic fauna of North America, which exclusively 
held the stage during basal Eocene time, in itself a 
very long epoch, by the ancestors of modern mam- 
mals, as shown in the accompanying diagram (fig. 12) 
and indicated precisely in the transition between the 
Phenacodus and Eohippus zones. The modernized 
mammals came in not suddenly or en masse, as we 
formerly supposed, but gradually, family by family, 
the first apparently being the swiftest and most vita- 
tive family — the horses (Eohippus). 

We infer that western Eiu-ope witnessed a similar 
replacement, for, although sparsely loiown, the basal 
Eocene life of western Europe was broadly similar to 
that of western North America. 

The archaic life of American basal Eocene time, 
first made known by Cope, then studied by Osborn 
and Earle, and finally given very full and precise 
geologic and zoologic determinations by Matthew, 



_ First appearance of the titanotheres in America. " Wind 
River" fauna of Cope. 

.First appearance of lophiodonts in America. "Lysite" 
fauna of Granger. 

.First appearance of tapirs in America. "Gray Bull" fauna 
of Granger. 

.First appearance of horses in America. "Sand Coulee" 
fauna of Granger. 

_ Phenacodus extremely abundant. "Clark Forli" and "Tif- 
fany" fauna of Granger. The closing phase of the reign of 
the archaic mammals of North America, Pantolambda, Cory- 
phodon, Phenacodus. 

Granger, and Gidley, affords the basis of our present 
knowledge of the wonderfully rich and varied fauna 
embraced within the four basal Eocene life zones. 

The precision with which we are now able to note 
the extinction or disappearance of the archaic mam- 
mals and their replacement, one by one, by members 
of modernized families is due especially to the ex- 
plorations of the American Museum of Natural His- 
tory, led by Granger with the assistance of Sinclair, 
and to the analyses of the fauna by Matthew and 
Granger in a series of researches which are classic not 
only for their precision but for the revelation of new 
and hitherto unsuspected affinities of the mammals 
of North America with those of South America and 
with the existing mammals of the oriental region of 
the Old World. 

Relation of the titanotheres to other quadrupeds. — In 
their broadest relations the titanotheres were mam- 
mals of the cohort Ungulata, which possess hoofs as 
distinguished from claws. We know that eleven great 
orders of ungulates (see accompanying table) were 
distributed through different parts of the earth during 
ancient and modern time. Of these eleven orders, 
which were the sources of the herbivorous quadrupeds 
of the world, only five have survived to the present 
time. 



INTRODUCTION TO MAMMALIAN PALEONTOLOGY 



13 



TTie eleven orders of Tertiary ungulates 

I. Archaic ungulates: 

I America and Eurasia. Originating in Cretaceous time and contemporaneous 
in Eocene time (Coryphodon) witli the titanotheres, becoming extinct in late 
Eocene time {Uintatherium and Eobasileus). 
I America, Eurasia, and possibly South America. For a short period contempo- 
raneous with the titanotheres, becoming extinct in the lower Eocene (Phena- 
codontidae) . 

II. Modernized ungulates: 

A. Primarily North American and north Eurasian: 

..,,,, . , , , f America, Eurasia, and subsequently South America. First appearing in early 

,' „ . , ^ , „ i.j_^_ _I,'""7 ~ I Eocene time. The Perissodactyla gradually gave way to the Artiodactj^la. 

The chalicotheres were in part contemporaneous with the titanotheres near 
the end of their Ufe period. 



4. Perissodactyla (horses, titanotheres, 
tapirs, rhinoceroses). 



B. Originally African-Asiatic ungulates: 

„ -H rv, ^ [First appearing on the African continent; subsequently, in part, entering 

„■ „ , ., , , , _ ~j ----------- 1 southern Eurasia and North America. None of these orders is known to 



Proboscidea (elephants and masto- 
dons). 



7. Sirenia (sirenians') . 



have been contemporaneous (in Europe) with the titanotheres or to have 

[ entered into competition with them. 

[Aquatic mammals, first known in Africa, possibly related to the same ancestors 

' I as the Proboscidea; believed to have sprung from ungulate ancestors. 

8. Embrithopoda (arsinoitheres) Known solely on the African continent; Oligocene. 

C. Distinctively South American ungulates: 

Q p iv^ . / .i.- j_N [Exclusively South American in history and evolution.'' None of these orders 

in T rl +• f f l'\ J entered into competition with the titanotheres. Part of them (Litopterna) 

,,'-i X , \- J.N I imitated the other orders of ungulates, and part (Toxodontia) evolved into 

11. Litopterna (extmct) ■ , & > f ^ 

[ unique forms. 



* A single jaw attributed to one of the aberrant Southi American ungulates has been found in the Eohippus- Corypho 
Basin, Wyo. 



I lite zone, "Sand Coulee beds" of Clark Fork 



Only three of the eleven ungulate orders shown 
in the table were living in the Rocky Mountain 
region when the titanotheres arrived — (1) the archaic 
Amblypoda, represented, as we have seen, by Cory- 
phodon, extremely smaU-brained, of very clumsy 
build, heavy-footed, in general proportions somewhat 
like the African rhinoceroses, RTiinoceros (Cerato- 
therium) simus and R. (Opsiceros) iicornis; (2) the 
Condylarthra, represented by a diminutive Phena- 
codus, also extremely small-brained, contrasting with 
Coryphodon in its small size and cursorial build, 
formerly but no longer believed to be ancestral to the 
higher ungulates; (3) the modernized Perissodactyla, 
including the ancestors of the horses {Eohippus), 
tapirs (Systemodon) , and lophiodonts (Heptodon). 

The newly arriving perissodactyl titanotheres 
equaled in size and resembled in their general cursorial 
limb structure the condylarths as well as the horses, 
tapirs, and lophiodonts. They were greatly surpassed 
in size by members of the Coryphodon family, some 
species of which were quadruple the size of the earliest 
known titanotheres. However, certain of the titano- 
theres of this stage (Eofitanops) exceeded the condy- 
larths (Phenacodus) in size. 

It is noteworthy that the archaic Condylarthra 
(Phenacodus) were numerically preponderant in the 
Phenacodus zone, just prior to the arrival of the 
earliest perissodactyl horses. There was doubtless an 
incessant competition between all these modernized, 
alert, large-brained perissodactyl ungulates and the 
archaic, small-brained ungulates {Coryphodon and 
Phenacodus), which were especially inferior in the 



mechanics of their foot structure. When, in the 
upper Eocene, the clumsily built Amblypoda reached 
the final phase of their evolution in the gigantic 
Uintatherium and Eohasileus, they apparently became 
suddenly extinct, and at the same time the titanotheres 
suddenly began to develop into more formidable 
animals. At no time in the Tertiary period was the 
earth populated in the same region with more than 
one type of very large quadruped. In the Northern 
Hemisphere the dominance of the amblypods (in the 
Eohasileus- Coryphodon epoch) was succeeded by the 
dominance of the titanotheres (in the closing titano- 
there epoch), and the titanotheres in turn, when 
they had reached their largest development, suddenly 
became extinct with no trace of a preliminary stage of 
decline. 

OID AND NEW SYSTEMS OF CLASSIFICATION 
OLD TERMINOLOGY RETAINED 

The studies for this monograph were begun by 
Professor Marsh under the old ideas of classification 
in mammalogy, derived from Linnaeus and his suc- 
cessors. These studies were continued by Osborn on 
the same old lines, as shown in his first paper on the 
titanotheres. (Osborn, 1896.107.) The discovery 
of adaptive radiation and of polyphyletic evolution, 
which was one result of the researches made for this 
monograph, has developed a new phyletic system of 
classification. Yet even in this new system it is 
necessary to adhere to the old Linnaean terminology, 
for the reason that Linnaean methods have been used 
during the long period of systematic description in 



14 



TITANOTHERES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



which the greater number of genera and species of 
titanotheres have been described; and the Linnaean 
generic and specific names can not be replaced unless 
two systematic names have been given to the same 
animal. Rather than introduce a new terminology 
we attempt to place each Linnaean species in its proper 
phyletic position — that is, in its true phylum — and to 
connect it with other species by intermediate or transi- 
tion stages, which are termed mutations, the "ascend- 
ing mutations" of Waagen as distinguished from the 
contemporaneous "mutations" of De Vries. 

LINNAEAN METHODS OF DEFINING SPECIES, GENERA, 
AND PHYLA OF TITANOTHERES 

Between 1847 and 1902 as many as 29 genera and 
67 species of Eocene and Oligocene titanotheres were 
defined, but of all the definitions given hardly a single 



Proceeding along these lines Marsh and Cope defined 
a number of genera of titanotheres, certain of which 
have since proved to be closely successive members of 
the same phylum and consequently members of the same 
genus. Osborn went to the opposite extreme in attempt- 
ing to reduce all the titanotheres to a single genus. In 
his paper of 1896, entitled "The cranial evolution of Ti- 
tanotJierium" (Osborn, 1896.110),hereached the wholly 
erroneous conclusion that there had been only a single 
distinct and definable genus of titanotheres — the origiaal 
Titanotherium of Leidy — and that all the variations 
among the titanotheres were of the rank of species, rep- 
resenting different stages of development. This has 
proved to be a greater error than that of Marsh, because 
it was based on the hypothesis that the titanotheres 
belonged to a single — monophyletic — line of descent. 




APPEARANCE AND EXTINCTION OF MAMMAL ORDERS IN NORTH AMERICA 



Archaic J\fa7n7na2s - soled, hlacTo. Mode?-nLze<t .MajnTnaZs - outiine 



Figure 12. — Diagram showing the gradual extinction of orders of archaic mammals (solid black) of earliest 
Eocene time and their gradual replacement during later Eocene time by the ancestors of modernized orders 
of mammals (outline), including related forms that are now extinct 



one has proved to be distinctive and valid. The 
main characters utilized in the old classifications by 
the chief contributors to the history of the Oligocene 
titanotheres — that is, by Leidy, Marsh, Cope, Scott, 
and Osborn — were the following: 

1. The presence or the absence and the number of incisor 
teeth (Cope and Marsh, in generic definition). 

2. The number of premolar teeth (Marsh, in generic definition). 

3. The development of the cingulum on the premolar teeth 
(Cope andMarsh, in generic definition). 

4. The presence of a second cone on the last superior molar 
(Marsh, in generic definition). 

5. The length and shape of the nasal bones (Cope, Marsh, 
Scott, and Osborn, in generic definition). 

6. The length and shape of the fronto-nasal horns (Cope 
and Marsh, in generic definition). 

7. The presence or absence of a trapezium in the carpus 
(Hatcher, in phyletic definition). 



RECOGNITION OF MANY LINES OF DESCENT) POLYPHYLY 
THE KEY TO INTERPRETATION OF THE FAMILY 

In January, 1901, a few months after the studies 
for this monograph were begun, all the data, observa- 
tions, skull sections, and measurements were assembled, 
and by July of the same year it was demonstrated by 
Osborn that at least four lines of separate descent 
are to be found among the lower Oligocene titanotheres, 
and this number has since been increased to five or 
eight. 

In 1902 Osborn established the fact that throughout 
lower Oligocene time, when the Titanotherium-he&vmg 
beds were being deposited, as many as eight more or 
less different phyla, or series, were independently 
evolving in the same region. Certain of these phyla 



INTRODUCTION TO MAMMALIAN PALEONTOLOGY 



15 



embrace one or more of the genera originally proposed 
by Pomel, Leidy, Cope, and Marsh. Other phyla 
correspond with certain genera — for example, Menodus 
Pomel (syn. Titanotherium Leidy), Brontops Marsh, 
Allops Marsh, Megacerops Leidy, BrontotheriumMarsh. 
These five generic names correspond to members of 
five phyla that persisted throughout a very long period 
of geologic time. The remaining phyla of titanotheres 
are branches that persisted only for a short time, so 
far as we know at present — for example, Diploclonus. 
As shown in the accompanying diagram (fig. 15) 
these generic phyla are branches of the family tree of 



has adopted in expressing the relationships and 
descent of the rhinoceroses, animals whose evolution 
presents in many respects analogies to the evolution 
of the titanotheres, especially in the modes of the 
evolution of horns, in the loss or retention of cutting 
teeth (incisors), and in the adaptations of limb struc- 
ture to swift and slow movement. 

RELATION OF THE PHYLOGENETIC CLASSIFICATION TO 
THE LINNAEAN CLASSIFICATION 

Linnaeus described one or more species of mammals 
geographically distributed in space (see table on p. 16), 




FiGUEE 13. — Phenacodus (A) and Coryphodon (B) drawn to the same scale 
Restorations made by Charles R. Knight under the author's direction. 



the titanotheres. When two of these branches run 
close together they may for convenience be united into 
a single subfamily. Thus, for purposes of description 
the graphic presentation of the titanothere family tree 
in the accompanying diagrams may be supplemented 
by the systematic subdivision of these animals into 
12 subfamilies and 24 genera, as shown on a subse- 
quent page. 

The free use of subfamily divisions to distinguish 
the branches of Eocene and Oligocene generic phyla 
from one another is similar to that which the author 



whereas the phylogenetic classification of the titano- 
theres covers species extending over both space and 
time. The geographic range of the existing red deer 
(Cervus) and of the extinct titanotheres lies within 
the same zoogeographic region — Holarctica, which 
includes Europe, Asia, and North America. The 
comparison is therefore significant. (See table on 
p. 18.) 

The classification presented in this monograph is 
more than phylogenetic: it is polyphyletic. Lin- 
naeus (1758.1), when he wrote the several editions 



16 



TITANOTHEEES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



of his "Systema naturae" (1735-1768), did not 
dream of the succession of species of mammals in 
time; he did not know of a single phylum, much less 
of polyphyla. Darwin's theory of descent and 
divergence implied the existence of phyla, but when 
he published "The origin of species" (1859.2) he also 
did not ls;now of a single phylum or a single direct line 
of descent. Waagen (1869.1) was the first to dis- 



tively rapid gain or loss of certain characters. This 
definition relates to the hard parts, which are pre- 
served in fossUization; the principle applies equally 
to characters of all kinds. 

In contemporaneous Linnaean genera and species 
we observe differences of many kinds, such as differ- 
ences in color and proportion, and, more rarely, we 
may note the presence or absence of simple characters 



A Contemporojieous LiruzcbeoJi Syste/n Zoology 



X 



c 



' Su2).\Fam..l 






JFAM I LY 

_A ^^ 

Sub. Fam,. 2 



)T 



"^ 






. Spec. 



Spec./ 



\ I -^^f^^^ /'^Si^.Fa.nv.4^ 

.' , I ' Gerv. Gerv) 

;■"■•• J \ ^ ^ ^ 

Spec./ \ y V -^ 



\ 



RECENT EXISTING PERIOD ^ --^ ' 

B Geologic /. Phylogenetlc SystejTh Ihleontolo^y 

/ ^^ ^ /T? 

/ — ~~-^'^^, \ \\ 



/ 



LOWER/ 
0LI60CENE ,^ / 






/A 



\ 



/! \ 

1 ^ 



\ 



UPPER^ , 
EOCENE \ 

\ >^j 
\ 

MIDDLE EOCEP _ 

\ 




\ 



\ 



^ jA V-.^\\ V:. V" ^ \\ \\ i i ,, I \\ I y'l 



EARLY EOCENE 



\ 



/ 



Figure 14. — Contrast between the Linnaean and phylogenetic systems of classification of sub- 
families, genera, and species 



cover a continuous phylum (namely, of ammonites) — 
that is, successive hereditary stages, which he named 
"mutations." Many direct phyla of invertebrate 
animals have since been made known. 

In this monograph we first learn the full meaning of 
a mammalian phylum — namely, a phylum is a con- 
tinuous geologic line of descent diverging from other 
phyla (1) in the gradual transformation of every 
character in size and proportion and (2) in the rela- 



of teeth, vertebrae, or claws. The "species" of Lin- 
naeus are now known to be actually superspecies and 
to include one or more modern species, subspecies, and 
geographic races and varieties, distinguished by differ- 
ences in coloring, habit, proportion, or otherwise. 
These differences are due in part to environment and 
in part to habit. They represent the different bodily 
effects produced on animals of similar ancestral stock 
under different environments, in which somatic changes 



INTRODUCTION TO MAMMALIAN PALEONTOLOGY 



17 



are rapid and conspicuous. They are in part hereditary 
(germinal) diflferences, which pass down for generations 
unmodified by habit or environment. 

For example, the American genus Peromyscus (the 
white-footed mouse), as studied by Osgood (1909.1), 



(dolichocephalic). Peromyscus may have been widely 
distributed from some common center during the last 
40,000 years, and during this long period there may 
have been both geographic or space evolution and 
geologic or time evolution, the evolution in time being 




Figure 15.- 



-The family tree of the titanotheres, showing the relation between the branches (phyla), sub- 
families, and genera, as known to science in 1919 



The shaded areas show connections that f 



3 well established; the dotted lines show gaps that remain to be filled by future discovery, especially 
in the Uinta formation of Utah. 



presents a continuous series of transition changes in 
color and form in species having a geographic range 
from Tehuantepec to Alaska. In the northern re- 
gions Peromyscus is larger and has relatively longer 
teeth and a skull that may be somewhat elongate 



comparable to that which we observe in the geologic 
phyla of the titanotheres. 

The existing genus Cervus affords another example, 
significant because its geographic range is similar to 
that of certain Oligocene titanotheres. 



18 



TITANOTHEEES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



Species and subspecies of the genus Cervus 

[Table prepared by Qeirit S. Miller, 1918] 



Name 


Habitat 


Nature of habitat 


Climate 


Cervus canadensis (American elk = 


New York and New Jersey southward to 


Open plains, badlands, 


Humid to extremely 


wapiti) . 


the Carolinas; central western States; 


sand hills; forests and 


arid. 




Nebraska, the Dakotas, and the coun- 


meadows. 






try farther west, across the Rockies. 






C. c. merriami (Merriam's elk) .- 


New Mexico and Arizona _. _ . 


Mountains and plateaus; 


Generally arid; for- 






forests and meadows. 


ests wet. 


C. nannodes (dwarf elk) _ . 


San Joaquin Valley, Calif., and adjoining 


Plains and tule swamps 


Generally arid. 




foothills. 






C. occidentalis (Olympic elk) 


Washington, Oregon, California; formerly 


Chiefly forested regions; 


Humid. 




south to San Francisco Bay. Van- 


some meadows. 






couver Island? 






C. xanthopygus (Bedford's deer; 


Manchuria and adjoining parts of Siberia 


Forests .. ... 


Do. 


Manchurian stag). 








C. sibiricus (Altai maral) 


Baikal, Saiansk, and Altai Mountains; 


Formerly forests and open 


Extremely humid to 




southern Siberia and northern Mon- 


timberless country; 


extremely arid. 




golia. 


even open high desert 
mountains. Now re- 
stricted to forests and 
meadows. 




C. songaricus (Tien-Shan stag) 


Tien-Shan Mountains ^ ._ 




Mostly arid(?) 


C. yarkandensis (Yarkand stag) 


Eastern Turkestan _ _ _ 




Do. 


C. macneilli (Kansu stag) 


Kansu and Szechwan border of Tibet 








Tibet.. - _ ... .-...- 








do 






C. wardi (Ward's stag) 


do 






C. hanglu (Kashmir deer; hangul, 


Vale of Kashmir and adjacent mountains. 


Chiefly forest; some open 


Humid. 


hanglu) . 




parks. 




C. bactrianus 


Russian Turkestan . . 




Chiefly arid. 


C. maral (maral) . 


Persia, Crimea, Caucasus 














C. e. atlanticus (Norwegian red 


West coast of Norwav _ 




Do. 


deer) . 








C. e. germanica (red deer). 


Middle Europe 




Do. 


C. e. bolivari (red deer of central 


Central Spain.. 






Spain). 








C. e. hispanicus (red deer of south- 


Southern Spain. 




Humid and semiarid. 


ern Spain) . , 
















C. corsicanus (Corsican stag) 


Corsica and Sardinia. _ . 




Semiarid. 


C. barbarus (Barbary deer) _ _ _ _ _ 


Morocco, Algiers, palearctic north Africa 




Chieflv arid. 











COMPARISON BETWEEN ZOOLOGIC AND PALEONTOLOGIC 

SPECIES 

The difference between zoologic and paleontologic 
species is represented in tlie accompanying diagram 
(fig. 16), showing the descent and relationship of cer- 
tain members of the dog family (Canidae). A theo- 
retic stem or central form is shown from which geo- 
graphic races have been given off horizontally, as it 
were, and the ascending mutations and species of the 
evolutionary line of development from the ancestral 
form have arisen geologically. 

It follows that in making an anatomic comparison 
between the existing geographic species and sub- 
species of such genera as Peromyscus or Cervus and a 
geologic phylum of species such as that of Menodus or 
Brontotherium the same comparative anatomical 
methods of measurement and observation should be 



employed. Direct measurements of the length and 
breadth of the skull should be recorded, by which 
indices (proportions of single structures like the skull) 
and ratios (proportions between different parts like 
the upper and lower segments of the limbs) should be 
established. 

The proportional changes technically known as 
dolichocephaly (elongation of the head), brachy- 
cephaly (broadening of the head), dolichopy (elonga- 
tion of the face), brachyopy (abbreviation of the 
face), dolichopody (elongation of the feet), brachy- 
pody (abbreviation of the feet), dolichomely (elonga- 
tion of the limbs), brachymely (abbreviation of the 
limbs) occur in geographic species and subspecies in 
their corresponding stages exactly as they occur in 
geologic phyletic time series. The chief difference is 
that in the geologic time phyla these differences of 



INTEODXJCTION TO MAMMALIAN PALEONTOLOGY 



19 



proportion may be followed through long periods of 
time from their incipient to their final stages, in 
which various climaxes of change of proportion are 
reached, such as extreme length or breadth of head or 
extreme length or shortening of the feet. 

PROPORTIONS OF THE SKULL IN BEARS AND IN 
TITANOTHERES 

In comparing the Eocene and Oligocene titanotheres 
with the modern bears {Ursus), for example, as 
studied by C. Hart Merriam (1918.1), we may note 
certain parallelisms. The members of each of the 
eleven subfamilies of titanotheres are distinguished by 
certain proportions of the skull — that is, they are 
broad-headed, round-headed, or long-headed — by 
the shape of the horns and the acceleration or retarda- 
tion in their development, by the presence or 
absence of cutting (incisor) teeth, by certain 
proportions of limb, according as they are 
swift-footed (cursorial), slow-footed (medi- 
portal), or heavy-footed (graviportal), and 
by other minor features. The methods ap- 
plied to the study of the existing bears may 
be applied to the study of the skull or other 
hard parts of the titanotheres. In the titano- 
theres, however, we may observe all these 
changes of proportion actually in progress 
from stage to stage as revealed by paleontol- 
ogy, whereas in the bears we can observe only 
certain structural forms, which, so far as our 
observation goes, appear to be fixed or com- 
pleted, although they undoubtedly represent 
stages in a state of actual progression. 



B. Bridger and succeeding titanotheres — Continued. 

6. Manteoceratinae; mesatioephalic to brachycephalic; 

accelerated development of the horns; mediportal 

{Manteoceras, Prolitanotherium) . 
5. Diplacodontinae; dolichocephalic; accelerated molar- 

ization of the premolars; imperfectly known 

(Diplacodon) . 
4. Telmatheriinae; mesaticephalic to dolichocephalic 

{Telmatherium, Sthenodecles) . 
3. Palaeosyopinae; brachycephalic; short-limbed {Palae- 

osyops, Limnohyops) . 
A. Wind River titanotheres ; face longer than cranium : 

2. Eotitanopinae; medium-limbed, mediportal (Eoti- 



1. Lambdotheriinae; light-limbed, cursorial {Lambdo- 
thenum) . 

The above scheme presents the eleven subfamilies 
of titanotheres as they were distinguished in 1914. 



GEOGRAPHIC DISTRIBUTION 
AT PRESENT TIME 



True R 

ZcoUfiical WlPES 




-GEOGRAPHIC DISTRIBUTION 
IN PAST TIME 



FEATURES 



DISTINGUISHING 
TITANOTHERES 



PHYLA OF 



The first application of changes of propor- 
tion to the arrangement of the subfamilies of 
titanotheres is the following synopsis, pre- 
pared in 1914: 

Proportions of skull and limbs; presence and absence of characters 
distinguishing the subfamilies {main phyla) of titanotheres 
lOsborn, 1914.409] 

B. Bridger and succeeding titanotheres; cranium longer than 
face: 
11. Brontotheriinae; mesaticephalic to brachycephalic; 
horns long, transversely flattened, and divergent 
(Brontolherium) . 
10. Megaceropinae; mesaticephalic to extreme brachy- 
cephalic; horns long, vertically placed; no incisor 
teeth (Megacerops (—Symborodon)). 
9. Brontopinae; brachycephalic; horns short, rounded, 
or oval; incisors persistent (Brontops {=Mega- 
ceratops), Diploclonus) . 
8. Menodontinae; mesaticephalic to dolichocephaUc; 
short triangular horns; incisor teeth reduced or 
wanting; feet and limbs long {Menodus {=Titano- 
therium), Allops). 
7. Dolichorhininae; mesaticephaUc to doUchocephalic; 
limbs, so far as known, short (Dolichorhinus, 
Mesatirhinus, Sphenocoelus, Metarhinus, Rhadi- 
norhinus) . 



Figure 16. — Theoretic descent of existing members of the dog family 
(Canidae) from a common ancestor 

A represents the ancestral type. Dots represent intergradations indicated by paleontologic 
observations (vertical lines) covering five periods of geologic time. A', B, B', C, and C rep- 
resent existing forms, and dots represent a few existing intergradations demonstrated by zoo- 
logic observations (horizontal lines). Heavy lines and the adjacent dots represent the phyla; 
also the past and present distribution of geographic (ontogenetic and environmental) sub- 
species, races, and intergrades. 



Since that time certain phyla have been condensed by 
the discovery of titanotheres that link together some 
of these subfamilies, and others have been expanded 
by the discovery of new subfamilies, such as the 
Rhadinorhininae. 

MUTATIONS OF WAAGEN 

Where the fossil material is abundant the genera 
and species are found to be connected by a series of 
intergradations. These intergradations, though con- 
tinuous, are measurable, and therefore a species is 
subdivisible into a series of intergrading forms. The 
monophyletic, systematic, or taxonomic unit division 
of these species is the mutation of Waagen, which is a 
subspecific stage in the development of one or more 
characters. Such an actual sequence of mutations of 
Waagen may be illustrated in the genus BrontotJierium, 
as indicated on the following page. 



20 



TITANOTHEEES OP ANCIENT WYOMING, DAKOTA, AND NEBEASKA 

Oligocene stages of titanotheres of the Brontotherium 'phylum in the Titanotherium zone 



Division of zone 


Stage 


Species 


Tiieoretic ascending 
mutations 






Brontotherium platyceras 

Brontotherium ramosum 

Brontotherium ourtum 


Species. 

Subspecies. 

Mutation. 

Do. 
Species. 
Subspecies. 
Mutation. 

Do. 
Species. 
Subspecies. 
Mutation. 

Do. 
Species. 
Subspecies. 
Mutation. 

Do. 














Upper. 








Do 










Subspecies. 
Mutation. 
Do. 








Species. 
Subspecies. 
Mutation. 
Do. 


Middle. 












Subspecies. 
Mutation. 
Do. 






Brontotherium hypoceras 




Lower. 




Subspecies. 
Mutation. 
Do. 








Subspecies. 
Mutation. 
Do. 



" Genus Titanops IVIarsli. 
ZOOLOGIC AND PALEONTOLOGIC NOMENCLATURE 

Significance of the table. — The sequence shown in the 
accompanying table, which presents what is believed 
to be a generic, monophyletic, or nearly single phyletic 
series of changes of form, evolving in a single geographic 
region of South Dakota, illustrates the manner in 
which the Linnaean binomial system and the muta- 
tion substages of Waagen may be adapted to express a 
phyletic sequence. The newer trinomial names of 
modern mammalogy and the subspecific names may 
be employed to connect the intergrading mutations. 

The most primitive species, Brontotherium leidyi, is 
so notably distinct in size and skull structure from the 
most advanced species, Brontotherium platyceras, that, 
if named by zoological standards, it might well be 



' Type of genus Brontotherium (Marsli). 

placed in a separate genus — in fact, several generic 
names have been suggested for members of this 
phylum, namely, Brontotherium, Titanops, Bronto- 
theridion (MS.) — but the subdivision of such a phylum 
into a number of genera would obscure the all-im- 
portant monophyletic unity, for such a phyletic genus 
is defined by its peculiar and distinct evolutionary 
tendencies. For example, the genus Brontotherium 
tends toward the evolution of flattened horns, a charac- 
teristic which begins in a very slight flattening of the 
posterior side of the horn, as observed in B. leidyi, and 
develops into the extraordinarily broad, flattened 
horns of B. platyceras. 

New phyletic meaning of species. — The species repre- 
sented by large collections of mammals like those of 
some of the phyla of the titanotheres, especially the 



INTBODUCTION TO MAMMALIAN PALEONTOLOGY 



21 



Brontops phylum, are so closely intergraded and con- 
nected by "ascending mutations" that the dividing 
lines between them can be drawn only arbitrarily, 
according to individual judgment. In the Brontops 
phylum, for example, the species Brontops hrachyce- 
pJialus grades imperceptibly into the species Brontops 
dispar through gradual transitions in a great number 
of characters, as may be seen in the Hatcher collection 
in the United States National Museum. There is no 
evidence of brusque transitions, saltations, or jumps 
in any structure, such as are presupposed in the 
mutation theory of De Vries. By contrast, the 
mutations of Waagen are intergradations between 
arbitrarily defined species, and through these muta- 
tions species and genera pass imperceptibly one into 
another. 

Evolutionary characters of each phylum. — Thus we 
reach a clear conception of a phylum of the titanotheres 
in its osteologic and dental characters. A phylum may 
be further defined as a succession of interbreeding 
(syngamic, Poulton) individuals of similar (synepi- 
gamic, Poulton) ancestry, which may or may not 
occupy a similar range of country (synpatric, Poul- 
ton), which follow in every structural character a sim- 
ilar line of evolution (synphyletic, Osborn) and adap- 
tation (syntelic, Osborn). 

In each horn, in each tooth, in every bone of the 
skull and skeleton, and by inference in all the hard 
parts as well as in all the soft parts, each phylum has 
its distinctive mode and rate of transformation in each 



character, as follows: (1) Distinctive hereditary pro- 
portion; (2) distinctive tendencies to change of propor- 
tion; (3) distinctive progressive changes of proportion; 
(4) distinctive retrogressive changes of proportion; (5) 
distinctive accelerations and retardations in ontogeny 
(individual development) ; (6) distinctive rates (veloc- 
ities) of progression and retrogression in phylogeny in 
each character. In each phylum are consequently 
developed distinctive and ever changing proportions 
and ratios between different single characters and 
groups of characters, measurable by indices and ratios. 
Such indices express the degrees of broad-headed, long- 
headed, broad-footed, short-footed structure and pro- 
portion, and so on. Each phylum has also its distinc- 
tive but constantly changing indices and ratios of 
teeth to skull, of skull to body, of body to limbs, etc., 
which also are constantly changing as we pass from 
the lower to the higher geologic levels. 

Old and new meanings of taxonomic terms. — In the 
following table a comparison is made between the old 
and the new meanings of the taxonomic terms used 
by mammalogists. The definitions given in the sec- 
ond column are those of the old "special creation" 
system — followed by Linnaeus — which is based on 
geographic distribution alone; the definitions given in 
the third column are those of the new phyletic sys- 
tem — that of Osborn — which is based on both geologic 
and geographic distribution. The new system was 
first used for the rhinoceroses (Osborn, 1900.192) and 
for the titanotheres (Osborn, 1902.208). 



Comparison oj ike Linnaean and the phyletic systems of taxonomic terms 



Term 


Definitions 


Old system 


New system 


Family 


A contemporaneous group of similar subfamilies 

A smaller contemporaneous group of similar genera__ 

A still smaller contemporaneous group of similar or 
related species. 

A group of related subspecies and geographic va- 
rieties. 

Nothing corresponding to the geologic mutation of 
Waagen. 


Contemporaneous and ancestral phyla that exhibit 
similar family tendencies of evolution. 

A branch composed of one or more phyla which exhibit 
similar generic tendencies of evolution. 

Part of a single phylum of successive species and muta- 
tions exhibiting similar tendencies. 

A series of ascending mutations. 

Geologic mutation (of Waagen); ascending substages 
within a specific phylum. 


Subfamily 

Genus __ 

Species 

Mutation 



Desired harmony of mammalian paleontology and 
zoology. — The methods employed by all zoologists, 
paleontologists, and anthropologists in their observa- 
tion and measurement of the hard parts of mammals 
should be the same. The methods pointed out above, 
first presented by Osborn (1914.412), are founded 
on the comparison in time of geologic ascending 
evolutionary phyla of mammals — such as the rhi- 
noceroses and the titanotheres — with contemporaneous 



geographic series of species, subspecies, and varieties 
that may be grouped within a single genus. What 
applies to the systematic terms used in the classifica- 
tion and description of animals applies with equal 
force to those used for single characters, for it is 
the cumulative sum of evolutionary change in a very 
large number of single characters which makes up 
the mutation of Waagen, the species, or the genus, as 
the case may be. 



22 



TITANOTHEEKS OF ANCIENT AVYOMING, DAKOTA, AND NEBEASKA 



SUMMARY OF DIFFERENCES BETWEEN OLD AND NEW 
SYSTEMS 

To sum up: (1) The Linnaean genus or species is 
defined (statically) by the presence of certain propor- 
tions and by the presence or the absence of certain 
characters, whereas the phyletic genus or species is 
defined (dynamically) by the progressive evolution of 
certain proportions and by the gradual gain or loss of 
certain characters; (2) the Linnaean genus or species 
was clearly distinguished from a related genus or 
species, whereas the phyletic genus or species may 
gradually fade into its ancestor or successor, and the 
point where we make the dividing line is largely arbi- 
trary; (3) consequently the phyletic genus actually 
has a new meaning, but to avoid innovation in nomen- 
clature we apply the phyletic term genus to a number 
of species having a wide range in time and space, in 
the same manner that Linnaeus applied the term 
genus to a number of species having a wide range in 
space only. 

STUDY OF THE EVOLUTION OF SINGLE CHARACTERS 

In the hard parts of living as of extinct animals 
only three kinds of changes are observable — (1) 
changes of proportion, which the author terms "allo- 
metrons"; (2) the appearance of absolutely new char- 
acters, which the author terms "rectigradations"; (3) 
the disappearance or retrogression of characters. 

Changes of proportion. — Changes of proportion 
(allometrons) make up by far the larger part of the 
evolution of the titanotheres, as of that of all other 
mammals. At least 95 per cent of the differences 
between the skeletons of Eotitanops horealis and 
Brontotherium plafyceras are due to changes of pro- 
portion, and not more than 5 per cent to additions of 
absolutely new characters, such as horns. Conse- 
quently a very careful study has been made of allo- 
metry — that is, of the methods of calculating, measur- 
ing, recording, and describing changes of proportion — 
and the result has been the discovery of a number of 
general principles that apply to all mammals, extinct 
and living, including man. Probably also the un- 
discovered causes of changes in proportions are the 
same in all mammals, but their discovery constitutes 
a very difficult problem. (See Chap. XL) In this 
difficult work the paleontologists may be greatly aided 
by the zoologists, especially by very precise field 
observers, such as Allen, Merriam, Miller, Osgood, and 
Sumner. 

Although the mammalogists have demonstrated that 
there is an apparently causal relation, direct or in- 
direct, between certain types of coloration and of size 
(harmonic increase or decrease) and the geographic 
environment, the relation between change of environ- 



ment and changes in proportion (disharmonic) is very 
obscure. It is known that a harmonic increase or 
decrease in size of the entire mammal is correlated 
with certain differences in habitat, often for the 
obvious reason that a favorable environment favors 
development of larger races, whereas an unfavorable 
environment dwarfs growth. It remains to be 
determined, however, whether certain environments 
induce uniformly similar disharmonic changes of pro- 
portion. Anthropologists, for example, have failed 
to establish a definite causal relation between environ- 
ment and the broad-headed (brachycephalic) or the 
long-headed (dolichocephalic) form of the human head. 

The chief contribution that the paleontologist has 
made to this obscure matter is to show that when a 
proportionate change of head form is once established 
in a certain direction there is a tendency to go to 
extremes, so that, for example, extremely long heads 
or extremely broad heads tend to evolve longer or 
broader heads. These evolutionary tendencies are 
illustrated in the titanotheres. 

Adaptive new characters. — The second mode of mam- 
malian evolution — by the appearance of absolutely 
new characters — lies in a field where the paleontologist 
has a great advantage over the zoologist, because in 
a series of fossils a new character (rectigradation) can 
be traced back to its incipient, rudimentary stage, in 
which it is so inconspicuous that it would not attract 
the attention of the zoologist. Many characters that 
eventually may exert a most profound influence on 
the evolution of a race — that may, in fact, dominate 
aU other characters — arise, so far as observed, from 
excessively minute beginnings. These origins of new 
characters are pointed out with great precision in 
Chapters V and VI, in which the evolution of the 
skuU and teeth is described in detail as observed in the 
Eocene and lower Oligocene titanotheres. This very 
precise study of the origin and evolution of similar 
characters in many different lines of descent has led 
to the important discovery that phyla differ less 
through the possession of this or that new character 
than through the different rates of evolution at which 
the same character arises and evolves. In one 
phylum a new character like the horns will arise in 
an early geologic stage and evolve very rapidly, 
whereas in a related phylum it will arise relatively late 
in geologic time and will evolve very slowly. Thus a 
phyletic genus is defined not only by the characters 
which it exhibits but by the rate of the evolution of 
these characters. This principle, again, is observable 
only thi-ough paleontology. 

The origin of new characters, as manifested in dif- 
ferent ways in the members of twelve subfamilies of the 
titanotheres and as indicated by comparison with the 



INTRODUCTION TO MAMMALIAN PALEONTOLOGY 



23 



origin of similar characters in other families of Peris- 
sodactyla, has accordingly been studied with great 
care. 

Retrogressive characters. — The retrogression or disap- 
pearance of characters is illustrated in the history 
of the titanotheres by the features enumerated below. 

1. Reduction of the canine teeth in many later titanotheres. 

2. Reduction and occasional loss of incisors. 

3. Reduction and frequent loss of first lower premolar. 



appearance in North America and western Europe of 
members of nine different families of Perissodactyla, 
the odd-toed ungulates, which were probably all de- 
scended from a common ancestral or stem form which 
lived in Upper Cretaceous time. The probable charac- 
ters of this stem form are fully described in Chapter X, 
where it is shown that the ancestral perissodactyl was 
a comparatively small and simple quadruped not ex- 




Perissodxictyls 



CRETACEOUS 



Peris sodaxtyL Stem, 



Figure 17.- 



-Successive invasion of nine perissodactyl families in Nortli America and western Europe between 
latitudes 40° and 50° 



Th2 chalicotheres (aberrant clawed perissodactyls with affinities to the titanotheres) are regarded as members of a separate superfamily, the 
Chalicotheroidea. Diagonal shading indicates the extent to which each phylum is represented by fossil remains. 



4. Reduction and loss of protoconule and metaconule in 
upper molars. 

5. Reduction of nasals and their coalescence with frontals. 

6. Reduction of the trapezium in later titanotheres. 

PHYLOGENY OF THE NINE TYPICAL FAMILIES OF THE 
PERISSODACTYLA 

The competition of the titanotheres through natural 
selection was naturally closest with other members of 
the order Perissodactyla. As shown in the ordinal 
phylogenetic tree (fig. 17), we observe the successive 



ceeding half a meter in height, and that it was origi- 
nally confined to a definite geographic area, feeding 
ground, and range, very possibly in northern Asia. 
The eight families that appear in North America and 
the paleotheres, which appear only in western Europe, 
were by no means equally distinct from one another. 
They were originally separated from the stem form 
not into nine branches but into five great main 
branches, termed superfamilies, as shown in Figure 17 
and in the accompanying table. 



24 



TITANOTHEEES OF ANCIENT WTfOMING, DAKOTA, AND NEBRASKA 

Phyla of the odd-toed ungulates 



Superfamilies 


Families 


1. Titanotheroidea 


1. Brontotheriidae : The titanotheres, known chiefly in North America and in 




eastern Europe. 


2. Chalicotheroidea 


2. Chalicotheriidae: The chalicotheres, first known in Europe and North America; 




then in Asia. 


3. Hippoidea: Horselike forms _ ._ 


3. Palaeotheriidae : The paleotheres, known in western Europe only. 

4. Equidae: The horses, first known in Europe; then simultaneously in North 

America and Europe; subsequently in Asia, Africa, and South America. 




4. Tapiroidea: Tapir-like forms 


5. Tapiridae: The tapirs, first known in North America; then in Europe and Asia. 




5. Rhinocerotoidea: Rhinoceros-like forms _- 


6. Lophiodontidae: The lophiodonts, known in North America and Europe. 

7. Amynodontidae: The amynodonts, aquatic rhinoceroses; first known in North 

America; then in Europe. 

8. Hyracodontidae : The hyracodonts, cursorial rhinoceroses; upper Eocene and 

OUgocene of North America only, so far as known. 

9. Rhinocerotidae : The rhinoceroses, the typical rhinoceroses; first known in 

North America and Europe; then in Asia and Africa. 



In North America the horses (Eohippus) were the 
first perissodactyls to arrive. They were followed by 
the tapirs (Systemodon) , which in turn were succeeded 
by the lophiodonts {Heptodon). It is possible that 
ancestral titanotheres were living iu northern parts 
of the American continent, but apparently thej^ did 
not reach the region near the fortieth parallel until 
it had become well populated with horses, tapirs, and 
lophiodonts. By middle Eocene time three more 
families had appeared — the paleotheres, in Europe 
only; the rhinoceros-like amynodonts (semiaquatic 
forms), which first appear in North America and 
subsequently in Europe; and the cursorial rhinocer- 
oses known as hyracodonts (Hyrachyus), which appear 
in North America only and preceded the amynodonts. 
Toward the beginning of upper Eocene time there 
first appear in North America, as well as in Europe, 
ancestors (Eomoropus) of the chalicotheres, animals 
closely related in tooth structure to the titanotheres, 
which were separated into a distinct order (Ancylo- 
poda) by Cope and are here regarded as forms some- 
what parallel to the Titanotheroidea. 

WIDE GEOGRAPHIC DISTRIBUTION OF THE 
PERISSODACTYLA 

We are first struck with the remarkably wide 
holarctic distribution of the perissodactyls in Eocene 
and lower Oligocene time, a fact which points to 
facihty of migration over the whole Northern Hemis- 
phere. Only one family, the paleotheres, is exclu- 
sively European, and one other, the hyracodonts, is, 
so far as known, exclusively North American. The 



titanotheres were formerly beheved to be exclusively 
North American, but two forms have been found in 
eastern Europe, which correspond very closely with 
the titanotheres of upper Eocene age from the Uinta 
Basin in northern Utah. 

Members of all the other perissodactyl families — 
the chalicotheres, tapirs, lophiodonts, amynodonts, 
and rhinoceroses — probably ranged freely to and fro 
over the great northern continent of Em-asia and 
North America combined, the geographic region 
known as Holarctica. 

The second important fact regarding the Peris- 
sodactyla is that, although the environment dining 
middle and upper Eocene time, after the extinction 
of the archaic imgulates — the Condylarthra and 
Amblypoda — was especially favorable to the existence 
of the Perissodactyla, this order reached its maxi- 
mum expansion in the lower Ohgocene epoch, when 
all the nine families were existing and apparently 
flourishing at the same time. It would appear that 
in upper Eocene and lower Oligocene time Holarctica 
was dominated by perissodactyls. This period was 
immediately followed by a period when either the 
environment was adverse to the existence of the peris- 
sodactyls or competition with other types of imgu- 
lates was disastrous to them, because at or before the 
end of the lower Oligocene epoch five perissodactyl 
families suddenly disappeared — the titanotheres, paleo- 
theres, lophiodonts, amynodonts, and hyracodonts. 
The aberrant chalicotheres, apparently through retreat 
to forested regions, survived in Europe and probably 
also in North America until the Pliocene epoch. 



INTRODUCTION TO MAMMALIAN PALEONTOLOGY 



25 



.^"^- 





Figure 18. — Outlines of the body form of the perissodactyls, drawn to the same scale 

The largest known member o( each family is selected for comparison. The smallest known stem forms of each family are illustrated 

in Chapter X. The animals are grouped according to their natural relationships, as indicated especially by the pattern of the 

molar teeth, as follows: 
Bhinocerotoid group: A, Mclamynodon; family Amynodontidae; graviportal; aquatic; lower Oligocene. B, Hyracodon, family 

Hyracodontidae; cursorial; middle Oligocene. C, Ceratotherium simum; living white rhinoceros; family Rhinocerotidae; 

graviportal. 
Tapiroid group: D, Tapirus terresiris; existing tapir; family Tapiridae; mediportal. 
Hippoid group: E, Palaeotherium; family Palaeotheriidae; lower Eocene; mediportal. F, Equus pTzewalskii; existing horse; family 

Equidae; cursorial. 
Chalicotheroid group: (?, MoTopus; family Chalicotheriidae; clawed perissodactyl; lower Miocene. 
Titanotheroid group: B, Brontotherium platyceras; family Brontotheriidae; graviportal; lower Oligocene. 

101959— 29— VOL 1 4 



TITAXOTHEEES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 




INTRODUCTION TO MAMMALIAN PALEONTOLOGY 



27 



Thus out of the nine original famiUes of the great 
order of Perissodactyla only three — the horses, tapirs, 
and rhinoceroses — have survived to the present time, 
and these during the glacial epoch were greatly 
reduced both in numbers and in geographic dis- 
tribution. 

The consideration of these facts raises the whole 
problem of the origin and adaptive radiation of the 
perissodactyls (see Chap. X) and the general problem 
of the causes of the extinction of the perissodactyls 
and of other quadrupeds (see Chap. XI). 



adaptive origin of new characters. The moment of 
origin of each new character is a very important 
moment in the history of that character. Does each 
new character arise fortuitously at this point or that, in 
an adaptive or inadaptive condition, or does each new 
character arise in a mechanically adaptive condition, 
although this condition may be merely incipient? 

The biologic purpose of the long and dry descrip- 
tions and tables of measurements given in Chapters 
V, VI, and VII of this monograph is to direct obser- 
vation continuously to this problem of the origin of 




Figure 20. — Periods of expansion and extinction of the perissodactyls and contemporary forms 

Showing that the expansion of the perissodactyls was coincident with the extinction of the archaic Condylarthra and Amblypoda and that the 
extinction of many perissodactyls was coincident with the expansion and adaptive radiation of the artiodactyls. 



CAUSES OF EVOLUTION 

There can be no doubt as to the survival value of 
certain finished types of tooth structure and Hmb 
structure (see pp. 880-881), a principle first formulated 
by the distinguished Russian paleontologist Kova- 
levsky (1873.1). Two important questions that the 
reader must keep in mind in considering the origin 
of innumerable new characters are (1) whether 
there is evidence of chance origins and chance rudi- 
ments of certain types of structure possessing suffi- 
cient survival value to establish themselves through 
the principle of the survival of the fittest, or natural 
selection; or (2) whether there is some other ortho- 
genetic principle at work causing the definite and 



new characters. Our general conclusions concerning 
these two questions are presented in Chapter XI. 

ADAPTIVE EVOLUTION AND OVEEEVOLUTION OF THE 
FORM OF SKULL, TOOTH, AND FOOT 

Whatever may be the causes of evolution its re- 
sults are definite. The visible evolution of all the 
hard parts of the body in herbivorous animals is 
originally mechanical and manifests general adapta- 
tion to two broad groups of purposes: 

1. Prehension of food (lips, teeth, and jaws); com- 
minution of food (teeth and jaws); conservation and 
transportation of stored food energy (body and limbs). 
These purposes involve all the mechanical changes of 
structure of skull and tooth. 



28 



TITANOTHEEES OP ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



2. Motion and locomotion; migration in search of 
food and to escape enemies; adaptation to perform 
the act of reproduction and to protect the young. 
These purposes involve all the mechanical changes of 
the structure of limb and body. 

The operation of the principle that, under the domi- 
nance of these modes of mechanical adaptation each 
organ, structure, and character is adaptively evolved 
for some special service to the organism is not invariably 
evident in respect to all changes in the proportion of 
characters. Certain characters of proportion, such as 
extreme broad-headedness or extreme long-headedness, 
seem to interfere with adaptation; they appear to be 
carried so far in one direction as to render the animal 
less adapted to survive than its less specialized ances- 
tral forms. In other words, certain tendencies of 
evolution may carry a phylum beyond its require- 
ments in adaptation. 

Aside from this question of the different degrees of 
survival or actual elimination value of certain tend- 
encies of evolution, there can be little doubt that in 
its origin and development each character, sooner or 
later, responds and reacts independently to the con- 
ditions of the environment, quite apart from the 
question as to the causes of such response. The teeth 
react to the kinds of food; the feet and limbs to the 
kinds of soil. 

The principles of the divergence of quadrupeds 
from each other in their independent adaptations in 
the skull, teeth, limbs, and feet are fully discussed 
elsewhere (see p. 123) in the treatment of the principle 
of adaptive radiation. Though they may have lived 
apparently in the same region and have been fossilized 
side by side in the same sediments, all distinct species 
of quadrupeds have locally different habits and habi- 
tats. The structure of the skull, jaws, and teeth re- 
sponds to their habits and tastes ; the structure of the 
feet and limbs responds to their habitats — the nature 
of the ground, etc. 

PHYLETIC DIVERGENCE IN THE EVOIUTION OF NEW 
PROPORTIONS IN HORSES AND IN TITANOTHERES 

All the families of an order of Perissodactyla start 
their career from a similarly proportioned ancestral 
stem form such as that described in Chapter X (p. 760) 
as the stem perissodactyl. Starting with the same 
complement of characters, divergence in proportions 
separates the families of perissodactyls more and more 
widely from one another. In the Equidae (horses), 
for example, the head form of the earliest known 
ancestor (EoMppus) is very similar to that of the 
earliest known ancestor (Eotitanops) of the family 
Brontotheridae. In both these primitive skulls the 
orbit is near the center of the head, and in the later 
forms it apparently moves backward or forward, but 
what really happens is that the skull is elongated in 
front of the orbit in the horse and is elongated behind 
the orbit in the titanothere. (See fig. 21.) 



A comparison of the forms shown in Figure 21 
with those shown in the following figures will demon- 
strate the marked similarity of the lower Eocene 
forms and the very wide divergence of the modern 
forms. The skulls of the ancestral tapir, horse, and 
titanothere {Systemodon, Eohippus, and Eotitanops) 
are in many ways much alike, the chief differences 
consisting in (1) the details of the characteristics of 
the dentition, (2) the relative position of the orbits, 
(3) the depth of the head through the back part 
of the lower jaw, and (4) the size of the muzzle. 
The primitive titanothere prophetically suggests the 
titanothere characters in the relatively heavy muzzle 
and stout lower jaw. The primitive horse Eohippus 
prophetically suggests the modern horse in the taper- 
ing form of the slender lower jaw and in the general 
contour of the skull, except that the eye is placed near 
the middle of the head, as in other primitive perisso- 
dactyls. The primitive perissodactyl Systemodon, 
regarded by Osborn as an ancestral tapiroid, had a 
somewhat longer, more pointed muzzle but was 
otherwise very similar to the contemporary horse 
Eohippus. 

These differences of proportion between the facial 
region in front of the orbit and the cranial region 
behind the orbit are partly correlated in adaptation 
to the elongation (hypsodonty) of the crowns of the 
grinding teeth. In the horse and in most of the rumi- 
nant artiodactyls the face is elongated to accommodate 
the vertically elongated (hypsodont) grinding teeth. 
In the titanotheres, which are browsing animals, 
and in the browsing rhinoceroses of India and of 
Africa the orbit is directly above the grinding teeth 
and the cranium is slightly elongated, as shown in 
Figure 22. Thus it may be stated as a general prin- 
ciple of skull evolution that in browsing ungulates 
the cranium tends to be elongated and the face tends 
to be abbreviated, whereas in grazing ungulates, 
like the white rhinoceros of Africa, in which the grind- 
ing teeth are elongated, the face is elongated, and 
the cranium is abbreviated. 

It follows that these respective proportions of the 
region in front and back of the eyes are adaptive; 
they are part of the general correlation of skull 
proportions with the functions of the grinding teeth 
employed in the prehension of food, as provided for 
chiefly in the shape of the upper and lower lips, 
which are obtrusible and flexible both in the browsing 
rhinoceroses and in the grazing horse, which occasion- 
ally browses. When the horse is browsing it extends 
its lips very much in the manner of the browsing 
rhinoceros, except that in the rhinoceros the independ- 
ent motion and the pointing of the upper lip are more 
extreme. In the grazing white rhinoceros the upper 
lip is extremely broad and square. The animal 
subsists largely on grasses, which it crops with its 
square lips, exactly in the manner that the horse 



INTEODUCTION^ TO MAMMALIAN PALEONTOLOGY 



29 



crops grass with its lips and front teeth. In all 
the rhinoceroses cropping front teeth are atrophied, 
the four pairs of incisors and the canines being 
reduced to a single large pair on either side and being 
thus analogous to those of certain titanotheres. 

From these comparisons we deduce the structure 
of the mouth parts in the titanotheres as restored by 
Gregory. (See p. 704.) We also deduce the various 
adaptations to the browsing and grazing habit re- 
spectively in the different genera of titanotheres, for 
undoubtedly some were purely browsers and others 



of the face, with a relatively short skull, and with a 
very powerful neck, a feature that is also especially 
characteristic of the titanotheres. 

Thus there is a general resemblance between the 
side profile of Brontotherium platyceras and that of the 
Indian rhinoceros, which is due to analogous mechan- 
ical evolution, through the principles known as homo- 
plasy, parallelism, or convergence. The titanotheres 
pass through a long lower and middle Eocene phase of 
tapir-like analogies, but when, in middle Eocene time, 
horns begin to appear the head region develops 




Figure 21. — Phyletio divergence in the evolution of new proportions in horses and in titanotheres 

Lower Eocene ancestral horse Eohippus (A) and lower Eocene ancestral titanothere Eotilaijops (C) (both with the orbit in the same relative 
position on the skull) compared with a modern horse (B) with face extended in front of the orbit and a titanothere of the latest stage (D) with 
slvuU extended behind the orbit. Thus two very similar heads (A, C) become increasingly dissimilar (B, D). Scales various. 



tended toward grazing. Thus the orbits, the face, 
the grinding teeth, the front teeth, the lips, and the 
bones supporting these structures are respectively 
transformed in adaptation to the function of prehen- 
sion and to browsing or grazing habits. The front 
part of the skull of the rhinoceros, with its terminal 
dermal horn, is comparable to that of the large-horned 
titanotheres, with their terminal bony horns. It will 
be observed that the entire front part of the head of 
the rhinoceros, in adaptation to the great strain of 
the horn used as a weapon of offense and defense, is 
correlated with a flat or a concave line along the top 



rhinoceros-like analogies. Similar analogous phases 
also occur to a greater or less extent in the feet of the 
rhinoceros and the titanothere. 

On comparing the heads of the types of perisso- 
dactyls, ancient and modern, we observe that different 
modes of feeding and of offense and defense guide the 
dominant adaptations in evolution. The evolution 
operates under the principles of anatomical correla- 
tion and compensation, gain or loss in one part being 
mechanically balanced by gains and losses in every 
other part. This process includes the principle of 
physiologic compensation, whereby loss of function in 



30 



TITANOTHERES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 





Figure 22. — Contours of the head and of parts of the mouth in browsing and grazing perissodactyls 



A, Asiatic rhinoceros {Rhinoceros ' inikusi, chiefly a browser; B, black rhinoceros ot Africa (R. (.Opsiceros) bicornis), chiefly a browser; C, white 
rhinoceros of Africa (R. (Ccraioiherium) simum), chiefly a grazer; D, domestic horse {Equns caballus), chiefly a grazer; E, American 
tapir ( Tapirus tenestTis), a browser. 



' The generic terminology of the rhinoceroses is not yet fully agreed upon by zoologists. The family tree, like that of the titanotheres, 
is polyphyletic. 



INTBODUCTION TO MAMMALIAN" PALEONTOLOGY 



31 




Figure 23. — Heads of lower Eocene and modern jjerissodactyls, showing changes of proportion and of the lip 

structure 

Based on materials in the American Museum of Natural History. Scales various. A, Head of the lower Eocene tapiroid Sijstemodon, very sim- 
ilar to that of Eohippus and of Lambdotherium; B, head of middle Eocene tapir Hdaleies, in which a prehensile upper lip first appears; 
C, head of the modern tapir Tapirus, whose prehensile upper lip forms a short proboscis; D, head of middle Eocene cursorial rhinoceros 
Hyrachyus, still of primitive proportions; E, head of existing white rhinoceros {EhiTicccrcs ( Caalothcriuw) simum) with extremely 
broad, grazing type of lip structure. 



32 



TITANOTHERES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



one part is taken up by some other part. For exam- 
ple, the loss of the function of the incisors in the pre- 
hension of food is compensated for by changes in the 
form and function of the lips. 



upper lip like that of the tapir necessitates space for 
the superior retractor muscles, which curl the lip 
upward and backward. An example of the results of 
the evoUition of the lower jaw may be seen by compar- 




FiGtTRE 24. — Restorations to the same scale of the heads of some of the principal t3'pes of titanotheres 

Drawn by Charles E. Knight, after models made by him under the author's direction. About one-seyenteenth natural size. A, 
Brontops roiustus Marsh, oblique yiew, middle Titanotherium zone; B, Menodus giganleus, upper Titanotherium zone; C,Megacerops 
copei Cope, partly oblique side yiew, summit of the Titanotherium zone of Colorado; D, Broniotherium platyceras Scott and Osborn, 
the final stage in the eyolution of the horns of the titanotheres, summit of the Titanotherium zone of South Dakota; E, Protitano- 
iherium sp., summit of the Eocene. 



With the evolution of the lips the structure of the 
anterior parts of both the upper and lower jaws, of 
the anterior teeth, and the anterior nasal openings is 
closely correlated. The development of a prehensile 



ing Eotitanops gregoryi and Brontotherium {medium) 
gigas, the whole jaw of the former hardly exceeding in 
length a single posterior grinding tooth of the latter. 
(See fig. 25.) 



INTRODUCTION TO MAMMALIAN PALEONTOLOGY 



33 



EVOIUTION OF THE LIMBS AND FEET OF THE 
TITANOTHERES 

The feet of the titanotheres, like their skulls, pass 
through a lower Eocene tapir-like phase, which is 
followed by a middle and upper Eocene rhinoceros- 
like phase and finally they attain a structure similar 
to that of the rhinoceroses, as shown in Figure 26, 
except that all the titanotheres, like the existing 
tapirs, retained four distinct and functional digits 
in the fore foot. 

The fore foot of the tapir resembles the fore foot of 
the lower Eocene titanothere except that in the latter 
D. II, III, IV, V were all of nearly equal size, as 
shown in the diagram (B). This is known as the 
mediportal stage, for it is adapted to carrying a 
moderate amount of weight. The 
foot of the rhinoceros (C, C, C) 
is like that of the upper Eocene 
and lower Oligocene titanotheres 
except that in these there were 
four weight-bearing digits instead 
of three. This is known as the 
graviportal type of foot, in which 
a large cushion pad is developed 
at the back to relieve the shock 
of impact, and the end phalanges 
of the digits are incased in the 
horny sheath in front. In the 
tapir and rhinoceros the main 
weight passes directly through the 
center of the median phalanx 
(D. Ill), but in the tetradactyl 
titanotheres the main weight 
passes between D. Ill and D. IV. 
The concentration of the weight 
on the central digit of the horse 
and its resultant monodactylism, 
correlated with the expansion of the horny hoof and 
the contraction of the pad, is part of the evolution 
of a cursorial type of foot, which presents the widest 
contrast to the graviportal type. 

In addition to comparing the head structure it was 
found necessary to compare the foot and limb struc- 
ture of the titanotheres with that of all the other 
perissodactyls — not only the bony parts but the 
musculature. The work done on the musculature led 
to an exhaustive study of all that is known of the 
muscular anatomy of the members of the three exist- 
ing families of perissodactyls. This study, which was 
directed by William K. Gregory, formed the basis of 
the restoration of the muscular anatomy of the giant 
Brontops rolustus presented in Chapter VIII (pp. 722, 
723). This restoration of an extinct animal is the first 
that has been based upon exact comparative study. It 



presents the titanothere as a superb example of the 
graviportal type of musculature and skeleton, sur- 
passed only by the existing elephants. 

The study of the structure of the foot led to a special 
investigation of the proportions of the limb bones in 
the ungulates. This investigation, directed by Osborn 
and cooperated in by Gregory, resulted in the striking 
discovery that the proportions of the upper and lower 
segments of the limbs and of the feet are invariably 
adjusted, first, to the weight that the limb must carry, 
and second, to speed of locomotion. These propor- 
tions are evolved, quite irrespective of ancestry, in 
adaptation to different modes of progression. Thus 
similar proportions of limb segments are observed not 
only in all mammals but in reptiles as well. A study, 



'^^-^Ga^;^^' I 




Figure 25. — Lower jaws of the first and the last of the titanotheres 

One-sixth natural size. A, EotUanops gregoryi, a small-jawed species from the Wind River formation (lower 
Eocene); B, BrontotTierium medium, from Chadron C level of Chadron formation (lower Oligocene). 

therefore, which was designed to disclose the habits of 
the titanotheres led to a thorough investigation of the 
principles of limb evolution in all the hoofed mam- 
mals in adaptation to various modes of locomotion 
and to various loads. This special study forms the 
subject of Chapter IX, in which acknowledgment is 
made to previous investigators. 

Not only the proportions of the upper and lower 
segments of the limbs but all the bones of the shoulder 
and pelvic girdles are gradually transformed from the 
subcursorial stages of Lambdoiherium and Eotitanops 
through the mediportal tapir-like stages to the gravi- 
portal stages of the ponderous Oligocene titanotheres. 
This transformation is continuous, not sudden; it is 
brought about gradually by the simultaneous and 
correlated modification of all the bones and muscles 
involved in locomotion. Function (habit) is evi- 



34 



TITANOTHEKES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



dently far more potent than ancestry (heredity) in 
the determination of general form, yet in comparing 
the limbs of all the members of the different perisso- 
dactyl families with one another we can generally, by 
some family characteristic inherited from the ancestral 
stem form, distinguish the tapir type, the rhinoceros 
type, the titanothere type, etc. In the limbs, as in 
the skull and teeth, the titanothere, rhinoceros, or 
tapu' ancestry respectively seems to keep the evolution 
of proportion and form within certain limits, so that, 
for example, the resemblance between the graviportal 
scapula of the titanothere and that of the rhinoceros, 
though it may be very close and deceptive, is never 
quite complete. The stages of muscular and skeletal 



the origin of new characters (rectigradations). In 
this problem of the origin of new characters in the 
titanotheres we have two principal subjects of study, 
namely, the origin of horns on the skull and the 
origin of cusps on the grinding teeth. 

In the evolution of the grinding teeth the titano- 
theres are very conservative; in them few new cusp 
elements originate, though several of the old cusp 
elements disappear. These animals thus present a 
striking contrast to the horses in the evolution of the 
grinding teeth, for in the horses a large number of 
new cusp elements are successively added. Yet the 
grinding tooth of the earliest titanotheres {Lambdo- 
therium and Eotitanops) is in general similar to that 




Figure 26. — Structure of the feet in extinct and living odd-toed ungulates (perissodactj-ls) 

A, Sole of the left fore foot of a tapir (Taphus ierresiris), showing the tripod-like arrangement of digits II, III, and IV, and 
the reduced condition of V; B, sole of the left fore foot of an Eocene titanothere (Mesatirkinus petersoni), restoration based 
on Princeton Museum specimen No. 10013; C, sole of the fore foot of a rhinoceros, showing the enlarged hoofs of the 
three digits (II, III, IV) ; C^, side view of same; C^ longitudinal section of same; D', sole of the fore foot of a horse, show- 
ing the expanded nail; D^, longitudinal section of same. The central pad (/} in A, B, and C is homologous with the 
relatively reduced pad or frog (/) in the foot of the horse (DO- All but B after Eber. 



evolution, arranged from latest to earliest, are as 
follows : 

4. Graviportal; ponderous, relatively slow-moving types, 
such as Brontolherium, Rhinoceros {C eratotherium) simum. 

.3. Mediportal; of moderate weight and speed, such as 
Limnohyops, Tapirus. 

2. Subcursorial; of light weight and relatively swift move- 
ments, such as Eolilanops of the lower Eocene. 

1. Cursorial; swift moving, Ught frame, such as Lanibdo- 
therium of the lower Eocene. 

ORIGIN OF NEW CHARACTERS AS DISTINGUISHED FROM 
CHANGES IN PROPORTION 

The continuous gradual changes of proportion in 
the head, trunk, and limbs (allometrons), as already 
outlined, present a problem distinct from that of 



of the earliest horses {Eohippus). In these lower 
Eocene contemporary mammals the grinding teeth 
are the same, cusp for cusp. In the horse all these 
cusp elements are preserved and utilized, and the 
highest degree of mechanical adaptation to the graz- 
ing habit is gradually evolved; in the titanotheres 
the browsing habit is generally conserved, and there 
is little marked increase of mechanical adaptation; 
in fact, mechanical inadaptation or imperfection of 
the grinders may have been one of the probable 
causes of the extinction of the titanotheres at a time 
when the conditions favorable to grazing gradually 
replaced those favorable to browsing. 

The adaptive radiation of the grinding teeth in the 
several families of the Perissodactyla from somewhat 



INTRODUCTION TO MAMMALIAN PALEONTOLOGY 



35 



similar ancestral forms is shown in Figure 29. The 
earliest members of every family had low-crowned 
(brachyodont) molar teeth, of relatively simple 
pattern, composed of six principal cusps ranged in 
three pairs — an external pair, the paracone and meta- 
cone; an intermediate pair, the protoconule and 
metaconule; and an internal pair, the protocone 
and hypocone. 

In the titanotheres, chalicotheres, paleotheres, 
and horses the internal pair of cusps assume the 
conical, rounded shape (bunoid), whereas the 
two external cusps assume the double crescentic 
shape (selenoid), together forming a W, hence 
this type of tooth is termed bunoselenodont. 
These bunoselenodonts apparently formed origi- 
nally a natural group from which the horses 
(Eohippus), the titanotheres (Eotitanops), and 
the chalicotheres (Eomoropus) gradually diverged 
very early in Eocene time. This is shown in 
Figure 30. 

Another group of perissodactyls is the bunolo- 
phodonts, which includes the tapirs and lophio- 
donts, in which the internal and external pairs 
of cusps alike assume an elongate, crested, or 
lophoid pattern. This group has two main 
branches, the tapirs and the lophiodonts. The 
tapirs as forest-seeking animals escaped fossiliza- 
tion and are rarely found; only isolated remains 
of them have been found in Europe and America; 
yet they constituted one of the most persistent 
of all the perissodactyl phyla. The lophiodonts 
were tapir-like animals, in which the posterior 
outer molar cusps were flattened and thus are 
intermediate in shape between the tapir tooth 
and the rhinoceros tooth. These animals doubt- 
less had a wide expansion in the luxuriant 
forests of Eocene France, and they attained 
very great size just before their extinction, 
which occurred contemporaneously with the 
extinction of the titanotheres in America — that 
is, in lower Oligocene time. Only one branch of 
the lophiodonts, the swift-footed Helaletinae, 
reached North America in lower Eocene time, 
soon after the arrival of the tapirs (Systemodon) 
and the horses (EoTiippus). 

The grinding tooth of the rhinoceroses is lopho- 
dont — that is, all the cusps are turned into elon- 
gate crests, of lophoid type, and the posterior 
outer cusps of the upper grinding teeth are 
elongated as well as flattened, producing an asym- 
metry of the cusps of the outer wall (ectoloph) of the 
crown. A grinding tooth of this kind is far more 
effective than that of the bunoselenodont titano- 
theres or of the bunolophodont tapirs. Such a tooth 
is a very efficient cutting instrument for an animal 
of either the browsing or the grazing habit. It is 
also capable of elongation (hypsodonty), and in 



two subfamilies of the rhinoceroses, the white rhinoc- 
eroses and the elasmotheres, the grinding teeth 
become hypsodont, greatly increasing the longevity 
and consequent reproductive power of each indi- 
vidual. 




Figure 27. — Restorations of nine species of titanotheres from the 

lower, middle, and upper Eocene and the lower Oligocene 

Drawn by Mrs. E. M. Fulda. About one-fiftietb natural size. 

The rhinoceroses gave off at least twelve distinct 
branches (phyla) and were thus more plastic in adapta- 
tion than the titanotheres. These branches became 
adapted to every habitat, aquatic as well as terrestrial, 
to every mode of locomotion — cursorial, mediportal, 
and graviportal — and to every kind of feeding — brows- 
ing and grazing. Like the titanotheres some of the 
rhinoceroses passed from the mediportal to the gravi- 



36 



TITANOTHBEES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



portal stage of locomotion. In doing so they acquired 
an entirely new set of proportions, which are shown 
in detail in Chapter IX. 

The teeth form the readiest means of distinguishing 
different branches and subbranches of the Perisso- 
dactyla from one another. The ancestral pattern, 
whether bunoselenodont or lophodont, is so marked 
and persistent that it is only partly modified through 



their evolution, and these give off one mediportal, 
forest-living branch, HypoMppus. The horses are 
paralleled by cursorial or subcursorial titanotheres, 
such as LamhdotJierium, by cursorial paleotheres 
{Palaeotherium and Paloplotherium) , mistakenly sup- 
posed by Huxley to be the ancestors of the horses, by 
two cursorial branches of the lophiodonts (the helale- 
tids and the chasmotheres), and by two cursorial 




Figure 28. — Evolution of the skeleton of the titanotheres 

A, First stage (subcursorial), lower Eocene, Lambdotlierium popoagicum; B, second stage (subcursorial), lower Eocene, Eotitanops 
horealis; O, intermediate stage (mediportal), middle Eocene, Palaeosyops leidyi; D, final stage (graviportal), lower Oligocene, Brontops 
Tobustus. From one twenty-eighth to one-thirtieth natural size. 

branches of the rhinoceroses (the triplopodines and the 
hyracodonts) . It is shown elsewhere (see Chap. IX) 
how the cursorial habit, independently assumed in 
each of these subfamilies, modified not only the limbs 
but the skull and the entire skeleton into analogous 
forms that simulate real affinity. In Figure 32 all 
these cursorial branches, independently evolving in 



analogous adaptation. The manner in which the 
skeleton and limbs similarly became adapted inde- 
pendently to various modes of locomotion and thus 
assumed analogous forms and proportions is no less 
remarkable than the independent adaptation of the 
teeth to similar kinds of food. 

Of the nine typical perissodactyl families the horses 
alone are cursorial through the entire period of 



INTRODUCTION TO MAMMALIAN PALEONTOLOGY 



37 





ProtitartotfieriuTrh emargrinaticm. 

Upper Eocene ( upper Uinta) 



Manteocems manteoceras 

Middle Eocene (upperBridger) 



Zimnohyops priscus 

M/ddle Eocene (lowerBridger) 





Eotitanops borecdzs 

Lower Eocene (Wind River) 

Figure 29. — Evolution of the skull and molar teeth in the titanotheres 

In EoUtanops the facial part of the skull is longer than the brain case (cranium). In Brontotherium the face is very short and the brain 
ease is very long. The horn swellings (H) first appear in Manteoceras and become very prominent in the succeeding stages. The top of 
the skull becomes deeply concave. The outer wall and the V-shaped cusps of the upper molar teeth (paracone, metacone) become 
very deep, while the inner cusps (protocone, hypocone) retain their low, conical form. The lower molars retain the W-shaped crown 
throughout, which increases considerably in depth. 



38 



TITANOTHEEES OP ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



different perissodactyl families, are indicated by dif- 
ferent kinds of shading. 

Forest-living habits among perissodactyls are some- 
what more rare, especially the extreme adaptation to 
forest living, consisting of relatively slow locomotion 
and marked special adaptation to browsing on the 
leaves of trees. Types that are more or less fully 



Aquatic branches of the perissodactyls are also more 
or less readily distinguishable. Among the titano- 
theres we have a group of swamp or river living forms, 
with short limbs and spreading feet, whose remains 
are preserved in many river-channel sandstones, 
namely, the genera DolichorMnus and Metarhinus, 
which are clearly distinguished from all other titano- 




FiGURE 30. — Adaptive radiation in tlie evolution of the upper molar teeth in the odd-toed hoofed mammals 

(perissodactyls) 

After W. D. Matthew. The earliest members of each family had low-crowned (brachyodont) teeth, of relatively simple pattern. In the 
titanotheres and paleotheres the internal cusps remain low and the two outer main cusps form a W. In the horses (hypsodont) the whole sur- 
face of the crown is thrown into complex crests and ridges and the crown becomes very long. In the tapirs (brachyodont) the molar crown 
takes the form of two sharp cross crests. A somewhat similar pattern is seen in the lophiodonts, e.Kcept that in this family (brachyodont) 
the outer cusps form an irregular outer wall. In the rhinoceroses (brachyodont to hypsodont) the outer wall (ectoloph) becomes very much 
flattened, elongate, and oblique, and the cross crests also become oblique. 



adapted to forest living are represented, we believe, 
among the chalicotheres, among certain forest-living 
horses {HypoJiippus), and among certain forest-living 
tapirs (Tapirus terrestris), all relatively slow in move- 
ment and all without conspicuous weapons of offense 
or defense, except that the chalicotheres, such as 
Moropus, are provided with heavy claws. 



theres by their apparent adaptations to river-border 
or aquatic life. Certain tapirs frequent river borders 
and swim freely for long distances, but they do not 
acquire distinctive aquatic adaptations. Among the 
rhinoceroses the pronounced aquatic division is the 
amynodonts, which have marked aquatic features 
about the head, simulating those of the hippopotami. 



INTRODUCTION TO MAMMALIAN PALEONTOLOGY 

The great family tree of the perissodactyls may be interpreted as shown below. 

Family tree of the perissodactyls 



39 



Primitive ancestors 


Ancient branches 


Families, extinct and living 




A. Bunoselenodont branch of basal Eocene 
time: inner cusps bunoid, conical; outer 
cusps selenoid, crescentic. 


1. Titanotheres. 

2. Chalicotheres. 

3. Paleotheres. 

4. Horses. 


Perissodactyls of Upper Cretaceous 
and basal Eocene time: four digits on 
the fore foot, three on the hind foot; 
six rounded cusps on the upper grind- 


B. Bunolophodont branch of basal Eocene 
time; inner cusps crested, outer cusps 
symmetrically crested and more or less 
flattened. 


5. Tapirs. 

6. Lophiodonts, mediportal and graviportal; 
confined to Europe. Helaletids, cursorial 
lophodonts; reaching America. 


ing teeth. 


C. Lophodont branch of upper Eocene 
time; inner cusps crested, outer cusps 
asymmetrical, greatly flattened. 


7. Amynodonts (aquatic). 

8. Hyraoodontidae (cursorial and medi- 
portal) . 

9. True rhinoceroses (mediportal and gravi- 
portal), variously adapted to browsing 
and grazing; distinguished by variations 
in the evolution of the horns. 



The mediportal structure, in which the skeleton 
and limbs are adapted to moderate speed and weight, 
embraces those intermediate stages in several different 
families in which there was moderate body weight 
and moderate speed, as in the tapirs. In the tapirs 
this is the last term of evolution, but in the titanotheres 
and in many rhinoceroses the mediportal stage is 
simply a gateway to the graviportal stage, in which 
the proportions of the limbs and trunk are adapted to 
weight bearing, more or less rapid progression, and 
active offense and defense. 

The interpretation of these phenomena of analogous, 
parallel, and convergent evolution under the princi- 
ple of adaptive radiation, presented on pages 121-127, 
simplifies the problem of the anatomy of the group 
as a whole as well as of the several adaptations 
seen in the skull, skeleton, limbs, and teeth. Each 
perissodactyl family appears to exhibit an innate 
potentiality to evolve in many different directions 
and thus to meet new conditions of life. In this 
sense each family is plastic. Here we are not wit- 
nessing the direct action of the environment: we 
are witnessing the direct response of the organism, 
through largely unknown causes, to develop its poten- 
tial heredity characters along certain new lines. If 
the supply of new potential characters is exhausted, 
if a mechanical stage is reached out of which no addi- 
tional stages can be developed, the animal will tend 
to become extinct unless it can retire to the recesses 
of the forests, as did the chalicotheres, and thus escape 
a struggle for existence in competition with more 
plastic forms, better adapted to the grazing life. The 
interpretation of these processes, however, has been 



the most difficult and baffling of all the problems that 
have arisen in the research made for this monograph. 
The interpretation of the modes and causes of the 
origin and evolution of new characters and of new 
proportions in response to new conditions of life 
(see pp. 834-849) is extremely difficult. Explanations 
that at first seem obvious appear on close analysis 
not to be explanations at all. As this monograph is 
the most exhaustive and most detailed study thus far 
made of any group of mammals it seems important 
to show the bearing of all the observations on each of 




Figure 31. — Three types of teeth of members 
of nine typical families of perissodactyls 

Bunoselenodont (A), bunolophodont (B), and lophodont (C) 
types of teeth displayed in the short-crowned Cbrachyodont) 
stage. 

the current theories of evolution. It appears that, 
as is fully set forth in Chapter XI, we are still very 
far from even a preliminary understanding of the 
causes of many of the processes of mammalian 
evolution. 

VELOCITY IN THE DEVELOPMENT OF CHARACTERS AND IN 
PHYLOGENY 

The earliest explanations of evolution were purely 
mechanical; we are now passing through a phase of 



40 



TITANOTHEEES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



chemical explanations; but it appears that we may 
be led to the adoption of certain physical conceptions 
and the use of certain physical terms (Osborn, 1917. 
462) for what has been described above as the rate of 
evolution of certain characters as distinguishing 
genera. For the term "rate" we will substitute the 
term "velocity." 

Ontogenetic velocity. — The velocity of the evolution 
of certain characters in embryonic development — in 
fact, throughout the whole course of individual 
development — has long been a very familiar feature 
of adaptation. From the embryo onward a char- 



isms, and we shall see that the most plausible explana- 
tion of it thus far offered is the theory of natural selec- 
tion proposed by Darwin. 

Phylogenetic velocity. — Another kind of velocity, 
however, may be noted in the perissodactyls and may 
be measured and calculated with great precision in 
the numerous phyla of titanotheres here considered. 
This velocity may be called phylogenetic velocity. Its 
postulation rests upon the fact that a given character 
may evolve much more rapidly in the members of one 
phylum than in the members of a related phylum, al- 
though the environment of both phyla may be the 




FiGiTRB 32. — The family tree of the perissodactyls, showing adaptive radiation of the nine families and thirty- 
five subfamilies 

Exhibiting their divergence In limb and foot structure into cursorial, forest-hving, mediportal, and graviportal types and in tooth structure into 

browsing and grazing types. 



acter may be either hurried along or slowed down in 
its rate of development, and in consequence it will 
appear in earlier or later stages of individual life. For 
example, certain adult proportions of the limbs are 
needed at birth in all cursorial animals; these adult 
proportions are consequently hurried forward during 
the foetal life, so that the animal is at birth able to 
run immediately with almost the same speed as the 
parent. This kind of velocity of development is 
called ontogenetic; it is appurtenant to every char- 
acter in every stage of its development, it is closely 
connected with the survival of certain young organ- 



same. For example, in twelve subfamilies of titano- 
theres we observe homogeneous characters evolving 
independently — the same cusps on the teeth, the same 
horns on the skull. How, then, do the subfamilies 
differ from one another? They differ because the 
evolution of each character in each phylum proceeds 
with its distinctive velocity. In a phylum that is 
evolving rapidly a certain character appears early in 
geologic time; in a phylum that is evolving slowly the 
same character appears late in geologic time. The 
titanotheres of one phylum may at a particular geologic 
period be completely hornless, whereas those of a con- 



INTBODUGTION TO MAMMALIAN PALEONTOLOGY 



41 



temporaneous phylum may have well-developed horns. 
In the former the horns may appear much later and 
may never acquire very great momentum in develop- 
ment. We can thus note the incipiency of the differ- 
ences between the short-horned titanotheres and the 
long-horned titanotheres. 

This principle of unequal phylogenetic velocity in 
the development of the same characters enables us to 
distinguish different genera and species. In one genus 
the development of the internal cusps of the premolar 
grinding teeth shows high velocity; in another genus 
it shows low velocity. Apparently these internal 
cusps are useful adjuncts of the tooth; they make the 
tooth more effective for grinding up food. Similarly 
the horns are useful adjuncts of the head in warding 
off enemies. Yet these characters evolve so slowly 
in certain phyla that it is unreasonable to believe that 
utility and natural selection are the prime causes of 
their evolution. There would seem to be physiological 
and physical (or chemico-physical) causes of these 
different velocities. It is the data on the different 
velocities of the developmen-t of the same characters in 
related phyla which give the principal biologic value to 
the long series of detailed measurements and justify 
the large number of figures that are presented in Chap- 
ters V and VI. This suggests a summary of the bio- 
logic aspects of the problems of this monograph and 
of the features that distinguish this particular field of 
biologic research. 

SUMMARY OF THE EVOLUTION OF THE TITANOTHEREG 

The known titanotheres were confined to a relatively 
small area near the fortieth parallel in western North 
America and to Europe and Asia. The direct lines of 
descent and the continuous changes in many branches 
in different or successive life zones were complicated 
by the occasional incursion of new families from out- 
side larger regions, probably from northern America 
and perhaps from northern Asia. (See appendix.) 
Nevertheless the localities in western North America 
where the remains of titanotheres have been found 
were apparently near the main geographic center of 
the evolution of the family, for the series of known 
fossils enables us to follow almost every step in the 
slow transformation of forms that were small and 
defenseless to forms that were huge and well armed. 

The remains of the titanotheres now collected repre- 
sent the most complete evolutionary series of mammals 
thus far discovered except those of the horses. The 
horses, however, are much less highly differentiated. 
In the titanotheres we see the growth of a great and 
vigorous family tree, giving off numerous branches 
(phyla), which diverge in characters and habits while 
retaining hereditary resemblances and certain heredi- 
tary trends and tendencies of transformation. Each of 
these branches is made up of slowly transforming 
successive stages (mutations of Waagen), which appear 
101959^29— VOL 1 5 



to be the more continuous and unbroken by sudden 
change the more thoroughly we explore the geologic 
levels where they successively occur. The evolution 
of the soft parts can only be inferred. The hard parts 
evolve in a variety of ways, chiefly through increase 
of size, through changes in proportion, through addi- 
tion of new parts, and in less measure through loss of 
parts. Actual addition or loss of parts in the titano- 
theres is rare; general increase in size is almost uni- 
versal, though in a few branches the size is diminished 
or arrested. 

Changes in the proportions (allometrons) of struc- 
ture were brought about by different velocities of 
phylogenetic evolution (acceleration and retardation) 
in the skeletal framework as a whole and in each of 
its parts. No less important is the definite and 
successive addition of new characters (rectigra- 
dations), each developing from infinitesimal begin- 
nings until it reaches a stage of usefulness and 
each apparently having its individuality (biocharacter) 
and its separate history. 

Throughout this wonderful transformation, which 
is in general adaptive, there were certain manifest 
germinal (hereditaiy) tendencies and certain unkno\vn 
interactions between these germinal changes and' the 
external, habitudrnal, and environmental influences. 
The more carefully we study the detailed characters in 
each branch the more evident it becomes that the 
causes of evolutionary development are neither exclu- 
sively external nor exclusively internal but are to 
be sought hypothetically in the interactions between 
germinal, habitudinal, and environmental forces. The 
changes in the proportions of the skeletal characters 
and the new elements added to the teeth and skull, 
which are the outward expressions of these hypothetic 
germinal and environmental reactions, become visible 
more or less contemporaneously but not simultane- 
ously in all members of the branches and sub- 
branches of the great family tree — that is, the 
same characters appear, but at different periods 
and with different velocities of development. The 
whole process is an orderly one, which is, however, 
not predetermined in the germinal constitution of the 
titanotheres but results from certain innate or germinal 
potentialities of evolution, which are evoked in response 
to certain environmental and habitudinal conditions. 

The struggle for existence, or natural selection, is 
operating continuously and more or less strongly on 
every single character according as its survival value 
is greater or less. In each successive geologic level 
we witness alterations of the family tree — its impover- 
ishment through the extinction of certain branches or 
its augmentation through the survival of other 
branches and the immigration of branches which 
evolved in other regions. The individual members of 
all the branches (with two exceptions) become more 
imposing and more diverse as time goes on. Finally, 



42 



TITANOTHEEES OP ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



at the climax of the general trend of transformation 
and at the very height of the grandeur of development, 
we observe the apparently simultaneous extinction of 
the whole titano there family, seemingly through failure 
to cope with changed environmental conditions or to 
compete successfully with other herbivorous types. 

This contribution to biology is therefore important 
chiefly as a study of the actual modes of evolution as 
observed in the skeleton and teeth of many different 
members of a great family of extinct animals which 
existed throughout a long period of geologic time — 
from the early Eocene through the early Oligocene — 
a time reckoned as hundreds of thousands of years. 
It is merely suggestive as to the causes of evolution. 

SECTION 3. BIBLIOGRAPHY OF LITERATURE CITED OR 
CONSULTED IN THE PREPARATION OF CHAPTER I 

Barrell, Joseph. 

1917.1. Rhythms and the measurement of geologic time: 
Geol. Soc. America Bull., vol. 28, pp. 745-904, 
Dec. 4, 1917. 
Darwin, Charles. 

1859-2. The origin of species by means of natural selection, 
or the preservation of favored races in the 
struggle for life, 502 pp. London, John 
Murray, 1859. 

FOBSTER-COOPER, C. 

1913.1. Thaumastotherium osborni, a new genus of perisso- 

dactyls from the upper Oligocene deposits of 
the Bugti Hills of Baluchistan (preUminary 
notice): Annals and Mag. Nat. Hist., 8th ser., 
vol. 12, pp. 376-381, October, 1913. 

1913.2. Correction of generic name [to Baluchitherium]: 

Annals and Mag. Nat. Hist., 8th ser., vol. 12, 
p. 504, November, 1913. 

Hayden, Ferdinand Vandiveer. 

1873.1. United States Geological Survey of the Territories; 
First, Second, and Third Annual Reports, 
reprinted in one 8vo volume, 1873. (First 
and second first issued in Rept. Commissioner 
of General Land Office for year 1867, Wash- 
ington, 1867; Third issued independently as 
Preliminary Field Report of U. S. Geol. Survey 
of Colorado and New Mexico, 1869.) 

Kovalevsky, Dr. Woldemar. 

1873.1. Monographic der Gattung Anthracotherium Cuv. 
und Versuch einer nattlrlichen Classification 
der fossilen Hufthiere: Palaeontographica, 
Band 22, Heft 3, pp. 131-210, Taf. 7-9, 1873; 
pp. 211-346, Taf. 10-17, 1874. 

Leidy, Joseph. 

1869.1. The extinct mammalian fauna of Dakota and 
Nebraska, including an account of some allied 
forms from other localities, together with a 
synopsis of the mammalian remains of North 
America: Acad. Nat. Sci. Philadelphia Jour., 
2d ser., vol. 7, pp. 1-472, 30 pis. 

Linnaeus, Caroltjs. 

1758.1. Systema naturae . . . Editio decima, reformata, 
Holmiae, 1758. 



Matthew, William Diller. 

1901.1. The Carnivora and Insectivora of the Bridger 
Basin, middle Eocene: Am. Mus. Nat. Hist. 
Mem., vol. 9, pt. 6, pp. 291-567, pis. 43-52, 
1901. 
Meek, F. B., 

1862.1 (and Hayden, F. V.). Descriptions of new lower 
Silurian (Primordial), Jurassic, Cretaceous, and 
Tertiary fossils collected in Nebraska Territory 
by the exploring expedition under the command 
of Capt. Wm. F. Raynolds, U. S. Top. Engrs., 
with some remarks on the rocks from which 
they were obtained: Acad. Nat. Sci. Philadel- 
phia Proc, vol. 13, pp. 415-447, 1862. 
Merriam, C. Hart. 

1918.1. Review of the grizzly and big brown bears of North 
America (genus Ursus), with description of a 
new genus, Vetularctos: U. S. Dept. Agr. Bur. 
Biol. Survey North Am. Fauna, No. 41, 136 pp., 
16 pis., Feb. 9, 1918. 
OsBORN, Henry Fairfield. 

1896.107. Titanotheres of the American Museum of 
Natural History: Am. Naturalist, vol. 30, 
No. 350, pp. 162-163, February, 1896. 
1896.110. The cranial evolution of Titanoiherium: Am. 
Mus. Nat. Hist. Bull., vol. 8, pp. 157-197, 
July 31, 1896. 
1900.192. Phylogeny of the rhinoceroses of Europe, 
Rhinoceros Contributions No. 5; Am. Mus, 
Nat. Hist. BuU., vol. 13, pp. 229-267, Dec. 11, 
1900. 
1909.321. Cenozoic mammal horizons of western North 
America, with appendix, Faunal lists of the 
Tertiary Mammalia of the West by William 
Diller Matthew: U. S. Geol. Survey BuU. 361, 
138 pp., 1909. 
1914.409. Recent results in the phylogeny of the titano- 
theres: Geol. Soc. America BuU., vol. 25, No. 
3, pp. 403-405, Sept. 15, 1914. 
1914.412. Rectigradations and allometrons in relation to 
the conception of the "mutations of Waagen" 
of species, genera, and phyla: Geol. Soc. 
America BuU., vol. 25, No. 3, pp. 411-416, 
Sept. 15, 1914. 
1917.462. The origin and evolution of life on the theory of 
action, reaction, and interaction. New York, 
Charles Scribner's Sons, 1917. 
Osgood, Wilfred H. 

1909.1. Revision of the mice of the American genus 
Peromyscus: U. S. Dept. Agr. Bur. Biol. 
Survey North Am. Fauna, No. 28, 285 pp., 
7 pis., map, Apr. 17, 1909. 
Waagen, W. 

1869.1. Die Formenreihe des Ammonites subradiatus, 
Versuch einer palaontologischen Monographic: 
Geognostisch-palaontologische Beitrage heraus- 
gegeben * * * von Dr. E. W. Benecke, 
Band 2, pp. 179-257, 1869. 
White, C. A. 

1868.1. First and Second annual reports of progress by 
the State geologist and the assistant and 
chemist on the Geological Survey of the State 
of Iowa, etc., 284 pp., Des Moines, 1868. 



CHAPTER II 



ENVIRONMENT OF THE TITANOTHERES AND EFFECT OF ADAPTIVE RADIATION ON THEIR 

VARIATION 



SECTION 1. GEOLOGY AND GEOGRAPHY 

CORRELATION OF EARLY TERTIARY EVENTS IN THE ROCKY 
MOUNTAIN REGION WITH THOSE IN WESTERN EUROPE 

The recorded history of the titanotheres extends 
from the upper horizons of the lower Eocene series 
(upper Ypresian or upper Wind River horizon) 
through the middle and upper Eocene to the top of 
the lower Oligocene (Sannoisian or Chadron horizon), 
covering a period estimated at 450,000 to 600,000 
years. This estimate is based on the assumption 
that 9,000 to 12,000 feet of sediment was deposited 
during the period from basal Eocene to lower Oligocene 



time and that the average rate of deposition was 1 
foot in every 100 years. 

The Eocene type formations (Wasatch, Bridger, 
etc.) of the Rocky Mountain region in North America 
have gradually acquired a time significance, similar 
to the stages (etages) into which the Eocene and 
lower Oligocene of Europe are divided, as shown in 
the following table. The correlation in time between 
France and America is close for some periods, as, for 
example, between the Sparnacian and lower Wasatch 
and between the Sannoisian and Chadron. For 
other periods the correlation is provisional, because 
the faunal relations are interrupted. 



Provisional correlation of European and American geologic stages and life zones of the tifanothere epoch 



Epochs 


Stages (Stages) ot Europe 


Type formations ot America 


Major type life zones 


Maximum 

thiclvuess of 

sediments in 

feet, deducting 

overlaps 


Lower Oligocene. 


Sannoisian. 


Chadron (Nebraska and South 
Dakota). 


(Extinction of titanotheres.) 
Tilanotherium-Mesohippiis. 


500 


Upper Eocene. 


Ludian. 


Uinta (northeastern Utah). 


Diplacodon-Protitanotherium-Epi- 
hippus. 


600 




Bartonian. 

Lutetian. 
Ypresian (upper). 


Bridger (southwestern Wyoming). 


Uintatherium-Manteoceras-Mesa- 
tirhinus. 

Palaeosyops paludosus-Orohippus. 

Eometarhinus - Trogosus - Palaeo- 
syops fontinalis. 




Middle Eocene. 


1,875 


Lower Eocene. 


Ypresian (lower). 
Sparnacian. 


Wasatch (western W3'oming). 


Coryphodon. 

(First titanotheres.) 


2,025 


Transition. 


Cernaysian. 






Thanetian. 


O 03 


Torrejon (northwestern 

New Mexico). 
Puerco (northwestern New 

Mexico) . 


Pantolambda. 
Polymastodon. 


6,000 


Basal Eocene. 


Total 11,000 


Cretaceous. ° 


Montian. 
Danian. 


Lance ( = in part Laramie and 
Denver)." 


Triceratops. 





" The United States Geological Survey classifles the Lance formation as Tertiary (?), the Laramie formation as Upper Cretaceous, and the Denver as Eo 
author of this monograph believes that the Lance formation is equivalent in part to the Laramie and Denver formations and that it is of Cretaceous age. 

43 



44 



TITANOTHEEES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



All estimates orgeologic time are highly provisional, 
because they involve two unknown quantities — the 
amount of overlap and the relative rate of deposition. 
The rate of the deposition of sediments varies enor- 
mously. For example, certain Fort Union sediments 
of Montana, aggregating 6,000 feet in thickness, are 
at present considered contemporaneous with Torre- 




Lambdotheruum popoaoicum £otUanoDS orinceps 



Figure 33.- 



-Outlincs of the bodies of titanotheres at different stages 
of evolution 



jon sediments of New Mexico, which aggregate only 
385 feet. It would therefore appear that sedimenta- 
tion in Montana was more than thirteen times as 
rapid as in New Mexico. The only sedimentary 
stage which appears fairly uniform in several geo- 
graphic localities is the Wasatch, which exhibits beds 
of approximately the same thickness in many different 
regions. 



If an average rate of deposition of a foot in a century 
is assumed, the period from basal Eocene to lower 
Oligocene time, inclusive, is estimated as not exceeding 
1,100,000 years, a moderate estimate considering the 
great biologic changes that took place in the titano- 
theres and other groups during this period. The 
epoch of the titanotheres is roughly estimated at 
500,000 years or more, during which they steadily 
increased in size, from the geologically earliest 
animals, which are no larger than a sheep, to some 
of the latest members of the race, which exceeded 
in size the largest rhinoceroses, standing over SJ^ 
feet at the shoulders. 

The recorded history of the titanotheres is 
nearly unbroken, but there have been two evolu- 
tionary gaps, one between the lower and the mid- 
dle Eocene, which was filled in 1918 by explora- 
tions of the Huerfano (Osborn, 1919.494), and one 
between the upper Eocene and the lower Oligo- 
cene, which will be filled by the exploration of 
the upper part of the Uinta formation (theoretic 
faunal zone 16, still unknown). The record also 
shows sudden transitions caused by invasions of 
animals from other regions. 

The geographic range of the titanotheres was 
probably continent wide in America and also ex- 
tended across Asia into the Balkan region of south- 
eastern Europe. In the relatively small Rocky 
Mountain and western plains region, where most 
of the fossil remains have been discovered, we 
observe the successive invasion of new kinds of 
titanotheres, which had apparently evolved pre- 
viously in other regions, probably in areas to the 
north and east. 

The geologic age of the little-known European 
titanotheres is somewhat uncertain. The type 
and only known specimen of Brachydiastemaihe- 
rium, an animal about the size of Diplacodon, is 
recorded fi'om a formation in eastern Hungary 
that was originally assigned to the lower Eocene, 
but this animal is in a stage of evolution corre- 
sponding to that of the uppermost Eocene titano- 
theres of America, and the same European forma- 
tion has yielded remains of a primitive rhinoce- 
ros (ProTiyracodon) of upper Eocene or even lower 
Oligocene type. Brachydiastematlierium is there- 
fore probably not of lower Eocene age. The 
animals described as Menodus rumelicus and 
Titanotherium hohemicum are in all respects sim- 
ilar to American titanotheres of lower Oligocene 
age, but as the localities and horizons from which 
these fragmentary specimens were obtained are in 
doubt they may be imported American fossils to which 
a European origin has been erroneously imputed. 

The correlation of the chief geographic, geologic, 
climatic, and faunistic events during the Tertiary 
period in the Eocky Mountain region with those in 
western Europe has been studied by the author con- 
tinuously during the last 20 years, with the coopera- 



ENVIRONMENT OF THE TITANOTHERES 



45 



tion of Depgret in France and of Matthew, Merriam, 
Granger, Brown, Peterson, Douglass, Riggs, Darton, 
Stanton, Berry, Knowlton, and others in this country. 
The theoretic correlations reached are shown in the 
accompanying tables (pp. 43, 48). The comparison of 
similar stages in the evolution and migration of floras 
and faunas is partly independent of changes in the 
surface of the earth and in climate and is partly 
related to them. The general succession (Osborn and 
Matthew, 1909.321; Osborn, 1910.346) of the four 
Eocene and Oligocene life phases of North America is 
as follows: 

Phase IV (lower Oligocene) , approximation. — A similar mam- 
mal fauna in western America and western Europe. Extinction 
of archaic fauna and invasion of modern fauna. 

Phase III {upper and middle Eocene), estrangement. — Inde- 
pendent mammal fauna of western America and western Europe; 
gradual diminution of archaic fauna. 

Phase II (lower Eocene), approximation. — Closely allied and 
similar fauna of western America and western Europe; first 
invasion of modernized fauna. 



place this after the first Rocky Mountain (Laramide) 
revolution in post-Laramie time — that is, after the 
end of typical Laramie deposition in Colorado. 
Others, among them the author of this monograph, 
place it at the time of the extinction of the great land 
and marine reptiles of Europe and America — that is, 
after Lance time.'' The Fox Hills formation, which 
underlies the Lance, represents the end of uniform 
widespread marine sedimentation. At some places 
the Fox Hills is continuous with overlying fresh- 
water deposits laiown as Laramie; at others it is con- 
tinuous with overlying deposits known as the Lance. 
Thus Laramie time and Lance time, in our opinion, 
are in part the same — that is, they overlap at some 
places. 

Lance and Fort Union flora. — New physiographic 
and climatic conditions arose during the initial period 
of the Rocky Mountain uplift, when uplands and 
plateaus were formed. Knowlton and Berry have 
shown that the Fort Union flora extends back into 




land areas 



Forme 



Known fossil areas 



iigration areas 

Figure 34. — Map showing the known areas (black) and the hypothetical areas (oblique lines) 
of titanothere migration and habitat 



Phase I (basal Eocene) , approximation. — Partly similar archaic 
mammal fauna of western America and western Europe. 

Final Mesozoic phase. — Gradual extinction of the upper 
Cretaceous dinosaur fauna and appearance of ancestors of the 
archaic Eocene fauna. 

This alternate approximation and estrangement of 
the mammal life of western America and western 
Europe points to periods during which conditions 
favored intermigration and intervening periods when 
geographic, climatic, or forest barriers may have stood 
between these widely separated regions. The basal 
Eocene American forests — those of the Fort Union 
epoch, for example — were very luxuriant and were 
unfavorable to migration. 

lATE CRETACEOUS AND EARLY TERTIARY CLIMATES 

End of the Cretaceous period. — The initial point in 
the correlation of geologic time in both the Eastern 
and the Western Hemisphere is-the end of Cretaceous 
deposition. (See table on p. 48.) Some geologists 



Lance dinosaur time, regarded by the author as late 
Cretaceous. The Lance flora is prevailingly a rela- 
tively warm temperate flora as compared with the 
antecedent Laramie and other Upper Cretaceous 
floras in the same region, and the climate in Lance 
time was about like that of the present Atlantic Coast 
States from North Carolina southward. In the Rocky 
Mountain province (Berry, 1914.1, pp. 153-154), in 
the zone of transition from the Cretaceous to the 
Eocene, a large number of local floras appear, such 
as those in the Arapahoe and Denver formations of 
Colorado, the Livingston formation and the Lance 
formation ("Hell Creek beds") of Montana, and the 
typical Lance formation of Wyoming. The forma- 
tions in which they occur consist of lacustrine, 
fluviatfle, and terrestrial deposits eroded from the 
rising land area of the Rocky Mountain province. 
These early so-called post-Laramie floras are said to 

8 The United States Geological Survey classifies the Lance formation as Terti- 
I ary (?). The author of this monograph regards it as Cretaceous. 



46 



TITANOTHEEES OF ANCIENT "VA^yOMING, DAKOTA, AND NEBRASKA 




1. Sweet Grass County, Mont. Fort 

Union formation. 

2. P. T., San Juan Basin, N. Mex. and 

Colo. Puerco and Torrejon forma- 
tions and "Tiffany beds." 

3. W., near Evanston, Wyo. Typical 

Wasatch group. 
4 Big Horn Basin, Wyo. Wasatch for- 
mation. 

5. W. R.. Wind Eirer Basin, Wyo. 

Typical Wind River formation. 

6. Beaver Divide, Wyo. Eocene and 

Oligocene section. 

7. H., Huerfano Basin, Colo. Typical 

Huerfano formation. 

8. B., Bridger Basin, Wyo. Typical 

Bridger formation. 

9. W. K., Washakie Basin, Wyo. Typi- 

cal "Washakie formation" of Hay- 
den. 

10. U., Uinta Basin, Utah. Typical 

Uinta and older Eocene deposits. 

11. Wh. R., White River, S. Dak. Typi- 

cal White River group. 

12. Powder River and Pumpkin Buttes, 

Wyo. Fort Union and Wasatch 
formations. 

13. F. U., Fort Union, N. Dak. Typi- 

cal Fort Union formation. 

14. P., Red Deer River, Alberta. Paska- 

poo formation. 



Oligo 
Oligocene flood plain 

Figure 35.— General geologic sketch map of the Rocky Mountain region, showing existing topography and drainage areas and 

their relation to areas of Eocene and lower Oligocene sedimentation 
Each of the numbered areas e.'cccpt 13 and 14 is also represented in geologic section in this chapter. Topography after the United States Geological Survey, 1911 

(See tables on pp. 48, 57, .=.8.) 



ENVrROlSTMENT OF THE TITANOTHERES 



47 



be distinct from those of the true Laramie and to be 
more closely allied to those of the true Fort Union 
above. 

The true Fort Union floras of basal Eocene (Thane- 
tian) age include between 500 and 600 species of trees, 
which were apparently derived from areas farther 
north, certainly not from areas farther south. 
These forests, which were contemporaneous with the 
Puerco and Torrejon mammals, indicate a climate 
in the Rocky Mountain region between the fortieth 
and fiftieth parallels that was far from tropical, yet 
moderately warm and humid, with mild winters, favor- 
able to the growth of palm, fig, and camphor trees, 
as well as other warm-temperate trees, including gink- 
gos and sequoias. This flora, which is characteristic 
of the early uplift period of the Rocky and Uinta 
Mountains in Colorado and Wyoming, indicates a 
somewhat cooler climate than that of the subsequent 
lower Eocene (Green River) epoch in the same region 
and a much cooler climate than the subtropical climate 
of the South Atlantic States in early Eocene time. In 
fact, both in the Rocky Mountain region and farther 
south the American climate became milder and more 
tropical as the Eocene epoch advanced. 

EOCENE GEOGRAPHY OF WESTERN NORTH AMERICA AND 
ITS RELATION TO FAUNAI MIGRATIONS 

GEOGRAPHIC DIVISIONS AND THEIR BEARING ON 
MIGRATION 

The main topographic features of western North 
America were established between late Cretaceous and 
middle Eocene time. In late Cretaceous and early 
Eocene time certain routes of migration connected 
the animal life of the central Rocky Mountain 
region with that of Eurasia and probably with 
that of South America. The key to these routes 
of migration and to the geographic distribution of 
these animals is afforded by the results of researches 
made since 1853 by the geological surveys of the United 
States and Canada. The foundation of the descriptive 
geologic history of the Rocky Mountain region is laid 
in the report of F. B. Meek and F. V. Hayden (Meek 
and Hayden, 1862.1). 

The entire Cordillera region extends from Bering 
Strait to the Isthmus of Tehuantepec, a distance of 



4,500 miles, and has an average width of 500 to 600 
miles. The main geographic divisions of the Cor- 
dilleran region, named in order from east to west, are 
the following : 

Rocky Mountain Range, Bering Sea to Colorado, 
including — 

Front or eastern range, facing the Great Plains. 
Rocky Mountain basins between the eastern and 
western ranges, forming the central north and 
south migration routes of mammals. 
Westerly ranges, facing the interior plateaus. 
Central interior plateaus, intermontane belt region 
(main migration routes of herbivorous mammals) : 
Northern interior plateaus, Alaska to Washington. 
Columbia Plateau. 

Nevada-Sonora Plateau (Great Basin). 
Colorado Plateaus. 
Mexican Plateau. 
Pacific mountain system, British Columbia: 
Sierra Nevada. 

Pacific mountain basins between the Sierra 
Nevada and the Coast Ranges. Coastal 
migration routes of mammals. 
Pacific Coast Range. 

A transverse section of the Cordillera on the 41st 
parallel exhibits clearly the main confines of these 
mountain ranges, basins, and plateaus. The great 
plateaus and the mountain basins may have pre- 
sented bordering forests and central grassy plains 
and jungles, interspersed with swamps, marsh lands, 
rivers, and lakes similar to those in the plateau 
and mountain (Kenya, Kilimanjaro) region of equa- 
torial Africa to-day. Migration from north to south 
or from south to north was possible along three 
routes. 

Our only knowledge of the late Cretaceous and 
Eocene mammal life of North America is afforded 
by the remains of mammals of the Rocky Mountain 
basins and foothills from Alberta to northern New 
Mexico. During the Oligocene epoch the life of the 
Columbia Plateau is revealed in the John Day forma- 
tion of Oregon. The life of the Great Plains first 
appears in the lower Oligocene formations in South 
Dakota, Wyoming, Nebraska, and Colorado, which 
border the Rocky Mountains on the east. The Eocene 
mammalian life of the country that stretches east- 
ward from the Rocky Mountain Front Range to the 
Mississippi and the Atlantic coast is entirely unknown. 



48 TITAKOTHEEES OF ANCIENT WYOMING, DAKOTA, AND NEBEASKA 

Correlation of late Cretaceous and early Tertiary stages in Europe and in North America 



Epochs 



European stages 



Rocky Mountain and Plains formations 



Changes in flora and climate 



Chief forms of reptile and mammal 



Upper Eocene. 



Ludian. 
Bartonian. 



Uinta formation {Diplacodon 
zone), upper part of "Wa- 
shakie" formation (Washakie 
B), and (?) upper part of 
Bridger formation (Bridger 
E). . 



Ancestors of horned titan- 
otheres. 



Middle Eocene. 



Lutetian. 
Upper Ypresian. 



Lower Eocene. 



Lower Ypresian. 
Sparnacian. 



Transition. 



Cernaysian. 



Basal Eocene. 



Uppermost Creta- 
ceous." 



Upper Cretaceous. 



Danian. 
Maestrichtian. 



Lower part of Bridger forma- 
tion (Bridger A, B, C, and D), 
lower part of "Washakie" 
formation (Washakie A), and 
upper part of Huerfano for- 
mation (Huerfano B). 



Rapid evolution of titano- 
theres (upper Bridger). 



Post- Wasatch and post-Green River uplift, Uinta Mountains, Utah.- 



Wind River, Green River, and 
Wasatch formations and low- 
er part of Huerfano forma- 
tion. 



Green River flora, show- 
ing affinity to tropical 
flora of the south; 
climate warmer than 
Fort Union. 



Post-Fort Union mountain uplift, Montana and Colorado. 



Fort Union, Torrejon, and Pu- 
erco formations. Swamp, la- 
goon, forested flood-plain 
sediments; lignitic and coal 
' deposits. 



Lance (upper part) , Denver and 
Arapahoe formations. Ris- 
ing land area of Rocky Moun- 
tain region; brackish-water 
estuarine, fluviatile, and chan- 
nel sediments. 



Fort Union flora of mod- 
ernized types. 



Appearance of titanothe- 
res (Wind River time). 

Appearance of modernized 
families (lower Wasatch 
time) . 



Archaic mammals of Pu- 
erco, Torrejon, and Fort 
Union time. 

Extinction of the dino- 
saurs and large marine 
reptiles. 



Fort Union flora. Warm 
and humid climate 
similar to that of south- 
eastern coastal States; 
mild winters, flora not 
tropical. Low-lying 
forested swamps in the 
plateau region. Open 
flood plains surround- 
ing the mountain 
slopes. 



Triceratops-Tyrannosaurus 
fauna. 

Mammals of Lance time. 
Ancestors of Puerco and 
Torrejon placentals, mar- 
supials, multitubercu- 
lates. Paskapoo mam- 
mal fauna of Alberta 
(more recent). 



Beginnings of Laramide revolution; Rocky Mountains (Colo.), Uinta Mountains (Utah), 
Wasatch Mountains (Utah). 



Uppermost of the conformable 
series sediments of Rocky 
Mountain and Plains region: 
Laramie formation ( = low- 
er part of Lance). 
Fox Hills sandstone. 
Pierre shale. 



Edmonton flora of Al- 
berta (similar to Fort 
Union) . 

Laramie flora transitional 
to modern. 

Upper Cretaceous flora. 
Climate warmer than 
Fort Union. 



Edmonton dinosaur {Lep- 
toceratopa) fauna (suc- 
ceeding Belly River), of 
Fox Hills (?) age; Ojo 
Alamo (N. Hex.) dino- 
saur fauna similar to 
Judith River fauna; 
Judith River (Mont.) 
and Belly River (Al- 
berta) dinosaur fauna; 
Monoclonius of Pierre 
age. 



■■ The Lance formation is classified by the United States Geological Survey as Tertiary (?) and the Denver and Arapahoe formations as Eocene. 

Note.— Near the end of Cretaceous time the chief uplift of the Laramide revolution in the Rocky Mountains began in the Front Range of the Colorado Rockies 
after the Laramie and before the Arapahoe. In the northern (the Montana) Rooky Mountains the chief uplift occurred at the end of the Fort Union. In southern Colo- 
rado and in northern New Mexico uplifts occurred both before the Puerco and after the Torrejon. (Ransome, 1915.1, pp. 360-3C2.) 



ENVIKONMENT OF THE TITANOTHERES 



49 




Figure 36. — Map of western North America showing supposed routes of migration of animals 
This map shows the general early Tertiary topography of the Great Plains, mountain ranges, northern and southern plateaus, and coast basins 
and illustrates the supposed lines of Asiatic migration from the north and South American migration from the south. Modified after 
F. L. Hansome (1915.1). 



50 



TITANOTHEKES OF ANCIENT WYOMING, DAKOTA, AND NEBEASKA 




Figure 37. — Map showing the orogeny of the western mountain and plateau region 
After F. L. Ransome (1915.1). Key to the numerals is given in Figure 36. 



ENVIRONMENT OF THE TITANOTHEKES 



51 



CHARACTER OF THE MOUNTAIN-BASIN, PLATEAU, AND 
PLAINS REGIONS 

The geographic history of the mountain-basin 
region and of the Plains region presents some resem- 
blances and some contrasts. Both regions were 
subject to slowly progressive elevation during this 
period. Nearly all the Eocene deposits of the moun- 
tain basins were laid down in broad, fiat valleys and 
on mountain plateaus, which were drained largely 
by the same great river systems that drain them 
to-day, whereas those of the Plains region were 
widely scattered over broad flood-plain areas in 
which the rivers frequently changed their courses, the 
present river courses being cornparatively modern. 
In the mountain basins, from the basal Eocene of 
the Fort Union, Puerco, and Torrejon formations to 
the summit of the upper Oligocene as represented 
in the John Day formation of the Columbia Plateau, 
the older Tertiary rocks were at very few places 
worked over into newer deposits, but at many places 
deposition was continuous. Despite continuous ero- 
sion since Oligocene time large areas of the historic 
Eocene sediments of the mountain-basin region have 
been preserved in their original purity and con- 
tinuity for the geologist and paleontologist. By 
contrast, in the Plains region large areas of the 
original Oligocene strata were in part worked over 
to form Miocene strata, and part of these in turn 
were eroded to form Pliocene strata; again all three 
contributed to the Pleistocene strata; and finally all 
four are now contributing to the alluvium of the 
Great Plains. 

EOCENE TOPOGRAPHY IN THE ROCKY MOUNTAIN REGION, 
MONTANA TO NORTHERN NEW MEXICO 

By middle Eocene time the topography of the Rocky 
Mountain region from Montana to northern New 
Mexico had become broadly similar to that of to-day. 
The existing sharply sculptured ranges of the Big 
Horn, Wasatch, Uinta, and San Juan Mountains are 
remnants of much loftier ranges, which had their 
birth in late Cretaceous and early Eocene time. The 
two great drainage systems of the region — (1) Big 
Horn, Yellowstone, and Missouri Rivers on the north 
and (2) Green, White, San Juan, and Colorado Rivers 
on the south — were probably well established at the 
end of Eocene time. 

According to Ransome (1915.1) and Lindgren 
(1915.1) the general uplift of the land in the Rocky 
Mountain region near the end of Cretaceous time 
was not uniform at different points either in its incep- 
tion or in its intensity. Apparently the earliest move- 
ment occurred after the deposition in the Denver 
Basin of the conformable series of Cretaceous beds 
that is now called the Laramie formation, which over- 
hes the Fox Hills sandstone. The Front Range of 
central Colorado arose at this time, before the deposi- 



tion of the Arapahoe formation of Colorado (Ran- 
some, 1915.1, p. 361). Andesitic tuffs and flows occur 
in the Denver formation, which immediately overlies 
the Arapahoe. At the south end of the Rocky Moun- 
tains, in northern New Mexico, great uphfts occurred 
both before and after the deposition of the basal 
Eocene Puerco and Torrejon formations. In con- 
trast, in the typical Rocky Mountains of Montana the 
principal uplift appears to have taken place at the 
end of Fort Union time — that is, subsequent to basal 
Eocene time. In the Park Range province of Colorado 
there was uplift and vigorous erosion at the end of 
the Cretaceous period and renewed uplift after the 
deposition of the lower Eocene Wasatch and Green 
River sediments. 

The separate history of the great mountain ranges 
in the basin region also shows that the upward move- 
ments began early in Eocene time. The Big Horn 
Range of northern Wyoming (Darton, 1906.1) arose 
as an anticline from the nearly horizontal strata of 
the Plains to a height of 9,000 feet in early Eocene 
time. Its uplifted peaks were truncated, and the larger 
features of the present topography were outlined. 
The major uplift of the Wind River Mountains, which 
produced a broad, low, somewhat broken anticline, 
also took place in early Eocene time (Fisher, 1906.1). 
In the Wasatch Range of western Wyoming, an east- 
ward-dipping monocline cut off along its western side 
by a great fault, there was only a slight uplift at the 
end of the Jurassic, the main uplift taking place at the 
end of the Cretaceous (Boutwell, 1907.1). Subse- 
quent movement took place in post-Eocene time. 
East of the Wasatch Range is the exceptional east and 
west anticline of the Uinta Mountains, which extends 
eastward and westward as a broad central plateau, 
150 miles long and 30 miles wide, forming a dividing 
line between the Bridger and Uinta Basins. The for- 
mation of the Uinta arch began at the end of the 
Cretaceous period (Emmons, 1907.1, p. 302), as is 
shown by the fact that the flanking Tertiary beds lie 
unconformably over the upturned edges of the older 
strata, which stand at angles of 30° or more. The Eo- 
cene formations — the Wasatch, Green River, Bridger, 
and Uinta — are upturned against the flanks of the 
Uinta Mountains, in a position which means that the 
continued rise of the mountain mass has dragged up 
the edges of the adjoining beds. 

Powell estimated that the summit of the Uinta 
anticline rose 25,000 feet above the level of the ad- 
jacent country — the Bridger and Uinta Basins. This 
altitude is equivalent to that of the Himalaya Moun- 
tains. Certainly in Eocene time the Uinta was a 
lofty, majestic range. The Colorado Front Range arose 
between the time of the deposition of the Laramie and 
Arapahoe formations, to the south, and the San Juan 
Mountains arose at the end of Cretaceous time and 
again after the deposition of the basal Eocene Puerco 
and Torrejon formations. 



52 



TITANOTHEEES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 




ENVIRONMENT OF THE TITANOTHEKES 



53 



The entire topography of the mountain-basin region 
was thus broadly defined at the end of the Cretaceous 
period and was accented by uplifts during and after 
Fort Union (Puerco and Torrejon) time; also after 
Wasatch and Green River time, following which, from 
the present Canadian border to northern New Mexico, 
there was a continuous very gradual uplift. In gen- 
eral this uplift was earlier and more rapid in Colorado 
and New Mexico — that is, it occurred before the Fort 
Union epoch — and more retarded in Montana, where 
it occurred after the Fort Union epoch. In the Huer- 
fano Basin the upturn of the western edge of the 
Huerfano beds amounts to 84°, and although this 
uplift is local it indicates a considerable movement 
in the Sangre de Cristo Range after Wind River 
time (W. Granger, letter, 1919). Ransome (1915.1, 
p. 362) believes that a large part of the Rocky 
Mountain uplift followed the deposition of the Fort 
Union formation. 

CONTRAST IN PHYSIOGRAPHIC CONDITIONS EAST AND 
WEST OF THE ROCKY MOUNTAIN FRONT RANGE 

During and after the deposition of the conformable 
Cretaceous formations (such as the Fox Hills and the 
Laramie) the country bordering the Rocky Mountain 
range on the east presented a marked physiographic 
contrast to that lying within the Rocky Mountain 
basins. Sedimentation east and west of the Rockies 
was not contemporaneous. 

East of the Rockies. — On the east flanks of the Front 
Range great river flood-plain systems began in the 
north in Pierre time and extended toward the south 
after Fox Hills time. Thus on the western borders 
of the present Great Plains region rivers had long been 
spreading out sand over their flood plains in Alberta, 
forming such deposits as the Belly River sandstone in 
Pierre time and the Edmonton sandstone in Fox Hills 
time, and extending southward through Montana to 
deposit the Judith River sandstone in Pierre time, the 
Laramie formation of Colorado, the "Hell Creek beds" 
of Montana, the great Lance sandstones of Converse 
County, Wyoming, and the Denver and Arapahoe 
formations of Colorado after Fox Hills time. 

The fact that the Lance sandstones were laid down 
at the end of Cretaceous time ^ is shown by the 
remains of the horned and carnivorous dinosaurs found 
in them, especially Triceratops and Tyrannosaurus. 
At about the same time Triceratops alticornis flour- 
ished east of the Front Range of Colorado, during the 
deposition of the Denver formation, wliich overlies 
unconformably (by erosion and uplift) the Laramie, 
the topmost formation of the "conformable Cretaceous 
series." These great flood-plain deposits, correlated 
both by their dinosaurs and by flora of the older Fort 

' The United States Geological Survey classifies the Lance formation as Ter- 
tiaryC?), but the author regards it as of Upper Cretaceous age. 



Union type, mark the beginning of the Rocky Moun- 
tain revolution as it affected the country to the east. 
At certain localities, notably along Hell Creek, Mont., 
south of the Missouri, these fans of much disturbed 
channel sand and gravel are contemporaneous with 
undisturbed beds that appear to be lithologically 
exactly like those of the Fort Union; consequently 
Fort Union sedimentation began in some regions early 
in post-Laramie time. 

This long period of mountain erosion and sedi- 
mentation east of the Rockies came to an end either 
through heavy forestation or high-gradient river ero- 
sion, which deposited materials farther east. It is a 
very significant fact that in the region east of the Rocky 
Mountains, between South Dakota and northern New 
Mexico, only sparse lower Eocene sediments (Huer- 
fano A and Cuchara) are known between Fort Union 
(basal Eocene) and Chadron (lower Oligocene) time, 
whereas in the region west of the Front Range sedi- 
mentation continued through the entire Eocene epoch. 

West of the RocTcies. — In the mountain-basin region 
from southern Montana to New Mexico the condi- 
tions during Lance time were very different from those 
that prevailed east of the Rockies. There was ap- 
parently erosion and rapid transportation rather than 
deposition. Within the mountain basins — except 
around Medicine Bow, near Laramie, and around the 
Agathaumas sylvestris locality, near Black Buttes, 
Wyo. — relatively few deposits of Lance age {Tricera- 
tops zone) have thus far been identified by means of 
fossils. The Evanston formation, above the Adaville 
formation, in the typical Wasatch section of south- 
western Wyoming, according to Berry, contains plants 
of Fort Union and of Wasatch rather than of Denver 
age. Similarly the oldest Eocene deposits of the San 
Juan Basin (the Puerco and Torrejon) are comparable 
with the Fort Union and not with the older Lance 
formation; they overlie unconformably beds of prob- 
able Montana age. In brief, few deposits of Lance 
time (Triceratops zone) have thus far been identified 
within the mountain-basin region, although they may 
be found hereafter. At many places the oldest sedi- 
ments of the mountain basins lie upon the eroded sur- 
faces of unquestioned Cretaceous and older formations 
with pronounced unconformity. 

Physiographic conditions again changed, apparently, 
for after Lance time sedimentation began vigorously 
in the mountain-basin region and continued through- 
out the Eocene until it formed deposits having a com- 
bined thickness of 9,000 to 11,000 feet. (See table on 
p. 43.) Not until Oligocene time, when the deposi- 
tion of these mountain-basin beds probably ceased, 
was great fluviatile and flood-plain sedimentation re- 
sumed east of the Front Range, forming the lower 
Oligocene Chadron beds. 



54 



TITANOTHEEES OP ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



LATERAL AND MAIN RIVER SYSTEMS IN THE MOUNTAIN- 
BASIN REGION 

The great mounfain-hasin valley. — The contour lines 
of the basal Eocene and lower Eocene sediments of 
the mountain-basin region in northern New Mexico 
and Montana are very illuminating. They show the 
presence of a series of broad, relatively level basins — ■ 
a chain of flat uplands or valleys — in which the prod- 



UINTA BAS 



Period of volcanic 
^ dust eruption in 
^ soutlnern Wyoming, 
Q> Utah. and Colorado, 
sandstones washakie basinQ, mingled with erosion 
'^ products. 

Chiefly dacite 
> tuffs and 
sandstones 




Figure 39. — Chronologic relations of formations in the mountain-basin region 

This diagram exhibits the overlapping of sediments and the falls of volcanic ash in eight widely separated areas (Nos, 
2-5 and 7-10, flg. 35), which, when combined, cover the entire Eocene epoch. 

ucts of erosion and the volcanic dust that were gath- 
ered by streams from the surrounding mountains were 
spread wide, indicating that although the mountain 
streams had high gradients and great erosive power 
the larger rivers had low gradients and little trans- 
porting power. The uniform elevation of the moun- 
tain-basin region at the north and the south and the 
low river gradients were favorable to sedimentation. 
We observe, moreover, that in basal Eocene time the 



conditions of climate and of sedimentation were some- 
what uniform in the Puerco and Torrejon deposits 
of the San Juan Basin in New Mexico, laid down by 
tributaries of Colorado River, and in the typical Fort 
Union deposits of Montana, laid down by tributaries 
of Missouri River. The rates of sedimentation were 
different. Similar basal Eocene sediments probably 
underlie some of the Wasatch (lower Eocene) deposits 
in the intermediate basins of 
Wyoming and Utah, for they 
have been exposed in the San 
Juan Basin only by the removal 
of the overlying Wasatch. A 
new sedimentary phase was 
begun in Wasatch time, and a 
third phase in Bridger time. 

The contrast in the physio- 
graphic conditions east and 
west of the Front Range has 
a very important bearing upon 
the paleontologic records. The 
mountain-basin sediments af- 
ford a marvelous and almost 
unbroken record of mammalian 
evolution in the Eocene, but 
little or nothing in the Oligo- 
cene, doubtless because large 
areas of Oligocene sediments 
have been eroded away. Only 
two spots remain — Bates Hole 
and Beaver Divide, in Wyo- 
ming. 

Piedmont, flood-plain, and la- 
custrine deposits. — King led the 
earlier geologists in presenting 
the theory that the mountain 
basins were once filled with a 
chain of lakes. This theory was 
adopted by Marsh, Cope, Scott, 
and Osborn. Leidy, as early as 
1869, cast doubt upon the lake 
theory as applied to the White 
River group east of the moun- 
tains. The lake theory has grad- 
ually been replaced by the flood- 
plain theory through the studies 
of Haworth (1897. 1), Gilbert 
(1896.1), Matthew (1899.2), 
Davis (1900.1), Johnson (1901.1), and Hatcher (1902.3). 
For the highly diversified mountain-basin region 
throughout the very long period of the Eocene, with 
its considerable climatic vicissitudes, no single theory of 
deposition is adequate. We have seen that in the 
basal Eocene, during Fort Union, Puerco, and Torrejon 
(Thanetian) time, there were doubtless great level 
areas, heavily forested, with dense undergrowth, favor- 
able to the formation of peat and lignitic deposits 



Period of mountain 
erosion of granitic, 
calcareous, rhyolitic 
early volcanic and 
sedimentary areas 



ENVIRONMENT OP THE TITANOTHEEES 



55 



and subject to heavy silting of fine sediments from 
annual floods. These were like the flooded areas of 
the forest belt in the Amazon delta. Such still-water 
areas were contemporaneous with areas in the pied- 
mont regions close to the mountains, where stream 
erosion was active. The conditions that prevailed in 
general during Wasatch (Sparnacian) time are nearly- 
paralleled by those now found in the flood plains of 
Parana, Paraguay, and Uruguay Eivers, which are 
carrying down vast masses of gravel, sand, and clay from 
the mountain chains of Brazil, as reported by John Ball 
in his " Notes of a naturalist in South 
America "(1887.1). The annual rain- 
fall in these mountains ranges from 100 
to 136 inches, and it rapidly disinte- 
grates the yieldingrocks and discharges 
a vast quantity of detrital matter over 
the broad plains of Argentina and 
Uruguay. The mountain streams 
have thus built up wide, level areas 
in these countries, and the lower 
rivers, ploughing their channels 
through the vast deposits over which 
they must make their way, extend 
their banks with every increment 
and thus continually make additions 
to the outskirts of the formation 
they are depositing. In this way 
deposits covering an area of 200,000 
square miles have been formed from 
the mountains of Brazil. 

The period of flood-plain and 
piedmont deposition in the Rockies 
was followed by the great lacustrine 
period of Green River time and of 
Wind River (Ypresian) time, in which 
the climate was much warmer. In 
the same region there ensued the 
flood-plain period of the Bridger. 

Eocene basin deposition of another 
kind and climatic change are indi- 
cated in the widespread horizontal 
' ' white layers "that divide the Bridger 
into several geologic and faunistic 
levels. These white layers indicate 
periods of lagoon leveling by annual 
uniform flooding and evaporation, similar to that of the 
existing playa lakes of the Great Basin in Nevada. 

In middle Eocene time new conditions of foresta- 
tion and erosion and the presence of volcanic atmo- 
spheric dust in the Bridger and Washakie Basins are 
indicated. Sinclair showed (1906-1909) that the 
Bridger formation was composed chiefly of volcanic 
material that has been more or less rearranged by 
stream action, and that clouds of volcanic dust 
doubtless filled the atmosphere during the Bridger 
epoch (middle and upper (?) Eocene). This interest- 
ing discovery was confirmed by thorough analyses 



made by Johannsen in 1914. The rocks of the upper 
and middle Eocene formations consist chiefly of 
volcanic tuff. Although the minerals of this tufl' are 
those of a dacite (quartz andesite), the quartz grains 
may be of sedimentary origin and the volcanic rock 
may be andesite (Johannsen, 1914.1, p. 210). The 
presence of dacite tuffs in the lower Bridger levels (B 
and C) indicates that the atmosphere was charged 
with volcanic dust, which also settled upon the con- 
temporaneous deposits of the Washakie Basin, 100 
miles to the east, as well as on the Uinta Range, 60 



.25, 26a, 26 b 




STACK MT^ 



CO > 

< / U 



LOWER BROV 



FiGURK 40. — Section of deposits near Barrel Springs, Washakie Basin, southern 
Wyoming (No. 9, fig. 35) 

Showing alternating beds of tuff, siliceous and calcareous deposits, and sandstone. Johannsen (1914.1), after 
Granger, with modifications. The numbers refer to lithologic specimens examined by Johannsen. 

miles to the south. Thus during middle and upper 
Eocene time the atmosphere over the present Bridger, 
Washakie, and Uinta region was at times charged with 
volcanic dust. Specimens of lower and basal Eocene 
rocks indicate sediments of more normal type, and 
whatever volcanic material they contain is so much 
altered by re-sorting and mixing with normal sedi- 
ments that it is not clearly recognizable. 

The manner in which the layers of dacite and glass 
tuffs alternate -with the heavy river-channel sand- 
stones is clearly displayed in the analysis of sediments 
from the Washakie Basin by Johannsen. Tuffs are 



56 



TITANOTHEEES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



also scattered, but more sparingly, through the earlier 
Torrejon, Wasatch, and Wind River formations, along 
with river-borne material derived from the decay and 
erosion of older rocks. 

SECTION 2. EOCENE AND LOWER OLIGOCENE 
FORMATIONS AND FAUNAL ZONES 

FIRST FAUUAI PHASE (BASAL EOCENE) 

SEVENTEEN LIFE ZONES 

Largely as the result of explorations and researches 
made for this monograph, the major Eocene and Oligo- 
cene type life zones that were recognized by Leidy, 
Cope, and Marsh up to the year 1900, such as the 
" Coryphodon beds," " JJintaiherium beds, "Diplacodon 
beds," and " TitanotJierium beds," have gradually been 
differentiated, through the work of Osborn, Granger, and 
Matthew, into 16 known life zones, each distinguished 
by the presence of a highly varied mammal fauna and 
by the appearance or disappearance of certain groups 
of mammals and reptiles. There is also one theoretic 
life zone, between known upper Eocene and known 
lower Oligocene time, making 17 in all. Each of these 
life zones corresponds with a series of sediments rang- 
ing in thickness from 300 to 600 feet. Many of them 
correspond with changes in climate, temperature, and 
forestation, and some of them are clearly defined and 



sharply demarcated from others. A single generic 
name, such as Coryphodon, rarely suffices to distin- 
guish them, because many genera and even certain 
species may survive for long periods of time. 

Each of these faunal zones is defined paleontologic- 
ally by one or more of the life forms it contains, geo- 
logically by the locality v/here it is best preserved, to 
some extent botanically by the flora it contains, and 
lithologically by the character of its rocks as shown 
by microscopic analyses. Thus, for example, we have 
the typical upper Wind River zone — the "Lost Cabin " 
or LamhdotJierium-Eotitanops-CorypJiodon zone, a rather 
cumbersome designation, which indicates that only 
during this period did these three types of mammals 
exist together. In this zone Lambdotherium is the 
most distinctive genus. Sediments in different geo- 
graphic basins are correlated in such a manner as to 
present the whole life story of the Eocene epoch, as 
shown in the accompanying diagram. Of the two 
chief faunistic gaps that have been recognized, that 
between the Wind River and the Bridger has now been 
filled by explorations of the Huerfano, so that there 
remains only that between the Uinta and the White 
River. These 16 known life zones will doubtless be 
multiplied to 20 or more by future discovery. They 
are shown in the following table: 



ENVIfiONMENT OF THE TITANOTHEEES 

Synopsis of life zones 



57 



Epoch 


Life zones 


Horizon 


Characteristic species and genera 




17. Titanotherium-Mesohippus. 




Brontops robustus, Menodus gigan- 


Lower Oligocene. 


Chadron B 


teus, Brontotherium platyceras, 
Allops crassicornis. 

Brontops dispar, Menodus trigono- 
ceras, Allops marshi, Brontothe- 
rium hatcheri. 

Brontops braohycephalus, Menodus 






heloceras, Brontotherium leidyi. 




16. Theoretic zone (no fauna) 

15. Diplacodon-Protitanotherium- 

Epihippus. 
14. Eobasileus-Dolichorhinus 

13. Metarhinus 


Uinta C 2. 

Uinta CI 


Diplacodon, Protitanotherium, Epi- 


Upper Eocene. 


Uinta B 2 and Washakie 

B 2. 
Uinta B 1 and Washakie 

B 1. 


hippus, Protoreodon. 
Eobasileus, Dolichorhinus. 

Metarhinus, Amynodon. 




12. Uintatherium-M a n t e c eras - 

Mesatirhinus. 
11. Palaeosyops paludosus-Orohip- 

pus. 
10. Eometarhinus-Trogosus-Palaeo- 

syops fontinalis. 


Washakie A and Bridger 
C and D. 


Uintatherium, Manteoceras, Mesati- 
rhinus. 
Palaeosyops paludosus, Orohippus. 




Bridger A and Huerfano 
B. 


Palaeosyops fontinalis, Eometarhinus. 


Lower Eocene. 


9. Lambdotheri um-Eotitanops- 

Coryphodon. 
8. Heptodon-Coryphodon-Eohippus. 

7. Systemodon-Coryphodon-Eohip- 

pus. 
6. Eohippus-Coryphodon. 


Huerfano A, Wind River 

B, and Big Horn E. 
Big Horn (Wasatch) D 

and Wind River A. 
Big Horn (Wasatch) C__. 

Big Horn (Wasatch) B-._ 


Lambdotheri um, Eotitanops, Cory- 

phodon, Meniscotherium. 
Heptodon, Eohippus, Coryphodon. 

Systemodon, Eohippus, Coryphodon. 

Eohippus, Pelycodus, Coryphodon. 


Transition basal Eo- 
cene to lower Eocene. 


5. Phenacodus-Nothodectes-Cory- 
phodon. 


Big Horn (Wasatch) A___ 


Phenacodus, Nothodectes, Corypho- 
don, Champsosaurus. 








Pantolambda, Tetraclaenodon, Claen- 








odon. 






Torrejon A._ . . .- -- 


Deltatherium, Mioclaenus, Haplo- 


Basal Eocene. 






conus. 








Polymastodon, Oxyclaenus. 














Puerco A._ _ 


Ectoconus, Champsosaurus. 










Cretaceous." 


Triceratops-Tyrannosaurus 


Lance and Denver forma- 
tions. 


Triceratops, Tyrannosaurus, Champ- 
sosaurus, Meniscoessus. 






Judith River and Belly 
River formations. 


Monoclonius, Deinodon, Eodelphis. 









» The United States Geological Survey classifies the Denver formation as Eocene and the Lance formation as Tertiary (?) . 
101959— 29— VOL 1 6 



58 



TITA^TOTHERES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 

Lower Tertiary geologic horizons and life zones and their hoofed mammals 



Epoch 


Geologic horizon 


Chiet life zones. (Titanotheres, horses, and 
other mammals.) 


Chief titanotheres and other perissodactyls 


Oligocene. 




Leptauchenia, Miohippus, and 
Oreodon. 


Extinction of titanotheres 


Chadron A, B, and C. 


17. Titanotherium-Mesohippus. 


Giant titanotheres — Menodus, 
Brontops, Brontotherium, etc. 


Upper Eocene. 


Uinta C. 


16. Theoretic zone (fauna un- 

. known). 
15. Diplacodon-Protitan othe- 
rium-Epihippus. 


Protitanotherium, early horned ti- 
tanotheres. 




Washakie B and Uinta B. 


14. Eobasileus-Dolichorhinus. 
13. Metarhinus. 


Dolichorhinus cornutus, Mesati- 
rhinus. 




Bridger C and D and Washakie A. 


12. Uintatherium-Manteoceras- 
Mesatirhinus. 


Manteoceras, ancestors of Ohgo- 
cene titanotheres. 


Middle Eocene. 


Bridger B. 


11. Palaeosyops paludosus-Oro- 
hippus. 


Palaeosyops and numerous other 
titanotheres. 




Bridger A and Huerfano B. 


10. Eometarhinus-Trog osus- 
Palaeosyops fontinalis. 


Palaeosyops fontinalis (primitive). 




Big Horn E ("Lost Cabin"), Wind 
River B (typical Wind River of 
Cope), and Huerfano A. 


9. Lambdotherium-Eotitanops- 
Coryphodon. 


Appearance of titanotheres 


Lower Eocene. 


Big Horn D ("Lysite") and Wind 
River A. 


8. Heptodon-Coryphodon-Eo- 
hippus. 






Big Horn C ("Gray BuU"). 


7. S3'stemodon-Coryphodon- 
Eohippus. 


Earliest tapiroids, Tapiridae. 




Big Horn B ("Sand Coulee"). 


6. Eohippus-Coryphodon. 


Earliest Equidae (horses). 


Transition. 


Big Horn A ("Clark Fork") of 
Wyoming and " Tiffany beds " of 
southwestern Colorado. 


5. Phenacodus-Nothodectes- 
Coryphodon. 


Earliest Phenacodus (condylarths) . 
Earliest Coryphodon. Notho- 
dectes, similar to Pleisiadapis of 
Cernay. 




Upper horizon of Torrejon forma- 
tion. 


4. Pantolambda. 


Ancestors of the Amblypoda. 


Basal Eocene. 


Lower horizon of Torrejon forma- 
tion. 


3. Deltatlierium. 


^ 




Upper horizon of Puerco forma- 
tion. 


2. Polymastodon. 


M ultitub erculata. 




Lower horizon of Puerco forma- 
tion. 


1. Ectoconus. 


Earliest known Taligrada. 



ENVIEONMENT OP THE TITANOTHEEES 



59 




Figure 41. — Eocene and lower Oligocene mammalian life zones in eleven typical correlated areas in New 
Mexico, Colorado, Utah, Wyoming, South Dakota, and Montana, located as shown on the general geologic 
map (fig. 35) 

Arranged by Osborn (1919) after original studies made in the field, chiefly by Granger, but also by Hatcher (Oligocene), Hills, Peterson, and 
Gidley (Eocene) . The 16 Icnown life zones numbered 1 to 15 and 17 are indicated in the diagram by darlc horizontal lines. The nonfossiliferous 
areas are indicated by light oblique lines. These life zones and sections also correspond with the detailed geologic sections in this chapter. 
The United States Geolojical Survey classifies the Lance formation as Tertiary (?), Eocene (?). The author of this monograph regards it 
as Cretaceous. 



60 



TITANOTHEEES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



BASAL EOCENE TIME IN MONTANA AND NEW MEXICO 

Fori Union formation of Montana. — The typical Fort 
Union formation of Hay den (Meek and Hayden, 1862.1, 
p. 433), at the junction of Yellowstone and Missouri 
Rivers, lies east of the center of an ancient forested 
swamp in which was laid down the upper part of 
Hayden's "Great Lignite Group." One of the most 
interesting results of discoveries made in 1901 (Doug- 
lass, 1902.1) is revealed in an exposure of the Fort 
Union in Sweet Grass County, Mont., near the head- 
waters of the Musselshell, containing a rich fauna of 
the archaic species of basal Eocene animals, some of 
which are identical with those found on the head- 
waters of San Juan River, in northern New Mexico, 
a thousand miles to the south. Both lie near the 
one hundred and seventh meridian. The presence in 
large numbers of animals belonging to similar species 
shows that uniform climatic and physiographic con- 
ditions existed in this great mountain-basin region 
for a very long time, because similar generic fonns of 
life (Olaenodon, Pantolambda) persist through 3,000 
feet of Fort Union sediments. The remains of the 
oldest of these mammals are found immediately above 
the dinosaur-bearing beds at a level which is here 
identical with that of the Lance formation; and the 
present opinion is that sedimentation may have been 
continuous throughout Upper Cretaceous and basal 
Eocene time in this region in Montana. 

The mode in which these Fort Union beds were 
formed has not yet been positively determined, but 
the masses of fresh-water shells which they contain in 
certain localities indicate that they were in part laid 
down in shallow lagoons and swamps, which were in 
some places crossed by river channels. At some 
places the beds contain multitudes of leaves, which 
give us a complete record of the forest life of the time. 
Vast areas of warm-temperate and more hardy trees 
were interspersed with areas where swamp vegetation 
accumulated rapidly enough to form great beds of 
lignite. Amid the glades of these forests there wan- 
dered swamp turtles, alligators, and especially the 
choristoderan reptiles of the characteristic genus 
Champsosaurus. 

Puerco and Torrejon formations of New Mexico. — 
A southern center of this archaic mammal life is the 
type locality of the Puerco formation, on the divide 
between the Rio Grande and the San Juan, in north- 
western New Mexico, a formation described by Cope 
(1875.1) as the "Puerco marls." Cope listed the 
first mammalian fauna from those beds in 1881, 
opening a new epoch in mammalian paleontology. In 
1885 he assigned to the formation a thickness of 850 
feet and distinguished it from the underlying beds, 
which he supposed to be Laramie but which have 
since been divided into the Qjo Alamo sandstone, the 
Kirtland shale, and the Fruitland formation, all 
probably of Montana age, older than Laramie. The 



Puerco of Cope appears to be a single formation geo- 
logically, deposited with apparent conformity between 
the upper and lower divisions, but it is sharply divided 
faunistically into two main life zones, a lower, which 
retains the name Puerco, and an upper, to which the 
name Torrejon was given by Wortman in 1895 
(Osborn and Earle, 1895.95, pp. 1-3A). In 1910 
Gardner (1910.1) applied the name Nacimiento 
group to both divisions. In 1897 Matthew (1897.2) 
separated the mammal fauna of the two levels, and 
in 1912 and 1913 Sinclair and Granger (1914.1) estab- 
lished in this group no less than four faunistic levels, 
which are shown in the accompanying section (fig. 
43). Two faunistic levels were observed by Wortman 
in the Puerco, and two distinct faunistic levels are dis- 
tinguished by Granger, Sinclair, and Matthew in the 
Torrejon. 

These four successive changes in the archaic fauna 
occurred during a period of continuous sedimentation, 
for no unconformity has been observed between the 
Puerco and Torrejon. The rate of deposition of the 
800 feet of Puerco and Torrejon sediments was rela- 
tively slow as compared with that of the deposition 
of the 6,000 feet of the corresponding Fort Union sedi- 
ments to the north. As the mammals distributed 
through 4,000 feet of the northern part of the Fort 
Union deposits correspond chiefly with those of the 
Torrejon, it appears possible that the underlying 
Puerco fauna may belong in part in upper Lance time. 
We observe that the Fort Union was deposited upon 
the Lance continuously, without recognized notable 
unconformity, whereas the Puerco lies upon the eroded 
surface of the Ojo Alamo, which, because of its 
dinosaur fauna, is considered of probable Judith River 
and Belly River age. 

The close resemblance of the crestless trachodont 
dinosaur, Kritosaurus navajovius, from the Ojo Alamo, 
to a corresponding form from the Belly River forma- 
tion of Alberta also suggests a close correlation in 
time.* 

In 1912 and 1913 Sinclair and Granger thoroughly 
explored the basal Eocene deposits of the San Juan 
Basin, with the results enumerated above. 

SUMMARY OF FAUNAE EVENTS OF BASAL EOCENE TIME 

In addition to the four fossiliferous zones observed 
tn the Puerco and Torrejon formations, all distinc- 
tively basal Eocene, there is an overlying zone in the 
"Tiffany beds," beyond the border of Colorado, deter- 
mined by Gidley (1909) and Granger (1916). These 
beds contain a fifth fauna, which is strictly interme- 
diate between basal Eocene and lower Eocene. This 
transitional basal-lower Eocene zone is described on 
pages 64-65. The basal Eocene mammalian life 



> See Parks, W. A., The osteology of the trachodont dinosaur Kritosaurus incur- 
limanus: Univ. Toronto Studies, Qeol. series, 1920. 



ENVIRONMENT OF THE TITANOTHERES 



61 




Figure 42. — Section of Upper Cretaceous and basal Eocene (Fort Union) deposits in 
Sweet Grass County, Mont. 

After Stanton (1909.1), Stone and Calvert (1910.1), and Gidley (1919). This very significant exposure (No. I, flg.35) is in 
an outlying area of the Fort Union formation and its mammal fauna corresponds broadly with that of the Torrejon 
formation of northwestern New Me-xico, although the section has not yet been divided into separate life zones. 
It affords the most satisfactory means of correlating the Fort Union and Puerco and Torrejon formations. The 
United States Qeological Survey classifies the Lance formation as Tertiary(?), but the author of this monograph re- 
gards It as of UpperTCretaceous age. 



62 



TITANOTHERES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 




Figure 43. — Section of Eocene deposits in the San Juan Basin, northwestern New Mexico (No. 2, 
fig. 35), showing the base of the Puerco formation resting upon the eroded surface of the Ojo 
Alamo sandstone, as observed by Sinclair and Granger (1914.1) 

This section displays the close geologic continuity of the Puerco and Torrejon beds, which are subdivided faunistically into four 
distinct life zones, named, in ascending order, (1) Ecloconus and (2) Poly mastodon zones, Puerco formation; (3) Deltatherium and 
(4) Fantolambda zones, Torrejon formation. The Ojo Alamo sandstone is perhaps of Judith River age. 



ENVIRONMENT OF THE TITANOTHEEES 



63 



of the Puerco and Torrejon formations in northwestern 
New Mexico and southwestern Colorado is accord- 
ingly related as follows: 

Transition epoch: 

5. Phenacodus-Nothodectes-Conjphodon zone. Represented 

in the "Tiffany beds" of southwestern Colorado and 

in the Wasatch formation (horizon Big Horn A, 

"Clark Fork") of Big Horn Basin, Wyo. 
Basal Eocene epoch: 

4. Pantolamhda zone. Represented in the upper levels of 

the Torrejon formation of northwestern New Mexico 

and in the upper part of the Fort Union formation of 

Montana. 
3. Deliatherium zone. Represented in the basal part of 

the Torrejon formation and in part of the Fort 

Union formation of Montana. 
2. Polymastodon zone. Represented in the upper part of 

the Puerco formation of northwestern New Mexico. 

Not yet recorded in the Fort Union formation. 
1. Ectoconus zone. Represented in the lower part of the 

Puerco formation. Not yet recorded in the Fort 

Union formation. 

BASAL EOCENE FAUNAL ZONES 

ZONES 1 AND 2: ECTOCONUS AND POLYMASTODON ZONES 

[Puerco fauna; part of Thanetian of Europe] 

No equivalent of the most ancient Puerco fauna has 
thus far been discovered in the Fort Union beds of 
the North or in Europe; it is at present unique. 

Puerco mammals and reptiles. — The Puerco mammals 
are extremely archaic, mostly Meseutheria (Osborn) 
or paleoplacentals (Matthew), representing groups of 
placentals that became extinct during the Eocene. 
The Puerco contains no remains of modern orders or 
families of mammals except three, one (Miacidae) 
which is related to the doglike Carnivora, a second 
which is related to the primitive Insectivora, and a 
third which is related to the primitive Edentata. No 
rodents or lemuroid primates have been discovered, 
and certainly no perissodactyl or artiodactyl ungu- 
lates were in this region at this time. Matthew 
(1914.1, p. 383) is of the opinion that most of these 
archaic placentals have "no known predecessors in the 
Lance formation. 

About 10 per cent of the fauna consists of rodent- 
like multituberculates, an extremely ancient order re- 
lated to the existing monotremes or to the marsupials. 
These animals are nearly related to ancestral forms in' 
the Lance. Didelphiid marsupials are also present. 

Similarly the reptiles all belong to families that 
originated in Belly River or Pierre time (Upper 
Cretaceous) or earlier. The Choristodera (Champso- 
saurus) became extinct in basal Eocene time. Note- 
worthy is the absence of the prevailing Tertiary 
families of chelonians (Emydidae, Testudinidae), 
which, with the modernized mammals, first appear in 
the lower Eocene. 

On comparing the life of the Puerco with that of 
the Lance we find a mammalian fauna that indicates 
no very wide gap in time — a fauna that is somewhat 



more ancient than the Torrejon and known Fort 
Union, also more ancient than the Cernaysian and 
upper Thanetian of France. It is therefore probable 
that the Puerco corresponds with the lower Thanetian 
of France, but its life has no known equivalent either 
in Europe or in this country. 

The opinion of Cope that the ancestry of the 
modernized mammals should be sought among these 
Puerco forms lacks adequate confirmation. The op- 
posite opinion — that the Puerco mammals are not 
ancestral to the modern mammals — was developed by 
Osborn (1893.82, 1894.89) when he applied to them 
the name Mesoplacentalia (Meseutheria), indicative of 
their archaic or Mesozoic characteristics. They repre- 
sent the first known adaptive radiation of the placen- 
tals into archaic flesh eaters and herbivores. We note 
the presence of three families of archaic Carnivora 
(Creodonta) and remote relatives {Psittacotherium) of 
the Edentata. Among the archaic ungulates we find 
one varied family (Periptychidae) of the Amblypoda 
(Taligrada) and two families (Phenacodontidae, Mio- 
claenidae) of the Condylarthra. 

Puerco sedimentation and physiography . — The Puerco 
formation is not separated from the overlying Torrejon 
formation by any lithologic or stratigraphic break. 
(Sinclair and Granger, 1914.1, p. 308.) The absence 
of erosional unconformity between the Puerco and 
Torrejon was also observed by Gardner (1910.1, pp. 
722-723) and by Bauer (1916.1, p. 277). That the 
Puerco and Torrejon formations represent a very 
long period of geologic time is demonstrated by the 
recorded 6,000 feet of Fort Union sediments, which 
have yielded the Torrejon fauna alone; and, like the 
Fort Union, they represent a very long period of uniform 
conditions of climate and sedimentation. The pres- 
ence of fish, crocodiles, turtles (Trionyx), and other 
genera in the same strata with the bones of mammals 
and at the same level shows conclusively that these 
deposits were formed by water. That the streams 
were of low gradient is shown by the complete absence 
of pebbles in the Puerco and by the wide horizontal 
extent of some of the clay bands. Bogs, apparently 
formed in back waters in the channels, were filled with 
accumulations that preserved impressions of the leaves 
of figs (Kc-Ms), plane trees (Platanus), poplars {Populus), 
relatives of the bread fruit (Artocarpus) , and numerous 
shrubs (Paliurus, Viburnum). The quantity and vari- 
ety of these plant remains, together with the abundant 
large drift logs in the clays, indicate a heavy growth 
of vegetation along the streams. The species of Ficus. 
Paliurus, Viburnum, and Artocarpus are also found in 
the Denver and Raton formations of eastern Colorado ; 
and other species indicate Fort Union age (Knowlton, 
cited by Sinclair and Granger, 1914.1, p. 306). The 
mode of occurrence of the fossils in the still-water 
clays and occasionally in the river-channel sandstones 
shows that some of the skeletons may have been 



64 



TITANOTHEEES OF ANCIENT "WYOMING, DAKOTA, AND NEBRASKA 



washed into the streams during heavy rains and 
scattered by the action of crocodiles, carnivores, tur- 
tles, and fish. Other skeletons show traces of gnaw- 
ing, probably by small Ptilodontidae, which proves 
that many of the bones lay for some time on the surface 
of the ground before reaching the streams or being 
covered in flood time by water-borne sediments. 

ZONES 3 AND 4: DEITATHERIUM AND PANTOLAMBDA ZONES 
[Torrejon and Fort Union faunas; in part Thanetian of Europe] 

The mammals of the Torrejon formation of north- 
western New Mexico, whose remains are found in a 
stratum about 250 feet above the base of the Puerco 
mammal-bearing level, are somewhat larger, con- 
siderably more diversified (perhaps because more fully 
known), and of slightly more progressive type. They 
show very close affinity to the Fort Union mammals of 
Montana and some affinity to the Cernaysian forms 
discovered in the conglomerat de Cernay, near Rheims, 
France. 

The multituberculates, which occur in the Holarctic 
region in upper Triassic (Rhaetic) time, now make 
their last appearance abundantly; of the Ptilodonti- 
dae, Ptilodus (or Neoplagiaulax) is found in New 
Mexico, Montana, and Cernay; the large Polymasto- 
don that distinguishes the upper Puerco zone does 
not recur. 

Here also are five families of archaic carnivores 
(Creodonta), among which, in the Miacidae, there is a 
genus (Didymidis) which appears to lead through the 
civet and doglike forms of the lower and middle 
Eocene into forms related to the modern Carnivora. 
Among the three Torrejon families of Insectivora the 
existing Centetidae (tenrecs) are possibly related to 
the genus Palaeorydes , a very primitive form resem- 
bling the modern Cape golden moles {(JhrysocMoris of 
South Africa, Necrolestes of South America). The 
ancestors of the modern edentates are highly diversi- 
fied (Edentata, Ganodonta) and include slothlike 
animals, indicative of present or former migrations 
into South America. Of the families of archaic ungu- 
lates two (Phenacodontidae and Mioclaenidae) repre- 
sent the Condylarthra, and two (Periptychidae and 
Pantolambdidae) represent the Amblypoda. Of the 
Amblypoda Pantolambda cavirictus, which is also 
found in the Fort Union, is very characteristic. Of 
the bearlike Creodonta (Arctocyonidae) Claenodon 
ferox, which is closely related to the Arctocyon of the 
Thanetian of France, occurs also in the Fort Union 
of Montana. 

Most of these mammals of the Torrejon, like those 
of the Puerco, were ancient adaptive radiations of 
the Mammalia. They were small-brained, had de- 
fective foot structure, and were unfitted to compete 
with the ancestors of the modernized mammals, 
which begin to appear immediately above the Noiho- 
dedes zone. Six families approached extinction at the 



end of the Torrejon — the Plagiaulacidae of the Multi- 
tuberculata; the Oxyclaenidae of the Carnivora; the 
Conoryctidae of the Edentata; the Periptychidae and 
Pantolambdidae of the Ambyploda (which, however, 
are related to the succeeding coryphodons) ; and the 
Mioclaenidae of the Condylarthra. The Plagiaula- 
cidae and Oxyclaenidae, however, survive into the 
early Wasatch, the Periptychidae into the "Tiffany 
beds." Torrejon time thus ends with the extinction 
of a large number of families of archaic mammals, 
though several families survived, passing into the 
succeeding lower Eocene. 

Unconformities of the Torrejon with the underlying 
Puerco have not been found. (Sinclair and Granger, 
1914.1, p. 312; also Gardner, 1910.1, p. 722, and Bauer, 
1916.1, pp. 273-277.) There is no doubt about the 
aqueous origin of either the Puerco or the Torrejon 
deposits. The Torrejon carries less petrified wood 
than the Puerco, but it contains Z7mo-bearing beds, 
which occur repeatedly in the gray clays, and abundant 
shells of land moUusks {Pupa), which are found in the 
clays that contain bones of mammals. Lithologically, 
the Torrejon closely resembles the Puerco, except that 
gravels of quartzite, jasper, red shale, etc., occur in 
some of the channel sandstones. Mammals appear 
principally in the zones filled with small rusty calca- 
reous concretions, which occur in clays that range in 
color from red mottled with green to gray. The upper 
boundary of the Torrejon is everywhere marked by 
the presence of Tetradaenodon (ancestor of PJiena- 
codus) and of the two amblypods PeriptycTius rhabdo- 
don and Pantolambda. The total thickness of the 
Torrejon differs at different places, ranging from 240 
to 660 feet, whereas the approximately contempora- 
neous Fort Union of Montana, which possibly also 
represents the Puerco, attains a thickness recorded as 
nearly 6,000 feet. 

The top of the Tori'ejon is in unconformable contact 
with sandstone that indicates a cycle of deposition of 
coarse sediments and alluvial fans, attributed to 
Wasatch time. 

SECOND FAUNAI PHASE (LOWER EOCENE) 

TRANSITIONAL BASAL EOCENE FAUNAS 

ZONE 5; PHENACODUS-NOTHODECTES-COETPHODON ZONE 

[Base of Wasatch formation of Big Horn Basin, first Wasatch life zone, Big Horn 
A; Cernaysian of Europe] 

The first Wasatch life zone is represented in the 
"Tiffany beds" of southwestern Colorado, in the basal 
part of the Wasatch formation (horizon Big Horn 
A= "Clark Fork") of the Big Horn Basin, Wyo., and 
probably in the summit of the Fort Union formation 
of Montana. In southwestern Colorado, near the 
headwaters of the San Juan, are the "Tiffany beds" 
of Granger, which contain a fauna characterized by 
the last appearance of PeriptycTius and by the first 
appearance of Phenacodus and of Coryphodon, a genus 



ENVIKONMENT OF THE TITANOTHEEES 



65 



characteristic of the Sparnacian of France. Notho- 
dectes of the "Tiffany beds" is particularly interest- 
ing because of its structural affinity to Plesiadapis 
of the Cernaysian of France. The multituberculates 
are represented in Wyoming by Ptilodus (" Sand 
Coulee" and "Clark Fork"?). Of the four specimens 
of ptilodontids from Wyoming, one found by Granger 
was from the Big Horn B horizon ("Sand Coulee 
beds"). Three found by Stein were probably from 
the same horizon but may have been from the under- 
lying Big Horn A horizon (the "Clark Fork beds"). 
Undoubtedly ptilodontids occur in the "Clark Fork," 
but we can not furnish any positive evidence (W. 
Granger, 1919). 

This mammal fauna as a whole actually resembles 
that of the Torrejon more closely than that of the 
lowest overljang horizon (Big Horn B, "Sand Coulee") 
of the Wasatch. A significant discovery in the No- 
thodectes zone is Zanyderis, a bat showing affinities 
with the vampires (Phyllostomatidae) of South 
America. 

The NotTiodedes zone ("Tiffany" and "Clark Fork") 
is basal Eocene, as indicated by the absence of the 
four orders Primates, Perissodactyla, Artiodactyla, 
Rodentia; it is lower Eocene, as indicated by the 
presence of Phenacodus and Coryphodon. 

The mammalian life of the "Clark Fork" beds in 
the Big Horn Basin of Wyoming, to the north, is very 
similar (Granger, 1914.1, p. 204) to that of the "Tif- 
fany beds" in Colorado. These "Clark Fork beds," 
500 feet in thickness, are characterized by the pre- 
dominance of the Condylarthra (PJienacodus and Ec- 
tocion), remains of which constitute three-fourths of 
the fossils collected from them. The amblypod un- 
gulates are represented by CorypJiodon and by the 
first appearance of an animal (Eohathyopsis) ancestral 
to Baihyopsis, of the Wind River formation, which in 
turn is ancestral to the horned UintatJierium of the 
Bridger formation. Among the Reptilia is the last 
surviving Champsosaurus from the Fort Union and 
the Cretaceous, a distinctively basal Eocene type. 

EARLY EOCENE TIME 

General correlation. — Lower Eocene (Wasatch) time 
began, it may be said, with the first appearance of 
Coryphodon and Phenacodus in the "Clark Fork" and 
"Tiffany beds" described above as the Phenacodus- 
Nothodectes- Coryphodon zone, in which is found the first 
phase of the Coryphodon fauna. The modernization 
occurred later, in the "Sand Coulee beds" (Eohippus 
zone), which overlie the "Clark Fork." 

The Sparnacian of Europe is broadly parallel with 
part of the Wasatch formation {Coryphodon zone) of 
America. It is typified in France by the deposits of 
Soissons, Meudon, and Vaugirard; in England by the 
Woolwich beds, which contain a rich flora. In these 
fluviomarine, lagoon, and lacustrine deposits of Europe 
mammals are rare, and homotaxis with America is 



afforded through the large coryphodons, the perisso- 
dactyl Lophiodon, and the creodonts Palaeonictis and 
Pachyaena. This sparse European fauna, which in its 
early stages lacks Equidae (Hyracotherium) , has almost 
a counterpart in that of the Nothodectes zone of the 
Rocky Mountain region. 

The two upper zones of the lower Eocene (Wasatch) 
are correlated with the Ypresian of Europe. 

Wasatch and Sparnacian floras. — According to Berry 
(1914.1, p. 148) the earliest Eocene beds of Europe 
(Sparnacian and Ypresian stages) contain the flora 
found in the Oldhaven, Woolwich, and Red beds of 
England, largely unstudied, and the small flora from 
the Paris Basin recently described. The Woolwich 
beds have yielded the fig (Ficus), the locust (JRohinia), 
the tulip tree (Liriodendron) , and Grevillea, a pro- 
teaceous plant now confined to Australia. Berry 
believes (letter to the author, April 1, 1918) that in 
lower Wasatch time the Fort Union flora persisted over 
the Rocky Mountain basin region. This belief implies 
that the climate was then prevailingly warm-temperate 
but that there were occasional incursions of trees of 
subtropical type. 

Sedimentation during Wasatch time. — As the Sparna- 
cian stage of Europe, which is equivalent to part of the 
Wasatch, derives its name from Epernay (Latin Spar- 
nacum), so the Wasatch stage of mammalian life 
derives its name from the typical Wasatch group of 
Hayden in western Wyoming, a single mammal-bearing 
member of which is the Knight formation (Veatch, 
1907.1), 1,750 feet in thickness, containing Cope's 
types of Eohippus index, E. vasacciensis, Phenacodus 
primaevus, Coryphodon radians, C. semicinctus, C. 
latipes. These species of mammals do not represent 
the oldest Wasatch fauna; they are of the same age as 
the species found at the "Lysite" horizon (life zone 
No. 8) of the Big Horn Basin. 

Among the chief sources of Wasatch mammals are 
the following: 

Feet 

1. Knight formation (Veatch), top of typical Wasatch 

group (Hayden), southwestern Wyoming; red and 
yellow sandy clays 1,750 

2. Wasatch formation, Big Horn Basin, Wye; red, 

brown, and gray sandstones and clays 2, 025 

3. Wasatch formation ("Bitter Creek" of Powell and 

"Vermilion Creek" of King), Washakie Basin, 
Wyo.; red and gray clays and sandstones 4, 000-5, 500 

4. Wasatch formation of the San Juan Basin, N. 

Mex 1,500 

5. Wasatch formation of the Uinta Basin, Utah (White, 

1878) 2,000 

6. Wasatch formation of the Powder River Basin, Pump- 

kin Buttes, Wyo 2,400 

The estimate made by King (1878.1) of the thick- 
ness of the sediments in the Washakie Basin (4,000- 
5,500 feet) is considered high (Granger). It is inter- 
esting to note that the mean thickness (about 2,300 
feet) of the Wasatch sediments in the six areas listed 
above exceeds somewhat that of the Bridger formation 
(1,875 feet). 



66 



TITANOTHERES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



The earliest Wasatch sediments are those in the 
Big Horn and Clark Fork Basins of northern Wyoming, 
from which we obtain the whole range of lower Eocene 
fossil mammals, beginning with the end of basal 
Eocene time. 



fossils. Douglass found a considerable faima in the 
Wasatch of the Uinta Basin. Systemodon occurs 
there. Wortman has reported (letters) a Coryphodon 
from the Wasatch of the Washakie Basin. To the 
south, in the San Juan Basin, there were laid 
down, over the Torre j on, thick beds of sand 
and fluviatile sediments, which form the New 
Mexico Wasatch. These beds, which are 
divided by Granger (1914.1) into an upper 
("Largo') and a lower ("Almagre") divi- 
sion, have a combined thickness of 1,500 feet, 
throughout the greater part of which mam- 
malian fossils are found. These Wasatch beds 
in New Mexico have much the same general 
appearance as the Wasatch in other localities, 
consisting of red, gray, and ocherous bands 
of shale and sandstone, without evidence of 
unconformity throughout the series. The most 
recently identified Wasatch sediments are 
those of Pumpkin Buttes, in the Powder 
River Basin, Wyo. 

The correlation of the faunal horizons in 
these sedimentary areas by the species of 
mammals which they contain was determined 
with remarkable precision by the American 
Museum expedition under Granger, as shown 
in the accompanying table (p. 67). 

Wasatch pJiysiograpliic and climatic condi- 
tions favorable to modernized fauna. — All the 
Wasatch sediments indicate a profound and 
somewhat abrupt change in the physiographic 
and climatic conditions of the mountain-basin 
region from those that prevailed during Fort 
Union, Puerco, and Torrejon time. In general, 
still-water sedimentation in level forests and 
lagoons ceased. Fluviatile, flood-plain, fluvial- 
fan, and channel deposits containing a larger 
percentage of coarser materials were wide- 
spread. There is evidence of open stretches 
of country exposing sand, gravel, and clay, 
subject to occasional desiccation and aridity. 

The Wasatch of the Big Horn Basin repre- 
sents the filling of an intermontane trough of 
downwarp. (Sinclair and Granger, 1912.1, 
p. 66.) Materials were transported by streams 
Figure 44.— Columnar section of Cretaceous and Eocene sediments ex- f ^.^^ ^^le surrounding mountains, as shown by 
posed -along Bear River near Evanston, in extreme southwestern ^^e lithology of the gravel, sand, and clay. The 




Wyoming (No. 3, fig. 35) , sliowing tlie typical Wasatcli group of Hayden 
(1869). Chiefly after A. C. Veatch (1907.1) 



Mammals similar to those in zone 8 (Eohippus, Phe-nncodus, lieplodon, and Coryphodon) occur 
in a narrow fossiliferous stratum which may be referred to the Heptodon- Coryphodon- Eohippus 
zone. Above are Oreen River and Bridget beds; below are 4,600 feet of beds (without mammals) 
belonging to the Wasatch group (Fowkes and Almy formations), which are imderlain by the 
Evanston formation, containing Fort Union plants, and the AdaviUe formation, containing 
Montana plants and invertebrates. The author of this monograph regards the Evanston 
formation as uppermost Cretaceous. 

Similar heavy and continuous sedimentation also 
occurred during Wasatch time, in both the northern 
and the southern Uinta region, in the Bridger and 
Washakie Basins on the north, and in the great Uinta 
Basin south of the mountains. Few of these vast 
masses of sediment have thus far yielded mammalian 



underlying Fort Union was uplifted before 
sedimentation began, and the synclinal basin 
was inclosed more or less completely to the east, 
south, and west by anticlinal mountains. 
Erosion from the mountain rocks represents 
all the members of the typical section from 
the Archean to the Fort Union, usually 
by stream transportation and deposition in river 
channels and over broad flood plains. No beds of 
volcanic ash have been found, nor is there evidence 
of transportation by wind. The deposits of clay show 
a more or less regular alternation of red and bluish- 
gray layers, which may be due to climatic changes. 



ENVIRONMENT OF THE TITANOTHEKES 



67 



The excess of iron salts in the red clays may have 
accumulated and oxidized to hematite during dry 
climatic cycles; the blue clays were probably deposited 
in a moister climate, which is less favorable to the 
concentration and oxidation of the iron. Similar 
alternations of red and blue clays in the desert basins 
of Lop and of Sewistan have been described by Hunt- 
ington, who also associates the colors with the recur- 
rence of moist and arid climatic cycles. Sinclair and 
Granger (1914.1) ascribe the color banding of the 
Wasatch and Wind River clays to a similar cause — the 
alternation of moist and dry climatic conditions — but 
they have not found any other evidence of excessive 
aridity, the fauna of the red and blue bands being 
the same. The fact that the blue clays of the Wasatch 
are here and there lignitic and are at some places 
associated with skeletal remains suggests that they 
may have been formed during cycles of rather abun- 
dant rainfall, when the surface of the intermontane 
basin was prevented from drying out rapidly. That 
these climatic and physiographic conditions were not 
local is shown by similar color banding in the Wasatch 



of all the mountain-basin regions. The name "Ver- 
milion Creek" was applied by King to the Wasatch 
because of the red color of the rocks through which 
that creek flows in southern Wyoming and north- 
western Colorado. 

Microscopic examination of the feldspars in the 
Wasatch deposits of the Big Horn Basin does not favor 
the idea of luxuriant subtropical forests and a warm, 
humid climate, with the formation of a deeply decayed 
humus, but rather suggests a dry, not necessarily arid 
climate, with rapid changes of temperature, favorable 
to splintering of the ledges of hard rock; rapid trans- 
portation of the fragments for short distances; and 
burial of these beyond reach of carbonated waters. 

A cursorial ungulate fauna. — This conception of a 
drier lower Eocene climate in the basins during 
Wasatch time accords with the successive appearance 
in this region of four families of the modernized types 
of perissodactyl mammals — horses, tapirs, lophio- 
donts, and titanotheres — with light, cursorial limb 
and foot structure adapted to rapid locomotion and 
wide seasonal migration. 



Correlation of lower Eocene life zones of Wyoming and New Mexico {after Granger, with modifications) 



New Mexico (Wasatch — 
"Largo" and "Almagre") 



'Largo" (typical). 
Eohippus, Menis- 
cotherium, Am- 
Vjloctonus. 



'Almagro" (typical). 
Eohippus, Anaco- 
don. 



Unconformit3' be- 
tween Wasatch and 
Torrejon. In south- 
ern Colorado "Tif- 
fany" (typical) . 
No perissodactyls. 



Torrejon. No peris- 
sodactyls. Fauna 
more primitive 
than in "Clark 
Fork." 



Evanston (typical Wasatch) 



Green River. 



Knight (typical) . 
Heptodon, Eohip- 



Wind River Basin (typical 
Wind River) 



Lambdotherium zone 
("Lost Cabin"; 
typical) . H y- 
rachyus, Eotita- 
nops, Lambdothe- 
rium, Heptodon, 
Eohippus, Menis- 
cotherium. 



Heptodon zone ("Lj'- 
site"; typical) . 
Heptodon, Eohip- 
pus. 



Big Horn and Clark Fork Basins 



Lambdotherium zone ("Lost 
Cabin"). Lambdothe- 
rium, Heptodon, Eo- 
hippus, Ambloctonus. 



Heptodon zone ("Lj'site"). 
Heptodon, Eohippus, 
Anacodon. 



Systemodon zone ("Gray 
Bull," typical). Syste- 
modon, Eohippus. 



Eohippus zone ("Sand Cou- 
lee," typical). Eohippus 
(abundant), etc., first ar- 
tiodactyls, rodents, and 
primates. 



Phenacodus zone ("Clark 
Fork"; typical). No 
perissodactyls, artiodac- 
tyls, rodents, or primates. 
Fauna more advanced 
than in Torrejon. 



End of lower Eocene. 



First titanotheres ap- 
pear. 



First lophiodonts ap- 
pear. 



First tapirs appear. 



First horses appear. 



Arrival of modern- 
ized mammals. 



End of basal Eocene. 



Archaic mammals 
onlv. 



68 



TITANOTHEEES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



A very significant fact, clearly presented in tlie table 
on page 67, is that these small, light-limbed, cursorial 
ungulates appear not simultaneously but at successive 
horizons. At the lowest level are the horses (Eohip- 
pus) ; at a higher level the pseudo tapirs (Systemodon) ; 
at a still higher level the lophiodonts (Heptodon) ; and 
then, toward the end of the lower Eocene, the titano- 
theres {Lambdoiherium) . 




Figure 45. — Generalized section through Upper Cretaceous and basal and 
lower Eocene deposits near Pumpkin Buttes, Powder River Valley, AVyo. 
(No. 12, fig. 3.5) 

Adapted from C. H. Wegemann (1917.1). 

Though the results of our observations may be modi- 
fied by further discoveries the successive rather than 
simultaneous appearance of these advancing waves of 
perissodactyl migration is what a study of modern 
migrations should lead us to expect. All these animals, 
as shown elsewhere in this monograph, have similar 
cursorial foot structure, which indicates extensive areas 
of dry land and open meadow, in which the small, 
defenseless Herbivora could easily escape the attacks of 
the Carnivora. 



Habitat of Wasatch mammals. — The conditions that 
prevailed in Wasatch time have been determined very 
interestingly by Loomis in his "Origin of the Wasatch 
deposits" (1907. 1, pp. 356-364). In adaptation to 
various habitats the known species of vertebrates are 
divided as follows: Aerial, 3 per cent; cursorial, 
terrestrial, and arboreal, 75 per cent; amphibious, 12 
per cent; aquatic, 10 per cent. The light-limbed 
horse, Eohippus, typical of a plains 
or partly open country, alone makes 
up 32 per cent of the total collections 
from the Systemodon zone ("Gray 
Bull" horizon). All the other odd- 
toed ungulates are light-limbed, in- 
cluding the tapiroids {Systemodon), 
lophiodonts (Heptodon), and primitive 
titanotheres {LamhdotJierium) , as well 
as the surviving archaic condylarths 
(PJienacodus and Edocion). The feet 
of all these animals indicate dry rather 
than swampy or forested land, because 
they are more slender than those of 
the modern tapir. On the other hand, 
the coryphodons were certainly marsh 
dwellers and perhaps in part stream 
dwellers. The small percentage of 
species of truly aquatic animals, such 
as crocodiles, fishes, and turtles, whose 
remains are mingled with those of the 
prevailing land animals, probably be- 
came stranded in lagoons far from the 
rivers. The presence in the rivers of 
rather large fishes is shown by the re- 
mains of the large Clastes. Remains 
of river-living turtles (Trionyx) have 
also been found in the Wasatch. 

LOWER EOCENE FAUNAL ZONES 
ZONE 6: EOHIPPUS-CORYPHODON ZONE 

[Second Wasatch life zone. Big Horn B; lower Sparnacian 
of Europe] 

Below the EoMppus-CorypTiodon 
zone in the Clark Fork Basin of Wyo- 
ming lies the first Wasatch life zone 
{PJienacodus - Nothodectes - CorypJiodon 
zone) described on pages 64-66. 
Near the head of the Big Sand 
Coulee, on the Clark Fork of the Yellowstone, which 
adjoins the Big Horn River basin on the west, 
is a series of about 200 feet of red-banded shales, 
which overlie the Phenacodus zone ("Clark Fork 
beds," transition basal Eocene) and contain a mam- 
malian fauna that is radically different from that of 
the underlying "Clark Fork." These beds (the 
"Sand Coulee beds " of Granger) mark the first appear- 
ance in the Rocky Mountain basin region of four 
modernized orders of mammals — the lemuroids. 



ENVIEONMENT OF THE TITANOTHEEES 



69 



rodents, artiodactyls, and perissodactyls. Of the 
Perissodactyla only one family occurs, the Equidae, 
represented by a prhnitive specific form of EoJiippus 
(E. horealis). There are two or possibly three species 
of Eohippus in these "Sand Coulee beds," which are 
not yet separable from the species found in the "Gray 
Bull" horizon above. Here also occurs Palaeanodon, 
an ancestral armadillo, which left descendants in lower 
and middle Eocene time. 

This lower Eocene horizon, described by Granger 
(1914.1, p. 205), appears to constitute the beginning 
of Sparnacian time in the Rocky Mountain region. It 
contains the oldest known modernized fauna (perisso- 
dactyls, artiodactyls, rodents, etc.) found in America. 
The antiquity of these beds is indicated by the last 
recorded appearance of the primitive order Multi- 
tuberculata, as represented by remains of Ptilodusf sp. 
The horizon is also distinguished by the absence of 
tapirs (Systemodon) . Here occur the first known 
species of the primitive lemuroid Notharctidae (Pely- 
codus) and the peculiar ungulate Hyopsodus, now re- 
garded as a condylarth. No other exposures con- 
taining this very primitive Wasatch fauna have thus 
far been discovered. 

ZONE 7: SYSTEMODON-COEYPHODON-EOHIPPUS ZONE 
[Third Wasatch life zone, Big Horn C; upper Sparnacian of Europe] 

The "Gray Bull beds" of Granger (1914.1, pp. 203, 
204), in the Big Horn Wasatch, lie at a horizon that 
is distinguished by the presence of the earliest tapirs — 
the pseudotapirs (Systemodon). These beds were at 
first called the Ralston,^ a name that had been pre- 
occupied. They are exposed principally in the Clark 
Fork and Big Horn Basins south of the Yellowstone 
(PI. V, B) and are at least 600 feet thick. They may be 
correlated with part of the "Almagre" of the Wasatch 
of New Mexico. As this is the first appearance of the 
tapirs, and as their remains are mingled with those 
of horses, this horizon may be known as the Syste- 
modon-Corypliodon-EoMppus zone. These beds are 
exposed chiefly along the south side of GreybuU 
River, where they extend over many miles. From 
this horizon was made the larger part of Cope's col- 
lection from the lower Eocene of the Big Horn 
Basin, including the classic skeleton of Phenacodus 
primaevus, as well as the skeleton of P. copei '° and 
that of Eohippus, besides many species of CorypTiodon. 
One of the most common forms is the pseudotapir 

>"Ealston" was the name given by Sinclair and Granger (1912) to the Clark 
Fork beds. "Clark Fork" was substituted by Granger (1914) because "Ralston" 
had been previously used otherwise. Sinclair and Granger (1912) referred the beds 
between the "Lysite" and the "Ralston" to the "Knight" formation. Granger 
(1914) separated the " Knight beds " into two horizons, which he called " Gray Bull " 
and "Sand Coulee." The "Gray Bull" and the overlying "Lysite" of Buffalo 
Basin constitute the "Big Horn Wasatch" of Cope and Wortman. The "Gray 
Bull" is exposed almost entirely in the Big Horn Basin, although a small area of it 
overlies the " Sand Coulee" beds at the head of Big Sand Coulee in the Clark Fork 
Basin (Granger, 1919). 

1 'The type of Phenacodus wortmani is from Wind River. Cope's reference of the 
small Big Horn skeleton to this species is not correct. Granger (1915) renamed the 
skeleton P. copei. 



Systemodon, which includes the species S. tapirinum, 
and it is noteworthy that this genus, which is in- 
directly related to true tapirs, does not appear in 
the overlying beds. 

ZONE 8: HEPTODON-COEYPHODON-EOHIPPUS ZONE 
[?ourth Wasatch life zone. Big Horn D and Wind River A ; lower Ypresian of Europe] 

To zone 8 belong the "Lysite beds" (PI. V, A) of 
the Big Horn Basin Wasatch, Wyoming (Big Horn D); 
the lower level of the Wind River formation (Wind 
River A) ; a part of the Knight formation of the typical 
Wasatch group; and parts of the "Almagre" and 
"Largo" of the New Mexico Wasatch. In this life 
zone Heptodon takes the place of Systemodon, which 
disappears or is not thus far recorded. The grace- 
ful lophiodont Heptodon appears at the very summit 
of the underlying "Gray Bull beds," is abundant in 
the "Lysite," and continues into the "Lost Cabin," 
its presence being one of the means of correlating the 
fauna of these beds with that of the typical Wasatch 
group in the Knight formation. This Knight fauna 
occurs in the CorypJiodon-henrmg layer, which Cope 
describes as 500 feet above the base of this division 
of the typical Wasatch of the Evanston region, or 
about the middle third of the formation according to 
Granger. 

The typical Heptodon zone (= "Lysite") of the 
Wind River beds, 350 feet in thickness, is distinguished 
by the absence of titanotheres (LamidotJierium, 
Eotitanops), which are very abundant in the super- 
imposed "Lost Cabin beds." The "Lysite" or 
Heptodon zone in the Big Horn Basin is 400 feet thick. 
Anacodon, one of the arctocyonid creodonts, which 
has flattened or pavement-like teeth, is characteristic 
of the Heptodon zone. This zone is faunistically but 
not lithologically separated from the overlying Lamh- 
dotJierium zone. 

ZONE 9: 1AMBD0THEEHTM-E0TITAN0PS-C0RYPH0D0N ZONE 

[Fifth Wasatch life zone. Big Horn E, Wind River B, and Huerfano A; upper 
Ypresian of Europe] 

Geology and fauna. — To zone 9 belong the typical 
Wind River of Hayden and of Cope in the Wind 
River Basin, Wyo. ( = the "Lost Cabin" of Granger 
and Sinclair); the "Lost Cabin" (Granger) of the 
Big Horn Basin Wasatch; part of the "Largo beds" 
(Granger) of the San Juan Wasatch of New Mexico; 
part of the Green River lacustrine formation of 
Wyoming; and the lower level of the Huerfano for- 
mation (Hills) of Colorado or Huerfano A. This is 
the typical Wind River life of all the literature of 
Cope. (See PI. VI.) 

The Lamldotherium life zone is distinguished by the 
arrival in the Rocky Mountain basin region of the 
first titanotheres, which are abundantly represented 
in remains of the smaller, cursorial Lambdotherium 
and the larger, mediportal Eotitanops. It includes the 



70 



TITANOTHEEES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 




FiGUHE 46. — Composite section of the Eocene deposits of tlie Big Horn and Clark Fork Basins, Wyo. 

This section contains ttie entire Big Horn Basin Wasatch of Cope's descriptions, which is now divided into very clearly defined 
ascending life zones, as follows: 5, Phenacodus-Nothoiectes-Coryphodon zone; 6, EoMppus- Corijphodm zone; 7. Systemodon-Corypho- 
don- Eohippus zone; 8, Heplodon- Coryphadon-Eohippus zone; 9, Lambdotliermm-Eotitanops-Coryphodon zone. A few characteristic 
species of mammals from each horizon are indicated in the right-hand column. Chiefly after Granger (1918). 



ENVIEONMENT OF THE TITANOTHERES 



71 



last surviving species of Coryphodon and of the con- 
dylarth Phenacodus among the archaic ungulates. 
The presence of the condylarth Meniscotherium 
serves to correlate the Wind River with the upper 
levels ("Largo beds") of the Wasatch of New Mexico. 
While the Wind River life on the whole represents 
a continuation of that of the preceding stages of the 
Wasatch, with which it possesses several genera and 
eleven species in common, it also includes nine new 
genera that survive in the Bridger formation of 
middle Eocene time. The Wind River marks the 
end of the lower Eocene, the last period of certain 
highly distinctive lower Eocene forms like Cory- 
phodon, but it is also prophetic of the middle Eocene 
in the presence of lemuroids like Notharctus, Anapto- 



somewhat like a slender, diminutive tapir in body 
proportions. In skull structure and dentition Eoti- 
tanops foreshadows the true titanotheres of the 
middle Eocene; its feet are more slender than those of 
its successors, and it was doubtless a more agile animal. 

The special life conditions surrounding these early 
titanotheres are more fully set forth in the descriptions 
of the Wind River titanotheres in Chapter V, section 3. 

Olimate and physiography during the deposition of 
the Wind River and Green River sediments. — For Wind 
River life in general the reader is referred to section 3 
of this chapter. Here we may speak of the whole 
basin region. 

While fluviatile and flood-plain sediments were 
being deposited in the Wind River Basin of northern 







f!t» - ""."-i**^*, 





Figure 47. — A typical "Lost Cabin" locality, on the north side of AlkaU Creek about 8 miles east of Lost 

Cabin, Wind River Basin, Wyo. 

Lambdotherium-Eotitanops- Coryphodon zone (Wind River B). A characteristic view of tlie red-banded beds that have yielded the greater part of 
the fauna of the Lambdotherittm zone. (Compare PI. VI, B.) After Granger (1910.1), Am. Mus. negative 17792. 



morphus, and Shoshonius; of true doglike or civet-like 
carnivores like Viverravus and Vulpavus; or of rodents 
like Sciuravus and Par amy s. Remains of Equidae 
are rather rare and are represented by several species 
of Eohippus, of which E. venticolus is the most pro- 
gressive, and those of titanotheres, especially Lamb- 
dotherium, are very abundant. 

Lambdotherium, one of the earliest titanotheres, 
was a small, light-limbed form, about the size of a 
coyote {Canis latrans). It represents a distinct 
cursorial side branch of the titanothere family, re- 
sembling the contemporary horses and lophiodonts 
in its light limb and foot structure. Eotitanops 
("the dawn titanothere") was a true and very primi- 
tive titanothere about the size of a sheep {Ovis aries), 



Wyoming there lay to the south a large, shallow lake, 
covering about 5,000 square miles, in which were 
deposited 800 feet of impure limestone at the base, 
followed by about 1,200 feet of thin, fissile calcareous 
shale. (King, 1878.1, p. 381.) The deposition of 
these lake sediments (Green River) began near the 
end of Wasatch time. They contain abundant and 
well-preserved remains of insects and fishes. The 
presence of sting-rays and other fishes of marine or 
coastal type indicates that these originally marine 
forms had become landlocked, as did the existing 
marine survivors in the Caspian Sea and Lake Titicaca. 
Many of the fishes of the Green River shales are related 
to forms now found chiefly in the southern continents, 
especially South America. 



72 



TITA.NOTHERES OP ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



Green River forests. — Our present knowledge of the 
Green River flora, which, according to Berry (1914.1, 
p. 164) was mid-Eocene, indicates a considerably 
warmer climate than that of the basal Eocene Fort 



forests differed from the tropical forests of the Georgia 
coast in the presence of genera Hke Bex, Juglans, 
Myrica, Planera, Quercus, Rhus, Salix, and Zizyphus, 
most of which are temperate types. Thus the Green 




MESOZOIC and PALEOZOIC 



Pe/ycoc/us 

Sh osh on fas 
/^bsaroh/L/S 
' Didyrnic^iS 
\//verr3\/us 

Miacis 
Vu/pavus 
Pafripfe/is 
Pro/i'mnocyon 
S/nopa 
Trif-emnoc/on 
^Hapa/odecfes 
CynodonfomyS 
Diacodon 
Pariofops 
D/de/pnodaS 
,Pa/seosinopa 
{Esfhonyx 
' Paramys 
Mysops 
? Palaeanodon 
5/y/inodon 
Phenacodus 
Pcfoconus 
Menisco^her/um 
Pyopsodus 
Coryphodon 
Baihyops/s 
Eoffranops 
Lambdoiher/um 
Hyrachyus 
Hep^odon 
fiobippus 
[O'Scodex/s 

Pe/ycodus 

Omomys 

Te fortius 

Abs3ro/<ius 

Didymicii's 

Vulpa vus 

Oxyasna 

S/nopa 

Hapa/odecfes 

Cynodon fomys 

En'fomo/esfes 

■acodon 
Esfhonyx 
Paramys 
PPa/asanodon 
Phenacodus 
Pyopsodus 
Coryphodon 
Pephodon 
Eobippus 
Wasdi-chia 
Diacodexis 



Figure 48. — Section through the Wind River formation (lower Eocene) near Lost Cabin, Wind River Basin 

Wyo. (No. 5, fig. 35) 

A complete list or genera from each horizon is given in the right-hand column. First appearance of the primitive titanotheres Lamidotherium 
and Eotitanops. Chiefly after Granger (1910.1) and Sinclair and Granger (1911.1). 



Union, for it includes such types as Acrostichum and 
Arunio, which are also represented in the contempo- 
raneous Eocene flora of Georgia, as well as the genera 
Ficus and Sapindus. The Green River lake-border 



River exhibits a commingling of warm-temperate and 
tropical trees such as are now found in subtropical 
forests in regions where there is a mean annual temper- 
ature of about 14° C, uniform humidity, and an 



ENVIEONMENT OF THE TITANOTHEHES 



73 



annual rainfall exceeding 200 centimeters. This flora 
is not very difl'erent from that found in the upper 
Ypresian of France. 

These forests are so interesting in respect to the 
environment of the first titanotheres which appeared 
in North America that the principal genera cited by 
Berry may be quoted in full. The figures 
appended to the names of the genera 
show the number of species in each genus. 



(Sinclair and Granger, op. cit., p. 105.) There is 
evidence also of an uplift of the Big Horn Range 
subsequent to the deposition of the Wasatch. In 
the Wind River Basin material washed down from 
the mountains continued to be spread over the basin 
floor by streams until the end of upper Eocene time. 



Acrostichum, 1. 
Alnus, 1. 
Ampelopsis, 1. 
Aralia, 1. 
Arundo, 2. 
Brasenia?, 1. 
Cheilanthes, 1. 



Eucalyptus?, 1. Myrica, 1. 

Ficus, 4. Phragmites, 1. 

Ilex, 2. Planera, 2. 

Juglans, 3. Quercus, 2. 

Leguminosites, Rhus, 1. 

1. Sabal, 1. 

Lygodium, 1. Salix, 2. 



Cissus, 1. Manicaria, 1. Sapindus, 1. 

Cyperus, 1. Musophyllum, Sphaeria, 1. 

Equisetum, 1. 1. Zizyphus, 2. 

The Green River flora is the only de- 
scribed middle Eocene flora known from 
latitude 40°. The nearly contempora- 
neous Claiborne flora of Georgia shows 
(Berry, op. cit., p. 161) that the main 
elements of the modern flora of tropical 
America reached at least as far north as 
latitude 33° and, in the middle Eocene, 
probably several degrees farther north. 

Wasatch and later events. — In areas that 
lay north of the great lake in this region 
in Wasatch time were laid down the sedi- 
ments of the Wind River and Big Horn 
Basins, the deposition of which began in 
the first phase of Wasatch time and prob- 
ably continued into middle Oligocene 
time. (Sinclair and Granger, 1911.1, p. 
85.) The Wind River sandstones in the 
vicinity of the Beaver Divide are stream- 
channel deposits, probably laid down in 
broad, shifting streams of low gradient 
which flowed across clay-covered flats, 
into which they sunli their channels or 
over which in seasons of flooding they 
spread coarse detritus. The materials 
composing these sandstones were derived 
from the granites and other pre-Ter- 
tiary rocks of the surrounding moun- 
tains. Below the Lambdoiherium zone 




Figure 49. — Map showing cluster of lower, middle, and upper Eocene sedi- 
mentary basins in southwestern Wyoming and northern Utah, exhibiting 
parts of areas of the Wasatch, Wind River, Bridger, and Uinta formations 
at other localities, interstratified with Extensive areas of the Wasatch are purposely omitted. After Osborn and Matthew (1909.321), U. S. 

Wi T~» • 1 1 ^ A ,1 Geol. Survey Bull. 361. Arrows show lines along which sections were taken 

md River clays and sandstones, there 



are layers of white volcanic tuff, 13 feet thick, in- 
dicating the presence of active volcanoes. The floor 
of the Big Horn Basin, to the north, was modified 
by erosion that took place subsequent to the main 
uplift of the Big Horn Mountains, which occurred 
after the deposition of the Fort Union formation. 
101959— 29— VOL 1 7 



Fluviatile and flood-plain deposition is indicated 
throughout Eocene time. The lignitic shales that 
cap the Lamhdof-herium zone of the Big Horn Basin, 
containing fresh-water mollusks (Planorhis) and crus- 
taceans (Entomostraca), are certainly both fluviatile 
and palustrine. 



74 



TITANOTHEEES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



TRANSITIONAL LOWER TO MIDDLE EOCENE DEPOSITS 
HUERFANO FORMATION OF COLORADO (LOWER AND MIDDLE EOCENE) 

While the lacustrine and flood-plain Green River 
and Wind River formations were being deposited in 
Wyoming there were accumulating in southeastern 
Colorado the lower fossiliferous beds of the Huerfano 
formation, described by Hills (1888.1), explored by 
Osborn and Wortman in 1896 and by Granger and 
Olsen in 1918, and now known as Huerfano A. The 
deposition of this formation apparently began near 
the end of Wasatch time and extended into early 



although part of its fauna is doubtless transitional 
from the summit of the underlying lower Eocene. 

In the upper half of the Huerfano formation (Huer- 
fano B) are found mammals that are also characteristic 
of the lower Bridger (A) . The imperfectly known life 
of the upper level includes the tillodont Trogosus and 
two kinds of small titanotheres, one (Eometarhinus) 
resembling Metarhinus and the other Palaeosyops fon- 
tinalis of Bridger A; also a horse {OroMppus?) and ani- 
mals resembling the Bridger genera Hyrachyus, Hyop- 
sodus, Microsyops, as well as more ancient genera — 
the creodonts Amhloctonus and Didymidis — which 




Figure 50. — Sketch map of the region of the Huerfano and Cuchara formations in southern Colorado 
After Hayden (1880), Hills (1888.1), and Granger (1918). 



Bridger time. Among the mammals of the lower 
Huerfano, which corresponds with the upper Wind 
River (="Lost Cabin"), are the rare Coryphodon, the 
small-limbed titanothere Lambdotherium, EoTiippus, 
Oxyaena, Didymidis, and Heptodon, a purely upper 
Wind River (="Lost Cabin") fauna. 

The whole Huerfano formation is 3,500 feet thick, 
and a large part of it (see fig. 51) lies below horizon 
A. (Granger, 1918.) Huerfano B, although it lies 
immediately above Huerfano A, contains the genus 
Palaeosyops, a distinctive middle Eocene form. Con- 
sequently Huerfano B is placed at the base of the 
middle Eocene and is correlated with Bridger A, 



suggest a fauna more ancient than that of Bridger B, 
corresponding perhaps with the still unknown fauna 
of Bridger A. It appears probable that the Huerfano 
will give us a complete faimistic transition between the 
end of Wasatch and the beginnuig of Bridger B (middle 
Eocene) time. 

WIND RIVER BEDS AND THEIR FAUNA 

The discovery of the geologic section at Beaver 
Divide, between Wind River and Sweetwater River, is 
one of the most significant recently made in the study 
of Rocky Mountain basin geology. Here deposition 
without angular unconformity extends from the third 



ENVIRONMENT OF THE TITANOTHEEES 



75 




Figure 61. — Section of the Huerfano formation in southeastern Colorado as exposed west of 
Gardner, Huerfano Basin 

Thickness 3,500 feet. Near the summit is a Wind River B fauna (.Lambdotherium zone), and above that a Bridger A fauna 
(.Palaeosyops fontinalis zone). After Granger (1918). 



76 



TITANOTHEKBS OF ANCIENT AVYOMING, DAKOTA, AND NEBEASKA 




Oj^eodoTL zoTie 



Oreodon 
Cylindrodon 
> Caenopus 
Ischyromys 
Poebro therium 



Menodus he/oceres 

?I)ipioicodo7v zoTze^ 

Amynodon ? anficjuus 
Protoreodon 
Camelodon 
^ Pro tifan other/ um 



Zamhdotherhwv zone 



Lambdofherium 
Coryphodon, Phenacoo/us_, 
Hepfodon, Eohippus 



Figure 52.— Section of exposures from lower Eocene to lower Oligocene at Green Cove, 
on Beaver Divide, at the southwestern border of the Wind River Basin, Fremont 
County, Wyo. 

Includes deposits in Wind River, Bridger (?), Uinta (?) and White River time. Chiefly after Granger (1910.1). This 
is a most significant section, for the base ot the TiimolUnum zone (Chadron A) uncontormably overlies beds 
originally referred to Uinta C 1 (Diplacodon zone). 



ENVIEONMENT OF THE TITANOTHEEES 



77 



Wasatch Heptodon-Coryphodon-EoMppus zone through 
the Wind River Lamhdotherium-Eotitanops-CorypTiodon 
zone upward into the Oreodon zone of Ohgocene time. 
This is the only undoubted Eocene-Oligocene sedi- 
ment thus far determined in the Rocky Mountain 
basin region. Its total thiclmess is 1,080 feet, and it 
represents relatively slow sedimentation. There is a 
single period of erosional unconformity at the end of 
the upper Eocene. 

The life of the Wind River beds of this section is 
distinctly of upper Wind River ("Lost Cabin") time, 
corresponding with Wind River B and Huerfano A, 
for it includes the titanothere Lambdotherium popo- 
agicum, a CorypTiodon, two species of Equidae {Eohippus 
craspedotus and E. venticolus), and two species of 
Heptodon (H. calciculus and H. ventorum), which are 
characteristic of closing Wasatch time. The presence 
of remains of garpikes and crocodiles in this fauna 
shows that these deposits were fiuviatile and indicates 
that Wind River shales were of flood-plain origin, 
though they include many channel fillings of coarse 
arkose. 

We thus have glimpses of a faunistic period broadly 
corresponding with the lower Ypresian of France, cer- 
tainly extending from Wyoming to Colorado, and 
probably spreading much more widely in the Rocky 
Mountain and the adjacent Plains region. Though it 
includes surviving members of the older Wasatch life 
and incoming members of the succeeding Bridger life, 
the Wind River and Huerfano life stands directly 
intermediate between these; in fact, the representa- 
tives of archaic families destined to become extinct 
and those of modernized families destined to populate 
the earth are very nearly balanced, including 21 genera 
(30 species) of archaic mammals and 22 genera (36 
species) of modernized mammals. 

Simultaneously with the decline of the coryphodons 
the uintatheres reappeared in the genus Bathyopsis, 
ancestral to the giant Uintatherium, which character- 
izes Bridger C and D. 

THIRD FAUNAI PHASE (MIDDLE AND UPPER EOCENE) 

CORRELATION OF AMERICAN ZONES WITH EUROPEAN 

STAGES 

There is strong evidence of uniform and favorable en- 
vironment and persistent evolution throughout middle 
and upper Eocene time in the Rocky Mountain basin 
region. The changes show progressive modification 
and adaptation rather than breaks by migration or 
extinction. Both the archaic and the modernized 
families increased in size and variety. The surviving 
archaic mammals appear to have flourished and in- 
creased, especially in size and muscular power. Near 
the very end of Eocene time only two new famihes of 
quadrupeds appear, the ancestral camels (Camelidae) 



and the oreodonts (Oreodontidae), whereas in western 
Europe new families repeatedly appear from the south. 




east, and north. The general correlation of the Euro- 
pean stages and the American zones is given on page 78. 



78 TITANOTHEEES OF ANCIENT WYOMING, DAKOTA, AND NEBEASKA. 

Correlation of middle Eocene and upper Eocene American life zones and European stages 



Epoch 


American life zone 


Approximate European 
stage 




15. Diplacodon-Protitanotherium-Epihippus zone (Uinta C). 


Ludian. 


Upper Eocene. 


14. Eobasileus-Dolichorhinus zone (Washakie B 2, Uinta B 2). 
13. Metarhinus zone (Washakie B 1, Uinta B 1). 


Bartonian. 




12. Uintatherium-Manteoceras-Mesatirhinus zone (Washakie A, Bridger C and 
D). 


Lutetian. 



Middle Eocene. 



I 11. Palaeosyops paludosus-Orohippus zone (Bridger B). 

i 10. Eometarhinus-Trogosus-Palaeosyops fontinalis zone (Bridger A, Huerfano B) . 



TYPICAL BBIDGER FORMATION; MIDDLE EOCENE 
(LUTETIAN AND BARTONIAN OF EUROPE) 

Character of sediments. — The Bridger formation, the 
most important and the most thoroughly explored of 
the Eocene Tertiary, occupies a great area north of 



ditions. The Bridger formation attains its maximum 
thickness of 1,875 feet near the Uinta Mountains and 
thins out northward. Beyond the margins of the 
Green River lacustrine deposits the Bridger overlies 
upper members of the Wasatch group. 




Figure 54. — Map showing the Eocene sediments encircling the Uinta Mountains of southwestern Wyoming 

and northern Utah 

Modified after King (1876.1). U, Uinta Basin, typical Uinta formation of King and Marsh. (The area mapped includes older and possibly 
younger rocks than the true Uinta formation— Z)/piacoion zone.) B, Bridger Basin, typical Bridger formation of Hayden. WK, Washakie 
Basin, typical " Washakie" formation of Hayden. G, Green River formation. W, Typical Wasatch group of Hayden. X, Type locality 
of Coryphodon and associated Wasatch fossils. C, Cretaceous. 



the Uinta Mountains and east of the Wasatch. Unlike 
the Wasatch, the lower Bridger (horizon B) is unique; 
no contemporaneous fossiliferous deposition is known. 
At the base its sediments pass gently into the Green 
River shales, and the lower levels of Bridger A show 
gradual transition from lacustrine to flood-plain con- 



Unlike the lower Eocene Wasatch and Wind River 
sediments the Bridger is not composed chiefly of 
material derived by erosion from the adjacent moim- 
tains (Sinclair, 1906.1, p. 278) but consists of great 
series of deposits of volcanic ash and dust, solidified 
into tuffs, which weather into picturesque and in 



ENVIRONMENT OF THE TITANOTHEHES 



79 



places highly colored "badlands." Apparently the 
greater part if not all of these tuffs were distributed 
from unlinown eruptive volcanic centers by wind; but 
at four periods they were deposited in great shallow 
playa lakes and partly worked over by stream, delta, 
and flood-plain deposition. From the general absence 
of coarse materials such as would be transported by 
streams of high gradient, it is inferred that the Bridger 
formation accumulated in a relatively level area. 
(Sinclair, 1906.1, p. 279.) 

Exploration of the Bridger formation. — The Bridger 
formation has been explored almost continuously by 
geologists and paleontologists, first by Hayden (1869- 
1871), next by King (1878), who regarded the Bridger 
as an ancient lake basin deposit, then by Osborn and 
Scott (1877-1878), and again by Endlich (1879). 
In 1902 the American Museum parties, guided by 
Matthew and Granger, under the direction of Osborn, 



undertook to determine whether the Bridger can be 
divided into a series of life zones. After four years of 
careful geologic field work by Granger and Matthew 
(1902-1905), who had at hand the level record of every 
specimen, the Bridger was subdivided lithologically 
and faunistically into five levels, A to E. Bridger A 
is relatively barren. Of these levels A and B were 
grouped into the lower Bridger (Palaeosyops paludosus- 
OroMppus zone), characterized by the absence of 
Uintatherimn, and C and D, the upper Bridger 
{Uintatherium- Manteoceras- Mesatirhinus zone), distin- 
guished by the appearance and great abundance of 
TJintaiherium. Similar faunistic surveys in the 
Washakie Basin, east of the Bridger Basin, and 
in the Uinta Basin, south of the Uinta Mountains, 
have given very complete correlation of the local 
subdivisions of the section as follows: 



Correlation of middle and upper (?) Eocene sections of the Uinta, WashaTcie, and Bridger Basins 



Uinta Basin 


Washakie Basin 


Bridger Basin 


Life zones 


Uinta C. 


Absent. 


Absent. 


Theoretic zone (No. 16); fauna unknown. 
15. Diplacodon-Protitanotherium-Epihippus. 


Uinta B. 


Washakie B. 


Bridger E (barren beds). 


14. Eobasileus-Dolichorhinus. 
13. Mctarhinus. 


Uinta A (barren). 


Washakie A. 


Bridger D. 
Bridger C. 


12. Uintatherium-Manteoceras-Mesatirhinus. 


Barren beds. 


Barren beds. 


Bridger B. 


11. Palaeosyops paludosus-Orohippus. 


Bridger A. 


10. Eometarhinus-Trogosus-Palaeosyops fontinalis. 



Volcanic ash deposits. — The petrographic analysis of 
the rocks of the Bridger formation serves to support 
their correlation with the deposits of the Washakie 
Basin, to the east, and of the Uinta Basin, to the south. 
The recognition by Sinclair (1906.1, pp. 273-280) of 
the fact that the entire Bridger series was in large 
part originally volcanic dust and the later careful petro- 
graphic analysis by Johannsen (1914.1) led to the 
conclusion that the Bridger rocks are largely tuffs 
perhaps modified in part by sufficient transportation to 
add the numerous grains of quartz they contain, and 
that these grains may be of sedimentary origin 
although the material of the tuffs is mostly andesitic. 
Johannsen's analysis of the Bridger rocks is essentially 
as follows: 
Bridger D. Irregular grains of quartzite, feldspar, hornblende, 

etc. : dacite tuff. 
Bridger C. Fragments of quartz and hornblende; groundmass of 

glass tuff. 



Bridger B. Smith's Fork; fragments of quartz, feldspar, horn- 
blende: ?dacite tuff. 

Bridger B. Church Buttes; fragments of quartz, feldspar, etc.: 
altered tuff, probably dacite tuff. 

Bridger A. North of Church Buttes, fragments of quartz, feld- 
spar, hornblende. No glass tuff seen. 

Thus the Bridger is composed chiefly of dacite tuff, 
of altered dacite, and of glass tuff containing irregular 
grains of quartz, feldspar, and hornblende, which 
are at some places contained in a groundmass made up 
of entirely coarse angular particles of stringy glass 
full of bubbles. The Huerfano formation of Colorado, 
which is in large part older than the Bridger, is com- 
posed largely of glass tuff. The deposits in the 
Washakie Basin, east of the Bridger Basin, are com- 
posed chiefly of dacite and glass tuffs. 

Playa lalce deposits. — Conspicuous features of the 
Bridger formation are four hard "white layers,'' 
which were laid down at intervals in the series of beds. 



80 



TITANOTHEBES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



Some of these "white layers" have been traced over 
many square miles. They are composed of tuffaceous 
shale and marl or of calcareous shale and are in places 
filled with fresh-water shells. They mark periods 
during which the deposition of volcanic dust was less 
rapid, when the Bridger Basin was temporarily base- 
leveled and the waters rose into wide, shallow playa 
lakes, in which sedimentation was slow. That these 
four relatively thin "white layers," which are vari- 




FlGURE 



-Geologic section of the entire Bridger formation in the 
Bridger Basin, Wj'o. 



Shows the division by the four chief "white layers" and the main divisions by three principal 
zones — A, Palaeosyops fontinalis zone; B, Palaeosyops paludosus-Orohippus zone; C and 
Utniatherium zone. 

ously known geographically as the "Cottonwood 
white layer," the "Burnt Fork white layer," the 
"Lone Tree white layer," and the "upper white layer," 
correspond with long periods of geologic time is shown 
by the marked faunal differences that separate them, 
which indicate that extensive migration occurred 
before and after the deposition of each of these layers, 
but especially the first, which separates the lower 
from the upper Bridger life zone. 

Life environment in Bridger time. — From observa- 
tions made by Hay (1905.1, pp. 327-329) while he was 



collecting fossil turtles in the Bridger in 1903, he con- 
cluded that the Bridger deposits were almost solely 
the result of fluviatile and flood-plain action, that this 
basin was a nearly level country, which was probably 
covered with vegetation and well forested. The dis- 
tribution of fossil remains in all parts of the Bridger 
area indicates that the animals lived near the places 
where they became buried and that they were chiefly 
such as may inhabit well-wooded regions. The river- 
channel beds, which are composed of coarse ma- 
terials, show that streams with rapid currents 
traversed the basin. These streams were bor- 
dered by swamps in which were formed beds of 
impure lignite, or by fresh-water bays in which 
the shells of fresh-water mussels accumulated. 
The finer deposits indicate shallow, muddy 
bays, in which the remains of the larger quad- 
rupeds are occasionally found in positions 
indicating that they had been mired in a 
standing posture. The old stream channels 
have yielded remains of several species of 
bowfins (Amiidae), garpikes (Lepidosteus) , 
and siluroids. Crocodiles were numerous and 
diversified. The reptiles suggest that the 
climate was Floridian, or south temperate, and 
we may picture a partly open, partly forested 
country, somewhat similar to the existing 
bayou region of the Mississippi Delta of 
Louisiana. Analysis of the Testudinata by 
Hay (1908.1) has also afforded a clear idea of 
the physiographic conditions in Bridger time. 
The soft-shelled river turtles (Trionychoidea) 
were represented by at least 25 species, and 
there are now in the world only 26; the 
Bridger rivers and brooks fairly swarmed 
with these creatures, some of them equal in size 
to the largest existing Asiatic species. There 
are indications of 4 species of the family 
Emydidae (order Cryptodira), as compared 
with the 12 species now living in the Missis- 
sippi Valley. The genus Baptemys, of the 
same order, has its nearest relatives at present 
in Central America, and a third genus 
(Anosteira) is reported by Lydekker in the 
upper Eocene of England. The presence of 
: and D, extcusive stretches of land is indicated by the 
true land tortoises (Testudinidae) of the genus 
Hadrianus, including giant tortoises nearly 3 feet 
long, which probably lived on dry lands bordering 
the sluggish Bridger streams. The ancient Lower 
Cretaceous order Amphichelydia is also represented 
here by four species belonging to two genera. 

The environmental adaptations of the animals of 
the Bridger Basin were classified by Matthew (1901.1, 
pp. 309, 310) as follow?: 
Land animals: 

1. Aerial: Remains of birds rare and fragmentary, as in 
nearly all geologic formations. 



ENVIHONMENT OF THE TITANOTHEEES 



81 



2. Arboreal: Primates, many Carnivora, and some Insec- 

tivora and Rodentia. Out of 1,007 specimens, belong- 
ing to 46 genera, 13 genera (184 specimens) were 
certainly arboreal and 11 genera (485 specimens) 
were probably arboreal. 

3. Terrestrial (cursorial and ambulatory) : Some of the 

carnivores and all the ungulates (17 genera, 314 
specimens). Also some lizards and chelonians. 

4. Fossorial: Certainly fossorial, 3 genera (S specimens). 

Some of the insectivores may also have been fossorial. 

5. Amphibious: One insectivore (Pantolestes) (1 genus). 

Probably certain carnivores. 
Water animals: 

6. Fresh- water: Numerous crocodiles, aquatic turtles, 

fish, and fresh-water mollusks. 

7. Marine: No marine animals. Contrast this lack of 

types with the types of fish in the preceding Green 
River formation. 

The Bridger life thus included many arboreal, 
terrestrial, and aquatic forms, the last mostly reptiles, 
fishes, and invertebrates. The slow-moving, ambula- 
tory quadrupeds form a relatively large proportion of 
the mammals, but the cursorial types, such as the 
Equidae (Orohippus), are relatively rare; also the fos- 
sorial types. The Bridger life seems to be that of a 
partly forested flood plain. The remains of large 
mammals are so numerous as to indicate abundant 
open, gladed areas, comparable to the partly forested 
and partly open delta regions along certain rivers of 
modern time. 

The foot structure of the Bridger quadrupeds gives 
less certain evidence of an open plains country, 
favorable to cursorial types, than that of the Wasatch 
(lower Eocene) quadrupeds of the same region. 

No impressions of leaves from the Bridger forests 
have been discovered. It is probable that the forests 
of Green River type, described on pages 72-73, per- 
sisted into Bridger time and that the climate then 
was warm-temperate, almost subtropical. 

The faunal history of the Bridger as a whole shows 
a gradual reduction in the number of archaic mammals 
of Mesozoic stock and a steady increase in the number 
of their competitors among the modernized mammals, 
the numerical relations between these two groups in 
upper Bridger time being as follows: 

Genera Species 

Archaic mammals 15 35 

Modernized mammals 57 146 



Duration oj the Bridger epoch. — Matthew (1909.1), 
following the earlier geologists, believes that the 
lacustrine conditions in Green River time arose from 
the uplift of the Uinta Mountain range, which blocked 
the basin and caused the formation of the great lake 
in which the material that formed Green River shale 
was laid down. As the river gradually cut its way 
through the east end of the Uinta Range the lake 
gave way to the broad Bridger flood plain, in which 
was deposited the volcanic ash washed down from 
the slopes of the Uinta Mountains to the south, and 



the deposit was worked over and sorted by the streams 
that flowed across the plain. The Bridger Basin was 
subject to intermittent overflow, which gave rise to 
large but shallow lakes of clear water. If we should 
assume that the Bridger formation occupied one- 
tenth of total estimated Eocene time — 90,000 to 
100,000 years — the fossiliferous beds, which are 1,100 
feet thick, have accumulated at an average rate of 
12 inches per century. This estimate would allow 
110,000 years for the deposition of the Bridger forma- 
tion exclusive of the "white layers " formed at intervals 
when deposition was arrested. The titanothere re- 
mains of the Bridger indicate a long period of evolu- 
tion, but not so long as that of the Chadron (lower 
Oligocene). 

Chief localities and exposures of the Bridger formation in ike 
Bridger Basin 

Bridger E: 

Uppermost exposures: 

Sage Creek Mountain. 
Henrys Fork Table. 
Twin Buttes. 

Bridger D: 

Upper exposures: Level 

Twin Buttes D 1-5 

Spanish John's Meadow D 1-5 

Cat Tail Spring D 1-5 

Henrys Fork, Burnt Fork post office D 1-5 

Henrys Fork, Lone Tree post office D 1-5 

Summers Dry Creek D 1-5 

Henrys Fork Hill D 1-5 

Sage Creek Spring D 1-5 

Lane Meadow D 1-5 

Bridger C: 

Lower exposures: 

Henrys Fork, Lone Tree post office C 4-5 

Lane Meadow C 3-5 

Spanish John's Meadow C 3-5 

Henrys Fork Hill C 3-5 

Twin Buttes C 1-5 

Dry Creek C 1-5 

Henrys Fork, Burnt Fork post office C 1-5 

Church Buttes, third bench C 1-3 

Bridger B: 

Upper exposures : 

Cottonwood Creek Typical B 4^5 

Millers ville, 6 miles southeast of B 4-5 

Cottonwood Creek, middle of B3 

Grizzly Buttes B 3 

Church Buttes B 2-3 

Lower exposures: 

Cottonwood Creek B2 

Grizzly Buttes Typical B 2 

Exposure B, 5 miles south of Granger B 2 

Millersville B 1-2 

Cottonwood Corral, Blacks Fork ,. B 1-3 

Exposure A, 5 miles south of Granger B 1 

Church Buttes B 1 

Bridger A: 

Hams Fork Bluff; Granger to Opal, 25 miles. 

Mouth of Big Sandy Creek {IPalaeosyops -fontinalis 

zone). 
Big Muddy exposures between Carter and Granger. 
Blacks Fork Bluffs, east of Granger. 



82 



TITANOTHERES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



ZONE 10: EOMETAEHINUS-TEOGOS0S-PALAEOSYOPS FONTINAIIS ZONE 
[Bridger A and Huerfano B ; lower Lutetian of Europe] 

The lower Bridger (Bridger A and B) corresponds 
with the "calcaire grossier superieur" of the Paris 
Basin. The correlation of Bridger A with Huerfano 



of some 200 feet of calcareous shale alternating with 
tuff (Matthew, 1909.1), which are exposed principally 
around the eastern, northern, and western margins of 
the Bridger Basin. It is therefore supposed that 
Bridger A, which passes down into the lacustrine 
Green River shales, is partly of lacustrine, partly of 




Figure 56. — Map of the Bridger Basin, Wyo. (No. 8, fig. 35) 

Showing the principal topographic features, Twin Buttes and Henrys Fork Table, and a diagrammatic section of the Bridger formation (A, B, C, D, and E) capped by 
the Bishop {"Wyoming") conglomerate (W). After Matthew and Granger, 1902, 1909. 



B has recently been established through the discovery 
in each of the mammalian species Palaeosyops (Lim- 
nohyops) fontinalis Cope. (Osborn, 1919.494.) In 
these beds vertebrate fossils are rare and include, 
besides the titanothere above mentioned, remains of 
crocodiles, turtles, and fishes only. Bridger A consists 



fluviatile origin and is transitional both geologically 
and in its fauna between Green River (upper Wind 
River) and Bridger B time. Sinclair describes this 
horizon as consisting of "buff and pale-green tuffaceous 
shales and sandstones, often containing in enormous 
numbers shells of Paludina and Unio." 



ENVIRONMENT OF THE TITANOTHEEES 



83 




z 

LJ 
O 
O 
liJ 

U 

_J 
Q 
Q 



no 
q: 

DQ 



B3 



Pa/aeosyops 
majo/; neofype 

{?)Limnohyops 
laevidens, type 



Palaeosypps palicdosits- 



Pa/aeosyops major, ref. 
Limnohyops monoconus, type 
L/mnohyops matfhewi, type 

L/mnohyops priscus, type 



Limnohyops 
monoconus 




OroTiippus typicus 



GRIZZLY BUTTES FAUNA 

Notharctus 

Harpagolestes 

Hyradiyus agrarizis 

MetadieiroTnys dasypus 

Orohippus atavus 

TUlothjeriLawfodiens 



A5 



,'' Pa/aeosyops 
fonf/na/is 



cr 

Uo 

q: 

CD 



NO MAMMALS 



oo 



cc 



^ LL t_> 

u bj F 

UJ > < 

a ^ 5 

o '^ e 



Figure 57. — Section of the lower part of the Bridger formation in the Bridger Basin, Wyo. (No. 8, fig. 35), 

showing the succession of the species of titanotheres and other mammals 
The section is 650 feet thick. The principal geologic features are represented in the center. After the studies of Osborn, Granger, and Matthew. 



84 



TITANOTHEBES OP ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



ZONE 11: PAIEOSYOPS PALUDOSUS-OEOHIPPUS ZONE 
[Bridger B; upper Intetian of Europe] 

The richly fossiliferous deposits belonging to the 
Paleosyops paludosus-OroMppus zone (Bridger B) 
are exposed chiefly in the northern area of the Bridger 
formation, near Fort Bridger, along Blacks Fork and 
its tributaries. They represent a very long period and 
consist of 450 feet of tuffs and sandstones (fig. 4) 
divided into two principal escarpments, which are 
separated by Cottonwood Creek valley. 

In this zone the titanotheres and other mammals 
undergo notable progressive evolution, and there is a 
marked succession of species. (See fig. 57.) 

The succession of the species of titanotheres in 
Bridger B, in descending geologic order, is as follows: 



Limnohyops monoconus Os- 
born, type. 

Limnohyops matthewi Osborn, 
type. 

Palaeosyops paludosus Leidy. 

Palaeosyops paludosus, re- 
ferred. 

Palaeosyops paludosus, type. 



?Mesatirhmus Junius Leidy. 

Palaeosyops major Leidy, hy- 
potype. 

Limnohyops laevidens Cope, 
type. 

Palaeosyops major Leidy, re- 
ferred. 

Limnohyops monoconus Os- 
born, referred. 

The species of titanotheres found in Bridger B 
belong exclusively to the subfamily Palaeosyopinae 
and represent the two generic branches Palaeosyops 
and LimnoTiyops, closely related animals with broad 
spreading feet and heavy limbs, slow in gait. The 
reference to Mesatirhinus of the species P. Junius 
Leidy is somewhat doubtful. The lower half of 
Bridger B at Grizzly Buttes (PI. VII, B) , an escarp- 
ment along Smiths Fork, is by far the richest collect- 
ing ground in the Bridger Basin; thousands of speci- 
mens have been taken from it, including many more 
or less complete skulls and skeletons, all recorded 
from Bridger B 2. Beds at a slightly higher level, in 
Bridger B 2 and in Bridger B 3, on the escarpment 
along Cottonwood Creek, have yielded a number of 
complete skeletons, including those of several species 
of Equidae (OroTiippus), a variety of catlike and dog- 
like creodonts {Limnocyon, Harpagolestes), abundant 
small civet-like creodonts (Viverravus, Sinopa), an- 
cestral canids (Miacis, Uintacyon), a surviving (?) 
condylarth (Hyopsodus) ; also ancestral Edentata 
{Metacheiromys, armadillo-like) and the rodent-like 
tillodonts {Tilloiherium fodiens, Trogosus). They 
have also yielded many rodents (Paramys, Sciuravus), 
as well as a rich primate fauna of lemuroids (Noth- 
arctus). The entire fauna has been very carefully 
reviewed and analyzed by Matthew (1909.1, pp. 
298-302). 

Rich as is the fossil life of the lower Bridger, many 
mammalian subfamilies and many genera and species 
are lacking which occur abundantly in the upper 
Bridger. Noticeable is the absence of uintatheres 
(Uintatherium) and of three important genera of ti- 
tanotheres {Manteoceras, Telmatherium, Mesatirhinus), 
which appear abundantly in the upper Bridger. 



The "Cottonwood Creek white layer," marking 
the summit of Bridger B, indicates a long period of 
shallow lake flooding of the Bridger Basin during 
which the large amblypod uintatheres and the more 
advanced titanotheres entered the basin. Vintaihe- 
rium is not found in Bridger B, but it occurs at the 
very base of Bridger C, the lowest level of the upper 
Bridger. 

ZONE 12: UINTATHERIUM-MANTEOCERAS-MESATniHINUS ZONE 
[Bridger C and D, Wasliakie A, and Uinta A; part of Bartonian of Europe] 

The fauna of zone 12 in the Bridger Basin, which 
includes deposits 725 feet thick (Bridger D, 375 feet; 
Bridger C, 350 feet; see fig. 58), may be clearly dis- 
tinguished from that of zone 1 1 (lower Bridger = Bridger 
B and A) by its content of the remains of the animals 
listed below: 

Titanotheres : 

Palaeosyops robustus Leidy. 

Palaeosyops copei, type. 

?Telmatherium validum, type. 

Manteoceras manteoceras. 

Mesatirhinus petersoni, type. 

Palaeosyops leidyi, type. 

Limnohyops laticeps, type. 

Mesatirhinus megarhinus, type. 

?Telmatherium cultridens. 
Other mammals : 

Hyrachyus princeps (cursorial rhinoceros). 

Patriofelis ferox (catlike creodont). 

Isectolophus latidens (tapir). 

Uintatherium robustum (four-horned amblj-pod) . 

Notharctus crassus (large lemuroid). 

Pantolestes natans (aquatic insectivore) . 

Homacodon vagans (primitive artiodactyl) . 

LTintatherium mirabile (amblj'pod uintathere). 

Orohippus sylvaticus (primitive equine). 

Bridger C. — The lowest beds of the horizon Icnown 
as Bridger C are exposed at the foot of Sage Creek 
Mountain, along the southern slope of Henrys Fork 
Table; also at the foot of Twin Buttes and along the 
slopes north of Twin Buttes. They consist of 350 feet 
of gray and greenish-gray tuffs, divided into a lower 
and an upper half by the "Burnt Fork white layer" 
and bounded above by the "Lone Tree white layer." 
After careful analysis of the fauna of Bridger C, Mat- 
thew concluded (1909.1, p. 304) that its marked dis- 
tinction from the fauna of Bridger B was due to the 
immigration of several new genera into the Bridger 
Basin. Among these especially are the titanothere 
genera Manteoceras, TelmatJieriurn, and MesatirJiinus, 
which appear to be really newcomers and not in any 
sense descendants of the lower Bridger genera Palaeo- 
syops and Limnohyops. The two genera last named, 
however, are represented in Bridger C by distinctly 
new specific forms, much more progressive than those 
in Bridger B. Thus Bridger C is characterized both 
by marked evolutionary changes in mammals that pass 
over from the lower levels and by the introduction of 
a fauna that is more or less new. Of this new fauna 



BNVIKONMENT OP THE TITANOTHERES 



85 



TJintaiherium is closely related to the ancestral BatJiyop- 
sis, which is found in the Wind River Lambdotherium 
zone and in the long antecedent first Wasatch zone. 
The pseudotapir Isedoloiyhus is related in tooth struc- 
ture to Systemodon, which is characteristic of the third 
Wasatch zone. We are therefore disposed to regard 
the life of the upper Bridger ( Uintatherium) zone as 
the result of a local immigration from the adjacent 
Rocky Mountain or Plains region into the Bridger 
Basin, and not as the result of a continental immigra- 
tion such as is made manifest in the lower Eocene. 

Bridger D. — Upon the "Lone Tree white layer" lie 
the 375 feet of strata that form Bridger D, in which 
are found five faunistic levels, D 1 to D 5. The fos- 
siliferous sediments of this closing period of the 
Bridger consist of 350 feet of "gray and greenish-gray 
sandy and clayey tuffs, with one or more ash beds," 
including the upper "white layer," which lies about 75 
feet below the top of the formation. Among the tita- 
notheres of this zone are descendants of species of 
Palaeosyops, Limnohyops, Manteoceras, and Telma- 
therium, which continue to increase in size and which 
represent advancing mutations that are exhibited in 
the comparative measurements shown in the tables on 
pages 304, 313, 341, 364. It is noteworthy that there 
is no very marked faunistic change in the species of 
titanotheres that persisted from Bridger C to Bridger D. 
For example, Manteoceras manteoceras persists from the 
lower to the higher levels, and Mesatirhinus peter soni is 
recorded in both C 2 and D 3. Exceptions to this 
slow evolution are seen in two species — Palaeosyops 
copei, which represents in certain characters an ad- 
vanced stage of evolution allied to a stage found in the 
lower sediments of the Washakie Basin, and Telma- 
therium validum, assigned to Bridger D, which shows a 
distinct advance upon Telmaiherium cultridens, as- 
signed to Bridger C 5. 

Bridger E. — Bridger E is theoretically correlated 
with Washakie B and Uinta B (upper Eocene). The 
topmost beds of the Bridger formation, 500 feet thick, 
include sediments that are almost barren of fossils, 
but the few fragments of mammals they have yielded 
are of undoubted Bridger age. The 500 feet of soft 
banded tuff containing at intervals thick layers of 
volcanic ash indicate increasingly active volcanism. 
The layers of gypsum found at this horizon were 
probably deposited in playa lakes (Sinclair, 1906.1), 
like those in the Humboldt Basin of the present time. 
The dark-red bands in Bridger E may indicate an arid 
climate. The correlation of Bridger E with Washakie 
B, to the east, is purely conjectural, for neither con- 
tains determinable remains of mammals. Matthew 
(1909.1, p. 306) attributes the paucity of life in this 
zone to violent volcanic eruptions, observing that the 
thick and generally unsorted beds of ash indicate great 
volcanic activity and that the large amount of gyp- 
sum and the absence of fossils might be due to the 



consequent destruction of vegetal and animal life, 
which converted the region into a barren plain that 
was alternately submerged and desiccated. 

The UintatJierium zone in the Washakie Basin 
(Washakie A) is described on pages 85, 87, in the 
description of the deposits of that basin. The barren 
deposits in the Uinta Basin (Uinta A) that correspond 
to the Uintatherium zone are described on pages 
91-92, in the description of the Uinta Basin. 

WASHAKIE BASIN, WYO. 
STRATIGRAPHY OF THE BASIW 

Deposits and faunal zones. — The Washakie Basin 
lies about 50 miles east of the Bridger Basin, and the 
two contain similar volcanic sediments. The basin 
was described by Hayden in 1869-70 (1871.2, p. 73), 
and more fully by Cope in 1873 (1873.4). Its fau- 
nistic levels were studied by the Princeton expedition 
(Osborn and McMaster, 1881.8) and by expeditions 
of the American Museum of Natural History, under 
Wortman (1893, 1895) and Granger (1906). Granger 
(1909.1, pp. 13-32) gave the first complete and accu- 
rate description of the geology of the Washakie Basin 
in his "Faujial horizons of the Washakie formation 
of southern Wyoming" (1909.1, pp. 13-32). King 
treated the deposits of the Washakie Basin as of 
Bridger age and of lacustrine origin. Osborn (Osborn 
and McMaster, 1881.8) favored the theory of separate 
deposition, and Scott (1899.1) showed that where the 
fauna of the Washakie Basin departs from that of the 
Bridger it approaches that of the Uinta. The dis- 
covery of the true upper Bridger fauna in horizon A 
of the Washakie Basin was due to the American 
Museum expeditions of 1893, 1895, under Wortman. 

The Washakie Basin, with its vivid coloring and its 
alternation of hard and soft layers of tttft' and sand- 
stone, affords the most picturesque geologic views to 
be found in the Rocky Mountain Eocene basins. 
Haystack Mountain ("Mammoth Buttes" of Cope), 
a long ridge of badlands near the north end of the basin, 
which in places rises 400 feet above the plain, forms 
the northern border of an extensive semicircular 
"central basin" that has the appearance of a gigantic 
crater. The floor of this basin is rather level and 
regular, being broken only by a few low tables and 
buttes, which have long been preserved from erosion 
by their capping of hard sandstone, though their 
sides are trenched by innumerable deep, vertical- 
walled canyons, which present a great variety of 
architectural forms that are illuminated by brilliant 
coloring. 

Washalcie A {TJintaiherium zone, middle Eocene). — ' 
The "lower brown sandstone" of the Washakie Basin, 
known as Washakie A (fig. 60), contains the fauna of the 
Uintatherium- Manteoceras- Mesatirhinus zone. It was 
deposited contemporaneously with the upper Bridger 
(Bridger D), to the west, and probably with the non- 



86 



TITANOTHEKES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



3 lIj 




u 

z 

UJ 

u 

O 

Ld 

U 

-J 
Q 
Q 



D3 



TITANOTHERES GEOLOGIC SECTION OTHER MAMMALS 



Pa/aeosyops robustus 
Mesai/rh/nusjan/us 



Pa/aeosyops robusius 



Pa/aeosyops cope/, type 

" robustus 
Mesat/r/j/nuspetersor?/, type 



Pa/aeosyops /e/dy/ 

Manteoceras manteoceras 

Limnohyops /at/ceps 




Pa/aeosyops /e/'c/yJ 




No t/i a/ 'c/./JS CI -assi fs 
Pairiofclis fcro.x- 



Z//'/ilcU/ior/u/n 
leic(ya/iurrh 

(?JJsec/o/op/ias latldens 



~~ Uintat/ieriL/yn. 

HENRYS FORK 
H/LL 
L ONE TREE WH/TE Z A YER 



UirifafJteri'iim att/rrps 
'^'Elac/i oce? -as "par vu/ri 



C5 



fPj Pa/aeosyops /e/oy/ 
'Je/mather/umcu/ir/dens,type 
/^esat/rh/nus megarb 's 



xoTie -- 



Pa/aeosyops /e/dyi, type 



CD 




C3 



C2 



BURNT FORK WH /TE LAYER 

? Limnohyops /at/ceps,type ^L,^,j^j_ i— j_j.^ ' 



Te/matber/um cu/tr/c/ens 
/^esatirb/nus peterson/ 



Pa/aeosyops granger/.tjpe 
?/i^anteoceras manteoceras 




Homacodon vaf/ans 



Jfyraehyus inipe/ia/L-i 
Oro/i/ppiis sy/vat/C7/s 



COTTONWOOD W/i/TE LAYER BENCH 



UintaiJi eriurn 



^^ B5 



_SANO^.O.fii£ 

Palaeosyops paZudosits- 
Ir^roTiippiis :zoTie 

\ ^ ' / COTTONWOOD 



CREEK BENCH 



Figure 58.— Section of the upper part of the Bridger formation in the Bridger Basin, Wyo. 

Shows the vertical distribution of the titanothere species on the left, the principal geologic features in the center, and the distribution of the 

other species of mammals on the right. Principally after Osborn, Granger, and Matthew. 



ENVIRONMENT OP THE TITANOTHERES 



87 



fossiliferous Uinta A, to the south. Its contempo- 
raneity with Bridger D is established through the 
common presence of the following species: 



Uintatherium robustum. 
Uintatherium mirabile. 
Manteoceras manteoceras. 
Notharctus tenebrosus Leidy. 
Hyrachyus princeps Leidy. 
Sinopa. 



Palaeosyops copei Osborn. 
Mesatirhinus megarhinus. 
Mesatirhinus petersoni. 
Hyopsodus. 
Paramys. 



I bench, which constitutes the lower rim of the basin 
I on its northern border. This "lower brown sand- 
I stone" passes at a low angle southward beneath the 
floor of the basin. Below it, and apparently conform- 
able with it, lie gray sandy shales, which are pro- 
visionally referred to the Green River but which 
were probably laid down in lower Bridger time 
(Bridger A and perhaps Bridger B). As these deposits 
I show no marked evidence of erosion it seems probable 



fShale? 
Nos. 35,36,37 

[Sandstone 



Nos. 25, 26a, 26 b 



Nos. 20,21, 22 




STACK MT^ 



LOWER BROV 



Figure 59. — Diagrammatic vertical section of deposits near Barrel Springs, Washakie 
Basin, southern Wyoming 

Shows the alternation of tuffs, siliceous, calcareous, and sandstone materials. Johannsen (1914.1), after 
Granger, with modifications. The numbers refer to lithologic specimens examined by Johannsen. 



This fauna of the Uintatherium zone occurs in 260 
feet of Washalde A, which is composed largely of 
altered eruptive rocks, probably dacite tuffs, of cal- 
careous and siliceous shales, and of glass tuffs mingled 
with grains of quartz, hornblende, feldspar, according 
to the analysis of Johannsen (1914.1, p. 214). 

The "lower brown sandstone" layer yields a rich 
fauna of uintatheres. This layer forms a persistent 



that the Washakie Basin was filled wi^h a lake in 
Green River time whUe Bridger A was being deposited 
to the west. 

Washakie B {Metarhinus and Eohasileus-DolicJio- 
rhinus zones, upper Eocene). — The upper Eocene 
Washakie B horizon is described on pages 89-90, in 
the description of upper Eocene faunal zones 13 and 
14, to which it belongs. 



TITANOTHEEES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 




Figure 60. — Diagrammatic horizontal section of the Washaliie Basin, southern Wyoming, from north to 



After Granger (1909.1). This section stiows tlie Uintatheriiim- Manteoceras zone (Washakie A, lower brown sandstones), discovered by the American Museum in 
1893; Metarhinus zone (Washalcie B 1), base of the original "Washakie" formation of Hayden and Cope; DolichoThinus-Eobasileus zone (Washakie B 2), sumrnit 
of the original "Washakie" formation of Hayden and Cope; "Adobe Town," rougtJy eroded area in which Amymion antiquus, Achaenodon, etc., were dis- 
covered by the Princeton espedition of 1878. The numbers show locations of lithologic specimens studied by Johannsen. 




FiGTJEE 61. — Sketch map of the Washakie Basin region, in southern Wyoming 

After Granger (1909.1) from Clarence King (1876.1). The shaded area is the "Washakie" formation of Hayden, mapped by King and 
the United States Geological Survey as the Bridger formation. 



ENVIRONMENT OF THE TITANOTHERES 

Mammalian life of the WashaMe Basin 



89 





Washakie A (Uintatherium-Manteooeras-Mesatirhinus zone). 


Washakie B 1 and B 2 (Eobasileus-Dolichorhinus zone and Meta- 




These forms are found also in Bridger C and D, to the west 


rhinus zone). These forms are found also in Uinta A and B, 
south of the Uinta Range. 


Primates : 






Lemnroids 


Notharctus sp. 
Hemiacodon sp. 








Rodents 


Paramys cf. P. delicatus. 


Paramys leptodus, type. 








Paramys grangeri. 


Carnivores : 






Creodonts 


Thinocyon cledentis, type. 


Harpagolestes immanis. (Giant creodont of the 






Patriofelis ferox. 


family Mesonychidae.) 




Sinopa rapax var. lania, type. 






Synoplotherium lanius, type. 




Miacids (doglike 


Miacis washakius, type. 


Limnocyon potens. (An oxyaenid creodont.) 


carnivores) . 


Miacis medius. 
Oodectes? pugnax, type. 




Condy larths 


Hyopsodus cf. H. despiciens. 
Uintatherium grande, type. 




Ambly pods 


Eobasileus cornutus, type. (Giant amblypod 






Uintatherium speirianum, type. 


with the front horn directly above the eyes; 

first appearance.) 
Eobasileus galeatus, type. 
Eobasileus furcatus, type. 
Eobasileus pressicornis, type. 


Artiodactyls 


Homacodon sp. 


Achaenodon insolens, type. (First of the elo- 








theres.) 






Achaenodon robustus, type. 






?Protylopus sp. (A cameloid form.) 


Perissodactyls : 






Titanotheres 


Palaeosyops copei? (Last of Palaeosyops. 
Palaeosyops sp. 

Manteoceras manteoceras, type. ("Prophet- 
horn" titanotheres.) 
Manteoceras washakiensis, type. 






Mesatirhinus megarhinus, type. (Ancestor of 


Metarhinus earlei, type. (Fluviatile type.) 




DoHchorhinus.) 


DoUchorhinus hyognathus, type. (Dohchorhi- 




Mesatirhinus petersoni. 


nus cornutus stage.) 
Dolichorhinus vallidens, type. 


Rhinoceroses and rhi- 


Hyrachyus sp. (Cursorial rhinoceros.) 


Hyrachyus sp. (Cursorial rhinoceros of Bridge 


noceratoids. 


Triplopus cubitalis. 


type.) 
Triplopus sp. 
Amynodon antiquus, type. (First of the amyno- 

donts (aquatic rhinoceroses).) 


Chalicotheroids 


Eomoropus amarorum, type. (Forest-living 
ancestral chalicothere; ancestor of Moropus.) 




Lophiodonts 


Helaletes sp. 

Desmatotherium guyoti, type. 
Dilophodon minusculus, type. 
Dilophodon minusculus? 





ZONES 13 AND 14: METARHINUS ZONE AND EOBASHEUS-DOIICHORHmUS 
ZONE 

[Uinta B 1 and Washakie B 1; Uinta B 8] 

The great life division known as Washalde B, 380 
feet in thickness, contains a new and dominant fauna, 
which is not represented at all in Bridger D or Wash- 
akie A. It is significant that this unit is divided into 
two zones by its fauna, exactly as Uinta B is divided 
into two zones, the Eoiasileus-DoIichorMnus zone 
(Washakie B 2 = Uinta B 2), and the Metarhinus 
zone (Washakie B 1 = Uinta B 1 = (in part) Bartonian 
of Europe). Certain of the older mammalian families 
and genera (as Uintatherium) begin to disappear and 
101959— 29— VOL 1 S 



new generic and specific forms replace them. Con- 
spicuous among these is the amblypod Eobasileus, 
first described from this region by Cope, which re- 
places Uintatherium. A full list of this fauna is given 
above. Among the distinctive forms are the fol- 
lowing : 



Eobasileus cornutus Cope, 

type; 
DoHchorhinus hyognathus 

Scott and Osborn, type. 
Uintatherium speirianum 

Osborn, type. 
Triplopus cubitalis Cope, type. 



Eomoropus amarorum Cope, 

type. 
Amynodon antiquus Scott and 

Osborn. 
Achaenodon insolens Cope. 
Metarhinus earlei Osborn, 

type. 



90 



TITANOTHEKES OP ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



Washakie A is characterized by "rusty brown nodu- 
lar sandstones," and Washakie B by "coarse white, 
pink, and sabnon-colored sandstones" and by "the 
extremely coarse green sandstones or feldspar con- 
glomerates. The rocks first recognized as sandstones 



GEOLOGIC LEVELS OFSPECIES 



SUMMIT OF HAYSTACK M 



Zp^y\Eobas/7eus comufus, type 

"°?S-<5-v>.\ EobcLsileiLS- 
ict)_i-s^E=^ Dolichorhimis 
zone 



ZONAL LEVEL APPROXinATE 



Eomoropus amarorum, type 
Lepforeodon marsh/, type 



\Do/ichorhinus hyoqnathus, type level 
(A.M. No. 13164, Co/U906) 



'■ a.^i^?l%'^'^^^i^hMelarl>/nus earlei, type 



L£V£L OF 

S^r^ A my no don 

_=ti -"^ —^''^pnt/quas type 

ADOBE TOW/V 

Achaenodon insolens, type 




-■-'■'■ -~ '^-^M anleoceras washak/ensis, type level 
Winlalherium speirlanum, type level 



Mesatirhinus megarhinus 

type 

Palaeosyops cope/ 

Manteoceras manteoceras 

type 

Uintafherium sp. div. 



Figure 62. — Columnar section of the Washakie Basin, Wyo. 

life zones 

Shows the principal genera of the lower and upper life zones and the actual level of certain characteristic species. 
Chiefly after Granger (1909.1). This section includes the Uintaiherium zone (Washakie A), lower brown sand- 
stones; Meiarhinus zone (Washakie B 1); and Eobasileus-DolichorMnus zone (Washakie B 2) , upper gray-green 
beds. ' Numbers in column show position of lithologic specimens examined by Johannsen. 

prove to be interspersed with dacite and glass tuffs." 
(Johannsen, 1914.1, p. 215.) The sandstones, which 
were derived from granite by erosion, consist of grains 
of quartz, hornblende, and feldspar embedded in a 
shghtly devitrified groundmass. (See PI. IX.) 



The composition of these sediments indicates the 
presence in this region of active volcanoes, which 
were discharging great clouds of dust. Unlike the 
sediments of the Bridger Basin the sediments of the 
Washakie Basin were deposited in rather turbulent 
water and contain none of the 
"white layers" that indicate 
the still water that prevailed 
in the upper Bridger. Turbu- 
lent water is not favorable 
to the preservation of the 
remains of small mammals. 
Only one of the smaller 
perissodactyl ungulates 
(Triplopus) has been found, 
and no remains of Equidae. 
The first aquatic rhinoceroses 
{Amynodon) belong to a river- 
frequenting type; the first of 
the entelodonts {Achaenodon) 
is also a river-frequenting 
form; the first of the forest- 
dwelling chalicotheres {Eomo- 
ropus) also occurs. Thus the 
Washakie Basin has preserved 
for us mainly the larger 
swamp and river-border fauna 
but has yielded little record 
of either the arboreal or 
plains -living cursorial fauna 
of the time. 

In the fauna of the Wa- 
shakie Basin (a list of which 
is given in the table on p. 89) 
the large hoofed animals pre- 
dominate, especially those 
adapted to stream borders, 
swampy land, rivers, and 
streams. A small fauna of in- 
sectivores, lemuroids, carni- 
vores, and ancestral artiodac- 
tyls doubtless abounded, but 
the environment was unfavor- 
able to the preservation of 
such remains, and the micro- 
fauna has been found only 
rarely. The small titanothere 
MetarUnus is highly distinc- 
tive of this Washakie B 1 life 
zone. (Kiggs, 1912.1.) 

Uinta .B.— Exactly the 
same physiographic condi- 
tions prevailed at the same time in the great basin 
south of the Uinta Mountams while the sediments 
known as Uinta B were being deposited. These sedi- 
ments, which have a combined thickness of 800 feet, 
contain exactly the same riparian fauna, including a 



35), showing 



ENVIRONMENT OF THE TITANOTHERES 



91 



AMVNODON SANDSrONE 



large number of identical species, and therefore con- 
stitute an extension of the Eobasileus-DolichorMnus 
and Metarhinus life zones to the south. The fauna 
and deposits of Uinta B are more fully described on 
pages 91-99, in the description of the Uinta Basin. 

UINTA BASIN, UTAH 

PHYSIOGRAPHIC, CUMATIC, AND VOLCANIC CONDITIONS IN THE UINTA 
BASIN DURING MIDDLE (?) AND LATER EOCENE TIME 

It is a striking fact that the later Eocene sediments 
in the Uinta Basin are composed mainly of altered 
eruptives, probably dacite 
tuffs, as indicated by analyses 
of nine samples by Johannsen 
(1914.1, pp. 212-214). The 
rocks of the lower levels ^ 
described as "brown sand- 
stones" comparable in litho- 
logic appearance to Washakie 
A, contain a large element of 
tuff and consist microscop- 
ically of irregularly broken 
and rounded fragments of 
quartz, lime-soda feldspar, 
hornblende, biotite, and frag- 
ments of andesite or basalt in 
a brown groundmass, which 
is chiefly chlorite but contains 
some calcite. On the lower 
levels (in Uinta A) brown is 
the prevailing color, as in 
Washakie A. In Uinta B 
sediments of this color pass 
into pinkish-brown and red- 
dish-brown sediments, and in 
Uinta C into pale-green and 
gray fine-grained rocks con- 
taining considerable glass. 
Many rocks that look like 
sandstones prove under the 
microscope to resemble flow 
breccias. 



Uinta A as now defined is entirely unfossiliferous 
but is here correlated with the middle Eocene fossil- 
iferous horizon A of the Washakie Basin {Uintathe- 
rium zone). 

Uinta B 1 (in some previous reports included in 
Uinta A) contains a rich river-border fauna, like that 
of Washakie B 1. 

Uinta B 2 (formerly constituting all of Uinta B) 
contains a larger land and river-border fauna, like 
that of Washakie B 2. 




BARREN 



{DiplacodoTV- 
EpUtippus 
zone 



EobcLsiZeus - 

DolichorTuruLS 

zoTte 



TYPICAL U/NTA MEADOW FAUNA 



TRANSITION FAUNA 



^:metarhinus sandstone" 

'XfluviatiTe' ^=r 

£^?^ ^_^1 MetarhiniMS 




FLU VI ATI LE FAUNA 



Figure 63. — Diagrammatic section of the Uinta formation exposed in tiie nortii wall of 
White River Canyon 3 miles below mouth of Evacuation Creek, Utah 



GEOLOGIC HORIZONS IN THE UINTA 
BASIN 



Uinta fauna of Marsh 100 feet above "Amynodon sandstone." 



The deposits of horizons 
A and B of the Uinta Basin 
are not those of the typical Uinta formation of Marsh 
(1871.3), of King (1878), or of Scott and Osborn 
(1891.1), all of which belong to Uinta C, the Diplaco- 
don zone; they form the lower part of the section 
(Uinta A and Uinta B), determined by the American 
Museum expedition of 1894 under Peterson (Osborn, 
1895.98) and successively explored with remarkable 
results by Peterson, Douglass, and Riggs, whose obser- 
vations and exact records of the vertical distribution 
of genera and species have firmly established the 
stratigraphy of the Uinta Basin section as presented 
in Figure 65. (See PL IX.) 



After observations of Peterson, Douglass, and Riggs. Uinta A, columnar sandstones, unfossiliferous; Uinta B 1, MetarUnus 
zone capped by ** Metarhinus sandstone," containing a fiuviatile fauna; Uinta B 2, Eobasileus-DaUchorhinus zone, capped 
by "Amynodon sandstone," containing a transition fauna; Uinta C, Diplacodon-EpiMppus zone, containing the typical 



Uinta C contains the typical Uinta (Diplacodon) 
fauna. 

The sediments in the Uinta Basin between the 
Diplacodon zone and the Green River formation were 
classified by White (1878.1, p. 37) as Bridger, although 
no fossils were found in it, and wore treated as con- 
temporaneous with the Bridger deposits. We now 
know that the sediments that form Uinta B were cer- 
tainly laid down after Bridger C and D had been 
deposited, but they may have been contemporaneous 
with the unfossiliferous Bridger E. During the 
American Museum explorations of 1893-94 Peterson 



92 



TITANOTHERES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



discovered 150 fossil mammals in the lower series, I by the subsequent explorations and publications of 
which were first correlated by Osborn (1895.98, p. 72) 1 Douglass (1909.1) and Eiggs (1912.1). The strati- 



N. 



Mesatirhirvus superior, type 
Metarhinus ripartus , •■ 




r Sthenodectes 



Anvynodon.CF>inte^-medzzts A 



Figure 64. — Section of the Uinta formation (No. 10, fig. 35) from Kennedy's Basin to White River 

Canyon, Utah 

This section includes Uinta A, the barren sandstones; Uinta B 1, the Metarhinus zone capped by prominent bluffs of "Metarhinus sand- 
stone"; above this Uinta B 2, the Eobasileus-Dolichorhinus zone, capped by the "Amynodon sandstone." After E. S. Eiggs (1912.1); see 
also F- B. Weeks (1907.1). 



with the typical "upper Washakie," now known as 
Washakie B. The determination of the stratigraphy 
as well as the faunistic succession has been modified 



graphic order of the later Eocene deposits of the Uinta 
Basin and the correlated fauna may be presented as 
follows : 



Later Eocene deposits and fauna in the Uinta Basin, Utah 



Formation and nature of deposits 



Geographic conditions and mammalian fauna 



Uinta of King, Marsh, and White: Diplacodon elatus beds of 
Marsh; horizon C of Peterson, Douglass, and Riggs. Dacite 
tuffs and sandstones, grayish and greenish. Ferruginous. 
Thickness, about 600 feet. 

Uinta B 2 of Peterson and Osborn: Doliehorhinus cornutus 
zone of Osborn (1895.98). Amynodon beds of Riggs (1912. 1, 
p. 22). Coarse brownish dacite tuffs and sandstones, 
capped at the summit by the "Amynodon sandstone," 
immediately underlying Uinta C. Thickness, 285 feet. 



Uinta A of Peterson and Osborn, in part [Telmatotherium 
megarhinum beds of Osborn = Metarhinus fluviatilis zone, 
Osborn, upper Metarhinus zone of Riggs]: Capped by the 
"Metarhinus sandstones" of Riggs, with underlying coarse- 
grained brownish dacite tuffs and sandstone ledges; channel 
beds, varying in thickness from 5 to 30 feet, containing 
abundant remains of Metarhinus. Thickness, 266 feet. 

Uinta A of Peterson and Osborn, lower levels (lower Metarhi- 
nus zone of Riggs) : Capping of columnar sandstones, under- 
lain by friable sandy shales, interspersed with ledges. 
Thickness, 585 feet (Douglass, 1913). Unfossiliferous. 
Underlain by Green Ri\'or formation. 



Meadow, forest, and river fauna. Large titanotheres: Diplacodon 
elatus, Protitanotherium emarginatum, etc. Artiodactyla: 
Protoreodon, Leptotragulus, primitive camels. Small equines 
(Epihippus uintensis). Other large and small members of the 
true Uinta fauna. No traces of Amblypoda. 

Fauna chiefly stream border and fluviatile and some small 
forms. Last uintathere (Eobasileus). Aquatic rhinoceros 
(Amynodon intermedius) abundant. Entelodonts (Protelothe- 
rium uintense). Rare cameloids (Protylopus) . Rare eden- 
tates (Stylinodon). Numerous long-headed titanotheres (Doh- 
chorhinus cornutus, D. fluminalis, Sthenodectes). In the 
upper levels, first long-horned titanothere (Eotitanotherium of 
Peterson) ; ancestral Symborodon-like titanotheres (Rhadinorhi- 
nus). Titanotheres e.xtinct at this level or not recorded from 
it are Mesatirhinus and Metarhinus. 

Abundant fluviatile and forest fauna, of small variety. Tita- 
notheres: SmaU lowland varieties of Metarhinus very abund- 
ant, including several distinct specific forms; also the long- 
headed Doliehorhinus superior, the short-headed Sphenocoelus, 
Metarhinus earlei, M. riparius, M. fluviatihs, Rhadinorhinus, 
Doliehorhinus longiceps, an ancestral form of Dohchorhinus 
cornutus. The amblypods Eobasileus or Uintatherium. The 
large creodont Mesonyx obtusidens. 

No fossil mammals certainly recorded by Peterson, Douglass, or 
Riggs from this level. 



ENVIEONMENT OF THE TITANOTHEEES 



93 



U 



DQ 



CD 

< 
h 
Z 



100 



300' 
400' 



Dolichorhinus 
fluminalis 




Dolichorhinus y- 
cornufus, type ' 

Sfhienodectes 
incisivus, iype 

Dolichorhinus 
heterodon 

Dolichorhinus 
hyognafhus 
(cornulus) 

Sfhenodectes 
incisivus 

R had in orhinus 
diploconus 

Dolichorhinus 
longiceps, type 



Dolichorhinus I =-,^^-==2:=". 



T Diplcuxtdort 



Amynodon skel-.ATn.Mus. N9J933 



26"ATrvynocLoTh sandstoTve^j^ArrvyTzodon, irvtermediics 
ProtelotheriuTTh idntense 



; Eobasileus- 



400 



super/or, type 
Melarhinus 

n'parius, lype 

Melarhinus 

earlei 

Telmalolherium 

D olich orhinus 
longiceps 

Rhadinorhinus 
abbotti 

Melarhinus 
fluvialilis, lype 

Sphenocoelus 
uinlensis, lype 

Melarhinus 
cristalus, lype 

(?Dolichorhinus 
longiceps) 

Melarhinus 
ripar/us 



1: 



I EobasileiLS 
< StyLLnodorv 
I ProtylopzLS 




sWidstone^^ Harpccg'olestes 
Eobasileits iiznte.nsis, type 
FieldMus. 12170 



iEobcLsiLeiis 

Triplopics 

MesoTvyx obtusidens 

Crocodilus 
? TriplopiLS 



i< 



500 



NO MAMMALS RECORDED 



Figure 65. — Section of the Eobasileus-Dolichorhinus and Melarhinus zones in tlie Uinta Basin, Utah, show- 
ing stratigraphic distribution of species of titanotheres 

The species of titanotheres are shown in the left-hand column, the geologic strata in the middle column, other characteristic mammals in the 
right-hand column. After observations made by Peterson, Douglass, Eiggs, and Osborn. 



94 



TITANOTHEEES OF ANCIENT WYOMING, DAKOTA, AND NEBEASKA 



The researches of Peterson, Douglass, Riggs, and 
Osborn prove that Washakie B and Uinta B comprise 
two distinct faunal divisions — a lower, Uinta B 1 
{MetarTiinus fluviatilis, M. earlei zone), probably cor- 
responding with the lower levels (B 1) of Washakie B, 
and an upper, Uinta B 2 {Eohasileus-DolichorJiinus 
(cornutus) Tiyognathus zone), probably corresponding 
with the upper levels (B 2) of Washakie B. 

UINTA B 1 (METARHINUS ZONE = ZONE 13) 

Riparian fauna. — The fauna of the MetarTiinus zone 
was evidently that of a riparian lowland and was in 
part fluviatile or aquatic, as is indicated by its adapta- 
tions to aquatic and lowland life, which are inde- 
pendently developed in members of several different 
families. These adaptations are indicated by some of 
the specific names, such as MetarTiinus riparius, M. 
fluviatilis, two diminutive titanotheres, and DolicTio- 
rTiinus fluminalis. The animal last named is a short- 
limbed swamp-dwelling form, a fit companion of the 
river-seeking rhinoceros Amynodon intermedius , which 
begins to show aquatic adaptations in the structure of 
the orbit. The generic aspect of this fauna is almost 
identical with that of Washakie B, with the single 
exception that near the summit of Uinta B the ente- 
lodont ProtelotTierium replaces AcTiaenodon of Washa- 
kie B. The fauna contains a single new titanothere, 
RTiadinorTiinus , which is closely related to MetarTiinus. 
Some of the specific forms are identical with those of 
Washakie B and some exhibit more recent phases of 
evolution, which may be represented in the unfossilif- 
erous upper levels of Washakie B. We consequently 
reach the broad generalization that Washakie B 1 and 
Uinta B 1 were not only contemporaneous sediments 
but that they indicate the prevalence of similar 
physiographic and climatic conditions at this time on 
the north and the south sides of the Uinta Range. 

River-cTiannel fauna. — Remains of the small titano- 
there MetarTiinus have been found in ancient river 
channels, as determined by Riggs. This genus is by 
far the most distinctive fossil of this life zone and is 
apparently confined to it, although at certain levels 
primitive species of DolicTiorTiinus {D. longiceps) are 
found in equal abundance (Riggs, 1912.1, p. 20). 
This life zone, which is 400 feet thick, is composed 
chiefly of massive ledges of sandstone alternating 
with layers of sandy shales or indurated clays. In all 
the ledges there are traces of cross-bedding, and at many 
places there are beds of coarser river sand containing 
pebbles of quartzose material, sandstone, and clay 
shale. In these beds are found disarticulated bones 
of mammals, as well as the branches and at some 
places the trunks of trees, all pointing to the action of 
swiftly flowing streams that swept through a flood 
plain. Many skulls are found embedded in gravel, 
with their narial or orbital cavities filled with pebbles 
such as could be carried only by rapidly flowing water. 
Another evidence of stream action lies in the dissocia- 



tion of the parts of single skeletons. Whole skeletons 
have exceptionally been found but little disturbed, 
lying in a fine-grained homogeneous sandstone, ap- 
parently deposited in quieter water, such as deep 
pools or eddies. Remains of the long-headed titano- 
there DolicTiorTiinus are found only in the heavy sand- 
stones, so that this animal was apparently confined to 
the vicinity of streams. Supposed river-frequenting 
species of MetarTiinus, always found in sandstone, in- 
clude M. fluviatilis, M. riparius, and M. earlei. The 
species last named is found also in the lower levels of 
Washakie B, north of the Uinta Mountains. As we 
ascend in this MetarTiinus zone we find, according to 
Riggs (1912, p. 24), increasing numbers of upland 
forms. The "MetarTiinus sandstone" ledge that caps 
this zone has yielded the type specimens of Doli- 
cTiorTiinus superior, MetarTiinus riparius, and M. earlei, 
the last-recorded appearance of these animals in the 
Uinta Basin. 

UINTA B 2 (EOBASILEUS-DOLICHORHINUS ZONE = ZONE U) 

In the beds of the Eohasileus-DolicTiorTiinus zone the 
river sandstones and channel deposits gradually give 
place to shales and clays, indicating physiographic 
changes in this part of the Uinta Basin. In the lower 
100 feet of bluish or grayish shales, which are overlain 
by 40 feet or more of fine red clays, little evidence of 
mammal life is found, but certain thin beds contain frag- 
ments of Eohasileus. As we rise in the formation the 
gray "clays" begin to yield a mixed fauna of lowland 
and plains forms, including Protylopus and Stylinodon, 
together with remains of DolicTiorTiinus (cornutus) 
TiyognatTius and Amynodon intermedius. The massive 
"Amynodon sandstone," which forms the summit of 
this life zone, yields the type specimen of the long- 
headed titanotheres DolicTiorTiinus (cornutus) Tiyog- 
natTius, D. fluminalis; also of StJienodectes incisivus. 
This is the last appearance of the genus DolicTiorTiinus 
in the Uinta Basin. Doubtless the massive "Amyno- 
don sandstone" terminated the active period of 
fluviatile and flood-plain deposition in this locality. 
The D. (cornutus) TiyognatTius zone yields the large 
enteledont RrotelotTierium uintense, which is inter- 
mediate between AcTiaenodon insolens of Washakie B 
and ElotTierium of the White River group. This 
sandstone contains also the lophiodont DesmatotTie- 
rium guyoti, which is a forerunner of Colodon of the 
White River (Oligocene) group. 

ZONE 15: DIPLACODON-PEOTITANOTHERIUM-EPIHIPPnS ZONE 
[Uinta C I; Xudian of Europe] 

To zone 15 belong the Uinta of King and Marsh, the 
"Brown's Park beds" of Powell, and the Uinta(?) of 
the Beaver Divide, Wind River Basin. This zone 
(Uinta C 1) is correlated with the European stage 
that was named Ludian, after the "marnes de Ludes" 
in the Paris Basin, a stage typified by the "gypse de 
Montmartre," made famous by the classic researches 



ENVIRONMENT OF THE TITANOTHEEES 



95 



of Cuvier. The lower Ludian yielded the type speci- 
men of the equine LophiotJierium, a horse in the same 
stage of evolution as the diminutive American Epihip- 
pus of the Uinta. The American beds contain a rich 
titanothere fauna. They include the "Diplacodon 
beds" of Marsh (1877.1, p. 354) and contain the 
robust titanothere Protitanotherium, which is inter- 
mediate between the "prophet-horn" Manteoceras and 
the horned titanotheres of the lower Oligocene; also the 
type of Protitanotherium superbum, an animal greatly 
exceeding in size the earlier Oligocene titanotheres. 
Of great interest is the survival of the ancestral genus 
Manteoceras in the species M. uintensis, a genus first 
occurring in the upper Bridger, and the Bridger genus 
TelmatJierium in the species T. ultimum. 



plains fauna (Hypertragulidae and Camelidae) rep- 
resented respectively by genera believed to be ancestral 
to the tragulids (Leptotragulus) and to the camels 
(Protylopus, Camelomeryx) ; also members of the 
oreodonts (Protoreodon), and the agriochoerids {Agrio- 
choerus). The diminutive tylopod Protylopus has 
been selected as the possible ancestor of the great 
family of American camels. 

The fauna found near the base of the true Uinta 
thus includes a considerable light-limbed meadow 
and plains element, transitional to the plains fauna 
of the lowest Oligocene of the White River group. 
The occurrence of this fauna near the base of Uinta 
C indicates that the Uinta formation probably passes 
up into lower Oligocene time. The beds represent 




Figure 66. — Badlands near the mouth of White River, Uinta Basin, Utah (No. 10, fig. 35) 

Wortman and Peterson collecting. This view shows the typical Uinta formation (Uinta C 1) and the Diplacodon zone in the fore- 
ground, with Uinta C 2 (unfossiliferous) in the distance. After Osborn (1910.346). Am. Mus. negative 17663. Compare Plate 
XII, B. 



The amphibious rhinoceros Amynodon occurs in the 
species A. antiquus. It should be noted that the 
remains of all these large mammals were found not 
far above the base of Uinta C, and that all the speci- 
mens in the chief collections of small Artiodactyla 
(Protoreodon, Leptotragulus) and of Perissodactyla 
(Triplopus, LopModon, Isectolophus , a tapiroid, Epi- 
Jiippus) (Peterson) were obtained from the lower 
levels of Uinta C. With Epihippus were found the 
only primate that has been discovered in the Uinta 
Basin, NotJiarctus? uintensis, a lemuroid, and the 
supposed condylarth or insectivore Hyopsodus. The 
few surviving ancient creodonts are represented by 
Oxyaenodon and by the giant Harpagolestes uintensis. 
Especially important is our first knowledge of the 



a considerable change in local physiographic condi- 
tions from those of Uinta B. The fine-grained soft 
material, composed of altered eruptives, probably 
dacite tuffs, is of much the same texture as the char- 
acteristic " Titanotherium beds" (Chadron formation) 
of South Dakota, except as to its color, which is brick- 
red ; in fact, a reddish tinge prevails throughout the 
sediments of Uinta C 

During this latest part of the Eocene epoch the 
titanotheres of the Rocky Mountain basin south of the 
Uinta Mountains distinctly approach in character the 
titanotheres of the Great Plains. The appearance 
in this layer and near the summit of Uinta B of two 
or three entirely new forms of titanotheres (Eotitano- 
tJierium, Diplacodon, Protitanotherium) is less indica- 



96 



TITANOTHBEES OF ANCIENT AVYOMING, DAKOTA, AND NEBEASK.V 



tive of new migrations into the Rocky Mountain 
region than of new physiographic conditions favor- 
able to the fossilization of some of the upland and 
meadow Herbivora that had been evolving in the 
adjacent Plains region but had not mingled with the 



fluviatile, swamp, and forest-border fauna that 
inhabited the Uinta Basin in Uinta B time. 

The following summary of the later Eocene faunas 
of the Uinta Basin should be examined in connection 
with Figures 63-66. 



Composite section of mammalian faunas of tlie late?' Eocene sediments of the Uinta Basin 

[After Peterson, Osborn, Riggs, and Douglass] 



Uinta C (true Uinta formation = Diplacodon 
zone); 600 feet. Badlands like those of 
South Dakota, but of brick-red color. 
Brownish and reddish ferruginous sand- 
stones and clays (Peterson). 



Uinta B 2 (Eobasileus-Dolichorhiniis zone) ; 
300-400 feet. Section along gilsonite vein 
No. 2 (Riggs). Includes "Amynodon sand- 
stone," gray and greenish clays, ferruginous 
sandstones, bluish and greenish shales. 
Two red layers with fossiliferous sandstone 
between (Douglass). Supposed base of 
horizon B 2. 



Uinta B 1 (Metarhinus zone = upper Meta- 
rhinus zone of Riggs) ; 400 feet. Section on 
divide between White River Canyon and 
Coyote Basin (Riggs) . Also section 3 miles 
below mouth of Evacuation Creek (Riggs) : 
"Metarhinus sandstone." 
"Eobasileus sandstone" = massive ledges 
of reddish sandstone, alternating with 
layers of sandy shales. 
Indurated clays. 



Uinta A (lower A of Peterson, lower Meta- 
rhinus zone of Riggs) ; 500 feet (Riggs) ; 585 
feet (Douglass). Section in north wall of 
White River Canyon (Riggs) : 

"Columnar sandstones, about 300 feet 
thick, weathering as bold cliffs, or but- 
tresses along the canyon of White 
River. Color slightly more grayish 
than the underlying shales, but brown 
predominates (Riggs). 
"Two hundred feet friable sandy shales, 
weathering in steep slopes, with hori- 
zontal outcroppings of nodular or 
sandy layers, or by massive ledges of 
limited extent" (Riggs). 

Green River (?) formation. Shaly gray sand- 
stone of lacustrine origin. 



Titanotheres 



Diplacodon elatus, type. 
Protitanotherium emarginatum. 
Protitanotherium superbum, type. 
Telmatherium ultimum, type. 
Manteoceras uintensis, type. 



Eotitanotherium osljorni, type. 
Dolichorhinus cornutus ( = hyogna- 

thus), type. 
Dolichorhinus fluminalis, type. 
Dolichorhinus heterodon, type. 
Sthenodectes incisivus, type. 
?Rhadinorhinus diploconus, type. 
Dolichorhinus longiceps, type (near 

base) . 



Dolichorhinus superior, type. 
Metarhinus riparius, type. 
Metarhinus earlei. 
"Telmatherium," large jaw. 
Dolichorhinus longiceps. 
Rhadinorhinus abbotti, tj-pe. 
Metarhinus fluviatilis, tyjje. 
Dolichorhinus longiceps, skeleton. 
Metarhinus riparius (from base). 
Sphenocoelus. 

Heterotitanops parvus. (=?Meta- 
rhinus), from base of B 1. 

None. 



Other forms of life 



First oreodonts. 

Cameloids. 

Dichobunids. 

Aquatic rhinoceros (Amynodon). 

Small perissodactyls (tapiroids, lophio- 

donts, Epihippus). 
Last creodonts, Meson\-chidae and 

Oxyaenidae. 
No amblypods found. 

Last ambly pod (Eobasileus). 

First cameloid (Protylopus). 

Last taeniodonts (Stylinodon). 

Aquatic rhinoceros (Amynodon). 

Primitive entelodont (Protelotherium 
uintense) . 

Remains of plants and fishes; oc- 
casionally plentiful in sandstone 
ledges (Peterson) . 

Last primates ("Notharctus"). 

Giant creodonts, Mesonychidae (Har- 
pagolestes) . 

Giant creodont (Harpagolestes). 

Crocodiles. 

Turtles. 

Giant amblypod (Eobasileus). 

Light-limbed perissodactyl (Triplopus). 

Creodonts, Mesonychidae (Mesonyx). 



None. 



No mammals (Peterson, Riggs). 

Fragments of turtles. 

Unios. 

Remains of plants and occasionally 

large tree trunks in sandstone ledges 

(Peterson). 



Remains of plants, fishes, and insects 
in the shales (Peterson). 



ENVIEONMENT OF THE TITANOTHERES 



97 



SUMMARY OF FAUNAS OF UINTA B AND C 

Though the whole later Eocene section of the 
Uinta is 1,900 feet thick it includes 500 feet of un- 
fossiliferous beds both at its base and at its summit, 
so that the fossiliferous beds cover only about 900 feet. 
The stages of evolution are best measured in the suc- 
cessive species of DolichorJiinus, which are found both 
at low and at high levels in the fossiliferous part of 
the section. 

The archaic mammals that play so large a part 
through lower and middle Eocene time diminish in 
number and approach extinction at the end of Eocene 
time. The numerical inferiority of the waning archaic 
mammals and the rapid increase in the numbers of 
modernized mammals are indicated in the following 
table, prepared in 1910: 

Transition in mammalian life at end of Eocene time 





Genera 


Species 


Archaic mammals: 


.2 
1 
5 


6 


Condylarthra ( H vopsodontidae) 


2 




5 








8 


13 


Modernized mammals: 

Primates _ 


2 
1 
3 
3 
9 


?3 


Rodentia __ 


3 


Carnivora (Miacidae) 


4 

4 


Perissodactyla 


16 




18 


30 



The Amblypoda culminate in the gigantic Eobasileus, 
which disappears at the end of Uinta B, when the 
gigantic creodont Mesonychidae and the catlike 
Oxyaenidae appear for the last time. It is note- 
worthy that these animals attain their largest size in 
this, their waning period. The lemuroid primates 
are found in greatly diminished numbers as compared 
with those in the Bridger, possibly because the con- 
ditions were unfavorable to the fossilization of re- 
mains of arboreal animals; in fact, we know nothing 
of the forest or the arboreal fauna during the entire 
period of Washakie B and Uinta B because of pre- 
vailing fluviatile conditions of deposition. 

ADAPXrVE RADIATION OF THE TITANOTHERES IN THE UINTA BASIN 
GENERA AND SPECIES HEPRESENTED 

Through these 650 feet of fossiliferous sediments 
the titanothere fauna of the Uinta Basin is revealed 
as extraordinarily large and varied, no less than 11 
genera and 22 species having been described. The 
animals range in size from the small Metarhinus flu- 
viatilis, some of which were not so large as a tapir, 
to the huge ProtitanotJierium superbum. 



The titanothere Metarhinus is abundant and char- 
acteristic in Uinta B 1, ranging from the base to the 
summit but not extending into Uinta B 2 as here de- 
fined. (In previous reports horizon B 2 has been 
included in Uinta A.) According to Riggs (1912.1, p. 
27) the genus includes two phyla — the first comprising 
the small MetarJiinus fluviatilis Osborn and M. riparius 
Riggs, with long, narrow skull; the second including 
the broad-skulled forms M. earlei Osborn (which is 
also found in Washakie B) and M. cristatus Riggs. 
Metarhinus was a companion of its long-skulled rela- 
tive Dolichorhinus in and near the rapidly flowing 
streams, its remains being usually found in coarse 
and semigravelly sandstones. (Riggs, op. cit., p. 24.) 
In Uinta B 2 rapid streams, apparently the favorite 
haunt of Metarhinus, were less abundant than in 
Uinta B 1 (Riggs, op. cit., p. 25), which partly ac- 
counts for the apparently sudden disappearance of 
these animals from the sediments. 

Sphenocoelus uintensis, which is also probably from 
the Metarhinus zone (Uinta B 1), is known only from 
the hinder half of a skull. This strange animal is 
clearly a member of the Metarhinus-Dolichorhinus 
group and may be closely related to the long-skulled 
Metarhinus riparius. The Metarhinus series as a 
whole is clearly related to the older and more primitive 
Mesatirhinus megarhinus of Washakie A and Bridger 
C and D, which is also structurally ancestral to 
Dolichorhinus . 

The name Heterotitanops parvus Peterson has been 
applied to the skeleton of a very young animal from 
Uinta B 1. It was found, articulated, in a hard sand- 
stone concretion and lower down in Uinta B 1 than 
any mammalian remains heretofore described from 
that horizon. (Peterson, 1914.2.) In the opinion of 
Gregory the characters of the deciduous dentition 
and of the facial region of the skull of this animal 
indicate that it probably represents the newly born 
young of some of the Metarhinus-Rhadinorhinus group. 

Rhadinorhinus is distinguished from Metarhinus by 
its tapering nasals and by the reduced infraorbital 
process of the malar bones. One species, R. abhotti 
Riggs, is found in Uinta B 1 , and another, R. diploconus 
Osborn, is recorded from Uinta B 2. Riggs suggests 
that Rhadinorhinus was an upland rather than semi- 
aquatic form. Gregory noted in 1902 that it fore- 
shadows the long-horned titanothere Megacerops 
(Symiorodon) of the lower Oligocene in the abbrevia- 
tion of the face and in the characters of the dentition. 

The long-skulled Dolichorhinus is represented by 
two species in Uinta B 1 (one of which, D. longiceps 
Douglass, extends into the base of Uinta B 2) and by 
four species in Uinta B 2. The most primitive species, 
D. superior, is in general intermediate in structure 
between the ancestral Mesatirhinus and the later 
species of Dolichorhinus. The most advanced species, 
D.fluminalis Riggs, is from the upper levels of Uinta 



98 



TITANOTHEEES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



B 2. The allied D. cornutus is believed to be specifi- 
cally identical with D. hyognathus of Washakie B. 
In Uinta B 1 remains of Dolicliorliinus, as noted by 
Riggs, are frequently found associated with those of 
Metarhinus in coarse pebbly sandstone laid down in 
rapid streams; but in Uinta B 2 they are frequently 
found in lenticular sandstones, which were apparently 
deposited in quiet water, for they show little evidence 
of water currents, and which yield skulls associated 
with mandibles and parts of skeletons. Riggs accord- 
ingly infers that Dolichorliinus inhabited the low 
grounds near quiet waters rather than the swift 
currents preferred by Metarhinus. 

Contrasting with the elongate and straight-sided 
skull of Bolichorhinus is the broad, robust skull of 
StJienodedes incisivus (Douglass), a titano there with 
exceptionally massive incisor and canine teeth and 
broad, low-crowned upper molars. The type skull 
was found by Douglass in a thick deposit of sandstone 
and small gravel, evidently of stream origin, near the 
middle of Uinta B 2, whereas Riggs's specimen was 
found in lenticular sandstones at about the same level. 
Sthenodectes shares many characters in common with 
the Bridger genera Manteoceras and Telmaiherium and 
appears to be an advanced member of that macrodont 
group. 

With the possible exception of Rhadinorhinus all 
the titanotheres so far noted from Uinta B 1 and B 2 
belonged to aberrantly specialized side lines, which 
are not found in later formations and apparently 
became extinct. 

One titanothere recorded from near the summit of 
Uinta B 2, namely, Eotitanoiherium osborni Peterson 
(1914.1 ; 1914.4), is highly progressive toward the giant 
plains-living types of the uppermost Eocene (Uinta C) 
and of the lower Oligocene. This animal, represented 
by parts of two skeletons in the Carnegie Museum, 
surpasses even ProtitanotTierium of Uinta C in the 
development of a pair of large oval bony protuber- 
ances above the eyes. One of the most remarkable 
features of Eotitanotherium is the progressive sub- 
molariform character of its third and fourth upper 
molars, which are, indeed, slightly more advanced in 
type than those of certain lower Oligocene titano- 
theres. The animal was about as large as a rhinoceros, 
and throughout the skeleton are mingled the earlier 
characters of its Eocene predecessors with progressive, 
plains-living adaptations prophetic of some of the 
Oligocene titanotheres, especially those of the meno- 
dontine group. 

The titanothere fauna of the true Uinta (Uinta C), 
though less extensive in genera and species than that 
of Uinta B 1 and B 2, is none the less of prime impor- 
tance to the historian of the family. Thus the titano- 
theres of the true Uinta include, first, certain conserva- 
tive phyla {Manteoceras, Telmatherium) , which repre- 
sent the little-altered descendants of genera either of 



Uinta B 1 and B 2 or of Washakie and Bridger types; 
second, two very progressive and different phyla, 
Diplacodon and ProtitanotTierium, of uncertain relation- 
ships, which appear to be immigrants from other 
localities. 

ADAPTIVE RADIATION OF PHYLA 

There seem to be at least four contemporary phyla, 
representing wide local adaptive radiation : 

First, the robust, short-limbed forms, one of which, 
Manteoceras uintensis Douglass, found in gray sand- 
stone in the red beds of the lower portion of Uinta C, 
is considerably larger than the more primitive species 
of Manteoceras in the upper Bridger and Washakie A 
but is little modified otherwise. Its horn swellings, 
if developed at all, were not large, and it is strongly 
macrodont in type, like Telmatherium and Sthenodectes. 

Second, the long-limbed, long-headed, relatively 
hornless Telmatherium, which is distinguished espe- 
cially by its deep malar bones and the high sagittal 
crest and is represented in Uinta C by the great 
Telmatherium ultimum and the gigantic T. altidens. 
T. ultimum is practically hornless, having only the 
slightest rugosity at the naso-frontal junction in the 
type skull. Accordingly the species Manteoceras 
uintensis and Telmatherium ultimum and the genus 
Sthenodectes, while advancing in the direction of the 
Oligocene type in various characters, are apparently 
excluded from direct ancestry to the later types by 
certain specializations, such as marked enlargement 
of the incisors and canines, and by the lack of develop- 
ment of effective horn swellings. 

Third, Diplacodon elatus Marsh of Uinta C, a 
progressive titanothere, which is known chiefly from 
the upper dentition and takes its generic name from 
its submolariform third and fourth premolars. The 
precise relations of this animal are still in doubt. 
The premolars and molars may have been derived 
from the type represented by Rhadinorhinus diploconus 
of Uinta B 2, which is the only one of the older titano- 
theres that has the dentition and skull at all like 
those of Diplacodon. In other respects the Diplacodon 
dentition suggests that of the lower Oligocene titano- 
there Menodus trigonoceras , and in still another respect 
it resembles that of Eotitanotherium of Uinta B 2, though 
it differs from that genus in the more molariform con- 
dition of the third upper molar. 

Fourth, Protitanotherium emarginatum Hatcher, 
which is known from the facial part of the skull and 
the lower jaw of the type specimen. It is a large 
animal, which, so far as Imown, approaches the 
Oligocene type of Brontops. It has oval horn swellings 
which are less protruding than those of the type of 
Eotitanotherium; its nasals are wide distally, in con- 
trast to the tapering nasals of Eotitanotherium; its 
canines are very stout and acutely conical; its stout 
upper incisors form a flattened arch. Altogether it 
seems to represent a phylum distinct from Eotitano- 



ENVIRONMENT OF THE TITANOTHEEES 



99 



therium and of doubtful relationship both to earlier 
or to later titanotheres, although it was possibly 
derived from Manteoceras. Of the same phylum is 
Protitanotherium swperhum, a gigantic animal, with a 
jaw 24 inches long and premolars and molars of very 
progressive type. It is much larger than the smaller 
titanotheres of the lower Oligocene. 

Another titanothere of uncertain relationship is 
BrachydiasUmatherium from Transylvania, eastern 
Hungary (now Rumania). The geologic age of this 
animal is not certain, but it is in an upper Eocene 
stage of evolution as compared with the titanotheres 
of America. 



FAUNA UNREPRESENTED 



The sequence of titanothere species in the Uinta 
Basin illustrates the vagaries of the fossil records of 
the Rocky Mountain basin region caused by local 
physiographic changes; each kind of sedimentation 
exhibits only a part of the fauna. For the entire 
period covered by the lower sediments of the Uinta 



Basin little or no knowledge of the small terrestrial 
fauna has come to light, none of the arboreal fauna, 
and none of the plains and upland fauna, in contrast 
with the surprisingly extensive knowledge of the 
fluviatile and the swamp-dwelling fauna. Gradually 
conditions changed, and Uinta B 2, as we ascend, 
affords an increasing knowledge of the cursorial 
meadow fauna; but it is not until Uinta C (true Uinta) 
that local conditions became favorable to the pres- 
ervation and fossilization of the small cursorial mam- 
mals of the artiodactyl and perissodactyl divisions 
of the ungulates. The sudden appearance of these 
animals might be attributed to immigration, but it 
is equally probable that they were all evolving in the 
same region or in the adjacent Plains region. Thus 
the data do not necessarily suggest immigration or 
migration; these animals may have been brought into 
the field of observation by changing conditions of 
fossilization. The manner in which these numerous 
phyla of titanotheres enter this field is shown in the 
following table: 



Geologic and geographic range of phyla {here "subfamilies" and certain genera) of titanotheres 



[Showing their successive immigration from the north and their evolution in their i 
column; the later immigrants are named in order from bottom to top. 



ew habitat. The earliest immigrants are those named at the bottom ot the first 
The difference in the length of the blaclc bars has no significance] 



Phylum 


Wind River B « 
(" Lost Cabin ") 
(middle Eocene) 


Bridger (upper middle Eo- 
cene) 


Washakie 
(upper 
Eocene) 


Uinta (upper- 
most Eocene) 


Chadron (Oligo- 
cene) 




A' 


B 


C = 


D » 


E 


A ' 


B 


A 


B" 


C 


A 


B 


C 














































































1 
























1 











































































^^^ 


___ 


1 
























— 






















































— 





■^ 























---- 






















— 




^ 








































! 









































































































» Wind River B = Huerfano A. 
I Bridger A = Huerfano B. 



■ Bridger and D = Washakie A. 
' Washakie A = Bridger C and D. 



• Uinta B = Washakie B. 

'Diplacodon, Eotitanotherium, Protitanotherium. 



ZONE 16: THEORETIC UINTA C 2 

Titanotheres have thus far been determined from 
only the lower 100 feet of Uinta C. They are at 
present only partly known. When fully known we 
shall probably find close generic if not specific corre- 
lation between the upper fauna (now unknown) of 
Uinta C and the fauna of the lower levels (Chadron 
A) of the White River group. The passage from 
Eocene to Oligocene time probably occurs within the 
period of Uinta C (true Uinta) deposition. Scott is 
disposed to put all of Uinta C in the Oligocene. 



COMPOSITE EOCENE AND LOWER OIIGOCENE SECTION AT BEAVER 
DIVIDE, WIND RIVER BASIN, WYO. 

Most of the Oligocene sediments in the Rocky 
Mountain basin region have been eroded away. The 
only locality where fossil-bearing lower Oligocene sedi- 
ments still overlie those of the upper Eocene is on the 
southern border of the Wind River Basin, Wyo., 
where the true Titanotherium zone overlies sediments 
containing a fauna similar to that of the Diplacodon 
zone (Uinta C). The geologic section observed at this 
point by the American Museum expedition of 1909 



100 



TITANOTHEHBS OF ANCIENT WYOMING, DAKOTA, AND NEBEASKA 



under Granger and N. H. Brown, who discovered this 
fauna in 1908, is as follows: 

Oreodon zone = Brule for 



Summit of lower Oligocene 

mation 

Base of lower Oligocene, Titanotherium zone=Chadron 
formation 

Upper Eocene, Diplacodon zone = Uinta(?) formation — 

Middle Eocene (?), unfossiliferous = Bridger (?) forma- 
tion 

Lower Eocene, Lamhdotherium zone = upper part of 
Wind River formation 



Feet 
540 




O/'eodoTL zone 



A single tooth of either Diplacodon or Protitano- 
therium has been found at Beaver Divide, Wyo. 
The correlation with Uinta C rests upon Camelodon 
arapahovius Granger, a species somewhat more pro- 
gressive than Protylopus of Uinta C and somewhat 
more simple than Leptotragulus , characteristics that 
combine to place it among ancestral camels, in the 
Camelidae. In certain characters it agrees with Lep- 
totragulus profedus of the Titanotlierium zone of Pipe- 
stone Springs, Mont. The Amy- 
nodon found here agrees with the 
species A. antiquus, originally 
determined in Washakie B 
( = Uinta B). Two specimens 
of Protoreodon are referable to 
P. parvus, from the base of Uin- 
ta C or the summit of Uinta B. 
Above this Diplacodon (?) 
level is a very marked erosional 
unconformity between the up- 
per Eocene and the lower Oli- 
gocene; broad, shallow valleys 
(Sinclair and Granger, 1911.1, 
p. 99), indicating fairly mature 
topography, were excavated in 
the sediments of the Diplaco- 
don(1) zone. After these val- 
leys were cut the first deposits 
laid down were fine-grained 
buff-colored tuffaceous shales. 
In this tuff the American Mu- 
seum exploring party of 1909 
found a skull of Menodus heJoce- 
ras, which belongs to the lower 
level of the Titanotherium zone, 
corresponding with Chadron A. 
The volcanic ash comprising 
the sediments of the Oreodon 
titanothere zone, a few feet 
thick, is covered with a mud 
flow of volcanic material 46 feet 
thick, above which lies 540 feet 
of fine, wind-blown buff ash and 
dust. No clays have been found 
at this middle Oligocene horizon , 
which corresponds in age with 
the Brule formation of the 
White River group — only wind- 
laid ash and coarse gravel, 
perhaps deposited by torrents 
during occasional heavy rains. 
None of these sediments ap- 
pear to have been much dis- 
turbed by water, and Sinclair 
Diagrammatic section of deposits at Green Cove, Beaver Divide, Wyo. ^^^ Grander (1911.1 p. 114) 



Oreodon 
Cyllndrodon 
Caenopus 
Ischyromys 
Poebro therlum 



Menodus he/oceras 

? DiplcLcodoTz zone 

Amynodon ? anfiquus 
Protoreodon 
Camelodon 
Pro titan o ttierium 



Lamhdotherium zone 



Lambdotherium 
Coryphodon, PlienacoduSj 
hieptodon , Eohippus 



Figure 67 ^^^ ^^„„ „ ^„ , 

(No. 6, fig. 35), from the Lambdotherntm zone (Wind River) to the Oreodon zone """ .""'"""" t"*h lie e that 

they accumulated under a drier 



(White River) at the summit 



Chiefly after Granger (1910.1). 



ENVIEONMENT OF THE TITANOTHEEES 



101 



climate than that which prevailed in Eocene time. 
These upper sediments contain a true Oreodon zone 
fauna. 

FOURTH FAUNAI PHASE (LOWER OLIGOCENE) 

LOWER OLIGOCENE MAMMALS 

COERELATION OF EUROPEAN AND AMEEICAN FORMS 

The lower Oligocene mammals represented by the 
fossils thus far discovered are listed below. 
Peculiar to Europe: 

Paleotheres. 

Anoplotheres. 

Oenotheras. 

Gelooids. 

Amphicyonids. 

Viverrids. 

Cricetines (hamsters). 

Theridomyids. 

Sirenians (Hahtherium). 

(Horses not recorded.) 
Common to Europe and North America: 

Titanotheres (central Europe). 

Chalicotheres. 

Rhinoceroses (aceratheres and diceratheres) . 

Amynodonts. 

Anthracotheres. 

Suillines. 

Entelodonts. 

Opossums. 

Hyaenodonts. 

Canids (dogs). 

Mustelids (martens). 

Machaerodonts (saber-tooth cats). 
Peculiar to North America: 

Horses. 

Hyracodonts (rhinoceroses) . 

Oreodonts. 

Camelids. 

Hypertragulids. 

Leptiotids. 

Chrysochlorids? (inseotivores) . 

Ischyromyids (rodents). 

Leporids (hares). 

ZONE 17: TITANOTHERIUM-MESOHIPPUS ZONE 
[Chadron A, B, and C; Sannoisian of Europe] 

The forms that constituted this rich world of lower 
Oligocene mammalian life were distributed through 
the Rocky Mountain basin region, but the sediments 
that contained the fossils have been eroded away 
except in a few isolated areas, such as those along 
Pipestone Creek, Mont.; at Beaver Divide, Wyo., 
south of the Wind River Basin; and at Bates Hole, 
Wyo. The areas in which these sediments were 
deposited lie east of the Rocky Mountains, in Sas- 
katchewan, North Dakota, South Dakota, and Colo- 
rado. The chief fossil-bearing sediments exposed are 
in the localities shown below. 

Recorded thickness of the Titanotherium zone in thirteen exposures 
of lower Oligocene deposits 

Feet 

1. Cypress Hills, Saskatchewan (Lambe, 1908) 50-500 

2. Pipestone Creek, Jefferson County, Mont. 

(Douglass, 1903) 300 + 

3. White Butte, N. Dak. (Douglass, 1903) 120 



4. Big Badlands, S. Dak. (Hatcher, Darton) (typical 

area of Titanotherium zone) 180 

5. Goshen Hole (Scotts Bluff), southeastern Wyoming 

(Darton), maximum thickness 200 

6. Hat Creek, South Fork, Cheyenne River, Dawes 

County, Nebr 100 ± 




102 



TITANOTHEEES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 




ENVIRONMENT OF THE TITANOTHERES 



103 



7. Near Dickinson, S. Dak. (Douglass) 40-50 

8. Pine Ridge, S. Dak. (Darton) 30-60 



9. Beaver Divide, Wyo. (Granger) 

10. Bates Hole, Natrona County, Wyo 

11. Adelia, Sioux County, Nebr. (Darton), about 

Between Platte River and Arkansas River drainage 

(Darton). ("Monument Creek group" 
of Hayden; Castle Rock conglomerate 

of Richardson, 1912.1) 300 

Horsetail Creek, northeastern Colorado 
(Matthew,1901.1), not over 



46 

(?) 



12. 



13 



between the upper fauna (now unknown) of Uinta C 
and the fauna of the lower levels (Chadron A) of the 
White River group. The passage from Eocene to 
Oligocene time probably occurs within the period of 
deposition of Uinta C. Scott is disposed to put all of 
Uinta C in the Oligocene. 

General Section of the Tertiary rocks of Nebraska. 



100 

The deposits at these localities, some of 
them indicated on the accompanying map, 
represent only the exposed parts of the lower 
Oligocene deposits of the great flood-plain sys- 
tem now known as the Chadron and corre- 
lated formations, the larger part of which is 
covered by the Brule and Arikaree formations. 
This flood plain extends 325 miles north and 
south and 300 miles east and west. We do 
not know whether it was wholly continuous. 
Such an area would embrace 97,500 square 
miles, which would not exceed the present 
Andean flood plains. 

At the base of these sediments in South Da- 
kota and northern Colorado there are abundant 
remains of titanotheres, certain of which are 
in stages of evolution no more advanced than 
those found at the base of Uinta C, Diplacodon 
zone. Consequently the faunistic relation be- 
tween the titanotheres living in the mountain 
basins and those living on the Plains remains 
to be solved by future discovery. This rela- 
tion may be revealed in the "missing" faunal 
zone. At present we may divide the life zones, 
in descending order, as follows: 

17. Titanotherium zone: 

Chadron C, levels 3, 2, 1: 
Brontops robustus. 
Menodus giganteus. 
Brontotherium platyceras. 
Chadron B: 

Brontops dispar. 
Menodus trigonoceras. 
Brontotherium hatched. 
Chadron A, levels 1, 2, 3: 

Brontops brachycephalus. 
Menodus heloceras. 
Brontotherium leidyi. 
16. Theoretic zone of Uinta C (upper levels, or Uinta 

C 2): Unknown or "missing." 
15. Diplacodon zone of Uinta C (lower levels, or Uinta C 1) : 
Protitanotherium emarginatum. 
P. superbum. 
Diplacodon elatus. 

It is very important to recall the fact that 
titanotheres have thus far been determined from only 
the lower 100 feet of Uinta C, that they are only 
partly known, and that when fully known we shall 
probably find a close generic if not specific correlation 



NameB. 


SUBDIVISIONS. 


Thick- 


LOCALITIES. 


Foreig:n 
Equiva- 
lentB. 


S 

3 


Fine loose sand, with some 
layers of limestope, — contains 
bones of Canis, Felis, Caxtor, 
Equus, Mastodon, Testudo, &c.., 
some of which are scarcely dis- 
tinguishable from living spe- 
cies. Also Helix, Physasucclnea, 
probably of recent species. All 
fresh water and land types. 


o 


On Loup fork of 
Platte River ; extend- 
ing north lo Niobrara 
River, and south to 
an unknown distance 
beyowd the Platte. 


a 


§ 

> 
s 
s 


White and light drab clays, 
with some beds sandstone, and 
local layers limestone. Fossils, 
Oreodon, Titanotherium, Cliaro- 
potamus, Rhinoceros, Anchithe- 
rium, Hycenonodon, Afachairodus, 
Trionyx, Testudo, Helix, Plan- 
orbis, Limncea, Petrified wood, 
&c. &c. All extinct. No 
brackish water or marine re- 
mains. 


o 

a 

o 

8 

O 


Bad Lands of White 
River ; under the 
Loup River beds, on 
Niobrara, and across 
the country to the 
Platte. 


a 

o 

o 


.£■2 

Pig 
Ti p. 


Light gray and ash colored 
sandstones, with more or less 
argillaceous layers. Fossils, — 
fragments of Trionyx, Testudo, 
with large Helix, Vivipara, 
Petrified wood, &c. No marine 
or brackish water types. 


o 
o 

§ . 

-21 

r-t 


Wind River valley. 
Also west of Wind 
River Mountains. 


»- 


5 

'S 

13 

"§ 

3 

D 


Beds of clay and sand, with 
round ferruginous concretions, 
and numerous beds, seams and 
local deposits of Lignite ; great 
numbers of dicotyledonous 
leaves, stems, &c. of the genera 
Platanus, Acer, Ulmus, Populus, 
&o., with very large leaves of 
true fan Palms. Also, Helix, 
Mclania, Vivipara, Corbicula, 

Unio, Ostrea, Potamomya, and 
scales Lepidotus, with bones of 

Trionyx, Emys, Compsemys, 

Crocodilus, &c. 


a 

u 
o 

o 
o 
o 


Occupies the whole 
country around Fort 
Union, — extending 
north into the Britisli 
possessions, to un- 
known distances ; 
also southward to 
Fort Clark. Seen un- 
der the White River 
Group on North Plat- 
te River above Fort 
Laramie. Also on 
west side Wind River 
Mountains. 


§ 



Figure 70. — Facsimile of the Meek and Hayden Tertiary section of 1862, 
showing original definitions of White River group and Wind River 
formation 

The deposits are now known to include the following: 

"Loup River beds" (lower Pleistocene fauna listed). The area includes deposits of the Plio- 
cene and Miocene (Ogalalla formation of Darton). 

"White River group," including lower Miocene (Arilcaree formation of Darton) and Oli- 
gocene (Brule and Chadron formations of Darton). The " Clioeropotamus" is Ancodus 
amcricoKKs.the ancodont of the Chadron formation (Titanotherium zone). 

"Wind River deposits" (summit of the lower Eocene). 

"Fort Union or Great Lignite group" (basal Eocene). 



OLIGOCENE FLOOD-PLAIN SEDIMENTATION IN THE 

WESTERN GREAT PLAINS REGION 

CONDITIONS OF DEPOSITION 

A very long period of extremely slow sedimentation, 
east of the Rocky Mountains of Wyoming and Colo- 
rado, began in lower Oligocene time and extended 
without interruption to lower Miocene time, laying 
down the great deposits originally described as the 
White River group by Meek and Hayden (1862.1, 
p. 433) in the following language: 



104 



TITANOTHEBES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



White River group * * * Wliite and light drab clays, 
with some beds sandstone, and local layers limestone. Fossils: 
Oreodon, Titanotherium, Choeropotamus, Rhinoceros, Anchithe- 
rium, Hyaenodon, Machairodus, Trionyx, Testudo, Helix, 
Planorbis, Limnaea, petrified wood, &c., &o. All extinct. No 
brackish- water or marine remains * * * I^OOO feet or 
more * * * Badlands of White River; under the Loup 
River beds, on Niobrara, and across the country to the Platte. 
* * * Miocene. 

This original definition appears to include all that 
has been determined subsequently and mapped by 
the United States Geological Survey (Darton, 1905) 
under three formations, namely, Chadron, Brule, and 
Arikaree, as shown in the accompanying illustration 



east. This fact is significant. It would appear, as 
stated at the beginning of this chapter, that sedimen- 
tation in this region was suspended after Denver, 
Lance, and Fort Union time. The Eocene gradients 
were so high that there were long periods of erosion, 
during which large areas of Upper Cretaceous beds 
were laid bare in the region that now includes North 
and South Dakota, western Nebraska, and Colorado, 
so that the lowest Oligocene sediments of the White 
River group, composing the Titanotherium zone 
(Chadron A), lie in gentle valleys of ancient formation 
that range in age from the Algonkian to the Denver 
formation and Dawson arkose. In Hayden's typical 




Figure 71. — Map showing tributaries of Cheyenne River, S. Dak., from the southeast and tlie type locality 
(X) of the " Titanotherium beds" of Hayden (Chadron formation), on Bear Creek; also principal collecting 
ground of Hatcher (dotted area) , the chief fossiliferous area in the Big Badlands 



(fig. 69). Meek and Hayden did not, however, specif- 
ically define the upper limit of theii' White River group, 
and all the fossils listed by them as characteristic 
of the White River group apparently came from beds 
now classified as Oligocene. The name White River 
group has therefore for years been restricted to the 
beds of Oligocene age (Brule and Chadron formations). 
This great flood-plain deposition was preceded by 
a long period of erosion in Eocene time. No sedi- 
ments of Wasatch, Bridger, or Uinta age have been 
found on the Plains east of the Front Range of the 
Rocky Mountains, except in a small area of Huerfano 
sediment which lies within a mountain basin farther 



locality of the White River group — the Mauvaises 
Terres of early explorers — the Big Badlands between 
the Cheyenne and the White River of South Dakota — 
the underlying beds are composed entirely of the 
Pierre (Upper Cretaceous). At some places (Loomis, 
1904.1, p. 432) the rivers depositing the Titanotherium- 
bearing beds washed out along theu" banks masses of 
the Pierre shale that contained characteristic Pierre 
fossils — Baculites and the bones of Cretaceous rep- 
tiles — and redeposited them in Oligocene sediments. 
On this level, the gently undulating surface of the 
Pierre, east of the Rocky Mountains and the Black 
Hills, meandered broad, sluggish streams, whose chan- 



ENVIEONMENT OF THE TITANOTHEEES 



105 



nels ranged in width from a few hundred feet to half 
a mile. Beside these stream channels there were 
lagoons and areas of back water, some of them spread- 
ing into shallow lakes but none into vast sheets of 
fresh water. Savannas were interspersed with grass- 
covered pampas traversed by wide, meandering rivers 
that frequently changed their course. In these chan- 
nels were deposited conglomerates and river sandstones, 
marked by cross-bedding, as well as calcareous grits 
In the shallow lagoons and back waters were deposited 
the fine clays and layers of fuller's Qarth. The de- 
posits of gypsum represent periods of evaporation. 
In the lower part of the Titanotherium zone the de- 



it spread over the great area on which it has left its 
traces by the deposition of its peculiar sediments. 
* * * The basin-like character of this formation 
is most admirably shown." In the same memoir, 
Leidy (1869.1, p. 25) expressed some doubt as to the 
lacustrine theory, observing: "It is a remarkable cir- 
cumstance that among the large quantity of fossil 
bones brought from the Mauvaises Terres and sub- 
mitted to the examination of the author, certainly 
amounting to several tons in weight, there was de- 
tected no trace of remains of birds or fishes; and the 
same may be said of reptiles, except one species of 
turtle." 





Juan Ot/l^^ Ul 'lo he.iLi' ol UjJ.yiU-J 




Figure 72. — Type locality of the " Tiianoiheriuvi beds ol llaj deu, oii Bear Creek, S. Dak. 

Panoramic view, connecting at X. Upper section, looking southeastward, up Bear Creek; lower section, looking northwestward, down Bear Creek. Am. Mas. 

negatives 104722-104726. 



posits consist chiefly of fine flood-plain or overflow 
sediments interspersed with river sandstones and 
conglomerates, perhaps locally lacustrine, and occa- 
sional layers of volcanic ash. 

This theory that the deposits of the western Great 
Plains region are of flood-plain and fluviatile origin 
has gradually replaced the older lacustrine theory that 
they were laid down in great fresh-water lakes. The 
lacustrine theory originated with Hayden, who, in his 
geologic introduction to Leidy's memoir of 1869 
(1869.1, p. 18), observes: "One of the most interesting 
features in regard to this great fresh-water lake is the 
evidence of its growth from a germ, as it were, until 
101959^29— VOL 1 9 



The lake-basin theory was generally adopted by 
geologists and paleontologists, reaching its apex in 
King's development of the lake theory both for the 
Plains and the mountain region. Johnson (1901.1), 
Gilbert (1896.1), Haworth (1897.1), and especially 
Davis (1900.1) reviewed the whole subject broadly in 
a critical way, developing the theory of fluviatile and 
flood-plain origin. Fraas (1901.1), Hatcher (1902.3), 
and more recently Darton (1905.2) set forth strong 
evidence for the theory of deposition in river channels, 
flood plains, back waters, lagoons, and shallow lakes. 
Among paleontologists Matthew (1899.2; 1901.1) 
was the first to attack the lacustrine theory of the 



106 



TITANOTHEEES OP ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



origin of the Brule clay of the White River group and 
to advance reasons for believing that the sandstones 
were formed by river and flood-plain sedimentation 
and the clays in part by back water and lagoon and 
chiefly by eolian sedimentation. His paleontologic 




Porcupine Butte 
CANIC ASH LAYER 



t-:<Wjg ,, 



"■""'saite-,,,. 




jifery'cocha 



Steneofiber 
Prom erycoch ce 



Figure 73. — Panoramic section of the Big Badlands of South Dakota 



Modified from United States Geological Survey Bulletin 361, PI. in. (Osborn and Matthew, 1909.321.) View 
southeastward from Cheyenne River, along line indicated on Figure 09, toward Porcupine Butte, across 
the Chadron, Brule, and Arikaree formations. This section illustrates the intrusion of river-channel 
deposits (the " T'/VaTioMfriiim sandstones," "JV/c/amyTiodon sandstones," and "Pro^ncfras sandstones") and plete skeletons at present knOWn maV 
river-channel conglomerates in "clays" of the ritanof/iOTitm and Orfodoji zones. It shows also the charae- , j-j+ufi f 1^" 

teristic erosion forms of these different layers. (See map forming fig. 69, vicinity of section B.) be COUntcd On the nngerS 01 One nana. 



The Testudinata as analyzed by Hay (1908.1) 
furnish evidence that during lower Oligocene time the 
Great Plains region was prevailingly dry land. In 
the sediments of the White River group there occur 
eight species of the Testudinidae, including one of 
the land tortoises, Stylemys, and one 
species of Testudo. Testudo hrontops 
Marsh occurs in the Titanotherium 
zone (Chadron formation) and is 
generally found in the White River 
deposits of Colorado. Of water-living 
forms the White River group of South 
Dakota has furnished one species of 
small turtles related to the Chelydridae 
and now confined to Central America. 
In 1904 Loomis (1904.1) advocated the 
flood-plain origin of the "Titanotherium 
beds" and described two new river-hv- 
ing reptiles — Chrysemys, similar to the 
Emys lativertehralis Cope of the 
Wasatch; and Alligator prenasalis 
(Loomis), recently found by Loomis 
in the beds of Indian Draw, the first 
appearance of this genus in the 
Tertiary. 

The nature of the sandstone or the 
clay in which their remains are found 
makes it impossible to separate the 
mammals of the Chadron formation 
(Titanotherium zone) into plains- 
dwelling and river-dwelling forms, 
because during floods both were swept 
into the streams, the skeletons being 
dissociated and the skulls and jaws 
separated. Doubtless also the remains 
of decaying carcasses were pulled apart 
by crocodiles and garpikes. Only three 
complete skeletons with skulls have 
been found intact, namely, the famous 
Brontops rohustus of the Yale Museum, 
the Brontops rohustus of the American 
Museum, and the Menodus trigonoceras 
of the Munich Museum. "For every 
even approximately complete skele- 
ton," observes Hatcher (1902.3, p. 
124), "there are scores of isolated 
skulls and other bones. Taking Titano- 
therium as an example, I have myself 
collected nearly 200 skulls of this 
animal, while the number of fairly com- 



analysis showed that the fine Brule clays contain 
chiefly terrestrial and plains animals, whereas the 
river-channel sandstones that traverse these clays, 
although contemporaneous, contain chiefly forest and 
fluviatile animals. 



SOUTH DAKOTA IN TITANGTHEEIUM TIME 

The best description of the conditions in the South 
Dakota region while it was inhabited by titanotheres 
is that given by Hatcher (1902.3, pp. 125-127), 
based on his own keen observations, which extended 



ENVIRONMENT OP THE TITANOTHEBES 



107 




75 



100 



over many years of arduous exploration for remains 
of titanotheres: 

The distribution, state of preservation, nature, and character 
of the animal and plant remains found in the clays and sand- 
stones, as well as the distribution of the latter, absolutely pre- 
clude the possibility of their having been deposited in a vast 
lake and favor the presence of streams meandering through 
low, broad, level, open or wooded valleys subjected in part at 
least to frequent inundations, con- 
ditions very similar to those at 
present prevailing in the interior of 
South America, about the head- 
waters of the Orinoco, the Amazon, 
and the Paraguay and Parana 
Rivers. 

Now it is evident that if such 
conditions prevailed in this region 
during the deposition of the White 
River beds there should remain cer- 
tain evidences concerning it, such 
as flUed-in river channels and small 
lagoons with their characteristic 
deposits and remains of the animal 
and vegetable life peculiar to each. 
Moreover, some indication at least 
of the forests should remain and be 
found somewhere in this vast region. 
With these and many other points 
constantly in mind the writer passed 
a considerable portion of the seasons 
of 1900 and 1901 in exploring these 
deposits. Particular attention was 
given to ascertaining whether or not 
they contained an aquatic fauna 
and flora. The sandstone lenses 
were especially examined with ref- 
erence to this, for whether the de- 
posits as a whole were of lacustrine 
origin or not, there could be little 
doubt as to the aqueous origin of the 
sandstones. Though for the most 
part remarkably barren of acjuatic 
life, remains of Trionyx, fishes, and 
crocodiles were found, and in one 
locality the casts of unios were ob- 
served in great numbers. A search 
in the clays of the Titanotherium 
and overlying Oreodon beds was re- 
warded with greater success, for 
numerous thin layers of limestone, 
varying in thickness from a fraction 
of an inch to a foot or more and 
always of limited areal extent, were 
discovered at many horizons rich 
in the remains of fresh-water plants 
and MoUusca, such characteristi- 
cally shallow-water forms as Chara, 
Ldmnaea, Physa, and Planorbis 
occurring in the greatest abundance. 



plants and MoUusca as are Chara and P/ji/sa at various horizons 
throughout the White River series, and in the very midst of the 
region which was supposed to have been occupied by a great 
lake, and intercalated with the clays which advocates of the 
the lake theory maintain were deposited in the deep and quiet 
waters, would appear to preclude the possibility of the existence 
of such a lake in White River times. Moreover, remains of 
forests were found at several places and at different horizons 



O 



''Leptcaicheniev 

y^-^^^^--^-^ zone 



.^rr-f^ Promerycochoerus 
zone 



~^^s^^^^^ zo7ie\ 

pN CHANN EL 5A NDSTONESh 



^: =--^^^=^^ Oreo doTh 
^^^^^^ f^^ zorieX 
j^~^-^2r-^^-^^ = (upper) _\ 

r^AL TERN A tTnG RED A NiTgRA Y LA yEPS 




^^^^^^^^^^Tttcuxotherizurv 

~Z One^r^'DDL E BEDS 



Miohippus 



Mefamynodon 
Meso/iippus bairdi 



Brontops robusfus 



Brontops d/'spar 



Brontops brachycephalus 




Figure 74. — Section of the Big Badlands of South Dakota showing the chief faunal zones 
of the Oligocene (White River group. No. 11, fig. 35) and the Miocene 

The Chadron formation ( Tilanotheriiim zone) is shown as determined by the surveys of Hatcher; the divisions of the 
Brule formation (.Oreodon and Leptauchenia zones) were first established by Wortraan's observations; above is the 
Arikaree formation of Darton (PromeTycochoerus zone). 



I have submitted these 
MoUusca to Drs. Dall, Pilsbry, and Stanton, and all have 
assured me that they belong to species inhabiting swamps 
and small ponds and could not have lived in the midst of a 
great lake; while Dr. Knowlton, who has examined the 
plants, finds in great abundance the stems and seeds of 
Chara, which, as aU know, is distinctly an inhabitant of small 
springs, shallow ponds, and brooks. The presence of these 
thin limestone layers with such characteristically swamp 



throughout these beds. At various localities in the Hat Creek 
basin in Sioux County, Nebr., I discovered remains of the 
silicified trunks of trees and seeds belonging especially to 
Hicoria and Celtis. These were found at various horizons from 
the middle Titanotherium beds to the very top of the Loup 
Fork. And in South Dakota, some 12 mUes north of White 
River, opposite the mouth of Corn Creek, I discovered the 
remains of a not inconsiderable forest. Here in the upper 
Titanotherium beds and lower Oreodon beds there occur, actu- 



108 



TITANOTHEKES OF ANCIENT WYOMING, DAKOTA, AND NEBRASEA. 




Chadron. Fonnatioii 
(TitanotheT-ibUTL 7,OTie} 



C OLORAD O 

Castle Rock 



K A N" S A S 



100 120 140 160 MILES 



Figure 75. — Map showing principal exposures of tlie Chadron formation {Titanotherium zone) of Montana, Soutli 

Dakota, Wyoming, Nebraska, and Colorado 

Chiefly after Darton, 1905, United States Geological Survey. Includes the exposures at Castle Rock (Castle Rock conglomerate), south of Denver, in 
Weld and Logan Counties, Colo., where the early collections of Marsh and Cope were made: iu the outlying localities of Bates Hole and Hat Creek, 
Wyo., from which the Reed and Hatcher collections were made; and in the Big Badlands of Pennington, Custer, and Washington Counties, S. Dak. 
The Titanotherium zone was first observed by Hayden at point X on map and was first subdivided faunistically by Hatcher. 



ENVIRONMENT OF THE TITANOTHEEES 



109 



ally by hundreds, the silicified stumps and partially decayed 
trunks of trees, weathering out of the fine clays of these deposits. 
It was noticeable that only the knots and lower stumps had 
been preserved. Nothing like complete trunks were to be 
observed, and the entire aspect was that of the remains of a 
dead and decayed forest on the margin of some streams, where 
only the less destructible knots and stumps would endure 
sufficiently long to be finally covered up and preserved. In this 
same region there were discernible certain strata which seemed 
to indicate that during the deposition of these beds there has 
been at several horizons an accumulation of vegetable mold 
or humus, and on Dry Creek, some 5 miles northeast of Chadron, 
in Dawes County, Nebr., I observed near the base of the 
Oreodon beds a stratum of some 2 feet of dark-colored humus, 
clearly indicating that this region had not been occupied by a 
great lake while this stratum was being deposited. 

Hatcher concludes that the sandstone, the con- 
glomerate, and a part of the clay were deposited in 
river channels and that the lenses of limestone, which 
are rich in remains of aquatic plants and moUusks, were 
formed in shallow ponds and lakes that were scattered 
over the higher tablelands and the broad flood plains, 
where most of the finer clays were deposited by 
occasional inundations in the rivers and by wind. 
These conditions are similar to those now prevailing 
about the sources of Parana and Paraguay Rivers in 
central South America. ' There the rainy season 
extends from October to April, and the heaviest rains 
fall near its end, when the small rivers from the 
highlands are flooded and pour their waters over the 
flood-plain. The water, however, takes a long time to 
spread over the plain, and it is there highest in July 
and August and lowest in February. The flood plain 
of the Paraguay is 1.50 miles wide and broadens up- 
stream. The flood plains of the upper Paraguay, the 
Amazon, and the Orinoco are confluent. Here we 
have a group of regions that are together probably 
larger than that occupied by the great White River 
group during Oligocene time — namely, 97,500 square 
miles. 

RAPID FIUVIATIIE SEDIMENTATION IN THE CYPRESS HILLS, 
SASKATCHEWAN 

While the conditions thus described existed in the 
Big Badlands of South Dakota, the streams were 
much more active at places in areas to the south and 
north. "That the Cypress Hills Oligocene deposits 
were the result of rapidly flowing water from the west 
is evident," observes Lambe (1908.1, p. 7). He con- 
tinues : 

The thick basal beds of rounded pebbles represent the work 
of a strong transporting force, such as would be supplied by 
a turbulent stream of considerable size carrying eastward 
material from the Rocky Mountains. The sands show false 
bedding as a result of varying currents. With the accumulation 
of material eastward, and consequent reduction of the trans- 
porting force, beds of finer material were deposited at a higher 
level and probably on extensive areas of overflow. 

The beds that yield the most fossils are composed of 
a fine conglomerate, which on disintegration has freed 
the fossils. Associated beds of a rich brown coarse 



sand have also yielded some interesting remains. The 
vertebrate fauna of this region in Saskatchewan 
includes about 37 genera and 58 species, comprising 
among the fishes Amia, Lepidosteus, and catfishes; 
among the reptiles land tortoises, chelydrids, leather- 
backs (Trionyx), anguid lizards, palaeophid snakes, 
and true crocodiles; among the mammals opossums, 
anthracotheres, elotheres, agriochoerids, camels, tragu- 
lids (Leptomeryx) , horses (Mesohippus), hyracodonts, 
true rhinoceroses (aceratheres), titanotheres (several 
lower Oligocene types), sciurid and ischyromyid 
rodents, beavers, hares, hyaenodonts, dogs {Oynodictis, 
DapTiaenus), and cursorial saber-tooths (Dinictis). 

These Saskatchewan beds are not only more than 
twice as thick as those in South Dakota but they have 
afforded a truer picture of the highly diversified rep- 
tilian and mammalian life during the time represented 
by the Titanotherium zone. The species of titano- 
theres that they contain belong chiefly to the lower 
(Chadron A) and middle (Chadron B) levels of the 
Titanotherium zone of South Dakota. 

SLOW SEDIMENTATION IN SOUTH DAKOTA 

As compared with the 500 feet of fluviatile deposits 
of the Titanotherium zone in Saskatchewan the bare 
180 feet of sediments that represent the Titanotherium 
zone in South Dakota are very misleading as to the 
length of geologic time they represent. Deposition in 
South Dakota must have been extremely slow. The 
finer materials that border the river channels and 
compose the clays must have accumulated very 
gradually. That a very long period of geologic time 
elapsed while these sediments were being laid down is 
evident also from the great span of evolutionary 
change indicated by members of each phylum of the 
titanotheres found in this region. On the lowest 
levels are found primitive small-horned titanotheres, 
inferior in size to the smaller existing rhinoceroses; on 
the highest levels are found gigantic animals, of almost 
elephantine proportions, armed with great, powerful 
horns. As a basis for estimating the time required for 
the deposition of the South Dakota sediments, com- 
parison may be made with existing conditions along the 
River Nile, which between Aswan and Cairo is build- 
ing up its bed at the slow average rate of 10 centimeters, 
or 0.32 foot, per century. (Lyons, 1906.1, p. 334.) 
At this rate the deposition of the 180 feet of "Titano- 
therium beds," if composed entirely of fine clays, would 
have required some 55,000 years. On the other hand, 
if we apply Humphreys and Abbot's estimates for the 
Mississippi River system, namely 0.5 foot in 100 years, 
about 36,000 years would have been necessary for the 
deposition of the fine clay materials of the Titano- 
therium zone. The present author inclines to the 
opinion that the lower Oligocene evolution of the 
titanotheres demands a period of not less than 55,000 
years, which would correspond with the present rate 
of sedimentation in the flood plain of the Nile. 



no 



TITANOTHERES OF ANCIENT WTOMING, DAKOTA, AND NEBRASKA 



GEOGRAPHIC DISTRIBUTION OF THE CHADRON FORMATION 

The following list of localities at which remains of 
titanotheres have been collected includes some 
isolated spots where the Chadron formation is rec- 



ognized by a few bones or a single skull, as well as 
points in the classic areas of the Great Plains where 
the history of the titanotheres is chiefly recorded. 



Localities at wMcli tlie principal types and collections of Oligocene titanotheres Jiave heen obtained 

South Dakota 



Eegion 



' Mauvaises Terres of Nebraska," 
Big Badlands of Chej'enne and 
White Rivers of South Dakota. 
This region, the one most exten- 
sively explored, commonly known 
as the Big Badlands, lies between 
White and Cheyenne Rivers, 
southwestern South Dakota, ex- 
tending over the border into Ne- 
braska and Wyoming, including 
the basin of Hat Creek. The 
lower Oligocene has been well dif- 
ferentiated in this region, and the 
records are generally definite. 



Explorations 



Successively explored by mem- 
bers of the American Fur Co. 
(1845), John Evans (1853), 
Meek and Hayden (1853), and 
by more recent explorers: 
Hatcher (for the U. S. Geologi- 
cal Survey, U. S. National 
Museum, and Yale University 
collections, 1886, 1887, 1888, 
1902), Garman (for the Muse- 
um of Comparative Zoology, 
1885), Wortman (for the 
American Museum of Natural 
History, 1892, 1894), Gidley 
and Thomson (for the Amer- 
ican Museum of Natural 
History, 1902), Thomson (for 
the American Museum of 
Natural History, 1904), Dar- 
ton (for the U. S. Geological 
Survey, 1905). 



Locality 



Bear Creek _ 



Indian Draw. 



Quinn Draw_ 



Corral Draw 



Type 



Menodus (Titanotherium) proutii 
(Owen, Norwood, and Evans), 
Diploclonus (Megacerops) tyleri 
(Lull), Brontotherium (Tita- 
nops elatus) gigas Marsh, Men- 
odus (Menops) varians (Marsh) 
Brontotherium tichoceras 
(Scott and Osborn), B. doli- 
choceras (Scott and Osborn), 
B. platyceras, B. leidyi Osborn, 
Brontops validus Marsh, Al- 
lops crassioornis Marsh, A. 
serotinus Marsh. 

Brontotherium medium Osborn, 
B. hatcheri Osborn, Mega- 
cerops (Symborodon) copei 
Osborn, Allops (Megacerops) 
marshi Osborn. 

Diploclonus bicornutus (Osborn) , 
Brontops brachycephalus Os- 
born, Brontotherium (Titano- 
therium) ramosum Osborn. 

Allops walcotti Osborn. 



Nebraska 



White River, northern Nebraska 

Hat Creek, Sioux County, Nebr. 
Hat Creek, a branch of the South 
Fork of Cheyenne River, rises in 
the canyon in the north front of 
Pine Ridge, Sioux County, and re- 
ceives numerous branches, also 
heading in this front. 

Big Cottonwood Creek, Sioux Coun- 
ty, Nebr. The exposures of the 
Titanotherium zone at the head of 
Big Cottonwood Creek are coex- 
tensive with those of the Hat 
Creek basin, which lies north of 
this locality. Much of Hatcher's 
collecting was done in exposures 
on the low divide connecting Big 
Cottonwood Creek with the ex- 
posures in the Hat Creek basin. 
Adelia is a station on the Chicago, 
Burlington & Quincy R. R., on 
the outskirts of this particular 
region. 



Marsh and Clifford (for Yale 
University, 1874). 

Hatcher (for the U. S. Geologi- 
cal Survey, 1886; for the Car- 
negie Museum, 1900), Peter- 
son(forthe Carnegie Museum, 
1901, 1902). 



Hatcher (for the U. S. Geo- 
logical Survey, 1886; for the 
Carnegie Museum, 1900). 



Dry Creek_ 
Hat Creek. 



Brontops rdbustus Marsh. 
Brontops dispar Marsh. 



ENVIEONMENT OF THE TITANOTHEKES 



111 



Localities at which the principal types and collections of Oligocene titanotheres have been obtained — Continued 

Colorado 



Region 



Explorations 



Locality 



Type 



Northeastern Colorado, Logan and 
Weld Counties, exposures south 
of the Pawnee Buttes escarpment 
and some distance north of the 
South Platte River. The lower 
Oligocene was differentiated and 
explored by Cope in 1873, but it 
has hardly been touched by any 
subsequent work (Horsetail Creek 
beds of Matthew). 

Lower Oligocene has been recognized 
at other points in Colorado, as in 
the vicinity of Akron, but no 
adequate collections have been 
made for the identification of 
species. 



Castle Rock conglomerate (upper 
part of " Monument Creek beds"), 
Colorado. 



Marsh (for the Yale Museum, 
1870), also field collectors. 



Cope 1873, Matthew, Brown, 
Martin (for the American 
Museum of Natural History, 
1898), Matthew, Brown (for 
the American Museum of 
Natural History, 1901). 



Darton (for the U. S. Geological 
Survey, 1905, 1906). Richard- 
son (for the U. S. Geological 
Survey, 1912). 



Probably in Weld 
County, Colo.; 
near Gerry's ranch, 
Colo.; also 4 
miles south of 
Pond Springs, 
Colo. Collector, 
Devendorf. 



Horsetail Creek , 
Colo. 



Brontotherium gigas Marsh, 
M e n o d u s (Brontotherium) 
ingens Marsh, Brontotherium 
(Titanops) curtum Marsh. 



Megacerops (Symborodon) acer 
Cope, M. (S.) altirostris Cope, 
M. (S.) bucco Cope, M. (S.) tor- 
vus Cope, Menodus (Symboro- 
don) trigonoceras Cope, M. (S.) 
heloceras Cope, M. (S.) hypo- 
ceras Cope, Megacerops riggsi 
Osborn. 



Wyoming 



Beaver Divide, Fremont County, 
Wyo. 

Bates Hole, Natrona County, Wyo. 
Exposures lying between Bates 
Hole, at the north end of the 
Laramie Plains, and Beaver 
Divide, at some distance to the 
west, have been casually examined 
by W. H. Reed and W. D. Mat- 
thew, who have recognized a 
lower Oligocene fauna, but no 
S3'stematic exploration has been 
made. A number of well pre- 
served specimens of titanotheres 
(Univ. Wyoming Mus.) were 
found in this area. 



Granger (for the American Mu- 
seum of Natural History, 
1910). 

Reed (for the University of 
Wyoming, 1907, 1908). 



Pipestone Creek and Thompson 
Creek, Jefferson County, Mont. 
In southwestern Montana, at 
Pipestone Springs, McCarty 
Mountain, north of Dillon, and 
elsewhere, small scattered expo- 
sures. A considerable fauna, of 
small species, has been described 
from these beds. 



Douglass (for the Carnegie Mu- 
seum, 1899, 1901, 1903), 
Matthew (for the American 
Museum of Natural History, 
1902). 



Saskatchewan 



Swift Current Creek, Cypress Hills, 


McConnell and Weston (1883), 




Menodus angustigenis, M. sel 


Saskatchewan. 


Weston (1888, 1889), Lambe 




wynianus, Megacerops syceras 




(1904). 




M. assiniboiensis, M. primi- 
tivus. 



112 



TITANOTHERES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



The Chadron formation was especially explored by 
Darton, under whose direction a map showing its gen- 
eral distribution Csee fig. 8) was prepared. His de- 
scription of this map may be paraphrased as follows: 

The most extensive outcrops are in the soutliwestern portion 
of South Dalcota, in a district known as the Big Badlands. 
These extend along the valley of White River and in the ad- 
jacent divide between White River and Chej'enne River 
West of the latter the formation caps many of the divides 
extending to and up the eastern slopes of the Black Hills. The 
formation extends eastward to the vicinity of longitude 100°, 
and it extends westward along the foot of Pine Ridge through 
Dawes and Sioux Counties in Nebraska and Converse County, 
Wyo., to the north end of the Laramie Range. The formation 
probably underlies a large portion of western Nebraska, but it 
only reaches the surface at the foot of Pine Ridge and along the 
north Platte Valley west of Scotts Bluff. Isolated outcrops 
are also reported at Valentine and Lone Pine. The formation 
appears extensively in northeastern Colorado, on both sides of 
the valley of South Platte River. There are prominent ex- 
posures west of Akron, south of which the formation extends 
across the greater part of Washington County. South of 
Denver an extensive area caps the divide between the South 
Platte and the Arkansas Rivers, at the foot of the Rocky 
Mountains. The deposits in this area have been designated 
the Monument Creek formation. This consists of two members 
of which the upper [now called Castle Rock conglomerate; 
(Richardson, 1912.1)] has recently yielded Titanotherium 
remains. Other outlying areas of the formation occur in Bates 
Hole west of the Laramie Range [Wyoming], in Butte County, 
S. Dak., and in the southeastern corner of Montana. 

An important additional exposure of the Titanothe- 
rium zone discovered by the American ]\luseum expe- 
dition of 1909 is at Beaver Divide (Wagonbed Spring), 
in the southern rim of the Wind River Basin, near 
Hailey, Wyo. Here a deposit containing a skull of 
a primitive Oligocene titanothere was found overlying 
a deposit of upper Eocene age containing Amynodon. 

The thickness of the Chadron formation varies, but 
in some places it reaches 180 feet. It consists of 

clays, sands, gravels, and sandstones, clay predominating greatly 
over the other materials. * * * This clay is of pale-greenish 
color, weathering in typical badland form and often having the 
peculiar character of fuller's earth. At the base of the formation 
there is usually a bed of gravel and sand merging upward into 
sands and sandy clays, which in the Big Badlands and western 
Nebraska are often of a reddish color. At various horizons 
through the formation there are beds of sandstone from a few 
inches to 4 feet in thickness and of local extent. Ordinarily 
these coarse materials exhibit current bedding and from their 
character and relations are clearly the products of running 
water. Beds of volcanic ash occur extensively in the Chadron 
formation in the Big Badlands and at intervals along the 
northern front of Pine Ridge (South Dakota). 
Hatcher observes (1893.1, pp. 206-207): 
The clays- greatly predominate, consist of very fine particles, 
and are quite compact. In places they are composed almost 
entirely of pure kaolin, but they often contain a considerable 
portion of sand. Near the bottom of the beds the color is often 
red or variegated, due to the presence in them of small quanti- 
ties of red oxide of iron; but the prevailing color is a very char- 
acteristic and delicate greenish white. * * * Owing to 
the extreme minuteness of the particles forming the clays and 



the absence of sufficient cementing material in them, in most 
places they readily yield to the action of water and are quite 
rapidly eroded. The clays of the Titanotherium beds were 
probably derived from two sources, viz, from the Cretaceous 
clays and shales and from the kaolinization of granite feldspars. 

The sandstones are never entirely continuous and never more 
than a few feet thick. They present everj- degree of compact- 
ness, from loose beds of sand to the most solid sandstones. 
They are composed of quartz, feldspar, and mica and are evi- 
dently of granite origin. When solidified the cementing sub- 
stance is carbonate of lime. 

The conglomerates, like the sandstones, are not constant, are 
of very limited extent, never more than a few feet thick. They 
are usually quite hard, being firmly held together by carbonate 
of lime. A section of the beds taken at any point and showing 
the relative position and thickness of the sandstones, clays, and 
conglomerates is of little [stratigraphic] value, since these vary 
much at different and quite adjacent localities. 

These descriptions by Darton and Hatcher reveal 
a wide contrast between the composition of the 
Titanotherium-heen-ing beds and that of the upper 
Eocene deposits of the mountain-basin region. 

COMPARISONS OF BASINS IN WESTERN UNITED STATES WITH THE FIOOD 
PLAIN OF THE NUE 

The flood-plain deposition of the Nile, which has 
been very carefully studied, also throws light on the 
mode of formation of parts of the Chadron formation. 
The following passages are taken from "The physi- 
ography of the River Nile and its basin," by Capt. 
H. G. Lyons (1906.1, pp. 241, 311, 334): 

When rivers already loaded with sediment emerge from their 
mountain valleys of high slope on to a level plain under arid 
climatic conditions where the water table is at somie distance 
from the surface their water sinks in almost at once instead of 
flowing on the surface and therefore deposits its load of sediment 
as an alluvial fan. This fan is built up most rapidly at its 
head, and as the floods of successive years come down new tem- 
porary channels are formed which divide and reunite, forming 
a network of channels, each b}' deposit building up banks for 
itself, which are probably cut through in the next season of the 
summer rainfall. 

While it is the finer silt which is deposited in the irrigation 
basins, on the shelving banks of the river, and on such parts of 
the flood plain as are annually flooded, it is the bottom load 
which is deposited in the bed of the river itself, and this con- 
sists of the coarser sand which the current can not carry so 
readily as the finer material. If the Nile mud is treated by 
levigation so as to remove the finest particles of clay and sand 
the residue is a fine whitish-gray sand, such as is seen forming 
sand banks in the Nile wherever the conformation of the river 
is such that the velocity of the fiood current is reduced at that 
point. 

In this Aswan-Cairo reach of the Nile, then, we have to do 
with a river which is fiowing with a low slope through an alluvial 
plain which it has formed and which, if uncontroUed, it annually 
floods, depositing on the flood plains part of its load of silt as 
the velocity of flood water is diminished. 

The Nile between Aswan and Cairo follows a depression in 
which it has gradually deposited a considerable thickness of 
alluvial mud, and now it meanders on the flood plain which it 
has formed. In earlier times side channels followed the lower 
margins of the valleys, and lagoons and swamps existed in the 
same part of the valley. 



ENVIRONMENT OF THE TITANOTHERES 



113 



FAUNAL DIVISIONS IN THE CHADRON FORMATION 
THEEE FAUNISTIC LEVELS DETERMINED 

In the series of sediments that were deposited on 
the uneven surfaces of the Pierre shale and that con- 
sist of fine clays, which were traversed and at many 
places secondarily eroded by river channels and which 
were very gradually accumulated during an extremely 
long period, we should not expect to find anything re- 
sembling clearly defined stratification or horizontal 
and vertical succession of species and genera. Never- 
theless, we owe to the genius and the untiring explora- 
tion of Hatcher a division of the Chadron formation 
into lower, middle, and upper levels, which we shall 
designate Chadron A, Chadron B, and Chadron C, 
and which correspond to similar divisions of the deposits 
of the Rocky Mountain basins. 

In his paper of 1893 (1893.1), "The Titanoiherium 
beds," Hatcher remarked that these beds were so 
named by Meek and Hayden in 1857 from the genus 
TitanotJierium, established by Leidy in 1852. Al- 
though we are obliged to replace the generic name 
TitanotJierium by Menodus it seems best to retain 
Titanotherium as the historic zonal name for these sig- 
nificant beds. 

The thickness of the "Titanotherium beds" at dif- 
ferent localities in Wyoming, Colorado, the Dakotas, 
and Saskatchewan, as recorded above, varies, having 
a maximum of 500 feet and a minimum of 30 feet. 

Hatcher, accepting a total of 180 feet as the maxi- 
mum thickness of these beds in the Big Badlands of 
South Dakota, assigned 50 feet to the lower level, 100 
feet to the middle, and 30 feet to the upper (1893.1, 
p. 210). During the field seasons of 1886, 1887, and 
1888 Hatcher collected for the present monograph 
material including 105 nearly complete Titanotherium 
{Menodus) skulls and parts of numerous skeletons, as 
well as disarticulated bones, besides remains of many 
other associated animals. Early in the season of 
1886 it became apparent that certain forms of skulls 
were characteristic of certain horizons in the "Titano- 
therium beds." This fact indicated the desirability of 
keeping, so far as possible, an exact record of the 
horizon from which each skull or skeleton was taken. 
From actual measurement the vertical range of the 
titanotheres in the Big Badlands was found to be 



about 180 feet. For convenience in keeping a record 
of horizons the beds were divided into three divisions 
of 60 feet each, and each of these three divisions was 
subdivided into three divisions of 20 feet each. As 
each skull or skeleton was dug out a separate letter or 
number was given to it and it was assigned to that 
subdivision of the beds from which it was taken. 

STRATIGEAPHIC DISTRIBUTION OF SPECIES OF OLIGOCENE 
TITANOTHERES 

In 1888 Hatcher drew up a manuscript table for 
Professor Marsh in which the lower, middle, and 
upper divisions of the "Titanotherium beds" were 
each subdivided into three levels, and in which he 
placed the letters assigned to many of the skulls 
found by him. In 1901 Hatcher revised this table for 
Osborn for use in the present monograph. In the 
summer of 1902 the United States Geological Survey 
sent Messrs. N. H. Darton and J. B. Hatcher to the 
Big Badlands of South Dakota for a resurvey of the 
localities where some of the skulls were found by 
Hatcher in order to determine precisely the elevation 
of these localities above the Pierre shale, at the 
base of the beds. Prof. Eberhard Fraas, of Stuttgart, 
accompanied the party and made some interesting 
observations on the mode of deposition of these beds. 
(Fraas, 1901.1.) This experienced geologist con- 
cluded that the "Titanotherium beds" consisted of 
river and flood-plain deposits whose surfaces were 
exposed during the dry seasons of the year; that 
parts of the overlying Brule clay — the beds in the mid- 
dle Oreodon zone — were deposited in shallow lakes, the 
dissolved materials, of varying concentration, giving 
rise to the banded layers; and that the reddish strata 
of the Oreodon zone (Brule) were formed of eolian 
loess. 

In the following table the results of records and ob- 
servations made by Hatcher, indicated by the abbre- 
viation J. B. H., are supplemented by the results of a 
few observations made by N. H. Darton of the United 
States Geological Survey, E. S. Riggs of the Field 
Museum, W. H. Reed of the University of Wyoming, 
and Walter Granger of the American Museum. The 
species are arranged in the five generic phyla deter- 
mined by Osborn, namely, Brontops, Allops, Menodus 
{= Titanotherium) , Megacerops { = Symborodon) , Bron- 
totherium. 



114 



TITANOTHERES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 

Geologic succession of Oligocene titanotheres in the Ohadron formation 



[Levels (above Pierre shale except as otherwise indicated) mostly taken from J. B. Hatcher's field records of 1886, 1887, 1888. Genera and species of fossils determined by 

H. r. Osborn and W. K. Gregory] 





Level 


Brontops 


Allops 


Menodus 


Megacerops 


Brontotherium 


Classification 
uncertain 














Brontotherium platyceras. 
















Skull 12161, Field Mus. 
















Upper levels (E. S. 












Menodus giganteus. 




Riggs). 
Brontotherium medium. 












Field Mus. skull 














P5927. Near top of 




Skull w, Nat. Mus. 4256 












upper Titanothe- 




(type); "from the ex- 












rium beds (E. S. 




treme top of the Bronto- 






3 


Brontops dispar?. Skull G, Nat. Mas. 
4248. Record and level uncertain 
(J. B. H.). 


Allops serotinus. 
Skull I, Nat. Mus. 
2151. -l-80feet. 
Same locality as H. 


Riggs). 
Menodus giganteus. 
Univ. Wyoming 
skulls. Upper beds 
(W.H.Reed). Bates 
Hole, Wyo. 




therium beds. Oreodon 
teeth were found in dig- 
ging it up" (J. B. H.). 
4-81 feet, "well up in red 
clays" (J. B. H.). 
Brontotherium curtum. 


Skulls. 










Menodus giganteus. 




Skull Y', Nat. Mus. 




o 








Skull r, Nat. Mus. 




1211. 4-93.3 feet. Skull 




e 








1212. 




q, Nat. Mus. 4946. 4-89 




3 










Megacerops copei. 


feet. Skull g, Nat. Mus. 




1 






Allops serotinus. 




Skull V, Nat. Mus. 


4244. 




O 






Skull H, Nat. Mus. 
4251. 4-77 feet; 34 




4711. -1-65.4 feet, 
possibly In C2 (J. B. 






S 












P 




Brontops dispar. Skull p, Nat. Mus. 


feet below top. 




H.). ?SkullL',Nat. 






s 




1217. 






Mus. 4700. 






d 




Brontops dispar?. Skull d, Nat. Mus. 

4696, 
Brontops sp. Mounted skeleton, Am, 


Allops serotinus 


?Menodus giganteus. 
Skull G', Nat. Mus. 
4291. "From mid- 




?Brontotherium hatcheri?. 
Univ. Wyoming skull 1. 


Skulls R', W'. 




2 


Mus. 618. "Very high, 8-10 feet 
from top." (J. B. H.) "32 feet be- 
low the 3-foot siliceous limestone 
layer at top of Titanotherium series." 
(N. H. Darton, 1901.) 


Skull i, Nat. Mus. 
4938. 


dle beds or toward 
the top" (J. B. H.). 




"Upper beds." Bates 
Hole, Wyo. 














Megacerops acer. 


Brontotherium medium?. 










Allops crassicornis. 


Menodus proutii. 


Univ. Wyoming 


Skull N', Nat. Mus. 






1 


?Brontops dispar. Skull h, Nat. Mus. 


Skull Z', Nat. Mus. 
4289. "J. B. H. in- 


Skull e, Nat. Mus. 


skull 2. "Upper 
beds about two- 


4699. Level? 


Skulls M', U'. 






4944. 




4701. Level doubt- 




Brontotherium gigas. 










clined to place this 


ful (J. B. H.). 


thirdsupfrombase" 


Skull H', Nat. Mus. 










skull higher up." 




(W. H. Reed). 
Bates Hole, Wyo. 


4262. 








Brontops dispar. Skull D', Nat. Mus. 




Menodus giganteus. 


Megacerops bucco. 
Skull 0', Nat. Mus. 


Brontotherium medium?. 








4706. Level essentially correct (J. B. 




Skull r, Nat. Mus. 


Skullu, Nat. Mus. 4716. 






3 


H.). 




1220 (very large). 


4705. Level rather 


Level? 


Skulls S', C, F', 






Brontops robustus. Type skeleton. 




Menodus trigonoceras. 


doubtful (J. B.H.). 
4-46,7 feet. 


Brontotherium hatcheri. 


E'. 






Yale Mus. 12048. 60 feet below 




Skull 0, Nat. Mus. 


Type skull a, Nat. Mus. 








summit (J. B. H.). 




4257. 


1216. 














Megacerops sp. (or 






g 




Brontops dispar. Skull f, Nat. Mus. 


Allops marshi?. 




Brontotherium 






g 




4703. Level certainly B2 (J. B. H.). 


Skull t, Nat. Mus. 




hatcheri). Skull Q', 


Brontotherium hypoceras. 




3 




Skull D (type), Nat. Mus. 4941. 


4942. 




Nat. Mus. 4255. 


Skull 1, Nat. Mus. 


Skulls T', P', C 


^ 


2 


Hat Creek. Levelcorrect (J. B. H.). 


Allops marshi?. 




"Certainly in mid- 


4273(?). Level proba- 


B'. 




Skull K, Nat. Mus. 4290 (type of 


Skull A', Nat. Mus. 




dle beds, perhaps in 


bly correct (J. B. H.). 




1 




B. validus). 


1215. 




middlelevel"(J. B. 






'f> 










H.). 






Id 
















s 




Brontops dispar. Skull (erroneously 












m 


1 


lettered P). Nat. Mus. 4245. Skull 
J, Nat; Mus. 4738. Hat Creek, 
Lower B, probably correct (J. B. H.). 
Brontops brachycephalus?. Skull M, 
Nat. Mus. 4259. -(-.55.6feet(J.B.H.). 




Menodus trigonoceras. 
Skull G', Nat. Mus. 
4291. 






Skulls N, A, B 
No.? (a large 
skull). 






Level B, probably correct (J. B. H.). 














Brontops brachycephalus. Skull i", 
















Nat. Mus. 4258. -f71.4 feet (J. B. 
















H.); -f48.5feet(N.H. Darton). 













ENVIKONMENT OF THE TITANOTHEKES 

Geologic succession of Oligocene titanoiheres in the Chadron formation — Continued 



115 





Level 


Brontops 


A Hops 


Menodus 


Megacerops 


Brontotherimn 


Classification 
uncertain 




3 


Brontops brachycephalus. Skull X', 

Nat. Mus. 1214. Level probably 

correct (J. B. H.). 
Brontops brachycephalus?. Skull m, 

Nat. Mus. 4940. Level probably 

correct (J. B. H.). 








Brontotherium ?hypoceras. 
Skull K', Nat. Mus. 
4702. Level very doubt- 
ful (J. B. H.). 


Skulls V, I'. 


B 

s 


2 


Diploclonus tyleri. Type skull. 35 
feet above Pierre shale and 165 feet 
below top of formation (R. S. Lull). 


Allops marshi?. 
Skull E, Nat. Mus. 
1213. Level proba- 
blycorrect (J.B.H.). 






Brontotherium leidyi 
(type). Skull R, Nat. 
Mus. 4249. Level cor- 
rect (J. B. H.). 




1 


1 


Brontops dispar??. Skull P, Nat. 
Mus.? (not 4245). "This skull in 
pieces found July 4, 1887, right at 
base of beds" (J. B. H.). 

Brontops brachycephalus. Skull c, 
Nat. Mus. 4261 (type) . Lower levels 
(J. B. H.). Skull b, Nat. Mus. 4947 
(marked 1991). +14.4 feet; 130.6 feet 
below top (J. B. H.). Very young 
skull (new born?), Univ. Wyoming. 
"From extremely low level" (W. 
H. Heed). Bates Hole, Wyo. 


Allops walcotti (type) . 
Skull Q, Nat. Mus. 
4260. "Probably 
lower beds, level A, 
fine-grained sand- 
stones" (J. B. 11.). 


Menodus heloceras. 
Skull, Am. Mus. 
14576. At base of 
beds near Hailey, 
Wyo. (W. Granger). 




Brontotherium leidyi. 
Skull, Carnegie Mus. 93. 
Hat Creek, Nebr., 15 or 
20 feet from bottom of 
lower beds (J. B.H.). 





HATCHER'S COHECTIONS, 1886-1888 

According to a report delivered orally by Hatcher to 
the author in 1901, the collections made by him in 1886 
included 24 skulls, some from Hat Creek, Nebr., and 
some from the Big Badlands of South Dakota, which 
were designated in his records and field notebooks by 
the letters A, B, C, etc., but which now bear United 
States National Museum numbers. In 1887 Hatcher 
collected from Sioux County, Nebr., mostly from Big 
Cottonwood Creek (adjoining Hat Creek), a second 
series of skulls, which he similarly designated by the 
letters A to K. Later in the same season he moved 
camp to the South Dakota Badlands and collected the 
skulls L to Z and a to w. Thus during the season of 

1887 he collected 45 skulls. During the season of 

1888 he collected another series of 24 skulls in the 
South Dakota Badlands, which he designated by the 
letters a', b', c', to z' but which Professor Marsh 
later relettered A' to Z' . Subsequently the catalogu ers 
of the United States National Museum assigned 
numbers to all these skulls. These revisions of the 
records have caused confusion, so that it is now doubt- 
ful whether certain skulls that bear capital letters and 
United States National Museum numbers belong to 
the series of 1886 from Hat Creek, Nebr., or to the 
series of 1887 from Big Cottonwood Creek and the 
South Dakota Badlands. Such uncertainty, of course, 
involves equal uncertainty as to the localities and 
geologic levels at which the specimens were obtained, 
but nearly all uncertainties have been settled by Dr. 
W. K. Gregory through careful examination of all 
the available evidence, with the assistance of Mr. 
J. W. Gidley, of the United States National Museum. 
The above table is based on these original and revised 
records. 



This remarkable collection, now preserved in the 
United States National Museum, constitutes the 
reference standard as specifically determined by the 
author with the assistance of Messrs. Gidley and 
Gilmore and includes the skulls and jaws indicated 
below, which are enumerated in detail under the 
respective genera in Chapter VI : 

Allops phylum: 24 skulls and lower jaws in four specific stages. 
Diploclonus phylum: 1 skull in one specific stage. 
Brontops phylum: 58 skulls and jaws in three specific stages. 
Brontotherium phylum: 42 skulls and jaws in nine specific 



Megacerops phylum: 7 skulls and jaws in three specific stages. 
Menodus phylum: 26 skulls and jaws in four specific stages. 




Figure 76. — Section showing the results of stratigraphic 
leveling in the Chadron formation (Titanotherium zone) in 
the badlands of White River, S. Dak., in June, 1901, by 
N. H. Darton 
The results are affected by dip, by unconformity, and by variation in the thickness 
of the beds. In determining the dip the beds showing the nearest reliable con- 
tacts of the Chadron with the Pierre formation were selected for all the levelings, 
and as most of the distances determined were short and were measured along the 
strike of the low-dipping beds the angle of the dip is unimportant . The Chadron 
formation lies on a smooth plane of unconformity, and its basal member is gen- 
erally continuous but was doubtless laid down against a sloping shore, and the 
layers are not synchronous throughout its extent. Nearly all the bones listed in 
the text, however, were found in an area so small that this unconformity is unim- 
portant. The variation in the thickness of beds is the most important factor 
aflecting the determination of the stratigraphic levels and one that could not be 
accurately determined, for the beds present so much variation in character that 
they can not be followed for a distance long enough to afford a basis for strati- 
graphic subdivision of the formation. A horizon 30 feet above the base of the 
formation at one point may represent a horizon 45 feet above it at another point; 
thus a bone found at A may have been deposited at the same time as a bone 
found at B. 



116 



TITANOTHEEES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



SOURCES OF ERROR IM DETERMINING STRATIGEAPHIC LEVELS 

It should be borne in mind that owing to the great 
difference in the thickness of the " Titanotherium 
beds" in different locaUties and to the irregular topog- 
raphy of the Pierre shale upon which the beds rest 
it often happens that the base of these beds at one 
point may correspond to the middle Titanotherium 
zone at others, so that an exact stratigraphic subdi- 
vision of the Chadron formation over wide areas is 



some 165 feet below their summit. Hence this skull 
is assigned to Hatcher's level A 3, although its large 
size and progressive structure would lead one to infer 
that it came from the upper Titanotherium zone 
(Chadron C). 

Notwithstanding these discrepancies we are able 
to follow the evolution of five separate phyla of 
titanotheres, from the small animals of the lower 
Titanotherium zone (Chadron A), which have small 




Adaptive 

radiatiorL of the 

subfaTnilies of 

PerLssodcLctyls 

Haiiits OTui habitats 



AQUATIC Kv-3 MEDIPORTAL WpM GRAVI PORTAL ESS 



Figure 77. — The family tree of the Perissodactyla 

Adaptive radiation of the 9 families and 35 subfamilies. Their divergence in limb and foot structure into cursorial, forest-living, mediportal, 
and graviportal types, and in tooth structure into browsing and grazing types, is indicated by respective symbols. 



not at all possible. In spite of such opportunity 
for error only a few well-authenticated records (such 
as that of the type of Brontops dispar) appear to con- 
tradict Hatcher's statement that the titanotheres of 
advanced structural development are confined to 
the upper levels of the beds. One such striking 
exception is recorded by Prof. R. S. Lull (1905.1), 
who states that he found the type of Diploclonus 
tyleri at a point only 35 feet above the Pierre shalei 
at the base of the Titanotherimn beds, which was there 



horns, through intermediate types to the latest forms, 
which have highly specialized skulls, from the top 
of the " Titanotherium beds." This evolution was 
rather even and regular in the phyla Brontotherium 
and Menodus {= Titanotherium) , but in the Brontops 
phylum it appears that some of the primitive types 
of the lower zone survived with little change into the 
middle zone (Chadron B), and that other primitive 
types evolved gradually into the more specialized 
species of the middle and upper zones. 



ENVIEONMBNT OP THE TITAN0THEEE5 



117 



Height in feet above Pierre shale at which remains of titanotheres were found as determined in 1901 hy J. B. 

Hatcher and N. H. Barton'^ 



Skull Y' (?). "Big flat-horned skull in National Museum"; Nat. Mus. 1211 (?) 

Skull Q. Indian Draw. Probably skull "small q" was meant (Nat. Mus. 4946, Brontotherium curtum), 

' ' wrongly lettered Q" 

Large-horned red skull. Nat. Mus. 4256, Brontotherium medium (type) 

Skeleton. Am. Mus. 518, Brontops robustus? 

Skull v. Indian Draw. Nat. Mus. 4711, Megacerops copei (type) 

Long-horned skull. West branch of Indian Draw. Brontotherium ramosum 

Skull M. Near Middle Corral Draw. Nat. Mus. 4259, Brontops brachycephalus 

Skull "F." Quinn Draw, South Dakota. ?Nat. Mus. 4258, Brontops brachycephalus 

Skull O'. South side of west fork of Corral Draw. Nat. Mus. 4705, Megacerops "bucco," female 

Skull "I." Quinn Draw, South Dakota. Nat. Mus. 2151, "AUops serotinus," female 

Skull "H." Quinn Draw, South Dakota. Nat. Mus. 4251, Allops serotinus (type) 

Skull "little F." Quinn Draw, South Dakota. Nat. Mus. 4703, Brontops dispar 

Little skull "B." On fork of west fork of Corral (?Quinn, J. B. H.) Draw. Probably skull b, Nat. Mus. 

4947, Brontops brachycephalus, female, aged 



81 
C) 
65.4 



"93. 3 



81. 



n 



55. 6 
71.4 
46. 7 
80 
77-34 
62 



65. 3 
62. 3 
55. 6 
48. 5 
46. 7 
43.5 
40.7 
39.0 

14.4 



• In a letter to the author, dated July 31, 1901, Hatcher expressed grave doubts as to the accuracy of these levels on account of practical difficulties encountered in the 
field. 

» From horizon ot skull O', Nat. Mus. 4705, to horizon of this skull there is a vertical upgrade of 46.6 feet. 

' 27 leet above skull V. 

•I Very high, 8 to 10 leet from top of titanothere zone (Hatcherl. 

• 32 leet below the 3-foot sihceous limestone layer at top of Titamtherium zone. The Pierre shale contact was far away, and although it was on a line of levels the dip 
in interval could not be ascertained precisely (Darton). 



MAMMALIAN LIFE OF THE LOWER OLIGOCENE TITANOTHEEIUM ZONE 

The most highly characteristic feature of the Ohgo- 
cene mammals as a whole, compared with the Eocene 
mammals, is their decided modernization, which is 
shown in the following table giving the percentages 
of the modern and the archaic families of the Oligo- 
cene Plains fauna as compared with those of the Eocene 
mountain-basin fauna. 

Percentages of modern and archaic families in Eocene, Oligocene, 
and Miocene time 



Basal Eocene 

Lower and middle Eocene 

Upper Eocene 

Lower Oligocene 

Miocene 



Modern fami- 
lies or those 
closely related 
or ancestral 
to modern 
families 



Archaic fami- 
lies supposed 
to be wholly 
extinct and 
not closely 
related to 
modern types 



This modernization of mammalian life is in part 
real and in part apparent, because the Plains fauna 
presents for the first time the full aspect of the upland, 
plains, and meadow life, especially the smaller and 
larger herbivorous ungulates. This life is, however, 
only partly revealed in the Titanotherium zone, in 
which conditions for the fossilization and preserva- 
tion of the land fauna were less favorable than in the 
overlying Oreodon zone (Brule clay). In fact, re- 
mains of the small ungulates, such as the horses of the 
period (Mesohippus) , are very rarely preserved in 
either the coarser or the finer sediments of the Chad- 
ron of South Dakota but are found more abundantly 
in the sediments of Pipestone Creek, Thompson 
Creek, and other areas in Montana and in the Swift 
Current Creek area of Saskatchewan. The entire 
Titanotherium zone fauna as listed by Osborn and 
Matthew (1909.321, pp. 103, 104) contains representa- 
tives of 6 orders and 24 families of mammals, which 
are of interest and value as showing the principal 
types of mamnxals that were in competition with the 
titanotheres in the struggle for existence. 



118 



TITANOTHERES OP ANCIENT WYOMING, DAKOTA, AND NEBRASKA 

Fish, reptile, and mammal Jauna contemporary with the titanotheres 



Common name or comparable form, habits or habitat, 
nature of deposits, etc. 


Classific name 


Region inhabited 




PISCES 
Actinopterygii: 
Amiidae — 












Do. 






Do. 


Gar pikes (Lepidosteus) ; rivers and streams. - 


Lepidosteidae — 


Do. 


Catfislies (siluroids) ; rivers and streams 


Siluridae — 


Do. 






Do. 






Do. 


Crocodiles and alligators; rivers and streams __ 


REPTILIA 

Crocodilia: 

Crocodilidae — 


Do. 






South Dakota. 




Squamata: 

Anguidae — 






Helodermoides tuberculatus Douglass 


Montana. 
Do. 




Palaeophidae — 




Subterrestrial ; wet and forested places 


Chelonia: 

Dermatemydidae — 

Xenochelys formosa Hay 


South Dakota. 




Emydidae — 








Land tortoises; characteristic of uplands, open 
country. 


Testudinidae — 


Do. 










Do. 


Soft-shelled turtles; aquatic; remains found in 
fluviatile sandstones. 


Trionychidae— 

PIat3'peltis leucopotamica Cope 


Saskatchewan, South Da- 




MAMMALIA 

Marsupiaha: 

Didelphidae — 


kota. 








Cursorial, predacious, like the Thylacinus of 
Tasmania; resembling modern wolves in 


Ferae : 

Hyaenodontidae — 

"Pseudopterodon" minutus" (Douglass) 


Montana. 


Comparable with hyena of Africa; powerful 




Saskatchewan, South 


jaws. 


Hyaenodon cf. H. crucians Leidy 


Dakota. 
Do. 




Canidae — 






Daphaenus cf. D. hartshornianus Cope 

Daphaenus cf . D. felinus Scott ** 


Saskatchewan. 
Do. 






South Dakota. 


Analogous to the marten and mongoose 


Nothocyon "^ paterculus (Matthew) 


Montana. 
South Dakota. 




Mustelidae — 

Bunaelurus inf elix 




Analogous to the marten and polecat 


Montana. 



" This is, in fact, an undescribed genus, more primitive than Hyaenodon and Pterodon, allied apparently to Sinopa and Tritemnodon. 

>> If Mr. Lambe's figure is correct this can hardly be Z>. felinus; it agrees much better with D. dodgei Scott. 

' CynodicHs is not applicable to the American OlJgocene species that have been called by that name. Nothocyon is next in priority among available names. The type, 
however (N. geismarianus) , is a large and rather specialized species from the John Day formation. It maybe necessary to separate the small species from the middle and 
lower Oligocene under the name Pseudocynodictis (Schlosser). 



ENVIKONMENT OP THE TITAN OTHERES 

Fish, reptile, and mammal fauna contemporary with the titanotheres — Continued 



119 



Common name or comparable form, habits or habitat, 
nature of deposits, etc. 


Classific name 


Region inhabited 


Analogous to the leopard and cheetah 


MAMMALIA — Continued 

Ferae — Continued. 
Felidae— 


South Dakota. 












Possibly like the modern Gymnura of the East 


Insectivora: 

Leptictidae — 


Montana. 




Do. 






Do. 






Do. 






Do. 






Do. 


Allied to the Solenodon of the West Indies(?) _ 


?Solenodontidae — 


Montana. 


Fossorial, like the Cape golden moles 

Burrowing; analogous to marmots, prairie 

dogs, and the like. 
Squirrels and ground squirrels (spermophiles). 


? Chry sochloridae — 


Montana, Wyoming. 

Montana. 

Saskatchewan. 

Montana, Wyoming. 

Assiniboia. 

Montana. 


"Xenotherium" "^ unicum Douglass 

Nothocyon "lippincottianus" (Cope) 

Glires (Rodentia) : 
Ischyromyidae — 

Titanotheriomys veterior (Matthew) 

Titanotheriomys ' ' Ischyromys typus Leidy ' ' _ 




Do. 




Prosciurus ?saskatohewensis (Lambe) 


Saskatchewan. 




Castoridae — 


katchewan. 
Do. 


Found in Chadron clays; like pocket mice; 
Perognathus. 

Rabbits; remains found in the Chadron clays-. 


Heteromyidae — 

Adjidaumo(Gymnoptychus) minor Douglass- 
Adjidaumo (Gymnoptychus) minimus Mat- 
thew. 

Leporidae — 


Montana. 
Do. 

Do. 




Do. 






Saskatchewan. 


Grazing, upland rhinoceroses; cursorial; found 
in the Chadron clays. 


Perissodactyla: 

Hyracodontidae — 








Amphibious rhinoceroses; found in the channel 


Amynodontidae — 


South Dakota. 
Do. 


Small rhinoceroses of rather slender propor- 
tions, probably of browsing habit; remains 
found chiefly in clays. 


Rhinocerotidae — 


Do. 




Do. 


Leptaceratherium trigonodum Osborn 


Do. 

Assiniboia, South Dakota. 












Do. 




Caenopus cf. C. platycephalus Osborn 


South Dakota. 


Very small, slender-limbed horses, cursorial; 
grazers and browsers. Abundant in clays; 


Lophiodontidae — 

Colodon ( = Mesotapirus) occidentalis Leidy. 
Equidae — 


Dakota. 
South Dakota. 

Colorado, South Dakota. 


Mesohippus westoni Cope 


Saskatchewan. 






Do. 












Do. 




Mesohippus precocidens Lambe 


Saskatchewan. 



I Name preoccupied by Xenotherium Ameghino, 1904, a genus of edentates. 



120 



TITANOTHEEES OP ANCIENT WYOMING, DAKOTA, AND NEBRASKA 

Fish, reptile, and mammal Jauna contemporary with the titanotheres — Continued 



Common name or comparable form, habits or habitat, 
nature of deposits, etc. 


Classiflc name 


Region inhabited 


Suillines abundant and characteristic. Diffi- 
cult to place. Small, compact, dldactyl 
feet and fairly long limbs, cursorial. Ribs 
and abdomen small. Common in clays and 
sandstones. 


MAMMALIA — Continued 

Artiodactyla: 

Entelodontidae — 


South Dakota. 




Do. 








South Dakota. 








Dicotylidae ( = Tagassuidae) — 


South Dakota.- 




Leptochoeridae — 

Stibarus montanus Matthew -- 


Montana. 


Analogous to pigs. Occur chiefly in the clays-_ 


Anthracotheriidae — 


South Dakota. 




Do. 


Peccary-like, but of grazing habits. Rather 


Oreodontidae ( = Agriochoeridae) — 


Montana. 




Do. 


Browsing. Agriochoerus partly arboreal 
proportions like the larger cats. 




Do. 


Oreodon (= Merycoidodon) hybridus Leidy. 
Oreodon (= Merycoidodon) buUatus Leidy_- 

Oreodon (= Merycoidodon) affinis Leidy 

Oreodon ( = Merycoidodon) " culbertsonil 
Leidy." 

Agriochoerus maximus Douglass 

Agriochoerus minimus Douglass 


South Dakota. 

Do. 

Do. 
Saskatchewan. 

Montana. 

Do. 
South Dakota. 






Saskatchewan. 


Analogous to existing chevrotains of Africa — 


Hypertragulidae — 

Trigenicus socialis Douglass 

?Trigenicus mammifer Cope 


Montana. 
Saskatchewan. 
Do. 






Do. 






South Dakota. 






Do. 


Grazing, upland, cursorial, like the smaller 
antelopes of Africa and the guanacos of 
South America. 


?Heteromeryx transversus Cope 

" Anthracotherium pygmaeum" Lambe " 

Camelidae — 


Saskatchewan. 
Do. 


?Leptotragulus profectus Matthew 


Montana. 









« Based upon a part of a "right upper molar," which from Mr. Lambe's figure appears to be a left lower molar, probably of a hypertragulid comparable 



to Heieromeryx. 



NOTES ON THE HABITAT OF THE FAUNA OF THE CLAY AND 
SANDSTONE AS A WHOLE 

Matthew was the first to distinguish between the 
upland forms, found chiefly in the clays (flooded 
plains), and the lowland and aquatic forms, found in 
the sandstones (river channels). The following dis- 
criminations have been made: 

1 . Typical grazing group oj open plains. — Hyracodon, 
Oreodon, Mesohippus, Eotylopus, Poehrotherium. Note 
the cropping front teeth, associated with delicately 
cut and progressively long-crowned grinders, small, 
compact feet, and, except in Oreodon, long, slender 
limbs. Colodon may belong here. 

2. Browsing group of hush country and forest. — 
Titanotheres, Metamynodon, Caenopus, lEntelodon, 
1 Anthracotherium, lAncodon, ''.Agriochoerus. All large- 



sized fighting beasts, with coarse, heavy enamel on 
cheek teeth; front teeth adapted to lip browsing. 
Metamynodon may very likely have been amphibious ; 
the others probably were not. Entelodon is somewhat 
of an enigma; Sus is the nearest analogue but not a 
close one. 

3. Small hush or forest-dwelling browsers. — Hetero- 
meryx, Leptomeryx, Trigenicus. Analogous to the 
modern tragulines and probably of similar habits. 

4. Carnivora. — The hyaenodonts are analogues of 
the wolves. The ancestral canids are analogues of the 
mustelines and viverrines. True mustelines are 
scarce. Dinictis is the only cat. 

5. Rodentia. — Rabbits much like modern "cotton- 
tails" of the Great Plains. Heteromyids have ap- 
peared, but no true mice (Muridae) imtil the middle 



ENVIHONMENT OF THE TITANOTHEEES 



121 



Oligocene. Ischyromyids are abundant and include 
terrestrial (?) and arboreal (?) forms; whether fos- 
sorial forms existed or not is not proved. Eutypomys, 
though referred to the Castoridae, is not at all analo- 
gous to the modern beaver but rather to a large 
squirrel or spermophile. 

6. Insedivora. — The leptictids have rather sharp- 
cusped teeth and are intermediate in type between 
opossums and tree-living erinaceids. The moderate 
wear of the teeth is evidence against the theory that 
their food was worms or other terrigenous forms. 
There are no obvious arboreal adaptations in the 
limbs and feet; perhaps they may have been semi- 
arboreal. Their survival, unaltered as to cheek 
teeth, from the basal Eocene is suggestive of some 
special protection, such as spines. As for the zalamb- 
dodonts, they may have been fossorial, Xenotherium 
being molelike, but the evidence is insufficient. 

7. Marsupialia. — Rare. Precisely like small opos- 
sums in the structure of the teeth. 

8. Aves. — No birds have been recorded in this 
fauna, although they probably existed and may have 
been even numerous and varied. 

9. Reptilia. — Crocodiles and trionychids occur in 
the sandstone lenses; probably they were aquatic 
forms analogous to modern crocodiles and soft-shell 
turtles. In the clays Testudo occurs; also Xenochelys, 
probably similar in habits to modern land tortoises 
and marsh turtles. The lizards are apparently analo- 
gous to the Gila monster and to some of the swift- 
footed anguid lizards. Burrowing amphisbaenids 
occur in the Oreodon zone but have not yet been dis- 
covered in the lower Oligocene; no doubt they formed 
part of the fauna; also other lizards and many snakes. 

10. Batrachia. — No batrachians have been recorded, 
but there is no reason to suppose that they were 
absent or rare. 

11. Pisces. — A few fragments of fresh-water fishes, 
similar to those characteristic of muddy rivers of 
to-day, are recorded from the Swift Current beds in 
Canada. They will doubtless be found in the sand- 
stones and other stream deposits of the Titanotherium- 
bearing beds of the United States. 

SECTION 3. ADAPTIVE RADIATION, PRIMARY AND 
SECONDARY, THROUGH CHANGE OF ENVIRONMENT 
A CAUSE OF THE DIVERSIFICATION OF THE TITANO- 
THERES 

HABITAT OF THE UNGULATES 

The present geographic features of modern equato- 
rial Africa, consisting of a high central plateau, river 
borders, savannas, and forests, exhibit a close parallel 
to what we believe were those of the known titanothere 
region of North America in Eocene and lower Oligocene 
time. These conditions may also be compared with 
those found in the existing flood plains at the head- 
waters of the great rivers of South America east of 
the Andes in the warm temperate and subtropical 
but not in the tropical belt. 

101959— 29— VOL 1 10 



Adaptive radiation: Favorite habitats of existing 
perissodactyls and elephants 

[See fig. 78] 
RHINOCEROSES 

Rhinoceros sondaicus. Java. Typically a forest dweller, 
occasionally found in alluvial swamps. A browser. 

Rhinoceros (Dicerorhinus) sumatrensis. Hilly forest districts 
of Sumatra. A browser. 

Rhinoceros {Opsiceros) bicornis. Bush-covered country and 
open plains; forested foothills in the dry season. Fairly abun- 
dant on the top of the Aberdare, British East Africa (elevation 
9,000 feet). A browser, feeding on shrubs, roots, leaves, etc. 

Rhinoceros unicornis. Grassy jungles of India. A grazer. 

Rhinoceros (Ceratotherium) simus. Savannas and grassy 
plains, with swamps or water holes for wallowing. A grazer. 

TAPIRS AND ELEPHANTS 

Tapirus roulini. Pinchaque tapir of the high region of the 
Andes and Cordilleras. A browser. 

Elasmognalhus bairdi. A hill dweller, seeking lowlands 
during rainy seasons. A browser. 

Tapirus terrestris. A forest dweller. Lowlands of Brazil 
and Paraguay. A browser, feeding on palm leaves, fruits, 
water plants. 

Tapirus indicus. Lowlands and forests of India. A browser. 

Elephas (Loxodonta) africanus. Less typically a forest 
animal than E. indicus. Savannas, dry country, and forests. 
Ranges from the seacoast to points beyond the alpine heath 
zone of Mount Kenia and the bamboo belt of other African 
mountains. Ascends and descends steep places with wonderful 
facility. A browser and grazer. 

HORSES AND ZEBRAS 

Equus burchelli. Essentially a plains dweller; often found 
in sparse savannas. 

E. grevyi. Grevy's zebra. Low plateaus, thorn bush and 
feather grass country that has gravelly soil. Essentially a 
dweller in open plains and savannas. 

E. quagga (extinct). The quagga. A karoo dweller. Fre- 
quents open, arid plains. 

E. zebra. Mountain zebra. Hilly and mountainous country. 

E. przewalskii. Przewalski horse. Gobi Desert. A steppe 
dweller. 

ASSES 

Equus asinus. Abyssinian ass. Wiry hedge and upland 
country. 

E. hemionus kiang. The kiang. Desolate plains in the 
vicinity of lakes and rivers. High table-lands of Tibet (15,000 
feet) . Coarse wiry pasture and rough hard yellow grass. 

E. asinus somalicus. Striped African ass. Borders of the 
Nubian Desert. 

E. hemionus onager. Persian wild ass. Migrates from the 
plains to the hills in summer. The onager of Persia. 

POLYPHYLY AMONG HOOFED MAMMALS 
THE TITANOTHEEES AND OTHER EXTINCT FORMS 

It is astonishing to find within relatively small 
geographic areas both Eocene and Oligocene remains 
of many kinds of titanotheres, which lived close 
together under very similar climatic conditions, the 
more so because the known geographic distribution of 
the titanotheres in Eocene time is confined to the cen- 
tral Rocky Mountain region and extends only from the 
Wind River Basin of Wyoming on the north to the 
White River Basin of Utah on the south, a distance of 



122 



TITANOTHEEES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 






kl 

^ ^ 


■5 


1^ 

■<5 ^ 


~ci ^ 


\ 


ro 




.fo 


1 


~^ 


S 




^ 




f:; 


^ 



/2.O00 to /O.OOO 




6.000 fo 2,000 



/oo' 






^ ^ ^ 



« ^ 
I S 



>0 



Vi 



I 






I 






/-/yr3<zoc/o/?/s 



D. s'ama- 

frensis 
(Suma/ra) 



/?. i/n/cor/7/s 

O. 6 /cor/7 is 
C/Hfr^'ca) 

/Icerafheres 
(£i/rope, /is/ 3. M /l/rjer/c^) 



/?. SO/7c/a/CUS 



/l/77y/?oc/o///s 



Me/a/77y/?oc/o/7/s 



Z~. a/^r/C(9r7us- 



£'. 3/'/-/C3/?tJS 



fSat^ y/7n3/7S o/~ /^/r/ c^) 



^/ep//as //7£7/ 



Ps/ae o/r?3sfo - 



A^as? oa^o/7/s 



fFar ?s/s o/'/ne//'^) 



^//>p opo fa/77/ 



Z?C/^0/7^ 






N 



£ as/'nus 
{/Africa) 

£.przew3/sA/ 
(y^sia s/eppesj 

£. ze.&r^ 

fj^fr/ca p/3:^e.aus) 



NypoAippic/S— 



ttypc h/pp/c/s 



£. zebrs 



£. iurc/ie/// 
£.^revy/ 
£. ^cjsgga 



■5 



Tap/rus 

C/t/g/7 

(Soc/M 



/Indi'^s) 



A/7?e r/ca) 



Tap/rus 

ba/rc// 
^oL//h /imer/'ca) 



Tap/ras 

/nd/cus 



77 smer/canus 

(/.Oiv/anc/.r S. Amer/ca) 



Te/ma'fheres 
(H/^h p/3ins) 

Menodonfs— 



Symboroc/onfs 
(Hi'//y regions) 



Man/eoceraf/nes 

Bronfofheres 
^f/oocf p/aJ/7s) 

-Menodonts 
Bron fop/n es 



DoZ/chorh/nes 

Mefarhines 

Pa/aeosyop/nes 



Figure 78. — Geographic cross section showing the nature of the habitats of the larger existing ungulates 
and of the titanotheres as illustrating adaptive radiation 

The upper row shows the present geographic distribution ot the ungulates in continental Africa and the theoretic geographic features of the 
Rocky Mountain region in Eocene and Oligocene time— namely, high valleys, plateaus, foothills, plains, river valleys, flood plains, bot- 
tom lands, and river and lake borders. The second row shows the corresponding present distribution of the plant foods of different 
types of browsing and grazing, cursorial, graviportal, and semiaquatic quadrupeds. The four next lower rows show, in descending order, 
the corresponding adaptive radiation of the rhinoceroses, extinct and living; of the elephants and mastodons and the typical aquatic hippo- 
potami and sirenians; of the plateau, plains, and forest types of horses; of the mountain, foothill, and lowland types of tapirs. The bot- 
tom row shows the theoretic adaptive radiation of the principal types of titanotheres— telmathcres and menodonts of the higher levels; 
symborodonts in the foothills; manteoceratines, brontothercs, brontopines, and menodonts on the flood plains; dolichorhines, metarhines, 
and palaeosyopines on the lowlands and river borders. 



ENVIRONMENT OF THE TITAN OTHEEES 



123 



480 kilometers (298 miles). The continental extent 
of the distribution of the titanotheres, which is still 
unknown, was undoubtedly far greater, including, 
perhaps, the larger part of the North American 
continent and certainly extending into Asia. In 
Oligocene time the known geographic distribution 
was somewhat larger, including an area extending from 
Colorado to southern Alberta and measui'ing from 
north to south about 1,200 kilometers (746 miles). 
Titanotheres lived also in eastern Europe, both in 
Transylvania' and Rumelia, also in Mongolia. 

Our present Icnowledge of the geologic horizons of 
the titanotheres is still extremely meager regarding 
certain strata. The extent of our knowledge is sum- 
marized below. 



Geologic horizons of the known genera and subgenera of the 

titanotheres 
Lower Oligocene; upper, middle, and lower levels: Brontops, 

Diplocloniis, Alloys, Menodus, Brontotherium, Megacerops. 
Upper Eocene; Uinta C (true Uinta) : Telmatherium, Man- 

teoceras, Diplacodon, Prolitanoiherium, Eotilanolherium. 
Upper Eocene; Washakie B and Uinta B: Metarhinus, Rhadi- 

norhinus, Mesatirhinus, DoUchorhinus, Manteoceras, Tel- 
matherium, Diploceras. 
Middle Eocene; upper part of Bridger formation : Mesatirhinus, 

Manteoceras, Palaeosyops, Telmatherium. 
Middle Eocene; lower part of Bridger formation: Limno- 

hyops, Palaeosyops, Eometarhinus. 
Lower Eocene; Wind River formation: Lamhdotherium, Eoti- 

tanops. 

As compared with what we observe among the hoofed 
animals living to-day these titanotheres certainly 
dwelt near one another under very similar conditions 
of climate but in different feeding ranges and local 
habitats; they sought the same watercourses, and 
their remains were entombed in similar deposits. As 
the whole tendency of discovery up to the present 
time has been to multiply the phyla, to separate and 
diversify the titanotheres, the probability is that many 
other kinds of titanotheres lived in other parts of 
North America and Asia. 

The evolutionary principle underlying these diversi- 
ties Osborn (1902. 214, p. 353) has called adaptive 
radiation, which is the application to paleontology of 
the idea of divergence as conceived and developed 
successively by the studies of Lamarck, Darwin, 
Huxley, and Cope." Radiation is a broader principle 
than divergence, because it implies evolution in every 
direction possible to the organism. The idea of 
radiating branches from central forms assists the 
imagination, because the known radiations of extinct 
animals must be supplemented by the unknown radia- 
tions, and it is most remarkable how these missing 
radii have been discovered in group after group of 
animals. Such adaptive "radiation" is either "con- 
tinental" — that is, it occurs where diversities in food, 

" See also Osborn, H. F., 1902.214; 1905.267; 1910. 345; 1910. 346; Stevenson- 
Hamilton, J., 1912. 1; Sclater, P. L., 1894. 1; Lydekker, K., 1893. 1; Gregory, J. W.^ 
1896.1; Blanford, W. T., 1888.1; Kobelt, W., 1902.1; Schimper, A. F. W., 1903.1; 
Lonnberg, E,, 1912. 1; Roosevelt and Heller, 1914. 1. 



soil, or climate prevail over large areas — or "local" — 
that is, it occurs where marked diversities prevail in 
relatively small areas. The radiation among the 
titanotheres in southern Wyoming and northern Utah 
seems to have been largely "local," indicating that the 
physiography of the mountain basin was highly 
diversified. 

One of the results of adaptive radiation is poly- 
phyletic evolution, the existence within families of a 
large number of independent minor branches that 
may pursue more or less divergent evolution in local 
or continental regions but that may come together in 
river and flood-plain basins, so that their fossil re- 

LIMBS AND FEET 



Short-limbed, plantigrade,] AMBULATORY 
pentadactyl, ung:uicu-> or 

late stem ^ ^^JTORRESTRIAL 



CURSORIAL 
Digitigrade 




OMNIVOROUS 

(Grass 
Herb 
Shrub 
Fruit 
^. „.„„o|..„.. ™<" 

"~-^^[Carr' 

MYRMECOPHAGOUS 
Dentition reduced 

Stem INSECTIVOROUS 

Figure 79. — Original radiation of the unguligrade 
Herbivora, Carnivora, and Inseotivora, showing 
the adaptations of teeth, limbs, and feet to various 
habits and environments 

mains are found in the same localities and deposits. 
Polyphyletic evolution has been discovered so fre- 
quently, among both the mammals and the lower forms 
of life, that it may be considered the rule and mono- 
phyletic evolution along single lines the exception. 
Some of the examples of polyphyletic evolution among 
extinct mammals that ha^'e been determined in com- 
paratively recent years are the following: 

Contemporaneous 
branches, or phyla 
Oreodonts (Cope, Wortman, Peterson, Matthew, Doug- 
lass) 7-9 

Lophiodonts (Osborn, Deperet) 5-7 

Anthracotheres (Stehlin, Deperet, Andrews) 6-8 

Rhinoceroses (Osborn) 8-9 

Horses (Osborn, Gidley, Matthew) 8-9 

Titanotheres (Osborn) 10-12 

Elephants and mastodons (Osborn) 7-10 



124 



TITANOTHERES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



THE EXISTING AFRICAN ANTELOPES 

The polyphyly among the titanotheres and other 
extinct Perissodactyla presents a marked contrast to 
the impoverished conditions among tlie existing mem- 
bers of the same order wlien we consider that in all 
parts of Asia and Africa only five kinds of existing 
rhinoceroses can be distinguished by the characters of 
the skeleton and teeth alone, that only six or eight 
kinds of horses, asses, and zebras in the same great 
region can be distinguished by their hard parts, and 
that, similarly, among the tapirs of Asia and South 
America only three kinds can be distinguished. This 
contrast between present monophyly and former poly- 
phyly is due to the fact that the order Perissodactyla, 
though formerly a dominant group, is now a declining 
group. 

In the existing Bovidae, especially those in the groat 
continent of Africa, we have a parallel to the ancient 
polyphyly of the titanotheres and other Perissodactyla. 
The Bovidae is a family that includes the cattle and 
antelopes and that is now in the highest stage of ra- 
diation and adapted to a great variety of physiographic 
and biotic conditions, as shown in the primary and 
secondary adaptations in the seven subfamilies of the 
African antelopes. 

The African antelopes: Subfamilies, habits, and environment 



Subfamilies and habits 



Antilopinae (browsers and grazers) 

Gazelles 

Pallahs (impalas) 

Springbucks 

Gerenuks 

Saigas 

Bubalidinae (mostly grazers) : 

Gnus 

Hartebeests 

Blesboks 

Sassabies 

Tragelaphinae (browsers and graz- 
ers): 

Elands 

Koodoos 

Bush bucks 

Bongos 

Situtungas 

Hippotraginae (grazers) : 

Roan antelopes 

Sable antelopes 

Gemsboks 

Addaxes 

Neotraginae (browsers and graz- 
ers): 

KUpspringers 

Oribis 

Dik-diks 

Cephalophinae (mostly browsers) ; 
Duikers 



Environment 



Plains and deserts. 
Thorny bush and glades. 
High veldts. 
Deserts and bush. 
Steppes. 

Open plains. 

Open forests or plains. 

Open rolling country. 



Open forests and flats. 
Stony hills. 
Forests. 
Dense forests. 
Swamps and lagoons. 

Thin forests. 
Rolling uplands. 
Open deserts. 
Waterless deserts. 



Hills, mountains. 
Thin forests. 



Dense forests and bush. 



The African anteloyes: Subfamilies, habits, and environment — 
Continued 



Subfamilies and habits 


Environment 


Cervicaprinae (grazers on suc- 
culent plants near water) : 


Open forests and stony 

hills. 
Reed swamps, river bor- 
ders. 
Open swampy plains. 
Slopes of hills. 















An incipient or attempted adaptation to a grazing 
life is seen in the teeth of certain titanotheres. Most 
titanotheres are browsers. Broadly speaking, her- 
bivorous animals that live on open plains are grass 
eaters and tend to become gregarious in habit and 
cursorial in locomotion, whereas those that prefer the 
shady depths of the forests are browsers, are of soli- 
tary habit, and are mediportal in locomotion. There 
are exceptions, such as the black rhinoceros (Rhi- 
noceros {Opsiceros) Mcornis), which now frequents the 
treeless plains of East Africa but which is habitually 
a browser, although it is at times a grazer. The long- 
necked giraffes are fond of rather dry and fairly open 
country and are not found in strictly forested regions, 
yet they are wholly browsers, being especially fond of 
the leaves of certain thorny acacias, notably Acacia 
girajfa, and the related short-necked okapi, which is 
found only in the dense forests of the Congo, is a 
browser. 

The principles of adaptation shown in the skull and 
teeth of Perissodactyla to browsing and grazing habits 
are described in Chapters V and VI of this monograph. 
The adaptation of the limbs of the Perissodactyla to 
speed and weight are described in Chapter VII. 

In general, the competition and range for food 
among hoofed animals is accompanied by lengthen- 
ing of the limbs from medium-paced (mediportal) 
types to either swift-moving (cursorial) types or heavy- 
bodied (graviportal) types. Similarly, adaptation of 
the grinding teeth to browsing habits is seen in the 
short-crowned (brachyodont) types, and transition to 
the grazing habit is accompanied by lengthening 
(hypsodontism) of the crowns of the grinding teeth. 
Such changes are accompanied by changes in the pro- 
portions of the head to adapt the action of the teeth 
to browsing or to grazing. We observe a passage 
from short-headed (brachycephalic) to long-headed 
(dolichocephalic) forms of skull. In adaptive radia- 
tion every possible combination of lengthening and of 
shortening of skull, tooth, limb, and foot may arise, 
as well as notable coincidences of structure in different 
forms, for similar kinds of food may be found and 



ENVIRONMENT OP THE TITANOTHEEES 



125 



similar feeding habits may be acquired in widely 
separated habitats or greatly different environments. 
Contrasts in structure, such as those shown below, 
are equally notably. 

Contrast in structure between browsing and grazing types 



Browsing types (brachyodont) 


Grazing types (hypsodont) 


Short-headed (brachycephalic) 


Long-headed (dolichocepha- 




lic). 


Straight-headed (orthooephalic) _ _ 


Bent-headed (cyptocepha- 




hc). 


Short-limbed (braehymelic, bra- 


Long-limbed (dohchomelic, 


ch ypodal) . 


dolichopodal) . 



Grazing on the harder 
siliceous grasses of dry 
plains and uplands 

Gazelle 

Addax 



Grazing and browzing on 
the tender leaves and 
twigs of plains, thin for- 
est and brush country 
Sable and roan 




Grazing and browzing 
on the tender grasses 
of moister land and 
swampy plains 
"Puku 
Cob 
Reedbuck 



Leaf, bark, and twig 

eaters in forests 

Duiker 



Browsing on tender 

leaves and shrubs 

of partly forested 

countries 

Bushbuck 

Waterbuck 



Figure 



Browsing on succulent 

aquatic plants 

of swampy lands 

Sitatunga 

Lechwe 

). — Adaptions in the structure of the skull and teeth 
of Herbivora to diverse habits of feeding 



Double or even multiple adaptive radiation is 
continually in operation, first, in the structure of 
skull and tooth, which is dependent on the nature of 
the food, and, second, in the structure of foot and 
body, which is dependent on the nature of the soil. 
Thus may arise cursorial (long-limbed) grazers (long- 
toothed), graviportal (heavy-limbed) grazers (long- 
toothed), or cursorial (long-limbed) browsers. There 
is no fixed law of correlation of structure of skull and 
tooth such as was supposed by Cuvier. The law of 
correlation as restated by Osborn (1902.214) is as 
follows : 

Structure of feet (correlated chiefly with structure 
of limb and body) and structure of teeth (correlated 
chiefly with structure of skull and neck) diverge inde- 
pendently in adaptation respectively to obtaining 



food (by feet) and eating food (by teeth) in different 
environments. Each structural feature is evolved 
directly to perform its own mechanical functions or 
purposes, yet in such a manner that each is consonant 
with the other. 

CONTINENTAL ADAPTIVE RADIATION OF THE AFRICAN 
ANTELOPES 

The African antelopes are divided into seven sub- 
families, all mediportal to cursorial in limb structure 
but widely different in tooth and skull structure, as 
shown in the table on page 124. 

The 133 or more species (Sclater, 1894.1) embraced 
in these seven subfamilies seek food and protection 
from enemies on the varied surface of the African 
continent in habitats including no less than 17 differ- 
ent kinds of country. 

Each type of habitat has food peculiarly favorable 
to certain feeding habits to which the structure of the 
teeth and skull is speciflcally adapted. Each type of 



CURSORIAL 
Distance carrying 
Gazelle 

Gemsbuck (Oryx) 
Addax 

SALTATORIAL 
Leaping and springing 
Springbuck 
MEDIPORTAL ^ Klipspringer 

Medium-weight bearing 
Hartebeest 
Gnu 

Sable and 
antelopes 

SEMI-ARBOREAL 
Progressively on 
branches of trees 
Impala 

GRAVIPORTAL 
Heavy-weight bearing 
Greater kudu 
E'and \ \ FOSSORIAL 

Digging and uprooting 



AMPHIBIOUS 
Swamp and river-living 
Waterbuck 
Puku 
Cob 
AQUATIC Reedbuck 

Partly fluviatile, largely 
feeding and seeking 
safety in the water 
Lechwe 
Sitatunga 

Figure 81. — Convergent adaptations in the structure of the 
limbs and feet of ungulates 

Certain gazelles are independent of watercourses. The adaptive radiations 
indicated above occur independently within different subfamilies. 

habitat demands modiflcations of limb, foot, and hoof 
structure for movement in the search for food and 
escape from enemies. 

The theory of the evolution of the antelope is that 
in mid-Tertiary time a divergent primary radiation 




126 



TITANOTHEEES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



divided them into seven subfamilies, each with its 
distinctive mode of life. During a long period of 
geologic time the Bovidae have undergone secondary 
radiations (A-Q, fig. 82), by which certain branches of 
these subfamilies have become adaptively convergent 
toward certain branches of other subfamilies through 
the adoption of similar habits and habitats. Thus, 
analogous genera and species arise independently in 
each subfamily. For example, waterless deserts were 
sought both by the addax, among the Hippotraginae, 
and by the gazelle, among the Antilopinae; reeds, 
river borders, and lagoons were sought both by certain 




Figure 82.- 



c 

-Adaptive radiation in the feeding habits of antelopes, as observed by 
Stevenson-Hamilton in 1912 
1-7, Primary radiations; A-Q, secondary radiations 



Cervicaprinae, such as the lechwes and kobs, and by 
certain of the Tragelaphinae, such as the situtungas. 



ADAPTIVE RADIATION IN THE FEEDING HABITS 
ANTELOPES 



OF 



The habits and habitats of the antelopes, as noted 
by Stevenson-Hamilton (1912), are as follows: 

1. Antilopinae. — The impalas {Apyceros melampus) eliiig to 
neighborhoods of dense thorny bush, to which they fly for 
refuge. More partial to brovv^sing than to grazing. Food con- 
sists largely of leaves and shoots, but they eat young and tender 
grass freely after early rains. Staple diet leaves and fruit of 
certain acacias, also twisted bean pods of the same. In March 



fruit of marula is eaten. Toward the end of the dry season 
they completely strip the bush of everything edible up to the 
extreme height which they are able to reach. * * * The 
springbucks (Antidorcas euchore) are typical of the high veldt 
fauna of South Africa. The only member of the gazelle group 
in this region. Love high, open tablelands. * * * Xhe 
typical African races of gazelles include 14 species. Grant's, 
Thomson's, Speke's, etc. Inhabitants of wide, open plains or 
sandy deserts. Largely independent of water. 

2. Bubalidinae. — Antelopes of large size, large, moi.st rhinari- 
um; including Buhalis (= hartebeest), Damaliscus (= bastard 
hartebeest), Connochaetes (= gnu, or wildebeest). Buhalis 
(= hartebeest), eight species, with everywhere same charac- 
teristics; frequent open or forest countrj' or treeless plains; 
essentially grass eaters; like to drink 
regularly. Young carried about eight 
months. * * * Damaliscus, bonte- 
buck {D. pycargus), blesbuck, tsessebe 
(sassaby) (both D. albifrons), grass-eat- 
ing antelopes, favoring rather open and 
fairly flat country, never hills or thick 
jungle, partiality for shady patches of 
bush or forest for shelter during the hot 
hours. * * * Connochaetes (gnu, or 
wildebeest), white-tailed or black (C 
()7iu) and brindle or blue (C. tautrinus). 
Prefer open, rolling country interspersed 
with thick thorn or other bush. Some- 
times remain in the open, bare spaces 
or plains where they can see for long dis- 
tances. Essentially grass-eating ani- 
mals. Pasture cropped closely. Socia- 
l5le, gregarious. 

3. Tragelaphinae. — Elands and bush- 
bucks, inyalas, kudus, situtungas. Elands 
{Taurotragus oryx), plains type, graze 
with horses, donkeys, and cattle but 
browse by preference, favoring the grass 
only when fresh and green, sometimes 
cropping the tops of young river reeds. 
Gestation period eight and a half months. 

* * * Bongo {Boocercus eurycoros), 
fond of the most dense forest, leaves and 
twigs of a certain kind of undergrowth, 
which grows from 6 to 8 feet in height. 
Young shoots all nipped off if bongo 
have been feeding. Also (?) bark eaters. 

* * * Bushbucks {Tragelaphus scrip- 
txis), forest dwellers, solitarj', nocturnal, 
prefer densely wooded gullies, or kloofs, 
of South Africa. Browsers on the leaves 
of various small shrubs and trees; eat 
grass sparingly when the latter is fresh 
and green; roots and tubers form fur- 

* * * Inyalas {Tragelaphus angasi), 
Exhibit intense localization, probably due 

to the presence of some peculiar foodstuff, limited in quantity 
but necessary to the health of the individual animal. Probably 
browse on various leaves, shrubs, and fruits, bean pods and 
acacias, fruit of the marula; grass eaten when it is young and 
of good quality. * * * Sitiitnnga.s (Tragelaphus spekei) , semi- 
aquatic animals, almost amphibious by nature. Great elonga- 
tion of hoofs. Strong swimmers. Rapid locomotion upon dry 
land very difficult. Frequent extensive reed and papyrus swamps 
bordering lakes and large rivers. * * * Kudus {Strepsiceros 
strepsiceros) , love stony or rather broken ground, covered with 
thorn scrub. Gregarious, more than most antelopes, a browser, 
subsisting chiefly on the leaves of thorn acacias and bush shrubs, 
together with the fruits of the marula and other trees. 



ther articles of diet, 
very local and rare. 



ENVIRONMENT OF THE TITANOTHERES 



127 



4. Hippotraginae. — Sable and roan antelopes Hippotragus, 
oryx and addax, distinguished by the presence of horns in both 
sexes and small rhinarium or bare space on the muzzle. Sable 
(H. niger) to a great extent, though not entirely, a grass eater. 
Prefers thin forest country, interspersed with alternate thickets 
for shade, and open vleis for grazing. Regular drinker, seldom 
found more than a few hours from water. Gestation period 
about 270 days. * * * Roan antelope {H. equinus) favors 
rather upland, rolling country, not too thickly wooded, such 
as the middle veldt, but when persecuted takes readily to 
forest or the same environment as the sable antelope. A 
grass eater, and drinks regularly * * * [Genus Oryx.] The 
gemisbuck (0. gazella) of South Africa separated from its nearest 
generic relative (0. beisa) of German East Africa by an interval 
of 1,500 miles. Fairly numerous in Kalahari Desert, main- 
taining its security owing to its independence of water, able 
to quench thirst from moist tubers and roots. Generally found 
in small troops. The beisa (0. beisa) inhabits the Kilimanjaro 
district, British East Africa, Somaliland and the Sudan, east of 
the Nile. Sometimes found in herds of 50. Period of gestation 
eight and a half to ten months. White oryx (0. leucoryx) is 
found west of the Nile. Essentially a desert animal and like 
the gemsbuck apparenth' associates in small parties. [Genus 
Addax.] The addax [A. nasomacidatus\ distantly related to 
both oryxes and roan and sable antelopes, pale sandy color. 
An inhabitant of waterless sandy deserts of northern Africa. 

5. Neotraginae. — Klipspringers {Oreotragus orcotra.gus), like 
the chamois, prefer small shrubs and grasses growing among the 
stones. Live on natural moisture of the grass and nightly 
dews. * * * Oribi (Ourebia) frequents open grass country 
or plains not too thickly forested. Grass feeders, seldom found 
any distance from water. Eight species. 

6. Cephalophinae. — Lovers of dense bush and forest of central 
and southern Africa. Thirty-eight species. Duiker (Cephalo- 
phus grimmi), solitary animal, fond of bush country. Never 
far from covert. Mainly browsers. Nibbles leaves and young 
shoots of various acacias, small shrubs. Grass consumed 
when. young and fresh. Red duiker (C natalensis), dense 
forests and bush. Blue duiker (C monticola), essentially a 
browser, favors shelter or dense covert. 

7. Ceruicaprinae. — Animals of large or medium size. Water 
buck (Cobus ellipsiprymiius), open forest country, eastern 
Africa, favor banks of large rivers, prefer succulent herbage, 
but are partial to rough and broken country, stony hillsides, 
and vicinity of fairly thick bush; grass feeders. During dry 
season frequent banks of streams for succulent herbage. 

* * * Sing-sing water buck {Cobus defassa), habits similar 
to above. * * * The lech we (C. lechwe) is smaller than 
the water buck. Hoofs elongated and pointed. Frequent 
great reed swamps and river borders, northern Rhodesia. 
Next to the situtunga, the most aquatic of all antelopes, stand- 
ing knee or even belly deep in large shallow lagoons. Come 
ashore to graze, food consisting of grass and young reeds. 

* * * Gray's water buck (C maria), frequent river bottoms 
and reedy grass. Stand in shallow water. * * * Puku 
(C. vardoni), less aquatic than the lechwe, approaching in this 
respect the water buck — that is, found close to but not in the 
water. Frequent swampy plains. * * * Uganda cob (C. 
thomasi), fond of open, rather swampy plains, near rivers or 
permanent water. Grazes on young shoots of grass. * * * 
Common reedbuck {Cervicapra arundinum) , lowlands of Natal 
and Zululand, Transvaal bush country, etc. Favors grassy or 
reedy valleys near streams or permanent water of some kind. 
Occasionally met with in thin bush. Food consists entirely of 
grass. Do not take to water when alarmed. * * * Moun- 
tain reedbuck {Ceruicapra fulvorufula), lower slopes of hills 
covered with rocks and loose stones, mingled with scattered 
bush and long grass. Grass eaters, at night descending from 
hills to nearest w-ater. Affecting sides rather than tops of 



hills. * * * Bohor reedbuck {Cervicapra redunca),, favor- 
ing open vleis and bush or swamp land. Like the neighbor- 
hood of water. * * * Gray rhebuck {Pelea capreolus) , 
unlike mountain reedbuck, frequent flat tops of the table 
mountains; common in South Africa as well as higher levels of 
the ranges. Grass feeders, and descending at night to drink 
after the manner of the mountain reedbuck. 

[Note vertical physiographic distribution of the genus 
Cervicapra.] 

CAUSES OF VARIATION AND POLYPHYLY AMONG 
QUADRUPEDS 

Change of physical environment. — A series of meteoric 
and biotic changes — that is, changes of season, of 
climate, or of rainfall, the appearance of new enemies, 
the introduction of new plants or the crowding out of 
old ones — will cause a change of food supply, which 
will cause a change of habitat, which in turn will 
cause a change of browsing or grazing habits that 
will affect locomotion — the use of the limbs in the 
search for food — and modify the form of the hoofs, 
because of the change of soil. The browsing mountain 
moose (Alces) of eastern Idaho, for example, has a 
hoof of very different form from that of the water- 
living forest moose of Maine. Among the new 
enemies that may appear are certain insect pests, 
such as flies or ticks, which may drive quadrupeds 
away from feeding ranges that are otherwise favorable 
into regions, perhaps not far distant, where food is 
scarcer and the general conditions are more adverse, 
and where, perhaps, the young are exposed to new 
dangers. 

Such changes may bring about (1) a change of 
habit or (2) a change in habitat or environment, 
either of which, as a general law, culminates in (3) 
change of function, followed by (4) change of struc- 
ture. (5) A change of function or habit certainly 
brings about a new "incidence" of selection or new 
set of causes tending to survival or extinction. 

Change of appetite. — Variations in appetite are un- 
doubtedly among the chief causes of local divergence. 
Stevenson-Hamilton (1912.1, pp. 97-158) noted the 
fastidious choice of food by each of the principal 
species of African antelopes, and other wild animals are 
very fastidious and seek an astonishing variety of food 
in the course of a single season. The predilection for 
certain kinds of food is very strong, and departures 
from it lead to adaptive radiation. Similarly Sampson 
(1905.1) records that the white-tailed deer (Odocoileus 
virginianus) browse on many kinds of plants in the 
course of a year. 

Local polyphyly through reunion of phyla. — Animals 
that have diverged through migration or through 
geographic segregation or separation may later be 
brought together in one region. For example, the 
mule deer {Odocoileus hemionus) and the white-tailed 
deer (0. virginianus), which have probably evolved in 
different regions of the United States, are now found 



128 



TITANOTHEEES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



together in the same region in the West. In Miocene 
time the American rhinoceroses were joined in the 
western plains by certain European rhinoceroses. 
Thus continental radiations from great countries like 
Africa, Asia, or America may pour some of their 
branches into a single small region, mingling many 
distinct phyla. 

Hypsodont or grazing types may mingle with brach- 
yodont or browsing types in the same locality through 
their choice of grasses or of shrubs as their principal 
article of diet. Independently in the same region in 
southern Wyoming two of the branches of the titano- 
theres {Telmatherium and DolichorTiinus) began to 
acquire long-crowned teeth, while two others {Palaeo- 
syops and Limnohyops) retained persistently short- 
crowned teeth. 

HABITS OF THE EHINOCEROSES PARALLEL TO THOSE OF 
THE OLIGOCENE TITANOTHERES 

Mingling of hrowsing and grazing rJiinoceroses in 
Africa. — In equatorial Africa the Nile is an insuperable 
barrier between two species of rhinoceros, the "white 
rhinoceros," which is confined to the west bank, and 
the "black rhinoceros," which ranges along the east 
bank; yet these two species were formerly found 
together in the same regions of South Africa. The 
large grazing "white rhinoceros," R. (Ceraiotheriwn) 
simus, has hypsodont teeth and grazes in the open 
country, particularly in the wide, grassy valleys, 
though it was frequently met on the high veldt of 
Matabele and Mashonaland, feeding at night or in the 
cooler parts of the morning and evening. Its food 
consists entirely of grasses. Its sight is bad, but its 
scent and hearing are acute. On the other hand, the 
smaller browsing "black rhinoceros," R. (Opsiceros) 
hicornis, which has brachyodont teeth, was formerly 
common on the slopes of Table Mountain and on the 
Cape Flats and closely overlapped R. (Ceratotherium) 
simus in certain parts of its range; it frequented bush- 
covered country more than open grass lands and was 
often found in rocky, stony districts. It is partly 
nocturnal in its habits. Its food consists entirely of 
leaves, twigs, and sometimes of the roots of certain 
bushes and shrubs, but seldom of grass (Roosevelt and 
Heller, 1914.1). Its adaptations are essentially those of 
a browser, for it prefers the twigs and small roots of 
certain shrubs which it finds on the treeless plains of 
East Africa (Stevenson-Hamilton, 1912.1). It has a 
considerable vertical geographic range,'- being found 
also on the high plateau near the glaciers of Mount 
Kenya. (J. W. Gregory, 1896.1, p. 267.) 

Habits of Asiatic rhinoceroses. — The existing species 
of Asiatic rhinoceroses differ in habitat; they do not 
mingle. Rhinoceros unicornis or indicus, which has 
relatively hypsodont grinders, frequents the swampy, 
grassy jungles of the plains of India. The R. sondaicus 

" Gregory attributes this range to the white rhinoceros, but his observation 
actually refers to the black rhinoceros, as Heller has pointed out. 



of Burma and Java has shorter grinders. As observed 
by Blanford (Lydekker, 1893.1, vol. 2, sec. 4, p. 
470), it "is more an inhabitant of the forest than of 
the grass, and although it is found in the alluvial 
swamps of the sudarbans, its usual habitat appears 
to be in hilly countries. It has been observed at 
considerable elevations both in Burma and Java." 
Indeed there is much evidence that it probably ascends 
occasionally to as much as 7,000 feet above sea level. 
Its food consists largely of twigs and smaller branches. 
The third species of Asiatic rhinoceros, the Sumatran 
rhinoceros {R. (Dicerorhinus) sumatrensis) , which has 
relatively short-crowned teeth, inhabits hilly forest 
districts and has been observed in Tenasserim at an 
altitude of 4,000 feet above the sea. According to 
Lydekker, it is a good swimmer and is said to have 
been seen swimming in the sea in the Mergui Archi- 
pelago, possibly traveling in search of new feeding 
grounds or to avoid certain unfavorable conditions. 

Thus we find among the rhinoceroses three lines of 
adaptation to habitat and to food radiation — first, both 
hypsodont (grass-loving) and brachyodont (browsing) 
forms; second, a considerable geographic vertical range 
both in R. (Ceratotherium) simus and R. sondaicus; 
third, the occasional assumption of semiaquatic habits. 

All these conditions were partly paralleled among 
the Oligocene titanotheres, which, however, attained 
no extreme hypsodontism. 

HABITS OF THE EXISTING TAPIRS PARALLEL TO THOSE OF 
THE EOCENE TITANOTHERES 

The Eocene titanotheres, although inferior in the 
structure of their grinding teeth, were nearest in form 
and in body adaptations to the existing tapirs. In 
the Tapiridae we find these principles of adaptive 
radiation — great vertical geographic range, including 
choice between upland and lowland habitat, and 
assumption of more or less aquatic life. The teeth 
are short-crowned (brachyodont), are crested (lopho- 
dont), and are superior in mechanism to the cone and 
crescent (bunoselenodont) grinders of the titano- 
theres. These principles are observed as follows: 

1. According to J. E. Gray (1872.1, p. 486) Tapirus 
pincJiaque ascends to very great heights in the Andes. 
M. Goudot "obtained a young female tapir at an 
elevation of about 1,400 meters — nearly up to the 
snow level on the Peak of Tolima in New Granada — 
about 1843." According to Gray (1872.1, pp. 487, 
488) Tschudi, in the "Fauna peruana" (p. 213), says, 
" This species of tapir [T. roulini] is found in Peru on 
the eastern slope of the Cordilleras at an elevation of 
7,000 or 8,000 feet, which is above the snow line. " 

2. On the other hand, the tapirs (T. hairdi) from 
Mexico and the Isthmus of Panama, which have 
been referred to the genus Elasmognathus by Gill, are 
more generally confined to the lower hills or occupy 
an intermediate habitat. Captain Dow observes 
(1867.1, p. 214): 



ENVIRONMENT OF THE TITANOTHEKES 



129 



Thus far all examples of T. bairdi [Elasmognathus] have been 
found exclusively on the Atlantic side of the Isthmus [of 
Panama], and north of the Chagres River. Their favorite 
haunts appear to be in the hills lying at the back of Sion Hill 
and the adjoining stations of the Panama Railway. It is only 
during the rainj' season that they seem to seek the lowlands, 
for it is only in that season they are captured. 

Similarly Tapirus {Elasmognathus) dowi was found 
in the highlands of Guatemala, Nicaragua, and Costa 
Rica. 

3. The opposite extreme from mountain-living 
habits is furnished by the typical South American 
tapir (T. terrestris), which inhabits the forest districts 
of Brazil, Paraguay, and the northern part of Argen- 
tina. This species is fond of gamboling in the water 
and rolling in soft mud and swims and dives like a 
capybara; it is not improbable that it may also walk 
along the beds of shallow rivers and lakes, as was 
observed to be the habit of a specimen of the Malayan 
tapir (Tapirus indicus). In Brazil, in districts remote 
from cultivation, the food of the tapir is composed 
largely of palm leaves, but at certain seasons of the 
year these animals subsist almost exclusively on 
fallen fruits, and in some districts swampy grasses 
and water plants form their chief food. 

VERTICAL GEOGRAPHIC RANGE OF QUADRUPEDS 

The rhinoceroses as a group have a wide vertical 
geographic distribution, ranging from sea level to the 
snow belt. The black rhinoceros, although it prefers 
the lower grassy plains, is found also on the high 
plateaus near the glaciers of Mount Kenya. (Gregory, 
1898, op. cit., p. 263. '') As above noted, the tapirs 
as a group range from sea level to the snow belt, 
8,000 feet above sea level. Some species are exclu- 
sively low-level forms {T. terrestris); others range 
from sea level well up into the mountains {T. iairdi); 
still others inhabit the higher Andes {T. pinchaque). 
The elephants also enjoy a wide vertical range; 
Elephas (Loxodonta) africanus is said to ascend and 
descend steep places with wonderful facility, and 
t footprints of the modern Asiatic elephants have been 
seen among the eternal snows of the highest mountains 
(Pohlig, 1891.1, p. 328). 

VERTICAL GEOGRAPHIC RANGE OF THE TITANOTHERES 

Thus, judging by analogy with the other Perisso- 
dactyla and from what we know to be true also of the 
horses, it is probable that the titanotheres enjoyed a 
considerable vertical geographic range in the Rocky 
Mountain region in Eocene time and that this may 
have entered into the causation of their local adaptive 
radiation. 

TEN CHIEF HABITAT ZONES OF MAMMALS 

Wide climatic and physiographic differences, if 
concentrated in a geographically restricted area, 
facilitate local adaptive radiation. For example, 

" Gregory inadvertently attributes this range to the white rhinoceros. 



grassy meadows favorable to shrubs bring grazers and 
browsers together. That much more extreme con- 
trasts are by no means unnatural is shown along the 
coasts of Mexico, where there is an abrupt transition 
from an extremely moist, warm lowland to a dry, . 
cool upland. Similarly abrupt transitions are ob- 
served in parts of the Andes and the Himalayas. 

It is consequently not difficult to account for the 
fact that seven or eight different phyla of titanotheres 
lived together in southern Wyoming and northern 
Utah in middle and upper Eocene time, for the entire 
region was varied and rnountainous. 

The life zones of mammals have been set forth 
admirably by Kobelt (1902.1) and should be studied in 
connection with the vegetation zones of Schimper 
(1903.1). Some mammals are strictly confined to 
their typical habitat zones — that is, they are intensely 
localized. Certain antelopes, such as Tragelaphus 
angasi, the inyala (Stevenson-Hamilton, 1912.1, p. 
135), probably feed upon only a single plant and are 
limited in range to its distribution. Many Herbivora, 
such as elephants, rhinoceroses, and horses, are very 
plastic and have great diversity of habitat in the 
course of the change in seasons and under varying 
conditions of competition. 

Life zones are defined by land and water, by mois- 
ture and aridity, by depression and elevation, by low 
and high temperature, by the distribution of insects, 
and especially by the presence of vegetation adapted 
to grazing or browsing. Life zones are therefore de- 
fined sharply in some places and feebly in others. 
The ten zones discriminated are described below. 

1 . Mountain or alpine liaiitat. — High mountains and 
mountain ranges with the snow and timber lines at 
altitudes of 6,000 to 12,000 feet or more. Thinly 
forested or tundra-like lands, adapted both to grazing 
and browsing ungulates having relatively short limbs 
and feet adapted to climbing. The Artiodactyla are 
represented by many forms, some of which range far 
above timber line, including goats (Capra), rupi- 
caprines {Rupicapra, NemorJiaedus, Oreamnos), moun- 
tain sheep (Ovis), vicunas {Lama vicunna) at certain 
seasons, Pudu deer {Pudua). The Perissodactyla that 
invade these high forest zones are only certain tapirs 
of the Andes {Tapirus pinchaque and T. roulini). 

2. Mountain forest habitat. — Lower mountain ranges 
and foothills, dry or well watered, well wooded, with 
river valleys. This zone includes the dry tropical 
woodlands (such as those of India), which are favorable 
to the larger ungulates; also the tropical rain forests 
(Asia, Africa, North America), generally unfavorable 
to large ungulates. In Asia the especial habitat of 
many deer, bovines, antelope, browsing perissodactyls, 
such as Rhinoceros sondaicus of Java, typically a forest 
dweller, R. {Dicerorhinus) sumatrensis of Sumatra. 
In the northern latitudes of North America, the typical 
home of the deer {Odocoileus), moose {Alces), wapiti 



130 



TITANOTHERES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



(Cervus), mountain caribou (Rangij'er) , at elevations 
of 2,000 to 8,000 feet. On these levels in South 
America are found among the Tapiridae T. {Elasmo- 
gnathus) bairdi, a hill dweller seeking the lowlands 



during the rainy season ; also T. (ElasmognatTius) dowi. 
In the equatorial belt of Africa both the high forests 
and lower forested foothills favorable to the growth 
of shrubs and trees attract also the elephants. 



Vertical distribution qf lije zones oj ungulates 













Alluvial bottom lands. 




Family or phylum 


Peaks and high- 
est mountain 
ranges; 6,000 
to 12,000 feet. 
Browsers 


High desert and drier 
uplands and plains, 
table-lands, plateaus, 
mesas; 5,000 to 15,000 
feet. Grazers 


Lower mountain ranges, 
foothills, well wooded 
and watered; forest 
lands; tributary river 
valleys; 2,000 to 8,000 
feet. Browsers 


Great plains and larger 
river valleys, broad 
grassy meadows, rolling 
country; sea level to 
6,000 feet. Grazers 


delta and flood-plain 
deposits, swamps and 
jungles, forests and 
partly forested low- 
lands; river or sea 
level. Browsers and 
grazers 


Rivers and lakes, 
river or lake level. 


Euminants 


Mountain sheep, 
goat, deer, and 
elk (summer). 


Pronghorn antelope 


Deer, moose, elk, cari- 
bou (winter). 


Buffalo and wapiti, or elk. 






Rbinocerotidae (re- 




Hyracodon nebrascensis. 


Rhinoceros sondaicus. 


Rhinoceros unicornis- in- 


Rhinoceros sondaicus. 


Metamynodon 


cent and extinct 




Three long toes. 


Java; typically a forest 


habits grass jungles. 


Occasionally seen in 


A mph ibious 


types). 




Rhinoceros (Geratothe- 


dweller. 


R. (Ccratotherlum) simus. 


alluvial swamps. 


rhinoceros. 






rium) simus. Meadows 


R. (Dicerorhinus) su- 


Large two-horned rhi- 


Aceratherium. Four- 








and sparse forests. 


matrensis. Inhabit- 
ant of hilly forest dis- 
tricts. 
R. (Opsiceros) bicornls. 
Two-horned " black " 
rhinoceros of Africa. 
Often seen on slopes 
of table mountains; 
feeds on roots, leaves, 
etc. 


noceros of Africa: inhab- 
itant of grassy valleys 
on high veldt. 

R. (Opsiceros) bicornls. 
Found on Cape flats, in 
bush-covered country. 

Coenopus.and Dlcerathe- 
rium. Three-toed ani- 
mals. 


toed rhinoceros. 




Equidae (horses, 
zebras, asses). 




Equus Demionus, E. ki- 


Equus zebra, mountain 


Equus hemionus onager. 








ang. Kianganddzigge- 


zebra. 


Migrates to the hills in 










tai. Inhabit table- 




summer. 










lands of Tibet, 16,000 feet 




E. asinus somallcus. In- 










high. Prefer desolate 




habits Nubian desert. 










places near lakes and 




E. quagga. The quagga 










rivers, and coarse wiry 




of South Africa; extinct. 










pasture of rough, hard 




A karroo dweller. 










yellow grass. 




E. burchelll. Burchell's 










E. onager. The onager of 




zebra; found north of 










Persia. 




Orange River; often seen 










E. hemippus. Syrian 




in sparse forests, but 










wild ass. 




predominantly a plains 










E. asinus. Feeds on wiry 




dweller. 










desert grasses. 




E. grevyi. Low plateaus 










E. zebra. Feeds on plains 




with gravelly soil. Seen 










grasses. 




in thick thorn bush and 










E. przewalskil. Inhabits 




tall feathery grass. Es- 










northern deserts. 




sentially an inhabitant 
of the open plains. 






Tapiridae (tapirs).. 


Tapirus roulini 


Tapirus roulini.. 


Tapirus bairdi. Hill 




Tapirus americanus. 




T. pinchaque. 


T. pinchaque. Tapir of 


dweller, seeking low- 




Common tapir of for- 






Inhabits -slopes 


the high regions of the 


lands at rainy season. 




ests and lowlands of 






of Cordilleras. 


Andes. 


T. dowi. 




Brazil and Paraguay. 
A forest dweller, feed- 
ing on palm leaves, 
fruits, and water 
plants. 
T. indicus. 




Proboscidea (ele- 






Elephas africanus. As- 


Elephas africanus. Less 






phants). 






cends and descends 
steep places with won- 
derful facility. 


typically a forest animal 
than E. indicus; found 
in comparatively open 
country; also in forests. 
E. indicus. Typically a 
forest animal. 






Sirenians, chalico- 










Macrotherium. 
Moropus. 


Manatee. 


theres, hippopo- 










Dugong. 


tami. 










Chalicotherium. 


Hippopotamus. 



ENVIRONMENT OF THE TITANOTHEEES 



131 



3. Boreal forest Tiaiitat. — Characteristic of north 
temperate zones with cold winters. The "temperate 
rain forests" of Schimper, partly interspersed with 
meadowlands. This zone includes the whole of primi- 
tive northern Europe and North America south of the 
tundra zone. In Asia it includes the whole of Siberia, 
grading on the south into the high "steppe" and high 
"plateau" regions and on the north into the Arctic 
tundras or barren grounds. It is the great boreal 
zone of North America, favored both by woodlands 
and meadows and by sufficient rainfall. The ungulates 
are very numerous, especially genera of Bovidae, 
Cervidae, and Suidae. 

4. Tundras and barren ground habitat. — In this low- 
lying, north circumpolar region trees are scarce or 
absent, except the willows and birches of the river 
bottoms, and the subsoil is frozen throughout the year. 
The ungulates are now represented only by the musk 
ox {Ovibos moschatus) and several species of reindeer 
(Rangifer); formerly by the mammoth and the horse 
in Alaska and Siberia during the period of greater 
forestation. 

5. Higher plains and plateaus. — Mesas, table-lands 
(as in Tibet and the Himalayas), and the desert 
plateaus of the Rocky Mountains and Andes, altitude 
3,000 to 6,000 feet or more; vegetation scattered, 
sparsely forested, both grasses and shrubs abundant; 
or rocky and open country with occasional forests. 
Climate generally severe in winter. This zone grades 
into the "high steppes" of Asia, the veldt of South 
Africa, the high plains of North America. It is mostly 
open country adapted to grazers with hypsodont teeth, 
long limbs, and slender feet, or to the cursorial and 
gregarious Herbivora. 

6. High steppe and desert habitat. — Treeless and arid 
wastes, steppes, and deserts of central Asia (such as 
the Desert of Gobi) or of Persia and Asia Minor, 
reaching an altitude of 6,000 feet, usually not so rich 
in flora and fauna as the high plateau. Climate 
extremely severe in winter. Inhabited chiefly by 
grazers. In Asia, among the Equidae we find the 
kiang {Eguus Jciang) of Tibet, the dziggetai {E. 
hemionus) of Mongolia, the wild horse {E. przewalslcii) 
of the Desert of Gobi or the Kobdo district of western 
Mongolia. The kiang of Tibet and Turkestan prefers 
desert places near lakes and rivers, seeking coarse, wiry 
pasture and rough, hard grasses. The dziggetai ranges 
from the lowland steppes of Turkestan to the high 
plateaus (1,680 meters) of the deserts of Mongolia. 
In this zone among the Artiodactyla we find the wild 
Bactrian camel {Camelus bactrianus), the saiga ante- 
lope {Saiga tartarica), and the Persian gazelle {Gazella 
gutturosa). 



7. Low desert habitats. — Steppes and sandy deserts 
of northern Africa, Syria, Arabia, Mesopotamia, and 
the northern borders of the Arabian Sea; rocky 
countries covered with sparsely vegetated areas and 
thin forests, scattered shrubs, and thorny bushes. 
Except in temperature and altitude this zone is like 
that of the high steppes; its vegetation is sought 
mostly by cursorial browsers and grazers with colora- 
tion of the desert; in Africa Gazella dorcas, Addax, 
Oryx leucoryx, and among the Equidae the north 
African wild ass {Equus asin'us), the Somaliland ass 
{Equus somaliensis) , the Assyrian E. hemippus, and 
the onager {E. onager), which grazes in the low deserts 
of Kutch and Rajputana. Neither the rhinoceroses 
nor the tapirs have ever had representatives in these 
low-lying desert belts. 

8. Plains habitat. — Great plains and larger river 
valleys; broad, grassy meadows bordering glades 
partly forested or not forested at all, extending from 
sea level to an altitude of 6,000 feet in northern 
latitudes. The tropical grasslands or savannas of 
Africa, the llanos of the Orinoco, the campos of 
Brazil, the semiarid karoos and veldts of South Africa 
are partly included in this zone, although they also 
approach the high steppe habitat. This zone is 
generally adapted to grazing, hypsodont types, mostly 
long-headed and cursorial. It is the natural habitat 
on the Great Plains of North America of the buffalo 
(Bison bison), of the pronghorn antelope (Antilocapra 
americana), and formerly of the wapiti (Cervus 
canadensis). Similarly on the plains of equatorial 
Africa are found numerous species of antelope (mostly 
grazers), oxen (grazers), giraffes (tfue browsers), the 
black rhinoceros, R. (Opsiceros) bicornis (browsers and 
grazers), and all species of zebra. The ungulates in 
this open country are either cursorial or graviportal 
and are well defended by horns. The Tapiridae 
have never been adapted to a country of this kind. 
The giraffes frequent the savanna and the thorn- 
forested country (xerophilous woodland of Schimper). 

9. Lower river valleys habitat. — Alluvial bottom 
lands, delta and flood-plain deposits, swamps and 
jungles, forested or grassy lowlands near rivers or sea 
level, typically the home of browsers rather than 
grazers, with feet and limbs adapted to soft soil, 

j limbs both of mediportal and graviportal type, with 
some cursorial types (such as situtungas) having 
spreading feet. The Artiodactyla include many 
bovines, some antelopes (such as situtungas), chevro- 
tains, suillines, the Liberian hippopotamus {Choer- 
opsis liberiensis) , and the primitive traguline (Dorca- 
therium) of West Africa. Among the Asiatic rhino- 
ceroses R. sondaicus, a browsing, brachyodont type, 



132 



TITANOTHEEES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



is occasionally seen in these alluvial bottoms. Simi- 
larly, the Sumatran rhinoceros, R. (DicerorJiinus) 
sumatrensis, also brachyodont, occasionally frequents 
such a region. Tapirus indicus inhabits this low 
forest belt in India, and T. terrestris is the common 
tapir of the forests and lowlands of Brazil and Para- 
guay. Among extinct forms the chalicotheres 
{Moropus, etc.) are found here. 

10. Aquatic, river and lalce iorder habitat. — Rivers, 
bayous, and lakes, frequented especially by aquatic 
browsing types with limbs adapted to swimming or 
to aquatic life and the teeth adapted to the softer 
kinds of food. Here we find the ungulates represented 
by their partly degenerate and specialized offshoots the 
sirenians, the Artiodactyla by the common hippopot- 
amus or the water buffalo of the Philippines. Either 
the lower river valleys or the rivers themselves were 
undoubtedly the habitat of the extinct rhinoceroses 
known as amynodonts; also, possibly, of the Miocene 
Teleoceras, a short-limbed river-frequenting animal. 
Among the titanotheres there are evidences of aquatic 
adaptation in species of the genus Mesatirhinus. 

CONCLUSIONS AS TO HABITATS OF THE TITANOTHERES 

We have no evidence that titanotheres formerly 
inhabited alpine, high steppe, or low desert regions. 
Neither the teeth nor the feet predispose us to specu- 
late upon such a habitat, nor have we any geologic evi- 
dence of it. There remain to be considered the "moun- 
tain," the "forest," the "boreal," or north temperate, 
the "plains," the "river valleys," the "rivers and la- 
goons." 

The earliest known types of titanotheres, which are 
subcursorial in limb structure, developed in a partly 
open and partly forested country, frequenting mead- 
ows, lower river valleys, and plains that were flooded 
during certain seasons of the year. There is reason 
to believe that one of their upper Eocene radiations 
(MetarJiinus) became amphibious or even aquatic. 
Some authors (Riggs, 1912.1, p. 36) believe that 
DolichorMnus , as well as the short-footed Palaeosyops, 
was semiaquatic. The habits of these animals are 
more 'fully considered in Chapter V. 

In Oligocene time the titanotheres entered the 
savanna-like Great Plains region of western North 
America, which was in part open country, in part 
country traversed by undulating rivers and by river 
bottoms bordered with forests. 

In dentition the titanotheres, both in Eocene and 
Oligocene phyla, are chiefly a browsing family, though 
they show incipient indications of adaptation to the 
grazing habit. 

SECTION 4. BIBLIOGRAPHY FOR CHAPTER II 

Ball, John. 

1887.1. Notes of a naturalist in South America, .xiii, 416 
pp., 1 map, London, 1887. 



Bauer, Clyde Max. 

1916.1. Contributions to the geology and paleontology 
of San Juan County, N. Mex. — 1, Stratig- 
raphy of a part of the Chaco River valley: 
U. S. Gaol. Survey Prof. Paper 98, pp. 271- 
278, pis. 6^71, Nov. 24, 1916. 

Maps western part of Puerco-Torrejon area and gives short 
suraraary of the two formations. 

Berby, Edward Wilber. 

1914.1. The Upper Cretaceous and Eocene floras of South 
Carolina and Georgia: U. S. Geol. Survey 
Prof. Paper 84, 200 pp., 29 pis., 12 figs., 1914. 

Blanford, W. T. 

1888.1. The fauna of British India, including Ceylon and 
Burma — Mammalia, 1888-1891. 

BouTWELL, John Mason. 

1907.1. Stratigraphy and structure of the Park City min- 
ing district, Utah: Jour. Geology, vol. 15, pp. 
434-458, 1907. 
Brown, Barnum. 

1914.1. Cretaceous-Eocene correlation in New Mexico, 
Wyoming, Montana, Alberta: Geol. Soc. 
America Bull., vol. 25, pp. 355-380, Sept. 15, 
1914. 

"Through this eastern exposure I have often found it im- 
possible to establish any definite line of demarcation between 
the two beds (Fox Hills and Lance)" (p. 3.5S). 

"The insensible gradation from marine through brackish- 
water into fresh-water sandstones is not confined to the 
eastern exposures of the 'Lance' on Hell Creek. The same 
transition is found on the border of the Lance formation on 
Alkali Creek, Sevenmile Creek, and Robber's Koost, all 
tributaries of the Cheyenne River in Weston County, 
Wyo." 

Calvert, William R. 

1910.1. See Stone, Ralph Walter, 1910.1. 

Clark, William Bullock. 

1891.1. Correlation papers — Eocene: U. S. Geol. Survey 
Bull. 83, 173 pp., 1891. 

Resume of work of various writers. Deposits of Bridger 
and Washakie Basins considered as one formation. Regards 
the Puerco as probably of Eocene age (p. 138). Eocene of 
the Atlantic coast. Gulf States, Pacific coast; historical sketch 
of the Eocene of the interior. Table showing relative posi- 
tion of interior Eocene deposits. Extensive bibliography. 

1896.1. The Eocene deposits of the middle Atlantic slope 
in Delaware, Maryland, and Virginia: U. S. 
Geol. Survey Bull.' 141, 167, pp., 40 pis., 1896. 

COCKERELL, THEODORE DrU AlISON. 

1906.1. The fossil fauna and flora of the Florissant 
(Colorado) shales: Colorado Univ. Studies, 
vol. 3, pp. 157-176, 5 figs., June, 1906. 
Birds, fishes, insects, mollusks, plants. 
CoMSTOCK, Theodore Bryant. 

1873.1. On the geology of western Wyoming: Am. Jour. 
Sci., 3d ser., vol. 6, pp. 426-432, 1873. 
Bridger classed as upper Miocene. 

Cope, Edward Drinker. 

1872.3. On Bathmodon, an extinct genus of ungulates: 

Am. Philos. Soc. Proc, vol. 12, pp. 417-420, 
1872. 

Describes the first mammal from this horizon (.Bathmoion) . 

1872.4. On a new genus of Pleurodira from the Eocene of 

Wyoming: Am. Philos. Soc. Proc, vol. 12, 
pp. 472-477, 1872. 

Gives a detailed account of the fossil-bearing beds along 
Bear River, near Evanston, Wyo. 



ENVIRONMENT OF THE TITANOTHEEES 



133 



Cope, Edward Drinker — Continued. 

1873.4. The monster of Mammoth Buttes: Penn Monthly, 
vol. 4, pp. 521-534, 1 pi., August, 1873. 

A popular account of the finding of tlie sltull of Eobasileus 
cornutus. 

1875.1. Report on the geology of that part of north- 
western New Mexico examined during the 
field season of 1874: U. S. Geog. Surveys 
W. 100th Mer. Ann. Rept. for 1875, pp. 61-97, 
pis. 2-6, 18 figs., 1875. 

Tlie original description of " Puerco marls. " Type locality, 
head of Puerco River. Gives section (p. 96) showing relation 
of Puerco and other beds in that vicinity. No mammalian 
fossils, but the marls are referred to the Eocene for stratigraphic 
reasons. 

1877.1. Report upon the extinct Vertebrata obtained in 
New Mexico by parties of the expedition of 
1874: U. S. Geog. Surveys W. 100th Mer. 
Rept., vol. 4, pt. 2, 370 pp., pis. 22-73, 1877. 

Extensive account of geology of the Wasatch beds and their 
fauna. Quotes former article (pp. 17, 18), but says the beds 
may represent Fort Union or the lignites of upper Missouri. 
The thickness of the Puerco is given as 500 feet. 

1879.1. The relations of the horizons of extinct Vertebrata 

of Europe and North America: U. S. Geol. 
and Geog. Survey Terr. Bull., vol. 5, pp. 33-54, 
1879. 

Correlation of Mesozoic and Cenozoic horizons of Europe 
and North America. 

1879.2. Second contribution to a knowledge of the Miocene 

fauna of Oregon: Am. Philos. Soc. Proc, vol. 
18, pp. 370-376, Dec. 30, 1879. 

John Day formation, Oligoeene. 

1880.1. The badlands of Wind River and their fauna: 

Am. Naturahst, vol. 14, pp. 745-748, October, 
1880. 
Eocene. 

1880.2. Observations on the faunae of the Miocene Ter- 

tiaries of Oregon: U. S. Geol. and Geog. 
Survey Terr. Bull., vol. 5, pp. 55-69, 1880. 
See also Paleont. Bull. No. 30, Dec. 3, 1878, 
and Am. Philos. Soc. Proc, vol. 18, pp. 63-78, 
Dec. 30, 1878. 
John Day formation, Oligoeene. 

1881.1. Mammalia of the lower Eocene beds : Am. Natural- 
ist, vol. 15, pp. 337-338, April, 1881. 

The first mammals are described, but they were not l^nown 
definitely at that time to be from the Puerco formation. 

1885.1. The Vertebrata of the Tertiary formations of 

the West: U. S. Geol. Survey Terr. Rept., 
vol. 3, XXXV, 1009 pp., 134 pis. (pis. l-75a), 
38 figs., 1885. 

Contains a general r63Um6 of the Wasatch. The deposits of 
the Bridger and W'ashaliie Basins and small area on White 
River in the Uinta Basin considered contemporary. Table 
of formations in this worli places Puerco as "post-Cretaceous," 
but in the text the author places it definitely in the Eocene. 

1885.2. The relations of the Puerco and Laramie deposits: 

Am. Naturalist, vol. 19, pp. 985-986, October, 
1885. 

states that the thickness of the beds near the type locality is 
850 feet. The author points out the distinctions from Laramie 
but considers the possibility of "post-Cretaceous" age. 

1885.3. The White River beds of Swift Current River, 

Northwest Territory: Am. Naturalist, vol. 19, 
p. 163, February, 1885. 
Oligoeene, White River. 



Cope, Edward Drinker — Continued. 

1886.1. The Vertebrata of the Swift Current Creek region 

of the Cypress Hills: Canada Geol. and Nat. 

Hist. Survey Ann. Rept., new ser., vol. 1, for 

1885, appendix I to article C, pp. 79-85, 1886. 

Oligoeene. 

Cdlbertson, Thaddeus a. 

1851.1. Journal of an expedition to the Mauvaises Terres 
and the upper Missouri in 1850: Smithsonian 
Inst. Fifth Ann. Rept., appendix 4, pp. 84-145, 
1851. 

Bear River [=Bear Creek] (p. 9.3), a southern tributary of 
the Cheyenne. First collection [in the Oreodon zone, Brule 
clays] (p. 94), rhinoceros skull (,A. oiddenlah) , several good 
heads, excellent teeth and jawbones, etc. Report to Baird 
(p. 105). 

Dall, William Healey. 

1892.2 (and Harris, G. D.). Correlation papers — The 
Neocene of North America: U. S. Geol. Survey 
Bull. 84, 349 pp., 3 pis., 43 figs., 1892. 

See especially chapter 6, on the supposed Neocene of the in- 
terior region, considered by States (pp. 280-317); table showing 
the vertical range of the Neocene of the interior (p. 279); 
map (p. 178); list of names applied to the Cenozoic beds and 
formations of the United States (p. 320). 

1898.1. A table of the North American Tertiary horizons 

correlated with one another and with those of 

western Europe, with annotations: U. S. 

Geol. Survey Eighteenth Ann. Rept., pt. 2, 

- pp. 327-348, 1898. 

Marine Tertiary horizons of the Atlantic coast'and the Gulf 
States correlated with one another, with those of the western 
United States, and with those of western Europe. 

Darton, Nelson Hokatio. 

1896.1. Catalogue and index of contributions to North 
American geology, 1732-1891: U. S. Geol. Sur- 
vey Bull. 127, 1045 pp., 1896. 

1903.1. Preliminary report on the geology and water 
resources of Nebraska west of the one hundred 
and third meridian: U. S. Geol. Survey Prof. 
Paper 17, 69 pp., 43 pis. (inch 9 maps), 23 figs., 
1903. 

Titanotherium zone (Chadron formation) of western 
Nebraska, along the North Platte, Scott Bluffs, Sioux County, 
etc. 

1905.1. Age of the Monument Creek formation: Am. Jour. 

Sci., 4th ser., vol. 20, pp. 178-180, 1905. 
Menodus {TiianoiheriuTTi) remains. Oligoeene. 

1905.2. Preliminary report on the geology and underground 

water resources of the central Great Plains: 
U. S. Geol. Survey Prof. Paper 32, 433 pp., 72 
pis., 18 figs., 1905. 

1906.1. Geology and underground waters of the Arkansas 

Valley in eastern Colorado: U. S. Geol. Survey 
Prof. Paper 52, 90 pp., 27 pis., 2 figs., 1906. 

" Monument Creek formation," containing Menodus (^Titano- 
iherium) of White River age (p. 34) . Nussbaum formation, 
of late Tertiary age (p. 34) . 

1906.2. Geology of the Big Horn Mountains: U. S. Geol. 

Survey Prof. Paper 51, 129 pp., 47 pis., 14 
figs., 1906. 

Brief reference to the Eocene rocks (p. C7). See especially 
Bridger [Wind River] formation (p. 70). 

Davis, William Morris. 

1900.1. The fresh-water Tertiary formations of the Rocky 
Mountain region: Am. Acad. Arts and Sci. Proc, 
vol. 35, pp. 346-373, 1900. 

History of opinion on mode of formation; evidence against 
lake-bed hypothesis and in favor of fluviatile origin. 



134 



TITANOTHEBES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



Dawkins, W. Boyd. 

1880.1. The classification of the Tertiary period by means 
of the Ivlammalia: Geol. Soc. London Quart. 
Jour., 1880, pp. 379-405. 

Tertiary and Quaternary horizons and faunas of Great 
Britain, France, and Italy correlatRd. 

Deperet, Charles. 

1893.1. Note sur la succession stratigraphique des faunes 
de mammifcres pliocenes d' Europe et du Pla- 
teau central en particulier: Soc. g<Sol. France 
Bull., 3d ser., vol. 21, pp. 524-540, 1893. 
1906.1. L'evolution des mammiferes tertiaires, importance 
des migrations, epoque miocene: Compt. Rend., 
vol. 143, No. 26, pp. 1120-1123, 1906. The 
evolution of Tertiary mammals and the impor- 
tance of their migrations (translation) : Am. 
Naturalist, vol. 42, pp. 109-114, 166-170, 
303-307. 

DOLLO, Loui.?. 

1909.1. The fossil vertebrates of Belgium [Correlation 
Bull. No. 2] (translation by W. D. Matthew) : 
New York Acad. Sci. Annals, vol. 19, No. 4, 
pt. 1, pp. 99-119, pis. 4-10, July 31, 1909. 

DouGL.ASs, Earl. 

1S99.1. The Neocene lake beds of western Montana and 
descriptions of some new vertebrates from the 
Loup Fork: Montana LTniv. thesis, 27 pp., 
4 pis., June, 1899. 

Geology, faunas, and correlation of White River, "Deep 
River," and "Madison Valley." "Loup Fork" horizons in 
Montana. Systematic descriptions of certain fossil camels, etc- 

1902.1. Fossil Mammalia of the White River beds of 

Montana: Am. Philos. Soc. Trans., new ser., 
vol. 20, pp. 227-278, pi. 9, 1902. 

"Pipestone beds," "Toston beds," "Blaclitail Deer Creelc 
beds." Geology and faunas; new genera and species of mam- 
mals. 

1902.2. A Cretaceous and lower Tertiary section in south- 

central Montana: Am. Philos. Soc. Proc, vol. 
41, No. 170, pp. 207-224, pi. 29, April, 1902. 

Sketch of the Jurassic and Cretaceous deposits. Probable 
relations of the "Laramie" and overlying beds. Fossil mam. 
mals of the Fort Union beds. Describes the Fort Union beds 
of Montana; considers them as of practically the same age as 
the Torrejon Tertiary. Places Puerco as Upper Cretaceous- 

1902.3. The discovery of Torrejon mammals in Montana: 

Science, new ser., vol. 15, No. 372, pp. 272-273, 
Feb. 14, 1902. 

First record of mammals from Fort Union beds of Crazy 
Mountains region. 

1903.1. New vertebrates from the Montana Tertiary: 
Carnegie Mus. Annals, vol. 2, No. 2, pp. 
145-199, pi. 2, 37 figs., November, 1903. 

"Sage Greek" (Eocene?), White River deposits, "Fort 
'Logan beds" (upper Oligocene), "Deep" and "Flint Creek" 
beds. New mammals described. 

1909.1. Preliminary descriptions of some new titanotheres 
from the Uinta deposits: Carnegie Mus. Annals, 
vol. 6, No. 2, pp. 304-313, pis. 13-15, 8 figs., 
August, 1909. 
Describes new faunas from liorizon B. 

Dow, John M. 

1867.1. Extracts from letters relating to Tapirus bairdi 
(read by P. L. Sclater) : Zool. Soc. London Proc, 
1867, p. 241, 1867. 



Earle, Charles. 

1895.1. See Osborn, Henry Fairfield, 1895.95. 

Emmons, Samuel Franklin. 

1907.1. Uinta Mountains: Geol. Soc. America Bull., vol. 
17, pp. 287-302, pi. 24, 2 figs., July 13, 1907. 

Endlich, Frederic Miller. 

1877.1. Report on the San Juan region: U. S. Geol. and 
Geog. Survey Terr. Ninth Ann. Rept., pp. 176- 
191, 1877. 

Tertiary (p. 189). Puerco beds of Animas Valley, southern 
Colorado (1,000 to 1,200 feet), are considered the basal member 
of the Wasatch. 

1879.1. Report on the geology of the Sweetwater district: 
U. S. Geol. and Geog. Survey Terr. Eleventh 
Ann. Rept., pp. 5-158, 1879. ' 

Refers to the lower Bridger exposed in the northern part o! 
the basin, Big Sandy^Creek, etc. (p. 132). Considers a portion 
of the "Wasatch" of Beaver Creek, Wyo., as parallelwith the 
Puerco marls. 
Filhol, Henri. 

1885.1. Observations sur le memoire de M. Cope intitul(5 
"Relations des horizons * * * d'animaux 
vertebres fossiles en Europe et en Amerique": 
Annales sci. g6ol., vol. 17, art. 2, pp. 1-18, pi. 
6, 1885. 

FiNLAY, George Irving. 

1916.1. U. S. Geol. Survey Geol. Atlas, Colorado Springs 
folio (No. 203), 17 pp., 3 maps. 
Laramie, Dawson, and Denver of Colorado; flora, fauna. 

Fisher, Cassius Asa. 

1906.1. Geology and water resources of the Big Horn 
Basin, Wyo.: U. S. Geol. Survey Prof. Paper 53, 
72 pp., 16 pis., 1906. 

Discusses briefly the character, thickness, and distribution 
of the Wasatch formation (p. 33) . 

Fraas, Eberhard. 

1901.1. On the aqueous vs. eolian deposition of the White 
River Oligocene of South Dakota (translation 
by H. F. Osborn): Science, new ser., vol. 14, 
No. 345, pp. 210-212, Aug. 9, 1901. 

" Titanotherlum beds" formed by river and flood-pli.in 
deposits exposed during dry season. Middle " Oreodon beds " 
deposited by a shallow lake with dissolved materials of varying 
concentration (cf. banded layers). Upper "Oreodon beds" 
formed by eolian loess. 

Gardner, James Heney". 

1910.1. The Puerco and Torrejon formations of the 
Nacimiento group: Jour. Geology, vol. 18, 
pp. 702-741, 1 pi., 9 figs., 1910. 

Gives historical review. Topography, structure, and physi- 
ographic record of the Puerco-Torrejon district described. 
Considers that an unconformity exists between the two for- 
mations, to which the group name Nacimiento is given. 

Gidley, James Williams. 

1904.1. See Matthew, William Diller, 1904.1. 

1917.1. [The 1910 collection near the Davis ranch, Powder 
River valley, Wyo.] In Wegemann, C. H., 
Wasatch fossils in so-called Fort Union beds of 
the Powder River basin, Wyo., and their bear- 
ing on the stratigraphy of the region: U. S. 
Geol. Survey Prof. Paper 108, p. 59, 1917. 

Gilbert, Grove Karl. 

1898.1. The underground waters of the Arkansas Valley 
in eastern Colorado: U. S. Geol. Survey 
Seventeenth Ann. Rept., pt. 2, pp. 553-601, 
pis. 56-68, figs. 45-49, 1896. 

Rocky Mountain deposits may be of fluviatile and not of 
lacustrine origin. 



ENVIKONMENT OF THE TITANOTHERES 



135 



Granger, Walter. 

1909.1. Faunal horizons of the Washakie formation of 
southern Wyoming: Am. Mus. Nat. Hist. 
Bull., vol. 26, pp. 13-23, pis. 2-6, 1 map, Jan. 
19, 1909. 

Divides "Washakie beds" into two horizons, characterized 
by fauna and lithology. Lower horizon=upper Bridger; 
upper horizon=lower and middle "Uinta" (Uinta A and B). 

1910.1. Tertiary faunal horizons in the Wind River Basin, 
Wyo., with descriptions of new Eocene mam- 
mals: Am. Mus. Nat. Hist. Bull., vol. 28, pp. 
235-251, pis. 20-23, 6 figs., July 16, 1910. 

Determines two distinct faunal horizons in Wind Hiver 
beds — the Lambdotherittm zone and an earlier horizon. 

1911.1. See Sinclair, William John, 1911.1. 
1912.1. See Sinclair, William John, 1912.1. 

1914.1. On the names of lower Eocene faunal horizons of 

Wyoming and New Mexico: Am. Mus. Nat. 
Hist. Bull., vol. 33, pp. 201-207, Mar. 31, 1914. 
Names "Clark Fork," "Sand Coulee," and "Gray Bull 
beds" of Big Horn, Wyo., and "Almagre" and "Largo beds" 
of New Mexico. Correlates the lower Eocene of New Mexico 
with that of the various Wyoming basins. 

1914.2. See Sinclair, William John, 1914.1. 

1917.2. Notes on Paleocene and lower Eocene mammal 
horizons of northern New Mexico and southern 
Colorado: Am. Mus. Nat. Hist. Bull., vol. 37, 
pp. 821-830, Dec. 5, 1917. 

1918.1 (and Matthew, W. D.). A revision of the lower 
Eocene Wasatch and Wind River faunas: 
Am. Mus. Nat. Hist. Bull., vol. 38, pp. 565- 
657, 1918. 

Gray, Dr. J. E. 

1872.1. Notes on a new species of tapir {Tapirus leucogenys) 
from the snowy regions of the Cordilleras of 
Ecuador and on the young spotted tapirs of 
tropical America: Zool. Soc. London Proc, 
1872, pp. 483-492, pis. 21-22. 

Gregory, John Walter. 

1896.1. The Great Rift Valley, 422 pp., London, John 

Murray, 1896. 

Harris, Gilbert Dennison. 

1892.2. See Dall, William Healey, 1892.2. 

Hatcher, John Bell. 

1893.1. The Titanotherium beds: Am. Naturahst, vol. 27, 
pp. 204-221, 3 figs.. Mar., 1893. 

General description. Accepts lacustrine theory of deposi- 
tion. 

1894.1. Discovery of Diceratherium, the two-horned 
rhinoceros, in the White River beds of South 
Dakota: Am. Geologist, vol. 13, pp. 360-361, 
May, 1894. 
Top of White River correlated with John Day formation. 

1895.1. On a new species of Diplacodon, with a discussion 
of the relations of that genus to Telmatherium: 
Am. Naturahst, vol. 29, pp. 1084-1093, pis. 
38-40, fig. 1, Dec, 1895. 

1902.3. Origin of the Oligocene and Miocene deposits of 

the Great Plains: Am. Philos. Soc. Proc, vol. 
41, pp. 113-131, 1902. 

Summarizes facts and accepts theory of small lakes, flood 
plains, river channels, and pampas as prevailing conditions 
during Oligocene and Miocene time. Gering, Arikaree, 
Ogalalla, Monroe Creek, Harrison, and "Nebraska" of Scott. 
Classification of the Oligocene and Miocene. "Lake-bed" 
hypothesis of origin disproved in favor of fluviatile, flood-plain, 
and eolian hypothesis . 



Haworth, Erasmus. 

1897.1. Physical properties of the Tertiary [of Kansas]: 
Kansas Univ. Geol. Survey, vol. 2, pp. 247-284, 
pis. 36-44, 1897. 

Rejects "lake-basin" hypothesis in favor of hypothesis of 
fluviatile origin of Tertiary of Kansas. 

Hay, (Jliver Perry. 

1905.1 The fossil turtles of the Bridger Basin: Am. 
Geologist, vol. 35, pp. 327-342, June, 1905. 

Evidence for flood-plain rather than lacustrine origin of the 
Bridger. Discussion of life and climatic conditions. 

1908.1 The fossil turtles of North America: Carnegie Inst. 
Wash. Pub. 75, 568 pp., 113 pis., 704 figs., 1908. 

H.\y, Robert. 

1889.1. Northwest Kansas, its topography, geology, cli- 
mate, and resources: Kansas State Board Agr. 
Sixth Bienn. Rept., pp. 92-116, 2 maps, 4 figs., 
1889. 
See especially discussion of the Tertiary geology of Kansas. 

Hayden, Ferdinand Vandiveer. 

1858.1. Notes on the geology of the Mauvaises Terres of 
White River, Nebr. : Acad. Nat. Soi. Phila- 
delphia Proc, vol. 9, pp. 151-165, 1858. 

Refers to Bear Creek, Pennington County, S. Dak. Type 
locality of Mesohippus bairdii. 

1862.1. See Meek, Fielding Bradford, 1862.1. 

1869.1. Geological report of the exploration of the Yellow- 

stone and Missouri Rivers, by F. V. Hayden, 
under the direction of William F. Raynolds, 
174 pp., 1 map, Washington, 1869. 

1871.2. Report of F. V. Hayden. In [Fourth Annual] 

Preliminary report of the United States geolog- 
ical survey of Wyoming and portions of con- 
tiguous territories, pp. 9-81, 1871. 

A general account of the topography and geology (type 
description) of the Bridger Basin (pp. 54-58) . Considers upper 
portion of Washakie Basin sediments, as either an extension 
eastward of the Bridger beds or as a separate deposit of the 
same age. Notes occurrence of vertebrate fossils. 

1873.1. PreUminary field report of the United States 
geological survey of Colorado and New Mexico: 
U. S. Geol. Survey Terr. Third Ann. Rept., 
pp. 105-251, 1869, reprinted 1873. [Reprinted 
in 1873 in First, Second, and Third Annual 
Reports of the Geological Survey of the Terri- 
tories. In the text of this monograph refer- 
ence is made to the reprinted edition.] 

Names and briefly describes "Green River shales," Bridger 
"group," Wasatch "group," and Bear River "group." Desig- 
nates Tertiary deposits between Creston and Bitter Creek 
along Union Pacific RaUroad as "Washakie group" (p. 190). 

1881.1. Geological and geographical atlas of Colorado and 
portions of adjacent territory, U. S. Geol. and 
Geog. Survey Terr., 1877, corrected to 1881. 

Heller, Edmund. 

1914.1. See Roosevelt, Theodore, 1914.1. 

Hills, Richard Charles. 

1888.1. The recently discovered Tertiary beds of the 

Huerfano River basin, Colo. : Colorado Sei. 
Soc Proc, vol. 3, pp. 148-164, 1 map, 1888. 

Beds first described. Upper half suspected to be of Wasatch 
age. 

1889.2. Additional notes on the Huerfano .beds: Colorado 

Sci. Soc. Proc, vol. 3, pp. 217-223, 1889. 

Mammals reported from upper division. Bridger age 
indicated. 



136 



TITANOTHEBES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



Hills, Richard Charles — Continued. 

1891.1. Remarks on the classification of the Huerfano 
Eocene: Colorado Sci. Soc. Proc, vol. 4, pp. 
7-9, 1891. 

Series divided into Huerfano, Cuchara, and Poison Canyon 
beds. Huerfano=Bridger; other two=lower Eocene. 
HovEY, Edmund Otis. 

1908.1. See Willis, Bailey, 1908.1. 
Irving, John Duer. 

1896.1 The stratigraphical relations of the Browns 
Park beds of Utah: New York Acad. Sci. 
Trans., vol. 15, p. 252, pi. 18, Sept., 1896. 
The beds in Browns Park valley assigned to the Pliocene. 
JOHANNSEN, ALBERT. 

1914.1. Petrographio analysis of the Bridger, Washakie, 
and other Eocene formations of the Rocky 
Mountains: Am. Mus. Nat. Hist. Bull., vol. 
33, pp. 209-222, 2 figs.. Mar. 31, 1914. 

Considers Bridger and "Washakie" rocks largely tufls 
modified by slight transportation. The older Eocene rocks 
are considered more strictly sedimentary. 

Johnson, Willard Drake. 

1901.1. The High Plains and their utilization: U. S. 
Geol. Survey Twenty-first Ann. Rept., pt. 4, 
pp. 601-741, pis. 113-116, figs. 300-329, 1901; 
Twenty-second Ann. Rept., pt. 4, pp. 631-669, 
pis. 55-65, figs. 236-244, 1902. 

Tertiary deposits of the Plains, of fluviatile and flood-plain 
origin. 

King, Cl.arence. 

1876.1. Geological and topographical atlas accompanying 
the report of the Geological Exploration of the 
40th Parallel, 1876. 
1878.1. Systematic geology: U. S. Geol. Expl. 40th Par. 
Rept., vol. 1, 803 pp., 21 pis., 12 maps, 1878. 

Gives the name "Vermilion Creek" to the Wasatch beds of 
southern Wyoming; considers them as lowest Eocene and 
unconformable with the overlying Green Eiver beds. The 
name "Uinta group" is given to the uppermost 400 feet of the 
sediments in the vallej' of White River; considered to lie un- 
cnnformably on lower beds and to represent uppermost Eocene 
Mammals collected by Marsh are listed. Area is mapped, 
and relationships of Bridger with other Eocene deposits of the 
basin are set forth. 

Knowlton, Frank Hall. 

1902.1. Fossil flora of the John Day Basin, Oreg.: U. S. 
Geol. Survey Bull. 204, i53 pp., 17 pis., 1902. 

Geology (pp. U-20, 102-108). Mascall formation referred 
to upper Miocene. 

1909.1. The stratigraphic relations and paleontology of 
the "Hell Creek beds," " Ceratops beds," and 
equivalents, and their reference to the Fort 
Union formation: Washington Acad. Sci. Proc, 
vol. 11, No. 3, pp. 179-238, Aug. 14, 1909. 

Kobelt, W. 

1902.1. Die Verbreitung der Tierwelt, 576 pp., Leipzig, 
1902. 

Lambb, Lawrence Morris. 

1908.1. The Vertebrata of the Oligocene of the Cypress 
Hills, Saskatchewan: Canada Geol. Survey 
Contr. Canadian Paleontology, vol. 3, pt. 4, 
65 pp., 8 pis., 1908. 

Leidy, Joseph. 

1869.1. The extinct mammalian fauna of Dakota and 
Nebraska, including an account of some allied 
forms from other localities, together with a 
synopsis of the mammalian remains of North 
America: Acad. Nat. Sci. Philadelphia Jour. 
2d ser., vol. 7, 472 pp., 30 pis., 1869. 



Lindgren, Waldemar. 

1915.1. The igneous geology of the Cordilleras and its 
problems. In Problems of American geology 
(Silliman Memorial Lectures, 1913), pp. 234- 
286, 1 map, Yale Univ. Press, 1915. 

Lonnbekg, Einar. 

1912.1. Mammals collected by the Swedish zoological 
expedition to British East Africa, 1911: K. 
svenska Vet.-Akad. Handlingar, Bd. 48, No. 5, 
1912. 

LooMis, Frederic Brewbteh. 

1904.1. Two new river reptiles from the titanothere beds: 
Am. Jour. Sci., 4th ser., vol. 18, pp. 427-432, 
4 figs., Dec, 1904. 

Flood-plain origin of the " Titanotherium beds." 

1906.1. The Tertiary of Montana: Carnegie Mus. Mem., 
vol. 2, pp. 203-224, pi. 22, 1905. 

Chiefly a description of Iciops, Xenotherium, and other lower 
White River mammals. 

1907.1. Origin of the Wasatch deposits: Am. Jour. Sci., 
4th ser., vol. 23, pp. 356-364, 3 figs.. May, 1907. 
Treats of the Big Uorn Basin Wasatch; divides the beds into 
three faunal levels, lists fossils from each level, and gives sec- 
tions. The Wasatch is considered a flood-plain deposit, the 
upper 1,000 feet of which appear to overlap in time the base 
of the Wind River. 

Lx'LL, Richard Swann. 

1905.1 Megacerops tyleri, a new species of titanothere 
from the Bad Lands of South Dakota: Jour. 
Geology, vol. 13, No. 5, pp. 443-456, pis. 3-4, 
1905. 

Lydekkeb, Richard (editor). 

?1893.1. The new natural history, vols. 1-4 (American 
reprint of "The Royal natural histor}'," pub- 
lished 1893-1896). 

Lyons, H. G. 

1906.1 The physiography of the River Nile and its basin, 
441 pp., 48 pis., 1 map, Egypt Survey Dept., 
1906. 

Rate of deposition (p. 334). 

McMaster, John Bach. 

1881.1. See Osborn, Henry Fairfield, 1881.8. 

Marsh, Othniel Charles. 

1871. 3. On the geology of the eastern Uintah Mountains: 
Am. Jour. Sci., 3d ser., vol. 1, pp. 191-198, 1871. 
Short account of the expedition to Uinta Basin in 1870. Con- 
siders Uinta Basin deposits synchronous with those of Bridger 
Basin on paleontologic evidence. Considers the fossils as 
indicating much greater age than Miocene of eastern Rocky 
Mountain basins. 

1875. 2. Ancient lake basins of the Rocky Mountain region: 
Am. Jour. Sci., 3d ser., vol. 9, pp. 49-52, Janu- 
ary, 1875. 

1877. 1. Introduction and succession of vertebrate life in 
America: Am. Jour. Sci., 3d ser., vol. 14, pp. 
337-378, 1877. 

Plate showing successive horizons named from characteris- 
tic genera. Names Diplacodon zone (p. 354). 

1891. 2. Geologic horizons as determined by vertebrate 
fossils: Am. Jour. Sci., 3d ser., vol. 42, pp. 336- 
338, October, 1891. 

1898. 1. The comparative value of different kinds of fossils 
in determining geological age: Am. Jour. Sci., 
4th ser., vol. 6, pp. 483-486, December, 1898. 

Value of a form depends upon its modiflability in accordance 
with changing conditions. 



ENVIBONMENT OF THE TITANOTHEKES 



137 



Matthew, 
1897. 2 



1899. 2. 



1901. 1 



1903. 1 



1906. 1 



1908. 1 



1909. 1. 



1909. 2. 
1914. 1 



1918. 1 



William Dillbe. 

. A revision of the Puerco fauna: Am. Mus. Nat. 

Hist. Bull., vol. 9, pp. 259-323, Nov. 16, 1897. 
Points out the distinct separation of species of upper and 

lower beds and adopts Wortman's proposed name, Torrejon 

for the upper beds. 

, A provisional classification of tlie fresh-water 
Tertiary of the West: Am. Mus. Nat. Hist. 
Bull., vol. 12, pp. 19-75, Mar. 31, 1899. 
Is the White River Tertiary an eolian formation? 
Am. Naturalist, vol. 33, pp. 403-408, May, 1899. 
Summary of the paleontologic evidence against the "lake- 
basin" hypothesis. 

Fossil mammals of the Tertiary of northeastern 
Colorado: Am. Mus. Nat. Hist. Mem., vol. 1, 
pt. 7, pp. 353-447, 1901. 

Stratigraphy of White River deposits ("Horsetail Creek," 
"Cedar Creek," and "Martin Canyon beds") and of "Loup 
Fork" formation ("Pawnee Creek beds"). Evidence as to 
mode of deposition (chiefly eolian); analysis of faunas: correla- 
tion of horizons: systematic descriptions. 

List of the Pleistocene fauna from Hay Springs, 
Nebr.: Am. Mus. Nat. Hist. Bull., vol. 16, pp. 
317-322, Sept. 25, 1902. 

Lists for comparison the faunas of Hay Springs (Nebr.), 
Silver Lake (Oreg.), and Washtucna Lake (Wash,). 

The fauna of the Titanotherium beds at Pipestone 
Springs, Mont.: Am. Mus. Nat. Hist. Bull., 
vol. 19, pp. 197-226, 19 figs.. May 9, 1903. 

(and Gidley, J. W.). New or little-known mam- 
mals from the Miocene of South Dakota: Am. 
Mus. Nat. Hist. Bull., vol. 20, pp. 241-268, 15 
figs., July 20, 1904. 

Upper Miocene "Loup Fork beds," geology and faunal list. 
Lower Miocene "Rosebud beds" (new name). New Carni- 
vora and Rodentia. 

Hypothetical outlines of the continents in Tertiary 
times: Am. Mus. Nat. Hist. Bull., vol. 22, pp. 
353-384, 7 figs., Oct. 25, 1906. 

A lower Miocene fauna from South Dakota: Am. 
Mus. Nat. Hist. Bull., vol. 23, pp. 169-219, 
26 figs., 1907. 

"Lower Rosebud" and "Upper Rosebud" deposits and 
faunas: comparison with American Oligocene and Miocene 
faunas. New Carnivora, Rodentia, Artiodactyla. 

Mammalian migrations between Europe and 

North America: Am. Jour. Sci., 4th ser., vol. 

25, pp. 68-70, January, 1908. 
The Carnivora and Insectivora of the Bridger 

Basin, middle Eocene: Am. Mus. Nat. Hist. 

Mem., vol. 9, pt. 6, pp. 289-559, pis. 44-52, 

118 figs., 1909. 
History of exploration. Stratigraphy and faunal divisions. 

Condition of deposition. 

See Osborn, Henry Fairfield, 1909. 321. 

Evidence of the Paleocene vertebrate fauna on the 

Cretaceous - Tertiary problem: Geol. Soc. 

America Bull., vol. 25, pp. 381-402, Sept. 15, 

1914. 
See Granger, Walter, 1918. 1. 



Meek, Fielding Bradford. 

1862. 1 (and Hayden, F. V.). Descriptions of new Lower 
Silurian (Primordial), Jurassic, Cretaceous, and 
Tertiary fossils, collected in Nebraska by the 
exploring expedition under command of Wm. F. 
Raynolds, with some remarks on the rocks from 
which they were obtained: Acad. Nat. .Sci. 
Philadelphia Proc, vol. 13, pp. 415-447, 1862. 
Wind River deposits considered intermediate in age between 
Fort Union and White River. 

101959— 2S— VOL 1 11 



Mercer, Henry Chapman. 

1899. 1. The bone cave at Port Kennedy, Pa., and its 
partial excavation in 1894, 1895, 1896: Acad. 
Nat. Sci. Philadelphia Jour., 2d ser., vol. 11, 
pt. 2, pp. 269-286, Feb. 4, 1899. 

Referred to the Pleistocene, but without comparison with 
other cave formations and faunas. 

Merriam, John Campbell. 

1901.1. A contribution to the geology of the John Day 
Basin [Oreg.]: California Univ. Dept. Geology 
Bull., vol. 2, pp. 269-314, pis. 6-8, fig. 1, 1901. 

Geology, faunas, and floras of the Cretaceous (Chieo and 
Kno.\ville), Eocene (Clarno), Oligocene (John Day), Columbia 
River lava, Miocene (Mascall), Pliocene (Rattlesnake), 
Quaternary (p. 2C9), 

NicKLBs, John M. 

1924.1. Geologic literature on North America, 1785-1918: 
U. S. Geol. Survey Bull. 746 (Bibliography), 
1167 pp.; Bull. 747 (Index), 658 pp., 1924. 

Osborn, Henry Fairfield. 

1878.3 (and Scott, W. B.). Palaeontologieal report of 
the Princeton Scientific Expedition of 1877: 
Princeton Coll. E. M. Mus. Geol. Archaeol. 
Contr., No. 1, 107 pp., Sept. 1, 1878. 

A general account of the Bridger badlands, with notes on 
analysis of the rocks. 

1881.8 (and McMaster, J. B.). A memoir upon Loxolo- 
phodon and Uintaiherium, two genera of the 
I suborder Dinocerata, accompanied by a strati- 

graphical report of the Bridger beds in the 
Washakie Basin by J. B. McMaster: Princeton 
Coll. E. M. Mus. Geol. Archaeol. Contr., vol. 
1, No. 1, pp. 5-54, 1881. 

Topography and geology described. Section given and im- 
portant fossil localities indicated. Osborn notes for first time 
difference in fauna between beds of the two basins and con- 
siders "Washakie" as somewhat later than Bridger. First 
stratigraphic section with geologic location of species. Error 
in stratigraphy. 

1887.30. See Scott, William Berryman, 1887.1. 

1887.37 (and Scott, W. B.). Preliminary report on the 
vertebrate fossils of the Uinta formation col- 
lected by the Princeton expedition of 1886: 
Am. Philos. Soc. Proc, vol. 24, No. 126, pp. 
255-264, 1887. 

1890.51. See Scott, William Berryman, 1890.1. 

1892.67 (and Wortman, J. L.). Fossil mammals of the 
Wasatch and Wind River beds, collections of 
1891: Am. Mus. Nat. Hist. Bull., vol. 4, pp. 
81-147, Oct. 20, 1892. 

Geology of the Big Horn Basin (Wortman), p. 135, Analysis 
and description of the fauna (Osborn) . Considers Wind River 
beds distinct from and successive to the Wasatch of Big Horn 
Basin. 

1893-82. Rise of the Mammalia [vice-presidential address 
before American Association for the Advance- 
ment of Science]: Am. Jour. Sci., 3d ser., vol. 
46, pp. 379-392, 448-466, November, Decem- 
ber, 1893; Am. Assoc. Adv. Sci. Proc, vol. 
42, pp 189-227, 1894. 

1894.89. A division of the eutherian mammals into the 
Mesoplacentaha and Cenoplacentalia [terms 
subsequently altered to Meseutheria and Ce- 
neutheria]: New York Acad. Sci. Trans., vol. 
13, pp. 234-237, June 4, 1894. 



138 



TITANOTHERES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



OsBORN, Henrt Fairfield — Continued. 

1894.90 (and Wortman, J. L.). Fossil mammals of the 
lower Miocene White River beds, collection 
of 1892: Am. Mus. Nat. Hist. Bull., vol. 6, 
pp. 199-228, pis. 2-3, July 28, 1894. 

Succession of species in the White River "Miocene" [Oligo- 
cene], 

1895.95 (and Earle, Charles). Fossil mammals of the 
Puerco beds, collection of 1892: Am. Mus. 
Nat. Hist. BuU., vol. 7, pp. 1-70, Mar. 8, 1895. 

Wortman (field notes, pp. 1, 2) divides the Puerco into upper 
and lower beds, with two thin yet distinct fossfl-bearing 
strata in the lower bed and one (?) thicker stratum in the 
upper. Gives the localities of both levels and estimates the 
thickness of the combined beds (upper and lower) at 800 to 
1,000 feet. 

1895.98 (and Peterson, O. A.). Fossil mammals of the 
Uinta Basin, expedition of 1894 (geologic 
levels by O. A. Peterson) : Am. Mus. Nat. 
Hist. Bull., vol. 7, pp. 71-105, 17 figs.. May 18, 
1895. 

Divides Uinta Basin deposits into three horizons; the three 
faunal levels (horizons A, B, C), with faunal lists. New 
genera and species, especially of Mesonyx, TelmatheTium, 
SpheTWCoelits, and Elotherium. 

1895.105 (and Wortman, J. L.). Perissodactyls of the 
lower Miocene White River beds: Am. Mus. 
Nat. Hist. Bull., vol. 7, pp. 343-375, pis. 8-11, 
Dec. 23, 1895. 
Oligocene. 

1897.126. The Huerfano lake basin, southern Colorado, 
and its Wind River and Bridger fauna: Am. 
Mus. Nat. Hist. BuU., vol. 9, pp. 247-258, 
Oct. 20, 1897. 

1900.182. The geological and faunal relations of Europe 
and America during the Tertiary period and 
the theory of the successive invasions of an 
African fauna: Science, new ser., vol. 11, No. 
276, pp. 561-574, Apr. 13, 1900. 

1900.187. Correlation between Tertiary mammal horizons 
of Europe and America, an introduction to the 
more exact investigation of Tertiary zoogeog- 
raphy, prehminary study with third trial sheet: 
New York Acad. Sci. Annals, vol. 13, No. 1, pp. 
1-72, July 21, 1900. 

1901.200. Correlation des horizons mammiferes tertiaires 

en Europe et en Am^rique: Cong. g^ol. internat., 
8« sess., Compt. rend., pp. 357-363, 1901. 

1901.201. See Fraas, Eberhard, 1901.1. 

1902.214. The law of adaptive radiation: Am. Naturalist, 
vol. 36, pp. 353-363, May, 1902. 

1905.267. Ten years progress in the mammalian palaeon- 
tology of North America: Cong, internat. 
zool., 6« sess. (Bern, 1904), Compt. rend., 
pp. 86-113m, pis. 1-15; Am. Geologist, vol. 
36, pp. 199-229, October, 1905. 

1907.294. Tertiary mammal horizons of North America: 
Am. Mus. Nat. Hist. Bull., vol. 23, pp. 237- 
253, 3 figs.. Mar. 30, 1907. 

1909.321 (and Matthew, W. D.). Cenozoic mammal 
horizons of western North America, with ap- 
pendix, Faunal lists of the Tertiary Mammalia 
of the West, by William DiUer Matthew: U. S. 
Geol. Survey BuU. 361, 138 pp., 2 pis., 14 figs., 
January, 1909. 

1910.341. The paleontologic correlation through the Bache 
fund: Science, new ser., vol. 31, No. 794, pp. 
407-408, Mar. 18, 1910. 



OsBORN, Hbnet Fairfield — Continued. 

1910.342. Correlation of the Cenozoic through its mam- 
malian fife: Jour. Geology, vol. 18, No. 3, pp., 
201-215, April-May, 1910; OutUnes of geologic 
history, pp. 251-264, Chicago Univ. Press 
July, 1910. 

1910.345. Paleontologic evidences of adaptive radiation: 

Pop. Sci. Monthly, vol. 77, pp. 77-81, July, 
1910. 

1910.346. The age of mammals in Europe, Asia, and North 

America, 635 pp., 220 figs.. New York, Mac- 

millan Co., 1910. 
1912.376. Correlation and paleogeography : Geol. Soc. 

America BuU., vol. 23, pp. 232-256, 1912. 
1919.494. New titanotheres of the Huerfano: Am. Mus. 

Nat. Hist. BuU., vol. 41, pp. 557-569, 7 figs., 

1919. 
Peale, Albert Charle.s. 

1876.1. Report on stratigraphy — Cenozoic formations: 

U. S. Geol. and Geog. Survey Terr. Eighth 

Ann. Rept., pp. 75-180, 13 pis., 5 maps, 1876. 

"Washakie" treated as distinct group. Contains table of 
localities, authorities, references, etc. (p. 140). 

1879.1. Report on the geology of the Green River district: 
U. S. Geol. and Geog. Survey Terr. Eleventh 
Ann. Rept., pp. 511-542, pis. 47-54, 1879. 

Peterson, Olof August. 

1895.1. See Osborn, Henry Fairfield, 1895.98. 

1914.1. A new titanothere from the Uinta Eocene: Car- 

negie Mus. Annals, vol. 9, pp. 29-52, pis. 6-10, 
figs. 1-14, 1914. 
Diploceras osborni, from horizon B. 

1914.2. A smaU titanothere from the lower Uinta beds: 

Carnegie Mus. Annals, vol. 9, pp. 53-57, pi. 11, 
figs. 1-2, 1914. 
Beterotitanops parvus, from horizon A. 

1914.3. Some undescribed remains of the Uinta titanothere 

Dolichorhinus: Carnegie Mus. Annals, vol. 9, 
pp. 129-138, figs. 1-7, 1914. 
From Uinta A. 

1914.4. A correction of generic name: Carnegie Mus. 

Annals, vol. 9, p. 220, 1914. 

PoHLiG, Hans. 

1891.1. Dentition und Kraniologie des Elephas aniiquus 
Falc. * * * Nachtrage: K. Leop. -Carol, 
deutsche Akad. Naturforscher Nova Acta, 
Band 57, pp. 285-459, 1891. 

Powell, John Wesley. 

1876.1. (and White, C. A.). Invertebrate paleontology of 
the Plateau province. In Powell, J. W., Report 
on the geology of the eastern portion of the 
Uinta Mountains and a region of country adja- 
cent thereto, pp. 74-135, U. S. Geol. and 
Geog. Survey Terr., 1876. 

Ransome, Frederick Leslie. 

1915.1. The Tertiary orogeny of the North American Cor- 
diUera and its problems. In Problems of Amer- 
ican geology (SilUman Memorial Lectures, 
1913), pp. 287-376, 1 map, Yale Univ. Press, 
1915. 

Richardson, George Burr. 

1912.1. The Monument Creek group: Geol. Soc. America 
BuU., vol. 23, pp. 267-276, 1 fig., 1912. 

Describes the Dawson arkose and Castle Rock conglomerate 
forming the "Monument Creek group" of Colorado and dis- 
cusses their relation to the Denver and Arapahoe formations 



ENVIRONMENT OF THE TITANOTHEEES 



139 



RiGGS, Elmer Samuel. 

1912.1. New or little known titanotheres from the lower 
Uintah formations: Field Mus. Nat. Hist. Pub. 
159 (Geol. ser., vol. 4, No. 2), pp. 17-41, pis. 
4-12, figs. 1-2, June, 1912. 

Discusses stratigraphy of lower part of lower horizon of Uinta 
Basin and gives general and detailed sections, with exact strati- 
graphic position of various species of titanotheres. 

Roosevelt, Theodore. 

1914.1. (and HeUer, Edmund). Life histories of African 
game animals, vols. 1-2, New York, Charles 
Scribner's Sons, 1914. 

Sampson, J. A. 

1905.1. A deer's bill of fare: Sierra Club Bull., vol. 5, 
pp. 194-210, 1905. 

SCHIMPER, A. F. W. 

1903.1. Plant geography upon a physiological basis (trans- 
lation by W. R. Fisher, revised and edited by 
Percy Groom and I. B. Balfour), 839 pp., 
Oxford, 1903. 

ScLATER, Philip Lutley. 

1894.1 (and Thomas, Oldfield). The book of antelopes, 
vols. 1-4, London, 1894. [Issued in parts dated 
consecutively 1894-1900.) 

Scott, William Bbrrtman. 

1878.1. See Osborn, Henry Fairfield, 1878.3. 

1887.1 (and Osborn, H. F.). Preliminary account of the 
fossil mammals from the White River forma- 
tion contained in the Museum of Comparative 
Zoology: Harvard Coll. Mus. Comp. Zoology 
Bull., vol. 13, pp. 151-171, pis. 1-2, September, 
1887. 

1887.2. See Osborn, Henry Fairfield, 1887.37. 

1888.1. The upper Eocene lacustrine formations of the 
United States (abstract) : Am Assoc. Adv. 
Sci. Proc, 1887, p. 217, March, 1888. 

1890.1 (and Osborn, H. F.). The Mammalia of the 
Uinta formation: Part I, The geological and 
faunal relations of the Uinta formation, by 
W. B. Scott; Part II, The Creodonta, Rodentia, 
and Artiodactyla, by W. B. Scott; Part III, 
The Perissodactyla, by H. F. Osborn; Part IV, 
The evolution of the ungulate foot, by H. F. 
Osborn: Am. Philos. Soc. Trans., new ser., 
vol. 16, pt. 3, pp. 461-572, pis. 7-9, 1890. 

Considers the " Washaliie" a later substage of the Bridger 
formation, and notes that several forms of animals found in the 
beds are more similar to the Uinta Basin stages than to the 
Bridger stages. Uinta considered top of Eocene, but strong 
affinities with the White River Oligooene shown in the fauna. 

1893.1. The mammals of the Deep River beds: Am. 

Naturalist, vol. 27, pp. 659-662, July, 1893. 

Preliminary description. 

1894.1. The later lacustrine formations of the West: Geol. 

Soc. America BuU., vol. 5, pp. 594, 595, 1894. 

"Nebraska formation," " Corsoryx beds." Type reference. 

1895.1. The Mammalia of the Deep River beds: Am. 

Philos. Soc. Trans., new ser., vol. 8, pp. 55-185, 
6 pis., 1895. 

Geology (pp. 55-63) . European homotaxis with Sanson and 
Simorre (middle Miocene). 

1895.2. The Tertiary lacustrine formations of America: 

Science, new ser., vol. 2, No. 42, p. 499, Oct. 
18, 1895. 

Tabular correlation of Tertiary horizons of Europe and 
America. 



Scott, William Berryman — Continued. 

1899.1. The selenodont artiodaotyls of the Uinta Eocene: 
Wagner Free Inst. Sci. Trans., vol. 6, pp. i-xiii, 
15-122, pis. 1-4, May, 1899. 

Angular unconformity between horizons B and C. White 
River beds homotaxial with Ronzon of France. Uinta 
compared with Paris gypsum (Lutftien). 

Scudder, Samuel Hubbard. 

1890.1. The Tertiary insects of North America: U. S. 
Geol. Survey Terr. Rept., vol. 13, 734 pp., 28 
pis., 1 map, 1890. 

Map of the Tertiary lake basin at Florissant, Colo. Geology 
of the deposits yielding Tertiary insects in America. Regards 
Florissant ( Amyzon) beds as Oligocene? Volcanic origin of the 
deposits. Now regarded as Miocene. 

1894.1. The effect of glaciation and of the glacial period 
on the present fauna of North America: 
Am. Jour. Sci., 3d ser., vol. 48, pp. 179-187, 
September, 1894. 

Sinclair, William John. 

1906.1. Volcanic ash in the Bridger beds of Wyoming: 
Am. Mus. Nat. Hist. Bull., vol. 22, pp. 273- 
280, pis. 35-38, July 31, 1906. 

General features of the geology. Lithologic and stratigraphic 
classification of the Bridger "group." The entire series of 
Bridger rocks is determined as of volcanic origin. 

1909.1. The Washakie, a volcanic ash formation: Am. 

Mus. Nat. Hist. Bull., vol. 26, pp. 25-27, 
Jan. 19, 1909. 

Determines the nature of the volcanic material of the " Wash- 
akie" to be different from that of the Bridger, which argues 
against contemporaneity of deposition in the two basins. 

1911.1 (and Granger, Walter). Eocene and Oliogocene of 
the Wind River and Big Horn Basins: Am. 
Mus. Nat. Hist. Bull., vol. 30, pp. 83-117, 
July 11, 1911. 

General account of Wasatch and later beds of the Big Horn 
Basin. Discusses origin and mode of deposition of the sedi- 
ments. 

1912.1 (and Granger, Walter). Notes on the Tertiary 
deposits of the Big Horn Basin: Am. Mus. 
Nat. Hist. Bull., vol. 31, pp. 57-67, Mar. 30, 
1912. 

Additional observations as to deposition, extent, and chron- 
ological subdivision of Big Horn sediments. Describes "Ly- 
site" and "Lost Cabin formations" in the Big Horn Basin. 

1912.2. Contributions to geologic theory and method by 

American workers in vertebrate paleontology: 
Geol. Soc. America BuU., vol. 23, pp. 262-266, 
June, 1912. 
1914.1 (and Granger, Walter). Paleocene deposits of the 
San Juan Basin, N. Mex.: Am. Mus. Nat. 
Hist. Bull., vol. 33, pp. 297-316, pis. 20-27, 2 
figs., June 3, 1914. 

Gives descriptions, measurements, and sections of Puerco 
and Torrejon formations at various points. Determines that 
mammal remains come from two layers in the Puerco and two 
in the Torrejon. Considers both formations of fluviatile origin . 
Lists important fossil localities. 

Smith, James Henry. 

1900.1. The Eocene of North America west of the 100th 
meridian (Greenwich) : Jour. Geology, vol. 
8, pp. 444^471, 1 map, 1900. 
R§sum6 of literature on these deposits (pp. 452-454). 

Stanton, Timothy William. 

1909.1. The age and stratigraphic relations of the "Cera- 
tops beds" of Wyoming and Montana: Wash- 
ington Acad. Sci. Proc, vol. 9, No. 3, pp. 239- 
293, 1909. 



140 



TITANOTHERES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



Stanton, Timothy William — Continued. 

1914.1. Boundary between Cretaceous and Tertiary in 
North America as indicated by stratigraphy 
and invertebrate faunas: Geol. Soc. America 
Bull., vol. 25, pp. 349-351, Sept. 15, 1914. 

Records Lance resting conformably on Fox Hills along 
Missouri River in tlie Dakotas. 

1916.1. Contributions to the geology and paleontology of 
San Juan County, N. Mex. — 3, Nonmarine 
Cretaceous invertebrates of the San Juan Basin; 
U. S. Geol. Survey Prof. Paper 98, pp. 309-326, 
pis. 79-83, 1916. 

Stevenson-Hamilton, Major J. 

1912.1. Animal hfe in Africa, 539 pp., iUus., London, 
Wm. Heinemann, 1912. 

Stone, Ralph Walter. 

1910.1 (and Calvert, W. R.). Stratigraphic relations of the 
Livingston formation of Montana: Econ. Geol- 
ogy, vol. 5, pp. 551-557, 1 pi.; pp. 652-669; 
pp. 741-764, 1 fig., 1910. 

Thomas, Oldpield. 

1894.1. See Sclater, Philip Lutley, 1894.1. 

Veatch, Arthur Clifford. 

1907.1. Geography and geology of a portion of south- 
western Wyoming, with special reference to 
coal and oil: U. S. Geol. Survey Prof. Paper 56, 
178 pp., 26 pis., 9 figs., 1907. 

Deposits of the period between the known Cretaceous and 
the known Eocene: Evanston formation (Eocene?) (p. 86); 
"Wasatch group (pp. 87-96); Green River formation (p. 97); 
Bridger formation (p. 99). Divides Wasatch group into three 
formations, only the uppermost of which (Knight formation) 
contains vertebrate remains. 

Weed, Walter Harvey. 

1896.1. The Fort Union formation: Am. Geologist, vol. 18, 

pp. 201-211, 1896. 

Reviews early descriptions of the Fort Union "group," 
gives two sections of Fort Union strata in Montana, and dis- 
cusses physical and faunal characters. 

Weeks, Fred Boughton. 

1902.2. North American geologic formation names, bibli- 

ography, synonymy, and distribution: U. S. 
Geol. Survey Bull. 191, 448 pp., 1902. 



Weeks, Fred Boughton — Continued. 

1907.1. Stratigraphy and structure of the Uinta Range: 
Geol. Soc. America BuU., vol. 18, pp. 427-448, 
6 pis., 1907. 

Describes the occurrence and relations of pre-Cambrian, 
Paleozoic, Mesozoic, and Tertiary formations and the geologic 
structure of the region. 

Wegemann, Carroll Harvey. 

1917.1. Wasatch fossils in so-called Fort Union beds of the 
Powder River Basin, Wyo., and their bearing 
on the stratigraphy of the region: U. S. Geol. 
Survey Prof. Paper 108, pp. 57-60, pis. 22-23, 
fig. 16, 1917. 

White, Charles Abiathar. 

1876.1. See Powell, John Wesley, 1876.1. 

1878.1. Report on the geology of a portion of northwestern 

Colorado: U. S. Geol. and Geog. Survey Terr. 

Tenth Ann. Rept., pt. 1, pp. 3-60, 1 map, 1878. 

General account of Uinta formation; thickness 1,200 feet, 
resting unconformably upon other Tertiary. Refers to ex- 
posures of Bridger "group" in the Uinta Basin (p. 37). 

Willis, Bailey. 

1908.1 (and Hovey, E. O.). Symposium on correlation. 
Section E, American Association for the Ad- 
vancement of Science and Geological Society 
of America: Science, new ser., vol. 28, No. 729, 
pp. 878-879, 1908. 

Wortman, Jacob Lawson. 

1882.1. The geology of the Big Horn Basin. In Cope, 
. E. D., Contributions to the history of the Verte- 
brata * * * Qf Wyoming: Am. Philos. Soc. 
Proc, vol. 20, pp. 139-142, 1882. 
1892.1. See Osborn, Henry Fairfield, 1892.67. 
1893.1. On the divisions of the White River or Lower 
Miocene of Dakota: Am. Mus. Nat. Hist. 
Bull., vol. 5, pp. 95-105, June 27, 1893. 

Division of White River into three zones; Titanotherium, 
Oreodon, and Protoceras. 

1894.1. See Osborn, Henry Fairfield, 1894.90. 

1895.1. See Osborn, Henry Fairfield, 1895.105. 

1903.1. Studies of Eocene Mammalia in the Marsh collec- 
tion, Peabody Museum, Part II, Primates: Am. 
Jour. Sci., 4th ser., vol. 15, pp. 163-176, 399- 
414, 419-436; vol. 16, pp. 345-368, pis. 11-12; 
vol. 17, pp. 23-33, 133-140, 203-214, figs. 100- 
146, 1903-4. 
Origin of tlie primates (vol. 15, pp. 419-436). 



U. S. GEOLOGICAL SURVEY 



MONOGRAPH 55 PLATE I 




A. ERUPTION OF THE CRATER OF TAAL, PHILIP- 
PINE ISLANDS, JANUARY, 1911 




Submerged layers 



of volcanic cinders, ashes, a 
C. Worce^er.) Comparable 



B. FLOODED AREA 
I mud, in vi^hich are entombed the bodies of men and th( 
> volcanic ash deposits of Bridger age in southern Wyoming 



of animals. (After De 



■D. S. GEOLOGICAL SURVEY 



MONOGRAPH 65 PLATE II 




A. OJO ALAMO, SAN JUAN COUNTY, N. MEX., LOOKING NORTH 
Contadl (indicated by arrows) betw^een Ojo Alamo sanditone and clay of Puerco formation is observed dire(5tly back of the 

trading store 





EjJqc 



'"*^*f*^/$0'^ 








;i. EXPOSURES OF PUERCO FORMATION EAST OF OJO ALAMO, N. MEX. 
sfts on eroded surface of Ojo Alamo sandftones; contaiS: indicated by dotted line. The dark Stratum at top to the right is the lower 
level i^£tocoyiu.s 2:one) of the Puerco formation 



CRETACEOUS AND BASAL EOCENE CONTACTS IN NEW MEXICO 

Photographs by W. J. Sinclair,'1913. (After Sinclair>nd Granger, 1914.1) 



U. S. GEOLOGICAL SURVEY 



MONOGRAPH 55 PLATE III 




/ rlij-(jf }!tt^4/> X -y 








A. UPPER BEDS OF TORREJON FORMATION, WEST FORK OF TORREJON ARROYO, SANDOVAL COUNTY, N. MEX. 
The Torrejon is overlain unconformably^y the basal sand^ones of the Wasatch(?) formation 




JPotyrn-fy-stodnTi zo/ie 

•■' iliiiii if^,. . 



^^T u ' . ■ < ^}*^mmi 




B. EXPOSURES OF PUERCO FORMATION 3 MILES EAST OF OJO ALAMO, N. MEX. 

BASAL EOCENE AND LOWER EOCENE CONTACTS IN NEW MEXICO 

Photographs by W. J. Sinclair, 1913. (After Sinclair and Granger, 1914.1) 



U. S. GEOLOGICAL SURVEY 



MONOGRAPH 55 PLATE IV 



jM-'Of ** * J ^ 




A. EOHIPPUS-CORYPHODON ZONE (LOWER PART OF WASATCH FORMATION, LEVEL BIG HORN B), LITTLE SAND 

COULEE 
Fir^ appearance of Eohipptis. Am. Mus. negative 18565 




B. PHENACODUS-NOTHODECTES-CORYPHODON ZONE (BASAL PART OF WASATCH FORMATION, LEVEL BIG HORN A), 

ABOUT 4 MILES NORTH OF RALSTON 
Am. Mus. negative 18563 

LOWER WASATCH STRATA RESTING ON BASAL WASATCH STRATA, CLARK FORK BASIN, 

PARK COUNTY, WYO. 



U. S. GEOLOGICAL SURVEY 



MONOGRAPH 55 PLATE V 




A. TYPICAL "LYSITE" LOCALITY, AT COTTONWOOD DRAW, NORTH OF LOST CABIN, WIND RIVER BASIN, WYO. 

Shows the Heptodon-Coryphodon-Eohippus zone (level Wind River A), with Paleojoic hills in the background. (After Granger, 

1910.1.) Am. Mus. negative 18393 





B. TYPICAL "GRAY BULL" LOCALITY, 4 MILES SOUTH OF OTTO, BIG HORN BASIN, WYO. 

Shows the Syilemodon-Coryphodon-Eohippus sone flevcl Big Horn C), with the excavation of the skeleton of Eohippus osboi 

in the foreground. Am. Mus. negative 18450 

EXPOSURES OF WASATCH FORMATION AND TYPICAL WIND RIVER DEPOSITS IN WYOMING 



TJ. S. GEOLOGICAL SURVEY 



MONOGRAPH 55 PLATE VI 




( SiA.'Cr'f'cu-vo JJ } 



I*, 







iMtoik^'' 




■Msiti-- 



»m 



^^f 



^^avjiKKsaitM 



A. A T^'I'ICAI. HL'ERFANO LOCALITY, 2 MILES WEST OF GARDNER, HUERFANO BASIN, COLO. 

Palaeosyops fontinalis zone (level Huerfano B). A cedar-covered ridge in midale diftance, and eruptive peaks in the background. The 
type of Eometarhinus and referred specimens of Palaeosyops fontinalis were found at this site. Am. Mus. negative 104715 



J^ cirrti do ih4^riz<^n^ zorve 




imm 




B. A TYPICAI- "LOST CABIN" LOCALITY, ON ALKALI CREEK, EAST OF LOST CABIN, WIND RIVER BASIN, WYO. 

Lamhdotheriunt sone (level Wind River B). The types of Lambdotherium po^oagicum, Eotitanops horealis, and E. gregoryi -were 

found at this site. Am. Mus. negative 18392 

TYPICAL HUERFANO FORMATION OF COLORADO AND WIND RIVER FORMATION OF 

WYOMING 



V. S. GEOLOGICAL STJBVET 



MONOGRAPH 55 PLATE VII 




A. HENRYS FORK TABLE, LOOKING NORTHWARD ACROSS HENRYS FORK, BRIDGER BASIN, WYO. 

[UintatheriUTti zone (levels Bridger C and D) and ^etarhinus zone (level Bridger E) -with Bishop ("Wyoming") conglomerate at the 
top. The Burnt Fork "white layer" (w and arrow) separates level Bridger C 2 from Bridger C 3. Am. Mus. negative 18152 




B. GRIZZLY BUTTES, SOUTH OF MOUNTAIN VIEW, UINTA COUNTY, WYO. 

Palaeosyops paludosuS'Orohippiis zione (level Bridger B). Excavation by Granger of the type skull of Limnohyops priscus (Am. Mus. 

11687). Am. Mus. negative 18089 

TYPICAL BRIDGER FORMATION (LEVELS UPPER C, D, E, AND LOWER B) OF WYOMING, 

MIDDLE AND UPPER (?) EOCENE 



n. s. cKoi.onirAL survey 



MONOGRAPH 55 PLATE VIII 




A. NORTHWEST POINT OF HAYSTACK MOUNTAIN, HEAD OF BITTER CREEK, SWEETWATER COLINTY, WYO. 

Eohasileus-Dolichorhinus and Metarhinus aones (levels Washakie B 2 and B 1). (After Granger.) Am. Mus. negative 18213. (Se 

figs. 60 and 61) 



Wash-akyoe £ 



WasTtaJtie- A 




B. VIEW SOUTHEASTWARD FROM LACLEDE STATION ON OVERLAND STAGE TRAIL, SWEETWATER COUNTY, WYO. 

Lower brown sandstones of \Jiy\tat\\,^rium 2;one (level Washakie A) in middle di^ance. Hayrack Mountain and the Eobasileus' 

1 the background. (After Granger.) Am. Mus. negative 18223 



:i sandilonei 



of \Jiy\taihe.rium 5c 
lis and Metarhinus : 



TYPICAL "WASHAKIE" FORMATION (LEVELS B 2 AND B 1 OVERLYING LEVEL A 1), WYOMING; 

MIDDLE AND UPPER EOCENE 



U. S. GEOLOGICAL SURVEY 



MONOGRAPH 55 PLATE IX 




A. COLUMNAR SANDSTONES, TOP OF LEVEL XJINTA A, WHITE RIVER CANYON, UINTA BASIN, UTAH 
(After Riggs, 1912.1.) Field Mus. negative 




B. PANORAMIC VIEW, WHITE RIVER CANYON, 4 MILES BELOW WAGONHOUND BEND, UINTA BASIN, UTAH 
Bluffs on right bank of river belong to the unfossiliferous level Uinta A. Photograph by Riggs. Field Mus. negative 

MIDDLE EOCENE OF NORTHERN UTAH (LEVEL UINTA A) 



tr. S. GEOLOGICAL SURVET 



MONOGRAPH 55 PLATE X 



A 7 11 

J\.myru)do7T^ sandstoTve 




JSobcLsile-us - 

DolicJioj'Tzisn 











A. NORTHERN BOUNDARY OF COYOTE BASIN, UINTA BASIN. UTAH 

Showing greenish clays of the 'Exibas^Xe^KS'■T>6iich<yr^^iy^,us zone (level Uinta B 2) capped by '^* Amyyiodotx, sandstone." These 
clays have yielded mo^ of the smaller mammaliin fauna of this middle horison of Uinta Basin. (After Riggs, 1912.1.) 
Field Mus. negative 




B. DIVIDE BETWEEN WHITE RIVER CANYON AND COYOTE BASIN, UINTA BASIN, UTAH 
Showing fossil-bearing sand^one of the Metarhinus ?one (level Uinta B 1). (After Riggs, 1912.1.) Field Mus. negative 

UPPER EOCENE OF NORTHERN UTAH (LEVEL UINTA B) 



U. S. GEOLOGICAL SURVEY 



MONOGRAPH 55 PLATE XI 



I (fanolht^/ iinii 




^\rn\ liOiioTL (?)aj'tiLou-u^s 



A. NORTH FACE OF BEAVER DIVIDE, WIND RIVER BASIN, WYO. 
View we^^vard from point near Wagonbed Spring, showing lo\ver Oligocene beds (.Titanotherium zone), with Menodus heloceras 
(level Chadron A), reding on upper Eocene {'Diplacodon sone?). Skull provisionally referred to Amynodon antiqutis was 
taken from left foreground. (After Granger, 1910.1.) Am. Mus. negative 18388 




-^ft'v"*.^- 















B. EXPOSURES AT WAGONBED SPRING, BEAVER DIVIDE, FREMONT COUNTY, WYO. 
Sho\ving contact between upper Eocene and lower Oligocene. The skull of Menodus heloceras came from the dra-w ju^ to the right 

of this view. Am. Mus. negative 18391 

LOWER OLIGOCENE OVERLYING UPPER EOCENE OF CENTRAL WYOMING 



V. S. GEOLOGICAL SURVEY 



MONOGRAPH 55 PLATE XII 




A. CONTACT BETWEEN TITANOTHERIUM ZONE (LOWER OLIGOCENE) AND PIERRE SHALE (CRETACEOUS), NEAR 
MOUTH OF CEDAR CREEK, BIG BADLANDS, S. DAK. 

Orcodo-n zone in the di^ance to the right. Am. Mus. negative 35997 









v> 



Uirvtco C J 



Dif^^a?^ 







B. BADLANDS SOUTH OF WHITE RIVER, UTAH 

Showing Diplacodon 5one (level Uinta C 1, upper Eocene) in foreground and level Uinta C 2 in diilance- (Compare fig. 66.) Am. 

Mus. negative 17665 




101959— 29— VOL 1 



D. S. GEOLOGICAL SURVEY 



MONOGRAPH 56 PLATE XIV 




EXPOSURES AT QUINN DRAW, BIG BADLANDS, S. DAK. 

Showing summit of lower Oligocene Chadron formation (Titanotherium zone) and, at the top, base of younger Brule formation (Oreodon :;one). The 
sandstone columns in the center indicate a river channel betvveen underlying and overlying claya. Am. Mus. negative 36012 



V. S. GEOLOGICAL SURVEY 



MONOGRAPH 55 PLATE XV 




A. SOUTH END OF SHEEP MOUNTAIN, NEAR HEAD OF CORRAL DRAW, BIG BADLANDS, S. DAK. 
Showing Oreodon zone (Brule formation). Am. Mus- negative 36006 




B. CEDAR CREEK, BIG BADLANDS, S. DAK. 
Showing Oreodon zone (Brule formation) overlying Titanotherium zone (Chadron formation). Am. Mus. negative 36013 

BRULE AND CHADRON FORMATIONS OF SOUTH DAKOTA 



CHAPTER III 
DISCOVERY OF THE TITANOTHERES AND ORIGINAL DESCRIPTIONS OF THE TYPES 



SECTION 1. HISTORY OF DISCOVERY 

Full descriptions of the geologic and geographic 
positions of the several types and kinds of titano- 
theres are given in Chapter II. The present chapter 
relates the history of the explorations and of the 
gradual discovery of the character and relations of the 
titanotheres. 

THE OLIGOCENE TITANOTHERES 

THE PIONEER PERIOD: PROUT, OWEN, EVANS, LEIDY 
(1846-1873) 

The Big Badlands of South Dakota and north- 
western Nebraska are even now practically unknown 
to most Americans. As these lands lie in an arid 
region far from navigable rivers — a region that was 
formerly occupied by hostile Indians and that offers 
little attraction to either the prospector or the set- 
tler — it is not surprising that their fossil wonders long 
lay hidden from the world. The fossil remains of the 
great animals described in this monograph were known 
to the Indians and referred to in their mythology as 
"thunder horses." (See Preface, p. xxi.) 

In 1846 Dr. Hiram A. Prout, of St. Louis, sent to 
Professors Dana and Silliman of Yale College a cast 
of a remarkable fossil that he had received from "a 
friend residing at one of the trading posts of the St. 
Louis Fur Co. on the Missouri River." Front's brief 
notes, together with a crude sketch of one of the lower 
molars, were accordingly published in the American 
Journal of Science and Arts. (Prout, 1846.1, pp. 
288, 289.) In a later communication Prout (1847.1) 
stated that this fossil (fig. 85) was discovered in the 
"Mauvais Terre, on the White River, one of the west- 
ern confluents of the Missouri." This was the famous 
specimen described by Prout as a "gigantic Palaeo- 
therium," which Leidy tells us (1852.1, p. 551) was 
"the first of the many mammalian remains which have 
been brought to the notice of the scientific world from 
the vast Eocene cemetery of Nebraska." It thus gave 
the first hint to scientists that "the region of Nebraska 
Territory of the United States appears to be as rich in 
the remains of Mammalia and Chelonia of the Eocene 
period as the deposits of the same age of the Paris 
Basin." (Leidy, 1852.1, p. 539.) 

The fossil jaw described by Prout represented an 
animal of great size. "The entire jawbone," he says, 
"must have been at least 30 inches long, which far 
exceeds in size the PalaeotJierium magnum." The 
reference to Cuvier's PalaeotJierium was, under the 
circumstances, very natural, because the lower molars 
of Front's specimen were surmounted by crescentic 
cutting surfaces somewhat like those of Palaeotherium. 



This discovery evidently attracted attention abroad, 
for in 1849 the French paleontologist Pomel (1849.1, 
pp. 73-75), after carefully considering Prout's descrip- 
tion and figures, stated that the fossil represented a 
new subgenus of paleotheres, for which he proposed 
the name Menodus giganteus, the generic name re- 
ferring to the crescents of the lower molars, the specific 
name to the great size of the animal. 

Meanwhile (in 1839, 1840-1849) the United States 
Government geologist. Dr. David Dale Owen, was 
making his extensive geologic reconnaissance of Wis- 
consin, Iowa, and adjacent States. In his final report 
(Owen, 1852.1, p. 194) he tells us that he was "de- 
sirous, if possible, to connect the geology of the Missis- 
sippi Valley, through Iowa, with the Cretaceous and 
Tertiary formations of the upper Missouri, a matter 
very important to the proper understanding of the 
formations of the intervening country, which it had 
been made my particular duty to explore." Finding 
it impracticable to explore the Missouri region himself 
he detailed to this work one of his assistants, Mr. John 
Evans. Late in the field season of 1849 Evans "finally 
reached that most curious unexplored region, the corner 
of the 'Badlands' (Mauvaises Terres), lying high up 
on White River, a locality which seemed likely, above 
all others, to furnish satisfactory information regard- 
ing the precise character and age of the Tertiary de- 
posits of the upper Missouri country." (Owen, 1852.1, 
p. 195.) 

From Evans's report (p. 197) Owen gives the fol- 
lowing description of the Mauvaises Terres of White 
River: 

To the surrounding country, however, the Mauvaises Terres 
present the most striking contrast. From the uniform, monoto- 
nous open prairie, the traveler suddenly descends, one or two 
hundred feet, into a valley that looks as if it liad sunk away 
from the surrounding world, leaving standing, all over it, 
thousands of abrupt, irregular, prismatic, and columnar masses, 
frequently capped with irregular pyramids and stretching up 
to a height of from one to two hundred feet or more. 

So thickly are these natural towers studded over the surface 
of this extraordinary region tliat the traveler threads his way 
through deep, confined, labyrinthine passages, not unlike the 
narrow, irregular streets and lanes of some quaint old town of 
the European continent. Viewed in the distance, indeed, these 
rocky piles, in their endless succession, assume the appearance 
of massive artificial structures, decked out with all the acces- 
sories of buttress and turret, arched doorway and clustered 
shaft, pinnacle and finial, and tapering spire. 

One might almost imagine oneself approaching some magnifi- 
cent city of the dead, where the labor and the genius of for- 
gotten nations had left behind them a multitude of monuments 
of art and skill. 

On descending from the heights, however, and proceeding 
to thread this vast labyrinth and inspect, in detail, its deep, 
intricate recesses, the realities of the scene soon dissipate the 

141 



142 



TITANOTHERES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



delusions of the distance. Tlie castellated forms which fancy 
had conjured up have vanished, and around one, on every 
side, is bleak and barren desolation. 

Then, too, if the exploration be made in midsummer, the 
scorching rays of the sun, pouring down in the hundred defiles 
that conduct the wayfarer through this pathless waste, are 
reflected back from ' the white or ash-colored walls that- rise 
around, unmitigated by a breath of air or the shelter of a soli- 
tary shrub. 

The drooping spirits of the scorched geologist are not per- 
mitted, however, to flag. The fossil treasures of the way well 
repay its sultriness and fatigue. At every step objects of the 
highest interest present themselves. Embedded in the debris 
lie strewn, in the greatest profusion, organic relics of extinct 
animals. All speak of a vast fresh-water deposit of the early 
Tertiary period and disclose the former existence of most re- 
markable races that roamed about in bygone ages high up in 



characters belonging now to the above three orders; for the 
molar teeth are constructed after the model of those of the 
hog, peccary, and babyroussa; the canines as in the bear; 
while the upper part of the skull, the cheek bones, and the 
temporal fossa assume the form and dimensions which belong 
to the cat tribe. Another, the Oreodon of Leidy, has grinding 
teeth like the elk and deer, with canines resembling the omnivo- 
rous thick-skinned animals, being, in fact, a race which lived 
both on flesh and vegetables and yet chewed the cud like our 
cloven-footed grazers. 

Associated with these extinct races we behold also, in the 
Mauvaises Terres, abundant remains of fossil Pachydermata of 
gigantic dimensions and allied in their anatomy to that sin- 
gular family of proboscidate animals of which the tapir may be 
taken as a living type. These form a connecting link between 
the tapir and the rhinoceros; while, in the structure of their 
grinders, they are intermediate between the daman and rhinoc- 





FiGUKE 83 — Mauvaises Terres, Nebraska. After David Dale On en, 1851 



the valley of the Missouri, toward the sources of its western 
tributaries, where now pastures the big-horned Ovis montana, 
the shaggy buffalo or American bison, and the elegant and 
slenderly constructed antelope. 

Owen continues (p. 198) with a popular description 
of the extinct animals found: 

Every specimen as yet brought from the Badlands proves to 
be of species that became exterminated before the mammoth 
and mastodon lived and differ in their specific character, not 
alone from all living animals, but also from all fossils obtained 
even from cotemporaneous geological formations elsewhere. 

Along with a single existing genus, the Rhinoceros, many new 
genera never before known to science have been discovered, 
and some, to us at this day, anomalous families, which com- 
bine in their anatomy structures now found only in different 
orders. They form, indeed, connecting links between the 
pachyderms, plantigrades, and digitigrades. For example, in 
one of the specimens from this strange locality, described by 
Dr. Leidy under the name Archiotherium, we find united 



eros; by their canines and incisors, they connect the tapir with 
the horse, on the one hand, and with the peccary and hog on 
the other. They belong to the same genus of which the labors 
of the great Cuvier first disclosed the history, under the name 
of Palaeotherium, in publishing his description of the fossil bones 
exhumed from the gypsum quarries of Montmartre, near Paris, 
but are of distinct species; and one at least, of this genus, dis- 
covered in the Badlands (Palaeotherium proutii), must have 
attained a much larger size than any which the Paris Basin 
afforded. In a green, argillo-calcareous, indurated stratum, 
situated within 10 feet of the base of the section, a jaw of this 
species was found, measuring, as it lay in its matrix, 5 feet 
along the range of the teeth, but in such a friable condition, 
that only a portion of it could be dislodged; and this, notwith- 
standing all the precautions used in packing and transportation, 
fell to pieces before reaching Indiana. 

A nearly entire skeleton of the same animal was discovered, 
in a similar position, which measured, as it lay embedded, 18 
feet in length, and 9 feet in height. But here, as in the former 
case, the crumbling condition of the bones rendered it impos- 



DISCOVERY OF THE TITANOTHERES AND ORIGINAL DESCRIPTIONS 



sible to disinter them whole; and the means of transportation to 
the Missouri were insufficient, even if these interesting remains 
could have been extracted in good condition. 




Figure 84. — "Vertical view of the posterior tooth belonging to the lower jaw of 
Mr. Prout's Palaeoiherium; natural size." After Prout, 1846 

Owen also gives (1852.1, p. 200) a tabular "Section 
of beds constituting the early Tertiary (Eocene) of 
the Badlands." This section, reproduced below, was 
doubtless taken by Evans. 



143 

The detailed description of the mammalian and 
chelonian fossils collected by Owen, Evans, and others 
was intrusted to Dr. Joseph Leidy, of Philadelphia, 
and was published in Owen's report 
of 1852 (1852.1, pp. 551, 552), 
already cited. In this publication 
Leidy describes Prout's original 
specimen and gives a poor figure 
of Evans's principal specimen, an 
imperfect lower jaw of a titanothere. 
He adopts provisionally the 
name Palaeotherium? proutii Owen, 
Norwood, and Evans but con- 
cludes his description of these fos- 
sils with the following significant 
remark : 



AU the preceding specimens, except, probablj', the latter two 
(fragments of upper molars), I suspect belong to a different 
genus from either Palaeotherium or A nchitherium, and should the 
suspicion prove correct, Titanotherium would be a good name 
for the animal, as expressive of its very great size. 



Section of beds constituting the early Tertiary {Eocene) 
of the Badlands (Mauvaises Terres) 



30 



[Numbered in descending order] 

1. Ash-colored clay, cracking in the sun; con- 

taining siliceous concretions 

2. Compact white limestone 

3. Light-gray marly limestone 

4. Light-gray indurated siliceous clay (not effer- 

vescent) 

5. Aggregate of small angular grains of quartz, 

or conglomerate, cemented by calcareous 
earth; slightly effervescent 8 

6. Layer of quartz and chalcedony (probably 

only partial) 1 

7. Light-gray indurated siliceous clay, similar 

to No. 4 but more calcareous, passing down- 
ward into pale flesh-colored indurated 
siliceous marly limestone (effervescent) ; 
turtle and bone bed 25 

8. White and light-gray calcareous grit; slightly 

effervescent 15 

9. Similar aggregate to No. 5 but coarser 8 

10. Light-green indurated argillaceous stratum 

(slightly effervescent) ; ' ' palaeotherian bed" - 20 

Some of the specimens brought back by 
Evans were referred to in a brief notice pub- 
lished by Owen, Norwood, and Evans (1850.1), 
in which the name "Palaeotherium proutii" 
was proposed "in compliment to Dr. Prout, 
of St. Louis." 

The next year (1850) after Evans's journey 
Mr. Thaddeus A. Culbertson visited, under 
the auspices of the Smithsonian Institution, 
the same region (Leidy, 1854.1, p. 12) 
and "made a good collection of its animal 
remains," including skulls of Oreodon culbertsoni 
and the titanothere upper premolars which Leidy 
afterward described (1852.2, p. 2) under the names 
Rhinoceros americanus and Eotherium americanus. 
The locality was Bear Creek, a dry tributary of 
Cheyenne Kiver. (See Chap. II, p. 104.) 





Figure 85. — Original figures of Prout's "gigantic Palaeotherium," the 
first titanothere discovered. After Prout, 1847 

A, "Fragment of the inferior maxillary of the left side," one-fourth natural size; B, last lower 
molar, left side, four-fifths natural size. 

Thus was proposed the name Titanotherium, which 
has figured so largely in the literature of American 
paleontology and was consequently chosen as the basis 
for the title of this monograph. 

Two years later Leidy (1854.1) gave a revised and 
extended description of the available remains of titano- 



144 



TITANOTHERES OP ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



theres, which was accompanied by excellent litho- 
graphs of Prout's and other fragmentary specimens. 
At this stage of knowledge the only parts figured 
under the name Titanotherium proutii included the 
lower molars, a lower premolar, a lower canine, a frag- 
mentary upper molar, and two upper premolars 
(Leidy's types of Rhinoceros americanus). Fragments 
of large upper molars were named Palaeotherium 
giganteum. 

The "palaeotherian bed" of Owen and Evans is 
referred to by Leidy (op. cit., p. 13) as the "Titano- 
therium bed." This appears to be the first use of this 
term, which was afterward, in the form " Titano- 
therium beds" (now Titanotherium zone), so widely 
used by geologists and paleontologists. 

An interval of 15 years in the literature of the sub- 
ject, from 1854 to 1869, is broken only by Prout's 
brief notice of an indeterminate molar (now lost) of a 
titanothere, which he named Leidy otherium. But 
during this seemingly barren interval Meek and 
Hayden were making their historical explorations 
(Merrill, 1906.1, pp. 585-592), which resulted in 
notable advances in our knowledge of the relations 
of the geologic deposits of the Rocky Mountains and 
Great Plains. They also brought back many verte- 
brate fossils, including specimens of Titanotherium. 

One of the specimens of titanotheres collected by 
Meek and Hayden included a nearly complete series 
of upper teeth. This specimen, which belonged to 
Prof. James Hall and is now No. 433 of the Hall 
collection of the American Museum of Natural 
History, was described and figured by Leidy in his 
memoir of 1869 (1869.1, pp. 206, 207, pi. 24) and was 
by far the best spec'men that had been described up 
to that time. Leidy referred it to his species Titano- 
therium proutii, but it probably belongs in the genus 
that Marsh afterward named Brontotherium. This 
specimen misled Leidy into assigning Titanotherium 
to the Artiodactyla. "From the form of its lower 
true molars, which were first discovered," he says, 
"it was supposed to be more nearly alHed with the 
Palaeotherium and was hence placed among the uneven- 
toed pachyderms, or Perissodactyla, but the nearly 
complete dentition of both jaws, since discovered, 
appears to indicate its position to be as above stated " — 
that is, it appeared to be "nearly allied with Chali- 
cotherium, and, like it, approximates the even-toed 
pachyderms, or Artiodactyla * * * with the Ru- 
minantia." 

In 1870 Leidy (1870.1, pp. 1, 2) described a frag- 
mentary fossil from Colorado that had been submitted 
to him by Doctor Hayden. We now know that this 
specimen consists of the horn cores and attached 
coossified nasal bones of a titanothere of some sort, 
but to Leidy, who knew practically nothing of the 
skull of the titanothere, it proved "singularly puzzling 
in character." He at first thought it might pertain 



to Titanotherium, "but in the state of extreme uncer- 
tainty as to its collocation, it may with equal proba- 
bility be referred to other genera, perhaps to Megalo- 
meryx, or it may have been an American species of 
Sivatherium. Under the circumstances it may be 
referred to a new genus, with the name of Megacerops 
color adensis ." 

This problematical fossil was redescribed and figured 
by Leidy in his memoir of 1873 (1873.1, p. 239). He 
states that the specimen "appears to correspond 
with that portion of the face * * * [of Siva- 
therium] which comprises the upper part of the nose, 
together with the forehead and the anterior horn 
cores." He compares the specimen with the corres- 
ponding parts of the Sivatherium, the rhinoceros, the 
tapir, and the mastodon. He decides that the frag- 
mentary horn core formerly attributed to Titano- 
therium may perhaps belong to another species of 
Megacerops. 

This erroneous determination, together with the 
previous assignment of Titanotherium to the Artio- 
dactyla, shows how greatly Leidy, even with all his 
skill and caution, was deceived by the lack of well- 
preserved and definitely associated feet and skulls, 
a lack which is felt to some extent even at the present 
time. 

Leidy's description of Megacerops may be regarded 
as marking the close of the first or pioneer period in 
the study of the titanotheres, a period characterized 
by (1) the chance discovery of "Prout's specimen," 
(2) the exploration of the White River badlands 
by Evans, Hayden, and others and the resulting 
knowledge of the general geologic age of the beds, (3) 
the description of fragmentary remains of titanotheres, 
chiefly teeth, by Prout, by Pomel, and by Leidy in 
successive publications, together with the beginnings 
of the systematic nomenclature, (4) the erroneous 
reference of Titanotherium to the Anoplotheriidae 
among the Artiodactyla. 

TAXONOMIC ARRANGEMENT AND COMPARISON 
WORK OF MARSH AND COPE (1870-1887) 

The second period in the study of titanotheres, 
which may be called the period of systematic descrip- 
tion, really began before the first period had closed 
(1873). 

From 1873 to 1891, inclusive, the literature of the 
Oligocene titanotheres is dominated almost exclu- 
sively by the explorations and systematic contribu- 
tions of Marsh and Cope. During this time Marsh 
described eight genera and fourteen species as new, 
and Cope described three genera and twelve species 
as new. The solution of the exact systematic and 
phylogenetic interrelations of these genera and species 
is one of the principal themes of Chapters IV to VII of 
the present monograph. 

In 1870 Prof. Othniel C. Marsh (1870.1) headed an 
expedition sent from Yale College to northern Colo- 



DISCOVERY OF THE TITANOTHERES AND ORIGINAL DESCRIPTIONS 



145 



rado, where he not only discovered and explored "an 
extensive outcrop of the true Mauvaises Terres, or 
White River formation," but also procured some mag- 
nificent specimens of titanotheres (including the types 
of Brontotherium gigas and B. ingens), which he de- 
scribed and figured three years later. Marsh was also 
able to solve the problem of the ordinal relationships 
of the titanotheres (1873.1, p. 486), showing that his 
Brontotherium gigas was a "true perissodactyl with 
limb bones resembling those of Rhinoceros. The genus 
is related to Titanotherium, and the two appear to form 
a distinct family, which may be called Brontotheridae." 
He was able in a very few words to throw a flood of 
light upon the characters of the skeleton, hitherto 
known chiefly from fragments : 

It closely resembles that in recent perissodactj^ls but shows 
some approach to the Proboscidea. The femur has a third 
trochanter, and its head a pit for the round ligament. The 
fibula is entire and slender. The astragalus is remarkably 
short. It has a deep groove on its upper surface, and the 
articular facets for the navicular and cuboid are nearly equal. 
In the manus there are four toes of nearly equal size, the first 
digits being rudimentary or wanting. There were three digits 
only in the pes, the first and fifth being entirely wanting. The 
toes were short and thick, as in proboscidians. The meta- 
carpals and metatarsals are longer than in the elephant, and 
the phalanges shorter. The foot was also more inclined. The 
carpal and tarsal bones are very short and form interlocking 
series. The tail was long and slender. 

An important point not touched upon in this com- 
munication was the presence or absence of horns. 

Prof. Edward D. Cope was not far behind Marsh in 
contributions to the literature of the titanotheres. 
Two years after Marsh had made his explorations in 
Colorado, Cope, in 1872, discovered a number of re- 
markable skulls (now in the Cope collection of the 
American Museum of Natural History) which, in bul- 
letins pubhshed in 1873 and 1874, he made the types 
of Symhorodon torvus, Megaceratops acer, M. heloceras, 
Symborodon hucco, S. dltirosfris, S. trigonoceras. He 
states (1873.2, pp. 2, 3) that "Leidy and Marsh have 
described two genera of this group, viz, Titanotherium 
and Brontotherium, but without certain indications of 
their possession of horns." He regards them as "all 
true perissodactyls and allied to the Rhinoceros and 
Palaeotherium." His genus Symhorodon, like Menodus, 
Titanotherium, and Brontotherium, was "established 
on mandibular rami only, which can not be certainly 
associated with crania," the last phrase suggesting one 
of the most troublesome and obdurate of titanothere 
problems, which from the first has caused confusion in 
the systematic nomenclature. Cope regarded the 
absence of incisors as one of the generic characters 
that separated Symhorodon from Titanotherium and 
Brontotherium, thus first raising the problem how far 
differences in the number of incisors may correspond 
to true generic differences. The discovery of so many 
more or less complete skulls enabled Cope to infer 



specific and generic characters from the variations in 
form of the horn cores, skull top, nasals, and zygo- 
matic arches. Thus the discoveries of Cope and 
Marsh, although they settled the ordinal relationships 
of the titanotheres, began to complicate the problem 
of their interrelationships. 

SUMMARY OF MAESH'S CONTRIBTTTIOKS 

In Marsh's paper "On the structure and affinities 
of the Brontotheridae" (1874.1) he developed further 
the family characters of the group, separating them 
from the Rhinocerotidae, "apparently their near 
allies," establishing the number of digits in the fore 
and hind feet and the general characters of the skull, 
lower jaw, vertebrae, and limbs. This paper is ac- 
companied by the first of a series of excellent litho- 
graphic plates, illustrating some of Professor Marsh's 
superb specimens of titanotheres from Colorado. 
Marsh contributed another short but pregnant article 
on the "Principal characters of the Brontotheridae" 
in 1876 (1876.1), and after that he published at in- 
tervals brief descriptions of supposedly new genera 
and species, not all of them accompanied by illustra- 
tions, until September, 1891, the date of his last 
published contribution to the subject. 

Marsh's most valuable contributions to our knowl- 
edge of the titanotheres may be summarized as follows: 
(1) He and his party explored the White River forma- 
tion in Colorado and collected from it many remark- 
ably fine specimens; (2) he demonstrated the ordinal 
position of the group, classifying its members as 
perissodactyls; (3) he recognized the fact that the 
titanotheres constitute a distinct family, which he 
named the Brontotheridae; (4) he made the illuminat- 
ing observation that his upper Eocene genus Diplaco- 
don served to connect the Oligocene Brontotheridae 
with the Eocene "Limnohyidae"; (5) he published 
many excellent lithographs and woodcuts, showing 
chiefly the skulls and dentition of titanotheres, but 
including also (1889) an excellent restoration of 
Brontops rolustus; (6) he supervised the preparation 
of a fine series of lithographic plates for the present 
work; (7) under the auspices of the United States 
Geological Survey he founded the present series of 
monographs on fossil vertebrates; (8) he began the 
preparation of the present monograph, although he 
left no manuscript for it; (9) he obtained for the 
National and Yale Museums their superb specimens of 
titanotheres, most of which were collected by his field 
assistant J. B. Hatcher, who in turn also made valuable 
scientific contributions to our knowledge of these 
animals. 

Marsh's detailed systematic work on the titanotheres 
was less fortunate than his broader contributions, 
owing chiefly to confusion in regard to features of the 
skull and jaw. After founding the genus and species 
Brontotherium gigas upon a lower jaw, he referred to the 



146 



TITANOTHERES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



same genus as the type of B. ingens, a skull that 
certainly belongs to another genus (Menodus). In 
consequence of this initial confusion he erected a new 
genus (Titanops) for skulls that should have been 
referred to Brontotherium. Many of his conceptions 
of the interrelations of the genera and species proposed 
by him were erroneous. Although recognizing the 
fact that the genera Brontops, Allops, and Teleodus 
were all allied to " BrontotTierium" — that is, as repre- 
sented by the skull of "BrontotTierium [Menodus] 
ingens" — he nevertheless thought thsLt Diplodonus was 
related to Titanops (the true BrontotTierium), and he 
referred to Menops (a near ally of his "BrontotTierium'' 
ingens) a well-preserved skull that is now known to 
belong to BrontotTierium proper. In fact, in common 
with Cope and others. Marsh apparently faUed to 
recognize the comparatively wide phyletic gap between 
the true BrontotTierium (his Titanops) and Cope's 
Symborodon on the one hand and the supposed genera 
Brontops, Allops, Menops, and Menodus (his "Bronto- 
tTierium ingens") on the other. Consequently his 
generic definitions are unsatisfactory, and he was 
certainly not overconservative in proposing new 
generic and specific terms. 

SUMMARY OF COPE'S CONTRIBUTIONS 

The next year (1874) after publishing his prelimi- 
nary descriptions of the several species of Symborodon 
and allied genera Cope (1874.2) gave full descriptions 
of these forms in his "Report on the vertebrate pale- 
ontology of Colorado," which was accompanied by 
eight lithographic plates. He presented a careful 
review of the general morphology of the skull, includ- 
ing the brain case and cranial antra, which was fol- 
lowed by a review of the work of preceding authors and 
by a tabular analysis and detailed description of the 
species of Symborodon. He recorded many interesting 
facts, such as the similarity of the dentition of Sym- 
borodon to that of Palaeosyops and of CTialicotTierium 
and the mingling of proboscidian and rhinoceros 
analogies in the limbs. He considered the indications 
that Symborodon possessed a short proboscis. In his 
tabular analysis of species he indicated the differences 
in the shape of the horns and noted that in S. trigono- 
ceras and S. Tiypoceras the upper premolars have a 
strong internal basal cingulum, whereas in S. bucco 
and S. altirostris the premolars are "without inner 
basal cingulum." 

Cope, like Marsh, failed to distinguish the sexes as 
well as the separate groups or phyla of titanotheres. 
His "S." trigonoceras , for example, is a Menodus, a 
member of an altogether different group from his "S." 
Tiypoceras, which is a BrontotTierium. 

After an interval of 12 years, in 1886, Cope (1886.1) 
described the first Canadian species, Menodus angusti- 
genis, basing it upon fragments discovered by Mc- 
Connell and Weston for the Geological and Natural 
History Survey of Canada. Three years later (1889.1, 



p. 153) he referred this form to a new genus, Hapla- 
codon, and in the same year (1889.2, pp. 628, 629) he 
described two other Canadian species. His review 
(1891.2, p. 17) of these forms and attempted revision 
of the nomenclature were involved and unsatisfactory. 
He recognized only two genera, Menodus and Symbo- 
rodon. The last species of titanothere described by 
him was his Menodus peltoceras (1891.1), which is 
probably a female of Marsh's BrontotTierium curtum. 

EEINTERPRETATION AND PHYLOGENETIC STUDY 

(OSBOEN, 1887-1919) 

STUDY OF CERTAIN FEATURES 

Before Marsh and Cope had ceased naming new or 
supposedly new genera of titanotheres a turn was 
given to the trend of study by a paper by Scott and 
Osborn (1887.1, pp. 157, 158), entitled "Preliminary 
account of the fossil mammals from the White River 
formation contained in the Museum of Comparative 
Zoology." This paper, which was a description of the 
interesting collection made by Mr. Samuel Garman 
under the auspices of Prof. Alexander Agassiz, reacted 
from the polynomial systems of Marsh and Cope and 
tended toward a mononomial system. In this paper 
the Perissodactyla were described by Osborn, the 
Artiodactyla and Carnivora by Scott. Before de- 
scribing the new titanothere material the authors 
noted the difficulty in deciding where to draw generic 
lines, a difficulty that is increased by the fact that the 
mandibles are seldom found with the skulls. 

As in Uintatherium, the variability in the various portions of 
the skull, especially in the region of the horns, is so extreme that 
no two skulls are found which are exactly alike. But the denti- 
tion, which is constant among the Dinocerata, here greatly com- 
plicates the problems of classification. The premolars vary in 
number, and the incisors, always of relatively small size and 
fairly constant in number in the upper jaw, vary from three to 
none in the lower jaw." In all the lower jaws found in Professor 
Cope's collection of Menodontidae from northern Colorado there 
are no incisors, and the mandibular symphj'sis is extremely 
narrow. In the lower jaws of the Cambridge and Princeton 
collections, which are all from the Nebraska and Dakota 
exposures, the symphysis is broad, and the incisors, where pre- 
served, are two in number, while in one of the Cambridge 
specimens no less than three incisor alveoli may be counted 
upon one side of the symphysis. 

We might infer from this that Symborodon can be clearly 
separated from Menodus by the absence of the lower incisors, 
accompanied by a narrowing of the symphysis; but Professor 
Cope has recently described a new species, M. angustigenis, 
from the Swift Current Creek region (Cope, 1886.1, p. 81c), 
which combines the narrow type of symphysis with the presence 
of two incisors. The separation of these genera is rendered 
still more improbable by the parallelism which exists between 
the skulls from the Nebraska and Colorado localities, especially 
in respect to the conformation of the nasal bones and the 
horns. The genus Symborodon is, however, provisionally 
adopted at present to include the species with a narrow man- 
dibular symphysis and no lower incisors. 

The genus Brontotherium Marsh (that is. Marsh's "Bronto- 
therium" ingens, not the true Brontotherium) can not be dis- 
tinguished from Menodus. It rests in part upon the premolar 

I' One of the Cambridge skulls, M. coloradensis, has but a single upper incisor. 



DISCOVERY OF THE TITANOTHERES AND ORIGINAL DESCRIPTIONS 



147 



formula |^, in the synopsis given by Professor Marsh (1876.1, 
p. 339), as distinguished from Menodus, with ?pm J^. One 
of the lower jaws of the Princeton collection, however, has the 
premolar formula 57^, demonstrating that the first lower pre- 
molar is a variable tooth and can not in this case be used in 
classification. The same rule applies to the second cone upon 
the last upper molar, the supposed generic character of Dicono- 
don Marsh. This is found in different species in all degrees 
of development, from a small prominence upon the basal cin- 
gulum to a well-developed cone {M. proutii) . 

From this evidence Osborn draws the following 
partly erroneous conclusion: 

Such characters as the invariable absence of lower incisors 
may subsequently be found to separate one genus of the Meno- 
dontidae from another; but our present evidence goes to show 
that they simply characterize the extremes of a closely related 
series of animals, from the same horizon, of which the inter- 
mediate forms are represented by numerous species. The 
safest basis of specific determination seems to be the correla- 
tion between the developnaent and proportion of the horns and 
of the nasals, the rule being that where the horns are long the 
nasals are short, and conversely. The number of the teeth 
does not at present seem to be absolutely constant, even within 
the limits of the species. 

The following determination of the species in the Cambridge 
collection is, for the above and other obvious reasons, provi- 
sional. The classification can be finally settled only when the 
lower jaws and skulls are found in association. 

Thus the validity of the several genera recognized 
by Marsh and Cope and of the chief criteria used by 
them as generic characters was called in question. 
The species are treated as belonging mainly to the 
single genus Menodus. Taking up the description of 
the new material, the authors mistakenly refer to 
Leidy's Megacerops coloradensis, a well-preserved 
skull, which at present is referred to Alhps marshi. 
They then describe two new species — "Menodus" 
tichoceras, based on a skull, and "Menodus" platyceras, 
based on a pair of bony horns. Both these species 
are at present referred to the true Brontotherium or 
flat-horned genus. The authors conclude their dis- 
cussion of the "Menodontidae" by presenting the 
first published restoration of the skeleton, made up 
of material in several museums, forming a composite 
animal representing Menodus proutii. In connection 
with a table of measurements arranged to show pro- 
gressive and correlated changes in the horns and nasals, 
they make the following remarks (op. cit., p. 16): 

The above measurements bring out very clearly the decrease 
in the proportions of the nasals pari passu with the gradual 
elongation of the horns. Another very interesting fact is 
brought out by the comparison of the transverse and longitudi- 
nal diameters of the horns at the base. As we pass from the 
short to the long horned types, through M. coloradensis, ticho- 
ceras, doUchoceras, and platyceras, there is a gradual rotation of 
the longer axis of the horn section from a fore and aft to a trans- 
verse plane, the species last named representing the extreme of 
the transverse type. 

The fuller development and more or less radical 
modification of the hypotheses put forward in this 
paper have been the subject of successive contribu- 
tions by Osborn, culminating in the present work. 



GEOLOGIC lEVEIS AND SUCCESSION OF TYPES (HATCHEE, 1886-1893) 

The work of Marsh and Cope had been exclusively 
descriptive and systematic. Osborn's observation of 
the correlated progressive reduction of the nasals 
and the enlargement and flattening of the horns 
seems to have been the first important application of 
evolutionary principles to tlie study of the Oligocene 
titanotheres. But materials for an exact knowledge 
of the phyletic succession, resting securely upon a 
knowledge of the precise geologic levels of a large 
series of specimens, had hitherto been entirely lacking. 
This all-important element of the time relations of 
the different species was largely supplied by the 
labors and study of J. B. Hatcher. In 1886, 1887, and 
1888 Hatcher spent 15 months in the White River 
beds of South Dakota and Nebraska, collecting 
material for Professor Marsh's monograph on the 
Titanotheridae. In an interesting article in the 
American Naturalist for March, 1893, Hatcher (1893.1, 
pp. 214, 215) tells us that he collected or purchased 
"nearly 200 complete skulls and many more or less 
complete skeletons," a part of which are now on 
exhibition in the National and Yale Museums. The 
superb Hatcher collection in the United States 
National Museum contains skulls and jaws of 157 
individuals; as completely listed in the generic 
sections of this monograph, it furnishes the classic 
standard of reference. Hatcher writes: 

Early in the season of 1886 it became apparent that certain 
forms of skulls were characteristic of certain horizons in the 
beds. This fact showed the importance of keeping, so far as 
possible, an exact record of the horizon from which each skull 
or skeleton was taken. From actual measurement the vertical 
range of the Titanotheridae was found to be about 180 feet. 
For convenience in keeping a record of horizons the beds were 
divided into three divisions of 60 feet each, and each of these 
three divisions was subdivided into three divisions of 20 feet 
each. The difi'erent skulls and skeletons, when dug out, were 
each given a separate letter or number, and this letter or num- 
ber was placed in that subdivision of the beds from which the 
skull or skeleton was taken. 

At present about 60 of these skulls and several more or less 
complete skeletons have been freed from their matrix. When 
studied in connection with the horizons from which they were 
taken, these remains show that a regular and systematic 
development took place in these animals from the base to the 
top of the beds. The most noticeable change which took place 
in the Titanotheridae was a gradual and decided increase in 
their size from the lowest to the uppermost beds, as is shown 
by the increase in the size of the skulls, fore and hind limbs, and 
other portions of the skeleton. Individuals found near the 
bottom of the beds are little, if any, larger than the living 
rhinoceros. From this they gradually increase in size as we go 
up until at the top we find a type described by Professor Marsh 
as Titanops, rivaling the modern elephant in size. 

This increase in size from the base to the summit of the beds 
was attended by a very marked development in certain portions 
of the skeleton, noticeable among which are the following: 
A variation in shape and an increase in the size and length of 
the horn cores as compared with the size of the skulls, attended, 
near the summit of the beds at least, by a decided shortening 
of the nasals. 



148 



TITAJSfOTHERES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



Hatcher was less fortunate in his observations on the 
evolutionary changes in the dentition, stating that 
"the number of incisors, though probably never 
constant, even in the same species, shows a tendency 
to decrease in skulls found near the summit of the 
beds," and concluding that "the number of incisors 
can hardly be considered as of either generic or specific 
importance in the Titanotheridae, where they are no 
longer functional and vary with individuals in the 
same species and with age in the same individual. 
The same may be said of the presence or absence of 
the first premolar." 

After noting certain other changes rightly believed 
by him to be progressive, such as (1) the loss of the 
trapezium, (2) the development of a postero-internal 
cone on the third upper molar, and (3) the flattening of 
the horns, Hatcher concludes his paper by giving a 
tabular paleontologic section of the " Titanotlierium 
beds," with a general description of the forms char- 
acterizing the three ascending divisions. As to the 
number of genera, he gives the impression that he 
•regards all the various species ascribed by Marsh and 
Cope to different genera (except Teleodus avus Marsh) 
as referable to the single highly variable genus 
Titanotherium Leidy. 

FIRST EUEOPEAn NOTICE (TOTJIA, 1892) 

The next important event is the discovery of a 
titanothere of Oligocene type in Europe, described as 
Menodus rumelicus by Toula (1892.1). This dis- 
covery, in connection with that of the Transylvanian 
BracJiydiastematJierium, described by Bockh and 
Maty in 1876 (1876.1), extended the known range of 
the titanotheres to the Old World. 

DISTINCTIONS OF SEX (OSBOEN AND WOETMAN, 1895) 

In 1895 Osborn and J. L. Wortman (1895.105) 
published a corrected restoration of Titanotlierium 
based upon the fine skeletal material secured by the 
American Museum field parties in 1892 and 1894. They 
ventured the conclusion that "it is probable that 
certain wide differences in the development of the 
horns, which have been assigned a generic value, 
are merely sexual characters. " 

MONOPHYIETIC INTEEPEETATION (OSBOEN, 1898) 

The extreme development of the erroneous theor}' 
that all the various species of Oligocene titanotheres 
belonged to the single genus Titanotherium and were 
practically monophyletic is worked out in a very 
elaborate way in Osborn's paper "The cranial evolu- 
tion of Titanotlierium," published in 1896. This 
'was the most comprehensive review of the subject 
that had hitherto appeared and was illustrated by 
numerous text figures and several folded plates. 
The direct observations were based chiefly on the 
large collection of titanotheres in the American 
Museum and to a less extent upon figures and descrip- 



tions previously published. Part I, the systematic 
introduction, includes a chronologic list of generic 
and specific terms, with references and a brief history 
of the progressive complication of the nomenclature, 
after which the author says (Osborn, 1896.110, p. 
162): 

It is obvious that the only method of clearing up this hetero- 
geneous list [of nominal genera and species] is first to establish 
certain laws of cranial development, and second to apply 
these laws to the distinction of genera and species in chrono- 
logical order. Examined in this way, the vast array of genera 
and species is resolved into one or possibly two genera and 
about fourteen definable species. 

Accordingly in Part II, "Principles of cranial and 
dental evolution," we find a study of the differences 
in size of skull, shape of horns, nasals, zygomatic 
arches, auditory meatus, cingula on grinding teeth, 
incisors, canines, second internal cone of last molar, 
etc., aU considered as indicating either specific or 
sexual or individual differences within the limits 
of a single genus, Titanotlierium. This is followed by 
the "Revision and definition of species," in which 
some 27 species, including the new T. ramosum, are 
discussed. The known species from the lower, middle, 
and upper beds are arranged in a single or monophy- 
letic series, beginning with the T. heloceras-trigonoceras 
ingens series, continuing with torvum, rohusfum, 
doliclioceras, elatum, amplum, acer, and culminating 
with ramosum and platyceras. 

This analysis, although wholly wrong in treating 
all the species as members of a monophyletic series, 
not only laid the foundation for the present evolution- 
ary and phylogenetic treatment of the group but 
established, as it were, the technique of investigation 

POIYPHYIETIC INTEEPEETATION (OSBOEN, 1902-1919) 

The reaction against the monophyletic theory was 
felt by the same author as a result of more extended 
research. In his paper of 1902 on "The four phyla 
of Oligocene titanotheres, " after acknowledging the 
services of the late Professor Marsh and admitting the 
incorrectness of the monophyletic theory, Osborn 
says (1902.208, p. 91): 

This second review is an abstract of a portion of the results 
obtained for the United States Geological Survey monograph 
"The titanotheres," now in preparation. It covers practically 
aU the type material in the Yale, National, American, and 
Harvard Museums, and advantage has been taken of the 
invaluable field observations by Hatcher of the levels on which 
the different skulls in the National Museum collection were 
discovered. The section method also has been very greatly 
extended and, taken in connection with the teeth and the 
detailed structure of the skull, has proved to be a sure criterion 
of specific and phjdetic character. 

Four important considerations had led Osborn to 
give up the monophyletic theory: First, from his 
phylogenetic studies on the rhinoceroses of Europe 
and America (Osborn, 1898.143; 1900.192) he had 
concluded that, contrary to earlier opinions, this 



DISCOVERY OF THE TtTANOTHERES AND ORIGINAL DESCRIPTIONS 



149 



group was in a high degree polyphletic, embracing 
many parallel phyla and having a wide adaptive 
radiation; second, the principle of dolichocephaly and 
brachycephaly (Osborn, 1902.207), as interpreted by 
him in the rhinoceroses and other groups, raised the 
presumption that similar differences would be found 
to distinguish genera and phyla among the titano- 
theres; third, he had learned to realize that the extent 
to which parallel and convergent evolution had oper- 
ated in many allied phyla had been but little appre- 
ciated by earlier writers, who had largely failed also 
to distinguish between persistent, progressive, and 
retrogressive characters; fourth, an examination of 
the titanothere skulls collected by Hatcher and now 
in the National Museum, which Hatcher had recorded 
exactly as to level, enabled him, with the aid of prin- 
ciples just stated, to distinguish several distinct phyla 
and to foUow them from the lower part through the 
middle and into the very top of the " Titanotherium 
beds." The characters of these phyla were summar- 
ized by Osborn as follows (1902.208, pp. 92-94): 

THE FOUR GENERA 

Titanotherium Leidy applies to .long-limbed animals with long 
skulls, persistently long and broad nasals, short triangular 
horns placed slightly in front of the e3'es, vestigial incisors ^o' 
large canine teeth. Known from the base to the summit of the 
[lower] Oligocene. 

Megacerops Leidy applies to titanotheres with broad skulls, 
nasals progressively shortening, short horns rounded or oval 
in section, shifting anteriorly, one or two pairs of incisor teeth, 
.|r}, medium-sized canine teeth. Known from the base to the 
summit of the [lower] Oligocene. 

Probably related to this are the subgenera of the t3'pes named 
Allops and Diploclonus by Marsh, differing from the above in 
horn characters. Known chiefly from the upper beds. 

Symborodon Cope includes titanotheres with skulls of varying 
proportion, nasals slender and progressively shortening, horns 
elongate and peculiar in being placed above the eyes instead of 
shifting forward, incisors vestigial l^, canines small, approx- 
imated. Known only from the middle and upper beds. 

Brontotherium iVlarsh embraces the largest titanotheres, with 
very broad zygomatic arches, nasals shortening while horns 
elongate and shift forward; incisors persistent, f in the males, 
canines stout and obtuse. 

Representatives of Titanotherium and Megacerops can be now 
continuously traced from the base to the summit of the [lower] 
Oligocene. Primitive species of Brontotherium also appear at 
the base, although the phyletio sequence through the middle 
to the upper beds is not so clear. Symborodon suddenly appears 
in the middle beds. 

Viewed in the light of the foregoing principles, the 
variations in the horns, nasals, incisors, cingula, etc., 
took on new meanings — biologic, phylogenetic, and 
systematic; so that, after more than half a centm-y of 
research (1846-1902) the systematic problem presented 
by the Oligocene titanotheres appeared in its main 
features to be solved. Subsequent research, however, 
has led to certain regrettable but apparently necessary 
changes in nomenclature: (a) The name " Megacerops" 
Leidy, as defined above, has been set aside for Brontops 
IVIarsh, for the reasons given below; (b) the name 



Titanotherium Leidy has been abandoned for the prior 
name Menodus Pomel; (c) the name Symhorodon Cope 
has been replaced by the prior name Megacerops 
Leidy. 

RECENT DISCOVERIES BY LULI, lAMBE, AND OTHERS 

There remain to be recorded the following contri- 
butions: (1) The description of Megacerops tyleri by 
Lull (1905.1), based upon a fine skull and lower jaws 
with associated limbs, discovered by the Amherst 
CoUege paleontologic expedition of 1903; (2) the 
description of Brontotherium hatcheri and Symhorodon 
copei by Osborn in 1908 (1908.318), based on skulls 
in the National ]Museum; (3) the description of Mega- 
cerops primitivus and M. assinihoiensis, based on frag- 
ments obtained from Saskatchewan, Canada, by 
Lambe in 1908 (1908.1); (4) W. K. Gregory observed 
(a) that there is an alliance between Brontops, Allops, 
and Menodus as these terms are now used by Osborn, 
indicated by certain intermediate forms between the 
extremely brachycephalic Brontops roiustus and the 
dolichocephalic Menodus giganteus, (b) that there is 
also an alliance between Brontotherium and Megacerops 
(Symhorodon) in spite of the differences in the incisors. 
Hence the former group — Brontops, Allops, Menodus — ■ 
has been called the menodontine group, and the latter 
group — Brontotherium, Megacerops — has been called 
the brontotheriine group. 

Possibly the most valuable general result of the 
study of the titanotheres has been the fact that it 
has made possible the close examination of an extensive 
evolutionary history, stretching from the lower Eocene 
to the summit of the lower Oligocene. IVTany observa- 
tions have been made on the precise modes of evolu- 
tion, especially with regard to the way in which char- 
acters first appear and subsequently develop. The 
results of this evolutionary study are set forth in 
Chapters V, VI, VH, and XI of the present work. 

THE EOCENE TITANOTHERES 

PIONEER DISCOVERIES 

WORK IN THE BRIDGER, WASHAKIE, AND UINTA BASINS BY lEIDY, 
MARSH, COPE, SCOTT, OSBORN, AND OTHERS (1870-1889) 

Prof. F. V. Hayden, in the course of his historic 
explorations in the fossiliferous beds of the Kocky 
IMountains and Great Plains, obtained at Church 
Buttes, near Fort Bridger, Wyo., a number of isolated 
teeth, which were described by Leidy (1870.2) under 
the name Palaeosyops paludosus. This was the first 
Eocene titanothere made laiown to science, 24 years 
after the discovery of Prout's "gigantic Palaeotherium" 
(Titanotherium) in South Dakota. Although Leidy 
noted that the lower molar of Palaeosyops "resembles 
in its constitution those of Palaeotherium, (Jlialicothe- 
rium, and Titanotherium," he did not classify the new 
genus with the titanotheres, for the reason that at 
that time he thought Titanotherium and Chalicotherium 
were allied to the Artiodactyla. (See p. 247.) Soon 



150 



TITANOTHERES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



afterward Leidy (1873.1, p. 27) described a skull, some 
teeth, and parts of the limb bones of Palaeosyops and, 
noting the similarities of this species to its supposed 
allies Tapirus and Palaeoiherium, correctly referred it 
to the perissodactyls — the odd-toed pachyderms. 
Three other species (P. major, P. Tiumilis, P. Junius) 
were also described by him from the Bridger beds 
upon very fragmentary material. 

In developing our knowledge of the Eocene titano- 
theres of the Bridger Basin, as in developing that of 
the Oligocene titanotheres. Marsh and Cope were not 
far behind Leidy. The first specimen of an Eocene 
titanothere described by Marsh, however (1871.2), 
was not recognized as such by him, as he mistook the 
isolated second lower premolar of a Bridger Palaeosyops 
for the fourth upper premolar of a dog and named it 
"Canis montanus." The next year (1872.1) he de- 
scribed some well-preserved remains under the name 
Palaeosyops laticeps and also founded the genus 
Telmatherium. Marsh's subsequent contributions to 
our knowledge of middle Eocene titanotheres were not 
especially significant, but in 1875 he described the 
very important genus Diplacodon from the upper 
Eocene Uinta beds of Utah and recognized its inter- 
mediate position both in time and in structural 
characters between his "Limnohyidae" (Palaeosyo- 
pinae) and Brontotherium. 

Cope's explorations of the Bitter Creek or Wash- 
akie Basin (middle Eocene) of Wyoming m 1872 led 
to his describing the species "Palaeosyops" vallidens 
and "Limnoliyus" laevidens, both represented by 
imperfect remains. The former is now known to 
belong to the long-headed genus Dolichorfiinus. 

The next year, 1873, Cope (1873.5) described the 
species Limnoliyus ( = Palaeosyops) fonfinalis from the 
lower levels of the Bridger formation (supposed 
Bridger A), which is the oldest middle Eocene titano- 
there yet discovered. His Palaeosyops diaconus, 
from the upper levels of the Bridger Basin, is prob- 
ably a synonym of Palaeosyops rohustus (Marsh). 

DISCOVERY IN HUNGAEY 

Shortly after these pioneer discoveries in America 
Bockh and Maty (1876.1) described a large lower 
jaw from Eocene deposits in Transylvania, in Hun- 
gary. The animal was supposed to be allied to 
Palaeoiherium and was named Brachydiastematherium 
transilvanicum. Its affinities with the Palaeosyops 
group long remained unnoted, and even to this day 
it is the only known specimen of its kind in Europe. 

PEINCETOM AND COPE-WOETMAN EXPEDITIONS 

The Princeton expeditions sent to the Bridger and 
Washakie Basins in 1877 and 1878 under Scott, 
Osborn, and Speir brought to light much valuable 
material of Palaeosyops and allied genera, especially 
the types of "Leurocephalus" cultridens and the pecu- 
liar form which was later described by Earle as 



Palaeosyops megarMnus. Thus by the end of 1878 re- 
mains of the genus Palaeosyops and its allies had 
been discovered in the middle Eocene Bridger and 
Washakie Basins and in the upper Eocene Uinta 
Basin. 

The next year (1879) Dr. J. L. Wortman, who was 
collecting for Cope, extended the kiiown range of the 
group into the lower Eocene Wind River formation of 
Wyoming, where he discovered the very primitive 
form which Cope in 1880 named Palaeosyops horealis 
and which is now recognized as approximately ances- 
tral to the middle Eocene titanotheres. Wortman 
also discovered a very small form, which was described 
by Cope in 1880 (1880.1) as Lamhdotherium popo- 
agicum and recognized as more or less closely allied to 
the Palaeosyops group. 

The next important expedition was that made by a 
Princeton party under Scott and Speir in 1886 into 
the Uinta Basin (upper Eocene). They collected 
skeletal material, referred at that time to Diplacodon, 
which was described by Osborn in 1890 (1890.51) and 
which demonstrated the intermediate characters of 
"Diplacodon" {Protitanoiherium) between the Oligo- 
cene and middle Eocene titanotheres. In the same 
publication Osborn also described "Palaeosyops" 
hyognathus, a species based on a jaw that is now known 
to represent the long-skulled genus DolichorMnus. 

FIRST SYSTEMATIC AND EVOLUTIONARY REVISION 
(EARLE, 1889-1891) 

Although Cope in 1884 (1885.1) had republished 
and partly extended the original descriptions of his 
own species, with lithographic figures, no satisfactory 
revision of the Palaeosyops group was possible at that 
time or for many years later. 

In 1889 Charles Earle, at the invitation of Prof. 
H. F. Osborn, began a careful study of the material 
in the Princeton Museum and other collections, and 
in 1892 he published a memoir "On the genus Palaeo- 
syops Leidy and its allies" (1892.1). Earle gave a 
very detailed description of the osteology of Palaeo- 
syops and of the first attempted reconstruction of the 
skeleton of an Eocene titanothere by Osborn. (See 
fig. 86.) Owing in part to the lack of sufficient well- 
associated material, in part to the confusing practice 
of the earlier writers in designating and founding 
species upon several specimens of doubtful specific 
association, Earle's revision of the species and genera 
was, as he himself recognized, by no means final. He 
rightly regarded as distinct the genera Lamhdotherium, 
Limnohyops, Palaeosyops, and "Telmatotherium," but 
as he showed in his tentative phylogenetic scheme, 
he, like other paleontologists at that time, did not 
appreciate the polyphyletic character of groups and 
consequently referred to a single main line of descent 
a number of forms that belong to widely separated 
phyla. 



DISCOVERY OF THE TITANOTHERES AND ORIGINAL DESCRIPTIONS 



151 



AMERICAN MUSEUM AND OTHER EXPLORATIONS OF THE 
EOCENE BASINS (1891-1895) 

The problems relating to Palaeosyops and its 
allies, which had been barely made evident by the 
pioneer discoveries and had now been partly formu- 
lated by Earle, were of course only particular results 
of the general explorations of the fossil-bearing forma- 
tions of the West. The early explorations had been in 
part reconnaissances, and their results were accordingly 
incomplete as regards both the nature of the material 
and the records of the stratigraphic levels at which the 
specimens were found, both absolutely prerequisite to 
a detailed knowledge of the phylogeny. 



exhibit a mounted composite skeleton of this animal. 
Much other material was also collected by the same 
party. All this material has been used profitably in 
the present monograph, especially the specimens 
representing the "prophet-horn stage," to which 
Doctor Wortman in a letter from the field applied the 
name Manteoceras. 

Another American Museum expedition, under Mr. 
O. A. Peterson, went into the Uinta Basin in 1894 
and examined two hitherto unexplored horizons (Uinta 
B 2 and Uinta B 1 of this monograph), which underlie 
the true Uinta (Uinta C). This expedition collected 
many new forms and worked out the faunal sequence 
of the three horizons indicated. Among the results 




Figure 86. — Osborn's first restoration of Palaeosyops paludosus Leidy 
This restoration is a composite one— the skull from the fine specimens in the Academy of Natural Sciences of Philadelphia, and the axial 



skeleton from the material in the Princeton Museum, 
twelfth natural size. 



The fore feet were afterward referred to Mesaiirhinus peterst 



The founding (in 1890) of the department of verte- 
brate paleontology in the American Museum of 
Natural History by Prof. Henry Fairfield Osborn and 
the consequent establishment of continuous and syste- 
matic exploration began a new era of exact investiga- 
tion not only of the titanotheres but of the whole 
series of vertebrate remains to be found in the Rocky 
Mountains and Great Plains regions, as well as the 
stratigraphic horizons at which they occur. 

The first of these expeditions, led by Dr. J. L. 
Wortman, procured some important skeletal material 
of "Palaeosyops" horealis from the Wind River forma- 
tion. Another expedition, sent out under Doctor 
Wortman in 1893, procured from the Bridger and 
Washakie Basins extensive material of the true 
Palaeosyops, enabling the American Museum to 



of this expedition, as reported in 1895, were the discus- 
sion by Osborn and Peterson (Osborn, 1895.98) of the 
three faunal levels (Uinta B 1, B 2, and C) and the 
description by Osborn of the specialized and interest- 
ing titanotheres named " TelmatotJierium" diploconum 
and T. cornutum. Wortman's "prophet-horn" skulls 
were referred to " Telmatotherium vallidens," so that 
animals showing a wide range of form were here 
erroneously included under a single genus. The 
very aberrant form SpJienocoelus was also described, 
but its ordinal and family positions were left "Incertae 
sedis," on account of the lack of the teeth in the type 
and the peculiar characters of the base of the skull. 

In the same year (1894) Mr. J. B. Hatcher, of the 
Princeton Museum, also went into the true Uinta 
area and discovered specimens representing the very 



152 



TITANOTHERES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



advanced stage which in 1895 (1895.1) he described 
as Diplacodon ejnarginatus. In a brief postscript 
to this description he noted the wide range of forms 
that had been erroneously grouped by Osborn under 
the genus " TelmatotTierium," and he formally proposed 
the generic names Manteoceras for the "prophet- 
horn" and Dolichorhinus for the long-skulled form. 




Figure 87. — Four stages in the origin and evolution of tlie 
horns in titanotheres 

After Hatcher's original plate (1895). A, Palaeosyops laiiceps (= Limnohyops lali- 
ceps), after Earle; B, Telmatotherium vallidens ( = Manteoceras manteoceras), after Os- 
born; C, Diplacodon emarginatus (,= PTOtitanotherium emarginatum), after Hatcher; 
D, Titanotherium varians (= Bronioihermin leidyi), after Marsh. One-eighth 
natural size. 

Both these terms, as well as the name Protitanotherium 
of Hatcher, have proved to be valid. Hatcher's separa- 
tion of these genera was a very important move toward 
a correct understanding of their phylogenetic rela- 
tions. He also figured a series of four stages ranging 
from the middle Eocene to the lower Oligocene, inclu- 
sive, showing the origin of the "horns." (See fig. 87.) 



INVESTIGATIONS AND EXPLORATIONS MADE IN PREPARA- 
TION FOR THE PRESENT MONOGRAPH (1900-1919) 

Between 1895 and 1900 no very important work on 
the Eocene titanotheres was done. By the end of the 
nineteenth century some 12 nominal genera and 25 
nominal species of the titanotheres had been proposed, 
but many of the real generic limits and phyloge- 
netic relations were still obscure except for the preg- 
nant suggestions of Hatcher. In 1900 Professor 
Osborn, at the invitation of Director Charles D. 
Walcott, undertook to revise and monograph the 
Eocene titanotheres in connection with the United 
States Geological Survey monograph on the Oligocene 
titanotheres that had been begun by Professor Marsh. 
The work on the Eocene titanotheres has proved to be 
by far the most difficult and most extensive part of 
this task. During the last 28 years Professor Osborn, 
with the assistance of Dr. W. K. Gregory, has studied 
the great and growing collection in the American 
Museum of Natural History and in other institutions 
and has set forth the results in several prelimmary 
articles and more fully in the present work. 

A long series of parties of exploration, beginnLng in 
1903, sent out from the American Museum by Osborn 
(1909.321) and conducted chiefly by Mr. Walter 
Granger, have carefully examined the various lower, 
middle, and upper Eocene basins of the West with 
special reference to the exact succession of species. 
This very precise work has shown that the Bridger 
and other formations are divided into a succession of 
zonal levels characterized by the remains of titano- 
theres and other mammals in different generic and 
specific stages of evolution. The stratigraphic rela- 
tions of the Eocene to the Oligocene deposits have also 
been in part explored. The results are fully set forth 
in this monograph. Although this work in the 
Eocene basins has been carried on chiefly by the Ameri- 
can Museum of Natural History, the Carnegie and 
Field Museums have sent expeditions into the Uinta 
Basin under Douglass (1909.1) and under Riggs 
(1912.1), which have yielded similar results as to 
specific and generic succession. 

The distinction of numerous independent Eocene 
phyla by Osborn has followed the discovery of the 
Oligocene phyla, some of which arise from those of the 
Eocene. 

Thus have been established secure bases of fact, 
first, for a general history of the early Tertiary faunas 
of the West; second, for a demonstration of the evolu- 
tion of certain phyla of titanotheres through long 
periods of time; and, third, for a consideration of the 
modes and factors of evolution of titanotheres in par- 
ticular and of mammals in general. 



DISCOVERY OF THE TITANOTHERES AND ORIGINAL DESCRIPTIONS 



153 



SECTION 2. ORIGINAL DESCRIPTIONS OF EOCENE 
TITANOTHERES 

FIVE RULES FOE DETERMINING THE NAMES OF 
TITANOTHERES 

The systematic revision of the Eocene and Oligo- 
cene titanotheres was begun in 1900 by the author 
with the cooperation of W. K. Gregory and has been 
continued up to the day of the deUvery of the manu- 
script of this volume to the Geological Survey. The 
great difficulties and the labor involved in determin- 
ing the correct prior names for the genera and species 
have been due to the imperfection of the fossil types, 
to loose methods of description and comparison, and 
to the mingling as cotypes of animals belonging to dif- 
ferent species or even to different genera. 

Experience has shown that the following five rules 
are absolutely necessary for future vertebrate paleon- 
tologic work. 

Rule 1. Accept the "law of priority," as defined by the 
International Committee on Nomenclature. 

In this revision the author has accepted as authori- 
tative the rules of nomenclature based upon the "law 
of priority," as defined by the "Code" of the Ameri- 
can Ornithologists' Union and by the recommenda- 
tions of the committee on nomenclature of the Twelfth 
International Congress of Zoology. Special acknowl- 
egments a,re due to the eminent authority Dr. J. A. 
Allen for frequent aid in deciding troublesome prob- 
lems of nomenclature. 

Rule 2. Determine the geologic level and characters of the 
type, as the starting point of monographic inquiry. 

Experience teaches that the characters of the holo- 
type specimen and the geologic level on which it was 
found afford the permanent facts to which all questions 
of nomenclature must be referred as the basis of mono- 
graphic investigation. 

Rule 3. Avoid confusion of characters of th& type and cotype 
or paratype. 

All the early systematic work on the titanotheres 
was done without regard to precise discrimination 
between the certain or permanent nomenclatural 
value of the holotype specimen and the uncertain 
value of "specific" characters based on cotype, para- 
type, and neotype specimens. 

For example, take the case of the classic species 
Palaeosyops paludosus Leidy. Leidy used as types 
the very fragmentary teeth from the lower levels of 
Bridger B, which first came into his hands; he later 
erroneously associated with these fragments, practi- 
cally as cotypes, other more complete specimens, which 
are now known to belong to two or three different 
species from higher geologic levels. Subsequently 
Leidy himself. Cope, Marsh, Scott, Osborn, and Earle 
all accepted Leidy's erroneous associations, and P. 
paludosus came to be known by certain of its falsely 
associated cotype and paratype characters instead of 
by its true type characters. 

101959— 29— VOL 1 13 



Thus the entire nomenclature of the subject became 
a mass of confusion, and the difficulties encountered in 
clearing it up have been almost insuperable. 

The rule is that specific definitions must be based on 
holotypes only, unless there is absolutely no possibility 
of doubt that the associated types are from the same 
geologic level and belong to the same species. 

Rule 4. Distinguish the different values and kinds of types. 

The use of the terms type (or holotype), cotype, 
paratype, lectotype, hypotype, neotype has been dis- 
cussed critically by Oldfield Thomas (1893.1, p. 241), 
by Schuchert (1905.1, pp. 9-14), and by Osborn 
(1918.473). The distinctions indicated below should 
be noted. 

Type, individual, or holotype. — -A holotype is a 
particular individual specimen "deliberately selected 
by the author of a species; or it may be the only 
example of a species known at the time of original 
publication. A holotype, therefore, is always a single 
individual but may embrace one or more parts, as the 
skin, skeleton, or other portions." (Schuchert, op. 
cit.) The holotype must usually be determined from 
the original description. 

Cotype, coordinate or equivalent type. — The term co- 
type is applied to specimens when an author's type 
description refers to remains of two or more individuals 
without selecting or distinguishing one as the holotype, 
so that all appear to be equally identified with the 
specific name given. 

Lectotype. — "Where the origina' diagnosis is with- 
out illustrations or is accompanied by figures based on 
two or more specimens, the first subsequent author is 
at liberty to select from these cotypes a type for the 
old species, adhering, so far as can be ascertained, to 
the intention of the original author. Such a type 
specimen is to be designated a lectotype ( = a chosen 
type)." (Schuchert, idem.) The practice of Osborn 
as to lectotypes in paleontology is either (a) to select 
the first individual specimen named by the original 
author, because the second individual specimen may 
belong to a distinct species, or (&) to select the speci- 
men to which the specific name obviously refers — 
for example, Cope's Menodus angustigenis. 

Hypotype and plesiotype. — As shown by Schuchert 
(idem), the terms hypotype and plesiotype have been 
used in two different senses to cover "supplementary 
types." They may well be dropped. 

Neotype. — A neotype is defined by Schuchert (idem) 
as a [new] "supplementary type selected by an [a sub- 
sequent] author, on which a species is to rest because 
of the loss of the original type, or where the original 
material still extant is so poor or fragmentary that from 
it the characters of the species can not be determined 
with certainty." Great care must be taken that the 
neotype comes from the same geologic level as the 
type. 



154 



TITANOTHERES OF ANCIENT "WYOMING, DAKOTA, AND NEBRASKA 



Summary. — The usage adopted in this monograph 

is as follows: 

Holotype (of original autlior) : The original individual type 
specimen selected by the author. 

Cotypes: Different individual specimens rightly or wrongly put 
together by the author as "types." 

Paratype (of original author) : Additional individual specimen 
or specimens noted by the author in the original description 
and used by him in defining the species. 

Lectotype (of subsequent author) : The specimen selected by 
a subsequent author, from among the "cotypes," for pur- 
poses of subsequent description or redefinition. This may 
be (a) the specimen first mentioned by the author, or (6) the 
specimen to which the specific name obviously applies. 

Neotype (of second or subsequent author) : A new specimen 
selected in a subsequent description because of the loss or im- 
perfection of the holotype or type. 

These five terms are all that are necessary in verte- 
brate paleontology. The terms plesiotype and hypo- 
type are discarded in this monograph because they are 
too indefinite. 

Monographic revision in the use of above terms. — 
Leidy founded the species Palaeosyops paludosus 
upon some isolated teeth from the low levels of 
Church Buttes. In the original description these 
teeth, which probably represent more than one in- 
dividual, were treated as coordinate or equivalent 
types or "cotypes." Out of this lot the second 
lower molar (m2), which was the first specimen men- 
tioned and described by Leidy, has been selected by 
Osborn in the present volume as the final standard, 
or "lectotype," of the species. 

In the same original description by Leidy of P. 
paludosus a second lot of teeth, from the high levels 
of Henrys Fork, were mentioned, and the characters, 
of these teeth entered into Leidy's original conception 
of the species. These teeth are now called "paratypes." 

In the present revision, since there is little doubt 
that Leidy's paratypes are not really conspecific with 
the specimen first mentioned (lectotype), Osborn 
has selected from the same geologic level. Church 
Buttes, a lower jaw in which m2 agrees most clearly 
with the lectotype m2 and which is to serve as a 
secondary type, or "neotype." 

It wUl be seen that cotypes, paratypes, or neotypes 
may sometimes be wrongly associated specifically 



with the holotype, in which case the specific name 
must cling to the holotype and lectotype as the 
ultimate standard means of identification. 

The first step toward permanence, therefore, is the 
settlement of the holotype characters, which is some- 
times an almost impossible task, owing to the poor 
quality of the holotype selected — for example, the 
holotype of Palaeosyops major Leidy, a jaw fragment 
without teeth; the holotype of P. humilis Leidy, a 
single deciduous premolar. 

Rule 5. Avoid mingling as types and cotypes specimens from 
different geologic levels. 

The mingling of types and cotypes from different 
geologic levels has been the second chief source of 
confusion. To cite a prominent instance. Cope's 
cotypes of Palaeosyops laevidens were two skulls col- 
lected at widely separated localities, and in his original 
description no regard was shown for their possible 
difl'erence of geologic age. It appears almost certain 
that the lectotype belongs to a lower level and is 
perhaps some thousands of years more ancient than 
the paratype. Similarly we have shown that the 
lectotype of Leidy's P. paludosus is from Bridger 
level B 1 or B 2; the paratypes are from level C 2 or 
C 3, a difference of geologic level representing a very 
long period of time, in which it is now certain that a 
very marked progressive evolution took place in teeth, 
skull, and skeleton. 

Our geologic leveling of the Bridger formation, 
described in Chapter II, has therefore not only 
afforded us the means of determining the evolutionary 
succession of the species of titanotheres but, if the 
localities of the types were properly recorded by the 
authors, it has enabled us to separate many er- 
roneously associated type specimens. The geologic 
levels of the materials recently acquired by the 
American Museum have been ascertained precisely; 
on the whole, the successive species correspond very 
closely with the successive levels — that is, in no case 
have different species in the same line of descent been 
found at the same level, although species in different 
lines of descent (that is, in different genera) are found 
in analogous stages of evolution. 



DISCOVERY OF THE TITANOTHERES AND ORIGINAL DESCRIPTIONS 
THE GENERA AND SPECIES OF EOCENE TITANOTHERES 



155 



The accompanying list shows, in chronologic order, 
the names assigned to Eocene titanotheres. The 
numbers in the first column indicate the chronologic 



order or rank of the systematic names, the roman 
numerals indicating generic names, the arable numerals 
specific names. 



Chronologic list of original descriptions of Eocene titanotheres 

[Generic names accepted in tliis work as valid are printed In small capitals; abandoned names are inclosed in brackets] 



I 
1 
2 
3 
4 
6 
6 
II 
7 
Ilia 
8 
9 
Illb 
10 
11 
12 
IV 
13 
V 

14 

VI 

15 

16 
VII 
17 
18 
19 
VIII 
20 
21 
22 
23 
24 
IX 
25 
X, 
26 

XI 
XII 

27 
28 
29 

30 
31 
XIII 
32 
33 
34 
35 
36 



1870 
1870 
1871 
1871 
1872 
1872 
1872 
1872 
1872 
1872 
1872 
1872 
1872 
1873 
1873 
1873 
1875 
1875 
1876 

1876 

1878 

1878 

1880 
1880 
1880 
1881 
1889 
1890 
1891 
1891 
1892 
1895 
1895 
1895 
1895 
1895 
1895 

1895 
1895 
1897 
1899 
1899 

1899 
1902 
1907 
1907 
1908 
1908 
1908 
1908 



Palaeosyops 

Palaeosy ops 

Palaeosyops 

[Canis] 

Palaeosyops 

[Palaeosyops] 

[Palaeosyops] 

Telmatherium 

Telmatheriu m 

[Limnohyus] 

[Limnohyus] 

[Palaeosyops] 

[Limnohyus] 

[Limnohyus] 

[Limnohyus] 

Palaeosyops 

DiPLACODON 

Diplaoodon 

Bkachydiastbmathebium 

Brachydiastematherium. 



[Leurocephalus]_ 
[Leurocephalu s] . 



paludosus.. 

major 

[montanus]_ 

[hu mills] 

Junius 

laticeps 



validus. 



robustus. 
vaUidens _ 



laevidens-. 
fontinalis. . 
[diaconus]. 



elatus_ 



transilvanioum. 



cultridens. 



[Palaeosyops] borealis 

Lambdotherium 

Lambdotherium 

[Lambdotherium] 

[Palaeosyops] 

LiMNOHYOPS 

[Palaeosyops] 

[Palaeosyops] 

Palaeosyops longirostris 

[Telmatotherium] diplooonum_ 

[Telmatotherium] [cornutum]_ 

Sphenocoelus 

Sphenoooelus uintensis 

PrOTITANOTHERIUM ': 

[Diplaoodon] emarginatus 



megarhinus. 
[minor] 



Manteoceras 

DoLICHORHINtrS- - 

[Palaeosyops] 

[Palaeosyops] 

[Telmatotherium], 



[Canis?] 

Manteoceras 

EOTITANOPS 

Lambdotheriu m_ 

Limnohyops 

Limnohy ops 

Limnohyops 

Palaeosyops 



Present determination 



Leidy.- 
do. 



do. 

Marsh. 
Leidy. . 
do. 



Marsh. 

do. 

do. 



do 

do 

Cope . 

Leidy (not Marsh). 

Cope 

do 



do 

Marsh 

do 

Bockh and Maty. 

do 



Osborn, Scott, and 

Spelr. 
do 



Cope. 



.do. 



popoagicum do__ 

brownianum do_. 

hyognathus Osborn.. 

Marsh. - 

Earle 

do._ 

do_. 

Osborn.. 

do_- 

do.. 

do.- 

Hatcher. 
do.. 



ultimus 

manteoceras 

[diploconum var. 
minus. 

[marshii] 

manteoceras 



pnmaevum. 

priscus 

matthewi 

monoconus. 



.! leidyi. 



do... 

do... 

Matthew. 

do... 

de- 



Hay. __ 

do. 

Osborn. 
Loomis. 
Osborn. 
do. 



.do. 
.do. 



Palaeosyops Leidy. 
Palaeosyops paludosus Leidy. 
Palaeosyops major Leidy. 
Palaeosyops major? Leidy. 
Palaeosyops sp. 
Mesatirhinus Junius (Leidy). 
Limnohyops laticeps (Marsh). 
Telmatherium Marsh. 
Telmatherium validum Marsh. 
Palaeosyops Leidy. 
Palaeosyops robustus (Marsh). 
Dolichorhinus vallidens (Cope). 
(Preoccupied.) 

Limnohyops laevidens (Cope). 
?Palaeosyops fontinalis (Cope). 
Palaeosyops robustus (Marsh). 
Diplaoodon Marsh. 
Diplaoodon elatus Marsh. 
Brachydiastematherium Bockh and 

Maty. 
Brachydiastematherium transilvani- 

cum Bockh and Maty. 
Telmatherium Marsh. 

Telmatherium cultridens (Osborn, 

Scott, and Speir). 
Eotitanops borealis (Cope). 
Lambdotherium Cope. 
Lambdotherium popoagicum Cope. 
Eotitanops brownianus (Cope). 
Dolichorhinus hyognathus (Osborn) . 
Limnohyops Marsh. 
Mesatirhinus megarhinus (Earle). 
Palaeosyops paludosus Leidy. 
Palaeosyops longirostris Earle. 
Rhadinorhinus diploconus (Osborn). 
Dolichorhinus hyognathus (Osborn) . 
Sphenocoelus Osborn. 
Sphenocoelus uintensis Osborn. 
Protitanotherium Hatcher. 
Protitanotherium emarginatum 

Hatcher. 
Manteoceras Hatcher. 
Dolichorhinus Hatcher. 
Telmatherium ultimum Osborn. 
Manteoceras manteoceras Hay. 
Metarhinus fluviatilis Osborn. 

Palaeosyops major? Leidy. 
Manteoceras manteoceras Hay. 
Eotitanops Osborn. 
Lambdotherium primaevum Loomis. 
Limnohyops priscus Osborn. 
Limnohyops matthewi Osborn. 
Limnohyops monoconus Osborn. 
Palaeosyops leidyi Osborn. 



156 



TITAJSrOTHERES OF ANCIENT "WYOMING, DAKOTA, AND NEBRASKA 

Chronologic list oj original descriptions oj Eocene titanotheres — Continued 

[Generic names accepted in this work as valid are printed in small capitals; abandoned names are inclosed in brackets] 



Bank Date 



37 
38 
39 

XIV 
40 
• XV 
41 
42 
43 
44 
45 
46 
47 
48 
49 
50 

XVI 
51 
52 
53 
54 
XVII 
55 
56 
57 
58 
59 
60 
XVIII 
61 

XIX 
62 
XX 
63 
64 
65 

XXI 
66 



1908 
1908 
1908 
1908 
1908 
1908 
1908 
1908 
1908 
1908 
1908 
1908 
1909 
1909 
1909 
1909 
1912 
1912 
1912 
1912 
1912 
1912 
1912 
1913 
1913 
1913 
1913 
1913 
1914 
1914 
1914 
1914 
1914 
1916 
1919 
1919 
1919 
1919 



grangeri 

copei 

washakiensis. 



petersoni. 



fluviatilis 

earlei 

intermedius. 

ultimum 

altidens 

superbum_ _ 

incisivum 

uintensis 

heterodon__ 
longiceps 



Palaeosyops 

Palaeosy ops 

Manteoceras 

Mesatirhinus 

Mesatirhinus 

Metarhinus 

Metarhinus 

Metarhinus 

Dolichorhinu s 

Telmatheriu m 

Telmatherium? 

Protitanotheriu m 

[Telmatherium?] 

Manteoceras 

Dolichorhinus 

DoUohorhinus 

Sthenodectes 

[Mesatirhinus] 

Metarhinus 

Metarhinus 

Dolichorhinus 

Rh ADINORHINUS 

Rhadinorhinus 

Eotitanops 

Eotitanops 

Eotitanops 

Lambdotherium 

Lambdotherium ' progressum. 

[ Diploceras] 

[Diplooeras] orborni 

[Heterotitanops] 

[Heterotitanops] I parvus 

EOTITANOTHERIUM I 

[Telmatherium?] j birmanicum 

Lambdotherium 

Eotitanops 

EOMETARHINUS 

Eometarhinus 



superior.. 

riparius 

cristatus__ 
fluminalis- 



abbotti.- 
gregoryi- 
princeps. 

major 

prisoum. 



magnum, 
minimus. 



huerfanensis _ 



Osborn. 
do- 



.do. 
.do. 
.do. 



.do_ 
.do_ 
.do. 



.do_ 
.do. 
.do. 



do... 

Douglass. 
do.._ 



do._ 

do.. 

Gregory. 
Riggs... 
do._ 



.do_ 
.do. 
.do. 



do. 

Osborn. 
do. 



-do. 
.do- 
.do. 



Peterson. 

do._ 

do__ 



.do_ 
.do. 



Pilgrim and Cotter . 

Osborn 

do 



.do_ 
.do_ 



Present determination 



Palaeosyops grangeri Osborn. 
Palaeosyops copei Osborn. 
Manteoceras washakiensis Osborn. 
Mesatirhinus Osborn. 
Mesatirhinus petersoni Osborn. 
Metarhinus Osborn. 
Metarhinus fluviatilis Osborn. 
Metarhinus earlei Osborn. 
Dolichorhinus intermedius Osborn. 
Telmatherium ultimum Osborn. 
Telmatherium altidens Osborn. 
Protitanotherium superbum Osborn. 
Sthenodectes incisivus (Douglass). 
Manteoceras uintensis Douglass. 
Dolichorhinus heterodon Douglass. 
Dolichorhinus longiceps Douglass. 
Sthenodectes Gregory. 
DoUohorhinus superior (Riggs). 
Metarhinus riparius Riggs. 
Metarhinus cristatus Riggs. 
Dolichorhinus fluminalis Riggs. 
Rhadinorhinus Riggs. 
Rhadinorhinus abbotti Riggs. 
Eotitanops gregoryi Osborn. 
Eotitanops princeps Osborn. 
Eotitanops major Osborn. 
Lambdotherium priscum Osborn. 
Lambdotherium progressum Osborn. 
Eotitanotherium Peterson. 
Eotitanotherium osborni Peterson. 
? Metarhinus. 
?Metarhinus sp. 
Eotitanotherium Peterson. 
Uncertain. 

Lambdotherium magnum Osborn. 
Eotitanops minimus Osborn. 
Eometarhinus Osborn. 
Eometarhinus huerfanensis Osborn. 



DISCOVERY OF THE TITANOTHERES AND ORIGINAL DESCRIPTIONS 



157 



OEIGINAI DESCRIPTIONS OF THE SPECIES 

Palaeosyops Leidy, 1870 

Cf. Palaeosyops, this monograph, page 312 

Original reference. — Acad. Nat. Sci. Philadelphia 
Proc, 1870, p. 113 (Leidy, 1870.2). 

Type species. — Palaeosyops paludosus Leidy. (See 
p. 319.) 

Generic characters. — Leidy, in his description of the 
fragmentary type, very properly refrained from at- 
tempting to distinguish generic from specific charac- 
ters. Generic characters are given below. 

Etymology. — TraXaio?, ancient; am, boar; ih\p, face 
(appearance). The name was probably suggested by 
the fact that the "upper true molars exhibit the outer 
part of a crown composed of a pair of lobes, exactly 
as in Hyopotamus." (Leidy.) 

Present determination. — The generic name is a valid 
one. 

Palaeosyops paludosus Leidy, 1870 

Cf. Palaeosyops paludosus, this monograph, page 319 

Original reference. — Acad. Nat. Sci. Philadelphia 
Proc, 1870, p. 113 (Leidy, 1870.2). 

Subsequent reference. — Leidy, Extinct vertebrate 
fauna of the Western Territories, p. 28, pi. 23, figs. 3-6 
(fig. 5 lectotype), 1873 (Leidy, 1873.1). 

Type locality and geologic horizon. — Church Buttes, 
near Fort Bridger, Bridger Basin, Wyo.; Palaeosyops 
paludosus-Orohippus zone (Bridger B 1 or Bridger 
B2). 

Leidy's cotypes. — M2, p*, m^, m^ (Nat. Mus. 759, 
758, 762). (Extinct vertebrate fauna, p. 28, 1873.) 
"The species Palaeosyops paludosus * * * -nras 
founded on a number of isolated teeth * * * qI^. 
tained by Professor Hayden at Church Buttes, Wyo." 
(Leidy.) (See fig. 88.) 

Characters. — Leidy (1870.1, p. 113) writes: 

The crown of a lower true molar [mj of the right side, the 
Osborn lectotype (fig. 88)] resembles in its constitution those of 
Palaeotherium, Chalicotherium, and Titanotherium, being com- 
posed of a pair of fore and aft conjoined pyramidal lobes with 
crescentic summits. It measures 16 lines anteroposteriorly and 
10 lines transversely. Fragments of upper true molars [m^ left, 
m' right] exhibit the outer part of the crown composed of a pair 
of lobes exactly as in Hyopotamus. The inner portion of the 
crown is composed of a pair of simple cones, broad and low, 
the front one considerably larger than the back one. One of 
the specimens in the entire condition of the crown measured 
about 22 lines fore and aft and 18 lines transversely. The crown 
of an upper premolar [p*] has its outer part composed of a pair 
of conjoined cones with acute summits and sides. The inner 
portion of the crown [p* of the opposite side] consists of a single 
broad, simple cone embraced in front and behind by a basal 
ridge. The anteroposterior diameter of the crown externally 
measures 9}/2 lines; the transverse diameter is an inch. 

Leidy's cotypes. — The first lot of specimens from 
Church Buttes (Bridger B 1), upon which the species 
was originally established, consist of a second lower 
molar (ma, Nat. Mus. 759; see Leidy, 1870.1, p. 113; 



1873.1, pi. 23, fig. 5); "of an upper fourth premolar 
nearly unworn" (p*, Nat. IVIus. 762; see Leidy, 1873.1, 
pi. 5, fig. 5); of the anterior half of a second upper 
molar (m^ Nat. IVIus. 758; see Leidy; 1873.1, pi, 23, 
fig. 6); and of the inner side of a premolar (p*) of the 
opposite side. This lot constitutes the cotypes, which 
are here refigured. Of these, the second lower molar 
agrees with the specimens described in this monograph 
as P. paludosus. The upper teeth do not certainly 
belong to the same animal; it appears necessary, 
therefore, to base the genus and species on the first 
specimen described in the original description, namely, 
the second lower molar, which may be taken as the 
lectotype. 

Leidy's paratypes. — Specimens of a second lot, 
from Henrys Fork, belonging to a much older individ- 
ual, were treated practically as paratypes of this species 
in the original notice; they were described in Leidy's 
memoir of 1873 (1873.1, pp. 29, last line, and 30), 
were figured in Plate 5, Figures 4, 6, 7, 8, 9, and are 




Figure 88. — Leidy's cotypes of Palaeosyops paludosus 

Specimens upon which the species was originally established. Hayden's collection 
of 1870. After Leidy, 1873; Nat. Mus. 758, 759, 762. Natural size. The second 
lower molar (Nat. Mus. 759) is the lectotype. 

preserved in the United States National IVTuseum. 
These are the specimens that Cope, IVEarsh, Osborn, 
Earle, and others may have taken for the types, but 
they are from a higher geologic level and may pertain 
to P. major or P. leidyi. A third lot of specimens, 
from Grizzly Buttes, included the "facial portion of a 
skull containing nearly all the molars and the canines 
of both sides." This specimen was treated virtually 
as a paratype by Leidy (1873.1, pp. 30-34, pi. 18, 
fig. 51, and pi. 4, fig. 3) and was described at length 
by him. It is probably but not certainly conspecific 
with the lectotype m2. 

Osiorn's neotype. — The determination of P. palu- 
dosus therefore rests positively on the second lower 
molar alone. To supplement this lectotype the 
present author has selected as a neotype a lower jaw 
(PI. LVI, B; fig. 268, C) with dentition, Am. IVIus. 1 1680, 
in which m2 agrees closely with the lectotype and with 
the measurements given by Leidy for P. paludosus 
(1873.1, p. 57 and pi. 5, figs. 10, 11). The locality 



158 



TITANOTHERES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



(Millersville) is about 10 miles distant from that of 
the holotype (Church Buttes), but the geologic level 
is believed to be identical, namely, Bridger B 1 . 

Etymology. — paludosus, marshy, dwelling in the 
marshes, probably because the remains were found in 
one of the supposed "ancient lake basins." 

Present determination. — Palaeosyops paludosus Leidy 
is a valid species, but the "P. paludosus" of other 
authors refers to related species of more recent geologic 
age (P. major, P. leidyi, P. rolustus). 

Palaeosyops major Leidy, 1871 

Cf. Palaeosyops major, this monograph, page 321 
Original reference. — Acad. Nat. Sci. Philadelphia 

Proc, 1871, p. 229 (Leidy, 1871.1). 

Subsequent reference. — Leidy, Extract vertebrate 

faima of the Western Territories, p. 45, pi. 20, fig. 

8, 1873 (Leidy, 1873.1). 





Figure 89. — Leidy's type (.holotypej of Palaeosyops major in 
the collection of the Academy of Natural Sciences of Phila- 
delphia 

Part of the right ramus of a lower jaw. After Leidy, 1873. One-half natural size. 
"The specimen is somewhat swollen and altered from disease and is one of those 
upon which the species was first indicated. Discovered by Dr. Carter at Grizzly 
Buttes." (Leidy.) 

Type locality and geologic liorizon. — Grizzly Buttes, 
Bridger Basin, Wyo.; Palaeosyops paludosus-OroTiip- 
pus zone (Bridger B 2 or B 3). Dr. J. Van A. Carter, 
collector. 

Holotype. — "A jaw fragment with the retained 
fragments of the true molars." This type is now in 
the collection of the Philadelphia Academy of Natural 
Sciences. (See fig. 89.) 

Characters (Leidy). — Size apparently "much larger 
than P. paludosus." 

The true molars occupied a space of 4J^ lines. The last 
molar measured IJ^ inches fore and aft and an inch trans- 
versely in front. " 

In his "Extinct vertebrate fauna" of 1873 (1873.1, 
pp. 45, 46) Leidy gives a fuller description of the very 
imperfect holotype and figures it on Plate 20, Figure 
8. He believed the jaw specimen to be 
in some degree abnormal in form, due to inflammation or 
some other affection connected with the second molar tooth. 



* * * In its proportions the jaw, in a normal condition, 
would appear to be of more robust character than in Palaeo- 
syops paludosus. * * * In its present state the base is 
more convex fore and aft than in the latter, and the alveolar 
border more ascending posteriorly. 

The remains of the molar fangs at the entrance of the alveoli 
appear to indicate teeth of the same form and construction as in 
Palaeosyops paludosus, for which reason the fragment was 
referred to the same genus. The true molars appear to have 
occupied a space of 4^ inches, though this is probably some- 
what exaggerated, as the interval occupied by the last inter- 
mediate molar appears proportionately somewhat too large. 
The crown of the last molar, which was clearly trilobate as in 
Palaeosyops paludosus, had an antero-posterior diameter of 2 
inches. 

Leidy's paratype, "consisting of the left ramus of the 
lower jaw, containing six molar teeth," was obtained 
by Doctor Carter "in Dry Creek Canyon, 40 mUes 
from Fort Bridger" (Bridger Basin, Bridger C) and 
together with a second similar specimen from the 
same locality is described by Leidy (1873.1, p. 46, pi. 
23, fig. 1; second specimen, fig. 2). 

The holotype, it is important to note, is from the 
low level (probably Bridger B 2) of Grizzly Buttes, 
but Leidy's paratype, which has the characters of the 
more progressive Palaeosyops leidyi Osborn, is from 
the higher level (Bridger C) of Dry Creek. The 
paratype is thus certainly not conspecific with the 
holotype. 

Osborn's neotype. — In order to supplement the 
characters of Leidy's imperfect holotype, the present 
writer has selected as a neotype a lower jaw (fig. 
268, C) with dentition (Am. Mus. 12181) from Cotton- 
wood Creek and from about the same level (B 3) 
as the holotype, with which it agrees closely. (See 
Chap. V.) 

Etymology. — major, in allusion to the larger size as 
compared with P. paludosus. 

Present determination. — The species P. major is 
believed to be a valid one. 

Canis montanus Marsh, 1871 
Cf. Canisf marshii Hay, below {Palaeosyops major?), page 178 

Original reference. — Am. Jour. Sci., 3d ser., vol. 2, 
p. 123, August, 1871 (Marsh, 1871.2). 

Type locality and geologic liorizon. — Grizzly Buttes, 
Bridger Basin, Wyo.; Palaeosyops paludosus-Oro- 
hippus zone (Bridger B, probably B 2). 

Marsh's cotypes. — "A last upper premolar tooth in 
good preservation, a canine, wanting most of the 
crown, and a number of the larger bones of a skeleton, 
all apparently of the same species, but pertaining to 
three individuals, differing somewhat in size " (Marsh). 
Of these materials the "last upper premolar" (first 
lower premolar) alone is described and measiu'ed, and 
it is also the first specimen mentioned. It should 
therefore be taken as the lectotype (Yale Mus. 11770). 

Characters. — "The last upper premolar * * * 
is robust, has a short compressed crown. The princi- 
pal cusp is conical, with subacute edges, the anterior 
being about twice the length of the posterior. Behind 



DISCOVERY OF THE TITANOTHERES AND ORIGINAL DESCRIPTIONS 



159 




Figure 90. — Leidy's 
type of Palaeosyops 
humilis 



the main cusp there is a large triangular tubercle, 
with its apex exterior to the fore and aft axis of the 
crown." (Marsh.) 

Anteroposterior diameter of last upper pre- 
molar 9 lines [19 mm.] 

Greatest transverse diameter of same 4.25 lines [8 mm.l 

Height of main cusp 6 lines [12.7 mm.] 

Height of posterior tubercle 3.75 lines [7 mm.] 

Synonym. — Canis? marsMi Hay was proposed in 
place of 0. montanus IVIarsh, name preoccupied by 
0. montanus Pearson (Hay, 
1899.1). 

Etymology. — montanus, dwell- 
ing in the mountains — that is, 
from the Eocky Mountain 
region. 

Present determination. — The 
type specimen of "Canis mon- 
specimen in the collection of tdnus" is a first lowcr premolar 
the Academy of Natural of somc Undetermined member 

S c i e n c e s of Philadelphia, <■ ,i n v 

After Leidy, 1S73. Natural 01 t^ie gCnUS FttlaeOSyOpS, pOSSl- 

size. Regarded by Leidy as jjly PaltteOSyOpS paludoSUS Or P. 

"A last upper molar of the . rm p , /-y • 

leftside. • • • Found by mc-jor. ihe reference to Cams 
Doctor Corson on the buttes ^^Q,s, doubtless made by rcason 

of Dry Creek Canyon." r ,i i . • ii n 

of the deceptive resemblance oi 
one of the lower premolars to the upper carnassial 
tooth of a dog. 

Palaeosyops humilis Leidy, 1872 
Cf. Palaeosyops major, this monograph, page 321 

Original reference. — Acad. Nat. Sci. Philadelphia 
Proc, 1872, p. 168 (Leidy, 1872). 

Subsequent reference. — Leidy, Extinct vertebrate 
fauna of the Western Territories, p. 58, pi. 24, fig. 8, 
1873 (Leidy, 1873.1). 

Type locality and geologic Tiorizon. — "Valley of Dry 
Creek 40 miles from Fort Bridger (Wyo.)." Doctor 
Corson, discoverer. "Buttes of Dry Creek Canyon," 
Bridger Basin; horizon probably Bridger C (Uinta- 
therium- Manteoceras- MesatirMnus zone) . 

Holotype. — "An upper molar." (See fig. 90.) 

Characters (Leidy). — "An upper molar tooth of this 
animal measures three-foiu-ths of an inch in diame- 
ter." In his later description Leidy recognized that 
the specimen belonged to the milk series. 

Etymology. — Tiumilis, lowly, small; in allusion to 
the small size in comparison with P. paludosus. 

Present determination. — This milk tooth probably 
pertains to the genus Palaeosyops, but comparison 
with P. major and P. leidyi leaves the species unde- 
termined. 

Palaeosyops Junius Leidy, 1872 

Cf . Mesatirhinus Junius (Leidy) , this monograph, page 388 

Original reference. — Acad. Nat. Sci. Philadelphia 
Proc, 1872, p. 277 (Leidy, 1872.3). 

Subsequent reference. — Leidy, Extinct vertebrate 
fauna of the Western Territories, p. 57, no figure, 
1873 (Leidy, 1873.1). 



Type locality and geologic horizon. — Fort Bridger, 
Bridger Basin, Wyo.; level not recorded. Dr. J. 
Van A. Carter, collector. 

Holotype. — -"Doctor Carter recently sent the writer 
several small fragments of the right side of a lower 
jaw, together with a sketch of a larger fragment of 
the left side, containing the last premolar and the 
succeeding molars." Of this type material only p4 
(right) and the posterior half of ma (right) were 
located (1906) in the collection of the Academy of 
Natural Sciences of Philadelphia. 

Characters. — Leidy writes: 

Intermediate in size to P. paludosus and P. humilis. Founded 
on portions of a lower jaw agreeing in character with the cor- 
responding parts of P. paludosus but smaller. Space occupied 
by the last premolar and the true molars, 4 inches. Antero- 
posterior diameter of last premolar, 8 lines; of last molar, 
173^ hnes. 

In the fuller description in his memoir of 1873, 
Leidy says: 

The specimens * * * appear to indicate a small species 
of Palaeosyops, though it is not improbable that they pertain 
to a small variety of P. paludosus. 

The parts agree closely with the corresponding parts of the 
lower jaw and teeth of the latter, except in size. They have 
been viewed as representatives of a species with the name of 
Junius. 




Figure 91. — Leidy's cotypes of Palaeosyops 
Junius 

Specimens in museum of Academy of Natural Sciences of 
Philadelphia; Bridger B(?), level doubtful. A, Eight fourth 
lower premolar (pO; B, posterior part of third lower molar 
Cms) . Natural size. 

The measurements of the teeth (fig. 91) in comparison with 
those of P. paludosus are as follows: 



Space occupied by the last pre- 
molar and molars 

Space occupied by the molars 

Breadth [anteroposterior] of last 
premolar 

Thickness [transverse] of last 
premolar 

Breadth [anteroposterior] of first 
molar 

Breadth [anteroposterior] of sec- 
ond molar 

Breadth [anteroposterior] of 
third molar 

Thickness [transverse] of third 
molar at middle 



39J^ 
8 

10 
12 

17 
7 



[Milli- 
meters] 



[102] 
[94] 

[17] 

[12] 

[21] 

[25] 

[10] 

[14] 



55 
46 



6M 
12J^ 
15 
19 

93^ 



[MUli- 
meters] 



[116] 
[96] 

[19] 

[12] 

[38] 

[32] 

[39] 

[19] 



160 



TITANOTHERES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



Etymology. — Junius, younger, in allusion to its small 
size. 

Present determination. — From the two teeth (p4 and 

part of ma) preserved it appears that this species 

probably pertains to the genus Mesatirhinus. It is 

smaller than Mesatirhinus megarhinus. No other 

material has been certainly identified with it. (See 

p. 388.) 

Palaeosyops laticeps Marsh, 1872 

Cf. Limnohyops laticeps Marsh, this monograph, page 311 

Original reference. — Am. Jour. Sci., 3d ser., vol. 4, 

p. 122, August, 1872, dated "July 18, 1872" (Marsh, 

1872.1). 

Type locality and geologic horizon.— NesiT Marsh's 

Fork, about 15 miles from Fort Bridger, Wyo. A. H. 

Ewing, discoverer. Level not recorded. 

Holotype. — "A__nearly complete skeleton" (Yale 

Mus. lioOO). 



Etymology. — latus, broad; caput (in compounds ceps), 
head; in allusion to the width across the zygomata. 

Present determination. — Marsh's accurate diagnosis 
of this excellent type was made before the generic 
characters of Palaeosyops were fully known. The 
species was subsequently chosen by Marsh as the 
type of the genus Limnohyops Marsh, and both the 
genus and the species are valid. 

Telmatherlum Marsh, 1872 

Cf. Telmatherium, this monograph, page 340 

Original reference. — Am. Jour. Sci., 3d ser., vol. 4, 
p. 123, August, 1872 (Marsh, 1872.1). 

Type species. — Telmatherium validum Marsh. (See 
pp. 160, 344 of this monograph.) 

Generic characters. — See T. validum (p. 340). 

Etymology. — reXfia, a pool, marsh (cf. "paludosus"); 
dripiov, beast. 




Figure 92. — Marsh's type of Palaeosyops laticeps 
Natural size. 



Characters. — Marsh writes : 



The teeth in this specimen have apparently the same general 
structure as those in the type of P. paludosus but differ in 
being nearly smooth, and this is not the result of age, as this 
individual was younger than the original of the larger species. 
The proportions, moreover, given for the molar described 
(Leidy, 1870.2, p. 113), "22 lines fore and aft and 18 trans- 
versely," would not apply to any of the series in the present 
specimen. The last upper molar of the latter has two well- 
developed internal cones. * * * The upper teeth form a 
complete series. The canine is large and broadly oval at its 
base. The outer incisor is the largest, and at its posterior 
edge the premaxillary is subtriangular in transverse section. 
The sagittal and occipital crests are strongly developed, and 
the coronoid process of the lower jaw is short and recurved. 

Measurements [Marsh] ^ 

Millimeters 

Length of entire upper molar series 155 

Anteroposterior extent of three true upper molars 94 [90] 

Anteroposterior diameter of last upper molar 36 [33] 

Transverse diameter [protocone to mesostyle] 40 

Anteroposterior diameter of upper canine at base 

[alveolar portion 28] 29 

Transverse diameter 22 



Present determination. — The generic term as re- 
defined in the present monograph is a valid one. 

Telmatherium validus Marsh, 1872 
Cf. Telmatherium validum, this monograph, page 344 

Original reference. — Am. Jour. Sci., 3d ser., vol. 4, 
p. 123, August, 1872; dated "July 18, 1872" (Marsh, 
1872.1). 

Type locality and geologic horizon. — "Near Henrys 
Fork of the Green River in Wyoming." (Bridger 
Basin, level C or D.) J. F. Quigley, discoverer. 

Holotype. — "The greater portion of a skull, with 
teeth" (Yale Mus. 11120). (See fig. 93.) 

Characters. — Marsh writes: 

The dentition of this genus, so far as known, appears to be 
similar to that of Palaeosyops; but the two may readily be dis- 
tinguished by the anterior portion of the skull, which in the 
present genus has the premaxillaries compressed, with an 
elongated median suture. The zygomatic arch is also much 
less strongly developed, and the squamosal portion of it is com- 
paratively slender. 



DISCOVERY OF THE TITANOTHERES AND ORIGINAL DESCRIPTIONS 

Limnohyus Marsh, 1872 



The upper molar teeth have the inner cones more elevated 
and more pointed than in Palaeosyops, and the basal ridge is 
well developed. The last upper molar has but a single internal 
cone. The upper canines are large, pointed, and have strong 
cutting edges. The outer incisors are the largest, and all these 
teeth have a strong inner basal ridge. The roof of the mouth is 
deeply excavated between the premolars. The nasals are de- 
curved laterally and much compressed. 



Cf. Palaeosyops, this monograph, page 331 

Original reference. — Am. Jour. Sci., 3d ser. 
p. 124, August, 1872 (Marsh, 1872.1). 

Type species. — Limnohyus rohustus Marsh. 



161 



vol. 4, 




Figure 93. — Marsh's type of Telmatherium validus 
Natural size. 



Measurements [Marsh] 

Millimeters 

Extent of upper molar series : 224 

Extent of upper true molars 130 

Anteroposterior diameter of last upper molar 54 

Anteroposterior diameter of last upper premolar 28 

Transverse diameter 33 

Anteroposterior diameter of upper canine at base 27 

Transverse diameter 22 

Etymology. — validus, strong, stout; perhaps in allu- 
sion to the large size of the upper canines. 



Generic characters (Marsh). — The term Palaeosyops 
is restricted to those specimens which, like P. paludo- 
sus, possess two inner cones on m^ 

The other specimens have but a single internal cone on the 
last upper molar, and for the genus thus represented the name 
Limnohyus is proposed. These genera may be distinguished 
from Telmatherium by the premaxillaries, which are short, 
stout, and depressed, with a small median suture. 

Etymology. — Xi^ufi?, a marshy lake; Cs, boar. 

Present determination. — Since the type species Lim- 
nohyus rohustus is now believed to be congeneric with 




Figure 94. — Marsh's type of Limnohyus rohustus 
Natural size. 



Present determination. — This is a valid genus and 
species. The name Telmatherium was amended to 
Telmatotherium by Marsh in 1880 (1880.1) in his 
"List of genera established by Prof. O. C. Marsh, 
1862-1879," and the amended form was accepted by 
Earle, Osborn, Hatcher, and later writers, but accord- 
ing to the rules of nomenclature now generally ac- 
cepted the amended form has no standing and the 
original form Telmatherium should be used. 



Leidy's Palaeosyops paludosus, Marsh's genus Limno- 
hyus becomes a synonym of Palaeosyops. 

Limnohyus rohustus Marsh, 1872 

Cf. Palaeosyops rohustus (Marsh), this monograph, page 331 

Original reference. — Preliminary description of new 

Tertiary mammals: Am. Jour. Sci., 3d ser., vol. 4, 

p. 124, August, 1872; dated "July 18, 1872" (Marsh, 

1872.1). 



162 



TITANOTHERES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



Type locality and geologic horizon. — Near Henrys 
Fork, Wyo.; Uintatherium- Manteoceras- Mesatirhinus 
zone (Bridger C or D). F. Mead, jr., collector. 

Holotype. — A fragmentary skull including nasals and 
palate with teeth (Yale Mus. 11122). 

Characters. — Marsh writes: 

The present species may be distinguished from those above 
described [Palaeosyops laticeps Marsh, Tdmatherium validum 
Marsh], especially by the strong basal ridge of the molars. On 
the last lower molar it extends entirely around the posterior 
lobe. The first of the upper true molars has the two inner 
cones nearly of the same size. The small intermediate median 





FiGTJRE 95. — Cope's cotypes of Palaeosyop. 



After Cope, 1884. Ai, Lett mandibular ramus, superior view Oeetotype, Am. Mus. 5098); Aj, internal view otsame; 
B, left upper premolars and molars (Am. Mus. 5097); Ci, third left superior molar, crown view; Ct, the same, buccal 
view (Am. Mus. 5099); Di, fourth superior premolar, crown view; Dj, the same, buccal view (Am. Mus.). 

tubercles are well developed on the upper molars, and all the 
teeth are strongly rugose, even in fully adult animals. The 
nasal bones contract anteriorly and are rounded in front. 
The outer margin is decurved and thickened. The premaxil- 
laries unite by a very short median suture, similar to that in 
Palaeosyops laticeps. The zygomatic process of the squamosal 
is stout but much compressed, thus differing widely from both 
the species already described. 

Measurements [Marsh] 

Millimeters 

Anteroposterior extent of last three upper molars 110 

Anteroposterior diameter of last upper molar 41 

Transverse diameter 43. 5 

Anteroposterior diameter of last lower molar 51 



Etymology. — robustus, robust; in allusion to the stout 
skull and dentition. 

Present determination. — The species is probably a 
valid one, referable to the genus Palaeosyops. 

Limnohyus Leidy (not Marsh), 1872 
Cf. Palaeosyops, this monograph, page 155 
Original reference. — Acad. Nat. Sci. Philadelphia 

Proc, 1872, pp. 240-242; published December 17, 

1872 (Leidy, 1872.1). 
^ As we have seen above, Marsh's genus Limnohyus is 

simply a synonym of Palaeosyops, which had been 
defined by Leidy as having 
"but a single lobe to the inner 
part of the crown " of the "last 
upper molar." In 1872 Leidy, 
after pointing out this fact, 
says that the name Limnohyus 
"might with propriety be 
applied to the animal with 
molars like those of Palaeo- 
syops except that the last upper 
one has two inner cones to the 
crown." This doubtless sug- 
gested Marsh's subsequent 
term Limnohyops. Lim- 
nohyus Leidy is thus preoccu- 
pied by Limnohyus Marsh, 
which is a sj'^nonym of 
Palaeosyops. 

Etymology. — Xifivrj, a marshy 
lake; Cs, boar. 
Palaeosyops vallidens Cope, 1872 

Cf. DoKchorhinus vallidens (Cope), 
this monograph, page 401 

Original rejerence. — -Pal. 
Bull. No. 7, dated "Aug. 22, 
1872"; Am. Philos. Soc. Proc, 
vol. 12, p. 487, 1873 (Cope, 
1872.1). 

Subsequent reference. — Ter- 
tiary Vertebrata, p. 699, pis. 
51, fig. 1; 52, fig. 3; 53, fig. 1; 
36, figs. 10, 10a, 11, 11a, 1884 
[1885] (Cope, 1885.1). 

Type locality and geologic 
horizon. — -"Mammoth Buttes, southwestern Wyoming, 
near the headwaters of Bitter Creek," Washakie Basin; 
Eobasileus-Dolichorhinus zone (Washakie B 2). 
Characters. — Cope writes: 

Represented by the dentition of one maxillary bone with 
other bones of one individual [Cope, Am. Mus. 5097]; a portion of 
the same dentition of a second [No. 5099]; with both rami of 
the mandible with complete dentition of a third [No. 5098]. 
The species is distinguished by the details of the dental struc- 
ture and by the superior size. It exceeds, in this respect, the 
Palaeosyops major Leidy; while the three posterior lower molars 
measure 4.5 inches in length, the same teeth of the present 
animal measure 6.25 inches. The last superior molar of an- 
other specimen measures 2 inches in length; in the third the 



DISCOVERY OF THE TITANOTHERES AND ORIGINAL DESCRIPTIONS 



163 



first true molar is 1.5 inches in length, while the last inferior 
molar is 2.25 inches long. The pecuharity in the structure of 
the superior molars consists in the existence of two strong 
transverse ridges, which connect the inner tubercle with the 
outer crescents, inclosing a pit between them. These are most 
marked on the premolars, where also is found the peculiarity 
of the almost entire fusion of the outer crescents into a single 
ridge. These united crescents are narrower than in P. major, 
and the summits of all the crescents are relatively more ele- 
vated. The number of inner tubercles is the same as in that 
species; all the teeth have very strong basal cingula, which 
rise up on the inner tubercle. The last inferior molar is rela- 
tively narrower than in P. major, and the posterior tubercle is 
larger and longer and is an elevated cone. 

From the foregoing description it will be seen that 
Cope based his specific description upon three speci- 
mens (cotypes) without designating any one of the 
three as being more typical than the others. (See 
fig. 95.) The name vallidens, however, from vallum, 
a wall, seems to refer to the peculiarity in the struc- 
ture of the superior molars, which 

consists in the existence of two strong transverse ridges, which 
connect the inner tubercle with the outer crescents. * * * 
These ridges are most marked on the premolars, where also is 
found the pecuharity of the almost entire fusion of the outer 
crescents into a single ridge. 

If we had nothing further to guide us we would thus 
be led to infer that the upper dentition (Am. Mus. 5097), 
which best shows these peculiarities alluded to in the 
name vallidens, should be regarded as the most typical 
of the three specimens and should be chosen as the 
lectotype. But in his "Tertiary Vertebrata" Cope 
(1885.1, p. 700) says: 

The bones containing the maxillary and mandibular teeth 
were not found together in any instance, so that it is possible 
that the different series may represent different species. No 
other species of the genus was, however, found in the locaUties 
to which the respective parts could be referred. Should these 
prove not to pertain together, the lower jaws may be regarded 
as typical of the species. 

As Cope was the "first reviser" of the species there 
seems to be no escape from the conclusion, if modern 
rules of nomenclature are to be followed, that the lower 
jaw (Am. Mus. 5098) must be treated as Cope's 
lectotype. 

Etymology. — vallum, wall or redoubt; dens, tooth; 
allusion as explained above. 

Present determination. — This little-known species is 
allied to but probably specifically distinct from Doli- 
cJiorTiinus Tiyognathus of Washakie B and Uinta B. 
It is also more primitive than that species (see below). 

Limnohyus laevidens Cope, 1873 

Cf. Limnohyops laevidens (Cope), this monograph, page 305 

Original reference. — Pal. Bull. No. 11 ("issued Jan. 
31, 1873"); Am. Philos. Soc. Proc, vol. 13, pp. 35, 
36, 1873 (Cope, 1873.5). 

Subsequent references. — Cope, On the extinct Verte- 
brata of the Eocene of Wyoming, observed by the 
expedition of 1872: U. S. Geol. and Geog. Survey 



Terr. (Hayden) Sixth Ann. Kept., p. 591, 1873 (Cope, 
1873.6); Tertiary Vertebrata, p. 701, cotype skull, pi. 
50, figs. 1, 2 (holo type) , fig. 3 (paratype), 1884 [1885] 
(Cope, 1885.1). 

Type locality and geologic horizon. — Type ("No. 1 "), 
Cottonwood Creek, Bridger Basin, Wyo.; Palaeosyops 
paludosus-Orohippus zone (Bridger B). Cotype ("No. 
2"), Bitter Creek, Washakie Basin, Wyo.; horizon 
uncertain. 

Cope's cotypes: "A cranium lacking the posterior 
part of one side and the lower jaw," from Cottonwood 
Creek ("No. 1," now Cope collection. Am. Mus. 5104). 
Also "a nearly complete cranium with dentition from 




Figure 96. — Cope's cotypes of Limnohyops laevidens 

After Cope, 1885. One-fourth natural size. A, Am. Mus. 5104, lectotype: Ai, 
" Cranium lacking posterior part of one side and lower jaw, from Cottonwood 
Creek" (Cope), "No. 1"; As, upper teeth of the same. B, Am. Mus. 6105, 
now retered to Palaeosyopst copei, right maxilla, p^m'. 

Bitter Creek" ("No. 2," now Cope collection. Am. 
Mus. 5105). (See fig. 96.) 

Cope's lectotype: Cope's first-mentioned specimen 
is the one from Bitter Creek (Washakie B?) (Am. Mus. 
5105), now referred to Palaeosyops? copei. But the 
"No. 1" of Cope's description and measurements and 
the specimen to which the name "laevidens" refers is 
unquestionably the skull Am. Mus. No. 5104, from 
Cottonwood Creek (level Bridger B), Bridger Basin, 
now referable to Limnohyops. Furthermore, in the 
"Tertiary Vertebrata" (Cope, 1885.1, pp. 701-703, 
pi. 50, figs. 1, 2) Cope definitely selects, describes, and 
figures this specimen as the type, again referring to 
the Washakie specimen as "No. 2" and admitting 
that its specific association with the other specimen 
was doubtful. We therefore follow Cope in regarding 



164 



TITANOTHERES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



"No. 1," the Bridger specimen (Am. Mus. 5104), as 
the lectotype. 

Characters.- — Cope writes: 

This species is one of the larger forms of the group originally 
represented by Palaeosyops, and which has turned out to be so 
numerous in species. [This statement refers apparently to 
"No. 2."] 



The anterior median small tubercle of the first true molar is 
wanting. The last true molar has but one interior cone. [All 
these statements apply evidently to " No. 1," the Bridger or Cot- 
tonwood Creek specimen.] 

The canine tooth is powerful and bearhke; the outer incisor 
is the largest. The premaxillary bones are short, and the side 
of the face elevated and plane to the convex nasal bones. Zygo- 
matic arch massive. 




Am. Mus. 5107. 



Figure 97.— 

After Cope, 1885. Ai, 
natural 



A3 



-Cope's type (holotype) of Limnohyus fontinalis 

Young skull seen from above, one-half natural size: As, the same, right side, one-half 

size; A3, right maxilla with dp<, m', m', natural size. 



The molars have the general form of those of L. robustus, 
but the second superior premolar has but one outer tubercle. 
The cingula are much less developed than in that species, 
those between the inner cones of the molars being entirely 
absent. These cones are low and, with the rest of the crowns 
of all the teeth, covered with smooth and shining enamel. 



Measurements [Cope, condensed and corrected] 

IVtillimeters 

Length of molar series (No. 1) 141 

Length of true molars 84 

Length of crown canine (anteroposterior) 20 

Length of crown last molar (anteroposterior) 30 

Width of crown last molar (transverse) 34 



DISCOVERY OF THE TITANOTHERES AND ORIGINAL DESCRIPTIONS 



165 



Etymology. — laevis, levis, smooth, shining; dens, 
tooth; apparently in allusion to the "smooth and 
shining enamel." 

Present determination. — This is a valid species refer- 
able to the genus Limnohyops. 

Limnohyus fontinalis Cope, 1873 

Cf. fPalaeosyops fontinalis Cope, this monograph, page 317. 

Original reference. — Pal. Bull. No. 11, "issued Jan- 
uary 31, 1873"; Am. Philos. Soc. Proc, vol. 13, pp. 35, 
36, 1873 (Cope, 1873.5). 

Subsequent references. — Cope, On the extinct Verte- 
brata of the Eocene of Wyoming, observed by the 
expedition of 1872: U. S. Geol. and Geog. Survey 
Terr. (Hayden) Sixth Ann. Kept., p. 594, 1873 (Cope, 
1873.6); Tertiary Vertebrata, p. 707, pi. 49, fig. 9; 
pi. 50, fig. 4; pi. 58a, figs. 4, 5, 1884 [1885], (Cope, 
1885.1). 

Type locality and geologic Tiorizon. — "Found by the 
writer on a bluff on Green River, near the mouth of 
the Big Sandy, Wyoming." ("Isolated patch lying 
northeast of the badlands." Probably Eometarliinus- 
Trogosus-Palaeosyops fontinalis zone (Bridger A).) 

Holotype. — A young, fragmentary skull (Cope col- 
lection. Am. Mus. 5107, retaining dp*, m', - - 
m^ of the right side. (See fig. 97.) 

Characters. — Cope writes: 

A small species agreeing with the P. paludosus 
in the two interior cones of the last superior 
molar. It is represented especially by a consider- 
able part of the cranium of an individual in which 
the last superior molar is not quite protruded, 
but with the other molars and last premolar of 
the permanent dentition in place. The enamel 
of these teeth is in accordance with the age, delicately rugose, 
and while the cingulum is present fore and aft, it is wanting 
internally and externally. The anterior median tubercle is 
present on all the true molars, and the bases of the acute inner 
cones are in contact. The sagittal crest is truncate, and the 
squamosal portion of the zygoma very stout. The nasal bones 
are together very convex in transverse section. 



Palaeosyops diaconus Cope, 1873 

Cf. Palaeosyops robustus (Marsh), this monograph, page 331. 

Original reference. — Pal. Bull. No. 12, p. 4, "pub- 
lished March 8, 1873" (Cope, 1873.1). 

Subsequent references. — Cope, On the extinct Verte- 
brata of the Eocene of Wyoming observed by the 
expedition of 1872: U. S. Geol. and Geog. Survey 
Terr. (Hayden) Sixth Ann. Kept., p. 593, 1873 (Cope, 
1873.6); Tertiary Vertebrata, p. 706, pi. 51, fig. 3, 
1884 [1885] (Cope, 1885.1). 

Type locality and geologic horizon. — Henrys Fork of 
Green River, Wyo.; Uintatherium- Manteoceras- Mesa- 
tirhinus zone (Bridger C or D). 

Holotype. — "Represented by parts of the two 
maxUlary bones, which present the crowns of the third 
and fourth premolars, and of the second and third 
true molars, with the bases of the other molars and 
premolars." (Cope collection. Am. Mus. 5106.) 
(See fig. 98.) 

Characters. — Cope writes: 

Belonging to the genus Palaeosyops as understood by Marsh — 
that is, with two cones on the inner side of the last superior 
molar. The species is as large as the Limnohyus major of 
Leidy but differs in the relative proportions of the teeth. 




Measurements [Cope] 

Millimeters 

Length of true molar series (2.75 inches) 67 

Length of last molar 25 

Width of last molar 26 

Etymology. — fontinalis, of or from a spring, hence 
original; in allusion to the primitive characters. 

Present determination. — Cope was in error in inter- 
preting the teeth of this skull, which belong to a very 
juvenUe animal, the teeth exposed being the last 
upper mUk tooth, dp*, the first and second molars, 
m*, m^. The cranial characters, too, are very juve- 
nUe. So far as they serve to guide us, the animal 
probably belongs to the genus Palaeosyops, and also 
probably to a distinct species, from a low geologic 
level, possibly Bridger A. 



Figure 98. — Cope's type (holotype) of Palaeosyops diaconus 
Left upper teeth. Am. Mus. 5106. After Cope, 1885. One-half natural size. 

Thus the last three molars have the same anteroposterior 
length, while the space occupied by four premolars is shorter. 
The anterior and posterior cingula of the true molars are very 
strong, but it is not weU marked on the inner side between the 
cones. The latter are acutely conic, and the median anterior 
tubercle is strongly developed. Although the wearing of the 
teeth indicates maturity, the enamel is coarsely and obtusely 
rugose. The fourth premolar differs from that of L. major 
in its smaller size relatively and absolutely and in the presence 
of a prominent vertical tubercle on the outer face, rising to the 
angle of the deep notch between the lobes. The third premolar 
is as wide as the fourth and about as large as the corresponding 
tooth in L. major, but different from it in the absence of tubercle 
and ridge that mark its external face. The first premolar has 
two roots, and the canine is large and short. 

Measurements [Cope] 

Millimeters 

Length of entire molar series 171 

Length of true molars 106 

Length of last molar (crown) 42 

Width of last molar (crown) 43. 7 

In comparison with Marsh's description of his P. laticeps, 
the measurements are all larger, and the enamel is as rugose as 
in L. major, instead of smooth. The shortening of the pre- 
molar series is greater in P. diaconus; thus in P. laticeps the 
two sets of molars are related as 94 to 61 millimeters; in the 
present one, as 106:65; were the proportions similar, the length 
of the premolar series should be 69 millimeters. 



166 



TITANOTHERES OF ANCIENT "WYOMING, DAKOTA, AND NEBRASKA 



Etymology. — 5t?, double; kcows, cone; because the 
third upper molar had two mner cones. 

Present determination. — The name P. diaconus Cope 
is probably a synonym of Palaeosyops rohustus 
(Marsh), as explaiaed in Chapter V of this mono- 
graph. 

Diplacodon Marsh, 1875 

Cf. Diplacodon Marsh, this monograph, pages 155, 439 

Original reference. — Am. Jour. Sci., 3d ser., vol. 9, p. 
246, March, 1875, "dated February 20, 1875" (Marsh, 
1875.1). 

Type species. — Diplacodon elatus Marsh. (See 
p. 439.) 

Generic characters. — Marsh writes: 

The genus here established presents characters in some 
respects intermediate between Limnohyus and Brontotherium. 
It agrees with the former in its complete dentition (44 teeth) 
and in the general form of the incisors, canines, and true molars. 
It resembles the latter still more closely in the premolar and 
molar teeth, and parts of the skeleton, especially in the verte- 
brae, and bones of the extremities. From the Eocene Limno- 
hyidae, already described, this genus is sharply distinguished 
by the last upper premolar, which has two distinct inner cones, 
thus agreeing essentially with the first true molar. This char- 
acter, which has suggested the name of the genus, is one step 



^9 



Characters. — The specific characters were not for- 
mally separated from the generic characters above 
given under Diplacodon. 

Measurements [Marsh] 

Millimeters 

Extent of upper molar series 242 

Extent of upper true molars 152 

Anteroposterior diameter of last upper molar 60 

Transverse diameter 59 

Anteroposterior diameter of [upper] canine, at base 32 

Height of crown 27 

Etymology. — elatus, lofty; apparently in allusion 
either to the large size or to the advanced stage of 
evolution. 

Present determination. — This important genus and 
species was based upon an excellent type. The genus 
and species are vahd. (See p. 439.) 

Brachydiastematherium Bockh and Maty, 1876 

Cf. Brachydiastematherium Bockh and Maty, this monograph, 
page 382 

Original reference — Mittheilungen aus Jahrb. K. k. 
geol. Anstalt, Band 4, pp. 125-150, 1876 (1876.1). 

Type species. — Brachydiastematherium transilvani- 
cum Bockh and Maty. (See p. 382.) 




Figure 



-Marsh's type of Diplacodon elatus 
One-third natuial size. 



toward the modern type of perissodactyl dentition. The dental 
formula of the genus is the same as Limnohyus, viz, incisors f , 
canines \, premolars |, molars f. In other respects the teeth 
most resemble those of the Brontotheridae. From this family 
Diplacodon differs widely in its dentition and the absence of 
horns. 

Etymology. — SlwXoos, double; aK-q, a point; 65ovs, tooth; 
because the upper premolars had two inner cones. 

Present determination. — This genus is certainly 
valid so far as it applies to the type species. (See 
p. 439.) 

Diplacodon elatus Marsh, 1875 

Cf. Diplacodon elatus Marsh, this monograph, page 439 

Original reference. — Notice of new Tertiary mam- 
mals: Am. Jour. Sci., 3d ser., vol. 9, p. 246, March, 
1875; dated "February 20, 1875" (Marsh, 1875.1). 

Type locality and geologic horizon. — "Upper Eocene 
beds of Utah"; horizon probably Diplacodon-Proti- 
tanotherium-Epihippus zone (Uinta C, true Uinta for- 
mation). 

Holotype. — A palate with dentition nearly complete, 
parts of the skull and skeleton (Yale Mus. No. 11180). 



Generic characters. — The generic characters mingled 
with the specific characters are given below. (See 
also p. 382.) 

Etymology. — jSpaxvs, short; didaTrnxa, an interval; 
dr]piov, beast; in allusion to the short diastema 
between the lower canines and first premolars. 

Present determination. — The type of this genus is an 
animal closely similar in size and in stage of evolution 
to the Protitanotherium superhum of the upper Eocene 
of Utah but differs in certain characters, which are 
probably of generic value. (See p. 382.) 

Brachydiastematherium transilvanicum Bockh and Maty, 1876 

Cf. Brachydiastematherium transilvanicum Bockh and Maty, 
this monograph, pages 382, 941 

Original reference. — -Mittheilungen aus Jahrb. K. k. 
geol. Anstalt, Band 4, pp. 125-150, pis. 17, 18, 1876 
(1876.1). Cf. Toula, Akad. Wiss. Wien Sitzungsber. , 
Band 101, pp. 612 et seq., 1892 (1892.1). 

Type locality. — Andrashaza (Siebenbtirgen), Hun- 
gary (Transylvania, eastern Hungary, about 150 miles 
northeast of Belgrade). Collected in 1871 by Dr. 
Alex. Pavay. 



DISCOVERY OP THE TITANOTHEEES AND ORIGINAL DESCRIPTIONS 



167 



Geologic level. — The specimen, according to Pro- 
fessor Koch (Bockh, 1876.1, p. 149), was found in 
"buntes Thongebilde" of "lower" Eocene age, but 
the assignment of a form of this advanced stage to a 
level so low appears unwarrantable, and later evidence 
indicates that the age of this specimen is more prob- 
ably upper Eocene. (See p. 382.) 

Holotype. — Anterior part of lower jaw, containing 
incisors, canines, four premolars, and one molar. 
Originally described and defined by Bockh (1876.1) 
as a palaeotherioid. (See fig. 100.) 

Generic and specific characters. — Bockh and Maty 
(p. 148) write: 



einer dreieckigen Emailzunge. An der Krone sammtlicher 
Zahne sind die Reste einer diinnen cementartigen Kruste zu 
sehen. 

The following measurements are taken from the 
original figures: 

Millimeters 

I3, anteroposterior 22 

I3, transverse 20 

C, maximum anteroposterior diameter (horizontal measure- 
ment near base) 38 

C, maximum transverse 31 

C, lieight of crown (estimated) 40 

Postcanine diastema (at top) 12 

Pi-p4, anteroposterior 107 

Pi, anteroposterior 18 





Figure 100. — Type (holotype) lower jaw of Brachydiastematherium transilvanicum 

After Bockh and Maty, 1876. A', Side view; K', inner side; A', rear view of mi; A<, front view of mi; A^, top view of jaw; A', outer view of right lower canine; A', 
section of root of right lower canine; A^, fragment of right lower incisor. Two-fifths natural size. 



I3, mit flachliegelformiger Krone, welche mit warziger 
Emailwulst versehen ist; Ci, mit kegelformiger Krone, welche 
gleichfalls eine warzige, starke Emailwulst besitzt; seine 
Wurzel ist iiberaus stark, lang und gerade. Die Zahnliicke ist 
sehr kurz; pi, deren erster am kleinsten, und seine nur eine 
Wurzel besitzende Krone stellt nur einen einfachen Kegel dar; 
die iibrigen drei wachsen gradatim und die warzige Wulst der 
Basis fehlt an der inneren Seite dieser letzteren. Die drei 
letzten Praemolare ahmen wohl die Form der entsprechenden 
Zahne der echten Palaeotherien nach, wirkliche Halbmonde 
an der Oberflache seiner abgewetzten Krone zeigt indessen nur 
der vierte Praemolar; an den demselben vorangehenden zwei 
Zahnen kann die Verzierung noch niclit als Halbmond bezeich- 
net warden. Die hintere Bucht des vierten Praemolares 1st 
durch eine Scheide in zwei Theile abgetheilt, und heizu ist 
der Keim auch schon beim dritten Praemolar zu bemerken; 
m (?)3, die innere Seite des ersten echten Molares zeigt gleich- 
falls keine Emailwulst, an der Mitte der hinteren Seite des 
hinteren Halbmondes vereinigt sich indessen die Wulst mit 



P2, anteroposterior 26 

P2, transverse (through anterior lobe) (estimated) 17 

P3, anteroposterior 31 

P3, transverse (estimated) 22 

P4, anteroposterior 38 

P4, transverse (estimated) 28 

Ml, anteroposterior 50 

Ml, transverse (estimated) 30 

Etymology. — transilvanicum, Transylvanian. 
Present determination. — The species is probably 
valid. 

Leurocephaius Osborn, Scott, and Speir, 1878 

Cf. Telmatherium Marsh, this monograph, page 341 

Original reference. — ^E. M. Mus. Geol. and Arch. 
Princeton Coll. Contr. No. 1, p. 42, pi. 4, 1878 (Osborn, 
Scott, and Speir, 1878.3). 



168 



TITANOTHERES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



Type species. — LeurocepTialus cultridens Osborn, 
Scott, and Speir. (See p. 341.) 

Generic characters. — Scott and Osborn write: 

Upper incisors acute, with strong posterior ridges, lower in- 
cisors compressed and laniariform, canines compressed, with 
serrated cutting edges; first upper premolar with rudimentary 
anterior lobe, last upper molar with rudimentary postero- 
internal cusp. Molars constructed as in Palaeosyops but higher, 
with sharper cones and more erect external lobes. Internal 
median valley very much deeper. Little or no depression at the 
forehead; zygomatic arch round, comparatively straight and 
does not project outward, and with obscure postorbital 
process. Premaxillaries short and straight. Mandible with 
nearly straight lower margin and shallow masseteric fossa; 
mental foramen single. 

Etymology. — Xeupos, smooth; Ke(j>a\r], head; in 
allusion to the smooth texture of the bone. 

Present determination. — LeurocepTialus is a synonym 
of Telmatherium Marsh. 




Figure 101. — Type (holotype) of Leurocephalus cultridens 

Right prematiUa, maxilla, and mandibular ramus. Princeton Mus. IQ027. Atter Osborn, 187S. One-third 

natural size. 

Leurocephalus cultridens Osborn, Scott, and Speir, 1878 

Cf. Telmatherium cultridens (Osborn, Scott, and Speir), this 
monograph, page 341 

Original reference. — E. M. Mus. Geol. and Arch. 
Princeton Coll. Contr. No. 1, p. 42, pi. 4, 1878 (Osborn, 
Scott, and Speir, 1878.3). 

Subsequent reference. — Earle, A memoir on the genus 
Palaeosyops Leidy and its allies: Acad. Nat. Sci. Phil- 
adelphia Jour., 2d ser., vol. 9, pp. 343-348, pi. 10, 
tig. 3, 1892; type (Earle, 1892.1). 

Type locality and geologic horizon. — Henrys Fork 
divide, near Fort Bridger, Wyo.; Vintatherium- 
Manteoceras- Mesatirhinus zone (Bridger C or D). 

Holotype. — "Established on specimen having a 
nearly complete dentition and portions of the cran- 
ium" (Princeton Mus. 10027). (See fig. 101.) 

Specific characters. — The specific and generic char- 
acters were not distinguished in the original descrip- 
tion. 

Etymology. — culter, a knife; dens, tooth; in allusion 
to the sharp-edged reciurved canines. 

Present determination. — This genus is a synonym of 
Telmatherium Marsh. The species is valid. (See 
p. 341.) 



Palaeosyops borealls Cope, 1880 ' 

Cf. Eotitanops horealis (Cope), this monograph, pages 156, 292 

Original reference. — Am. Naturalist, vol. 14, p. 746, 
1880 (Cope, 1880.1). 

Subsequent reference. — Cope, Tertiary Vertebrata, 
p. 703, pi. 58a, fig. 3, 1884 [1885], (Cope, 1885.1). 

Type locality and geologic horizon. — "Badlands in the 
upper drainage basin of the Big Horn River in western- 
central Wyoming"; Wind River formation, horizon 
not determined, probably Lambdotherium-Eotitanops- 
Coryphodon zone (Wind River B). J. L. Wortman, 
collector. 

Holotype. — "Founded on a portion of the right 
maxillary bone, which supports the three true molars 
and one premolar" (Cope collection. Am. Mus. 4892). 
(See fig. 102.) 

Characters. — Cope writes : 
Size of Limnohyus fontinalis, or much 
smaller than P. laevidens. Anterior median 
tubercle well developed; anterior and pos- 
terior cingula strong, not rising to inner 
cones. A low ridge extending outward and 
forward from posterior cone. Enamel 
smooth. Differs from P. junior Leidy in 
the presence of the intermediate tubercle 
and crest and in the weak ex-ternal cin- 
gulum. Length of true molar series 63 [mil- 
hmeters]; diameters of first true molar, 
anteroposterior, 19; transverse, 20. 

Etymology. — borealis, relating to 
Boreas; in allusion to the Wind River 
formation. 

Present determination. — The species 
is valid but generically distinct from 
Palaeosyops. It is the type of the 
genus Eotitanops Osborn. (See p. 289 .) 

Lambdotherium Cope, 1880 

Cf . Lambdotherium Cope, this monograph, page 279 
Original reference. — Am. Naturalist, vol. 14, p. 746, 

1880 (Cope, 1880.1). 
Subsequent reference. — Cope, Tertiary Vertebrata, 

p. 710, 1884 [1885] (Cope, 1S85.1). 




Figure 102. — Type (holotype) of Palaeosyops borealis 

Right upper part of right maxilla with p<-m3. Am. Mus. 4892. After Cope, 

18S5. Natural size. 

Type species. — Lambdotherium popoagicum Cope. 
(Seep. 281.) 

Generic characters. — Cope writes : 

Dentition much as in Limnohyus, excepting that there is a 
diastema in front of the second inferior premolar. Presence of 
first inferior premolar not ascertained. Fourth inferior pre- 
molar without posterior cusps. Superior molars with an 



DISCOVERY OF THE TITANOTHERES AND ORIGINAL DESCRIPTIONS 



169 



angular ridge extending inward from each inner cusp. Last 
inferior molar with heel. * * * The V-shaped crests of the 
inferior molars separate it from Hyracotherium. 

Etymology. — Xa/i/35a, lambda; dTjplov, beast; in allusion 
to the A-shaped crests of the lower molars. 

Present determination. — The genus is valid and is 
now referred to the Eocene titanotheres. 

Lambdotherium popoagicum Cope, 1880 

Cf. Lambdotherium popoagicum Cope, this monograph, page 281 

Original reference. — Am. Naturalist, vol. 14, pp. 
746, 747, 1880 (Cope, 1880.1). 

Subsequent reference. — Tertiary Vertebrata, p. 710, 
pi. 58b, figs. 7 and 7a, 1884 [1885] (Cope, 1885. 1). 

Type locality and geologic Tiorizon. — Badlands of 
Wind Elver, western-central Wyoming; Lambdo- 
iherium-Eotitanojps-Coryphodon zone (Wind River B). 

Type. — A lower jaw with dentition (Am. Mus. 4863). 
(See fig. 103.) 

Specific cliaracters. — Cope writes : 

The heels of the second and third premolars have 
a median keel; the third only has an anterior 
tubercle. The crest of the heel of the fourth forms 
an imperfect V. Heel of the last true molar small. 
No cingula; enamel smooth. Length of molar series 
80 [millimeters]; of true molars 44; of last molar 
19; depth of ramus at first premolar 21; at last 
molar 31. * * * About the size of Hyrachyus 
agresiis. 

Etymology. — popoagicum, in allusion to 
Popo Agie River, a tributary of Wind River. 

Present determination. — The species is 
valid. (See p. 283.) 

Lambdotherium brownianum Cope, 1881 

Cf. EoHtanops brownianus (Cope), this monograph, 
page 292 

Original reference. — U. S. Geol. and Geog. 
Survey Terr. Bull., vol. 6, p. 196, 1881 
(Cope, 1881.2). 

Subsequent reference. — Cope, Tertiary 
Vertebrata, p. 709, pi. 56a, fig. 10 (not the 
type), 1884 [1885] (Cope, 1885.1). 

Type locality and geologic horizon. — Badlands of 
Wind River, western-central Wyoming; Lambdo- 
therium- Eotitanops-CorypJiodon zone (Wind River B). 

Holotype. — "The greater part of a lower jaw," 
with p^, m'-m' (Cope collection. Am. Mus. 4885). 
(See fig. 104.) 

Characters. — Cope writes: 

Considerably larger than the L. popoagicum and about 
equal to the Tapirus ierreslris. The greater part of a lower 
jaw represents the species, and on this, unfortunately, only 
one of the premolar teeth remains. The three premolars are 
all two-rooted, and the posterior lobe of the last true molar is 
well developed. The inferior part of the e.xternal side of the 
ramus contracts or retreats rather abruptly posteriorly, below 
the last molar. It presents a slight external convexity below 
the second and third premolars. The alveolar line rises rapidly 
101959— 29— VOL 1 14 



posteriorly, so that the last true molar is quite oblique. The 
second (first) premolar has a considerable heel, which is narrow 
and elevated on the middle line. The principal cusp is large 
and compressed but obtuse and has no anterior basal tubercle. 

Measurements [Cope] 

Millimeters 

Length of six molars 90 

Length of true molars 55 

Diameters of second (first) premolar: 

Vertical 9 

Anteroposterior 12 

Transverse 6 

Length of base of first true molar 15 

Width of base of first true molar 9 

Length of base of third true molar 23 

Width of base of third true molar 11 

Depth of ramus at second premolar 30 

Depth of ramus at ma: 

At front of tooth 39 

At end of tooth 47 

Etymology. — "Dedicated to my friend Arthur E. 
Brown, superintendent of the Philadelphia Zoological 
Garden" (Cope). 





FiGUEE 103. — T3'pe (holotype) of Lambdotherium popoagicum 
Left mandibular ramus, with pj-ms. Am. Mus. 4863. After Cope, 1885. Natural size. 

Present determination. — The species is valid, 
generic reference is to Eotitanops. (See p. 292.) 



The 



Palaeosyops hyognathus Osborn, 1889 

Cf. Dolichorhinus hyognathus (Osborn), this monograph, 
page 409 

Original reference. — Am. Philos. Soc. Trans., new 
ser., vol. 16, p. 513, 1890 [author's reprint issued 
Aug. 20, 1889; O. P. Hay] (Scott and Osborn, 1890.51). 

Subsequent reference. — Earle, A memoir upon the 
genus Palaeosyops Leidy and its allies: Acad. Nat. Sci. 
Philadelphia Jour., 2d ser., vol. 9, pi. 11, figs. 10, 
11 [type], 1892 (Earle, 1892.1). 

Type locality and geologic horizon. — Washaliie, White 
River, northeastern Utah; Washakie B. 



170 



TITANOTHERES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



Eolotype and specific characters. — Scott and Osborn 
write : 

In the Washakie beds is found a large species, about the 
same size as P. vallidens Cope, which is provisionally referred 
to Palaeosyops (P. hyognathus, sp. nov., Princeton collection. 
No. 10273). This is represented by a lower jaw seven-eighths 
as large as the type mandible of Diplacodon. [See fig. 105.] 



;-->;x..i4:: 




Figure 104. — Cope's type of Lainhdotherium brownianum 
One-half natural size. 

As in the latter, the incisors form a close procumbent series; 
the tips forming a gently arched line when seen from above. 
The symphj'sis is extremely long (11 centimeters) and shallow; 
the canines are rather small and semiprooumbent. The molar- 
premolar series measures 24.5 centimeters, the last molar 
measures 6.5 centimeters, the transverse measurement outside 
of the canines is 9.6 centimeters; in Diplacodon elalus 
the same measurement is 10 centimeters. Unfortu- 
nately, the premolar crowns are broken; it is probable 
that one or two of the premolars will be found to be 
like the molars. The characters of the chin and sym- 
physis are significant of close relationship to Dipla- 
codon elatus. 

Etymology. — vs, boar; yvados, jaw; in allusion 
to the forward-pointing lower incisors and 
shallow mandibular symphasis. 

Present determination. — The species is valid. 
The generic reference is to DolicliorTiinus. 
(See p. 409.) 

Liitinohyops Marsh, 1890 

Cf. Limnohyops Marsh, this monograph, page 303 

Original reference. — Am. Jour. Sci., 3d ser., 
vol. 39, p. 525, 1890 (Marsh, 1890.1). 

Type species. — Palaeosyops laticeps Marsh 
(Am. Jour. Sci., 3d ser., vol. 4, p. 122, 1872). 
(Seep. 311.) 

Generic characters. — Marsh says: 

In 1872 the writer described a large mammal from the Eocene 
of Wyoming under the name of Palaeosyops laticeps. As the 
name Palaeosyops has since been restricted, this species must 
be regarded as representing a distinct genus, which may be 
called Limnohyops. In this form the last upper molar has two 
inner cones, and in Palaeosyops, as now defined, there is only one. 



Etymology. — Xiixfrj, shore; vs, boar; &}//, face. 
Present determination. — This is a valid genus and 
species. For fuller descriptions, see page 303. 

Palaeosyops megarhinus Earle, 1891 

Cf. Mesatirhinus megarhinus (Earle), this monograph, 
page 388 

.--''" "~\ Original reference. — Am. Naturalist, 

vol. 25, No. 289, pp. 45-47, 1 fig., Jan- 
;-''' uary, 1891 (Earle, 1891.1). 

\__,,— ,^ Subsequent reference. — Earle, A mem- 

oir on the genus Palaeosyops Leidy and 
its allies: Acad. Nat. Sci. Philadelphia 
Jour., 2d ser., vol. 9, pp. 320-329, pi. 
10, fig. 2; pi. 11, figs. 4, 5, 1892 
(1892.1). 

Type locality and geologic horizon. — 
Washakie Basin of Wyoming; level unde- 
termined, probably Uintatherium- Man- 
teoceras- Mesatirhinus zone (Washakie A). 
Type.— "A fine skull (No. 10008) in 
the Princeton collection" (Earle). (See 
fig. 106.) 
Paratype. — Earle writes: 

There is also another portion of a skull (No. 10041), probably 
belonging to this species, with the occiput well preserved, from 
the Bridger proper [Earle, 1891.1, p. 45]. This paratype probably 
belongs to a more advanced species of this genus. (See p. 388.) 





Figure 105. — Type (holotype) of Palaeosyops hyognathus 
Incomplete lower jaw. Princeton Mus. 10273. After Earle, 1892. One-fifth natural size. 

Specific characters. — Earle writes: 

Cranium: The characters of this skull are quite unique 
and depart widely from any of the species of the family that 
I have examined. The general form of the skull is broad and 
depressed. Its dorsal contour is very like that of Palaeotherium 
crassum — namely, there is no frontal depression, which is so 
characteristic of Palaeosyops paludosus, and the occipital 



DISCOVERY OF THE TITANOTHERES AND ORIGINAL DESCRIPTIONS 



171 



region is only slightly higher than the frontal. The temporal 
fossae are not deeply excavated, and the occipital crests are 
weakly developed when compared to this region of the slcull in 
Limnohyops. The occiput itself is high and rather narrow. 
The foramen magnum is wide, bordered by very large condyles. 
The auditory processes are widely separated. The post- 
tympanics are broad and heavy. The postglenoid is peculiar 
in form; it is very short and thick; its form is very different 
from other species in the collection. An internal glenoid 
process is present in this species. The mastoid was probably 
exposed. The form of the zygomatic arch is striking; it is 
very light, nearly straight, with the temporal portion strongly 
compressed. The malar portion is also peculiar; the malar 
insertion is very abrupt and strongly depressed, With the 
external part very broad, thin, and shelf-like. The infra- 
orbital foramen is not exposed. The form of the malar in this 
species is totally different from all other allied forms 
that I have examined. The orbit is very small, termi- 
nates anteriorly above the anterior border of the second 
superior molar; the postorbital processes are well 
marked. The facial region of the skull is very short, 
compared to the total length of the cranium. The 
nasals are very long and heavy; their distal portion is 
expanded and broader than the middle part. The nasal 
notches are very deep and high. The premaxillaries are 
triangular in outline; their symphysis is short and nar- 
row, with a prominent anterior keel. The canine al- 
veolus is very prominent. The palate is long and nar- 
row, the roof of the same being strongly arched. The 
posterior termination of the palate is at the second 
superior molar. The incisive foramina are not divided. 
Teeth: The crowns of the teeth in this skull are 
badly damaged, but enough remains to give the total 
measurements and the characters of the last molar. 
The superior molars in this species form a continuous 
series, being not interrupted by a diastema. The 
sections of the incisors are very small. The canines 
are also very small and diverge widely. Only the 
second and third molar of each side are partially pre- 
served. They have a square form with low crowns; 
externally they are totally without a cingulum. The 
external V's are rather wide and angular, in this respect 
approaching that of Telmalotherium. The last molar 
is without any intermediate conules. 

Measurements 

Millimeters 
Length of skull, from premaxillary symphysis to end 

postglenoid ' 285 

Length from orbit to premaxillary symphysis 125 

Length from orbit to postglenoid 160 

Depth of nasal notch 84 

Length of nasals 100 

Entire molar series 148 

Last superior molar: 

Anteroposterior 37 

Transverse 39 

Etymology. — fie-yas, great, pis, nose; in allusion to the 
length of the nasal bones. 

Present determination. — This is a valid species which 
has been made the type of the genus MesatirJiinus 
by Osborn. (See p. 388.) 

Palaeosyops minor Earle, 1891 
Cf. Palaeosyops paludosus, this monograph, page 319 

Original reference. — Acad. Nat. Sci. Philadelphia 
Proc. for 1891, p. 112, issued March 31, 1891 (Earle, 
1891.2). 



Subsequent reference. — Earle, A memoir upon the 
genus Palaeosyops Leidy and its allies: Acad. Nat. 
Sci. Philadelphia Jour., 2d ser., vol. 9, pp. 269, 331, 
332, 1892 (1892.1). 

Earle's cotypes. — In his original description Earle 
says (1891.2, p. 112), "P. minor embraces specimens 
which Leidy erroneously described as P. paludosus, pi. 
4, figs. 3-6, of Leidy's report for 1873." In his memoir 
(1892.1, pp. 269, 330) Earle refers again to Leidy's 
Plate 4, Figures 3-6, as the types of P. minor, but on 
page 332 he says, "We may consider as the type 
specimen" the "beautifully preserved mandible fig- 




FiGURB 106. — Type (holotype) of skull of Palaeosyops megarhinus 
Princeton Mus. 10008. After Earle, 1892. No scale given. 

ured by Leidy" (Leidy, 1873.1, pi. 5, figs. 10, 11); 
and again on page 387 he states that the specimen 
figured in his (Earle's) Plate 12, Figure 14, is "the 
type of this species and is in the collection of the 
Academy of Natural Sciences of Philadelphia." But 
this specimen is apparently the same one figured in 
Leidy's Plate 4, Figure 5. (See fig. 107.) 
Specific characters. — Earle writes: 

Second superior premolar with two external lobes, external 
lobes of last superior premolar equal. Intermediate conules of 
true molars reduced, a strong external cingulum present. 

Etymology. — minor, in allusion to the relatively 
small size. 

Present determination. — Of the first-mentioned speci- 
mens (Leidy, 1873.1, pi. 4, figs. 3-6) Figures 3 and 4 
represent an upper dentition, which is probably con- 
specific with P. paludosus as determined in this mono- 
graph; hence if this is taken as Earle's type P. minor 
becomes a synonym of P. paludosus. 



172 



TITANOTHERES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



On the other hand, Leidy's Plate 4, Figures 5 and 6, 
represent an upper dentition of uncertain specific ref- 
erence. The "beautifully preserved mandible" 
(Leidy's pi. 5, figs. 10, 11) is probably referable to P. 
paludosus. Hence we may regard P. minor as a syno- 
nym of P. paludosus. 



Type locality and geologic horizon. — Cottonwood 
Creek, Bridger Basin, Wyo.; Palaeosyops paludosus- 
Orohippus zone (Bridger B). 

Holotype. — "A jaw, No. 10275 [Princeton Mus.], 
associated with a well-preserved radius, ulna, and two 
metacarpals." 








Ci 

FiGUKJs 107. — Earle's ootypes of Palaeosyops minor in the collection of the Academy of Natural 

Sciences of Philadelphia 

Ai, Lett maxilla with root of canine and premolar-molar series. After Leidy, 1873. Two-thirds natural size. Aj, The same; 
outer view of premolar-molar series. B, Another specimen; left upper premolar-molar series. After Leidy, 1873. Two- 
thirds natural size. (A reversed view of this specimen, which is of uncertain specific reference, was figured by Earle 
as the type (Earle's pi. 12, fig. 14).) Ci, Left mandibular ramus with p3-m3. After Leidy, 1873 (pi. 5, fig. II). One-halt 
natural size. C2, The same, pj-ma; crown view. After Leidy, 1873 (pi. 5, fig. 10). One-half natural size. The last two 
specimens are referable to Palaeosyops paludosus. 



Palaeosyops longirostris Earle, 1892 

Cf. Palaeosyops longirostris Earle, this monograph, page 319 

Original reference. — Acad. Nat. Sci. Philadelphia 
Jour., 2d ser., vol. 9, p. 338, 1892 (Earle, 1892.1). 



Characters. — Earle writes: 

The type jaw of this species, with the parts of the skeleton 
associated with it, was referred by Scott and Osborn [Osborn, 
1878.3, pp. 37, 38] to our P. minor (equal, in part, to P. palu- 
dosus Leidy). After comparing Leidy's type specimen [prob- 



DISCOVERY OF THE TITANOTHERES AND ORIGINAL DESCRIPTIONS 



173 



ably the specimen figured in Leidy's memoir of 1873, pi. 5, 
fig. 11] with this jaw, I find that there is such a marked difference 
in some of its characters that I have to give it a specific rank. 
The following characters distinguish it from Leidy's type: (1) 
The great posterior extension of the jaw behind the last molar 
(this is a unique character of this jaw — I have not observed it 
in any other species of this subfamil}') ; (2) the symphysis is 
much more elongated than in P. minor; (3) the lower border is 
straighter and less inflected than in P. minor; (4) the posterior 

en'f 




Figure 108. — Earle's type of Palaeosyops longirostris 
Princeton Mus. 1027-5. One-foarth natural size. 

tubercle of the last inferior molar is much larger than in the 
last named species; (5) the V's of premolar 4 are not so well 
developed as in P. ininor, and there is also a well marked dif- 
ference in the size of the first molars of the two species. 

In this jaw the first true molar is con- 
siderably smaller than in P. minor. The 
canine is very large and semiprooumbent, 
its position in the jaw resembling that of 
T. hyognathus. 

Etymology. — longus, loia^g; rostrum, 
bill, snout (hence, in this instance, 
jaw); in allusion to the great 
posterior extension of the j aw behind 
the last molar. (Earle.) 

Present determination. — This prob- 
ably valid species is certainly refer- 
able to the Palaeosyopinae and 
probably to Palaeosyops. (See 
p. 319.) 

Telmatotherium diploconum Osborn, 1895 

Cf . Rhadinorhinus diploconus (Osborn) , this 
monograph, page 431 

Original reference. — Am. Mus. 
Nat. Hist. Bull., vol. 7, p. 85, fig. 6, 
1895 (Osborn, 1895.98). 

Type locality and geologic horizon. — 
NoTthe&sternlJtah;" Telmatotherium 
cornutum" beds, Eohasileus-Doli- 
chorhinus zone (Uinta B). 

Holotype. — "The type is a skull 
(No. 1863) [Am. Mus.] in which the 
nasals are wanting and the mid- 
region of the cranium was ci'ushed." 
(See fig. 109.) 

Characters. — Osborn writes: 

Superior premolar-molar series, 174 millimeters. A large 
hypocone upon last upper molar. Nasofrontal without horn. 
Long sagittal crest. Canines small, rounded. 

This species differs from T. megarhinum in the absence of the 
infraorbital shelf and in the presence of a large hypocone upon 
the last upper molar. The premolar-molar dentition is similar in 
size and form to that of T. cultridens, but there are the following 
important general differences: (1) Canines small and circular in 
section; (2) a very short diastema, if any, behind the canine; 



(3) a large hypocone upon m'; (4) the infraorbital foramen 
close beneath the anterior border of the molar [malar]. [Com- 
parisons with T. cultridens follow.] 

Etymology. — StxXoos, double; Kcofos, cone; in allu- 
sion to the presence of two internal cones on the third 
upper molar. 

Present determination. — The species is valid; it is 
now referred to the genus Rhadinorhinus. (See p. 431.) 

Telmatotherium cornutum Osborn, 1895 

Cf. Dolichorhinus hyogiiathus (Osborn), this monograph, page 409 

Original reference. — Am. Mus. Nat. Hist. Bull., vol. 
7, p. 90, figs. 10, 11, 1895 (Osborn, 1895.98). 

Type locality and geologic horizon. — Northeastern 
Utah; "Telmatotherium cornutum beds," Eohasileus- 
Dolichorhinus zone (Uinta B 2). 

Holotype and paratypes. — Osborn writes: 

The type of this species is a fine skull (No. 1851) [Am. Mus.], 
while several other well-preserved skulls from the same levels 
give us all the cranial characters and the superior dentition 
(Nos. 1850, 1847, 1848, 1852, 1837). [See fig. 110.] 




Figure 109. — Type (holotype) of Telmatotherium diploconum 

Superior and lateral views of skull. The nasals are broken off. Am, Mus. 1863. After Osborn, 1895. 
One-fourth natural size. 

Characters. — Osborn writes: 

Incisors f . Premolar-molar series, 208 millimeters. A nar- 
row diastema. Upper canines lanceolate. Long premaxillary 
symphysis. A well-developed nasofrontal protuberance. Top 
of cranium completely flattened. No sagittal crest. An 
infraorbital process upon malar. 

This species is remarkable for its very long flat-topped cranium 
and its incipient knoblike osseous horns borne chiefly upon the 
nasals but partly upon the frontals. These horns project 
laterally and rise slightly above the general surface, and are best 



174 



TITANOTHERES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



seen in the anterior view, Figure 110. Ttiese characters and 
the absence of the frontoparietal and interparietal sutures all 
point well toward Titanotherium, but the premolars are still 
absolutely simple, showing no trace of the postero-internal cusps 
which characterize Diplacodon elatus. 

Other striking peculiarities are the upward-arching mid- 
cranial region, the extremely long, narrow, and laterally de- 
curved nasals; the strong infraorbital shelf upon the molars 
[malars] (seen also in T. megarhinum) , the slender zygomatic 
arch, the low occiput, the backward extension of the posterior 
nares by the palatines, and the partial inclosing of the roof of 
the pharynx by the pterygoids. 



Sphenocoelus Osborn, 1895 

Cf. Sphenocoelus Osborn, this monograph, page 417 

Original reference. — Am. Mus. Nat. Hist. Bull., vol. 
7, pp. 98-102, 1895 (Osborn, 1895.98). 
Generic characters. — Osborn writes: 

The distinctive features of the skull may therefore be summed 
up as follows: Deep paired pits in the aUsphenoids, and orbito- 
sphenoids upon either side of the thin presphenoid [basisphenoid] ; 
a long alisphenoid canal; foramen ovale widely separated from 




Figure 110. — Type (holoty 
Side, front, and top views of skull. Am. Mus. 

This general description of character was followed 
by a more detailed description. 

Etymology. — cornutus, horned; in allusion to the 
osseous "horns." 

Present determination. — Comparison of the lower 
jaw with the type of Palaeosyops hyognathus Osborn 
indicates that the species T. cornutum is a synonym of 
P. hyognathus, a species which is now referred to the 
genus Dolichorhinus. 



pe) of Tebnatotherium cornutum 

1851. After Osborn, 1S95. One-fourth natural size. 

1 for. lac. medium; condjdes very broad; foramen magnum 
large; occipital crest extending anteriorlj' into a short sagittal 
crest with convex sagittal ridges; skull apparently long and 
narrow. 

Etymology. — <T4>riv, a wedge; koIXos, hollow; in allusion 
to the paired cavities in the basisphenoid bone. 

Present determination. — This is a valid genus of 
Eocene titanotheres related to the long-skulled 
Dolichorhinus. (See p. 417.) 



DISCOVERY OF THE TITANOTHERES AND ORIGINAL DESCRIPTIONS 



175 



Sphenocoelus uintensis Osborn, 1895 

Cf. Sphenocoelus uintensis Osborn, this monograph, page 419 

Original reference. — Am. Mus. Nat. Hist. Bull., vol. 
7, pp. 98-102, figs. 12-15, 1895 (Osborn, 1895.98). 

Type locality and geologic horizon. — Northeastern 
Utah; Metarhinus zone (Uinta B 1). 

Holotype. — "Represented by the posterior portion 
of a slvull" (Am. Mus. 1501). (See fig. 111.) 



convex sagittal ridges. The occiput is rather broad, and below 
it are two widely set occipital condyles which are directed 
obliquely downward and backward. On either side of these 
the exoccipitals extend down into obtuse paroccipital processes, 
which are closely joined to the post-tympanics. The external 
auditory meatus is open inferiorly. In front of this the post- 
glenoid process faces somewhat inward; the glenoid facet is 
L-shaped, two narrow arms extending out upon the squamosal, 
and a broad arm descending upon the postglenoid. The dis- 
tinctive feature of the zygoma is the presence of a deep depres- 
sion just behind the lateral arm of the glenoid facet. 




FiGUKE 111. — Type (holotype) of Sphenocoelus uintensis 
Posterior half of cranium. Am. Mus. 1601. After Osborn, 1895. 



, Basal view; Aa, top view; As, occipital view; 
natural size. 



fiew of left side. One-third 



Specific characters. — Osborn writes: 

This new genus is represented by the posterior portion of a 
skull, which is distinct from any cranium known to the writer. 
Its most distinctive feature is the presence of a pair of pits in 
the floor of the skull upon either side of the narrow presphenoid 
[basisphenoid]. These pits were at first mistaken for the for. 
lac. media, but more careful investigation shows that they are 
roofed over by bone and apparently do not communicate at 
all with the cranial cavity. The pit on the right side is per- 
fectly preserved and clearly exhibits these characters. The 
pits are 42 millimeters long, 14 millimeters wide, and' 2 milli- 
meters deep. 

The skull has a long, narrow cranium surmounted posteriorly 
by a sagittal crest, which diverges anteriorly into two decidedly 



Skull measurements 

Millimeters 

Width across zygomatic arches 230 

Height of occiput 142 

Breadth 117 

Breadth of occipital condyles 130 

Basioccipital to top of sagittal crest 114 

The foramina of the skull are related to those of the Peris- 
sodactyla, for there is a long alisphenoid canal, upon the outer 
side of the anterior opening of which is the foramen. Just 
behind the posterior opening of the canal is the foramen ovale, 
and between these foramina are the two pits above mentioned. 
This foramen is separated by a very wide plate of bone from the 
for. lac. medium, which is partly filled by the periotic mass. 



176 



TITANOTHERES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



Etymology. — uintensis, from Uinta, in allusion to the 
Uinta Basin. 

Present determination. — This is a valid genus and 
species. (See p. 419.) 

Protitanotherium Hatcher, 1895 

Cf. Protitanoiheriiim Hatcher, this monograj)h, page 374 
Original reference. — Am. Naturalist, vol. 29, p. 1084, 
December, 1895 (Hatcher, 1895.1). 



would seem that Professor Marsh's conclusion is entirely con- 
jectural, since his material does not show whether there were 
horns or not. The present skull has a well-developed pair of 
frontonasal horns, and, since it agrees in all the characters 
known to that genus, I have preferred to refer it to that genus 
rather than to propose for it a new one on the strength of this 
purely conjectural character ascribed to Diplacodon by Pro- 
fessor Marsh. Should future discoveries show that there are 
hornless forms with the same dental characters as Diplacodon, 
it will then be necessary to establish for the present specimen a 
new genus, which may be called Protitanotherium. 




FiGUKE 112. — Type (holotype) of Diplacodon emarginatus 

Facial part of skull and anterior part of mandible. Princeton Mus. 11242. After Hatcher, 1895. Ai, Az, Aj, Side, top, and front views of 
skull; Bi, B2, B3, side, top, and front views of mandible. One-fourth natural si:e. 



Type species. — Diplacodon emarginatus Hatcher. 
Generic characters. — Hatcher writes: 

In referring this skull to Diplacodon, I have been compelled 
to ignore certain characters ascribed to that genus by Professor 
Marsh. That author, in speaking of the relations of this genus 
to the Titanotheriidae (Brontotheridae) , in his original descrip- 
tion of the type specimen, says (Marsh, 1875.1, p. 24) : "From 
this family, Diplacodon differs widely in its dentition and the 
absence of horns." In describing Diplacodon as hornless, it 



Etymology. — pro, before; Titanotherium — that is, 
forerunner of Titanotherium. 

Present determination. — It is not yet settled whether 
Diplacodon elatus Marsh had horns or not, but it is 
now believed that even if this character is set aside 
D. elatus is generically distinct from D. emarginatus, 
and we may therefore regard Hatcher's Protitanothe- 
rium as a valid genus. 



DISCOVERY OF THE TITANOTHERES AND ORIGINAL DESCRIPTIONS 



177 



Diplacodon emarginatus Hatcher, 1895 

Cf. Protitanotherium emarginatum Hatcher, this monograph, 
page 377 

Original reference. — Am. Naturalist, vol. 29, pp. 
1084-1087, pi. 38, figs. 1-4, December, 1895 (Hatcher, 
1895.1). 

Type locality and geologic horizon. — Found by J. B. 
Hatcher "near the base of the Diplacodon elatus beds 
[Uinta C of Osborn], in the upper Eocene or Uinta of 
Marsh. The locality is about 8 miles north of White 
River and 25 miles east of Ouray Agency, Utah, and is 
locally known as Kennedy's Hole." 

Holotype. — A skull with lower jaw (Princeton Mus. 
11242). The anterior part only of the skull is well 
preserved. (See fig. 112.) 

Characters. — Hatcher writes: 

The present species is at once clistinguished from D. elatus by 
its greater size, as is shown by a comparison of the length of the 
premolar and molar series, which is* 310 millimeters in the 
former and 242 in the latter. 

In general appearance the cranium of D. emarginatus is re- 
markably like some of the smaller forms of Titanotherium. 

Etymology. — emarginatus, referring to the emargi- 
nate form of the distal end of the nasals. 

Present determination. — The species is valid but 
generically distinct from Diplacodon Marsh and is now 
referred to ProtitanotTierium Hatcher. (See p. 377.) 

Manteoceras Hatcher, 1895 

Cf. Manteoceras Hatcher, this monograph, page 362 

Original reference. — Am. Naturalist, vol. 29, p. 1090, 
1895 (Hatcher, 1895.1). 

Type species. — By designation Telmatotherium valli- 
dens (of Osborn, not Palaeosyops vallidens Cope) = 
Manteoceras manteoceras Hay ex Osborn, MS., 1902. 

Hotelier's description — Hatcher writes: 

The genus Telmatotherium as it now stands should be divided, 
since it embraces at least three distinct forms. The type of T. 
vallidens should be removed from that genus and made the 
type of a new genus. This new genus may be called Man- 
teoceras, as suggested by Wortman from the field; it would be 
distinguished from Telmatotherium by the absence of the infra- 
orbital shelf, the stronger and more expanded zygomata, and 
the concave superior aspect of the skull and incipient fronto- 
nasal horns. 

In the above passage the reference to "the type of 
T. vallidens" if taken by itself would lead one to regard 
"Palaeosyops" vallidens Cope as the type of the genus 
Manteoceras Hatcher. But a careful study of 
Hatcher's full text and a knowledge of the history 
of the subject proves that Hatcher had in mind the 
"Telmatotherium vallidens" of Osborn, not of Cope: 
because (a) Hatcher refers to his Plate 29, Figure 2, 
as "Telmatotherium vallidens," and this figure is 
copied from Osborn's " Telmatotherium vallidens," 
Figure 7; (6) these figures represent Wortman's 
original "prophet horn" skull, to which he had 
applied the name Manteoceras "in a letter from the 



field" (Osborn); (c) the generic characters assigned by 
Hatcher refer most clearly to this skull and are utterly 
inapplicable to Telmatotherium {"Palaeosyops") valli- 
dens Cope, in which only the dentition and not the 
skull is known. 

Thus the type of the genus Manteoceras Hatcher is 
Telmatotherium vallidens of Osborn not Cope, which 
is equivalent to Manteoceras manteoceras Hay ex 
Osborn MS. The generic name can not be credited 
to Wortman, because he never published it, although 
Osborn (1895.98), mentions it as a manuscript name. 

Etymology. — juavrtj, prophet; Ktpas, horn; in allu- 
sion to the incipient "horns" above the orbits. 

Present determination. — This valid genus is fully de- 
scribed on page 362. 

Dolichorhlnus Hatcher, 1895 

Cf. DoKchorhinus Hatcher, this monograph, page 396 

Original reference. — Am. Naturalist, vol. 29, p. 1090, 
1895 (Hatcher, 1895.1). 

Type species. — Telmatotherium cornutum Osborn. 
Characters. — Hatcher writes: 

The genus Telmatotherium as it now stands should be divided, 
since it embraces at least three quite distinct forms * * * 
The type of T. cornutum should also be made the type of a new 
genus which may be called Dolichorhinus; it would be dis- 
tinguished from Manteoceras and Telmatotherium by the 
reduced number of inferior incisors, presence of incipient 
horns, presence of infraorbital shelf, and position of posterior 
nares. 

Etymology . — SoXixos, long; pis, nose. 
Present determination. — This is a valid genus. 
p. 396.) 



(See 



Palaeosyops ultimus Matthew, 1897 (ex Osborn MS.) 

Cf. Telmatherium ultimum Osborn, 1908, this monograph, 
page 345 

Original reference. — Am. Naturalist, vol. 31, pp. 
57-58, 1897 (Matthew, 1897.1). 

Subsequent reference. — Bibliography and catalogue 
of the fossil Vertebrata of North America: U. S. 
Geol. Survey Bull. 179, p. 631, 1902 (Hay, 1902.1). 

Doctor Matthew had no intention of describing a 
new species. He merely stated incidentally that 
P. ultimus, as established in manuscript by Osborn, 
and P. paludosus both have a short-necked astragalus. 
No type was mentioned, and the single character 
given does not separate the species from P. paludosus. 
Hence "Palaeosyops ultimus Matthew" (cited by 
Hay, 1902, p. 631) remained a nomen nudum until 
the type was fixed by Osborn in 1908. (See p. 345.) 

Etymology. — ultimus, last, latest; in allusion to the 
relatively late geologic horizon and to the apparent 
extinction of the race. 

Palaeosyops manteoceras Matthew, 1899 (ex Osborn MS.) 
Cf. Manteoceras manteoceras Hay, this monograph, page 395 
Original reference. — Am. Mus. Nat. Hist. Bull., 
vol. 12, p. 47, 1899 (Matthew, 1899.1). 



178 



TITANOTHERES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



In this faunal list the present specific name is 
merely mentioned. No type is specified, and no 
characters are given, so that Palaeosyops manteoceras 
Matthew was a nomen nudum until the type was 
designated by Hay in 1902. (See p. 365.) 

Etymology. — /xavris, prophet; Ktpas, horn; in allusion 
to the incipient "horns" above the orbits. 

Telmatotherium diploconum var. minus Matthew, 1899 

(Nomen nudum) 

Original reference. — Am. Mus. Nat. Hist. Bull., 
vol. 12, p. 50, 1899 (Matthew, 1899.1). 

In Matthew's faunal list " Telmatotherium diploconum 
var. minus" is recorded but not defined, and no type 
is specified. Hence Telmatotherium diploconum minus 
is a nomen nudum. 

Etymology. — minus, less; because smaller than the 
tyjDical T. diploconum. 

Canis? marshii Hay, 1899 

Of. Palaeosyops major? Leidy, this monograph, page 321 

Original reference. — Science, new ser., vol. 10, p. 253, 
1899 (Hay, 1899.1). Founded on "Canis montanus" 
Marsh (see p. 158), which was preoccupied by Canis 
montanus Pearson. 

Etymology. ^-'Named in honor of Prof. O. C. Marsh. 

Present determination. — As explained above, the 
type of Canis montanus Marsh (not Pearson) and 
Canis? marshii Hay is a second lower premolar of an 
Eocene titanothere, probably Palaeosyops paludosus 
or P. major. Canis? marshii Hay is therefore either 
indeterminate or a synonym of P. paludosus or P. 
major. 

Manteoceras manteoceras Hay, 1902 (ex Osborn MS.) 

Cf . Manteoceras manteoceras Hay, this monograph, pages 365-370 

Original reference. — U. S. Geol. Survey Bull. 179, 
p. 632, 1902 (Hay, 1902.1). 

Lectotype. — A skull (Am. Mus. 1569) lacking the 
dentition, described and figured by Osborn as "Telma- 
totherium vallidens" (Osborn, 1895.98, pp. 87-90, figs. 
7-8). (See fig. 113.) 

Paratype. — An incomplete skull (Am. Mus. 1570) 
with dentition (op. cit., fig. 9) from the same locality 
and level. 

Type locality and geologic horizon. — Washakie Basin, 
Wyo.; discovered by J. L. Wortman, of the American 
Museum Bridger expedition of 1893, "in a brown 
layer of sandstone 3 miles north of the base of Haj''- 
stack Mountain, upon Bitter Creek" (op. cit., p. 87). 

Uintatherium- Manteoceras- Mesatirhinus zone (Wash- 
akie A). 

. Hay's type. — We have seen above that the name 
Palaeosyops manteoceras Matthew (ex Osborn MS.) 
was a nomen nudum, because no type had been desig- 
nated. The type was for the first time clearly indi- 
cated by Hay (1902.1, p. 632), who refers to this 



species Hatcher's (1895.1) Plate 39, Figure 2 (p. 368, 
this monograph) and Osborn's (1895.98) Figures 7-9 
(pp. 366, 368). These are clearly the same two "prophet 
horn" skulls (Wortman's first "Manteoceras" speci- 
mens) that had been at first erroneously referred by 
Osborn to "Telmatotherium vallidens Cope." Of these 
two skulls. Am. Mus. 1569 — that is, Osborn's Figures 
7, 8 and Hatcher's Figure 2 (copied from Osborn's 
fig. 7) — may be taken as the lectotype. 

The generic name Manteoceras and the specific 
name manteoceras were first brought together by Hay 
in the reference now under consideration. 

Specific characters. — In Osborn's original descrip- 
tion (Osborn, 1895.98, p. 87) these skulls were errone- 
ously identified as conspecific with the type of Palaeo- 
syops vallidens Cope, under the name "Telmatotherium 
vallidens Cope." The specific characters given by 
Osborn were as follows: 

Superior premolar-molar series, 184-220 millimeters. A 
narrow diastema. Molar cusps less elevated. A rudimentary 
nasofrontal tuberosity. Premaxillary symphysis short. Top 
of cranium flattened; very short bifid sagittal crest. 

Etymology. — fiavrLs, prophet; Kepas, horn; in allu- 
sion to the incipient "horns" above the orbits. 

Present determination. — The species is a valid one 
and is fully described on pages 365-370. 

Lambdotherlum primaevum Loomis, 1907 

Cf . Lambdotherium -primaevum Loomis, this monograph, page 283 

Original reference. — Am. Jour. Sci., 4th ser., vol. 23, 
p. 363, fig. 2, May, 1907 (Loomis, 1907.1). 

Type locality and geologic horizon. — Buffalo Basin, 
near Meeteetse, Wyo. "Wasatch beds of the Big 
Basin." Horizon regarded by Loomis as equivalent 
to the base of the Wind Eiver formation — that is, 
the Heptodon-Coryphodon-Eohippus zone (Wind River 
A). 

Holotype. — Amherst Mus. 254, "consisting of upper 
molars 1 and 2 of the right side and lower molars 1, 2, 
and 3 from the same side, the specimen being from the 
Buffalo Basin, near Meeteetse, Wyo. This species is 
fau'ly abundant at this horizon and is intermediate in 
size between L. hrownianum and L. popoagicum." 
(See fig. 114.) 

Characters. — Loomis writes: 

On the upper molars the parastyle, though strong, is not so 
well developed as in the foregoing forms; the paraoonule is well 
developed, but the metaconule is so annexed to the metaoone as 
to appear like a buttress of this cusp. The second molar 
measures 12 milUmeters transversely [anteroposteriorly] by 17 
millimeters lengthwise [transversely]. The robust lower molars 
have the protoconid markedly bifid, while the paraconid and 
hypoconid are each high crescents. The heel of the last molar 
is a high shallow basin completely surrounded by an outer rim. 
The three molars occupy 41 milhmeters. 

The brackets above indicate that in the foregoing 
description the measurements of the molar teeth 
have been inadvertently transposed. The description 



DISCOVERY OF THE TITANOTHERES AND ORIGINAL DESCRIPTIONS 



179 



should read: "Second superior molar, antei'oposterior, 
12 millimeters; transverse, 17 millimeters." 

Etymology. — primaevus, earliest in age; in allusion to 
the supposedly low geologic horizon. 

Present determination. — Provisionally recognized as 
a valid species. 



Mus. Nat. Hist. Bull., vol. 24, pp. 600, 601, 1908 
(Osborn, 1908.318). 

Type species. — Palaeosyops horealis Cope. 

Generic characters (Osborn, 1908.318, p. 601). — 
Superior molars subquadrate and rounded in form; 
conules reduced, sublophoid; m'-m^ 63 millimeters 




Figure 113. — Cotypes of Manteoceras manteoceras {Telmalotheriam vallidens) 

After Osborn. Ai, Composite Am. Mus. 1569, 1570; side view otslsull; As, Am. Mus. 1669 (lectotype), superior view ot slcull; B, Am. 
Mus. 1570, superior view of slcull. All one-fourth natural size. 



Eotitanops Ogborn, 1907 

Cf. Eotitanops Osborn, this monograph, page 289 

Original reference. — Am. Mus. Nat. Hist. Bull., vol. 
23, p. 242, 1907; type species designated (Osborn 
1907.294). 

Subsequent reference. — Osborn, New or little-known 
titanotheres from the Eocene and Oligocene: Am. 



(estimated). Inferior molars without metastylids. 
Hypoconulid of m' subconic. Fii'st inferior premolar 
present. Manus tetradactyl, functionally tridactyl 
with a tendency to mesaxonic structure. From Wind 
River formation. 

Etymology. — ^cbs, dawn; Tltclv, a titan; w^, face — ■ 
that is, first of the titanotheres. 

Present determination. — This genus is valid. (See 
p. 289.) 



180 



TITANOTHERES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



Limnohyops priscus Osborn, 1908 

Cf. Limnohyops priscus Osborn, this monograph, page 306 

Original reference. — Am. Miis. Nat. Hist. Bull., 
vol. 24, pp. 601-602, fig. 5, 1908 (Osborn, 1908.318). 

Type locality and geologic horizon. — Grizzly Buttes, 
Bridger Basin, Wyo.; Palaeosijops paludosus-Oroliippus 
zone (Bridger B 2). 





A 

Figure 114. — Type (holotype) of Lambdotherium 
primaevum 

Amherst Mus. 254. After Loomis, 1907. A, Right upper molars 1 and 
2; B, right lower molars (mi-ms). Natural size. 

Holotype. — A crushed skull with excellent dentition 
(Am. Mus. 11687), discovered by the American 
Museum expedition of 1903. (See fig. 115.) 




Figure 115. — Type (holotype) of Limnohyops priscus 
Am. Mus. 11687. Pi-m' left. After Osborn, 1908. One-half natural size. 

Characters. — Osborn writes: 

P'-m', 148 (type) to 161 millimeters. Distinguished from 
the contemporary Limnohyops laevidens Cope by its larger 
size and by the more progressive character of pm^-pm'. 
Second superior premolar obliquely elongate with a very rudi- 
mentary tritocone. Large hypocone on m'. 

Etymology. — priscus, ancient, in allusion "to the 
low geological level and primitive characters of this 
species." (Osborn.) 

Present determination. — The species and generic 
reference are probably valid. (See p. 306.) 




Figure 116.- 



-Tj'pe (holotype) skull of Limnohyops 
matthewi 



Am. Mus. 11684. After Osborn, 1908. One-fourth natural size. 

Limnohyops matthewi Osborn, 1908 

Cf. Limnohyops matthewi Osborn, this monograph, page 308 

Original reference. — Am. Mus. Nat. Hist. Bull., vol. 
24, p. 602, fig. 6, 1908 (Osborn, 1908.318). 



Type locality and geologic horizon. — Grizzly Buttes, 
Bridger Basin, Wyo. ; Palaeosyops paludosus-Orohippus 
zone (Bridger B 2). 

Holotype. — A skull (Am. Mus. 11684) lacking the 
anterior portion and dentition. Discovered by the 
American Museum expedition of 1903. (See fig. 116.) 

Specific characters. — Osborn writes: 

Intermediate in size between L. laevidens and L. monoconus. 
M' of small size with large hypocone and quadrate inner half 
Occiput very high and narrow. Cranial portion of skull greatly 
abbreviated, bringing post-tympanic and postglenoid processes 
into broad union. Temporal openings subcircular as defined 
by zygomatic arches. 

Etymology. — Named "in honor of Dr. W. D. 
Matthew, of the American Museum staff." (Osborn.) 

Present determination. — The species is probably 
valid. (See p. 308.) 




Figure 117. — Type (holotype) skull of Limnohyops monoconus 
Am. Mus. 11679. After Osborn, 1908. One-fourth natural size. 

Limnohyops monoconus Osborn, 1908 

Cf. Limnohyops monoconus Osborn, this monograph, page 309 

Original reference. — Am. Mus. Nat. Hist. Bull., vol. 
24, p. 603, fig. 7, 1908 (Osborn, 1908.318). 

Type locality and geologic horizon. — Grizzly Buttes, 
Bridger Basin, Wyo. ; Palaeosyops paludosus-Orohippus 
zone (Bridger B 2). 

Holotype. — A crushed skull with dentition (Am. 
Mus. 11679). Discovered by Mr. Quackenbush, of 



DISCOVERY OF THE TITANOTHERES AND ORIGINAL DESCRIPTIONS 



181 



the American Museum expedition of 1903. (See fig. 
117.) 
Specific characters. — Osborn writes: 

M^ without hypocone, roundly triangular in form, with 
broadly extended ectoloph and parastyLe. P^-m' 150, p'-m^ 
163 milUmeters. Condyle to incisive border 510. Occiput 
very high, cranium relatively elongated, with space (4 miUi- 
meters) between post-tympanic and postglenoid processes. 
Temporal openings as defined by zygomatic arches elongate. 




Figure 118. — Type (holotype) skull of Palaeosyops leidyi 
Inferior view. Am. Mus. 1544. After OsbDrn, 1908. One-fourth natural size. 

Etymology. — yibvo's, single; kccws, cone; named "in 
reference to the presence of but a single cone on the 
inner side of the third superior molar, an exceptional 
condition in the genus Limnohyops." (Osborn.) 

Present determination. — The specific and generic 
references are probably valid. (See p. 309.) 

Palaeosyops leidyi Osborn, 1908 

Cf. Palaeosyops leidyi Osborn, this monograph, page 323 

Original reference. — Am. Mus. Nat. Hist. Bull., 
vol. 24, p. 604, fig. 8, 1908 (Osborn, 1908.318). 

Type locality and geologic horizon. — Henrys Fork, 
Bridger Basin, Wyo.; Uintatherium- Manteoceras- Mesa- 
tirliinus zone (Bridger C 2 to C 4). Discovered by the 
American Museum expedition of 1893, under Dr. 
J. L. Wortman. 

Holotype. — A well-preserved skull (Am. Mus. 1544) 
associated with considerable portions of the skeleton. 
This specimen, which is associated with a considerable 



portion of the postcranial skeleton, is now mounted 
in the American Museum, the missing parts having 
been supplied from other individuals. (See p. 323; 
Pis. XXVII, L, LXI; and fig. 118.) 
Specific characters. — Osborn writes: 

Of larger size; total length of skull 415 millimeters; p'-m^, 158; 
P2-m3, 168; diastema behind canines; p-', p^ superior, with 
mesostyles. Barely defined sweUings representing the rudi- 
ments of osseous frontonasal horns. 

Etymology. — Named "in honor of Joseph Leidy, 
the discoverer of the family and [founder] of the 
genera Palaeosyops, Titanotherium, and Megacerops." 
(Osborn.) 

Present determination. — The species is probably 
valid. 

Palaeosyops granger! Osborn, 1908 

Cf. Palaeosyops grangeri Osborn, this monograph, page 335 

Original reference. — Am. Mus. Nat. Hist. Bull., vol. 
24, p. 604, fig. 9, 1908 (Osborn, 1908.318). 

Type locality and geologic horizon. — Twin Buttes, 
Bridger Basin, Wyo. ; Uintatherium- Manteoceras- Mesa- 
tirhinus zone (Bridger C 1). 

Holotype. — A palate and grinding teeth with por- 
tions of the lower jaw and skull (Am. Mus. 12189), 
American Museum expedition of 1904. (See fig. 119.) 

Specific characters. — Osborn writes: 

Exceeding P. robustus in certain dental proportions; p^-m^, 
165 millimeters. Fourth superior premolar enlarged (trans- 
verse, 31 mm.). Molars with extremely prominent parastyles 
and oblique ectolophs. 

Etymology. — Named "in honor of Mr. Walter Gran- 
ger, of the American Museum staff, whose explora- 
tions have transformed our knowledge of the Bridger 
animals. " (Osborn.) 

Present determination. — The species is probably 
^alid. (See p. 335.) 




Figure 119. — Tj'pe (holotype) of Palaeosyops grangeri 
; maxillary with p'-mi. Am. Mus. 12189. After Osborn, 1908. One-half natural size. 

Palaeosyops copei Osborn, 1908 

Cf. Palaeosyops copei Osborn, this monograph, page 336 

Original reference. — Am. Mus. Nat. Hist. Bull., vol. 
24, p. 606, fig. 10, 1908 (Osborn, 1908.318). 

Type locality and geologic horizon. — Lone Tree Hen- 
rys Fork, Bridger Basin, Wyo.; Uintatherium- Man- 
teoceras-Mesatirhinus zone (Bridger D 3). 

Holotype. — A series of superior grinding teeth (Am. 
Mus. 11708). (See fig. 120.) 



182 



TITANOTHERES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



Specific characters. — Osborn writes: 

Of more diminutive size (p'-m', 153 mm.), but the miost 
progressive species of Palaeosyops known in the evolution of its 
superior premolars and molars. Heavy oingula embracing the 
inner sides of the crowns. A rudimentary tetartooone on p^. 

Etymology. — Named "in honor of the late Prof. 
E. D. Cope, the describer of Lambdotherium, 'Palaeo- 



FiGTjRE 120. — Type (holotype) of Palaeosyops copei 
P'-ms, right. Am. Mus. 11708. After Osborn, 1908. One-half natural size. 

syops' horealis, and other species of Eocene titano- 

theres." (Osborn.) 

Present determination. — The species is probably 

valid. 

Manteoceras washakiensis Osborn, 1908 

Cf . Manteoceras washakiensis Osborn, this monograph, page 371 

Original reference. — Am. Mus. Nat. Hist. Bull., vol. 
24, p. 607, fig. 11, 1908 (Osborn, 1908.318). 

Type locality and geologic Tiorizon. — Base of Hay- 
stack Mountain, Washakie Basin, Wyo.; summit of 
UintatJierium- Manteoceras- Mesatirliinus zone (Washa- 
kie A). 

Holotype. — A well-preserved skull with dentition 
(Am. Mus. 13165). Discovered by Mr. Paul Miller, 
of the American Museum expedition of 1906. (See 
fig- 121.) 

Specific characters. — Osborn writes: 

Distinguished from M. manteoceras of a somewhat lower 
geological level by its more progressive characters, as follows: 
Canines short, obtuse, recurved; internal lobes of pm^, pm^ 
broadening, with shelf for development of deuterocone; p^ (ap. 
19 mm., tr. 17) with marked external convexities and a re- 
duced external cingulum; p^ (ap. 19, tr. 25) exhibits 
the tetartocone fold somewhat more conspicuously 
than in the most progressive Bridger level D speci- 
mens. ?■* (ap. 24, tr. 30) is progressive in transverse 
measurement and in the development of the tetar- 
tocone shelf. The molars are progressive in their 
large size (m^ ap. 42, tr. 48), in the strong develop- 
ment of the internal cingulum, and in the elongate 
ectolophs. 

Etymology. — washaJciensis; "so named be- 
cause it is a more recent phase, probably 
characteristic of the Washakie rather than of 
the Bridger." (Osborn.) 

Present determination. — The species and the generic 
reference are valid. (See p. 371.) 

Mesatirhinus Osborn, 1908 

Cf. Mesatirhinus Osborn, this monograph, page 387 

Original reference. — Am. Mus. Nat. Hist. Bull., vol. 
24, p. 608, 1908 (Osborn, 1908.318). 




Type species and geologic horizon. — Palaeosyops 
megarhinus Earle. Bridger Basin, Wyo., levels Bridger 
C and D; Washalcie Basin, Wyo., levels Washakie A 
and base of Washakie B. 

Specific characters. — Osborn writes: 

Titanotheres of small size (skull length 354^425 mm.) 
typically mesaticephalic, persistent or progressing to dolicho- 
cephalic. The horns when present incipient or rudimentary, 
chiefly borne on the nasals. An infraorbital shelf. Cranium 
with a sagittal crest. Humerus relatively abbreviated — that 
is, with refeYence to Palaeosyops — carpus and tarsus narrow, 
astragalus with elongate neck, the sustentacular distal and 
ouboidal facets continuous and forming a reversed L (j) ; meta- 
podials slender. 

Etymology. — fiiaaros, middle; pis, nose; because the 
length of the snout is moderate compared with that 
in the allied genus Dolichorhinus. 

Present determination. — The genus is valid. (See 
p. 387.) 

Mesatirhinus petersoni Osborn, 1908 

Cf. Mesatirhinus petersoni Osborn, this monograph, page 389 

Original reference. — Am. Mus. Nat. Hist. Bull., vol. 
24, p. 608, fig. 12, 1908 (Osborn, 1908.318). 

Holotype. — A skull with dentition (Am. Mus. 12184) 
from Cattail Spring, Bridger Basin, Wyo., levels 
Bridger D 3 and C 3. (See fig. 122.) The species is 
also recorded from Washalde Basin, Wyo., level 
Washakie A. 

Specific characters. — Osborn writes: 

Pm'-m^, 156 millimeters; m'-m', 90. Skull length, pre- 
maxillaries to condyles 412 (estimated) ; preorbital facial region 
more elongate (217). Other characters as in Mesatirhinus 
megarhinus — that is, broad occipital condyles, broad infra- 
orbital shelf on malar, etc. 

Comparison of this animal with the type of M. megarhinus 
can leave no doubt that we have to do here with a much more 
advanced stage of evolution. The skull is longer, the pre- 
orbital region especially. The grinding teeth occupy more 
space, and there is an average advance in all the rectigradations 
which proves that these differences in form and size are not 
merely due to fluctuations of size or differences of sex. 




ffK?- 



Figure 121. — Type (holotype) skull of Manteoceras washakiensis 
Left side. Am. Mus. 13165. After Osborn, 1908. One-flfth natural size. 

Etymology. — "The species is named in honor of Mr. 
O. A. Peterson, now of the Carnegie Museum, whose 
titanothere collections in the Uinta formation greatly 
extended our knowledge." (Osborn.) 

Present determination. — The species and generic ref- 
erence are valid. (See p. 389.) 



DISCOVERY OF THE TITANOTHERES AND ORIGINAL DESCRIPTIONS 



183 



Metarhinus Osborn, 1908 

Cf. Metarhinus Osborn, this monograph, page 420 

Original reference. — Am. Mus. Nat. Hist. Bull., vol. 
24, p. 609, 1908 (Osborn, 1908.318). 

Type species and geologic liorizon. — Metarhinus flu- 
viatilis Osborn. Washakie Basin, Wyo., level Washa- 
kie B; Uinta Basin, Utah, levels Uinta B 1 and B 2. 

Specific charac- 
t er s . — Osborn 
writes : 

Small tltanotheres 
(skull length 355 to 440 
mm.), persistently mes- 
aticephalic. Narrow, 
abbreviated preorbital 
region, premaxillary 
symphysis greatly 
elongated, and anterior 
narial openings deeply 
recessed in side view. 
Infraorbital shelf pres- 
ent, or wanting (M. 
diploconus) ; occipital 
condyles narrow. 
Grinding teeth sub- 
hypsodont; premolars 
progressive; hypoco- 
nulid of ms small, 
conic. 

Etymology. — iiera, 
after; rJiinus (that 
i s , MesatirMnus) . 
"The name alludes 
to the somewhat 
later geological ap- 
pearance of this 
genus as compared with MesatirMnus." (Osborn.) 
Present determination. — The genus is valid. (See 

p. 420.) 

Metarhinus fluviatilis Osborn, 1908 

Cf. Metarhinus fluviatilis Osborn, this monograph, page 421 
Original reference. — Am. Mus. Nat. Hist. Bull., vol. 
24, p. 609, fig. 13, 1908 (Osborn, 1908.318). 




Figure 122. — Type (holotype) skull 
of MesatirMnus petersoni 

Top view. Am. Mus. 12184. After Osborn, 1908. 
One-fourth natural size. 




Figure 123. — Type (holotype) skull of Metarhinus fluviatilis 



Lett side. Nasals broken off. 



Am. Mus. 1500. 
natural size. 



After Osborn, 1908. One-fourth 



Type locality and geologic liorizon. — Uinta Basin, 
Utah; Metarhinus zone (Uinta B 1). 

Holotype. — A skull (Am. Mus. 1500) discovered by 
the American Museum expedition of 1894 in horizon 
B 1 of the Uinta Basin. (See fig. 123.) 



Specific characters. — Osborn writes: 

Pm'-m^ — 144 millimeters. A relatively short (355 mm., 
estimated), broad (200 mm., estimated) skull. Eye sockets 
small and very prominent. Premaxillary symphysis elongate, 
grinding teeth subhypsodont, m' with a cingulum-hypocone in 
the type. 

Etymology. — fluviatilis, fluviatile. "The name is 
given in allusion to the possibly river-living or am- 
phibious habits of the animal." (Osborn.) 

Present determination. — The species and the generic 
reference are valid. For fuller specific distinctions 
see page 421. 

Metarhinus earlei Osborn, 1908 

Cf. Metarhinus earlei Osborn, this monograph, page 426 

Original reference. — Am. Mus. Nat. Hist. Bull., vol. 
24, p. 610, fig. 14, 1908 (Osborn, 1908.318.) 




Figure 124. — Type (holotype) .skull of 
Metarhinus earlei 

Top view. Am. Mus. 13166. After Osborn, 1908. 
One-fourth natural size. 

Type locality and geologic horizon. — North side of 
Haystack Mountain, Washakie Basin, Wyo.; Meta- 
rhinus zone (Washakie B 1 ) . 

Type. — A skull (Am. Mus. 13166) lacking the nasals, 
American Museum expedition of 1906. (See fig. 124.) 

Specific characters. — Osborn writes: 

Pmi-ni' = 167 millimeters. Skull proportions, length 380, 
breadth 230. Narrow occipital condyles. Extremely elongate 
premaxillar}' symphysis. A short sagittal crest. No hypocone 
on m^. 

This animal is readily distinguished from M. diploconus by 
(1) the infraorbital shelf of the malars; (2) the elongate premaxil- 



184 



TITANOTHERES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 




lary; (3) the absence of a double cone on m^ In many other 

respects it resembles M. diploconus, especially in its proportions. 

It is distinguished from M. megarhimis by (1) the elongate 

premaxillary symphy- 
sis, correlated with the 
long, narrow facial re- 
gion; (2) the narrowness 
of its occipital condyles. 
It is distinguished from 
M. fluviatilis by (1) its 
greatly superior size 
and (2) the lesser prom- 
inence of the orbits. 

Etymolo g y. — 
Named "in honor of 
Charles Earle, the 
first monographer of 
the genus Palaeo- 
syops and its allies." 
(Osborn.) 

Present determina- 
tion. — The species 
is probably valid. 
(See p. 426.) 

Dolichorhinus interme- 
dius Osborn, 1908 

Cf. Dolichorhinus inter- 
niedius Osborn, this 
monograph, page 405 

Original refer- 
ence. — Am. Mus. 
Nat. Hist. Bull., vol. 
24, p. 611, fig. 15, 
1908 (Osborn, 
1908.318). 

Type locality and 

Top view. Am. Mus. 1837. After Osborn, 1908. neoloqic llOrizOn. — 
One-fourth natural size. .^t. , -r» • i.^ 

Umta Basin, north- 
eastern Utah; Eobasileus - Dolicliorhinus zone 
(Uinta B 2). 

Hohtype. — A skull with dentition (Am. Mus. 1837), 
discovered by the American Museum expedition of 
1894. (See fig. 125.) 

Specific characters. — Osborn writes : 

Distinguished from D. hyognathus Scott and Osborn by (1) 
its inferior size (pm'-m^ 179, m'-m' 109 mm.); 
(2) premolars less progressive, with subconic 
deuterocones; (3) all oingula less robust; (4) 
nasals more pointed and less expanded distally; 
(5) infraorbital shelf of malar relatively narrow. 

Etymology. — "The name 'intermedins' 

is given because in some characters this 

species is intermediate between Mesati- 

rhinus petersoni and DolichorMnus Tiyo- 

gnathus, although on the whole it is Figure 126. — Type (holotype) skull of Telmatheriiim ultimum 

much more nearly allied to the latter." side ™w. Am. Mus. 2O6O. After Osbom, 19O8. One-flfth natural size. The skull has been some- 
//-j V s what deformed by pressure. 

Present determination. — The generic reference ap- I pointed. P,, pz laterally compressed, nonmolariform; ps, p4 
pears certain; the species is probably valid. (See submolariform; dolichocephalic, anterior portion of face 
p. 405.) I elongate. 



Figure 125. — Type (holotype) skull of 
Dolichorhinus inter medius 



Telmatherium ultimum Osborn, 1908 

Cf . Telmatherium ultimum Osborn, this monograph, page 345 

Original reference. — Am. Mus. Nat. Hist. Bull., 
vol. 24, p. 613, fig. 17, 1908 (Osborn, 1908.318). 

Type locality and geologic horizon. — Uinta Basin, 
northeastern Utah; Diplacodon-Protitanotherium-Epi- 
hippus zone (Uinta C, lower levels). 

Holotype. — A well-preserved skull with dentition 
(Am. Mus. 2060). Discovered by Mr. Peterson, of 
the American Museum expedition of 1895. (See fig. 
126.) 

Synonymy. — This species was mentioned by Mat- 
thew as Palaeosyops ultimus Osborn MS. (see p. 177), 
but as no type was indicated or specific diagnosis 
given the name remained a nomen nudum until a type 
was designated and a diagnosis given by Osborn in 
1908. 

Specific characters. — Osborn writes: 

P'-m*, 226 mm. Lateral superior incisors greatly en- 
larged, caniniform. Pm^- ^. * with Internal subcrescentic 
deuterocone ridges, with faint rudiments of tetartocones 
posteriorly. Ectolophs of premolars elevated and biconvex. 

Etymology. — ultimus, latest. "The specific name is 
given because this appears to be the last representative 
of the Palaeosyops-Limnohyops-Telmatherium group." 
(Osborn.) 

Present determination. — This species is certainly a 
valid one. The grounds for regarding it as allied to 
the genus Telmatherium are given on page 345. 
Telmatherium? altidens Osborn, 1908 
Cf . Telmatherium altidens Osborn, this monograph, page 351 

Original reference. — Am. Mus. Nat. Hist. Bull., 
vol. 24, p. 614, fig. 18, 1908 (Osborn, 1908.318). 

Type locality and geologic horizon. — Uinta Basin, 
northeastern Utah; Diplacodon-Protitanotherium-Epi- 
hippus zone (Uinta C). 

Holotype. — A lower jaw with dentition (Am. Mus. 
2025) discovered by the American Museum expedition 
of 1895. (See fig. 127.) 

Specific characters. — Osborn writes: 

Pmj-mj, 330 milUmeters; a wide diastema (70 mm.) behind 
the canines. Canines in male exceptionally elevated (76) and 




DISCOVERY OF THE TITANOTHERES AND ORIGINAL DESCRIPTIONS 



185 



The specific characters are more fully given on 
page 351 of this monograph. 

Etymology. — "The specific name refers to the high- 
crowned piercing canine." 




Figure 127. — Type (holotype) of Telmaiherium? altidens 
Lower jaw. Am. Mus. 2025. After Osborn, 1908. One-sixth natural size. 

Present determination. — The species is probably 
valid. The generic reference is somewhat less certain. 
(Seep. 351.) 

Protltanotherium superbum Osborn, 1908 

Cf. Protitanotherium superbum Osborn, this monograph, page 379 

Original reference. — Am. Mus. Nat. Hist. Bull., 
vol. 24, p. 615, fig. 19, 1908 (Osborn, 1908.318). 

Type locality and geologic Jiorizon. — Uinta Basin, 
northeastern Utah; Diplacodon-Pi'otitanotherium-E'pi- 
Jiippus zone (Uinta C) ; probably higher levels. 

Holotype. — A well-preserved lower jaw with denti- 
tion (Am. Mus. 2501). (See fig. 128.) 

Specific characters. — Osborn writes: 

Pi-m3, 318 millimeters. Canines in males very robust; pi 
double fanged; postoanine diastema abbreviated; premolar 
series relatively abbreviated; p2 with very large talonid and 
crescentic protoconid; ps, p4 with talonid heavy and promi- 
nent — that is, submolariform — but no entoconid. Ma with 
hypooonulid sharply constricted off at base. 

Etymology. — superbum, haughty, arrogant. "The 
name is given in reference to the great size and pre- 
sumed power of this Uinta titanothere, which con- 
siderably exceeds that of the smaller [lower] 
Oligocene titanotheres." (Osborn.) 

Present determination. — The species is 
probably valid. The generic reference is 
somewhat less certain. (See p. 379.) 

Telmatherium? incisivum Douglass, 1909 

Cf. Sthenodecies incisivus (Douglass), this mono- 
graph, page 354 

Original reference. — Carnegie Mus. Annals, 
vol. 6, No. 2, pp. 305-307, text figs. 1, 2, 3, pi. 
13, fig. 1, 1909; "issued November 6, 1909" 
(Douglass, 1909.1). 

Type locality and geologic horizon. — Uinta 
Basin, Utah, about 3 miles northeast of well 
2, from "a thick deposit of sandstone and 
small gravel evidently of stream origin, near the mid- 
dle of horizon B." Near the summit of Eohasileus- 
Dolichorhinus zone (Uinta B 2). Discovered by Mr. 
J. F. Goetschius. 

101959— 29— VOL 1 15 



Type. — A skull, lacking the ends of the nasals 
(Carnegie Mus. 2398). (See figs. 129, 130.) 

Specific characters. — Douglass writes: 

I think that this skull represents a different genus 
from Telmatherium, but I prefer to place it provi- 
sionally here rather than establish another genus. The 
skull is broad and short, but not high. The forehead 
is broad and flat. ' The premaxillaries are oblique, 
not transverse. The face is short and concave. 
Apparently there are vacuities anterior to the orbits. 
Beneath these there is a rounded angle on the malar, 
but there is no flattened shelf beneath the orbit. The 
zygomatic arch is spreading and moderately heavy. 
The sagittal crest is quite high and thin. The 
superior wings of the occiput are also thin. The brain 
case^is small; the outward-projecting zygomatic proc- 
esses of the squamosals shelf -like and broad anteropos- 
teriorly. The paroccipital processes extend laterally 
and are continuous with the paramastoid processes 
posterior to the external auditory meatus and the postglenoid 
process. The anterior portion of the opening of the posterior 
nares is between the anterior portions of the last molars. 
The teeth increase quite regularly in size from p2 to m'. The 
premolars have heavy cingula. The deuterocones on p" and 
p^ are oblong anteroposteriorly, while that on p* is high and 
conical. 

Measurements [Douglass] 

Millimeters 

Length of skull, basal 490 

Width of skull 330 

Length of dental series 295 

Length of molar-premolar series 212 

Transverse diameter of i' 21 

Anteroposterior diameter of i' 22 

Transverse diameter of i^ 27 

Anteroposterior diameter of i^ 25 

Transverse diameter of i' 22 

Anteroposterior diameter ofi^ 25 

Transverse diameter of canine 24 

Anteroposterior diameter of canine 27 

Transverse diameter of p2 22 

Anteroposterior diameter of p^ 20 

Transverse diameter of p^ 30 

Anteroposterior diameter of p^ 24 

Transverse diameter ofp* 37 

'T\ Anteroposterior diameter of p"" 27 




Figure 128. — Type (holotype) of Protitanotherium superbum 
Lower jaw. Am. Mus. 2501. After Osborn, 1908. One-sixth natural size. 

Anteroposterior diameter ofm' 44 

Transverse diameter of m^ 53 

Anteroposterior diameter of m' 46 

Transverse diameter of m^ 53 

Anteroposterior diameter of m' 46 



186 



TITANOTHERES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



Etymology. — incisivum, provided with incisors; in 
allusion to the great size of the superior incisors. 

Present determination. — The species probably repre- 
sents a genus distinct from Telmatherium (see p. 353) 
named Sthenodectes by Gregory. 




Type. — A skull lacking the posterior portion 
(Carnegie Mus. 2888). (See figs. 131 and 132.) 
Specific characters. — Douglass writes: 

The skull is high, the forehead broad, and the zygomatic 
arches spreading. The premaxillary region as seen from the 
front is broad, though the incisors are only moder- 
ately large. The canines are directed outward. The 
free nasals are short and moderately broad. Appar- 
ently the infraorbital foramen is not excessively large. 
The malar is rounded beneath the orbit and has no 
protuberance or shelf. The zygomatic arch is not 
very heavy and is only moderately deep anterior to 
the glenoid articular surface. It is not nearly so 
heavy as in Telmatherium uliimum. The opening 
of the posterior nares extends forward to the middle 
of the second molars. Their border is rounded and 
thickened. 

The incisors are moderately large but not cupped. 
They are arranged in an oblique line about halfway 
between a transverse and anteroposterior direction. 
The crowns of i' and i^ are low. The anterior faces 
are very convex. There are two posterior flattened 
surfaces separated by a rounded ridge. There are 
no cups, but the posterior portion forms a kind of 
ledge or keel. P is higher and is directed more down- 
ward. The posterior portion is flattened, and there is 
a low flat ledge behind the conical cusp. The canine 
has a moderately high curved crown, on which there 
are antero-internal and postero-external ridges, pass- 
ing downward from the base to the apex. There is 
also a narrow postero-internal ledge. 

Unless the skull is more crushed laterally than it 
appears to be, there is a sudden contraction posterior 
to the canine, so that the first two premolars are 
much nearer to the median line of the palate than are 
the canines. The diastema between the canine and 
p' is about 3 centimeters in length. 

P' is a simple oblong conical tooth, which has a 
small antero-internal depression, and a small ridge 
passes backward from the apex to the posterior por- 
tion of the rudimentary keel. P 2, 3, and 4 have low 
cusps. The teeth increase nearly uniformly in width 
and size from p^ to the last molar. The two outer 
elements in each are well defined and are subequal 
in size, although the anterior cusp is slightly the 
larger. The internal cusp on p^ is small, oblong 
anteroposteriorly, and is placed far back. The inter- 
nal cusp on p' is much larger and is crescent-shaped. 
On p* it is more nearly conical. There are rudi- 
mentary cingula on the inner faces of the last three 
premolars. The postero-internal cusp on m' is repre- 
sented by a low crescent-shaped ridge. 



FiGUBE 129. — Type (holutype) skull of Telmatherium? incisivum 

palatal view; A3, 



Carnegie Mus. 2398. After Douglass, 1906. Ai, Superior view; 
view. One-fifth natural size. 

Manteoceras uintensis Douglass, 1909 

Cf. Manteoceras uintensis Douglass, this monograph, page 372 
Type reference. — Carnegie Mus. Annals, vol. 6, No. 

2, pp. 307-310, text figs. 4, 5, pi. 13, fig. 4, 1909; 

"issued November 6, 1909" (Douglass, 1909.1). 
Type locality and geologic horizon. — Uinta Basin, 

Utah, about 5 miles northeast of well 2, from "gray 

sandstone in red Uinta beds. Lower portion of 

horizon C." Diplacodon-Protitanotherium-Epihippus 

zone (Uinta C) . 



Measurements [Douglass] 

Millimeters 

Is, lateral Lgjjgth of skull, anterior portion to glenoid 430 

Length of dental series 356 

Length of molar-premolar series ^-_-- 247 

Length of premolar series 106 

Length of molar series 141 

Transverse diameter of i' 16 

Anteroposterior diameter ofii 18 

Transverse diameter ofi^ 16 

Anteroposterior diameter of i^ 18 

Transverse diameter of i' 20 

Anteroposterior diameter ofi^ 22 

Transverse diameter of canine 22 

Anteroposterior diameter of canine 26 



DISCOVERY OF THE TITANOTHERES AND ORIGINAL DESCRIPTIONS 



187 




Figure 130. — Type (holotype) of Telmatherium? incisivum 
Lett upper teeth, crown view. After Douglass, 1909. One-half natural size. 



Millimeters 

Transverse diameter of p' 12 

Anteroposterior diameter of p' 22 

Transverse diameter of p^ 21 

Anteroposterior diameter ofp^ 28 

Transverse diameter of p^ 28 

Anteroposterior diameter ofp' 27 

Transverse diameter of p* 33 

Anteroposterior diameter ofp' 30 

Transverse diameter of m' 44 

Anteroposterior diameter of m' 40 

Transverse diameter ofm.^ 63 

Anteroposterior diameter ofm^ 55 

Transverse diameter ofm^ 56 

Anteroposterior diameter ofm^ 51 

Width of palate between canines? 68 

Width of palate between first premolars 54 

Width of palate between last molars 83 

Etymology. — uintensis, in reference to the 
Uinta Basin. 

Present determination. — The generic refer- 
ence to Manteqceras appears to be correct. 
The species is a valid one. 

Dolichorhinus heferodon Douglass, 1909 

Cf. Dolichorhinus heterodon Douglass, this monograph, 
page 416 

Original reference. — Carnegie Mus. Annals, 
vol. 6, No. 2, pp. 310-311, text figs. 6, 7, 
pi. 13, fig. 3, 1909; "issued November 6, 1909" 
(Douglass, 1909.1). 

Type locality and geologic horizon. — Uinta Basin, 
Utah, 6 or 7 miles northeast of well 2; from "upper 
part of horizon B or lower part of horizon C"; 
Eohasileus-DolicJiorJiinus zone (Uinta B 2). 

Type. — A skull lacking the front teeth and both 
zygomatic arches (Carnegie Mus. 2340). (See figs. 133 
and 134.) Discovered by Mr. J. F. Goetschius. 



The infraorbital foramen is large. The infraorbital shelf is 
represented by a protuberance, which is thickened on the free 





Figure 132. — Type (holotype) of Manteoceras uintensis 
Upper teeth. Carnegie Mus. 2388. After Douglass, 1909. One-third natural size. 

Specific characters. — Douglas writes: 

The skull is long, narrow, and moderately high. The face is 
short and the brain case long. The free nasals are long, the 
posterior opening of the anterior nares extending well backward 
toward the orbit. The lower borders of the nasals approach 
each other, but this is probably in part due to lateral crushing. 



Figure 131. — Type (holotype) skull of Manteoceras uintensis 

Carnegie Mus. 2388. After Douglass, 1909. Ai, Palatal view; Aj, view of right side. 
One-fifth natural size. 

outer surface. If there were horn cores 
above the orbit they ■v\'ere very small. The 
long brain case was apparently arched 
from before backward, the posterior de- 
scent to the crest of the occiput being very 
steep, though this may be somewhat ex- 
aggerated by crushing. The occipital con- 
dyles are very large. The median portion 
of the occiput above them is convex, while 
above this there is a large concavity. The 
postglenoid processes are not excessively large. 

The premolars are small, the last being very decidedly 
smaller than the first molar. The first premolar is not pre- 
served, but it was evidently a simple tooth. In the last three 
premolars there is a lobe or buttress on the antero-external 
portion of the tooth, which makes the anterior margin oblique. 



188 



TITANOTHERES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



The inner cusps (deuterocones) are low with rounded summits. 
They are more nearly opposite the postero-external than the 
antero-external cusi5. There are inner cingula on p' and p*. 
The antero-internal cusp in m^ is quite high and m' conical. 
The postero-internal cusp is due simply to an increase in height 
of the cingulum. 

Measurements 

Millimeters 

Total length of top of skull 500 

From anterior orbit to front of nasals 160 

Width of occiput 128 

Height of occiput 140 

Length of molar-premolar series 190 

Length of premolar series 75 

Length of molar series 115 

Length of p2 20 

Width of p2 16 



Dolichorhinus longiceps Douglass, 1909 

Cf. Dolichorhinus longiceps Douglass, this monograph, page 406 
Original reference. — Carnegie Mus. Annals, vol. 6, 

No. 2, pp. 312-313, text fig. 8; pi. 13, fig. 2; pis. 14, 

15, 1909; "issued November 6, 1909" (Douglass, 

1909.1). 

Type locality and geologic horizon. — Uinta Basin, 

Utah, "about 1^ miles east of well No. 2," from 





Figure 133. — Type (holotype) skull of Dolichorhinus heterodon 

Carnegie Mus. 2340. After Douglass, 1909. Ai, Palatal view; As, right lateral 
view. One-fifth natural size. 



Millimeters 

Length of p3 21 

Width of p3 20 

Length of p< 24 

Width of p^ 27 

Length ofm' 34 

Width of ml 35 

Length ofm^ 46 

Width of m2 42 

Length ofm^ 48 

Width of m3 42 



Figure 134. — Type (holotype) of Dolichorhinus heterodon 

Upper premolar series. Carnegie Mus. 2340. After Douglass, 1909. Slightly less 
than one-half natural size. 

"the lowest level at which fossils were 
found in horizon 'B' of the Uinta, about 
700 feet below the bottom of the Uinta 
red beds (horizon 'C')." Eohasileus- 
DolicJiorJiinus zone (Uinta B 2). 

Type. — A skull lacking the incisors, part 
of the dentition, and the basioccipital 
region (Carnegie Mus. 2347). (See figs. 
135 and 136.) 

Specific characters. — Douglass writes: 

Phis skull in general outline is very much like 
that of Dolichorhinus hyognathus, though broader. 
In describing it I prefer to point out the char- 
acters which distinguish it from that species. 
Apparently it is somewhat broader proportionally 
than that of D. hyognathus. The skull is some- 
what crushed, but it evidently was not flattened 
on top. The present specimen had no heavy 
protuberances or horn cores, though there may 




Figure 135. — Type (holotype) skull of Dolichorhinus longiceps 
Top view. Carnegie Mus. 2347. After Douglass, 1909. One-sixth natural size. 



Etymology. — crepos, difl^erent, or various; bbobs, tooth. 
Allusion not clear; name possibly given because no 
two teeth in the superior premolar-molar series are 
alike. 

Present determination. — The form is closely allied 
to D. intermedins, of which it may be the successor. 
Its specific separateness is somewhat doubtful. 



have been the slightest beginning of such. There is a 
rather narrow shelf, or lateral expansion of the malars, with 
rounded outer borders, beneath the anterior portion of the 
orbit, but it is not like the infraorbital process of D. hyognathus. 
The postorbital hook does not appear to have been long or 
prominent. Evidently the zygomatic arches extend laterally 
outward more than in the last-named species; the postglenoid 
processes are not nearly so heavy; the palate is broader; the 
top of the cranium, though there is no zygomatic arch, becomes 
narrower anterior to the crest of the occiput. 



DISCO'V'EEY OF THE TITANOTHERES AND ORIGINAL DESCRIPTIONS 



189 



The teeth are very similar to those of Dolichorhinus heterodon, 
so much so that, if only the teeth were known, they might be 
referred to that species. They, as well as the skul), are larger. 

Measurements [Douglass] 

Millimeters 

Length of top of skull 590 

Length of free nasals 150 

Length of skull posterior to anterior portion of orbit 393 

Width of skull at glenoid articular surface 267 

Width at infraorbital shelves 247 



Millimeters 

Length of p' 24 

Width of p3 ■. 25 

Length of p^ 27 

Width of p* 31 

Length ofm' 30 

Width of m', about 37 

Length of m^ 37 

Width of m2 44 

Length of m^, about • 41 

Width of m3, about 43 




A2 



FiGUKE 136. — Type (holotype) of Dolichorhinus longiceps 

Carnegie Mus, 2347. After Douglass, 1909. Ai, Palatal view of skull, somewhat less than one-third natural size; Aj, left lateral view of skull, somewhat less than 

one-third natural size; A3, crown view of right upper premolar series, one-half natural size. 



Length of molar-premolar series 192 

Length of premolar series 88 

Length-of molar series 112 

Length of p' 15 

Width of pi 11 

Length of p^ 20 

Width ofp2 20 



Etymology. — longiceps, in allusion to the long skull. 

Present determination. — For the reasons stated above 
it appears that this form is connected with the typi- 
cal D. Jiyognathus by a skull of intermediate char- 
acters. Its status as a distinct species is therefore 
somewhat doubtful. 



190 



TITANOTHERES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



Sthenodectes Gregory, 1912 

Cf. Sthenodectes, this monograph, page 353 
Original reference. — Science, new ser., vol. 35, No. 

901, p. 545, April, 1912 (Gregory, 1912.1). 

Subsequent reference. — Riggs, New or little known 

titanotheres from the lower Uinta formations: Field 

Mus. Nat. Hist. Pub. 159, Geol. ser., vol. 4, No. 2, 

p. ^8, June, 1912 (Riggs, 1912.1). 




Figure 137. — Type (holotype) skull of Mesatirhinus superior 
Field Mils. 12188. After Riggs, 1912. Side, top, and palatal views. Less than one-fourth natural size 

Type species. — Telmatherium? incisivum Douglass. 
Generic characters. — Gregory writes: 

This, genus is distinguished from Telmatherium ultimum Os- 
born by the following assemblage of characters: (1) The in- 
cisors are far larger and more advanced in evolution, i' being 
closely appressed to its fellowr in the median line, with anterior 
face elongate, antero-internal tip blunt, median basin large, 
posterior wall or cingulum very massive, i^ i^ extremely large 
with low recurved tips and very heavy posterior cingula. (2) 
The postcanine diastema is reduced or absent. (3) Superior 
premolars 2, 3, 4 are much more advanced than in T. ultimum, 



having very heavy internal cingula, pronounced external cin- 
gula, high slender internal cusps (deuterocones) ; p^ especially 
is in a relatively advanced stage as compared with T. ultimum. 

(4) The least tranverse diameters of p* and of the anterior lobe 
of m' are greater, that of m^ much less, than in T. ultimum. 

(5) The basicranial region differs in many details, such as the 
apparent junction of the postglenoid and post-tympanic proc- 
esses below the auditory meatus. (6) The occiput is low, 
with a sharp, long sagittal crest. (7) The forehead is 

relatively wide. (8) The nasals taper dis- 
tally. 

From Manteoceras (especially M. uinten- 
sis) the genus under consideration is dis- 
tinguished by (1) the form and size of the 
incisors and canines, (2) the much more 
advanced stage of evolution of the premo- 
lars, (3) the shorter anteroposterior diam- 
eter of m^, (4) the reduction of the post- 
canine diastema, (6) the arched and 
spreading zygomata, etc. 

From Dolichorhinus and Mesatirhinus it 
is separated by the shortness and relative 
breadth of the skull, the great size of the 
incisors, the relatively heavy zygomata, 
and many other details. 

Etymology. — adho^, strength, driKT-qs, 
a biter; in allusion to the great 
power and development of the in- 
cisors and canines. 

Present determination. — A valid 
genus, offshoot of the typical Telma- 
therium phylum. 

Mesatirhinus superior Riggs, 1912 

Cf. Dolichorhinus superior (Riggs), this 
monograph, page 405 

Original reference. — Field Mus. 
Nat. Hist. Pub. 159, Geol. ser., vol. 
4, No. 2, p. 26, pi. 6, June, 1912 
(Riggs, 1912.1). 

Type locality and geologic hori- 
zon. — White River divide, north- 
eastern Utah; upper " Metarhinus 
sandstones," summit of Metarhinus 
zone (Uinta B 1). (See fig. 137.) 

Holotype.— A skull (Field Mus. 
12188). 

Specific characters. — Riggs writes: 

Skull 485 by 255 millimeters, molar series 182 millimeters, 
nasals free to a point over last premolar, infra-orbital process 
present, arches slender anteriorly, nasals infolded at margins, 
sagittal area expanded, canines small, p^ and p' oblique to axis 
of series. Molars relatively small, strong hypocone on m^, pos- 
terior nares opening opposite the anterior margin of last molar. 

Etymology. — superior, in allusion to its large size 
and high stage of evolution. 

Present determination. — This is a valid stage im- 
mediately ancestral to the Dolichorhinus stage. 



DISCOVERY OF THE TITANOTHEEES AND ORIGINAL DESCRIPTIONS 



191 



Metarhinus riparius Riggs, 1912 

Cf. Metarhinus riparius, this monograph, page 429 

Original reference. — -Field Miis. Nat. Hist. Pub. 159, 
Geol. ser., vol. 4, No. 2, p. 28, pi. 7, fig. 1, June, 1912 
(Eiggs, 1912.1). 

Type locality and geologic horizon. — White River 
canyon and divide, northeastern Utah; "entire upper 
Metarhinus beds," base of Metarhinus zone (Uinta B 1). 




Figure 138. — Type (holotype) skull of Metarhinus riparius 
Field Mus. 12186. After Eiggs, 1912. About one-fourth natural size. 

Holotype.— Skull (Field Mus. 12186). (See fig. 138.) 
Paratype {"cotype"). — "Lower jaws" (Riggs, pi. 7, 
figs. 2, 3). 
Specific characters. — Riggs writes: 

Skull long and narrow (405 by 210 mm.). 
Anterior cranial region expanded, sagittal crest 
short. Interorbital region relatively narrow and 
rounded, rudimentary horn cores above orbits, 
canines large, molar series short (88-93 mm.), 
hypocone usually present on m', mandible 
straight in the ramus, lower canine long and 
recurved. 



Etymology. — cristatus, crested; in allusion to the 
high sagittal crest. 

Present determination. — A valid stage in the Meta- 
rhinus fluviatilis phylum. 

Dolichorhinus fluminalis Riggs, 1912 

Cf. Dolichorhinus fluminalis, this monograph, page 417 

Original reference. — Field Mus. Nat. Hist. Pub. 159, 
Geol. ser., vol. 4, No. 2, p. 33, pi. 10, 
figs. 1-3, June, 1912 (Riggs, 1912.1). 

Type locality and geologic horizon. — 
Uinta Basin, northeastern Utah; "Amy- 
nodon sandstone," summit of Eohasileus- 
Dolichorhinus zone (Uinta B 2). 

Holotype. — A fine skull. Field Mus. 
12205; collector M. G. Mehl. (See fig. 
140.) 

Specific characters. — Riggs writes: 

Skull small and narrow (520 by 230 mm.), 
facial region much shorter than cranial, nasals 
narrow and slightly tapering, posterior nares 
opening between hamular processes, postorbital 
process of jugal back of the last molar, molar- 
premolar series 171 millimeters; canines short and recurved, in- 
cipient horn cores in the form of high, narrow ridges. * * * 
The skull is slender, light and complex in structure as com- 
pared with the massive and rounded D. cornutus. The molar 
teeth are no longer in the crown than those of Metarhinus 



Etymology. — riparius, riparian, in allu- 
sion to the nature of the habitat. 

Present determination. — A valid species 
in the Metarhinus phylum. 

Metarhinus cristatus Riggs, 1912 

Cf . Metarhinus cristatus, this monograph, page 429 

Original reference. — Field Mus. Nat. 
Hist. Pub. 159, Geol. ser., vol. 4, No. 2, 
p. 28, pi. 9, fig. 3, June, 1912 (Riggs, 
1912.1). 

Type locality and geologic horizon. — 
White River canyon, northeastern Utah; 
"upper Metarhinus beds," lower section 
of Metarhinus zone (Uinta B 1). 

Holotype. — A skull, lacking the muzzle (Field Mus. 
12194). (See fig. 139.) 

Specific characters. — Riggs writes: 

Skull length approximately 380 millimeters, molar series 94 
millimeters. Frontal region broad, sagittal crest long and 
high, molars short-crowned, no hypocone on m', arches rela- 
tively heavy. Represented by a single skull lacking the 
nasals and the premaxillaries. 




Figure 139. — -Type (holotype) skull of Metarhinus cristatus 
Field Mus. 12194. After Eiggs, 1912. One-third natural size. 



earlei. The jugal process of the maxillaries arises at a point 
back of the last molar rather than beside it as in Z). longiceps. 
There is no offset in the palate between the last molars, though 
the primary position of the posterior narial opening is marked 
by a slight rugosity. 

D. fluminalis is most nearly related to D. intermedins. The 
skull exceeds in length the type of that species in the ratio of 
520:465 millimeters. The molar teeth are proportionately 
much smaller; the series measures relatively 99:109 millimeters. 



192 



TITANOTHERES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



The position of the posterior narial opening is the most distinc- 
tive character, appearing much farther back in D. fluminalis 
than in any other described species. The two forms agree more 
closely in the tapering form of the nasals and in the narrow 
recess separating them, from the maxillaries. 

Etymology. — fluminalis, pertaining to rivers; in 
allusion to the habitat. 

Present determination. — A stage in the DolichorTiinus 
phylum, not very clearly distinguished specifically 
from other progressive stacc?. 




,to» 




Figure 140. — Type (holotype) skull of Dolichorhinus fluminalis 
Field Mus. 12205. After Eiggs, ] 912. Side, top, and palatal views. About one-fifth natural size 

Rhadlnorhinus Riggs, 1912 

Cf. Rhadinorhinus, this monograph, page 430 

Original reference. — Field Mus. Nat. Hist. Pub. 159, 
Geol. ser., vol. 4, No. 2, p. 36, June, 1912 (Riggs, 
1912.1). 

Type species. — Rhadinorhinus ahbotti Riggs. 

Generic characters. — Riggs writes : 

Titanotheres with slender skulls, nasals deeply recessed later- 
ally and tapering, molars long-crowned, p-- ^- * subrectangular, 
a wide median area between the incisors, no infra-orbital 
process. The name Rhadinorhinus alludes to the tapering 
nasals which characterize this genus. 



Etymology. — pa8iv6s, slender; pis, nose. 
Present determination. — Probably a valid stage, an 
extreme offshoot of the Metarhinus phylum. (See 
p. 17, fig. 15.) 

Rhadinorinus abbotti Riggs, 1912 
Cf. Rhadinorhinus ahbotti, tliis monograph, page 430 

Original reference. — Field Mus. Nat. Hist. Pub. 159, 
Geol. ser., vol. 4, No. 2, p. 36, pi. 11, figs. 2, 3, June, 
1912 (Riggs, 1912.1). 

Type locality and geologic horizon. — 
Northeastern Utah; "upper Meta- 
rhinus beds," center of Metarhinus 
zone (Uinta B 1). 

Holotype. — A fine skull (Field Mus. 
12179). (See fig. 141.) 

Specific characters. — Riggs says: 

Length of skull 435 millimeters, molar- 
premolar series 168 millimeters, nasals shorter 
than premaxillaries, thickened at suture, and 
tapering toward a terminal rugosity. Arches 
slender, posterior nares open opposite middle 
of m2. Sagittal crest long and narrow. 
Hypocone of m' vestigial, diastema short. 

Etymology. — Named in honor of 
Mr. J. B. Abbott, of the Field 
Museum of Natural History. 

Present determination. — A valid spe- 
cific stage. 

Eotitanops gregoryi Osborn, 1913 

Cf. Eotitanops gregoryi, this monograph, 
page 291 

Original reference. — Am. Mus. Nat. 
Hist. Bull., vol. 32, p. 407, fig. 1; 
p. 411, fig. 4B, September 2, 1913 
(Osborn, 1913.400). 

Type locality and geologic horizon. — 
Type from Wind River Basin, Wyo., 
100 feet above Alkah Creek "red 
stratum . ' ' Lamhdotherium-Eotitanops- 
Coryphodon zone (Wind River B, 
"Lost Cabin"). 

Type. — An incomplete lower jaw, 

containing the right lower premolar- 

molar series (pa-ms), also fragments 

of left maxilla containing m", m^ (Am. Mus. 14889). 

(See fig. 142.) 

Specific characters. — Osborn writes: 

Of inferior size. P2-m3, 78.4 millimeters; mi_3, 49; P2-3 
with the internal cusps, paraconid and metaoonid, consisting 
of rectigradations of most rudimentary stage; hypoconulid of 
ms very small; m^ with a single internal cone, no hypocone. 

This very sharply defined species may represent a 
persistent primitive stage, because its recorded 
(Granger) geologic level, 100 feet above the Alkali 
Creek "red stratum," is higher than that of the 



DISCOVERY OF THE TITANOTHERES AND ORIGINAL DESCRIPTIONS 



193 



typical and relatively progressive E. horealis. Its 
primitive condition is shown in the comparison of the 
premolars with the same teeth in E. horealis (Cope) 
and E. princeps Osborn. 

The third inferior premolar is seen to be much less 
progressive than in E. princeps or even in Lamhdo- 




FiGURE 141. — Type (holotype) skull of Rhadinorhinus abboiti 
Field Mus. 12179. After Riggs, 1912. About one-fourth natural size. 



tJierium; the other premolars are also very primitive. 
P2 short, compressed, with a very rudimentary hypo- 
conid; ps laterally compressed, hypoconid distinct, 
paraconid, metaconid, and entoconid extremely rudi- 
mentary rectigradations. In the molar teeth, mi_3, 
the metastylid and entostylid are also in an extremely 
rudimentary or rectigradational stage. In ms the 
hypoconulid is small, subconic, external in position. 

Etymology. — Named in honor of Dr. W. K. Gregory, 
of the American Museum of Natural History, the 
colleague of the author in the preparation of this 
monograph. 

Present determination. — A valid specific stage. 

Eotitanops princeps Osborn, 1913 

('f. Eotitanops princeps, this monograph, page 295 

Original reference. — Am. Mus. Nat. Hist. Bull., vol. 
32, pp. 410-411, fig. 4E, September 2, 1913 (Osborn, 
1913.400). 



Type locality and geologic horizon. — Wind River 
Basin, Wyo.; Lamidotherium-Eotitanops-CorpJiyodon 
zone (Wind River B, "Lost Cabin," exact level not 
recorded). J. L. Wortman, collector. 

Type. — Am. Mus. 296, including lower jaw, femur, 
humerus, right manus, one cervical, three dorsal, and 
one caudal vertebrae. (See figs. 143, 
144.) 

Specific cJiaracters. — Osborn writes: 

Of still larger size, pa-ms 105 millimeters 
(estimated). Inferior premolar teeth some- 
what more complicated, as shown in the type 
specimen. P2 with elevated, distinct, but very 
rudimentary paraconid and metaconid; ento- 
conid very rudimentary; talonid narrow. P3, 
paraconid quite distinct, elevated; metaconid 
small, distinct; entoconid rudimentary ; talonid 
broad. P4, talonid broad; entoconid distinct. 
Hypoconulid of ma rounded, more robust. 
Ramus, larger and more robust. 

The more advanced development of the 
premolar rectigradations, the increased size of 
the teeth and of the jaw, the larger size of the 
hind feet in the referred specimen (Am. Mus. 
4902) combine to distinguish this specimen as 
a mutation or subspecific stage between E. 
horealis and E. maJQr. 

Etymology. — princeps, chief; in allu- 
sion to its comparatively large size. 

Present determination. — A valid spe- 
cific stage. 

Eotitanops major Osborn, 1913 
Cf . Eotitanops major, this monograph, page 296 

Original reference. — Am. Mus. Nat. 
Hist. Bull., vol. 32, pp. 412-413, figs. 
5D, 6, September 2, 1913 (Osborn, 
1913.400). 

Type locality and geologic horizon. — • 

From Alkali Creek, Wind River Basin, 

Lambdoiherium-Eotitanops-CorypTiodon zone 



Wyo.; 




Figure 142. — Type (holotype) teeth of Eotitanops gregoryi 

Am. Mus. 14889. After Osborn, 1913. A, Left m'-m'; B, right lower premolar 
series (P2-ms). Natural size. 

(Wind River B, "Lost Cabin"; exact level unre- 
corded). 

Type. — Am. Mus. 14894, a left median metatarsal; 
also the distal end of the tibia. (See fig. 145.) 



194 



TITANOTHERES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



Specific cJiaracters. — Osborn writes: 

Of superior size, Mts III 104 millimeters longitudinal, 16 
transverse, index 15. 

This ill-defined species indicates the existence in Wind River 
times of a relatively large, short-footed titanothere, v^hich is 




Figure 143. — Lower jaws of Lambdotherium and 
EoHlanops 

A, Lambdothenum popoagicum; B, Eotitanops gregoryi (holotype); 
C, Eoiiianops browniamis; D, Eotitanops boreatis; E, Eotitanops 
princcps (type). .One-fourth natural size. After Osborn, 1913. 

possibly ancestral to some of the short-footed middle Eocene 
types. The comparative measurements with the median 
metatarsal of E. borealis are as follows: 





E. borealis 


E. major 




Mm. 
86 
13 
15 
21 


Mm. 
104 


Width of shaft - 


16 




15 




25 







Etymology. — major, larger; in allusion to the supe- 
rior size of this animal compared with others of the 
same genus. 

Present determination. — A valid specific stage. 

Lambdotherium priscum Osborn, 1913 

Cf. Lambdotherium priscum, this monograph, page 286 

Original rejerence. — Am. Mus. Nat. Hist. Bull., vol. 
32, pp. 413-414, figs. 7A, 9A, September 2, 1913 
(Osborn, 1913.400). 

Type locality and geologic horizon. — -Wind River 
Basin, 3 miles east of Lost Cabin, Wyo.; Lambdo- 
therium-Eotitanops-Ooryphodon zone (Wind River B). 
Granger, American Museum expedition, 1905. 

Type. — Am. Mus. 12822, anterior portion of jaw 
with P2-P4, nil of right side, also ps, mi, m2 of left 
side. Rami fragmentary. (See fig. 146.) 

Specific cJiaracters. — Osborn gives the following 
description: 

P2-P4, 25 millimeters. Second and third lower premolars 
extremely simple, with rudimentary paraconid. Metaconid 
of p3 rudimentary, placed very low upon slope of protoconid; 
talonid narrow, depressed, with cingular rudiment of entoconid. 

The extremely simple or primitive structure of the second 
lower premolar clearly distinguishes this stage. 

A referred specimen (Am. Mus. 14908) is slightly more 
advanced in the structure of the second lower premolar, but is 
still much more primitive than the type of L. popoagicum. 

This specimen was found in the Wind River Basin, Dry 
Muddy Creek, 18 miles up (Granger, Am. Mus. expedition, 
1909). 

The measurements of these two specimens are: 



Type 
(No. 12822) 



Mm. 

Second to fourth premolar, inclusive 25 

Third premolar, anteroposterior 8 

Third premolar, transverse 5 

Fourth premolar, anteroposterior 9 

Fourth premolar, transverse 6. 5 

First molar, anteroposterior 11.5 

First molar, transverse 7. 5 

First to third molar, inclusive 



Beferred 
specimen 
(No. 14908) 



10 

7 
37 



Etymology. — priscus, ancient; in allusion to the 
primitive character of the species. 

Present determination. — A valid specific stage. 

Lambdotherium progressum Osborn, 1913 

Cf. Lambdotherium progressum, this mongraph, page 286 

Original rejerence. — Am. Mus. Nat. Hist. Bull., 
vol. 32, p. 415, fig. 8, September 2, 1913 (Osborn, 
1913.400).- 

Type locality and geologic horizon. — Wind River 
Basin, Wyo. (Alkali Creek, Buck Spring); Lambdo- 
therium- Eotitanops- Cor yphodon zone (Wind River B). 
Granger, American Museum expedition, 1909. 



DISCOVERY OF THE TITANOTHERES AND ORIGINAL DESCRIPTIONS 



195 



Type. — Am. Mus. 14917. Right ramus and sym- 
physis of jaw containing ps-mz of right side, also left 
canine. (See fig. 147.) 

Specific characters. — Osborn writes: 

P2-P4 16.5 millimeters. Second, third, and fourth lower 
premolars progressive. Rudiment of metaconid on p2. Pswith 




Figure 144. — Type (holotype) of Eotitanops 
Left lower grinding teetii. Am. Mus. 296. After Osborn, 1913. 

elevated metaconid subequal with protoconid, broad talonid 
with rudimentary entoconid. P4 with bifid metaconid and 
distinct entoconid. 

This species is readily distinguished from both 
L. priscum and L. popoagicum by the advanced con- 
dition of p3, which may be described as submolariform. 



pnnceps 
Natural size. 




AM 14894^ 

Figure 145. — Type (holotype) of Eotitanops 
major 
Metatarsal (A) and fragment of tibia (B). Am. Mus. 14894. After Osborn, 1913. 
A, Median metatarsal: A', posterior view; A', anterior; A', distal; A*, projdmal. 
B', Distal end of left tibia, anterior view; B^ the same, distal view. All one- 
half natural size. 

Measurements of type 

Millimeters 

Second to fourth lower premolar, inclusive 26 

Second premolar, anteroposterior 8 

Second premolar, transverse (trigonid) 4. 8 

Third premolar, anteroposterior 9 

Third premolar, transverse 6 

Fourth premolar, anteroposterior 9. 3 

Fourth premolar, transverse 7. 3 

First molar, anteroposterior 11.5 

First molar, transverse 8. 5 

Second molar, anteroposterior 12. 5 

Second molar, transverse 9. 5 

Etymology . — progressum , progressive . 
Present determination. — A valid specific stage. 

Diploceras Peterson, 1914 

Cf. Eolitanolherium, this monograph, page 435 

Original reference. — Carnegie Mus. Annals, vol. 9, 
Nos. 1-2, pp. 29-52, text figs. 1-15, pis. 6-10, 1914; 
"issued August 17, 1914" (Peterson, 1914.1). 

Type species. — Diploceras oshorni. 



Generic characters. — Peterson writes: 
Dentition: I|, C^, P-J, M|; premolar series proportionally 
long; p5 with two distinct internal tubercles; horn cores well 
developed; limbs relatively long and slender; tibial trochlea not 
extended back on the calcaneum. 
Astragalus high, with long neck, cal- 
caneal and cuboidal facets laterally 
located. 

Etymology. — SittAoj, double ; 
Kepai, horn. 

Present determination. — The 
name Diploceras being preoccu- 
pied, Eotitanotherium was later 
substituted. (See below.) The 
genus itself is probably related 
to the typical Diplacodon Marsh. 

Diploceras osborn! Peterson, 1914 

Cf. Eotitanotherium oshorni, this monograph, page 435 

Original reference. — Carnegie Mus. Annals, vol. 9, 
Nos. 1-2, pp. 29-52, text figs. 1-15, pis. 6, 7, 1914; 
"issued August 17, 1914" (Peterson, 1914.1). 

Type locality and geologic horizon. — On Duchesne 
River near Myton, Uinta County, Utah; Eohasileus- 
Dolichorhinus zone (upper levels of Uinta B 2). 

Type. — Front of skull, lower jaws, portion of pelvis, 
atlas, portion of axis, fragments of scapula and foot 
bones, No. 2859 (Peterson, figs. 2, 3, 4, 7, 12; pis. 6, 
7, 10). (See figs. 148, 149.) 

Paratypes. — Front of skull, No. 2858; vertebral 
column, fragments of ribs, bones of limb and foot, No. 
2860; crowns of two upper molars, No. 2860a; hu- 
merus, No. 2861; tibiae. No. 2862 (Peterson, figs. 1, 5, 
6, 8, 9, 10, 11, 13, 14, 15; pi. 8). 




Figure 146.- 



-Type (holotype) of Lamhdotherium 
priscum 



Am. Mus. 12822. Ai, Anterior part of lower jaw; As, inner view 
of right pj-p (reversed). After Osborn, 1913. Natural size. 

Specific characters. — Peterson writes: 

Alveolar borders of the premaxillaries extending well in front 
of the canines; nasals long and relatively thin, their anterior 
portion abruptly turned downward and convex on the anterior 
border; incisors well in front of the canines and relatively sub- 
equal in size; canines proportionally small. 



196 



TITANOTHERES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



Etymology. — Named in honor of Prof. H. F. Osborn. 

Present determination. — The genus is doubtfully sep- 
arable from Diplacodon Marsh, but the species differs 
in the more advanced development of the third upper 
premolar. 




FiGUKE 147. — Type (holotype) of Lambdotherium progressum 
Lower jaw. Am. Mus. 14917. After Osborn, 1913. Natural size. 
Heterotitanops Peterson, 1914 
Cf. Metarhinus, this monograph, page 420 
Original reference. — Carnegie Mus. Annals, vol. 9, 
Nos. 1-2, pp. 53-57, text figs. 1, 2; pi. 11, "issued 
August 17, 1914" (Peterson, 1914.2). 

Type species. — Heterotitanops parvus Peterson. 
Generic characters. — Peterson writes: 

Dentition: If?, C-}-?, P|?, M|. Deciduous dentition: If?, 
C-r, Mf?. Rapid increase in size of the deciduous upper 
cheek teeth from first to last tooth. D * with perfectly formed 
internal tubercles (proto- and hypocones) and the antero- 
external angle very greatly developed. Molars hypsodont. 
Ml with large conical proto- and hypocones, the external faces 
of the ectoloph less emarginated anteroposteriorly 
than in the titanotheres generally and the median 
vertical ridge of the ectoloph projecting forward to a 
greater degree. 

Etymology. — erepoj, other, different; Ttrdi', 
Titan; ajf, face; in allusion to its supposed 
possible relationship to such forms as Eoti- 
tanops. 

Present determination. — According to Dr. 
W. K. Gregory, who has studied the type 
specimen of Heterotitanops parvus, the animal 
probably represents a very young individual of 
Metarhinus or Rhadinorhinus. 

Heterotitanops parvus Peterson, 1914 

Cf. Metarhinus sp. or Rhadinorhinus sp., this mono- 
graph, page 198 

Original reference. — Carnegie Mus. Annals, 
vol. 9, Nos. 1-2, pp. 53-57, text figs. 1, 2, 
pi. 11, 1914; "issued August 17, 1914" 
(Peterson, 1914.2). ^"'""'° 

Type locality and geologic horizon. — White River, 
Uinta County, Utah; base of Metarhinus zone (Uinta 
B 1). The type specimen "was found articulated 
in a hard sandstone concretion, and lower down in 
horizon A ['^J of the Uinta sediment than any mam- 

" The upper or fosslliferous part of Uinta A of previous reports is Uinta B 1 of 
this monograph. 



malian remains hitherto described from that forma- 
tion." (Peterson.) 

Type. — Skull, lower javrs, vertebral column, ribs, 
limb bones, calcaneum, and astragalus of young indivi- 
dual (Carnegie Mus. 2909). (See figs. 150, 151, 152, 360.) 
Specific characters. — Not determined. 
Etymology. — parvus, poor, small. 
Present determination. — According to Dr. W. K. 
Gregory the type specimen probably represents a 
very young individual of an undetermined species of 
one of the previously described genera of Uinta 
Basin titanotheres, probably of Metarhinus. 

Eotitanotherium Peterson, 1914 

(To replace Diploceras Peterson, 1913, preoccupied) 
Cf. Eotitanotherium, this monograph, page 435 

Original reference. — Carnegie Mus. Annals, vol. 9, 
p. 220, September 12, 1914 (Peterson, 19U.4); Eotitano- 
therium, a new generic name to replace Diploceras 
Peterson. (See Peterson, 1914.1.) 

In my article entitled "A new titanothere from the Uinta 
Eocene" I employed the generic name Diploceras, having 
overlooked the fact that this name is already preoccupied, 
having been employed by Conrad as early as 1844 to designate 
a genus belonging to the Mollusca. For this name I now sub- 
stitute the name Eotitanotherium, which, after a diligent search 
of the literature, I believe is not preoccupied. (Peterson.) 

Etymology. — ^cos, dawn; Ttrav, a Titan; drjpiov, a 
beast. 




FiGUEE 148. — Type of Diploceras osborni 
I lower jaw. Carnegie Mus. 2859. After Peterson, 1914. One-fourth natural size. 

Present determination. — The genus is doubtfully 
separable from Diplacodon Marsh. 

Telmatherium? birmanicum Pilgrim and Cotter, 1916 

Cf. Telmatherium f birmanicum, this monograph, pages 196-199 
Original reference. — India Geol. Survey Records, vol. 
47, pt. 1, pp. 72-74, pi. 5, figs. 9-11, 1916 (Pilgrim and 
Cotter, 1916.1). 



DISCOVERY OF THE TITANOTHERES AND ORIGINAL DESCRIPTIONS 



197 



Type locality and geologic Jiorizon. — Myaing Town- 
ship of the Pakokku district, Burma; Pondaung sand- 
stone (upper to middle Eocene). 

Cotypes. — Pilgrim and Cotter write: 

This species is represented by five fragments of upper 
molars, two of which are ahnost identical in shape and com- 
prise the antero-internal quarter of two of the upper molars 
probably occupying successive positions in the maxilla and 



ably more behind the level of the paracone than is the case in 
the Chalicotheriidae; thirdly, because in pm^ there is a single 
large rounded and isolated inner cusp — the protocone, which is 
totally unconnected with the two main outer cusps — a condi- 
tion which never occurs in any chalicotheroid. In that family 
the protocone in the premolars is connected to the outer cusps 
either by a single or by a double crest. In addition to these 
specific differences, the general structure of the tooth is unlike 
that of any chalicotheroid that is known to us. 




Figure 149. — Type of Diploceras {Eolitaiiotherimn) osborni 
Palatal view. Carnegie Mus. 2859. After Peterson, 1914. One-half natural size. 



being either m^ and m' or m' and m', two other portions of 
the wall of the external crescents, and another an isolated proto- 
cone. A sixth fragment consists only of the internal half of 
what we take to be the last upper premolar. Three of these 
pieces are figured in Plate 5, Figure 11 [9-11]. (See fig. 153.] 

Systematic characters. — Pilgrim and Cotter write: 

It is obvious that these are not chalicotheroid; first because 
there is no trace of a protoconule, which in the Chalicotheriidae 
is always present between the protocone and the paracone, 
being invariably united to the latter by a transverse crest; 
secondly, because the protocone in our specimens lies consider- 



On the other hand, it approximates so nearly to that of many 
of the Titanotheriidae that we have no hesitation in assigning 
these fragments to that family. A careful comparison with the 
various known species of the Titanotheriidae convinces us 
that the Burmese fragments belong to a new species, but whether 
this is to be referred to one of the known genera of that family 
or whether it belongs to a new genus is a point which the 
material at our disposal is insufficient to enable us to deter- 
mine. We shall therefore do no more than indicate its prob- 
able affinities, leaving a definite conclusion to the future, 
when we may hope that more abundant material may come 
to fight. 



198 



TITANOTHERES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



One of the most crucial points which has presented itself to 
us for decision in connection with the material belonging to this 
species is the position in the jaw of the tooth (G. S. I. No. C. 



widening which we must assume to have taken place in m' of 
this species. Again the faint V-ing of the line which connects 
the two external crescents points to these being more closely 




FiGUEE 150. — Type (holotype) skeleton of Heterotitanops parvus 
Carnegie Mus. 2909. After Peterson, 1914. One-fourth natural size. 



315) figured in Plate 5, Figure II. Although in some respects 
this specimen reminds us of the last upper molar in some of the 




FiGtTBE 151. — Type (holotype) skull of Heterotitanops parvus 

Carnegie Miis. 2909. After Peterson, 1914. One-half natural size. 

upper Eocene members of the Palaeosyopinae, yet its small 

size as compared with the two other specimens of the upper 




connected than is the case in the last upper molar of a titano- 
there. On the other hand these features are such as the last 
upper premolar of that family 
would present, the only peculiar- 
ities being the rounded nature of 
the inner cone and the highly de- 
veloped cingula on the anterior 
and posterior margins of the frag- 
ment, dying away internally and 
apparently also on either side of 
the two main external cusps. 

It is evident that this simple struc- 
ture of pm* prohibits the possibility 
of this species being one of theTitan- 
otheriinae of the Oligocene, while 
on the other hand the increased 
development of the cingulum and 
the absence of an intermediate 
tubercle point to its representing one of the latest develop- 
mental stages of the Eocene subfamily of the Palaeosyopinae. 
A similar indication is afforded by the fragmentary upper 



Figure 152. — Type (holo- 
type) of Heterotitanops 
parvus 

Upper and lower teeth. Carnegie 
Mus. 2909. After Peterson, 
1914. 1, Deciduous upper pre- 
molars, first permanent molar; 
2, permanent mi. One-half 
natural size. 






Figure 153. — Cotypes of Telmatherium? birmanicum 

In the collection of the Geological Survey of India. After Pilgrim and Cotter, 1916. Naturalsize. A, "The antero-internal 
portion of a right upper molar, surface view"; B, "e-xternal portion of an upper molar, showing the gently rounded 
median fold, external view"; C, "internal portion of last upper premolar, surface view." 



molars militates against this view. Further, the almost 
rectangular shape of the inner portion of the tooth, which 
alone is preserved to us, is inconsistent with the external 



molars, in which the protocone is rather lofty and the only 
vestige of a protoconule is the presence of a minute row of 
fringing the protocone between it and the paracone. 



DISCOVERY OF THE TITANOTHERES AND ORIGINAL DESCRIPTIONS 



199 



These start from the prominent cingular protostyle and cul- 
minate in a more elevated portion some 13 millimeters to the 
rear, diminishing again behind this point. 

Attention may also be called to the presence in one of the 
specimens of a broad, gently rounded median fold iu the 
center of the external paraconal wall of the tooth, although in 
the other specimen no such fold is visible. According to 
Earle such a median rib is characteristic of all the early titano- 
theres, tending to vanish in the upper Eocene and being 
entirely absent in the Oligocene subfamily of the Titano- 




Measurements of inferior teeth Pi-m^ and superior teeth m'-m^ 

Millimeters 
P2-m3: Huerfano A. L. priscum (ref.), Am. Mus. 17526. 67 
Wind River B. L. popoagicum (type), Am. Mus. 

4863 69 

Wind River B. L. progressum (type), Am. Mus. 

14917 (estimated) 71 

Huerfano A. L. magnum (type), Am. Mus. 17527. 74 
Mi-m^: Huerfano A. L. priscum (ref.), Am. Mus. 17529. 21. 5 
Huerfano A. L. priscum (ref.). Am. Mus. 2688.. 22. 5 
Wind River B. L. popoagicum (ref.), Am. 

Mus. 14902 25 

Huerfano A. L. progressum (ref.), Am. 

Mus. 17530 23.5 

Wind River B. L. magnum (ref.). Am. 

Mus. 15600 27.5 

These measurements show that there is not a 
great range in size between the smaller and the 
larger animals referred to this genus. 

Etymology. — magnum, large. 

Present determination. — A valid specific 



Figure 154. — Type (holotype) of Lambdotherium magni 
Lower jaw. Am. Mus. 17527. After Osborn, 1919. Natural size. 

theriinae. In any case the external lobes are broad and flat 
and considerably elevated, hli;e those of the latest members of 
the Palaeosyops-Diplacodon phyla. 

Perhaps taking everything into consideration the present 
species shows greater affinities with Telmatherium than any 
other known titanotherid genus. 

Etymology.- — Mrmanicum, relating to Burma. 

Present determination. — Position uncertain. The 
very close beading and massive cones of the single 
grinding tooth figured suggest comparison with 
Palaeosyops, a progressive species like P. copei. 
These teeth might belong to a chalicothere, such as 
Macrotherium or Moropus, but the resemblance is 
not close. 

Lambdotherium magnum Osborn, 1919 

Cf. Lambdotherium magnum, this monograph, page 288 

Original reference. — Am. Mus. Nat. Hist. Bull., vol. 
41, p. 562, fig. 3, 1919 (Osborn, 1919.494). 

Type locality and geologic Jiorizon. — Lower horizon of 
the Huerfano formation (Huerfano A) of Colorado. 

Specific characters. — Osborn writes: 

Exceeding in size any other known lambdothere is the type 
jaw (Am. Mus. 17527) from the Garcia Canon, lower Huerfano, 
containing a complete inferior series, p2-m3 of both sides, 
represented in Figure 3. (1) These teeth exceed in length over 
all (74 mm.) those of the type of L. popoagicum, in which the 
same teeth measure 69 millimeters. (2) P3 has a rudimentary 
metaconid and paraconid, in the same stage of evolution as in L. 
popoagicum. (3) Of similar large size is a referred specimen. 
Am. Mus. 15600, from the Big Horn, west end of Tatman Moun- 
tain. "These_ referred grinders, m', m^, coincide closely in size 
with the type of L. magnum and may be regarded as a paratype. 
[See fig. 154.] 



stage. 

Eotitanops minimus Osborn, 1919 

Cf. Eotitanops minimus, this monograph, page 296 

Original reference. — Am. Mus. Nat. Hist. Bull., vol. 
41, p. 564, fig. 4, A, A\ 1919 (Osborn 1919.494). 

Type locality and geologic Jiorizon. — Two miles north 
of Gardner, Huerfano Basin, Colorado; from the lower 
level of the upper horizon of the Huerfano formation 
(Huerfano B). 

Specific characters. — Osborn writes: 

In reference to the fact that it is the smallest true titanothere 
known, these type lower molar teeth, pi-ms. Am. Mus. 17439 
(fig. 4, A, A'), * * * are assigned a new specific name 
on the following grounds: (1) The measurement of p4-m3 (53 
mm.) is much less than that (58) of the corresponding teeth 



E.minlmus, Type 



Figure 155. — Type (holotype) of Eotitanops minimus 

Lower teeth. Am. Mus. 17439. After Osborn, 1919. A, Lingual or internal view; 
A^, crown view. Natural size. 

in E. gregoryi; (2) the other characters are so similar to those 
of E. gregoryi as to suggest that this is a related form. [See 
fig. 155.] 

The accompanying figures (fig. 4, A, B, C) exhibit the 
dimensional proportions of the above species of Eoiilaiiops. 
It has been found from the large number of measurements of 
Eocene titanotheres that no single species exhibits so great 
a range of size. 

Etymology. — minimus, small. 

Present determination. — ^A valid specific stage. 



200 



TITANOTHERES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 
Eometarhinus Osborn, 1919 Eometarhinus huerfanensis Osborn, 1919 



Cf. Eometarhinus, this monograph, page 419 
Original reference. — Am. Mus. Nat. Hist. Bull., vol. 
41, pp. 568, 569, 1919 (Osborn, 1919.494). 
Generic characters. — Osborn writes: 

Small; ancestral to Metarhinus; with rudimentary frontonasal 
horn; nasals elongate; overhanging premaxillaries, decurved as 



Cf. Eometarhinus huerfanensis, this monograph, page 420 

Original reference. — Am. Mus. Nat. Hist. Bull., vol. 
41, pp. 567-569, fig. 6, 1919 (Osborn, 1919.494). 

Type locality and geologic horizon. — Huerf ano-Muddy 
divide, 3 miles west of Gardner, Huerfano Basin, Colo.; 
Huerfano formation, 205 feet below top (Huerfano B). 




Figure 156. — Type (holotype) skull of Eometarhinus huerfanensis 

11. Mus. 17412. After Osborn, 1919. A, Nasals, superior view; Ai, Aj, sections; B, skull, view of 
left side; C, right upper jaw and teeth. One-half natural size. 



in Metarhinus; no infraorbital shelf; characters apparently in- 
termediate between those of the Metarhinus and Mesatirhinus 
phyla. 

Etymology. — rjcos, dawn; Metarhinus, a genus of the 
middle Bridger beds; indicating an ancestral form of 
Metarhinus. 

Present determination. — This genus appears to be 
ancestral to the Dolichorhinus phylum. 



Type. — Anterior portion of skull (Am. Mus. 17412). 
(See fig. 156.) 

Specific characters. — Inferior in all measurements to 
Mesatirhinus megarhinus. Premolars with small deu- 
terocone. p'-m', 124 millimeters ; p'-p*, 53; m'-m', 72. 

Etymology. — huerfanensis, in allusion to type locality. 

Present determination. — A valid specific stage. 



DISCOVERY OF THE TITANOTHERES AND ORIGINAL DESCRIPTIONS 



201 



SECTION 3. ORIGINAL DESCRIPTIONS OF TYPES OF 
OLIGOCENE TITANOTHERES 

IIST OF GENERA AND SPECIES 

The following list shows that 49 species of Oligocene 
titanotheres in North America and Europe have been 
described and made the types of 17 different genera, 
of which seven are regarded as valid. The types come 
from many geologic levels. In finally determining the 
genera wo are reluctantly compelled to adopt Menodus 



Pomel in preference to Titanotherium Leidy, to adopt 
Megacerops Leidy although it is based on a poor type, 
and to reject Symhorodon Cope, because the genotype 
species belongs to Menodus. The genera that rest on 
the genotypic specimens are Brontotherium Marsh and 
Brontops Marsh. Diplacodon Marsh is the least 
soundly determined. It is close to Brontops and may 
represent a sport. Teleodus Marsh represents an in- 
ferior stage of the Brontops phylum, transitional to 
Protitanotherium. 



Chronologic list of the genera and species of Oligocene titanotheres 

[Generic names accepted in this work as valid are printed in small capitals; abandoned names are inclosed in brackets.] 





Date 




1846 


I 


1849 


1 


1849 


2 


1850 


II 


1852 


3 


1852 


4 


1852 


III 


1853 


6 


1854 


IV 


1860 


V 


1870 


6 


1870 


VI 


1873 


7 


1873 


VII 


1873 


8 


1873 


VIII 


1873 


9 


1873 


10 


1873 


11 


1873 


12 


1873 


13 


1873 


14 


1873 


1.5 


1873 


16 


1874 


IX 


1875 


17 


1875 


X 


1876 


18 


1886 


19 


1887 


20 


1887 


21 


1887 


XI 


1887 


22 


1887 


23 


1887 


XII 


1887 


24 


1887 


XIII 


1887 


25 


1887 


26 


1887 


XIV 


1887 



Present determination 



["Gigantic Palaeotheri- 
um."] 

Menodus 

Menodus 

[Palaeotherium(?)] 



giganteus. 
proutii 



[Titanotheriu m] 

[ Palaeothieriu m] 

[Rhinoceros] 

[Eotherium.] (Type Rhi- 
noceros americanus 
Leidy.) 

[Palaeotherium] 

[Leidyotherium] 

Megacerops 

Megacerops 

Brontotherium 

Brontotherium 

[Symborodon] 

[Symborodon] 

[Miobasileus] 

[ Miobasileus] 

Megaceratops 

[ Megaceratops] 

[Symborodon] 

[Symborodon] 

[Symborodon] 

[Brontotherium] 

[Symborodon] 

[ Anisacodon] 

[Anisaoodon] 

[Diconodon (not Anisa- 
codon).] 

[Menodus] 

[ Menodus] 



[maximum], 
[americanus] 



[giganteum]. 



coloradensis. 
gigas 



torvus. 



[ophryas] 

acer 

heloceras 

bucco 

[altirostris] . _ 
trigonoceras- 

[ingens] 

hypoceras 



[montanus]. 



angustigenis. 
tichoceras 



[Menodus]- 
[Menodus]_ 



dolichoceras _ 
platyceras 



101959 



Brontops 

Brontops 

Brontops 

[M'enops] 

[Menops] 

[Titanops] 

[Titanops] 

[Titanops] 

Allops 

—29— VOL 1 16 



robustus- 
dispar 



curtus.. 
[elatus]_ 



Prout- 



Pomel _ 
do_ 



Owen, Norwood, and 
Evans. 

Leidy 

do 



_do_ 
-do. 



do_ 

Prout_. 
Leidy- . 

do_ 



Marsh. 
do_ 

Cope.- 
do. 



.do_ 
.do. 
_do. 



.do. 
_do. 
.do_ 



do. 

Marsh. 
Cope.- 
Marsh. 

do. 

do. 



Cope 

Scott and Osborn. 



.do. 



Marsh. 

do. 

do. 



.do. 
.do. 
.do. 



.do. 
-do. 
.do. 



Menodus Pomel. 



Do. 



Menodus giganteus Pomel. 
Menodus proutii (Owen, Norwood, 
and Evans). 
Do. 
(Indeterminate.) 

Do. 
Subfamily Menodontinae, genus in- 
determinate. 

(Indeterminate.) 

Do. 
Megacerops Leidy. 
Megacerops coloradensis Leidy. 
Brontotherium Marsh. 
Brontotherium gigas Marsh. 
Menodus Pomel. 
Menodus torvus (Cope). 
(Indeterminate.) 

Do. 
Megacerops acer Cope. 
Menodus heloceras (Cope). 
Megacerops bucco (Cope). 
Megacerops acer Cope. 
Menodus trigonoceras (Cope). 
Menodus giganteus Pomel. 
Brontotherium hypoceras (Cope). 
(Indeterminate.) 
Menodus giganteus Pomel. 
Menodus giganteus? Pomel. 

?Brontops angustigenis (Cope). 
Brontotherium tichoceras (Scott and 

Osborn) . 
Brontotherium dolichoceras (Scott 

and Osborn). 
Brontotherium platyceras (Scott and 

Osborn) . 
Brontops Marsh. 
Brontops robustus Marsh. 
Brontops dispar Marsh. 
Menodus Pomel. 
Menodus varians (Marsh). 
Brontotherium Marsh. 
Brontotherium curtum (Marsh). 
Brontotherium gigas Marsh. 
Allops Marsh. 



202 



TITANOTHERES OP ANCIENT "WTOMING, DAKOTA, AND NEBRASKA 

Chronologic list of the genera and species of Oligocene titanotheres — Continued 

[Generic names accepted in ttiis work as valid are printed in small capitals; abandoned names are inclosed in brackets.] 



Present determination 



27 
XV 

28 
29 
XVI 
30 
XVII 
31 
32 
33 
34 
35 
36 
37 
38 
39 
40 
41 
42 
43 
44 
45 
46 
47 
48 
49 



1887 
1S89 
1889 
1889 
1890 
1890 
1890 
1890 
1891 
1891 
1891 
1891 
1892 
1896 
1902 
1902 
1902 
1902 
1905 
1908 
1908 
1908 
1908 
1913 
1916 
1916 



serotinus. 



selwynianus . 
syceras 



amplus. 



AUops 

[Haplacodon] " 

[Menodus] 

[Menodus] 

DiPLOCLONUS 

Diploclonus 

Teleodus 

Teleodus 

Allops 

Brontops 

[Titanops] 

[Menodus] 

[Menodus(?)] 

[Titanotherium] 

[ Megacerops] 

[ Megacerops] 

[Megacerops] 

Brontotherium 

[Megacerops] 

Brontotherium 

[Symborodon] 

[ Megacerops] j primitivus — 

Megacerops assiniboiensis 

[Titanotheriu m] | [bohemicum] _ 

Allops ' walcotti 

Megacerops j riggsi 



Marsh. 
Gope_- 
do_ 



avus 

crassicornis 

[validus] 

medius 

[peltoceras] 

rumelicus 

ramosum 

braehycephalus - 

bicornutus 

marshi 

leidyi 

tyleri 

hatcheri 

copei 



do_ 

Marsh. 
do- 



.do. 
.do. 
-do. 



do_ 

do_ 

Cope.. 

Toula.. 

Osborn. 
do. 



.do. 
.do. 
.do. 



LuU... 
Osborn. 
do- 



Lambe.- 
do.. 

Kiernik. 

Osborn.. 
do_. 



Allops serotinus Marsh. 
Allops sp. 

Diploclonus selwynianus (Cope). 
? Megacerops syceras (Cope). 
Diploclonus Marsh. 
Diploclonus amplus Marsh. 
Teleodus Marsh. 
Teleodus avus Marsh. 
Allops crassicornis Marsh. 
Brontops dispar Marsh. 
Brontotherium medium (Marsh). 
? Brontotherium curtum (Marsh). 
? Brontotherium rumelicum (Toula). 
Brontotherium ramosum (Osborn). 
Brontops braehycephalus (Osborn) . 
?Diploclonus bicornutus (Osborn). 
Allops marshi (Osborn). 
Brontotherium leidyi Osborn. 
? Diploclonus tyleri (Lull). 
Brontotherium hatcheri Osborn. 
Megacerops copei (Osborn). 
Teleodus primitivus (Lambe). 
Megacerops assiniboiensis Lambe. 
Menodus giganteus Pomel. 
AUops walcotti Osborn. 
Megacerops riggsi Osborn. 



« Genotype Menodus angastigenis, upper teeth only. See No. 18, above. 

PROUT'S DESCRIPTIONS OF A FRAGMENTARY lOWER JAW, 
THE FIRST TITANOTHERE MADE KNOWN TO SCIENCE 

"Gigantic Palaeotherium," Prout, 1846 

Original reference. — Am. Jour. Sci., 2d ser., vol. 2, 
pp. 288-289, 1 fig., 1846 (Prout, 1846.1). 

Subsequent references. — Leidy, Description of the 
remains of extinct IVIammalia and Clielonia from 
Nebraska Territory, in Owen, Report of a geological 
survey of Wisconsin, Iowa, and JMinnesota, p. 551, 
1852 [Tab. 9, figs. 3, 3a, is not Prout's specimen] 
(Leidy, 1852.1); The ancient fauna of Nebraska, pp. 
72, 114, pi. 16, fig. 1, 1853 (Leidy, 1854.1). 

Original description. — Dana and Silliman write: 

Gigantic Palaeotherium. — We have recently received infor- 
mation from Mr. H. A. Prout, of his discovery of the remains 
of a Palaeotherium in the Tertiary near St. Louis, and we are 
also indebted to him for a cast of the jaw, a view of the pos- 
terior tooth of which is represented below. Mr. Prout is pre- 
paring a memoir on the subject; and in the meantime we 
state the following facts from his letter. 

This fossil was found in the great northwestern Tertiary 
belt, which is deflected from the north by the Black Hills and 
which crosses the Missouri River at about latitude 43°. It 
was accompanied by several Baculites compressus, an Inocera- 
mus concentricus, a vertebra of a large fish, and some crystallized 
gypsum. [As noted later by Prout these were from the Creta- 
ceous and from another locality.] The entire jawbone, judg- 
ing from the decrease in size of the teeth, must have been at 
least 30 inches long, which far exceeds in size the Palaeotherium 
magnum. The face of the posterior tooth is 4^ inches in 



length; and from the posterior side of the last tooth to the 
anterior side of the antepenultimate molar of the same side 
the distance in the specimen is 11 inches. [See fig. 157.] This 
is the aggregate length, in the line of the jaw, of but three out 
of seven teeth; and with the most liberal allowance for decrease 
of size in the other four the whole of the seven could not have 
measured less than 16 or 18 inches, which is about one-half 
larger than in the P. magnum. 

Remarlcs. — This specimen was "the first of the 
many mammalian remains which have been brought 
to the notice of the scientific world from the vast 
Eocene cemetery of Nebraska" (Leidy, 1852.1, p. 551). 
It was the subject of Prout's second article cited below 
and was the type of Menodus giganteus Pomel and one 
of the cotypes of Palaeotherium? proutii Owen, Nor- 
wood, and Evans (1850.1) and of Titanotherium, 
proutii. 

" Fossil maxillary bone of a Palaeotherium," Prout, 1847 

Original reference. — Am. Jour. Sci., 2d ser., vol. 3, 
pp. 249, 250, 1 fig., 1847 (Prout, 1847.1). 

Subsequent references. — (See p. 204.) 

Prout's description. — The following notice, written 
by Dr. Prout himself, is a full description of the same 
lower jawbone mentioned in his letter of the preceding 
year: 

The palaeotherial bone here described was sent to me some 
time ago by a friend residing at one of the trading posts of the 
St. Louis Fur Co., on the Missouri River. From information 
since obtained from him, I learn that it was discovered in the 



DISCOVERY OF THE TITANOTHERES AND ORIGINAL DESCRIPTIONS 



203 



Mauvais Terre, on the White River, one of the western confluents 
of the Missouri, about 150 miles south of St. Pierre, and 60 
east of the Black Hills, at a point which would very nearly 



The fifth and sixth molars (first and second true molars) re- 
semble the one described, except that they want the third lobe, 
and the dentine area on the crown of each lobe is much larger. 




Figure 157. — "Vertical view of the posterior tooth belonging to the lower jaw of 
Mr. Prout's Palaeotherium" 
After Prout, 1846. Natural size. 



The sixth is 33^ inches from front to posterior side. The 
posterior lobe is 2 inches from the outer to the inner surface 
and 1% inches long in the line of the jaw. The whole distance 
on the jaw occupied by the three teeth is 11 inches. In the 




correspond with the intersection of latitude 43" with longitude 
26° west of Washington. 

The Baculites and the Inoceramus which accompanied it and 
which I at first supposed belonged to the same locality were 
found in another formation — probably the Cretaceous- 
distant about 100 miles, and included in the Grande 
Detour or Great Bend of the Missouri River. 

This fossil bone is a fragment of the inferior maxillary 
of the left side, consisting of the posterior part of the 
bone, together with the last three molar teeth. The 
ramus is much fractured and presents an irregular sur- 
face; yet the general direction of its outline may be 
made out. The length of this fragment is 15 inches, 
its depth from the liighest point of the ramus (a) to 
the lowest (h) is 9K inches: it narrows regularly forward 
so as to measure only 3}4 inches from the lower sur- 
face of the bone at (d) to the alveolar process of the 
antepenultimate tooth at (c). The inner surface of the 
bone is more uniform, being marked merely by depres- 
sions for the attachment of muscles. The alveolar por- 
tion is here very prominent and well rounded, the teeth 
being planted more than an inch from a vertical line which 
is tangential to the inner surface of the bone. It is 
covered in places with a concretionary matter which 
could not be removed without injury to the specimen; on 
analysis, this was found to consist chiefly of carbonate of 
lime, with some alumina, and a small proportion of silex. 
The last molar tooth has the three lobes of the Pa- 
laeotheria, as shown in Figure 2. The inner surface is 
nearly smooth and flat and shows no trace of lobes. 
The size of the tooth from posterior to anterior sides is 4}/^ 
inches, of which 1^ inches belong to the anterior lobe, the 
same to the middle, and 134 inches to the posterior. In 
an upper view the two larger lobes have a deltoid form, 
with the sides somewhat convex, and a rounded outer 
angle. The thickness through from the outer to the op- 
posite side is 15^ inches. The enamel of the inner side 
folds over the surface, covering nearly a semicircular space 
and leaving between it and the edge of the posterior en- 
amel a subcrescent-shaped space (deltoido-lunate) of den- 
tine, somewhat concave, which is nearly seven-eighths of 
an inch broad at its widest part. These crescent-shaped Figure 158. — Original figures of Prout's "gigantic Palaeotherium," the 
areas of the two lobes are connected by a continuous tract first titanothere discovered 

of dentine, nearly IJ^ lines wide at the narrowest part; 
and the same tract continues from the middle lobe to the 
posterior; upon the latter it does not widen over the in- 
terior, as the reflexed inner enamel covers the whole of the crown, 
excepting a narrow space adjoining the posterior enamel. The 
prominent points of the crown between the lobes project about 
half an inch; and probably much more in the perfect tooth. 




After Prout, 1847. A, "Fragment of the inferior maiillary of tlie left side," one-fourth natural 
size; B, last lower molar on the left side, four-fLIths natural size. 



largest Palaeotherium hitherto described, the P. magnum, the 
same teeth occupy a space scarcely one-third that of the Mis- 
souri animal. 

St. Louis, December 10, 1846. 



204 



TITANOTHERES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



POMEL'S GENUS MENODUS, BASED ON PROUT'S 
DESCRIPTION AND FIGURE OF THE FRAG- 
MENTARY LOWER JAW 

Menodus Pomel, 1849 

Cf. Menodus, this monograph, page 522 

Original reference. — Bibliotheque universelle de 
Geneve (Supp.) Arch. sci. phys. nat., vol. 10, pp. 
73-75, January, 1849 (Pomel, 1849.1). 

Type species. — Menodus giganteus Pomel. 

Original description. — Pomel writes : 

Ce fossile a 6t6 dScouvert k Mauvais-Terre sur la Riviere 
Blanche a 43° latitude nord et 26° longitude ouest de Wash- 
ington, sur le versant occidental du bassin du Missouri. C'est 
un fragment de mandibule portant les deux dernieres molaires 
et I'alveole de I'antepenultieme, qui montrent tous les caracteres 
du genre palaeotherium. La derniere molaire, la mieux oon- 
servee, indique une espece plus voisine des vrais palaeotherium 
(dont les P. magnum, medium, etc., sont les types) ou du sous- 



Malheureusement on ignore I'age du terrain ou ce fossile 
remarquable a 6t6 decouvert, quoiqu'il soit probable que c'est 
dans la serie des formations de I'epoque alluviale qu'il faudra le 
ranger. Cette difference d'age entre ce palaeothere et ceux de 
I'Europe ocoidentale, ne doit pas etonner, puisque Ton trouve 
dans I'Amerique du sud, dans des formations de meme age, un 
animal de la meme tribu (on pourrait dire du meme grand genre), 
le macrauchenia qui, lui aussi, est d'une taille superieure aux 
esptices d' Europe. On salt, du reste, que sa derniere molaire 
inf^rieure n'a que deux collines, comme dans le paloplotherium, 
et que ses membres sont assez greles, tandis qu'il est probable 
qu'un animal aussi gigantesque que ce nouveau palaeotherium 
a ete assez trapu. Nous proposons de designer cette forme 
animale fossile sous le nom de Menodus giganteus, en la consi- 
derant comme un sous-genre des palaeotherium. 

Etymology. — fxrivrj, the moon; 65ovs, tooth; in allu- 
sion to the crescents of the lower molars. 

Present determination. — Pomel proposed Menodus 
as a subgenus of PalaeotJierium, using the latter term 
in a very comprehensive sense, as later authors would 




genre plagiolophus, que des anchitherium et des paloplotherium, 
en ce que la troisieme coUine est bien d^veloppee, et forme un 
troisieme croissant k la couronne; les autres croissants sont un 
peu anguleux (croissants deltoides, dit I'auteur). La base de 
la couronne est entouree d'un petit bourrelet comme dans les 
palaeotherium d'Europe; mais si le dessin est exact, la maniere 
dont les croissants principaux se reunissent indiquerait quelque 
rapport avec ce qui existe chez les anchitheriums et les paloplo- 
theriums, cette partie dtant plus 6paissie. II serait n^cessaire 
d'en connaitre une molaire superieure pour fixer sa veritable 
place; nous serions porte a presumer toutefois, que ce palaeothe- 
rium est le type d'un sous-genre particulier; car independam- 
ment de la brievetiS du fAt de la couronne des molaires, sa taille 
est trop au-dessus de ceUe de nos plus grandes espfeoes euro- 
p^ennes, pour qu'on puisse admettre sans hesitation son identity 
subg^nerique avec ceUes-ci. En effet, I'arriere-molaire du 
palaeotherium magnum est k peine le tiers de celle de I'espece 
americaine: aussi cette derniere est-elle r^ellement colossale, 
mesurant 0m,116, dont 0m,032 appartiennent k la troisieme 
coUine; son ^paisseur est 0m,045. L'os mandibulaire est, 
comme on devait s'y attendre, tres-robuste; il a 0m,112 de 
diametre vertical entre les deux arriere-molaires; il s'elargit 
consid^rablement k la partie du bord inf^rieur situ^e sous la 
branche montante. 



Figure 159. — Type of Menodus giganteus 

Prout's original specimen. After Leidy, 1854. One-tlurd natural size. 

speak of a family. In 1873 Marsh (1873.1, p. 486) 
rejected the name Menodus on the ground that it was 
essentially the same word as Menodon Meyer, 1838, a 
genus of reptiles (Palmer, 1904.1, p. 410); but, as the 
two names are spelled differently, according to the 
modern rules of nomenclature' Menodus Pomel can 
not be rejected on that ground. As shown below, the 
type species Menodus giganteus rests upon Prout's 
specimen, of which an excellent figure was given later 
by Leidy (1854.1, pi. 16, fig. 1). 



Menodus giganteus Pomel, 1849 

Cf. Menodus giganteus, this monograph, pages 530, 535 

Original reference. — See genus Menodus, above. 

Type specimen. — As noted above, the species rests 
upon Prout's original specimen, which was figured by 
Prout in 1847 (1847.1, p. 249, and 1 fig.) and by 
Leidy under the name Titanotherium proutii in 1854 
(1854.1, pi. 16, fig. 1 only). The type may have been 
destroyed in the "great fire" of St. Louis. 



DISCOVEKY OF THE TITANOTHEEES AND ORIGINAL DESCRIPTIONS 



205 



Neotype (Osborn). — A carefully made model, based 
on Leidy's figures and measurements of the lower jaw, 
was compared with various specimens of Menodus 
untn an upper dentition was found (in a skull, Am. 
Mus. 505) which appears to fit very well the lower 
teeth of the type. Hence the skull (Am. Mus. 505) 
has been selected as a neotype of Menodus giganteus. 

Specific characters. — Not separated from the generic 
characters in Pomel's description. (See p. 530.) 

Etymology. — giganteus, gigantic ; because larger than 
the Palaeotherium magnum. 

Present determination. — Although Prout's original 
specimen, the type of Menodus giganteus Pomel, has 
been lost, Leidy's carefully executed figure of this 
specimen, together with his measurements and descrip- 
tions, reveals generic and specific identity with the 
dolichocephalic titanothere which Osborn in 1902 
designated (1902.208, p. 96) Titanotherium ingens 
Marsh. Titanotherium ingens is therefore to be 
regarded as a synonym of Menodus giganteus Pomel. 



Type. — From a study of the foregoing references it 
is evident that Owen, Norwood, and Evans intended 
the name Palaeotherium? proutii to cover both Prout's 
original specimen and "Owen's specimen," discovered 
by Evans, the lower jaw which was figured by Leidy 
in 1852 (1852.1, pi. 9, figs. 3, 3a) and is still preserved 
in the United States National Museum (No. 113; 
our fig. 160). Prout's specimen is the type of Meno- 
dus giganteus Pomel; hence, by the method of elimina- 
tion, Owen's specimen becomes the type of Palaeothe- 
rium? proutii Owen, Norwood, and Evans. 

Etymology. — Named in honor of Dr. Hiram Prout. 

Present determination. — "Owen's specimen" (Nat. 
Mus. 113) appears to represent a Menodus, of a stage 
slightly smaller than M. trigonoceras. (See p. 528.) 

Titanotherium Leidy, 1852 
Cf. Menodus, this monograph, page 522 

Original reference. — "Palaeotherium? proutii Owen, 
Norwood, and Evans," Owen, Eeport of a geological 



Y 



'<j. 



\y\ 




Figure 160. — Owen's specimens of Palaeotherium? -proutii 



After Leidy, 1862. A, Type of Palaeolheriumf proutii (Owen's specimen), Nat. Mus. 113. One-third natural size. Ttiis was the 
principal specimen referred to by Leidy in proposing the name Titanotherium (1852.1). B, Third left lower molar, another of 
Owen's specimens used by Leidy in describing Titanotherium. Two-thirds natural size. 



EARLY NOTICES BY lEIDY AND OTHERS, 1850-1870 
Palaeotherium? proutii Owen, Norwood, and Evans, 1850 

Cf. Titanotherium proutii Leidy 

Original reference. — Acad. Nat. Sci. Philadelphia 
Proc, vol. 5, p. 66, August, 1850 (Owen, Norwood, 
and Evans, 1850.1). 

Subsequent reference . — " Palaeotherium? proutii Owen, 
Norwood, and Evans," Leidy, Description of the 
remains of extinct Mammalia and Chelonia from 
Nebraska Territory, in Owen, Report of a geological 
survey of Wisconsin, Iowa, and Minnesota, pp. 551- 
552, tab. 9, figs. 3a, 3, 1852 [Owen's specimens, not 
Prout's] (Leidy, 1852.1); "Titanotherium proutii 
Leidy," The ancient fauna of Nebraska, pp. 72-73, 
pi. 16, figs. 1-3, 1853 (Leidy, 1854.1). 

Original description. — Owen, Norwood, and Evans 
state that 

These remarkable remains are thus named in compliment to 
Dr. Prout of St. Louis who first noticed them in the American 
Journal of Science and Arts. The generic characters, however, 
are not yet satisfactorily decided. 



survey of Wisconsin, Iowa, and Minnesota, p. 552 
1852 (Titanotherium) (Leidy, 1852.1V 

Subsequent reference. — " Titanotherium proutii 
Leidy," Leidy, The ancient fauna of Nebraska, pp. 
72, 114, 1853 (Leidy, 1854.1). 

Type species. — Palaeotherium? proutii Owen, Nor- 
wood, and Evans."* (See p. 526.) 

Generic characters. — Not separated by Leidy from 
specific characters. 

Etymology. — 'Yirkv, a Titan; driplov, beast. 

Present determination. — Leidy based the genus Ti- 
tanotherium collectively upon a number of specimens, 
including, first, Prout's original specimen; second, 
"Owen's specimen" (Nat. Mus. 113); and third, cer- 
tain other fragmentary specimens. Prout's specimen 
was already the type of Menodus giganteus Pomel, 
hence by elimination the genus Titanotherium rests 
upon the species Palaeotherium? proutii Owen, Nor- 

i« In his work of 1853 Leidy placed his own name after the speoihc name proutii, 
evidently following the practice of those who placed the name of the author of the 
genus after the specific name. 



206 



TITANOTHERES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



wood, and Evans, the type of which is the second 
specimen described by Leidy, namely, Evans's speci- 
men (Nat. Mus. 113). This specimen is believed by 




FiGUEE 161. — Type (holotype) of Palaeoiherium maximum 
Parts of the outer wall (the ectoloph) ot two upper molars. After Leidy, 1862. Natural size. 

Osborn to be congeneric with the type of Menodus 
giganteus Pomel. 

Proposal of the generic name TitanotJierium. — After 
describing under the name Palaeotherium? 
proutii the specimens made known by Prout 
and by Owen, Norwood, and Evans, Leidy /' 

(1852.1, p. 552) says: 

All the preceding specimens, except probably the 
latter two, I suspect belong to a different genus from 
either Palaeotherium or Anchitherium, and should the 
suspicion prove correct, Tilanotherium would be a 
good name for the animal, as expressive of its very 
great size. 

Palaeotherium maximum Leidy, 1852 

Original reference. — Leidy, in Owen, Report 
of a geological survey of Wisconsin, Iowa, 
and Minnesota, description of tab. 12 B, figs. 
3, 4, 1852 (Leidy, 1852.1). 

Type locality and geologic horizon. — White 
River, "Nebraska" [South Dakota]; Chadron 
formation {Titanotherium zone). 

Type. — Parts of the outer wall or ectoloph 
of two superior molars. Types now lost- 
(See fig. 161.) 

Characters. — Leidy writes: "I am at pres- 
ent very much inclined to consider these as 
belonging to a true species of Palaeotherium, 
which from its very great size might be 
appropriately named Palaeotherium maxi- 
mum." 

Etymology. — maximum, greatest — that is, 
greater than P. magnum. 

Present determination.— These fragments belong to 
a large Oligocene titanothere of wholly uncertain 
reference. 



Rhinoceros americanus Leidy, 1852 

Original reference. — Acad. Nat. Sci. Philadelphia 
Proc, vol. 6, p. 2, 1852 (Leidy, 1852.2). 

Subsequent reference. — Leidy, 
The ancient fauna of Nebraska, 
p. 76, pi. 17, figs. 1-4, 1853 CLeidy, 
1854.1). 

Type locality. — White River, 
"Nebraska" [South Dakota]. 

Type. — Two superior premolars 
belonging upon opposite sides of 
the jaw. Part of a collection pro- 
cured by Mr. Thaddeus A. Cul- 
bertson for the Smithsonian Insti- 
tution. Types not located. (See 
fig. 162.) 

Characters. — The proceedings 
of the Philadelphia Academy con- 
tain the following note: 

Dr. Leidy called the attention of the members to a fossil 
tooth and a fragment of a second, from the collection made by 
Mr. Culbertson in Nebraska Territory, which, he observed, 
belonged to a new species of Rhinoceros, or probably Acero- 





FiGURE 162. — Cotypes of Rhinoceros americanus 
Two upper fourth premolars. After Leidy, 1853. Natural size. 

therium. The former specimen is probably a third premolar, 
the latter a portion of the fourth. A great peculiarity in the 
teeth is the confluence of the inner lobes with each other and 
their separation to the base from the outer lobes. They 



DISCOVERY OF THE TITANOTHERES AND ORIGINAL DESCRIPTIONS 



207 



possess a remarkably strong basal ridge and indicate an animal 
larger than any species of existing Rhinoceros; the greatest 
transverse diameter of the third premolar being 2J^ inches; 
its anteroposterior diameter 1% inches. For the species the 
name Rhinoceros americanus is proposed. 

Etymology. — americanus, in allusion to the then 
novel fact that a supposed rhinoceros had once in- 
habited America. 

Present determination. — It does not seem possible 
to determine positively whether these isolated pre- 
molar teeth belong to Allops or to Menodus; the 
affinity to one or the other of these genera is indicated 
by the pronounced internal and external cingula and 
by the large tetartocone on p*. In view of the doubt 
and the disappearance of the type, it seems best to re- 
gard "Rhinoceros" americanus as indeterminate. 

Eotherium Leidy, 1853 

Cf. Menodus Pome!, this monograph, page 522 
Original reference. — Acad. Nat. Sci. Philadelphia 
Proc, vol. 6, p. 392, 1853 (Leidy, 1853.1). 



Present determination. — The specimens indicated 
were first chosen the types of Rhinoceros americanus. 
(See above.) The very pronounced internal and 
external cingula of the type (fig. 162), however, 
appear to indicate that they belong generically to 
Menodus. The genus Eotherium was subsequently 
treated by Leidy as a synonym of Titanotherium. The 
name Eotherium was subsequently (1875) applied by 
Owen to a genus of sirenians. 

Palaeotherium giganteum Leidy, 1854 

(Indeterminate) 

Original reference. — The ancient fauna of Nebraska: 
Smithsonian Contr. Knowledge, vol. 6, p. 78, pi. 17, 
figs. 11-13, 1853 (Leidy, 1854.1). 

Type locality. — White River, "Nebraska" [South 
Dakota]. 

Types. — Portions of the ectoloph of five molars "in 
the collections of Mr. Culbertson and Dr. Owen." 

Lectotypes (Osborn). — The fragmentary ectoloph 
figured in Plate 17, Figure 11, of Leidy's work. (See 
fig. 163.) 




Figure 163. — Cotypes of Palaeotherium giganteum 
Parts of the ectoloph of upper molars. After Leidy, 1853. Natural size. 



Subsequent reference. — Leidy, The ancient fauna of 
Nebraska, pi. 17, figs. 1-7, 1853 (Leidy, 1854.1). 

Type species (monotypic). — Rhinoceros americanus 
Leidy. (See above.) 

Generic description. — Leidy says: 

Of the huge Titanotherium proutii there are numerous small 
fragments of bones and teeth and also several entire superior 
molars, which have served to remove some of the obscurity 
in regard to the characters of the animal. From the last- 
mentioned specimens it appears that those which have been 
described as probably indicating a new species of Palaeotherium^ 
under the name P. giganteum (Ancient fauna of Nebraska, pi. 
17, figs. 11-13), belong to Titanotherium ■proutii, while several 
superior molars (ib., figs. 1-7), attributed to the latter, belong 
to a new genus associating characters of Rhinoceros and Palaeo- 
therium. For this genus and species, represented by Figures 
1-7, Plate 17, in the Ancient fauna of Nebraska, I propose the 
name of Eotherium americanum. 

Etymology. — rjois, dawn, dripiov, beast; possibly in 
allusion to the relatively early geologic age of the 
animal. 



Characters. — Leidy writes: 

The fragments, of which there are five, are only single ex- 
ternal lobes of the upper molars. These, externally, correspond 
to the description of Cuvier of the teeth of Palaeotherium. A 
conjoined pair of the lobes, forming the outer part of a tooth, 
"present the external face strongly inclined inward in descend- 
ing and divided by three salient ridges into two concavities, 
which are rounded toward the fangs and terminate in a tri- 
angular cusp at the masticating surface, the basal angles of 
which rest upon the termination of the salient ridges." The 
median ridge is a thick obtuse fold outward of the tooth, and 
the anterior and posterior ridges are acute, roughened offsets 
from the basal ridge, descending to the masticating surface. 

The measurements of the more perfect specimens are as 

follows : 

In. hnes 

Length of the longest lobe 2 4 

Length of a second specimen 2 

Breadth of the second specimen at the basal angles of the 

cusp 1 8 

Length of the shortest lobe 1 7 

Breadth of the shortest lobe at the basal angles of the 

cusp 1 3 



208 



TITANOTHERES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



Etymology. — giganteum, gigantic; in obvious allu- 
sion to the great size, which seems to have impressed 
all the early observers of Titanotherium. 

Present determination. — Leidy himself subsequently 
(1854.1, p. 157) transferred this species to T. proutii. 
It is indeterminate. 




Figure 164. — Type (holotype) 



coloradensis. Nasals and 



One-third natural size. 



After Leidy, 1873. A, Top view; B, front view; C, view of left side, 
Leidyotherium Prout, 1860 
(Indeterminate) 

Original rejerence. — Acad. Sci. St. Louis Trans., vol. 
1, pp. 699-700, 1860 (Prout, 1860.1). 

Subsequent rejerence. — Leidy, Extinct Mammalia of 
Dakota and Nebraska, p. 390, 1869 (Leidy, 1869.1). 

Type species. — None designated. 

Type locality. — The specimen was reported to have 
been obtained near Abingdon, in Virginia, but was 
later stated by Leidy (op. cit., p. 390) to be "a fossil 
from the Mauvaises Terres of White Kiver, Dakota." 

Type. — "The fragment of a large molar tooth." 



Generic characters. — Prout writes: 

The lobed or indented border of the enamel would seem to 
show that this animal was nearly allied to Titanotherium, while 
the great width and depth of the groove between the outer and 
what may have been the inner border of the tooth would sepa- 
rate it from this genus. * * * It is distinguished, more- 
over, from these [Lophiodon] by the greater length of the fangs 
and the comparative shortness of the enamel on 
the outer surface of the tooth. * * * j^ must 
have been a phytivorous pachyderm, as large if 
not larger than the Titanotherium. 

Etymology. — Named in honor of Joseph 
Leidy. 

Present determination. — No specific name 
■;. is given. Leidy treated the genus as syn- 
onymous with Titanotherium. It is an 
indeterminate member of the family. 
Megacerops Leidy, 1870 
Cf. Megacerops, this monograph, page 541 
Original reference. — Acad. Nat. Sci. Phila- 
delphia Proc, vol. 22, p. 2, 1870 (Leidy, 
1870.1). 

Subsequent reference. — Leidy, Extinct ver- 
tebrate fauna of the Western Territories, 
p. 239, pi. 1, figs. 2, 3; pi. 2, fig. 2, 1873 
(Leidy, 1873.1). 

Type species. — Megacerops coloradensis 
Leidy. 

Generic characters. — In the original refer- 
ence a detailed description of the type 
specimen of Megacerops coloradensis is given, 
comparisons being made with the anterior 
horn cores and nasals. of the Siwalik Sivathe- 
rium, with which it was thought possibly to 
be allied. Leidy concludes as follows: 

It is probable that the fossil may pertain to the 
same animal as the remains from the Mauvaises 
Terres of Nebraska, described under the name of 
Titanotherium, but in the state of extreme uncer- 
tainty as to its collocation, it may with equal 
probability be referred to other genera, perhaps 
to Megalomeryx, or it may have been an American 
species of the Sivatherium. Under the circum- 
stances it may be referred to a new genus, with the 
name of Megacerops coloradensis. 

Etymology. — jxkya^, great; Kfpas, horn; &4', 
face. 

Present determination. — Leidy's carefully 
executed figures of the type, in the opinion 
of the present writer (Osborn), reveal the 
generic relationship of this animal with that later de- 
scribed by Cope (1873.2, p. 4) as Megaceratops acer. 
Megacerops coloradensis Leidy, 1870 
Cf. Megacerops coloradensis, this monograph, page 544 
Original reference. — Acad. Nat. Sci. Philadelphia 
Proc, vol. 22, p. 2, 1870 (Leidy, 1870.1). 

Subsequent reference. — Leidy, Extinct vertebrate 
fauna of the Western Territories, pp. 239-242, pi. 1, 
figs. 2, 3; pi. 2, fig. 2, 1873 (Leidy, 1873.1). 

Type locality and geologic horizon. — Colorado ; Chad- 
ron formation {Titanotherium zone), level not ascer- 
tained. 



DISCOVERY OP THE TITANOTHERES AND ORIGINAL DESCRIPTIONS 



209 



Type. — Fractured horns and nasals. The present 
location of this type has not been determined. It is 
not in the collection of the Philadelphia Academy, 
nor is any record of its loan to be found. (See fig. 
164.) 

Characters of type. — Leidy's description is too long 
to quote here. The specimen may be described briefly 
as follows: Seen from above the nasals are of moderate 
length and taper toward the extremities; from the 
side and front they appear decidedly long and thin 
and are strongly decurved at the tip, at which point 
there is a median notch. The horns project forward 
and outward and pass from an elongate oval section 
at the base to rounded, transversely oval tips. The 
greatest diameter of the horns at the base is antero- 
posterior, with flattened outer and convex inner 
faces. The following approximate measurements are 
taken from Leidy's descriptions and figures: 

Millimeters 

Free width of nasals 108 

Free length 104 

Outside measurement of horns 140 

Etymology. — coloradensis , in allusion to the type 
locality. 

Present determination. — The type of Megacerops 
coloradensis, consisting of the osseous horns and nasals, 
is apparently distinct specifically from Cope's M. acer, 
M. hucco. 

SPECIES DESCRIBED BY MARSH AND COPE IN 1873-1876 
Brontotherium Marsh, 1873 

Cf. Brontotherium, this monograph, pages 555-557 

Original reference. — Am. Jour. Sci., 3d ser., vol. 5, 
p. 486, 1873 (Marsh, 1873.1). 

Type species. — Brontotherium gigas Marsh. (See 
below.) 

Generic characters. — Marsh writes: 

An examination of the remains, in the Yale Museum, of the 
huge mammals allied to Titanotherium has led to the discovery 
that two different animals have hitherto been referred to the 
species known as T. prouti. These animals are generically dis- 
tinct and probably are from separate geological horizons. 
The one here described differs from Titanotherium in its denti- 
tion, having but three lower premolars, the series being as fol- 
lows: Incisors 2, canine 1, premolars 3, molars 3. The animal 
was, moreover, a true perissodactyl, with limb bones resembling 
those of Rhinoceros. The genus is related to Titanotherium, and 
the two appear to form a distinct family, which may be called 
Brontotheridae. The present species is based on portions of 
three individuals, one of which has the lower jaws and en- 
tire molar series complete. They indicate an animal fully 
equal to T. prouti in size, and but little inferior in bulk to the 
elephant. The lower molars resemble those in the type speci- 
men of T. prouti, but the jaw below them is not so deep, and its 
lower margin is more nearly straight, descending but very 
slightly toward the angle. The front part of the lower jaws is 
somewhat suilline in form. The incisors are quite small, and 
the two next to the symphysis are separated from each other. 
There is a short diastema between the canine and the first 
premolar. [This is followed by remarks on the skeleton based 
on the "other specimens."] 

Etymology: fipovTi], thunder; drjpiov, beast. 



Present determination. — This was the most impor- 
tant contribution to the knowledge of the titanotheres 
made up to that time. The characters of the lower 
jaw and of the skeleton are correctly described, and the 
family is referred to the Perissodactyla. Subsequent 
research has shown that the genus Brontotherium is 
distinct from Menodus and Megacerops; "Brontothe- 
rium ingens," as used in later publications by Marsh, 
referred to the skull, the type of "B. ingens," and not 
to the jaw, the type of Brontotherium gigas. 

Brontotherium gigas Marsh, 1873 
Cf. Brontotherium gigas, this monograph, page 567 

Original reference. — Am. Jour. Sci., 3d ser., vol. 5, 
p. 486, 1873 (Marsh, 1873.1). 

Subsequent reference. — Principal characters of the 
Brontotheriidae : Am. Jour. Sci., 3d ser., vol. 11, pi. 
12, figs. 1-3, 1876 (Marsh, 1876.1). 

Type locality and geologic horizon. — Colorado; exact 
locality and level not published. Sargent, Griswold, 
and Marsh, collectors. 

Type. — "The present species is based on portions 
of three individuals, one of which has the lower jaws 
and entire molar series complete [lectotype]." Yale 
Mus. 12009. (See fig. 165.) 

Characters of type. — The specific characters were 
not separated by Marsh from the generic charcters. 
Measurements of the lower jaw were given, some of 
which (now verified) are as follows: 

Millimeters 

Length of lower jaw, from condyle to front of symphysis 634 

Depthof lower jaw, from top of coronoid process to angle 367 

Length of last lower molar 117 

Length of last lower premolar (Marsh gives this as 51) [49] 

Etymology. — yiyas, giant. 

Present determination. — This valid species is fully 
discussed in Chapter VI of this monograph (p. 567). 
Symborodon Cope, 1873 
Cf. Menodus, this monograph, page 525 

Original reference. — Pal. Bull. No. 15, p. 2, "issued 
August 20, 1873" (Cope, 1873.2). 

Subsequent reference. — Cope, Eeport on the verte- 
brate paleontology of Colorado, pi. 2, fig. 1; pis. 3, 4, 
1874 (Cope, 1874.2). 

Type species. — Symborodon tonus Cope. (See 
below.) 

Generic characters. — Cope writes: 

Dentition: I.? 0; C. 1; Pm. 3; M. 3; the canines slightly 
separated from each other, but not from the first premolar. 
Crowns of the premolars with L-shaped crescents as in Rhi- 
noceros; of the molars with completed crescents; the last molar 
with third posterior crescent. Symphysis mandibuli coossi- 
fied, crowns of canines not projecting, conic. * * * Xhe 
genus differs from Titanotherium and Brontotherium in the 
absence of incisors and from the former in the presence of but 
three premolars. If there had been a deciduous incisor on each 
side I was unable to detect any trace of it. 

Etymology. — (tvv, together; |Sop6s, devouring; oSous, 
tooth; in reference to the approximation of the op- 
posite canines toward the middle line. 



210 



TITANOTHERES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



Present determination. — Subsequent research has 
proved that this genus is a synonym of Menodus. It 
is fully described on page 522. 

Symborodon torvus Cope, 1875 

Cf. Menodus torvus, this monograph, page 525, Figure 166 
Original reference. — Pal. Bull. No. 15, p. 2, "issued 
August 20, 1873" (Cope, 1873.2). 




FiGUEB 165. — Type (lectotype) of Bronluihenum gigas 
Lower jaw, with nearly complete dentition. Yale Mus. 12009. After Marsh, 1876. One-sixth natural size 

Subsequent reference. — Report on the vertebrate 
paleontology of Colorado, p. 486, 1874. The jaw 
figured in Plate 2, Figure 1, is not the type of torvus 
(Cope, 1874.2). 

Type locality and geologic Tiorizon. — Horsetail Creek, 
Logan County, northeastern Colorado; Chadron for- 
mation {Titanotlierium zone), level not ascertained. 

Ootypes. — Cope writes: "The present genus is 
established on mandibular rami only, which can not be 



certainly associated with crania." These rami (Cope 
collection. Am. Mus. 6365, 6345) are accordingly 
CO types. In his "Report on the vertebrate paleon- 
tology of Colorado" Cope says, "I append a de- 
scription of the mandible, on which the species 
Symborodon torvus was established." Careful com- 
parison of Cope's original and subsequent descriptions 
and measurements shows that the species 
was established largely upon the lower 
jaw (Am. Mus. 6365, fig. 166) which 
is accordingly regarded as the lecto- 
type. 

Etymology. — torvus, wild, grim. 
Present determination. — The species is 
now regarded by Osborn as belonging in 
the genus Menodus. In size the type 
is intermediate between M. Jieloceras and 
M. trigonoceras. 

Miobasileus Cope, 1873 
(Indeterminate) 
Original reference. — Pal. Bull. No. 15, 
p. 3, "issued August 20, 1873" (Cope, 
1873.2). 

Subsequent references. — On some ex- 
tinct types of horned perissodactyls, 
p. 108, 1874 (Cope, 1874.1); Synopsis 
of new Vertebrata from the Tertiary of 
Colorado, p. 14, 1873 (Cope, 1873.3); 
Report on the vertebrate paleontology of 
Colorado, p. 490, 1874 (Cope, 1874.2); 
U. S. Geol. Survey Terr. Rept. for 1873, 
p. 490, 1874. 

Type species. — Miobasileus ophryas 
Cope. (See below.) 

Generic characters. — Not separated 
by Cope from specific characters. 
(Seep. 201.) 

Established on a cranium with nearly com- 
plete dentition but without mandibular ramus. 
Head elongate, concave in profile from the 
interorbital region to the supraoccipital crest. 
This is transverse and concave, the posterior 
borders of the temporal fossae extending behind 
it. These fossae leave a narrow flat vertex 
between them. Zygomatic arch stout and rather 
deep; a strong postglenoid process. Nasal bones 
very massive, their free portion elongate, 
hornless. A massive horn core rising from 
above each orbit, no superciliary angle or ridge. Orbit not 
inclosed behind. Of molar teeth only Pm. 2, M. 3, pre- 
served, the M. with two, the Pm. with one inner cone, and 
two outer continuous crescents. The latter send inward to 
one side of the cones a transverse ridge. Incisors and canines 
unknown. 

Char, specif. — Front concave transverse just behind between 
the horns. Latter massive and little compressed. Nasal 
bones convex longitudinally and transversely, slightly rugose. 
Transverse ridges of teeth with transverse expansions at their 
inner extremity, being thus T-shaped. 



DISCOVERY OF THE TITANOTHERES AND ORIGINAL DESCRIPTIONS 



211 



Millimeters 

Length from apex of nasals to occipital condyles (axial) 684 

Length from occipital cond3'les to femoris of palate 376 

Length from occipital condyles to end of palatine lamina 

pteryzoidea 270 

Length of four last molars 242 

Length of three last molars 195 

Length of last molar 68 

Width of palate at nareal notch 116 

Etymology. — Mio, Miocene; /SacriXeiis, king — that is, 
monarch of the Miocene. 

Present determination. — -The genus is indeterminate. 
(See M. ophryas.) 

Miobasileus ophryas Cope, 1873 
(Indeterminate) 

Original reference. — Pal. Bull. No. 15, p. 3, "issued 
August 20, 1873" (Cope, 1873.2). 

Subsequent references. — Cope, On some extinct types 
of horned perissodactyls, p. 108, 1874 (Cope, 
1874.1); Synopsis of new Vertebrata from the 
Tertiary of Colorado, p. 14, 1873 (Cope, 
1873.3) ; Report on the vertebrate paleontology 
of Colorado, p. 490, 1874 (Cope, 1874.2). 

Type locality and geologic Jiorizon. — Cedar 
Creek, Logan County, Colo.; Chadron forma- 
tion {Titanotherium zone), level not ascertained. 

Type. — A cranium with incomplete dentition, 
without mandibular ramus. (In a later descrip- 
tion Cope (1874.2, p. 490) remarks, "of molar 
teeth only pm 3-4, m 1, 2, 3, preserved. ") This 
type was left in the field and is now lost. 

Generic and specific cJiaracters (summarized 
from Cope). — Supraoccipital crest concave. 
Zygomatic arch stout and relatively deep. 
Nasal bones very massive, elongate, convex 
longitudinally and transversely; a massive horn core, 
little compressed, rising above each orbit. In a later 
communication Cope (1874.2, p. 491) gives the length 
from apex of nasals to occipital condyles as 664 
millimeters and observes: 

The dental characters of this species ally it to the S. trigo- 
noceras, but the form as well as the position of the horns is 
quite different. Instead of being triangular, a section of the 
base of these is elliptic. Extremity conical. 

Millimeters 

Length from apex of nasals to occipital condyles 684 

Length of three last molars 195 

Length of last molar _. 68 

Etymology. — 64>pvs, eyebrow*; possibly in allusion 
to the form of the orbit. 

Present determination. — Owing to the loss of the 
type and the uncertain character of the description, 
this genus and species is indeterminate. 

Megaceratops Cope, 1873 

Original reference. — Pal. Bull. No. 15, p. 4, "issued 
August 20, 1873" (Cope, 1873.2). 



Present determination. — This name Megaceratops was 
not proposed in order to denominate a new genus but 
was merely an emendation on etymologic grounds of 
Leidy's term Megacerops, of which it must be regarded 
as a synonym. 

Megaceratops acer Cope, 1873 
Cf. Megacerops acer, this monograph, page 545 

Original reference. — Pal. Bull. No. 15, p. 4, "issued 
August 20, 1873" (Cope, 1873.2). 

Subsequent reference. — Cope, Report on the verte- 
brate paleontology of Colorado, p. 488, pi. 7; pi. 8, 
fig. 3, 1874 (Cope, 1874.2). 

Type locality and geologic Jiorizon. — Horsetail Creek, 
northeastern Colorado; Chadron formation {Titano- 
therium zone), level not ascertained. 

Type. — "A single cranium without under jaw." 
Am. Mus. 6348. (See figs. 167, 170.) 




Figure 166. — Type (lectotype) jaw of Symhorodon torvus 
One-sixth natural size. 

Specific cJiaracters. — Cope writes: 

Top of head flat, forming a narrow plane between the temporal 
fossae; latter produced backward. Orbit not inclosed behind, 
an overhanging superciliary ridge. Nasal exceedingly short 
and massive, each supporting a large acute horn core, which is 
connected with its fellow by a ridge at the base and diverges 
widely from it with an outward and forward curve to the 
acutely compressed apex. Each horn core about 1 foot long. 
The top of the head is plane between the orbits, and little 
concave fore and aft. The zygoma is very deep, and the post- 
glenoid process well developed. End of nasal bones short and 
thick but flat. 

Measurements 

Millimeters 

Length of cranium (35 inches) 895 

Length from posterior rim temporal fossa to middle of super- 
ciliary ridge 345 

Width front between eyebrows 210 

Length horn core on inner side (10 inches) 254 

The elemental origin of the horn cores is probably different 
in this genus from that which exists in Miobasileus. 

Etymology. — acer, fierce, in allusion to the somewhat 
ferocious appearance. 

Present determination. — This valid species, which 
pertains to the genus Megacerops, is fully described on 
page 545. 



212 



TITANOTHERES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



Megaceratops heloceras Cope, 1873 

Cf. Menodus heloceras, this monograph, pages 524, 681 

Original reference. — Pal. Bull. No. 15, p. 4, "issued 
August 20, 1873" (Cope, 1873.2). 

Subsequent reference. — Cope, Report on the verte- 
brate paleontology of Colorado, pp. 487-488, 1874 
(Cope, 1874.2). 

Type locality and geologic horizon. — Horsetail Creek, 
northeastern Colorado; Chadron formation {Titano- 
therium zone), level not ascertained. 

Type. — "A cranium * * * with nearly com- 
plete maxillary dentition," anterior teeth and part of 
frontals wanting. Am. Mus. 6360. (See fig. 168.) 

Specific characters. — Cope writes: 

There is a prominent horizontal superciliary ridge without 
horns, and two short obtuse horn cores on the muzzle. These 




Figure 167. — Type (holotype) skuU of Megaceratops acer 

Am. Mus. 6348. After Cope, 1874. One-sixth natural size. 

diverge outward, the outer sides being flattened and the sum- 
mits contracted and truncate. They are mere rudiments of 
the horns seen in M. aceronsor [sic], M. coloradoensis. The 
molar teeth do not exhibit the T-shaped cross ridges seen in 
Miobasileus, and the two outer crescents are continuous with 
each other. 

Measurements 

Millimeters 
Length from posterior rim of temporal fossa to middle of 

osseous eyebrow 472 

Least width of parietal plane 104 

Superciliary width 260 

Elevation of horn core 50 

Etymology. — ^Xos, wart; xepas, horn; in allusion to 
the wartlike appearance of the horn. 

Present determination. — The species 
referable to the genus Menodus. (See p. 

Symborodon bucco Cope, 1873 
Cf. Megacerops bucco, this monograph, page 544 
Original reference. — Synopsis of new Vertebrata 
from the Tertiary of Colorado, p. 11, 1873 (Cope, 
1873.3). 



is valid 
524.) 



but 



Subsequent reference. — Cope, Report on the verte- 
brate paleontology of Colorado, pp. 484, 485, pis. 
2-4. 1874 (Cope, 1874.2). 

Type locality and geologic horizon. — Horsetail Creek, 
northeastern Colorado; Chadron formation {Titano- 
therium zone), level not ascertained. 

Cotypes. — In the original description Cope says the 
species is represented "by an imperfect cranium; by 
a cranium almost perfect, including very probably 
both mandibular rami, with entire dentition; a frag- 
mentary skeleton, including parts of cranium, teeth, 
and vertebrae; and by a series of cervical and dorsal 
vertebrae." Which of these cotype individuals thus 
mentioned shall we select as the lectotype? If we 
should take the first specimen mentioned, namely, the 
imperfect cranium (known to be Am. Mus. 6346), we 
find that since it consists of only the posterior portion 
it lacks most of the characters given in the specific 
description, except the single one of possessing ex- 
panded zygomata (hence the name hucco). On the 
other hand the "cranium almost perfect" (Am. Mus. 
6345) also has expanded zygomata and was evidently 
the chief specimen, since it furnished most of the 
specific characters and measurements given in the origi- 
nal description; moreover, in Cope's 
fuller report (1874.2) it was figured 
in Plates 2, 3, 4, imder the name 
Symhorodon hucco, and in the key to 
the species (p. 484), in which S. hucco 
is contrasted with S. altirostris, the 
diagnostic characters (referring to 
the horns, premolars, nasals, de- 
pressed cranium) are evidently from 
the "cranium almost perfect" (No. 
6345) rather than from the "im- 
perfect cranium." 

Lectotype. — From these clear indi- 
cations of the author's intention the 
skull (Am. Mus. 6345) may therefore be regarded 
as the lectotype. (See figs. 169, 170.) 

Specific characters. — Cope mentions especially the 
enormous buccal expansion of the zygomata, the char- 
acters of the horns, nasals, skull top, orbits, etc. 
Specific characters are fully given on page 544. 
Etymology. — hucco, one having extended cheeks. 
Present determination. — This species is provisionally 
regarded as a valid one. 

Symborodon altirostris Cope, 1873 

Cf. Megacerops acer, this monograph, page 545 

Original reference. — Synopsis of new Vertebrata 
from the Tertiary of Colorado, p. 12, 1873 (Cope, 
1873.3). 

Subsequent references. — Cope, Report on the verte- 
brate paleontology of Colorado, p. 486, pis. 5, 6, 8, 
fig. 1, 1874 (Cope, 1874.2); The Perissodactyla, pi. 33, 
fig. a, opposite p. 1062, 1887 (Cope, 1887.1). 



DISCOVERY OF THE TITANOTHERES AND ORIGINAL DESCRIPTIONS 



213 



Type locality. — Cedar Creek, Logan County, Colo. 

Type. — A cranium with premolar-molar teeth, zygo- 
matic arches fractured (Am. Mus. 6350). (See figs. 
170, 171.) 

Characters of type (summarized from Cope). — Nasal 
bones very short, broad, obtuse, massive, and "stand- 
ing on a plane above that of the front." Orbit very 
far forward. Horns straight, with approximated bases 



Present determination. — As shown (p. 545), there are 
reasons for regarding the type of S. altirostris as repre- 
senting a female skull of Megacerops acer. 

Symborodon trigonoceras Cope, 1873 

Cf. Menodus trigonoceras, this monograph, page 528 

Original reference. — Synopsis of new Vertebrata from 
the Tertiary of Colorado, p. 13, 1873 (Cope, 1873.3). 






1^ 


^^5 


^^^^^N^^^H 


H 




Hk ';^^v^H 






1 



and moderately divergent, subcylindrical at base and 
compressed inward and forward at the narrow apex. 
"The first premolar and two incisors are very insig- 
nificant; canines with short stout crowns." The pre- 
molars with two smooth cones. Many other charac- 
ters are given. 

Etymology. — altus, high; rostris, beak, snout; in 
allusion to the high position of the nasals. 



FiGUEE 168. — Type (holotype) skull of Megaceratops heloceras 
Am. Mus. 6360. After Cope. One-flfth natural size. 

Subsequent references. — Cope, Eeport on vertebrate 
paleontology of Colorado, 1874, p. 488, 1874 (Cope, 
1874.2); The Perissodactyla, p. 1065, figs. 29, 30, 1887 
(Cope, 1887.1). 

Type locality and geologic Tiorizon. — ^Horsetail Creek, 
northeastern Colorado; Chadron formation {Titano- 
therium zone), level not ascertained. 



214 



TITANOTHERES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



Cotypes. — Skull (Am. Mus. 6355) lacking all the 
teeth except m^; Am. Mus. 6356, anterior-inferior por- 
tion of skull, including horns, nasals, right zygoma, 
and teeth. Of these two cotypes we may regard No. 
6355 as the lectotype. (See fig. 172.) 

Specific characters (summarized from Cope). — A 
strong basal cingulum, on the inner side of the pre- 
molars, which is continued in a less prominent form 




Present determination. — This is a valid 
described on page 528, referable to Menodus. 



species, 



Brontotherium ingens Marsh, 1873 

Cf. Menodus giganteus, this monograph, page 530 

Original reference. — Am. Jour. Sci., 3d ser., vol. 7, 
pp. 85, 86, pis. 1, 2, January, 1874; "published Dec. 
30, 1873" (Marsh, 1874.1). 



■"-^fS^ ^ 








Figure 169. — Type (lectotype) skull of Symborodon bucco 
Am. Mus. 6345. After Cope, 1874. One-ninth natural size. The mandible in the upper figure probably does not belong with the skull. 



between the bases of the cones of the molars. Bases 
of cones of premolars strongly plicate. Horns tri- 
quetrous, dii-ected outward and upward. Squamosals 
not expanded, nasals elongate transversely plane. 

Etymology. — rpis, three; yuvla, angle; Kepas, horn; 
in allusion to the three-sided section of the horn. 



Subsequent reference. — Marsh, The principal char- 
acters of the Brontotheridae, p. 335, text figs. 1, 2, 
pis. 10, 11, 1876 (Marsh, 1876.1). 

Type locality and geologic horizon. — Colorado; Chad- 
ron formation {Titanotherium zone); exact locality 
and level not recorded. 



DISCOVERY OF THE TITANOTHERES AND ORIGINAL DESCRIPTIONS 



215 



Type. — A complete skull; premaxillaries with in- 
cisors and canines wanting; nasals and horns partly 
restored. Yale Mus. 2010. (See fig. 173.) 

Characters. — Marsh says: 



to the median line. The upper part of the horn cores is rugose, 
and the base contains large air cavities. The free extremities 
of the nasals are coossified and much elongated. They are 
rounded in front, slightly decurved, and the surface at the end 
is rugose. [Many other characters are listed.] 






Figure 170. — Type skulls of Symhorodon altirostris (1), S. bucco (2), and 
Megacerops acer (3) 
Front views. After Cope, 1874. One-sixth natural size. 



The most striking peculiarity of this cranium is the pair of 
huge horn cores on the nasals. They are about 8 inches in 
length and extend upward and outward. They are triangular 
at the base, with the broadest face external. The two inner 
faces of each core are separated by a ridge, which is continued | 



Millimeters 
Length of skull from occipital condyles to end of nasals (36 

inches) 915 

Distance on median line from occipital crest to end of 

nasals 762 



216 



TITANOTHERES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



Millimeters 

Expanse of zygomatic arches 558 

Least distance across vertex 157 

Space occupied by four upper premolars 162 

Space occupied by three upper true molars 266 

Space occupied by molar-premolar series 428 

Etymology. — ingens, vast. 

Present determination. — The species is a synonym of 
Menodus giganteus Pomel. 

Symborodon hypoceras Cope, 1874 
Cf. Brontotherium hypoceras, this monograph, page 562 

Original reference. — U. S. Geol. and Geog. Survey 
Terr. Ann. Rept. for 1873 (Hayden), p. 491 [no 
figure], 1874 (Cope, 1874. 2). 



cores of very different shape described below. (See 
fig. 174.) 

Specific characters. — Infraorbital foramen "fiat with 
a wide external face, instead of being a cylindric col- 
umn as in S. acer, altirostris, bucco, and ophryas." 
One of the horn cores "consists of the extremital 
part. * * * j^g section is a compressed oval 
narrowed in front; its profile with parallel outlines and 
a little recurved and not very rugose. Its size as 
compared with the rest of the skull is the smallest in 
the genus, and not more than half the proportions of 
the S. altirostris." Another fragment Cope deter- 
mined as a portion of the frontal bearing a "large 
osseous tuberosity, which consists of a mass of bone 




FiGTJRE 171. — Type (holotype) skull of Symborodon altirostris 
Am. Mus. 6350. After Cope, 1874. One-sixth natural size. 



Type locality. — ?Cedar Creek, Logan County, Colo. 
Type. — Cope writes: 

This species reposes on a fragmentary cranium only, which 
embraces nasal, maxillary, frontal, malar bones, etc., both 
zygomata, premolar, and parts of molar teeth. These frag- 
ments were taken out of the matrix by the writer and were 
found in juxtaposition. They represent parts of the same 
skuU and, as no other was found in the same bank, are prob- 
ably without admixture. 

The only remains representing this type which are 
now preserved in the American Museum of Natural 
History (Am. Mus. 6361) include two portions of the 
malar bones, a fragment of the orbit and infraorbital 
canal, a fragment of the alveolar region, and two horn 



coossified with the upper surface as in the horn of 
the girafi'e." Cope concluded that "it is probable 
that this species possessed two pairs of osseous proc- 
esses or cores on each side, the one on the nasal, the 
other on the frontal bone." The name "hypoceras" 
doubtless referred to the supposed presence of the 
second horn core (the rounded tuberosity) behind and 
below the oval-sectioned horn on the nasals. Cope 
gives 14 measurements, including the following: 

Millimeters 

Length from front of orbit to glenoid fossa (axial) 365 

Depth of malar below orbit 20 

Length of molars and last three premolars 293 

Length of last three premolars 110 

Diameter of horn core, transverse 38 



DISCOVERY OF THE TITANOTHERES AND ORIGINAL DESCRIPTIONS 



217 



Fixation of ledotype. — Cope's conclusion that the 
above-mentioned fragments "are probably without 
admixture" appears open to doubt. The "frontal 
tuberosity" referred to is shown by comparison with 
well-preserved material to be the horn core of the left 
side of an immature individual resembling Allops 
marshi, a reference favored by Cope's observation of 
the wide bridge over the infraorbital foramen, which 
contrasts with the narrow columnar bridge in Bron- 
totheriwn and Symhorodon. The oval-sectioned horn 
core which Cope supposed to be borne on the nasals is 
a right horn core of very different shape, agreeing closely 
with that in skull No. 4702, U. S. National Museum, 
which Osborn selected (1902.208, p. 106) as the neo- 
type of this species. The oval-sectioned horn core 
may, therefore, be regarded as the ledotype. 

Etymology. — invb, below; /cepas, horn; in allusion to 
the supposed presence of a low horn swelling on the 
frontal, behind the one on the nasal. 

Present determination. — As thus interpreted, hypo- 
ceras is a valid species of the genus BrontotJierium. 

Anisacodon Marsh, 1875 
Cf. Menodus, this monograph, page 522 

Original reference. — Am. Jour. Sci., 3d ser., vol. 9, 
p. 246, March, 1875 (Marsh, 1875.1). 

Type species. — Anisacodon montanus Marsh (see 
below). 

Generic characters (Marsh). — "Dentition: Incisors-?; 
canines y; premolars f ; molars f. No superior dias- 
tema. Strong inner basal ridge on upper pre- 
molars. Last upper molar with two inner cones. No 
postorbital process." 

Etymology. — iivtcT-os, unequal; (xktj, point; 66ous, 
tooth. Possibly in allusion to the unequal develop- 
ment of the two inner cones on the third upper molar. 

Present determination. — In view of the strong 
similarities to Menodus in the vestigial condition of 
the incisors, in the strong internal cingulum in the 
premolars, in the shape of the nasals, and in the second 
internal cone of the third molar, this genus is now 
regarded as a synonym of Menodus. 

Anisacodon montanus Marsh, 1875 
Cf. Menodus giganteus?, this monograph, page 537 

Original reference. — ^Am. Jour. Sci., 3d ser., vol. 9, 
p. 246, March, 1875 (Marsh, 1875.1). 

Type locality and geologic horizon. — "Northern 
Nebraska" (Big Badlands, White River, S. Dak.); 
Chadron formation {Titanotherium zone); exact local- 
ity and level not recorded. 

Type. — A fragmentary skull including the maxil- 
laries and fragmentary molar teeth. Yale Mus. 
10022. (See fig. 175.) 

Specific characters. ^Mursh. writes: 

This species is especially distinguished by the emargination 
of the extremity of the nasals, the short premaxillaries, and 
101959— 29— VOL 1 17 



the rectangular form of the last upper molar. The inner pos- 
terior cone of this molar is smaller than the one in front, and 
quite distinct from the posterior basal ridge. 

Measurements [selected from Marsh] 

Millimeters 

Width of nasals above end of premaxillaries 95 

Anteroposterior diameter of last upper premolar 43 

Anteroposterior diameter of penultimate upper molar 77 

Anteroposterior diameter of last upper molar 84 

Etymology. — montanus, dwelling in the mountains. 
Exact allusion uncertain, unless the badland topogra- 
phy of South Dakota is thought of as mountainous. 



\ 




Figure 172. — Type (holotype) skviH of Sy7nborodontrigonoceras 
Am. Mus. 6355. One-ninth natural size. 

Present determination. — In the form of its premolars 
and third molar as well as in its vestigial incisors this 
animal resembles Menodus giganteus; the emarginate 
nasals with processes on either side of the median 
notch also recall female Menodus skulls. Anisacodon 
(Diconodon) montanus is probably referable to Meno- 
dus cf. M. giganteus. 

"Diconodon non Anisacodon" Marsh, 1876 

Cf. Menodus giganteus, this monograph, page 530 

Original reference. — Am. Jour. Sci., 3d ser., vol. 11, 
p. 339, April, 1876 (Marsh, 1876.1). In this paper 
Marsh gives diagnosis of four genera of Brontothe- 
ridae. No. 4 is called "Diconodon Marsh (Anisaco- 



218 



TITANOTHERES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 





Figure 173. — Type (holotype) skull of Brontotherium {=Menodus) ingens 
Yale Mus. 2010. After Marsh, 1S74. About one-sixth natural size. 






Figure 174. — Type (lectotype) of Symborodon {= Brontotherium) hypoceras 
Am. Mus. 6361. One-half natural size. Fragment of right horn core: A', front view; A^, rear view; A', top view. 



DISCOVERY OF THE TITANOTHERES AND ORIGINAL DESCRIPTIONS 



219 



Ann. 



don). * * * Type D. montanus Marsh." The 
term Anisacodon had been preoccupied by Anisacodon 
Marsh, 1872, a genus of insectivores. 

Etymology. — Sis, double; kuvos, cone; 65ovs, tooth. 

Present determination. — See remarks under Anisaco- 
don, above. 

FIRST NOTICE OF CANADIAN TITANOTHERES BY COPE, 1886 

Menodus angustigenis Cope, 1886 

Cf. Megacerops angustigenis, this monograph, page 482, 
fig. 176, Ci 

Original reference. — Canada Geol. Survey 
Kept., new ser., vol. 1, p. 81, 1886 (Cope, 
1886.1). 

Subseguent references. — " Haplacodon 
angustigenis," The Vertebrata of the 
Swift Current River, II, p. 153, 1889 
(Cope, 1889.1); On Vertebrata from the 
Tertiary and Cretaceous rocks of the 
Northwest Territory, I, p. 13, pi. 5, figs. 
1, 2; pi. 6, figs. 2, 2a; pi. 7, figs. 1, la, la 
[bis], 1891 (Cop, 1891.2). 

Type locality and geologic liorizon. — 
Swift Current River, Assiniboia, Canada; 
Cypress Hills beds, level not determined. 
McConnell and Weston, collectors. 

Cope's cotypes. — Cope writes: 

This large mammal is represented by numer- 
ous specimens. I select for present description 
two maxillary bones from the same skull [fig. 
176, A] [Cope, 1891.2, pi. 5, figs. 1, 2], each of 
which contains the first [fourth] premolar and 
the true molars; and two lower jaws from second 
and third individuals [fig. 176, B]. One of these 
[op. cit., pi. 7, figs. 1, la, la [bis], our fig. 176 C, 
now regarded as the lectotype] consists of little 
more than the symphysis. The other [op. cit., 
pi. 5, fig. 2; pi. 6, figs. 2, 2a] includes part of 
the symphysis and the left ramus, which con- 
tains all the molar teeth except the first and last. 
[See fig. 176.] 

Lectotype. — Of these semingly coequal 
types or cotypes, which is to be regarded 
as the lectotype? The one mentioned 
first is "the two maxillary bones from 
the same skull," but the mandibular 
symphysis (op. cit., pi. 7, figs. 1, la, la 
[bis]), from which the species evidently 
takes its name (meaning narrow chin) , is 
certainly to be selected as the lectotype. 

Specific cTiaracters. — Cope's description 
is too long to quote here. He compared 
Menodus angustigenis with " Symhorodon trigonoceras" 
and other species and gave numerous measurements. 
Among the chief characters noted are the contracted 
shape of the mandibular symphysis and the square 
outline of the molars. 

Etymology. — angustus, narrow; gena, chin. 

Present determination. — As defined from the lecto- 
type the species is provisionally referred to Megacerops, 
although its generic reference is uncertain. 



The maxilla with the dentition belongs to a very 
different animal. It is apparently referable to Allops 
sp. (See below.) The lower jaw appears to be 
referable to Menodus cf . M. proutii. 

SPECIES DESCRIBED BY SCOTT AUD OSBORN IN 1887 

Menodus tichoceras Scott and Osborn, 1887 

Cf. Brontotherium tichoceras, this monograph, page 565 

Original reference. — Mus. Comp. Zoology BuL., vol. 
13, No. 5, p. 159, text figs. 3, 2; 5, 2; 6, 2, 1887 (Scott 
and Osborn, 1887.1). 




Figure 175. — Type (holotype) of Anisacodon montanus 

Yale Mus. 10022. A, Third right upper molar; B, fourth upper premolar and first and second molars; C, 
alveoli of upper canines and incisors; Di, nasals, top view; Da, nasals, front view. All one-half 
natural size. 



Type locality and geologic Tvorizon. — Big Badlands, 
South Dakota; exact locality and horizon not recorded. 
S. Garman, collector. 

Type.^&coit and Osborn describe the type as "a 
large skull with the dentition complete, lacking the 
upper part of the horns and the crest of the occiput." 
Now in the Museum of Comparative Zoology at Cam- 
bridge, Mass. (See fig. 177.) 



220 



TITANOTHERES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



Characters (abbreviated from Scott and Osborn). 
Dentition: I 2, C 1, P 4, M 3. The skull is described 
as 29 inches [736 mm.] in length; with a narrow and 
elevated anterior portion; nasals of medium length, 
with short, obliquely placed horns, zygomatic arch 
very massive, presenting a bulge in the posterior half 
which is much less prominent than in S. iucco. 

Etymology. — relxos, wall; Kipas, horn; possibly in allu- 
sion to the high connecting crest. 



Type locality and geologic horizon. — South Dakota; 
Chadron formation {TitanotJierium zone); exact local- 
ity and level not recorded. 

Type. — A skull incomplete in the supraoccipital 
region; zygomatic arch fragmentary; maxillary, pala- 
tine, and basioccipital regions much distorted. Now 
in the Museum of Comparative Zoology, Harvard 
University. (See fig. 177.) 

Characters. — Scott and Osborn write: 




Figure 176. — Cope's cotypes of Menodus angustigenis 

A, Right maxilla (subsequently made the type of Haplacodon angustigenis), three-sixteenths natural size; B, left halt of a lower jaw (now referred to Menodus 
sp.) , three-sixteenths natural size; C, symphysis mandibulae (leetotype), one-third natural size (Ci, front; Ci, right side; C3, under side). 



Present determination. — This species is provisionally 
referred to Brontotherium, but its exact position in 
that phylum is uncertain. (See p. 565.) 

Menodus dolichoceras Scott and Osborn, 1887 
Cf. Brontotherium dolichoceras, this monograph, page 572 

Original reference. — Mus. Comp. Zoology Bull., vol. 
13, No. 5, p. 160, figs. 3, 3; 5, 3; 6, 3, 1887 (Scott and 
Osborn, 1887.1). 



Dentition: I ?, C^-, P^, M^. Upper premolars with a faint 
internal cingulum. Nasal bones extremely short and obtuse. 
Horns extremely long and powerful, directed obliquely forward 
and outward, projecting beyond the nasals in side view. The 
section is suboval at the base, with the long axis obliquely 
transverse. Cranium very broad and saddle-shaped above 
the orbits, narrowing somewhat posteriorly. A prominent and 
overhanging superciliary ridge. Postglenoid and post-tym- 
panic processes united for a short distance. The skull which 
we have made the type of this species is much larger and more 
powerful than Professor Cope's type of jS. acer. The horns are 



DISCOVERY OF THE TITANOTHERES AND ORIGINAL DESCRIPTIONS 



221 



longer and more widely divergent at the base. The angle of 
inclination of the horns and the diminutive proportions of the 
nasals, as well as the form of the top of the cranium, all bring 
this specimen near S. acer and separate it from other known 
species. Unlike S. acer, the horns are not united by a ridge. 
[This is an error.] The specimen is incomplete in the supra- 
occipital region, the zygomatic arch is fragmentary, and the 
maxillary, palatine, and basioccipital regions are much dis- 
torted. 



Menodus platyceras Scott and Osborn, 1887 
Cf. Brontotherium platyceras, this monograph, page 578 
Original reference. — Mus. Comp. Zoology Bull., vol. 
13, No. 5, pp. 160, 161, fig. 4, 1887 (Scott and Osborn, 
1887.1). 

Subsequent reference. — The cranial evolution of Tita- 
notherium, p. 186, fig. 7A, 1896 (Osborn, 1896.110). 
The specimen figured is not the type. 





Figure 177. — Anterior part of skulls of (1) " Megacerops colorodensis" (not the type), now referred to Allops 
marshi; (2) Menodus tichoceras (type) (present determination, Brontotherium tichoceras) ; and (3) Menodus 
dolichoceras (type) (present determination, Brontotherium dolichoceras) 

Specimens in the Museum of Comparative Zoology, Harvard University. After Scott and Osborn, 1887. Greatly reduced in size. A, Side 
views: B, front views, showing the variations in the horns, nasals, and anterior nares; O, top views, showing the nasals and horns, and sections 
of the bases of the horns. 



Revised measurements 

IVIilUmeters 

Occipital condyles to nasal tips 690 

Free length of nasals 46 

Free breadth of nasals 90 

Outside measurement of horns 310 

Anteroposterior diameter of horns 85 

Transverse diameter of horns 125 

Etymology. — 56\ixoi, long; Kepas, horn. 

Present determination. — As shown in Chapter VI 
the present species probably pertains to Brontotherium 
rather than to Symhorodon. 



Type locality and geologic horizon. — Big Badlands, 
South Dakota; Chadron forma'tion (Titanotherium 
zone, Chadron C); exact locality and level not re- 
corded. S. Garman, collector. 

Type. — A pair of horns with the nasal bones at- 
tached. Now in the Museum of Comparative Zoology 
at Cambridge, Mass. (See fig. 178.) 

Neotype.SknW (Am. Mus. 1448). 

Characters. — Scott and Osborn write: 

Nasal bones e.xtremely short and obtuse, as in M. dolicho- 
ceras and M. acer. The inner [posterior] contour of the horns 



222 



TITANOTHERES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



is concave; they are greatly flattened anteroposteriorly, with 
a ridgelilie outer margin, and connected by a well-raised median 




Figure 178. — Type (holotype) horns of 
Menodus platyceras 

In the collection of the ]Museum of Comparative 
Zoology, Harvard University. After Scott and 
Osborn, 1887. Greatly reduced. A, Front view; B, 
cross section: C, side view. 

ridge. The posterior face is nearly plane, the anterior is con- 
vex, so that the section of the horn is plano-convex from base 



SPECIES DESCRIBED BY MARSH IN 1887 

Brontops Marsh, 1887 

Cf. Brontops, this monograph, page 482 

Original reference. — Am. Jour. Sci., 3d ser., vol. 34, 
p. 326, October, 1887 (Marsh, 1887.1). 

Type species. — Brontops rohustus Marsh. (See 
below.) 

Generic characters. — Marsh writes: 

The present genus is quite distinct from any of the forms 
previously described. * * * xhe skull is large and massive, 
with widely expanded zygomatic arches, and short and robust 
horn cores, projecting well forward. In general form it re- 
sembles the skull of Brontotherium but may be readily dis- 
tinguished from it by the dental formula, which is as follows: 
Incisors xi canines-}-; premolars |; molars |. 

The presence of four premolars in each ramus of the lower 
jaw is a distinctive feature in this genus. This character, with 
the single, well-developed incisor, marks both the known species 
[B. robusius, B. dispar]. 




FitiUHE 179. — Tj'pe (holotype) skeleton of Brontops, robusius 
Yale Mus. 12048. After Marsh, 1889. One twenty-fourth natural size. 



to tip. In side view the horns completely overhang the nasals 
and are slightly recurved. The long axis of the horn section is 
[almost or quite] directly transverse. 

Measure7nents 

Millimeter.'! 

Outside length of horns 315 

Transverse diameter of horns 125 

Anteroposterior diameter of horns . 67 

The type probably belongs to a young male in which the 
horns are not fully developed, because the horns increase in 
width and flatness and the basal section becomes more truly 
transverse, with age. 

Etymology. — irXarvs, flat; /cepas, horn. 
Present determination. — This valid species, described 
on page 578, belongs in the genus Brontotherium. 



Etymology. — Brontotherium; &^, face, "having the 
face or appearance of"; resembling Brontotherium. 

Present determination.— In 1902 Osborn (1902.208) 
treated Brontops as a synonym of Megacerops Leidy, 
but renewed examination of Leidy's figure of M. 
coloradensis indicates that it is not congeneric with 
Brontops, which is here regarded as a valid genus. 

Brontops robustus Marsh, 1887 
Cf. Brontops robustus, this monograph, page 492 
Original reference. — Am. Jour. Sci., 3d ser., vol. 34, 
p. 326, October, 1887 (Marsh, 1887.1). 

Subsequent references. — Restoration of Brontops ro- 
iustus: Am. Jour. Sci., 3d ser., vol. 37, pp. 163-165, pi. 



DISCOVERY OF THE TITANOTHERES AND ORIGINAL DESCRIPTIONS 



223 



6, 1889 (Marsh, 1889.1); skeleton and restoration, this 
monograph. Plates XCVI-CIII, CXCV-CCXXIX. 

Type locality and geologic horizon. — "Near the 
White River in northern Nebraska. " "The geological 
horizon is in the upper part of the Brontotheriwn 
beds [Chadron formation, Titanotherium zone] " 
(Marsh). "Upper levels of middle beds at least 60 
feet below the top of the upper beds" (Hatcher, 1901). 




Figure 180. — Type (holotype) lower jaw of Brontops dispar 
Nat. Mas. 4941. After Marsh, 1887. One-eighth natural size. 

Type. — A skull and skeleton, Yale Mus. 12048. 
(See fig. 179.) 

Specific characters. — Marsh did not formally sepa- 
rate the specific from the generic characters. He 
records the fact that the skull is large and massive, 
with widely expanding zygomatic arches and stout, 
robust horn cores, projecting well forward. 

Etymology. — roiustus, robust (that is, strong as an 
oak, rohur). 

Present determination. — The genus and species are 
valid. The species is described also on pages 492-499. 

Brontops dispar Marsh, 1887 

Cf. Brontops dispar, this monograph, page 488 

Original reference. — Am. Jour. Sci., 3d ser., vol. 34, 
pp. 327, 329, figs. 7, 8 (jaw); not figs. 5, 6 (skull), 
October, 1887 (Marsh, 1887.1). 

Type locality and geologic horizon. — Found on Hat 
Creek, Sioux County, Nebr., by J. B. Hatcher, May 
14, 1886; Chadron formation {Titanotherium zone), 
middle level. 

Type. — "A nearly complete skull with lower jaws 
and entire dentition." (Marsh.) Nat. Mus. 4941 
(skull D). (See fig. 180.) 

Characters. — Marsh writes: "The skull is less mas- 
sive and proportionately more elongate than in the 
type species, and the lower jaw more slender." In 
the same brief passage Marsh described a young skull 
(Nat. Mus. 4258) as belonging to the same species; 
this is a somewhat more primitive type (Brontops 



brachycephalus) belonging to a younger individual 
(p. 483). 

Etymology. — dispar, uneven, probably in allusion to 
the asymmetrical distortion of the type skull. 

Present determination. — The species is valid and is 
now referred to Brontops. 

Menops Marsh, 1887 
Cf. Menodus, this monograph, page 522 

Original reference. — Am. Jour. Sci., 3d ser., vol. 34, 
p. 328, October, 1887 (Marsh, 1887.1). 

Type species. — Menops varians. (See below.) 
Generic characters. — Marsh writes: 

The present genus is most nearly related to Diconodon and 
in its molar teeth agrees with that form. It differs in the 
presence of two upper incisors on each side. The superior 
dentition is as follows: Incisors, 2; canine, 1; premolars, 4; 
molars, 3. 

Etymology. — Menodus; &\p,i ace; resembling Menodus 
(cf. Brontops, above). 

Present determination. — The incisors are vestigial, 
the alveoli being very small. The skull presents re- 
semblance to both Menodus and Allops. The generic 
reference is to Menodus. 

Menops varians Marsh, 1887 

Cf. Menodus varians, this monograph, page 535 

Original reference. — Am. Jour. Sci., 3d ser., vol. 34, 
p. 328, fig. 9, October, 1887 (Marsh, 1887.1). 

Type locality and geologic horizon. — " Brontotherium 
beds of Dakota" (Chadron formation, Titanotherium 
zone); exact locality and level not stated. George 
A. Clarke, collector. 




FiGUKE 181. — Type (holotype) skull of Menops varians 
Yale Mus. 120G0. Front view. One-eighth natural size. 

Type. — A well-preserved skull (Yale Mus. 12060). 
(See fig. 181.) 

Specific characters.— Not separated by Marsh from 
generic characters. (See above.) 

Etymology. — varians, variant; allusion doubtful, but 
possibly to the somewhat aberrant character of the 
type skull. 

Present determination. — The species is valid and is 
referred to Menodus. 



224 



TITANOTHERES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



Titanops Marsh, 1887 

Cf. Brontolherium, this monograph, page 555 
Original reference. — Am. Jour. Sci., 3d ser., vol. 34, 
p. 330, October, 1887 (Marsh, 1887.1). 

Type species. — Titanops curtus. (See below.) 




Figure 182. — Type (holotype) skull of Titanops curtus 
Front view. Yale Mus. 12013. After Marsh, 1887. One-eighth natural size. 

Generic characters. — Marsh writes: 

This genus contains the largest members of the Brouto- 
theridae and some of the last survivors of the group. They are 
distinguished from all the other known types by the long, 
narrow skulls, lofty, flat horn cores, and short nasals. The 
upper dentition corresponds nearly to that of Brontotherium, 
but the upper molars have all two inner cones. * * * The 
nasals are the shortest known in the group. 

Etymology. — Titanotlierium; u^/, face — that is, like 
Titanotlierium. 

Present determination. — The genus is a synonym of 
Brontotherium Marsh. 

Titanops curtus Marsh, 1887 

Cf. Brontotherium curtum, this monograph, page 574 

Original reference. — Am. Jour. Sci. 3d ser., vol. 34, 
p. 330, fig. 11, October, 1887 (Marsh, 1887.1). 

Type locality and geologic horizon. — Colorado; e.xact 
locality not stated but recorded by Hatcher (1901) as 
from the upper Titanotlierium zone [of Chadron 
formation]. 

Type. — A complete skull with teeth (Yale Mus. 
12013). (See fig. 182.) 

Specific characters. — Not separated from generic 
characters by Marsh. 

Etymology. — curtus, short; in allusion to the short 
nasals. 

Present determination. — The species is valid and is 
referred to Brontotherium. 

Titanops elatus Marsh, 1887 
Cf. Brontotherium gigas, this monograph, page 567 
Original reference. — -Am. Jour. Sci., 3d ser., vol. 34, 
p. 330, fig. 12, October, 1887 (Marsh, 1887.1). 



Type locality and geologic horizon. — "Upper Titano. 
therium zone, South Dakota" (Chadron formation). 

Type.— A skull and jaw (Yale Mus. 12061). (See 
fig. 183.) 

Specific characters. — Marsh writes: 

The nasals are much longer, and the occipital crest much 
higher, than in the type species [T. curtus]. The zygomatic 
arches are unfortunately wanting, but the lower jaw is present, 
nearly in place. It shows no marked characters different 
from that of Brontops. 

Etymology. — elatus, lofty; possibly in allusion to the 
high stage of specialization. 

Present determination. — The species is synonymous 
with Brontotherium gigas Marsh. 

Allops Marsh, 1887 

Cf. Allops, this monograph, page 506 

Original reference. — Am. Jour. Sci., 3d ser., vol. 34, 
p. 331, October, 1887 (Marsh, 1887.1). 

Type species. — Allops serotinus. (See below.) 
Generic and specific characters. — Marsh writes: 

This skull in its general form resembles that of Brontotherium, 
but differs in having only a single upper incisor, and the last 
molar has the posterior inner cone more strongly developed. 

The superior dentition is as follows: Incisor, 1; canine, 1; 
premolars, 4; molars, 3. 

In the tj'pe specimen the canine is small, extending but 
httle below the premolars. There is no diastema. The upper 
premolars have a very strong inner basal ridge. The nasals 
are wide, expand forward in the free portion, and are notched 
in front. The entire length of the skull is 31 inches (79 centi- 
meters), the distance across the zygomatic arches 21 inches 
(53 centimeters), and the length of the horn cores about 10 
inches (25 centimeters). 




Figure 183. — Type (holotype) skull of Titanops elatus 
Front view. Yale Mus. 12061. After Marsh, 1887. One-eighth natural size. 

Etymology. — aXXos, strange; ciiA, face. 

Present determination. — Allops is intermediate be- 
tween Menodus and Brontops and is here regarded as 
a valid genus. (See p. 506.) 



DISCOVERY OF THE TITANOTHERES AND ORIGINAL DESCRIPTIONS 



225 



Allops serotinus Marsh, 1887 
Cf. Allops serotinus, this monograph, page 515 
Original reference. — Am. Jour. Sci., 3d ser., vol. 34, 
p. 331, October, 1887 (Marsh, 1887.1). 

Type locality and geologic horizon. — Quinn Draw, 
South Dakota, "near the top of the Brontotherium 
beds," Chadron formation {TitanotTierium zone). 




Figure 184. — Type (liolotype) skull of Allops serotinus 
After Marsh. Nat. Mus. 4251. One-seventh natural size. 

Type. — "A well-preserved skull and various other 
remains." U. S. Nat. Mus. 4251. J. B. Hatcher, 
collector. (See fig. 184.) 

Specific characters. — Not separated from generic 
characters in original description. 

Etymology. — serotinus, from sero{1), to bind, connect; 
possibly because the characters appeared to be more 
or less annectant with those of other species. 

Present determination. — The species is valid. It is 
described on page 515. 

CANADIAN SPECIES DESCRIBED BY COPE IN 1889 
Haplacodon Cope, 1889 

Cf. Allops, this monograph, page 506 
Original reference. — Am. Naturalist, vol. 23, p. 153, 
March, 1889 (Cope, 1889.1). (See p. 202.) 

Type species. — Menodus angustigenis Cope. The 
genus was founded on the characters of one of the 
several "types" of Menodus angustigenis, namely, a 



maxilla containing the fourth upper premolar and the 
three molars. 

Generic characters. — Cope writes: 

It differs from all the genera of the Menodontidae in the 
presence of but a single internal cusp of the first (posterior) 
superior premolar, a fact which renders it highly probable 
that the premolars which precede it in the maxillary bone 
were similarly constituted. It differs from all other genera of 
Lambdotheriidae and also from Diplacodon, to which it is 
allied, in the presence of but two inferior incisors on each side. 
It is not certain whether it possesses horns or not. 

Comparative measurements of the type of "Haplacodon" Cope, in 
millimeters 



M'-m' 

P', ap. by tr__ 
M', ap. by tr_ 
M\ ap. by tr_ 
M^, ap. by tr_ 



187 
38X62 
50X52 
66X62 
65X62 



169 
35X51 
45X51 
61X61 
60X61 



Etymology. — awXoos, simple; aKri, cone; 65ous, tooth; in 
allusion to the "single internal cusp" of the fourth 
upper premolar. 

Present determination. — The upper teeth agree 
closely in general characters with those of Allops and 
are intermediate in size between Allops walcotti and 
Allops marshi. 




A2 



A3 



Figure 185. — Type of Menodus selwynianus 

Coossifled nasal. Ai, Left side; Aj, upper side; A3, under side. Three-eighths 

natural size. 

Menodus selwynianus Cope, 1889 

Cf. Diploclonus selwynianus, this monograph, page 502 
Original reference. — Am. Naturalist, vol. 23, p. 628, 

July, 1889 (Cope, 1889.2). 

Subsequent reference. — On Vertebrata from the 

Tertiary and Cretaceous rocks of the Northwest 

Territory, I, p. 17, pi. 5, figs. 3, 3a, 3b, 1891 (Cope, 

1891.2). 



226 



TITANOTHERES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



Type locality and geologic horizon. — Swift Current 
River, Assiniboia, Canada; Cypress Hills beds, level 
not recorded. 

Type. — Coossified nasal bones detached from skull. 
Ottawa Mus. (See fig. 185.) 




Figure 186. — Type of Menodus syceras 
Coossified nasal and left horn eoie. After Cope. Ai, Leftside; A2, front; A3, section of left horn. One-half 



Characters of type. — Cope writes: 

Represented by a nasal process, which consists of the coossi- 
fied nasal bones, of peculiar form. They are elongate as com- 
pared with their width and are vaulted. The lateral borders 



are nearly parallel, and the extremity viewed from above is 
rounded. Owing to the thickness of the body, the profile 
descends abruptly at the extremity, and the convex surface is 
roughened as though for the attachment of some fixed body, 
tegumentary or muscular. From this tuberosity the surface 
descends steeply to a thin border. A short distance posterior 
to the extremity the lateral margins are 
decurved, forming the lateral walls of a 
deep longitudinal median gutter-like nasal 
meatus, which is deeper than in any other 
species. The horns are broken off, but the 
median inferior surface is so little recurved 
laterally that it is evident that the former 
were not only small but laterally placed. 
Length of fragment above, millimeters, 130; 
length of nasal border, 70; width at nasal 
notch, 80; width near extremity, 65; depth 
at apical tuberosity, 26. 

Additional ohservations. — The 
lower surface of the horns in the 
type exhibits a portion of the frontal 
sinus. The nasals are shorter than 
in the type of M. coloradensis. The 
measurements are as follows: 

Millimeteis 

Free length of nasals 80 

Free width of nasals 101 

Outside measurement of horns 67 

Anteroposterior measurement of horns 79 

Etymology. — "This species is dedi- 
cated to Dr. A. R. C. Selwyn, the 
accomplished Director of the Survey 
of Canada." (Cope.) 

Present determination. — The species 
is probably allied to Diploclonus 
hicornutus (Osborn). 

Menodus syceras Cope, 1889 

Cf. Megacerops syceras, this monograph, 
page 549 

Original reference. — Am. Natu- 
ralist, vol. 23, pp. 628-629, July, 
1889 (Cope, 1889.2). 

Subsequent reference. — Cope, On 
Vertebrata from the Tertiary and 
Cretaceous rocks of the Northwest 
Territory, I, p. 18, pis. 7, fig. 2; 8, 
figs. 4, 5, 1891 (Cope, 1891.2). 

Type locality and geologic hori- 
zon. — Swift Current River, Assiniboia, 
Canada; Cypress Hills beds, level 
not recorded. 
The nasal bones of three individuals 
present the characters above given." Of these we may 
select as the lectotype the specimen figured by Cope 
(1891.2, pi. 8) that shows the character from which 



DISCOVERY OF THE TITANOTHERES AND ORIGINAL DESCRIPTIONS 



227 



the name syceras is derived, in reference to the approxi- 
mation of the horns at their bases. Portion of right 
frontal, coossified nasals, and right horn. (See fig. 
186.) 

Characters of type. — Cope writes: 

It differs from the two species of that group now known, 
the M. proutii Leidy and the M. Hchoceras S. and 0., in the 
very close approximation of the basis of the horns and the pres- 
ence of a strong angle or ridge connecting them, so that the nasal 
bones are in a different plane from that of the front. The 
entire width of the skull at the basis of the horns is not greater 
than the length of each horn above the nasal notch. The 
horns are not long, and the section of their base is a longitudinal 
oval, flattened on the external side. Summit subround. The 
nasal bones are fiat, with broadly rounded extremitj', and are 
much wider than long. 

The width of the nasals at the base of the horns is 116 milli- 
meters; length of do. from do., 70; diameters of bases of horns; 
anteroposterior, 94; transverse, 67; length of horn from nasal 
notch, 160; width of muzzle at bases of horns inclusive, 160. 

Etymology. — aiiv, together; /cepas, horn; because the 
horns were set very near to each other at the base. 

Present determination. — M. syceras is at present 
indeterminate or possibly a synonym of M. angustigenis, 
both are provisionally referred to the genus Megacerops. 

LAST FIVE SPECIES DESCRIBED BY MARSH, 1890-91 
Diploclonus Marsh, 1890 

Cf. Diploclonus, this monograph, page 499 

Original reference. — Am. Jour. Sci., 3d ser., vol. 39, 
p. 523, June, 1890 (Marsh, 1890.1). 

Type species. — Diploclonus amplus. (See below.) 
Characters. — Marsh writes: 

One of the most marked features is seen in the horn cores, 
which are high, compressed transversely, and have a prominent 
knob on the inner superior margin about one-third of the dis- 
tance to the summit. Seen from the front the horn cores thus 
appear to be branched. It is probable that in life this feature 
was still more evident, and the covering of the horn core may 
have shown an actual division, but this can not be determined 
from the present specimen. There is a sharp ridge at the base 
of the horn cores on the outside. The nasals project but very 
little in front of the horn cores. The zygomatic arches are 
especially strong and widely expanded. The posterior nares 
have their front margin opposite the back of the last upper 
molars. 

There were apparently but two upper incisors — that is, one on 
each side — and no diastema exists behind the canines. The 
premolars have a strong inner basal ridge, and the last upper 
molar has two inner cones. This genus appears to be most 
nearly related to T'iianops, but the horn cores will distinguish 
it readily from all known forms of the Brontotheridae. 

Etymology. — 5t7rX6os, double; kXwv, a twig; in allusion 
to the branched appearance of the "horn core." 

Present determination. — The genus is now regarded 
as valid by Osborn. 

Diploclonus amplus Marsh, 1890 

Cf. Brontops amplus, this monograph, page 504 

Original reference. — Am. Jour. Sci., 3d ser., vol. 39, 
p. 523, June, 1890 (Marsh, 1890.1). 



Type locality and geologic horizon. — South Dakota; 
" Brontotherium beds" (= Chadron formation, or Titan- 
otherium zone). 

Type. — "Nearly complete skull, in good preserva- 
tion, but without the lower jaws." Yale Mus. 12015a. 
(See fig. 187.) 

Specific characters. — Marsh writes: 

The skull measures 28 inches from the front of the nasals to 
the back of the occipital condyles and 24 inches in greatest 
width across the zygomatic arches. The space occupied by the 
upper dental series is 13J^ inches, and by the true molars 8 
inches. 

Etymology. — amplus, broad, in allusion to the great 
breadth of the skull. 





FiGUEE 187. — Type skull of Diploclonus amplus 
After Marsh. One-eighth natural size. A', Side view; A', front view 

Present determination. — This species is cither an 
aberrant stage in the evolution of Brontops — that is, 
a sport — or a lateral phylum of Brontops. 

Teleodus Marsh, 1890 

Cf. Teleodus, this monograph, page 481 

Original reference. — Am. Jour. Sci., 3d ser., vol. 
39, p 524, June, 1890 (Marsh, 1890.1). 

Type species. — Teleodus avus Marsh. (See below.) 
Generic characters. — Marsh writes: 

The present genus is allied to Brontotherium but differs from 
it in having six lower incisors instead of four. It has the same 
number of inferior premolars and molars, and these teeth are 
similar in the two genera. From Diplacodon of the upper 



228 



TITANOTHERES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



Eocene the present genus may be distinguished by having 
only three lower premolars on a side instead of four. * * * 

Of the three lower incisors in place on each side, the middle 
one is the largest. There is a short diastema behind the 
lower canine, but no first premolar. The dental formula of 
the lower jaws is as follows: Incisors, 3; canine, 1; premolars, 
3; molars, 3. 

The space occupied by the lower dental series is 143^ inches, 
and by the last three molars 8H inches. 

Etymology. — reXeos, distant; dSovs, tooth; in allusion 
to the peculiar character of the incisors. 

Present determination. — The genus may either be 
valid or synonymous with an early stage in the evolu- 
tion of Brontops. 

Teleodus avus Marsh, 1890 

Cf. Teleodus avus, this monograph, page 481 

Original reference. — Am. Jour. Sci., 3d ser., vol. 39, 
pp. 523, 524, June, 1890 (Marsh, 1890.1). 

Type locality and geologic horizon. — " BrontotJierium 
beds of Dakota" ( = Chadron formation, or Titano- 

therium zone) ; exact 
geologic level not re- 
corded but probably 
lower beds (Chad- 
ron A). 

Type . — A lo wer j aw. 
Yale Mus. 10321. 
(See fig. 188.) 

Specific cliaracters. — 
Not separated by 
Marsh from the gen- 
eric characters. (See 
Type ol Teleodus avus p. 481.) 

Etymology. — avus, 
grandfather; in allu- 
sion to the primitive character of the animal. 

Present determination. — The species is probably 
valid. 

Allops crassicornis Marsh, 1891 

Cf. Allops crassicornis, this monograph, page 517 

Original reference. — Am. Jour. Sci., 3d ser., vol. 42, 
p. 268, September, 1891 (Marsh, 1891.1). 

Type locality and geologic horizon. — "BrontotJierium 
beds of South Dakota" ( = Chadron formation, or 
Titanotherium zone). Geologic level as recorded by 
J. B. Hatcher, collector, is the lower portion of the 
upper Titanotherium zone (Chadron C). 

Type. — A "nearly perfect skull of an adult but not 
old animal." Nat. Mus. 4289. (See fig. 189.) 

Specific characters. — Marsh writes: 

The skull is of medium size, with the zygomatic arches moder- 
ately expanded. The nasal bones do not project beyond the 
premaxillaries. The horn cores are very short and massive, 
with rounded summits, and thus form one of the striking fea- 
tures of the skull. The dentition is complete and in fine pres- 
ervation. The single incisor is quite small and situated close 




Figure 188.- 

Lower incisors and canines, 
natural size. 



to the canine. The latter is of moderate size and projects 
but little above the rest of the dental series. There is no 
diastema between the canine and the first premolar, which is 
small and has its inner face on a line between the canine and 
the second premolar. The second, third, and fourth premolars 
are large and have a strong inner basal ridge. The last molar 
has its anterior margin somewhat in advance of the front 
border of the posterior nares. 

The length of this skull on the median line is about 30 
inches, and the width across the zygomatic arches 23 inches. 
The width across the horn cores is 14 inches. The extent of 
the superior dental series is 16 inches. 

Etymology. — crassus, thick; cornus, horn. 

Present determination. — The species is valid. It is 
fully described on page 517. 

Brontops validus Marsh, 1891 

Cf. Brontops dispar, this monograph, pages 230, 488 

Original reference. — Am. Jour. Sci., 3d ser., vol. 42, 
p. 269, September, 1891 (Marsh, 1891.1). 

Type locality and geologic horizon. — The geologic 
level as recorded by J. B. Hatcher, collector, is the 
"middle portion of the middle Titanotherium beds, 
White River, S. Dak." (Chadron formation, horizon 
Chadron B). 

Type. — A "skull in fine preservation." Nat. Mus. 
4290 (skull K). (See fig. 190.) 

Specific characters. — Marsh writes: 

[The skull] agrees in its main characters with the other 
species of the genus but is particularly short and robust. The 
zygomatic arches are widely expanded, almost as much as in 
any skull of this group. The nasal bones have only a moderate 
extension in front and do not reach the end of the premaxil- 
laries. The free portion is broad and massive. The horn cores 
are of moderate size, nearly round in section, and have their 
obtuse summits directed somewhat backward. The occipital 
crest slopes forward and is expanded transversely. The length 
of this skull on the median line is about 26 inches. The great- 
est transverse diameter across the zygomatic arches is 22 inches, 
and across the summits of the horn cores 14 inches. 

Etymology. — validus, stout, brave. 
Present determination. — As shown on page 202, this 
species is probably synonymous with Brontops dispar. 

Titanops medius Marsh, 1891 

Cf. Broniotherium medium, this monograph, page 573 

Original reference. — Am. Jour. Sci., 3d ser., vol. 42, 
p. 269, September, 1891 (Marsh, 1891.1). 

Type locality and geologic horizon. — "Near the top of 
the Brontotherium beds of South Dakota" (Chadron 
formation, Titanotherium zone). J. B. Hatcher, col- 
lector. 

Type. — "One skull in fair preservation with the 
horn cores and dentition complete." Nat. Mus. 4256. 
(See fig. 191.) 

Specific characters. — Marsh writes: 

The free portion of the nasals is very small and projects but 
slightly beyond the anterior line of the horn cores. The latter 
are compressed anteroposteriorly and project laterally nearly 



DISCOVERY OF THE TITANOTHERES AND ORIGINAL DESCRIPTIONS 



229 



at right angles to the median line of the skull. The two in- 
cisors on each side are quite small and separated from each 
other and from the canine. There is a slight diastema behind 
the canine. The first premolar is small and triangular in out- 
line. The second premolar is of moderate size, and the third 
and fourth premolars have only an incomplete inner basal ridge. 
The width of this skull across the horn cores is 23 inches, 
and the distance from the end of the nasals to the front of the 



XAST SPECIES DESCRIBED BY COPE, 1891 
Menodus peltoceras Cope, 1891 

Cf. Brontotherium curtum, this monograph, page 574 

Original reference. — Am. Naturalist, vol. 25, p. 48, 
January, 1891 (Cope, 1891.1). 




FiGUKE 189. — Type skull of Allops crassicornis 
Palatal view. Nat. Mus. 4289. After Marsh. One-fifth natural size. 



posterior nares is 16 inches. The extent of the upper dental 
series is 17 inches. 

Etymology. — medius, middle; in allusion to the in- 
termediate character (between the species elatus and 
curtus) of this form. 

Present determination. — The species is probably 
valid and is referable to Brontotherium. 



Type locality and geologic horizon. — " Titanotherium 
beds of northern Nebraska" (Chadron formation). 

Type. — "Represented by the nasal region and the 
horn cores; the apex of one of the latter being broken 
away." Am. Mus. 10719. Dr. Hobart Hare, col- 
lector, Nebraska. Presented by the Museum of the 
University of Pennsylvania. (See fig. 192.) 



230 



TITANOTHERES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



Specific characters. — Cope writes: • 

The peculiarity of the species consists in the immense trans- 
verse extent of the horn cores and their complete fusion into an 
osseous wall which extends across the muzzle, forming a huge 
plate or shield. The superior border of this shield is moderately 
concave, a protuberant angle on each side representing the apex 
of each horn core. The nasal bones form a flattened protuber- 
ance much wider than long, which overhangs the nares. * * * 
Measurements: Elevation of horn-core plate at middle line 
behind, 180 millimeters; do. at lateral apex, 190 millimeters; 
total width of do. at middle, 300 millimeters. Projection of 
nasal bones beyond lateral base of horn-core plate, 20 milli- 
meters; width of nasal meatus at base of nasal bones, 65 milli- 
mieters; width of base of horn-core plate outside of nasal 
meatus, 90 millimeters. Anteroposterior diameter of base of 




, Figure 190. — Type (holot3'pe) skull of Brontops validus 
After Marsh. Nat. Mus. 4290. One-eighth natural size, 
horn core above side of and parallel to nasal meatus, 85 milli- 
meters. This species is nearest the M. platyceras S. and O., 
which has transverse compressed horn cores. They are, how- 
ever, distinct from each other, and not nearly so expanded 
transversely as in the present form. The M. pelioceras, in fact, 
carried a transverse shield on the end of its nose, which must 
have given it an extraordinary appearance. 

Etymology. — we\Tri, small shield; Kepa^, horn; be- 
cause the bases of the horns formed together a "huge 
plate or shield" extending across the muzzle. 

Present determination. — The type specimen (fig. 192) 
possibly represents a female of one of the long-horned 
species of Brontotherium, perhaps B. curtum, B. platy- 
ceras, or B. ramosum. The species is therefore practi- 
cally indeterminate at present. 



FIRST EUROPEAN OLIGOCENE SPECIES, DESCRIBED BY 
TOULA, 1892 

Menodus? rumelicus Toula, 1892 

Cf. Brontothermm rumelicum, this monograph, pages 660, 941 

Original reference. — Akad. Wiss. Wien, Math.-nat. 
Classe, Sitzungsber., Band 101, Abt. 1, pp. 608-615, 
1 pi., May, 1892 (Toula, 1892.1). 

Subsequent reference. — Ueber einen neuen Rest von 
Leptodonf (Titanotherium?) rumelicus Toula spec, pp. 
922-924, 1896 (Toula, 1896.1). 

Type locality and geologic horizon. — Near the railroad 
on the Jambol line near Kajali, northwest of Burgas, 
eastern Rumelia. Level, lower Oligocene (?). 
The formation from which the type was re- 
corded was correlated by Toula with the 
" Belvedereschotter." 

On account of the extreme rarity of titano- 
theres in Europe it seems important to 
note the published evidence concerning the 
provenience of the type and referred speci- 
mens of this species. According to Toula the 
specimens were received from his friend G. N. 
Zlatarski in Sofia. Toula does not state that 
Zlatarski himself collected the specimens. 
He states only that they must have come 
from near the railroad at Kajali, from the 
great heaps of material which had been dug 
up in the search for usable rubble ("taugli- 
chem Schotter"), and that these "Schotter- 
massen" should correspond at best with that 
isolated remnant of a formation at Lidscha, 
northwest of Burgas, of which he had already 
spoken in his first report on the geology of the 
eastern Balkans. He writes : "I have referred 
to these 'Schotter' as Belvedereschotter, and 
I believe, from the condition of preservation 
of the specimens from Kajali, and especially 
from the rusty sand grains still adhering to 
them, that they must be referred to the 
same kind of rock." Besides the specimens 
of titanotheres Toula records a lower molar 
and a canine of a "middle-sized rhinoceros" 
from the same locality. Later he received 
from the same locality, this also from Zla- 
tarski, a fragment of the lower jaw of a titanothere 
that included the symphyseal region (Toula, 1896.1, 
pp. 922-924). But Toula has not disproved the 
possibility that these specimens may have been im- 
ported from America, perhaps by laborers returning 
home from the western United States. (See p. 560.) 

Lectotype. — Third right lower molar and part of the 
right ramus of the lower jaw. (See fig. 193.) 

Paratypes. — A second right lower molar and a 
canine. 

Referred specimen. — The symphyseal region of the 
jaw with the roots of pi, p2, Ps, and the worn p4, 
in place. 



DISCOVERY OF THE TITANOTHEEES AND ORIGINAL DESCRIPTIONS 



231 



Specific characters. — Toula's description is too long 
to quote here. The principal characters revealed by 
his figures are, symphysis massive, canines (?) large, 
four lower premolars, lower molars with faint external 
cingula, hypoconulid of ms without strong internal 
crest. 

Etymology. — rumelicus, in allusion to Rumelia, the 
region in Hungary where the type was discovered. 

Present determination. — The species is probably 
valid, and its generic reference is probably to 
Brontotherium. 

SPECIES DESCRIBED BY OSBOEN IN 1896 AND 1902 
Titanotherium ramosum Osborn, 1896 

Cf. Brontotherium ramosum, this monograph, page 577 

Original reference. — Am. Mus. Nat. Hist. Bull., 
vol. 8, p. 1941, pi. 4, text fig. 13, 1896 (Osborn, 
1896.110). 

Type locality and geologic horizon. — "Upper 
Titanotherium beds, South Dakota." Chadron for- 
mation, Quinn Draw, Big Badlands, S. Dak. 

Type. — A complete male skull lacking incisive 
border. Am. Mus. 1447. (See fig. 194.) 

Characters of type. — Osborn writes: 

The distal spreading or branching of the horns is the 
character by which this species is designated. It differs 
from T. elatum in this character, but more especially in 
the great depth of the "connecting crest" and the ex- 
treme flattening of the horns, the section, as shown in 
diagram 1, being intermediate between that of the T. 
elatum and of T. plaiyceras. It is remarkable that the 
teetli in this large skull are relatively of 3mall size; the 
last upper molar has no second cone. 

Etymology. — ramosum, branched, in allusion to 
the "distal spreading or branching of the horns." 

Present determination. — The species is probably 
valid. 

Megacerops brachycephalus Osborn, 1902 

Cf. Brontops brachycephalus, this monograph, page 483 

Original reference. — Am. Mus. Nat. Hist. Bull., 
vol. 16, pp. 97-98, fig. 3 (not the type), 1902 
(Osborn, 1902.208). 

Type locality and geologic horizon. — Big Bad- 
lands, S. Dak.; Chadron formation, Chadron A, base 
or level A of lower Titanotherium zone. 

Type. — A complete skull (Nat. Mus. 4261, skull a), 
collected by J. B. Hatcher in 1887. (See fig. 195.) 

Specific characters. — Osborn writes: 

The type of this species is No. 4261, U. S. Nat. Mus. It 
includes very small, broad-skulled titanotheres with very rudi- 
mentary second internal cones upon the upper premolars; 
nasals elongate, narrowing anteriorly, as in Palaeosyops. Horns 
of anteroposterior oval section placed above orbits. It is 
represented in the National Museum by numerous skulls 
besides the type, all collected and recorded by Hatcher. One 
of these skulls was provisionally referred by him to Teleodus avus, 
from which this species is quite distinct. 



Etymology. — /3paxi's, short; Ki4>a\i), head, in allusion 
to the brachycephalic form of the skull. 

Present determination. — The species is probably 
valid. 

Megacerops bicornutus Osborn, 1902 

Cf. Diploclonus bicornutus, this monograph, pages 234, 501 

Original reference. — Am. Mus. Nat. Hist. Bull., 
vol. 16, p. 99, fig. 5, 1902 (Osborn, 1902.208). 




Figure 191. — Type (holotype) skull of Titanops medius 
After Marsh. Nat. Mus. 4256. One-eightli natural size. 

Type locality and geologic horizon. — Quinn Draw, 
White River, S. Dak.; exact level not recorded. Col- 
lected by J. W. Gidley, 1896. 

Type. — Skull and lower jaws (Am. Mus. 1476). 
(See fig. 196.) Paratype, skull (Am. Mus. 1081). 
One of these skulls (No. 1081) was first described by 
Osborn (1896.110, p. 176) as Titanotherium torvum ox 
rohustum. 

Specific characters. — Osborn writes: 

Horns directed anteriorly. Hornlets upon the inner and 
anterior midportion of the horn. Basal section of the horn 
slightly oval, subtransverse. Nasals narrow and relatively 
elongate. Sharp malar bridge in front of orbit. Orbit large. 



232 



TITANOTHERES OF AKCIENT WYOMING, DAKOTA, AND NEBRASKA 







Figure 192. — Type (holotype) nasofrontal shield 

of Menodus peltoceras 

Am. Mus. 10719. One-fourth natural size. 




FiGORB 193. — Cotypes of Menodus? rumelicus 

After Toula, 1892. Two-thirds natural size. The right lower molar 
(two upper figures) is the leetotype. 




Figure 194. — Type (holotype) skuU of Tiianoiherium ramosum 

After Osborn, 1896. Am. Mus. 1447. Ai, Side view, one-twelfth natural size; A2, top view, one- 
thirteenth natural size; A3, front view, one-thirteenth natural size. 



DISCOVERY OF THE TITANOTHERES AND ORIGINAL DESCRIPTIONS 



233 



This animal stands nearest M. selwynianus, though dis- 
tinguished by the greater size and slightly greater width 
of the nasals. The sharp malar bridge is the most abso- 
lute character. The two hornlets are possibly variations. 



Type locality and geologic horizon. — -Big Badlands 
(probably Cheyenne River badlands), S. Dak.; 
Chadron formation (TitanotJierium zone), exact level 
not determined. 




FiGTjHE 195. — Type skull of Megacerops brachycephalus 
Nat. Mus. 4261. One-fourth natural size. 



Etymology. — lis, twice; cornutus, horned; in allusion 
to the presence of small accessory horn swellings. 

Present determination. — The species is probably 
valid. 

Megacerops marshi Osborn, 1902 

Of. Allops marshi, this monograph, pages 511-515 

Original reference. — Am. Mus. Nat. Hist. Bull., vol. 
16, pp. 100-101, fig. 6, 1902 (Osborn, 1902.208). 
101959— 29— VOL 1-^18 



Type. — A complete skull (Am. Mus. 501). (See 
fig. 197.) 

Paratype. — Skull (Am. Mus. 1445). Collected by 
American Museum expedition, 1892. 

Specific characters. — Osborn writes: 

Type, skull No. 501; cotype, skull No. 1445, Am. Mus. 
Nasals elongate and square distaUy, horns short, of oblique 
oval basal section, overhanging the maxillae, or projecting 
forward or outward. Incisors, ?f . Canines short, tetartocones 



234 



TITANOTHERES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



of premolars moderately developed. These skulls were pre- 
viously confused by the writer with T. trigonoceras, from which 
they are readily separated by the horn section, which relates 
them to some of the primitive types of M. hrachycephalus and 
equally to M. robustus. The canines are more obtuse than in 




Figure 196. — Type (holotype) skull and lower jaw of Megacerops 

bicornutus 

Am. Mus. 14/6. After Osborn, 1902. One-eighth natural .size. 

M. dispar, and the superior incisors resemble those in Bronto- 
iherium rather than in M. robustus. 

Etymology. — Named in honor of the late Prof. O. C. 
Marsh, who estabUshed the remarkable collections of 
titanotheres in the Yale and National Museums, pro- 
posed the family name Brontotheridae, gave names to 
many of the genera and species, and projected the 
present monograph. 

Present determination. — The species is probably 
valid. 

Brontotherium leidyi Osborn, 1902 

Cf. Bronlhotherium leidyi, this monograph, page 558 

Original reference. — Am. Mus. Nat. Hist. Bull., vol. 
16, pp. 105-106, figs. 9, 10, 1902 (Osborn, 1902.208). 

Type locality and geologic horizon. — Big Badlands, 
S. Dak.; Chadron formation, lower levels of lower 
Titanotherium zone (Chadron A). 

Type.— A complete skull (Nat. Mus. 4249, skull R) 
collected by J. B. Hatcher in 1887. (See figs. 198, 
199.) 

Specific characters. — Osborn writes: 

Nasals elongate, narrowing anteriorly. Horns very short, 
slightly recurved, of transverse oval section. Canines stout 
and blunt. Premolars noncingulate, with rounded contours 
and weU-developed tetartocones. Incisors?^. 

Etymology. — Named in honor of Joseph Leidy, the 
first of the three great founders of American vertebrate 
paleontology, describer of Titanotherium, Megacerops, 
Palaeosyops, author of "The ancient fauna of Ne- 
braska" and of "The extinct mammalian fauna of 
Dakota and Nebraska." 

Present determination. — The species is probably 
valid. 



SPECIES DESCRIBED BY LULL IN 1905 
Megacerops tyleri Lull, 1905 

Cf. Diploclonus tyleri, this monograph, page 502. 
Original reference. — Jour. Geology, vol. 13, No. 5, 
pp. 443-456, pis. 3, 4, August, 1905 (Lull, 1905.1). 

Type locality and geologic horizon. — North side 
of Spring Draw Basin, about 10 miles from the 
mouth of Bear Creek, a tributary of Cheyenne 
River, S. Dak. Type specimen found 35 feet 
above the base of 200 feet of the Chadron 
formation {Titanotherium zone) lying upon Pierre 
deposits, "hence in the upper part of the lower 
division," as defined by Hatcher in 1893 (1893.1, 
p. 218). 

Type. — Skull, limbs, and many vertebrae of a 
single individual (Amherst Mus. 327). (See 
figs. 200 and 201.) Found by T. C. Brown, of 
the Amherst College paleontologic expedition of 
1903. 

Specific characters. — Lull writes: 

Horns well in front of orbits, directed somewhat 
forward and outward, an elongate oval in basal section 
with the long axes in line, rounded oval at the summit. 
Hornlets quite conspicuous, on the inner face of the 
horns midway between the base and summit. Con- 
necting crest low and inconspicuous. Nasals broad, 
well rounded in front, and but sHghtly arched beneath. 
Zygomata expanded and deep, with a well-rounded outer 
face. Dentition: Superior incisors represented by the deep 
and well-defined median alveoli and by the lateral teeth, 
which remain in place and which have hemispherical crowns 
which show little sign of wear. The canines are lanceolate, 
with a well-developed postero-internal cingulum. There is a 
short diastema in front of, and a longer one behind, the canine. 
Premolars with a smooth internal cingulum, less pronounced 
in the middle of the tooth, and with no external cingulum. The 
deuterocone is well developed, while the tetartocone, especially 
of premolar 4, is inconspicuous. 

The jaw is deep and robust, with the alveoli of two incisors, 
probably of the second and third, deep and distinct. There is 




Figure 197. — Type skull of Megacerops marshi 

After Osborn, 1902. Am. Mus. 501. One-tenth natural size. The lower 
jaw (Am. Mus. 516) figured with this skull does not belong with it. 
It is probably referable to Brontotherium leidyi. 



'no space between the lateral incisors and the canine, though 
between the two median alveoli a considerable gap occurs. 
There seems to have been a small diastema behind the lower 
canines, which are lanceolate, though with a less prominent 
cingulum, and not so strongly recurved as the upper ones. 



DISCOVERY OF THE TITANOTHERES AND ORIGINAL DESCRIPTIONS 



235 



Etymology. — Named in honor of Prof. John M. 
Tyler, of Amherst College, "a teacher of men, who, 
by his earnest efforts, as well as by his own generosity, 
was mainly instrumental in maldng possible the ex- 
pedition which secured the specimen" (Lull). 

Present determination. — This species is probably 
valid. It is discussed on page 502 of this monograph. 




Figure 198. — Type (holotype) skull of Brontotherium leidyi 

After Osborn, 1902. Nat. Mus. 4249. One-eighth natural size. The side view of this 
skull was figured by Marsh (Am. Jour. Soi., October, 1887) as Menops varians. 

SPECIES DESCRIBED BY OSBORN IN 1908 
Brontotherium hatched Osborn, 1908 

Cf. Brontotherium hatcheri, this monograph, page 563 

Original reference. — Am. Mus. Nat. Hist. Bull., vol. 
24, pp. 615-616, fig. 20, 1908 (Osborn, 1908.318). 

Type locality and geologic horizon. — South Dakota; 
Chadron formation, middle Titanotherium zone (Chad- 
ron B), lower levels. J. B. Hatcher, collector. 

Type. — A nearly complete skull (Nat. Mus. 1216, 
skull a) lacking the premaxillaries and anterior por- 
tion of the maxillaries. (See fig. 202.) 

Specific characters. — Osborn writes: 

If, Pf. Nasals moderately long (97 mm.), thin at the 
edges. Horns 250 -f millimeters, two-thirds the length oj 
B. gigas horns. Skull length (pm.x-condyles) , 710 (estimated), 
width across zygomata, 530 (estimated) . This species appears 
to represent an early phase of evolution of B. gigas. The horns 
are very round or convex in section and have a well-defined 
malar ridge on the lower outer portion. The connecting crest 
is relatively shallow, and the nasals are thin. The premolars 
are well advanced, the tetartocone of p' being well rounded and 
quite distinct. 

Etymology. — Named "in honor of the late J. B. 
Hatcher, who discovered many of Professor Marsh's 
titanothere types, brought together the great collec- 
tion of titanotheres ia the National and Yale Mu- 



seums, and placed the stratigraphic succession of the 
species upon a secure basis." (Osborn.) 

Present determination. — The species is probably 
valid. 

Symborodon copei Osborn, 1908 

Cf. Megacerops copei Osborn, this monograph, page 548 

Original reference. — Am. Mus. Nat. Hist. Bull., vol. 
24, pp. 616, 617, fig. 21, 1908 (Osborn, 1908.318). 

Type locality and geologic horizon. — South Dakota, 
Big Badlands, Indian Draw; Chadron formation, 
level probably middle Titanotherium zone (Chadron 
B). J. B. Hatcher, collector. 

Type. — A complete skull (Nat. Mus. 4711, skull 
V), collected by J. B. Hatcher, 1888. (See fig. 203.) 

Specific and generic characters. — Osborn writes: 

Incisors (type) persistent but greatly reduced; canines 
very small, reduced (28 mm.) ; premolars with cingula reduced 
or absent; tetartocones connected with deuterocones by a 
longitudinal ridge. Skull: nasals thin, short and broad in pro- 
portion, 80 by 125 millimeters; horns, S , 300, no connecting 
crest, transverse oval near summit; buccal processes of zygomata 
t? stout and conve.x; malar in front of buccal process very deep, 
beneath postorbital process stout, convex; occipital pillars 
not greatly expanded at the summits. 

Etymology. — Named in honor of the late Prof. 
E. D. Cope, prolific author of "The Vertebrata of the 
Tertiary formations of the West," original describer of 
Symborodon, founder of the "Cope collection," now 
in the American Museum of Natural History. 

Present determination. — The species is probably 
valid. 

CANADIAN SPECIES DESCRIBED BY LAMBE IN 1908 
Megacerops primitivus Lambe, 1908 

Cf. Teleodus primitivus, this monograph, page 482 

Original reference.— Contr. Canadian Paleontology, 
vol. 3, pt. 4, pp. 49-51, pi. 6, figs. 4, 5, 1908 (Lambe, 
1908.1). 




FiauRE 199.^Upper premolars of type 

skull of Brontotherium leidyi 

After Osborn, 1902. Nat. Mus. 4249. One-halt natural size. 

Type locality and geologic horizon. — "Oligocene 
deposits of the Cypress Hills," Saskatchewan. Col- 
lector, L. M. Lambe, 1904. 

Type. — Both halves of the lower jaw, with the denti- 
tion of the left side complete. Ottawa Museum. (See 
fig. 204.) 

Specific characters. — Lambe writes: 

Incisors, in three pairs, with a space between the inner pair; 
canines, of small diameter, apparently short; a diastema 



236 



TITANOTHERES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 








Figure 200. — Type (holotype) skull of Megacerops iyleri 

After Lull. Amherst Mus. 327. A, Dorsal aspect of skull; B, lateral aspect of skull and jaw; C, 
anterior aspect of horns and nasals; all about one-eighth natural size. D, Upper dentition (incisor, 
canine, and premolar series), one-fourth natural size. 



DISCOVERY OF THE TITANOTHERES AND ORIGINAL DESCRIPTIONS 



237 




Figure 201. — Right manus and right hind limb of the type of Megacerops tyleri 

Alter Lull, 1905. Amherst Mus. 327. A, Pro.dmal row of carpals, proximal aspect; B, distal row of carpals, proximal 
aspect (sc. /., scaphoid facet, lu.f., lunar facet, en. /., cuneiform facet); C, right manus; all one-fourth natural 
size. D, Right hind limb, one-eighth natural size. 



238 



TITANOTHERES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



between the canine and the first premolar; first premolar 
small; third premolar becoming molariform; fourth premolar 
molariform; symphysis long; symphyseal surface between 
canines narrow; jaw contracted at the diastema; external 
cingula moderately developed; internal cingula wanting; 
mental foramen beneath the second premolar; coronoid process 
short. 

Megacerops avus (Marsh), from the Oligocene of South 
Dakota, has three pairs of inferior incisors but only three pre- 
molars below on each side, and there is a short diastema behind 





P 


^ 


^B 




r 


T'tif' ^^ 


^^H 




L_ 


_ r^^ 


^^ 




Jr- , "^^B 


i^Tr^^H 


n 


W'^^^P" 


^H 


F 1 


B^ 


H^^' 


PI 


f 1 


^m 




kw ~ 


1 


i^m 




1 : 


-rf»r-"^ 


l^^l 




iH 


mn 


H 



Figure 202. — Type (holotype) skull of Bronlo- 
therium hatcheri 

Top view. Nat. Mus. 1216. After Osborn. 1908. One-tenth 
natural size. 

the lower canine. Its dimensions are greater than those of 
M. primitivus. These two species are apparently the only ones 
of the Oligocene titanotheres in which there are three pairs 
of incisors in the lower jaw. 

In the Cypress Hills specimen the crowns of the incisors are 
of a depressed spherical shape, with a tendency to come to a 
rounded central point above. The second incisor is the largest, 
and the first is slightly smaller than the third, which is the 
most upright. The first is more procumbent than the second. 
Between the inner pair is a very decided interval, leaving a 
space of 6.5 millimeters between the crowns of the two teeth. 
The crowns of the canines are broken ofi' (that of the right tooth 
being restored in fig. 5 of pi. 6) and the right first premolar is 
lost from its alveolus. * * * 

Keeping in mind the differences due to sex in titanotheres 
generally and the apparent variability, both specific and indi- 
vidual, of certain dental characters, such as the degree of devel- 
opment of the cingula, the presence or absence of the first 
premolar, the size of the canines, and the number of the incisors, 
M. primitivus is apparently a well-marked species, characterized 
principally, so far as known at present, by the breadth of the 
mandible anteriorly (as compared with M. angustigenis) and 
the presence of the fuU number of teeth, with a comparatively 
long diastema behind the canines. 

This species, for which the name primitivus is used, is regarded 
as representing a rather early stage in the development of the 



titanotheres. The general character of the dentition suggests 

the appropriateness of referring the species to the genus 

Megacerops. 

Measurements 

Millimeters 

Length of ramus 475 

Depth of same at posterior end of fourth premolar 74 

Depth of same at posterior end of second molar 81 

Depth of same from tip of coronoid process to lower 

border 247 

Maximum thickness of same beneath third molar 46 

Length of symphysis 144 

Distance apart of inside surface of base of canines" 31 

Length of premolar series 103 

Length of molar series 183 

Diameter of canines at base: 

Anteroposterior 18 

Transverse 16 

Diameter of second premolar: 

Anteroposterior 26 

Transverse 18 

Diameter of third premolar: 

Anteroposterior 32 

Transverse 23 

Diameter of fourth is premolar: 

Anteroposterior 35 

Transverse 27 

Etymology. — primitivus, primitive; in reference to 
the presence of three lower incisors. 

Present determination. — The species is probably 
valid. It is probably referable to Teleodus. 

Megacerops assiniboiensis nom. prov., Lambe, 1908 

Cf. Brontotherium curium, this monograph, page 574 

Original rejerence. — Contr. Canadian Paleontology, 
vol. 3, pt. 4, pp. 51-53, pi. 5, fig. 6, 1908 CLambe, 
1908.1). 




Figure 203. — Type (holotype) skull of Symborodon copei 
After Osborn, 1908. Nat. Mus. 4711. One-tenth-naturai size. 

Type locality and geologic horizon. — Oligocene de- 
posits of the Cypress Hills, Saskatchewan. Collection 
of 1904. 

" In the mandible of M. angustigenis (No. II) figured by Cope, op. cit. [1891.2] 
this measurement is about 18 millimeters, and in the symphysis of the jaw (No. I, 
also figured) a like measurement given, by the same authority, as 27 millimeters , 
should be 22 millimetei s. 

>' First premolar in Cope's description of M. angustigenis. 



DISCOVERY OF THE TITANOTHERES AND ORIGINAL DESCRIPTIONS 



239 



Type. — "A robust, short left mandibular ramus," 
lacking the posterior end. The three molars and the 




the fourth premolar and the first molar. The bone is massive 
and heavy throughout. The mental foramen is placed beneath 
the posterior root of the third premolar, farther back than in 
M. ■primiiivus. 

The cingula are very slightly developed. The external cingu- 
lum is present for a short distance only, on the anterior face 
of each of the four teeth, and in the third molar in advance of 
the heel. The only trace of an internal cingulum is to be seen 
in the third molar on the posterior slope of the heel. 




Figure 204. — Type (holotype) jaw of Megacerops primiiivus 
In the collection or the Ottawa Museum. After Lambe, 1908. A, Superior aspect, one-half natural size; B, lateral aspect, one-third natural size. 



fourth premolar are preserved, as well as part of the 
symphyseal region. Ottawa Museum. (See fig. 205.) 



The fourth premolar is fully molariform. The teeth are 
stout and of about the size of the corresponding ones in M. 




Figure 205. — Type (holotype) jaw of Megacerops assiniboiensis 
In the collection of the Ottawa Museum. After Lambe, 1908. One-third natural size. 



Characters. — Lambe writes: 

The jaw is much deeper, thicker, and relatively shorter than 
in angustigenis and primiiivus, and the teeth are much larger 
than in these species. It is narrow anteriorly, and the sym- 
physis extends back to a point in line with the division between 



marshi Osborn, but the jaw is relatively shorter than in this 
species. 

From the material available, the species, for which the 
provisional name assiniboiensis is proposed, can not be defi- 
nitely characterized. 



240 



TITAJSrOTHERES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



Measurements of ramus {type) 

Millimeters 

Depth of ramus at posterior end of fourth premolar 80 

Depth of ramus at posterior end of third molar 156 

Thiclcness of ramus above lower border beneath posterior 

end of first molar 55 

Vertical thickness of symphysis a little in advance of its 

posterior termination 53 

Vertical thickness of symphysis in line with front root of 

third premolar 31 

Space occupied by fourth premolar and the molars 260 

Diameter of fourth premolar: 

Anteroposterior 41 

Transverse 31 

Diameter of first molar: 

Anteroposterior 55 

Transverse 36 



SECOND EUROPEAN OIIGOCENE SPECIES, DESCRIBED BY 
KIERNIK, 1913 

Titanotherium bohemicum Kiernik, 1913 

Cf. Menodus giganteus, this monograph, page 530 

Original reference. — Acad. sci. Cracovie Bull., ser. B, 
vol. lOB, pp. 1211-1225, pi. 63, 1913 (Kiernik, 1913.1). 

Type locality. — Uncertain. The specimen, a frag- 
ment of the lower jaw containing the third right lower 
molar, was received with a lot of fossils from the dilu- 
vium near Prague. It was supposed to have come 
from the lime pits of Podbaba, near Prague, and to 
have been sold by one of the workers in the lime pits 




C D 

Figure 206. — Type of Titanotherium bohemicutn Kiernik 

Fragment of a lower jaw with third right lower molar. After Kiernik. A, Outer side view; B, inner side view; C, top view, showing 
the grinding surface of ms; D, front view, showing the exposed posterior roots of mj. Ahout one-fourth natural size. 



Diameter of second molar: 

Anteroposterior 71 

Transverse 41 

Diameter of third molar: 

Anteroposterior 99 

Transverse 43 

Space occupied by roots of third premolar (anteroposterior) 34 
Space between fourth premolars (twice the distance of 
fourth premolar from vertical plane through symphysis) - 60 
Etymology. — assiniboiensis , in allusion to the geo- 
graphic occurrence of the type. 

Present determination. — This species apparently be- 
longs in the Brontotheriinae. It is smaller than 
Brontotherium Tiatcheri. The nasals doubtfuUy referred 
by Lambe to this species suggest those of Bronto- 
therium curtum. 



to Herr Baumeister Kuchta (died 1910). He gave it, 
along with other prehistoric specimens, to Herr 
EoJanek, who in turn gave it to Herr Jira, who pre- 
sented it to the Institute for Comparative Anatomy 
at Prague. After carefully considering the possi- 
bility that the specimen might have been of American 
provenience the author, Herr Kiernik, inclines rather 
to the view that it really came from Bohemia, al- 
though not from Pfodbaba, but from the fresh-water 
Tertiary deposits of Tuchofitz (northwestern Bo- 
hemia). The well-known fauna of Tuchofitz is, 
however, of lower Miocene facies. 

Type. — A lower jaw fragment containing the third 
right lower molar. (See fig. 206.) 



DISCOVERY OF THE TITANOTHERES AND ORIGINAL DESCRIPTIONS 



241 



Characters. — Kiernik carefully compares the frag- 
ment with the types of Brachydiastemafherium tran- 
sihanicum Bockh and Maty, Menodus rumelicus Toula, 
and Titanotherium proutii Leidy. He shows that the 
third lower molar is much larger than that of either 
Brachydiastematherium or Menodus rumelicus, but 
that it is nearer in its measurements to the type of 
Titanotherium proutii, as indicated in the following 
table : 

Measurements of Titanotherium hohemicum, T. -proutii, and 
Menodus rumelicus, in millimeters 



T. bohemicum M. rumelicus 



Total length of the wear- 
ing surface 

Breadth of the first section 
(lobe) of the tooth 

Breadth of the second sec- 
tion (lobe) of the tooth... 



27 



T. proutii 



The author concludes that this species is widely 
distinct from the Ivnown European forms but that 
possibly it may eventually prove to be identical with 
either Titanotherium proutii or another species of the 
same genus. This, however, he considers unlikely, in 
view of its [supposed] European origin, so that he 
thinks he is quite justified in retaining the name 
Titanotherium hohemicum. 

Etymology. — hohemicum, in allusion to the country 
where the specimen was supposedly found. 

Present determination. — According to Dr. W. K. 
Gregory, who has compared a cast of the type of 
Titanotherium hohemicum with various American ti- 
tanotheres, the type specimen is closely similar to one 
in the American Museum of Natural History referred 
to Menodus giganteus (Am. Mus. 1007). It differs 
chiefly in the greater width of the anterior lobe of 
m3. It appears indeed to be specifically referable to 
Menodus giganteus, and it seems possible that it is 
in reality an American specimen which became mixed 
with the collection of fossils from Podbaba, near 
Prague. (Cf. pp. 230, 560, 941.) 

Measurements of Menodus bohemicus and M. giganteus, in 
millimeters 



M3, total length (estimated) 

M3, breadth of first lobe at base 

M3, breadth of second lobe 

M3, breadth of third lobe 

Center of protooonid to center of 

hypoconid 

Center of metaconid to center of 

entoconid 

Depth of jaw below front edge of ms-. 
Depth of jaw just behind ms 



108 
52 
47 
33 

39 

39 + 
111 
152 



M. giganteus 

(trigonoceras). 

Am. Mus, 

1007 



109 

47 
47 



111 
143 



FINAI OLIGOCENE SPECIES DESCRIBED BY OSBOEN IN 
1916-1919 

Allops walcotti Osborn, 1916 

See page 509 

Original reference. — Am. Mus. Nat. Hist. Bull., 
vol. 35, pp. 721, 722, fig. 1, 1916 (Osborn, 1916.433). 

Type locality and geologic horizon. — "Big Badlands," 
S. Dak., probably Corral Draw; Chadron formation 
{Titanotherium zone), lower levels (Chadron A). 

Type. — A nearly complete skull in the National 
Museum (No. 4260, skull Q). (See fig. 207.) 





Figure 207. — Type (holotype) skull of Allops walcotti 
Nat. Mus. 4260. After Osborn, 1916. One-eighth natural size. 

Specific characters. — Osborn writes: 

Premolars with small tetartocones; p'-m^ 285 millimeters. 
Incisors f . Horns elongate oval, no connecting crest. Mesa- 
ticephaUo. Nasals elongate, broad. Face relatively elongate. 

The type skull of this species (U. S. Nat. Mus. 4260) from 
level A is narrow and elongate, partly owing to lateral crushing. 
This feature conceals its resemblance to Allops marshi, which 
is apparent in other features — namely, (1) primitive, long 
nasals, (2) horns primitively short and obhquely oval, (3) large 
lateral incisor (ij) and small first (ij) or median incisor, (4) 
premolars accelerated, tetartocones more advanced than in 
Brontops robustus of level C. 



242 



TITANOTHERES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



Observations on the measurements oj AUops wal- 
cotti. — The type and only known specimen of this 
species exhibits the following comparison in measure- 
ments with skulls of B. hrachycephalus and Menodus 
Jieloceras, which show that the type of AUops walcotti 
has relatively large premolars and small molars. 

Measurements of AUops walcotti, Menodus heloceras, and Brontops 
brachycephalus, in millimeters 





A.walcotti, 
Nat. Mus. 
4260 (type) 


M. helo- 
ceras. Am. 

Mus. 

14576 


B. brachycephalus 




Nat. Mus. 
4940 


Nat. Mus. 
42S1 


Pi-m3 

Pi-p* 

Mi-m3. 


285 
112 
169 
640 
105 
100 
35X51 
60X61 


265 

170 

603 

132 

70 


265 
101 
160 


280 

" 104 

178 


Pmx-condyles 


680 




102 
32X51 
62X70 


85 




33X53 




68X73 







Etymology. — -"The species is named in honor of 
the Secretary of the Smithsonian Institution, Charles 
D. Walcott." (Osborn.) 

Present determination. — The skull is 'crushed later- 
ally but probably had a low zygomatic index — that is, 
it was mesaticephalic. While its reference to AUops 
requires confirmation, its nearer affinities appear to be 
with this genus rather than with Brontops or Menodus. 
The external cingula of the premolars are not as 
sharply defined as in other primitive members of the 
menodontine group. 

Megacerops riggsi Osborn, 1916 

See page 550 

Original reference. — Am. Mus. Nat. Hist. Bull., vol. 
35, p. 723, fig. 2, 1916 (Osborn, 1916.433). 

Type locality and geologic horizon. — Northeastern 
Colorado, Horsetail Creek; Chadron formation (Titano- 
therium zone), upper (?) levels. 



Type. — A nearly complete lower jaw in the American 
Museum (No. 6364). E. D. Cope, collector. (See 
fig. 208.) 

Specific characters (Osborn). — Of small size, smaller 
than any known individual of Megacerops or Bronto- 
therium. Very massive jaw with a small coronoid 




FiGUEE 208. — Type (holotype) jaw of Megacerops riggsi 
Am. Mus. 6364. After Osborn, 1916. One-sixth natural size. 

process and a very short symphysis. Premolar series 
greatly abbreviated (85 mm.). Premolars and molars 
with reduced external cingula. 

Measurements of type 

Millimeters 

Symphysis to condyle (estimated) 465 

Premolar-molar series (pi-ms) 282 

Premolar series (pi-pi) 85 

Molar series (mi-ma) 194 

Etymology. — Named "in honor of Mr. E. S. Riggs, 
of the Field Museum of Natural History, in recogni- 
tion of his discoveries of Eocene titanotheres." (Os- 
born.) 

Present determination. — The type of this species 
is a jaw in the Cope collection (Am. Mus. 4636), 
which was wrongly referred by Cope to his species 
" Symhorodon" acer. It represents a highly specialized 
and small form of Megacerops. 



Note. — For descriptions of upper Eocene and lower Oligo- 
cene titanotheres from MongoUa described by Osborn in 1923 
see appendix; also the final opinion regarding the titanotheres 
of eastern Europe, page 941. 



CHAPTER IV 

SYSTEMATIC CLASSIFICATION OF THE TITANOTHERES 



SECTION 1. 



PHYLETIC VERSUS LINNAEAN SYSTEM 
OF CLASSIFICATION 



NEO-LINNAEAN SYSTEMATIC DIVISIONS (ZOOLOGIC) AND 
EVOIUTIONARY PHYIA (PALEONTOIOGIC) 

As explained in the introduction, the Linnaean 
system was based on the theory of the special creation 
of all systematic divisions coinciding in geographic 
space, so that its application to our modern paleonto- 
logic phyla, which succeed one another over long 
periods of geologic time, is beset with great difficulties 
and has led to different uses of systematic terms by 
different authors. The present monograph employs 
a phyletic system which has been used by the author 
since 1892 in the classification of the Perissodactyla 
(Osborn, 1892.67, pp. 90-94). 

The taxonomic principle is that ancestral affinity is 
stronger than contemporary resemblance. Thus an 
animal that is directly ancestral to the titanotheres is 
placed in the family Brontotheriidae; an animal that 
is directly ancestral to BrontotJierium is placed in the 
subfamily Brontotheriinae; a series of ascending 
species in the same line are placed in the genus 
BrontotJierium; a series of "ascending mutations" may 
be placed within the single species BrontotJierium 
gigas. 

Such a vertical or phyletic application of the Lin- 
naean system involves, it is true, a departure from the 
traditional Linnaean methods, but in the author's 
opinion it is far preferable to the introduction of a 
new systematic terminology. If necessary the author's 
system may be distinguished as neo-Linnaean. It is 
an adaptation of the Linnaean system to phylogeny 
as revealed by paleontology. 

The degrees or steps in the evolution of neomorphic 
and heteromorphic characters, or rectigradations and 
allometrons, afford the real basis of our division of the 
great family tree of the titanotheres into branches 
and subbranches as follows: 

Family, a branch of the Perissodactyla having a large num- 
ber of similar characters and similar tendencies of evolution. 

Subfamily, a branch of the main family embracing one or 
more genera retaining certain similar characters and developing 
certain peculiar evolutionary tendencies. 

Genus, a branch of a subfamily or a stage of a subfamily 
distinguished by the prominent position of certain distinctive 
characters, which may be in widely different stages of develop- 
ment — for example, Brontotherium leidyi, B. platyceras. 

Species and subspecies, divisions distinguished by certain 
gradations in the development of characters common to the 
genus, also by certain rectigradations and allometrons. 

Ascending mutations, divisions distinguished by various 
intermediate stages of development of rectigradations and 
allometrons. 



These principles of phyletic classification as devel- 
oped and adopted in this monograph are also fully 
explained in Chapter I. 

Classification is simply a convenient and condensed 
expression of our knowledge of hereditary lines of 
descent. It is constantly shifting and changing with 
discovery. The final classification can be attained 
only after we have worked out all the lines of descent 
of this great family. In the meantime we may review 
the history of the successive attempts at classification 
made up to the present time. 

SUPEEFAMILY NAMES PROPOSED BY OSBORN (1898) AND 
HAY (1902) 

Superfamily Titanotherioidea Osborn, 1898 

Original reference. — Am. Mus. Nat. Hist. Mem., vol. 
1, pt. 3, p. 79, 1898 (Osborn, 1898.143). 

Osborn divided the Perissodactyla into five super- 
families : 

I. Titanotherioidea ("including the single family Titano- 
theriidae"), understood by Osborn to include both Eocene and 
Oligocene titanotheres. 

II. Hippoidea, including Equidae and Palaeotheriidae. 

III. Tapiroidea, including Tapiridae and Lophiodontidae. 

IV. Rhinocerotoidea, Including Hyracodontidae, Amyno- 
dontidae, Rhinocerotidae. 

V. Chalicotherioidea, Chalicotheriidae. 

Present determination. — Superfamily names are 
formed by adding oidea to the stem of the family 
name, and as Brontotheriidae is now regarded as valid, 
it appeared necessary to Doctor Hay to substitute for 
Osborn's term Titanotherioidea the term Bronto- 
therioidea, first used by Hay in 1902. 

Superfamily Brontotherioidea Hay, 1902 

Original reference. — U. S. Geol. Survey Bull. 179, 
p. 629, 1902 (Hay, 1902.1). 

The content of this term is as follows: 

Brontotherioidea : 
Brontotheriidae: 

Lambdotheriinae (Eocene titanotheres) . 
Brontotheriinae (Oligocene titanotheres). 

The content of the term Brontotherioidea Hay, 1902, 
is thus the same as that of Titanotherioidea Osborn, 
1898. 

FAMILY NAMES PROPOSED OR ADOPTED BY MARSH (1873), 
FLOWER (1875), COPE (1879-1889), AND OSBORN (1889) 

Family Brontotheridae Marsh, 1873 

Original reference. — Am. Jour. Sci., 3d ser., vol. 5, 
p. 486, 1873 (Marsh, 1873.1). 

Included genera. — Titanotherium Leidy and Bronto- 
therium Marsh. 

243 



244 



TITANOTHERES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



Family characters. — Not distinguished, but state- 
ment is made that Brontotherium was a " true perisso- 
dactyl with hmb bones resembling those of RM- 
noceros." Marsh gave the famUy characters fully in 
a paper entitled "On the structure and affinities of the 
Brontotheridae." He writes (Marsh, 1874.1, p. 82): 

Among the more marked characters of the Brontotheridae, 
which readily distinguished them from the Rhinocerotidae, 
apparently their near allies, may be mentioned the following: 
There are four short and thick toes in the manus, and three in i 
the pes. The skull supports a pair of large horn cores, placed 
transversely, as in modern artiodactyls.^' There are well- i 
developed canine teeth in both jaws. The molar teeth, above 
and below, are not of the Rhinoceros type but resemble those of 
Chalicotherium. 

Present determination. — As long as Brontotherium 
was regarded as a synonym of Titanotherium the term 
Brontotheriidae had no standing, but since Bronto- 
therium has been shown to be a good genus the term 
Brontotheriidae must be held valid. 

Family Limnohyidae Marsh, 1875 
Cf. Palaeosyopinae, this monograph, page 298 

Origin^ reference. — Am. Jour. Sci., 3d ser., vol. 9, 
p. 246, 1875 (Marsh, 1875.1). 

Present determination. — In defining the genus Dipla- 
codon, Marsh says: "From the Eocene Limnohyidae, 
already described, this genus is sharply distiuguished." 
The name Limnohyidae does not occur in Marsh's 
previous descriptions, and so far as one can judge the 
famUy had not been defined. As Limnohyus is a 
synonym of Palaeosyops the family name is invalid. 

Family Titanotherlidae Flower, 1876 

Cf. Brontotheriidae Marsh, this monograph, page 279 

Original reference. — Nature, vol. 13, p. 328, 1876 
(Flower, 1876.1). 

Present determination. — Flower regarded Bronto- 
therium as synonymous with Titanotherium and so 
naturally called the family Titanotherlidae; but siace 
Brontotherium is now regarded as valid, Flower's term 
becomes a synonym of Brontotheriidae Marsh. 

Family Chalicotherlidae Cope, 1879 

Original reference.- — U. S. Geol. and Geog. Survey 
Terr. Bull., vol. 5, p. 228, 1879 (Cope, 1879.1). 

Included genera. — "Limnohyus Leidy [ = Limnohyops 
Marsh], Palaeosyops Leidy, ' Leurocephalus S., O. & 
S.' [= Telmatherium cultridens], Menodus Pomel, Sym- 
lorodon Cope, Daeodon Cope, Chalicotherium Kaup, 
Nestor itherium Kaup." 

Present determination. — The titanotheres should 
never have been included in the same family with 
Chalicotherium. 

19 Ehinoceros pleuToceTOS Duv., from the Miocene of France, has a transverse pair 
of small horn cores on the nasals, not unlike those in DiTtoceras. R. mxnutua Cuv. 
has somewhat similar processes. 



Menodontidae Cope, 1881 
Cf. Brontotheriidae Marsh 

Original reference. — Am. Philos. Soc. Proc, vol. 19, 
pp. 378, 379, 397, 1881 (Cope, 1881.1). 

Present determination. — The name Menodontidae as 
applied to the Ohgocene titanotheres is invalid because 
antedated by Brontotheriidae Marsh. 

Family Lambdotheriidae Cope, 1889 
Cf. Lambdotheriinae, this monograph, page 279 

Original reference. — Am. Naturalist, March, 188 9 
p. 153 (Cope, 1889.1). 

Included genera. — From Cope's description it is 
plain that he intended to refer to the Lambdotheriidae 
not only the type genus Lambdotherium but all 
titanotheres with "but a single internal cusp on the 
first (posterior) superior premolar." He thus con- 
trasts the Lambdotheriidae with the Menodontidae 
( = Brontotheriidae). Cope then also referred to the 
family Lambdotheriidae an Oligocene genus "Hapla- 
codon" (= Megacerops angustigenis) . 

Synonymy. — The term Lambdotheriidae as used by 
Nicholson and Lydekker (1889.1, vol. 2, p. 1371) had 
the same connotation. It was apparently first limited 
to the genera Lambdotherium, Palaeosyops, and "Lim- 
nosyops" { = Limnohyops) by Flower and Lydekker 
(1891.1, p. 413) in 1891. Later authors, as Earle in 
1892 (1892.1) and Zittel in 1893 (1893.1, p. 300), used 
the term Palaeosyopidae or Palaeosyopinae to include 
the same genera. 

Present determination. — In this monograph the group 
under consideration is treated as a subfamily Lambdo- 
theriinae of the Brontotheriidae. 

Family Titanotherlidae Osborn, 1889 (1890?) 
Cf. Brontotheriidae Marsh, this monograph, page 279 

Original reference. — Am. Philos. Soc. Trans., new 
ser., vol. 16, p. 514, 1889 (1890) (Scott and Osborn, 
1890.1). 

Included genera. — Osborn writes: 

Palaeosyops has hitherto been referred to the ChaUcothe- 
riidae, but the discovery of the footbones of Chalicotherium 
by Filhol shows that the genera are widely separated. The 
discovery of the skeleton of Diplacodon, however, links Palaeo- 
syops very closely to Titanotherium. * * * It seems best 
to group the three genera [Palaeosyops, Diplacodon, and 
Titanotherium] in the single family Titanotherlidae. 

Present determination. — This was the first descrip- 
tion which included the true titanotheres of the 
Eocene and Oligocene without extraneous elements 
(Chalicotherium). The term is nevertheless pre- 
occupied by Titanotherlidae Flower, 1876, which is 
in turn a synonym of Brontotheriidae Marsh, 1873. 



SYSTEMATIC CLASSIFICATION OF THE TITANOTHERES 



245 



SUBFAMILY NAMES AND PHYLA PROPOSED BY STEIN- 
MANN AND DODERIEIN (1890), EARLE (1892), AND RIGGS 
(1912) 

Subfamily Falaeosyopinae Steinmann and Doderlein, 1890 

Original reference. — Elemente der Palaontologie, 
p. 777, 1890 (Steinmann and Doderlein, 1890.1). 

Included genera. — The authors divide the Chali- 
cotheriidae into three subfamihes — Falaeosyopinae, 
Brontotheriinae, Chalicotheriinae. The Falaeosyo- 
pinae include the genera Palaeosyops, "Limnohyus" 
{Limnohyops) , Diplacodon. 

Present determination. — Falaeosyopinae Steinmann 
and Doderlein, 1890, thus has priority over Falaeo- 
syopinae Earle, 1892. 

Subfamily Falaeosyopinae Earle, 1892 

Cf. Falaeosyopinae Steinmann and Doderlein 

Original type reference. — Acad. Nat. Sci. Fhila- 
delphia Jour., 2d ser., vol. 9, pp. 272 et seq., 1892 
(Earle, 1892.1). 

Included genera. — Lamidotherium, Limnohyops, 
Palaeosyops, Telmatherium, Haplacodon [Megacerops 
angustigenis]. 

Present determination. — Earle gives a detailed and 
accurate description of the subfamily characters 
(pp. 274-276). The term is preoccupied by Falaeo- 
syopinae Steinmann and Doderlein, 1890, and in its 
content is preoccupied by Lambdotheriidae Cope, 
1889. 

Subfamily Dolichorhinae Riggs, 1912 

Cf. Dolichorhininae 

Original reference. — Field Mus. Nat. Hist. Fub. 159, 
Geol. ser., vol. 4, No. 2, p. 25, June, 1912 (Riggs, 
1912.1). 

Included genera. — Middle Eocene titanotheres 
having nasals elongate and deeply recessed laterally, 
face shorter than cranium, an infraorbital process 
more or less developed, and molars only moderately 
expanded. 

This group is proposed in order to designate those 
long-nosed titanotheres which evidently sprang from 
a common stock and form a natural and homogeneous 
group. It includes the genera MesatirJiinus, Meta- 
rJiinus, Dolichorhinus, and RhadinorJiinus. 

DIVISION OF THE OLIGOCENE TITANOTHERES INTO FOUR 
CONTEMPORARY PHYLA, OSBORN (1902) 

Original reference. — Am. Mus. Nat. Hist. Bull., vol. 
16, pp. 91-109, February 18, 1902 (Osborn, 1902.208). 
Included genera. — Osborn writes: 

The Oligocene titanotheres consisted of at least four contem- 
porary phyla, to which the prior generic names Titanotherium, 
Megacerops, Symborodon, and Brontotherium may be applied. 

They represent an adaptive radiation for different local hab- 
itat, different modes of feeding, fighting, locomotion, etc., which 
took origin, in part at least, in the middle or upper Eocene. 
Europe and Asia also may have shared in this radiation, since 
titanotheres are now definitely known in the Balkan region. 



The main phyletic characters are analogous to those recently 
(Osborn, 1900, p. 231) determined among rhinoceroses; the 
great antiquity of the lines leading to the existing species of 
rhinoceroses necessitated the revival of a number of discarded 
generic names to distinguish them. Similarly the separateness 
of four of the titanothere phyla, throughout the Ohgocene and 
possibly from the Eocene, renders it desirable to revive certain 
generic names which in my first review I considered undefinable. 

Radiation involved three main sets of characters, two of 
which were correlated: 

First, doliohocephaly and brachycephaly, associated with 
numerous changes in the skull and teeth and, in at least two 
phyla, with longer and shorter limbs. 

Second, four distinct types in the shape and position of the 
horns, correlated with the structures of the nasals and frontals 
and indicative of different modes of combat among the males. 
(See fig. 209.) 

Third, canines of different form; and, finally, the presence 
of one or two pairs of functional incisor teeth, or the total 
degeneration of these teeth. 

Titanotherium Leidy applies to long-limbed animals with 
long skulls, persistently long and broad nasals, short triangu- 



MegaceropSy Upper Beds. 




Diplodonus, Upper Beds. 



Syiiiborodoti. Upper Beds, 



.Brontotkcriuv!, Upper Beds, 

Figure 209. — Characteristic basal sections of horns of 
Oligocene titanotheres 

l.ar horns placed slightly in front of the eyes, vestigial incisors, 
^:?, large canine teeth. Known from the base to the summit of 
the Oligocene. 

Megacerops Leidy applies to titanotheres with broad skulls, 
nasals progressively shortening, short horns rounded or oval in 
section, shifting anteriorly, one or two pairs of incisor teeth, 
^, medium-sized canine teeth. Known from the base to the 
summit of the Oligocene. 

Probably related to this are the subgenera of the types named 
Allops and Diploclonus by Marsh, differing from the above in 
horn characters. Known chiefly from the upper beds. 

Symborodon Cope includes titanotheres with skulls of vary- 
ing proportion, nasals slender and progressively shortening, 
horns elongate and peculiar in being placed above the eye 
instead of shifting forward, incisors vestigial, |^, canines small, 
appro.ximated. Known only from the middle and upper beds. 

Brontotherium Marsh embraces the largest titanotheres, with 
very broad zygomatic arches, nasals shortening while horns 
elongate and shift forward; incisors persistent, f in the males, 
canines stout and obtuse. 

Representatives of Titanotherium and Megacerops can now 
be continuously traced from the base to the summit of the 
Oligocene. Primitive species of Brontotherium also appear at 



246 



TITANOTHEKES OF ANCIENT WYOMING, DAKOTA, AND NEBRASKA 



the base, although the phyletic sequence through the middle 
to the upper beds is not so clear. Symborodon suddenly appears 
in the middle beds. 

The names of three of the genera thus recognized 
were subsequently changed, for the reasons given, as 
follows: For " Titanotherium Leidy" was substituted 
Menodus Pomel; for " Ilegacerops Leidy" was substi- 
tuted Brontops Marsh; for "Symborodon Cope" was 
substituted Megacerops Leidy; " Brontotherium Marsh" 
was permanently accepted. The phyla subsequently 
were called subfamilies. (See below.) 

RECLASSIFICATION OF THE EOCENE AND OLIGOCENE 
SUBFAMILIES BY OSBORN (1914) 

Original reference. — Geol. Soc. America Bull., vol. 25, 
pp. 403-405, Sept. 15, 1914 (Osborn, 1914.409). 

Reasons for reclassification. — Osborn makes the 
following statement : 

Recent discoveries have modified the author 's earlier opinions 
as to the lines of descent of the titanotheres, and still further 
changes are anticipated with increase of knowledge of the 
connections between upper Eocene, or Uinta, titanotheres and 
those of the lower Oligocene, or White River. 

The main lines of division are indicated in the proportions of 
the limbs, whether cursorial, mediportal, or graviportal; the 
proportions of the skull, whether mesaticephalic, brachy- 
cephalic, or dolichocephalic; the development of frontonasal 
horns, whether accelerated or retarded; the molarization of the 
premolar teeth, whether accelerated or retarded; the presence 
or absence of incisor teeth; the abbreviate or elongate, the tri- 
angular or oval form of the frontonasal horns as developed in 
Oligocene times. 

The new arrangement. — With these criteria the vari- 
ous phyla were distinguished in 1914 as follows: 

A. Wind River titanotheres, face longer than cranium: 

I. Lambdotheriinae, light-limbed, cursorial: 
Lambdotherium. 
II. Eotitanopinae, medium-limbed, mediportal: 
Eotitanops. 

B. Bridger and succeeding titanotheres, cranium longer than 
face: 

III. Palaeosyopinae, short-limbed, brachycephalic : 

Palaeosyops, Limnohyops. 

IV. Telmatheriinae, mesaticephalic to dolichocephalic: 

Telmatherium, Sthenodecies. 
V. Diplacodontinae, dolichocepahlic, with accelerated 
molarization of the premolars, imperfectly known: 
Diplacodon. 
VI. Manteoceratinae, mesaticephalic to brachycephalic, 
accelerated development of the horns, mediportal: 
Manteoceras, Protitanotherium. 
VII. Dolichorhinae, mesaticephalic to dolichocephalic, 
limbs, so far as known, abbreviate: 

Dolichorhinus, Mesatirhinus, Sphenocoelus, Meta- 
rhinus, Rhadinorhinus. 
VIII. Menodontinae, mesaticephalic to dolichocephalic, 
with abbreviate, triangular horns, with incisor teeth 
reduced or wanting, feet and limbs elongate: 
Menodus { = Titanotherium), Allops. 
IX. Brontopinae, brachycephalic, horns abbreviated, 
rounded, or oval, incisors persistent: 

Brontops {= Megaceratops^") , Diploclonus. 

" Error; should have been Megacerops. 



B. Bridger and succeeding titanotheres — Continued. 

X. Megaceropinae, mesaticephalic to extreme brachy- 
cephalic, horns elongate, vertically placed, no in- 
cisor teeth: 

Megacerops {^Symborodon). 
XI. Brontotheriinae, mesaticephalic to brachycephalic, 
horns elongate, transversely flattened and diver- 
gent: 

Brontotherium. 

The free use of subfamily divisions to express the 
distinct phyletic series is similar to that which the 
author adopted in the phylogeny of the rhinoceroses. 
More conservative usage would have divided the titano- 
theres into four subfamilies only. Of these names 
of phyla those assigned to Nos. II, IV, V, VI, VIII, 
IX, X, and XI had apparently not hitherto been pub- 
lished, and those assigned to Nos. I, III, and VII, 
although they had been used in previous publications, 
mostly by other authors, were now used in a more 
restricted sense. 

Other subfamilies awaited further study and the 
discovery of connecting forms, namely : 

Diplacodontinae = ancestors of Menodontinae or Bronto- 
theriinae. 

Eotitanopinae = ancestors of Palaeosyopinae. 
Rhadinorhininae = ancestors of Megaceropinae. 

Each subfamily name is carried back as far as possi- 
ble — that is, to the point, even very remote, where the 
subfamily characters and tendencies of evolution are 
first clearly and unmistakably manifested. 

SPECIES WRONGLY REFERRED TO THE TITANOTHERES 
Palaeosyops minor Marsh, 1871 (=Anchippodus minor) 

Original reference. — Am. Jour. Sci., 3d ser., vol. 2, 
p. 36, 1871 (Marsh, 1871.1). 

Type. — "A molar tooth, from the right lower jaw, 
and probably by some other less characteristic re- 
mains" from Grizzly Buttes, Bridger Basin, Wyo. 

Present determination. — This specimen was wrongly 
referred to Palaeosyops, as was recognized by Marsh, 
Cope, and others. The specimen pertains to the order 
Tillodontia. 

Helotherium procyoninum Cope, 1872 

Original reference.— Fed. Bull. No. 2, p. 466, 1872 
(Cope, 1872.2). 

Synonymy. — LambdotTierium procyoninum Cope, 
Tertiary Vertebrata, pp. 631, 711, pi. 24, fig. 22, 1884 
[1885] (Cope, 1885.1). 

"Syn.? of OroUppus pumilis," Hay (1902.1, p. 612). 

Hyracotherium procyoninum Matthew, Am. Mus. 
Nat. Hist. Bull., vol. 12, p. 45, 1899 (Matthew, 
1899.1). 

OroMppus sp. Granger, Am. Mus. Nat. Hist. Bull., 
vol. 24, p. 227, 1908 (Granger, 1908.1). 

Daeodon shoshonensis Cope, 1878 

Original reference. — Pal. Bull. No. 30, "December 3, 
1878" (Cope, 1878.1). 



SYSTEMATIC CLASSIFICATION OF THE TITANOTHERES 



247 



Type and geologic horizon. — "The terminal portion 
of the lower jaw of a huge mammal " (Am. Mus. 7387), 
from the Miocene of Oregon. 

Present determination. — The genus and species be- 
long in the family Entelodontidae (Peterson, 1909.1, 
p. 63). 

SECTION 2. CLASSIFICATION OF THE TITANOTHERES 
ADOPTED IN THIS MONOGRAPH 

SYNOPSIS OF THE CLASSIFICATION 

The natural classification or ancestral tree of the 
titanotheres is based on the characters of the skull 
and teeth, as set forth in Chapters V and VI, com- 
bined with those of the limbs and feet, as set forth in ' 
Chapter VII. The full definitions of the family and 
of the 12 subfamilies into which the titanotheres are 
now divided are presented in Chapters V and VI, of 
which the following classification is a synopsis. It 
should be compared with the phylogenetic tree given 
m Chapter X (p. 769). Each of the chief phyla has 
H subfamily name. 

A. Wind River titanotheres, face longer than cranium: 

I. Lambdotheriinae, light-limbed, cursorial: 
Larnhdoiherium. 
II. Eotitanopinae ( = ?Palaeosyopinae), medium- 
limbed, mediportal: 
Eoiitano-ps. 

B. Bridger and succeeding titanotheres, cranium longer than 

face: 

III. Palaeosyopinae ( = ?Eotitanopinae), short-limbed, 

brachycephalic: 

Palaeosyops, Limnohyops. 

IV. Telmatheriinae, mesatioephalic to dolichocephalic: 

Telmatherium, Sthenodectes. 
V. Manteoceratinae ( = Brontopinae), mesaticephalic to 
brachycephalic, accelerated development of the 
horns, mediportal: 

Manteoceras, Protitanoiherium, Brachydiasie- 
ntatherium. 
VI. Dolichorhininae, mesaticephalic to dolichocephalic; 
limbs, so far as known, abbreviate; facial region 
downturned: 

Eomeiarhinus, Dolichorhinus, Mesatirhinus, 
Sphenocoelus, Metarhinus. 
VII. Rhadinorhininae ( = ?Megaceropinae), mesatice- 
phalic, facial region cyptocephalic, upturned: 
Rhadinorhinus. 
VIII. Diplacodontinae ( = ?Menodontinae, =7Bronto- 
theriinae), dolichocephalic, with accelerated 
molarization of the premolars, imperfectly 
known: 

Diplacodon, Eotitanoiherium. 
IX. Brontopinae (= Manteoceratinae), brachyce- 

phalic, horns abbreviated, rounded or oval, 
incisors persistent; premolars retarded: 

Teleodus, Brontops { = Megacerops), Diplo- 
clonus. 
X. Menodontinae ( = ?Diplacodontinae), mesatice- 
phalic to dolichocephalic, with abbreviate, tri- 
angular horns, with incisor teeth reduced or 
wanting, feet and limbs elongate, premolars 
accelerated: 

Menodus {= Titanotherium) , Allops. 



B. Bridger and succeeding titanotheres — Continued. 

XI. Megaceropinae ( = ?Rhadinorhininae), mesatice- 
phalic to extreme brachycephalic, horns elon- 
gate, vertically placed, no incisor teeth: 
Megacerops {=^ Symhorodon) . 
XII. Brontotheriinae ( = ?Diplacodontinae), mesatice- 
phalic to brachycephalic, horns elongate, trans- 
versely flattened and divergent, premolars 
accelerated: 
Bronlotherium. 

Suggestions as to resemblance or the affinity between 
subfamilies are given above in parentheses, and the 
families are arranged according to the general geologic 
sequence. One of these suggestions of ancestral 
affinity is now apparently well established, namely, 
that the Manteoceratinae are ancestors of the Bron- 
topinae. 

I. TITANOTHERES OF LOWER EOCENE TIME 

(Face elongate) 
Group I. Hornless: 

1. Subfamily Lambdotheriinae Osborn. "Lamb- 

dotheres." (Lower Eocene titanotheres. 
Long-headed, very small; body and limbs 
slender and cursorial; face longer than cra- 
nium, slender.) Pages 

Genus Lambdolherium Cope 168,279,690 

Species priscum Osborn 194, 286, 590 

primaevum Loomis 178, 283, 590 

popoagicum Cope 168, 281, 590 

progressum Osborn 194, 286, 590 

magnum Osborn 199,288,590 

2. Subfamily Eotitanopinae (— ?Palaeosyopinae) 

Osborn. " Eotitanopines. " (Lower Eocene 
titanotheres of intermediate size. Head of 
medium length; body and limbs less slender 
and cursorial than in the lambdotheres; gait 
submediportal; face longer than cranium.) 

Genus Eoiitanops Osborn 179, 289, 591 

Species gregoryi Osborn 192, 291, 593 

brownianus (Cope) 169,292 

borealis (Cope) 168,292,593 

princeps Osborn 193,295,593 

major Osborn 193, 296, 597 

minimus Osborn 199,296 

II. TITANOTHERES OF MIDDLE AND UPPER EOCENE TIME 
(Face abbreviate) 

Group II. Retarded horn rudiments: 

3. Subfamily Palaeosyopinae (=? Eotitanopinae) 

Steinmann and Doderlein. "Palaeosyo- 
pines." (Titanotheres larger than tapirs. 
Broad-headed, skull and limb proportions be- 
coming stout; skull broad; zygomata progres- 
sively brachycephalic; grinders small; nasals 
tapering distaUy; face shorter than cranium; 
feet abbreviate, brachypodal; gait gravi- 
portal.) Pages 

Genus Limnohyops Marsh (mesaticephalic 

to brachycephalic) 170,303,612 

Species prisons Osborn 180, 306 

laevidens (Cope) 163,305 

matthewi Osborn 180, 308 

monoconus Osborn 180, 309, 614 

laticeps Marsh 160, 311, 618 



248 



TITANOTHERES OF ANCIENT "WYOMING, DAKOTA, AND NEBRASKA 



Group II. Retarded horn rudiments — Continued. 

3. Subfamily Palaeosyopinae — Continued. Pages 

Genus Palaeosyops Leidy (bracliycephalic, 

hyperbrachy cephalic) 157, 312, 619 

Species ffoniinalis Cope 165,317 

longirostris Earle 172,319 

paludosus Leidy 157, 319 

major Leidy 158, 321, 620 

grangeri Osborn 181, 335 

leidyi Osborn 18 1, 323, 620 

robustus (Marsh) 161, 331, 626 

copei Osborn 181, 336, 629 

4. Subfamily Telmatheriinae Osborn. "Telma- 

theres." (Middle and upper Eocene titano- 
theres of larger size. Heads of medium length, 
with large cingulate incisors and heavy, sub- 
lanceolate canines; grinders large; mesatice- 
phaUc {Telmatherium) or subbrachycephalic 
(Sthenodectes) ; of mediportal gait.) 
Genus Telmatherium Marsh (mesaticephalic, 

narrow sagittal crest) 160, 340 

Species cuUridens (Osborn, Scott, and 

Speir) 168, 341 

validum Marsh 160, 344 

altidens Osborn 184, 351 

ultimum Osborn 184, 345 

Genus Sthenodectes Gregory (mesaticephalic 

to subbrachycephalic) 190, 353 

Species incisivus (Douglass) 185, 354 

Group III. Accelerated horn rudiments : 

5. Subfamily Manteoceratinae (Brontopinae) Osborn. 

" Manteoceratines " (prophet-horn), "bronto- 
pines." (Precociously horned titanotheres, of 
the same stock as the Dolichorhininae. Skull 
mesaticephalic, face abbreviate; feet abbre- 
viate, brachypodal, gait graviportal; premolars 
retarded, incisors rounded; ancestral or related 
to the Brontops phylum of the Oligocene.) 
Genus Manteoceras Hatcher (horns rudi- 
mentary) 177, 362, 631 

Species manteoceras Hay 177, 365, 631 

washakiensis Osborn 182,371 

uintensis Douglass 186, 372 

Genus Protitanotherium Hatcher (horns elon- 
gate, oval, more prominent than in Man- 
teoceras) 176, 375 

Species emarginatum Hatcher 177, 377 

superhum Osborn 185, 379 

Genus Brachydiastematherium Bockh and 
Maty (large size; upper Eocene of Tran- 
sylvania) 166, 382 

Species transilvanicum Bockh and 

Maty 166, 382 

6. Subfamily Dolichorhininae Riggs. "DoUcho- 
rhines" (long-snouted). (Middle and upper 
Eocene titanotheres. Typically dolichocepha- 
lic and dolichopic; nasals typically long and 
expanding distally; precocious horn rudiments; 
infraorbital shelf usually conspicuous.) 
Genus Eometarhinus Osborn (ancestral to 

Metarhinus, primitive, mesaticephalic). 200, 419 
Species huerfanensis Osborn 200, 420 



Group III. Accelerated horn rudiments — Continued. 

6. Subfamily Dolichorhininae — Continued. Pages 

Genus Mesatirhinus Osborn (ancestral to 

Dolichorhinus; subdolichocephalic) . 182,387,636 

Species Junius (Leidy) 159, 388 

megarhinus (Earle) 170, 388 

peiersoni Osborn 182, 389, 641 

Genus Dolichorhinus Hatcher (extremely 
dolichocephalic, cyptocephalic; becoming 

extinct) 177, 396, 645 

Species superior (Riggs) 190, 395, 405 

longiceps Douglass 188,406,651 

vallidens (Cope) 162, 401 

heterodon Douglass 187,416 

intermedins Osborn 1 84, 405 

hyognathus (Osborn) _ 169, 173, 409, 646 

Jluminalis Riggs 191, 417 

Genus Metarhinus Osborn (dwarfed, aber- 
rant, mesaticephalic) 183, 420, 648 

Species earlei Osborn 183, 420 

fluviatilis Osborn 183, 421 

cristatus Riggs 191, 429 

riparius Riggs 191, 429 

Genus Sphenocoelus Osborn (little known; (?) 

branch of Mesatirhinus) 174, 417 

Species uintensis Osborn 175,419 

7. Subfamily Rhadinorhininae ( = ?Megaceropinae) 

Osborn. " Rhadinorhines " (slender - nosed) . 
(Middle Eocene titanotheres. Mesaticephalic, 
cyptocephalic; infraorbital shelf reduced.) 
Genus Rhadinorhinus Riggs (nasals short, 
pointed; possibly ancestral to the Oligo- 
cene Megacerops; cyptocephalic) 192, 430 

Species abbotti Riggs 192, 430 

diploconus (Osborn) 173, 431 

Group IV. Short-horned: 

8. Subfamily Diplacodontinae (=?Menodontinae, 

Brontotheriinae) Osborn. " Diplacodonts. " 
(Upper Eocene ancestors of the Oligocene sub- 
family Menodontinae. Heads probably mesa- 
ticephalic; grinding teeth foreshadowing the 
menodont type.) 
Genus Diplacodon Marsh (horns well devel- 
oped) 166,439 

Species eZa^Ms Marsh 166,439 

Genus Eoiitanotherium Peterson (horns well 

developed) 196, 435, 656 

Species osborni Peterson 195, 435, 656 

in. TITANOTHERES OF LOWER OIIGOCENE TIME 

(Face extremely abbreviate) 

Group I. Short-horned: 

9. Subfamily Brontopinae (Manteoceratinae) Osborn. 

" Brontopines. " (Lower Oligocene and possibly 
middle to upper Eocene titanotheres. Progres- 
sively brachycephalio, with short-crowned teeth 
and moderately short feet; horns short, sub- 
oval; incisor teeth persistent, rounded crowns, 
one or two pair; premolars with retarded 
tetartocones.) Pages 

Genus Teleodus Marsh (with three lower 

incisors; basal Oligocene) 227,481 

Species avus Marsh 228, 481 

primitivus (Lambe) 235, 482 



SYSTEMATIC CLASSIFICATION OF THE TITANOTHERES 



249 



Group I. Short-horned — Continued. Pages 

9. Subfamily Brontopinae — Continued. 

Genus Brontops Marsh (with two or one 

lower incisors; lower Oligooene) . 222, 482, 664-676 

Species brachycephalus (Osborn) 231, 

483, 675, 676 

dispar Marsh 223, 488, 664 

robustus Marsh i___ 222, 492, 666 

fangustigenis (Cope) 219,482 

Genus Diploclonus Marsh (with internal 

branching horns; lower Oligocene) 227, 

499, 675-678 

Species ftyleri (Lull) 234, 502, 675 

fbicornutus (Osborn) 231,501 

amplus Marsh 227, 504 

selwynianus (Cope) 225,502 

10. Subfamily Menodontinae ( = ?Diplacodontinae) 
Osborn. "Menodonts." (Lower Oligocene 
and possibly upper Eocene titanotheres. Heads 
of medium width, progressively elongating 
(Menodus) or broadening (Allops); horns 
short, trihedral in section; incisor teeth vesti- 
gial; grinding teeth long-crowned with promi- 
nent cingula; premolars with accelerated 
tetartocones.) 
Genus Menodus Pomel {— Titanolherium 

Leidy) 204, 522, 681 

Species heloceras (Cope) 212, 524, 681 

iorvus (Cope) 210,525 

proutii (Owen, Norwood, and 

Evans) 205,526 

trigonoceras {Cope) 213,528,683 

varians (Marsh) 223,535 

giganieus Pomel 204, 530, 687 

Genus Allops Marsh 224, 506, 678 

Species walcotli Osborn 241, 509 

marshi (Osborn) 233,511,678 

serotinus Marsh 225, 515 

crassicornis Marsh 228,517,679 



Group II. Long-horned: Pages 

11. Subfamily Megaceropinae ( = ?Rhadinorhininae) 

Osborn. "Megaceropines," "symborodonts." 
(Relatively small, long-horned titanotheres, 
possibly descended from Rhadinorhinus. Of 
lower Oligocene age. Horns precociously 
evolved, with little or no connecting crest; 
head mesaticephalic to brachycephalic, oypto- 
oephalic; narrow-lipped; premolars with pre- 
cocious tetartocones; grinding teeth without 
cingulum; vestigial incisor teeth.) 
Genus Megacerops Leidy { = Symborodon 

Cope) (horns rounded, erect) 208,541,691 

Species riggsi Osborn 242, 550 

assiniboiensis Lambe 239, 549 

copei (Osborn) 235, 548 

acer Cope 211, 545 

bucco (Cope) 212, 544 

coloradensis Leidy 208, 544 

?syceras (Cope) 226, 549 

12. Subfamily Brontotheriinae ( = ?Diplacodontinae) 

Osborn. "Brontotheres." (Lower Oligocene 
titanotheres. Primitively dolichocephalic, pro- 
gressively mesaticephalic and brachycephalic, 
slightly cyptocephalic; broad-lipped; very pre- 
cocious development of the horns; accelerated 
development of internal cones of superior pre- 
molars; prominent cingulate incisor teeth in 
males.) 
Genus Broniotherium Marsh (horns