THE LIBRARY OF YORK UNIVERSITY Date Due - York Form — Cooper Graphics Digitized by the Internet Archive in 2014 https://archive.org/details/pleistoceneofmidOOhayo THE PLEISTOCENE OF THE MIDDLE REGION OF NORTH AMERICA AND ITS VERTERRATED ANIMALS Published by the Carnegie Institution of Washington Washington, October, 1924 THE NKW YORK JMH -.U, LENOX A NO iSjuUEH FOUNDATIONS * 1047 L CAENEGIE INSTITUTION OF WASHINGTON Publication No. 322 a Copies of this book OCT 15 1924 JUDD & DETWEILER WASHINGTON, D. C. CONTENTS PAGE Finds of Xenarthra in middle region of North America 1-9 Louisiana 1 Texas 1 Oklahoma 4 Missouri 4 Kansas 5 Colorado 6 Iowa 6 Nebraska 8 Minnesota 9 South Dakota 9 Finds of mastodons in middle region of North America 10-46 Louisiana 10 Texas 12 Arkansas 22 Oklahoma 23 New Mexico 24 Missouri 24 Kansas 33 Colorado 35 Iowa 36 Nebraska 43 Minnesota 44 South Dakota 45 Montana 46 Finds of Elephas boreas in middle region of North America 47-56 Texas 47 Missouri 47 Kansas 48 Iowa 49 Nebraska 55 Minnesota 56 Finds of Elephas columbi in middle region of North America 57-84 Louisiana 57 Texas 57 Oklahoma 66 New Mexico 67 Missouri 69 Kansas 69 Colorado 72 Iowa 74 Nebraska 78 Wyoming 81 South Dakota 82 Montana 83 Alberta 84 Saskatchewan 84 Finds of Elephas imperator in mid- dle region of North Amer- ica 85-101 Texas 85 Oklahoma 94 Kansas 95 Colorado 96 Iowa 97 Nebraska 98 Wyoming 101 Montana 101 Finds of undetermined species of elephants in middle region of North America... 102-121 Louisiana 102 Texas 102 Oklahoma 108 New Mexico 108 Missouri 109 Kansas 110 Colorado Ill Iowa 112 Nebraska 117 Wyoming 118 Minnesota 118 South Dakota 120 North Dakota 120 Montana 120 Saskatchewan 121 Alberta 121 Melville Island, Franklin 121 Finds of Equidse in middle region of North America 122-154 Louisiana 122 Texas 123 Arkansas 134 Oklahoma 135 New Mexico 137 Missouri 137 Kansas 140 Colorado 143 Iowa 145 Nebraska 150 Wyoming 152 South Dakota 153 Montana 153 Alberta 154 Manitoba 154 Finds of Tapirida? in middle region of North America 155-156 Louisiana 155 Texas 155 Oklahoma 156 Finds of peccaries, Tagassuidae, in middle region of North America 157-159 Texas 157 Arkansas 157 Missouri 157 Kansas 157 Iowa 158 Nebraska 159 Finds of camels, Camelidae, in mid- dle region of North Amer- ica 161-169 Texas 161 Oklahoma 163 New Mexico 164 Missouri 164 Kansas 165 'Colorado 166 Iowa 167 Nebraska 168 III CONTENTS. IV PACK Studs of camels, Camelidas, in mid- die region Of North Amer- ica— Continued. South Dakota 168 Montana 16!) Finds of deer, O&OCOtteus, in middle region of North Amer- ica 170-171 Louisiana 170 Texas 170 Arkansas 170 Oklahoma 170 Missouri 171 Kansas 171 Towa 171 Finds of the wapiti, Cervu8, in mid- dle region of North Amer- ica 173-174 Arkansas 173 Oklahoma 173 Missouri 173 Kansas 173 Iowa 174 Finds of reindeer, Rangifer, in mid- dle region of North Amer- ica 176-177 Iowa 176 Minnesota 177 Finds of musk-oxen, Ovibovinrr, in middle region of North America 178-185 Arkansas 178 Oklahoma 178 New Mexico 179 Missouri 179 Kansas 181 Colorado 182 Iowa 182 Nebraska 184 Minnesota 185 Alberta 185 Finds of extinct species of bison in middle region of North America 186-200 Louisiana 186 Texas 186 Oklahoma 190 Missouri 191 Kansas 192 Colorado 195 PAGE Finds of extinct species of bison in middle region of North America — Continued. Iowa 196 Nebraska 198 Minnesota 199 Manitoba 200 Finds of Bison bison in middle re- gion of North America 201-209 Texas 201 Oklahoma 203 Missouri 203 Kansas 203 Colorado 204 Iowa 204 Nebraska 207 Wyoming 208 Minnesota 208 South Dakota 209 Alberta 209 Finds of the giant beaver, Casto- r aide 8, in middle region of North America . . . 210-212 Texas 210 Oklahoma 210 Kansas 210 Iowa 210 Nebraska 211 Minnesota 212 South Dakota 212 The Pleistocene of the middle region of North America and the vertebrate remains occur- ring in it 213-317 Louisiana 213 Texas 221 Arkansas 249 Oklahoma 253 New Mexico 256 Missouri 259 Kansas 266 Colorado 271 Iowa 275 Nebraska 301 Wyoming 305 Minnesota 306 South Dakota 308 North Dakota 310 Montana 311 Manitoba, Saskatchewan, Al- berta 313 ILLUSTRATIONS Plates Map 1. Pleistocene Xenarthra in mid- dle region of North America. 2. Pleistocene mastodons in mid- dle region of North America. 3. Pleistocene mastodons in Texas. 4. Pleistocene mastodons in Iowa. 5. Elephas boreus in middle region of North America. 6. Elephas boreus in Iowa. 7. Elephas eolumbi in middle re- gion of North America. 8. Elephas eolumbi in Texas. 9. Elephas eolumbi in Iowa. 10. Elephas imperator in middle re- gion of North America. 11. Elephas imperator in Texas. 12. Elephas sp. indet. in middle re- gion of North America. 13. Elephas sp. indet. in Texas. 14. Elephas sp. indet. in Iowa. 15. Pleistocene horses in middle re- gion of North America. 16. Pleistocene horses in Texas. 17. Pleistocene horses in Iowa. Map 18. Pleistocene tapirs in middle re- gion of North America. 19. Pleistocene peccaries in middle region of North America. 20. Pleistocene camels in middle re- gion of North America. 21. Pleistocene species of deer, Odo- coileus, in North America. 22. Pleistocene reindeer, Rangifer, in middle region of North America. 23. Pleistocene deer, Cervus, in middle region. 24. Pleistocene musk-oxen in mid- dle region of North America. 25. Extinct bisons in middle region of North America. 26. The existing bison in middle region of North America. 27. Giant beaver, Castoroides, in middle region of North Amer- ica. 28. Geological map of Louisiana. 29. Glacial map of Minnesota. Text-figures page Fig. 1. Geological section in Louisiana 214 2. Geological section across Tule Canyon. Texas 232 3. Geological map of region between Pecos River and Rio Grande 235 4. Map of region along Brazos River near Bryan, Texas 244 5. Part of plastron of box-tortoise, Terrapene impressa 245 V PREFACE. In a previous publication, Carnegie Institution of Washington (No. 322), issued February 24, 1923, the writer dealt with the Pleistocene and its verte- brated animals of the region east of Mississippi Eiver and northward. The present volume relates to the vertebrate palaeontology of the region west of the Mississippi, including the States of New Mexico, Colorado, Wyoming, and Montana and the British Provinces lying northward. The Pleistocene geology of the various sections is dealt with as far as it has seemed likely that the geology might throw light on the palaeontology or that the palaeontology might throw light on the geology. The writer's acknowledgments are due to many institutions and persons. First of all, he acknowledges his debt to the Carnegie Institution of Wash- ington for its support. Two States, Texas and Iowa, have furnished numerous remains of Pleistocene vertebrates. The Pleistocene geology of both States is extremely interesting and their geological surveys are being actively prose- cuted. Dr. E. H. Sellards has read the manuscript on the Pleistocene geology of Texas, with benefit to the writer and to the manuscript. To Dr. W. C. Alden, of the U. S. Geological Survey, Dr. G. F. Kay, state geologist of Iowa, Dr. James H. Lee, assistant state geologist, and Professor A. C. Trow- bridge, the writer's thanks are due for reading portions of the manuscript on the Pleistocene geology of Iowa. Dr. Willis T. Lee, of the IT. S. Geological Survey, perused the pages on which is discussed the geology of New Mexico; and W. A. Johnston, of the Geological Survey of Canada, gave the writer the benefit of his wide knowledge of the glacial geology of Manitoba, Saskatch- ewan, and Alberta. Dr. C. L. Baker, formerly connected with the Geological Survey of Texas, has furnished the writer much information regarding the Pleistocene of Texas. It seems proper that mention should be made here of two errors in the writer's former volume (publication 322 of the Carnegie Institution of Wash- ington) to which his attention has been called. On the base map used to display the distribution of certain vertebrate species and genera, as Map 1, etc., the writer failed to have represented the very old drift lying south of the Wisconsin moraine in Pennsylvania and New Jersey. This old drift is, however, represented in figure 10, on page 308, and on Map 6A. A more serious error occurs on Map 38. In northwestern Illinois the Wis- consin drift is represented as continuing to the Mississippi, from the middle of Henry County northward. The Illinoian drift occupies this region from the Mississippi and the driftless area to the Bloomington moraine. The drift there ought to have been represented by small crosses. A correct statement regarding the drift of the area in question is made on page 335 of the volume referred to. Oliver P. Hay. October 1, 1924. VII THE PLEISTOCENE OF THE MIDDLE REGION OF NORTH AMERICA AND ITS VERTEBRATED ANIMALS. FINDS OF XENARTHKA IN THE MIDDLE REGION OF NORTH AMERICA. LOUISIANA. (Map 1.) (Petite Arise, Iberia Parish (1).* — In the salt mines at this place have been discovered many remains of Mylodon, and a claw of Megalonyx has been reported. Mylodon appears to have been first announced by Doctor Leidy (Proc. Acad. Nat. Sci. Phila., 1884, p. 22). In 1889 (Trans. Wagner Inst., vol. ii, p. 33, pi. v, figs. 1-4), he described and figured remains of Mylodon harlani. From the Joor collection, made in 1890, Cope, in 1895 (Proc. Amer. Philos. Soc, vol. xxxiv, p. 458), described two new species, M. sulcidens and M. renidens. Since that time, Stock (Bull. Dept. Geol. Univ. Calif., vol. vm, p. 319), who has had abundant materials obtained from the asphalt deposits at La Brea, California, has concluded that Cope's species were not well founded and are variations of Mylodon harlani. The materials collected by Joor belong to Tulane University, New Orleans, where the writer was permitted to see such portions as were accessible at that time. In 1895, Dr. Joseph F. Joor (Amer. Naturalist, vol. xxix, p. 397) stated that, at an earlier date, Captain Dudley Avery, of Petite Anse, had sent to the Smithsonian Institution a claw found at the salt mines. This was identi- fied as belonging to Megalonyx. TEXAS. (Map 1.) Wheeler County (1). — The U. S. National Museum contains a skull of a species of Nothrotherium, which was secured from Baylor University, having been presented by some one now unknown. It is reported to have been found, at a depth of 40 feet, in digging a well somewhere in Wheeler County. The skull has been described by the writer (Proc. U. S. Nat. Mus., vol. Li, p. 107, pis. vi, vn) under the name Nothrotherium texanum. The animal undoubt- edly belonged to the Pleistocene and to an early stage. Rock Creek Canyon, Briscoe County (2). — In 1893 (4th Ann. Kep. 'Geol. Surv. Texas, p. 78), Cope announced that he had secured a part of a femur of a species of Mylodon from the Equus beds of the Staked Plains of Texas. This was certainly at Rock Creek, or thereabout. The bone was referred with doubt to Mylodon sodalis. In 1915 (Amer. Jour. Sci., vol. xxxix, pp. 614, 616), Troxell reported that the Yale expedition into the Panhandle region of Texas had secured remains of Mylodon harlani at the head of Rock Creek * Numbers in parentheses after the italic headings indicate the numbers on the map, and point out the provenance of the specimens. 1 2 XENAKTHRA. Canyon. Doctor Troxell informed the writer that the locality is in the south- western quarter of section 207. It is probable that the femur secured by ("ope belonged to M. harlani. A list of the species of vertebrate fossils ob- tained in the Rock Creek region will be found on page 239. Professor Lull, in 1915 (Amer. Jour. Sci. ser. 4, vol. xxxix, p. 328), re- printed TroxelPs list of vertebrates from the head of Rock Creek and stated that at least four individuals of Mylodon harlani had been found. He re- vised the species and described and figured many parts of the skeleton. Terlingua, Brewster County (3). — In 1917 (Bull. Dept. Geol. Univ. Calif., vol. x, p. 138) Stock stated that bones of Nothrotherium had been sent to Dr. J. C. Merriam from this place. With these bones were some sparse re- mains of a fossil horse (p. 134). Hawley, Jones County (12). — From Dr. Mark Francis the writer received, for examination, about 15 plates of the carapace of a glyptodon, which were found in Hawley. These had been sent to Doctor Francis by R. J. Magee. of the same place. They appear to have belonged to the species Glyptodon petaliferus, but the animal seems to have been larger than the one mentioned below. One plate has a diameter of 60 mm. and a thickness of 25 mm. ; another has a diameter of 50 mm. and a thickness of 44 mm. Naturally, the specific identification is doubtful. Some of these plates were sent to the U. S. National Museum. All were found in digging a well and at a depth of 56 feet. It is possible that a Pliocene deposit had been reached. Wolfe City, Hunt County (4). — The U. S. National Museum has part of al skeleton of a species of Glyptodon, found near Wolf City. It has been described by the writer (Proc. U. S. Nat. Mus., vol. Li, p. 107, pis. iii-v) under the name of Glyptodon petaliferus Cope. The finder, 0. S. Shelton, states that the skeleton was found in 1908 on the bank of Middle Sulphur Creek, about 9 feet from the surface. The bones lay on a bed of gravel and were overlain with clay. Quite certainly the deposits belong to the Sheridan or Equus beds. Near Quintan, Hunt County (15). — From D. A. Saunders, of Greenville, Hunt County, the U. S. National Museum received in 1923 a well-preserved proximal phalange of the left index digit of Megalonyx jeffersonii, found about 3 miles north of Quinlan, in a locality described as "locality No. 3, some 12 miles south and 2 miles east of Greenville, on the H. T. Weathers farm, in the bed of a nameless creek, midway between the Wieland road and the Weathers oil-well/' The bone is slightly smaller than that described and figured by Leidy (Smitbson. Contrib. Knowl., vol. vn, p. 32, pi. x, figs. 15, 16, 17). The length is 3 inches and the other dimensions are in proportion. At the same locality were found fragmentary and water-worn teeth of an undetermined species of horse. Speegleville, McLennan County (5). — The collection of Baylor Univer- sity, Waco, Texas, contains the distal end of a femur of Mylodon, found by J. K. Prather near Hog Creek, 3 miles northwest from Speegleville. The animal probably belonged to M. harlani. Temple, Bell County (14). — Some unimportant remains seen by the writer, belonging to a species of Mylodon, were unearthed by W. F. McGregor in the TEXAS. 3 gravel-pits along Leon Kiver. A list of the associated species is given on page 228. Bastrop, Bastrop County (13).— In October 1922, Mr. Hugh H. Duval sent to the U. S. National Museum part of a bone of a ground-sloth, which was identified by J. W. Gidley as that of a Mylodon. With this was a hind- most molar of Mammut americanum. The Mylodon bone and mastodon tooth were found at a depth of 6 feet in a gravel-pit on the second terrace at Bastrop. This terrace is about 50 feet above ordinary low water. Pittbridge, Burleson County (6). — In July 1899, Mr. Dansby located near Pittbridge a part of the left ramus of the lower jaw of a megatherium found in the bank of Brazos River just after a time of high water. The jaw presents the base of the hinder, or fifth, tooth, the inner faces of the two teeth next in front, and a little of the socket of the second tooth. The following measure- ments are given in millimeters : Front of socket of second tooth to rear of fifth 143.0 Hinder tooth, fore-and-aft diameter 43 . 5 Hinder tooth, transverse diameter 37 . 5 Fourth tooth, fore-and-aft diameter 43.0 Fourth tooth, transverse diameter 47.0 Third tooth, fore-and-aft diameter 39.0 Depth of socket of fifth tooth 100.0 Near this jaw was found the lower end of the radius of Megatherium. In February 1921, Doctor Francis sent the writer a part of a dermal scute which was recognized as having belonged to a species of Clilamijtherium. In March he sent photographs of a right ramus of a lower jaw which had been found at Pittbridge. A comparison of this with figures of the one described by Dr. E. H. Sellards, from Florida, showed that the Texas specimen belonged to C. septentrionale. The bone is complete, but all of the teeth except two are missing. See Journal Mammalogy (vol. 3, Feb. 1922, p. 22, pi. n) for an account of the jaw, by Professor A. R. Cahn. Hidalgo Falls, Washington and Brazos Counties (7). — B. F. Shumard, the geologist, reported (Trans. St. Louis Acad. Sci., vol. n, p. 141) that remains of Megalonyx had been found at this place, associated with Mastodon, Elephas, Equus, Crocodilus (probably Alligator), and Testudo. The geology of the locality is discussed on page 245. Sour Lake, Hardin County (8). — In 1868, Leidy (Proc. Acad. Nat. Sci. Phila., vol. xx, p. 174) reported the discovery here of a fragment of a tooth of Megalonyx validus. For a list of other vertebrates and a discussion of the geology, see page 242. San Felipe, Austin County (9). — W. M. Carpenter, in 1846 (Amer. Jour. Sci., vol. i. p. 249), reported the finding at this place of two claws of an animal belonging to the Me gather iidce. He thought they might belong to Oryctotherium, which is now regarded as a synonym of Mylodon. Leidy (Jour. Acad. Nat. Sci. Phila., vol. vn, p. 413) regarded the animal as prob- ably a Mylodon. For a list of the species found at this locality, see page 246. "Fifty miles above Mouth of Brazos River," Fort Bend County (10). — In 1856 (Proc. Bost. Soc. Nat. Hist., vol. VI, pp. 51, 53), Wyman gave an ac- count of fossil remains found about 50 miles above the mouth of Brazos River and presented to the society. The fossils included 2 molars of Mammut ameri- 4 X i:\ ARTHRA. canum, a tibia of Megatherium and a molar and symphysis of the lower jaw, and a femur of an elephant. . A deseription and measurements of the tibia were given. These remains were found in the bed of Brazos River at a stage of low water. It is unfortunate that the locality was not more exactly re- ported. We do not know whether the distance was taken in a straight line or along the windings of the river. In either case the bones were probably located within the area occupied by the Beaumont formation. One can not now be certain whether or not the river had cut down through this into the older Lissie. Nueces County (11). — In 1888 (Amer. Naturalist, vol. XXII, p. 345), Cope described a species of Glyplodon as G. petaliferus, found in Nueces County by William Taylor. The specimen which formed the type was only half of one of the plates which compose the armor of the animal. This was figured in 1889 (Amer. Naturalist, vol. xxm, p. 662). The half scute had been found in the beds which furnished Equus crenidens (= E. giganteus) and E. harcencei (— E. tau) ; but the exact locality is not known. For (12) see page 2; for (13) page 3; for (14) and (15) page 2. OKLAHOMA. (Map 1.) Mulhall, Logan County (1). — In 1917 (Bull. Geol. Soc. Amer., vol. xxvin, p. 212), Troxell stated that he had found remains of a giant sloth in a de- posit near Mulhall. Nothing has been published on the collection since that time. For the associated genera, see page 255. MISSOURI. (Map 1.) Pomme de Terre River, Benton or Hickory County (1). — In 1843, Harlan (Amer. Jour. Sci., vol. xliv, p. 69, pis. i-iii) described the remains of a large ground-sloth, to which he gave the name Orycterotherium missouriense. This was afterwards found to be the same animal as Mylodon harlani. The bones formed a part of the collection which Albert Koch stated he had gathered on the Pomme de Terre River in Benton County. The position of this lo- cality, which appears to have been possibly in Hickory County, is discussed on page 25. Statements by other observers are referred to in connection with Koch's discoveries in Gasconade County, which call in question the correct- ness of some of his utterances. The materials which Harlan described con- sisted of 24 teeth, fragments of 2 maxillae, 2 mandibles, 2 humeri, an ulna, 2 radii, 5 wrist-bones, 4 claw-phalanges, 2 tibia?, an astragalus, parts of 2 or 3 clavicles, parts of ribs, 4 elements of the sternum, and a considerable part of the pelvis. The writer does not know what has become of these bones. Harlan stated that this collection included remains of the ox (probably Bison) , deer, elk, and Megatherium. The last-named genus occurs in South Carolina, Georgia, Texas, and New Jersey. It may. of course, have made its way north into Missouri ; but it would be interesting to have additional materials from this State. Its occurrence on the Pomme de Terre was not mentioned by Leidy in his monograph (Smithson. Cont. Knowl., vol. VII., KANSAS. 5 art. 3). The discovery of Mylodon at the locality mentioned suggests that a part of the deposits belonged to the middle or early part of the Pleistocene. Unfortunately we do not know on what parts of each species the determina- tions were based. Kimmswick, Jefferson County (2).— Professor W. C. Mills, of Ohio State University, has shown the writer two teeth of Megalonyx jeffersonii, which he obtained at Kimmswick. Both of the teeth belong to the upper jaw, one being the front tooth, the other the third or fourth. The front tooth had a height of 61 mm., a fore-and-aft length of 39 mm., and a thickness of 17.5 mm. The other tooth was 51- mm. high, 27 mm. wide, and 18 mm. fore-and-aft. Mastodon bones in great numbers have been found here, probably elephant re- mains; also a horse tooth, and a horn-core resembling that of Bodtherkim bombifrons. The locality had probably been a resort for various species dur- ing a large part of the Pleistocene epoch. Although it has been worked for specimens by various people, nobody appears to have thought it worth while to keep accurate records. From Dr. W. F. Parks, of St. Louis, Missouri, the U. S. National Museum, in July, 1921, received 8 upper teeth of Megalonyx jeffersonii, found at Kimmswick, including both of the large anterior teeth. It is probable that the teeth did not all belong to the same individual. A list of species from this locality is given on page 263. KANSAS. (Map 1.) Seneca, Nemaha County (1). — In 1895 (Kansas Univ. Quart., vol. in, p. 175, figs. 1, 2), Williston figured and described a xenarthran fibula which had been found, as reported, in a well dug in Seneca. The bone was referred with doubt to Mylodon harlani. It was again mentioned by Williston in 1897 (Univ. Geol. Surv. Kansas, vol. n, p. 304, figs. 12, 13) as Mylodon. Tn 1913 (Mem. Mus. Comp. Zool., Harvard, vol. xl, p. 339, pi. iv, fig. 21), Dr. G. M. Allen described and figured the fibula of his Mylodon garmani, found somewhere in the region of Hay Springs, Nebraska. He called atten- tion to the bone described by Williston and showed that it was about 25 mm. longer than the largest fibula of M. garmani. However, there was a differ- ence of 31 mm. between the right and left fibulas of the type specimen of Allen's species. At any rate, it is impossible to say at present to what species the Seneca specimen belonged. This fibula is different from that of Mega- lonyx. Williston reported that the bone had been discovered at a depth of 30 feet below the surface. This locality is within the area of the Kansan drift, and it seems probable that this drift sheet had been penetrated and that the Aftonian had been reached. Harper Township, McPherson County (2). — In 1892 (Trans. Amer. Philos. Soc, vol. xvii, pp. 1-10, pis. i-v), Josua Lindahl described a fine skull of a species of Megalonyx, which he named M. leidyi. This was found in a sand- pit in Harper Township. The writer has not been able to ascertain the lo- cality more definitely. The skull is now in the collection of Bethany College, 6 XENARTHRA. Lindborg, Kansas, where the writer has seen it. Aceording to statements made, a tooth of a horse was found at the same point. The locality is within the area described by Haworth and Beede (Univ. Kansas Geol. Surv., vol. u, pp. 285-296) as belonging to the Equus (Sheridan) beds. Meade (or Clark) County (3). — Professor F. W. Cragin, in his account of the geology of Meade and Clark Counties (Colorado Coll. Studies, vol. VI, p. 53), reported that the Meade gravels contained, among other species, Mega- lonyx leidyi. The writer has studied Cragin's collection, but did not find anj remains of this species. He found, however, teeth which do not seem to differ from those of Mylodon harlani. For a discussion of the character and age of the deposits and a list of the species found, see page 270. COLORADO. (Map 1.) Walsenburg, Huerfano County (1). — In 1909 (Univ. of Colorado Studies, vol. vi, p. 309, pis. I, n), Dr. T. D. A. Cockerell described a fine skull of a ground-sloth, which he referred to Mylodon, but did not determine specific- ally. This skull was found by E. A. Lidle on his farm one mile south of Walsenburg. It was presented to the University of Colorado, where it still remains. This specimen, as well as the one from Nebraska, described by Barnum Brown as Paramylodon nebrascensis, is referred by Dr. Chester Stock to Mylodon harlani. IOWA. (Map 1.) Mills County (1). — At the Cleveland meeting of the American Association for the Advancement of Science, in 1888, Professor J. E. Todd exhibited a claw phalange of a large mammal which had been found in this county. It was afterward determined by Doctor Leidy to be a claw of Megalonyx. The claw was discovered in a deposit of sand which underlies the drift; the exact place where it was found has not been given. Calvin (Bull. Geol. Soc. Amer., vol. xx, p. 353) stated that these sands belong to the Aftonian interglacial. It is impossible to say to which species of Megalonyx this specimen belongs. Avoca, Pottawattamie County (7). — In the collection of fossil vertebrates in the University of Iowa, with catalogue No. 396, is the distal two-thirds of the right humerus of Megalonyx jeffersonii, which was found by Eay Leonard in the dump of a dredging-machine at work in the East Branch of Nishnabotna River, in the southeast quarter of northeast quarter of section 10, township 77 north, range 39 west. This is near Avoca. Missouri Valley, Harrison County (2). — From the Cox gravel-pit at Mis- souri Valley, Calvin reported a damaged right tibia, which he referred to Megalonyx jeffersonii. The writer published a description of this bone in 1914 (Iowa Geol. Surv., vol. xxm, p. 130, fig. 30). If the section taken at the middle of the length is compared with that of the species just named (op. cit., p. 123, fig. 20), it will be seen that they are very different and probably do not belong to the same genus; nor does the tibia belong to Mylodon. From this pit was obtained a large ungual phalange of Mylodon harlani, later figured and described by Calvin (Bull. Geol. Soc. Amer., vol. xx. IOWA. 7 p, 353, pi. xxvi) and the writer (Iowa Geol. Surv., vol. xxm, p. 142, pi. viii, figs. 1, 2, text-figs. 40-42). From the same pit was obtained by Professor Shimek a lumbar vertebra, which appears to belong to a species of Megalonyx. This has been described by the writer (op. cit., p. 130, pi. vi, figs. 1, 2). On page 296 the location and geology of this pit are described, and a list of the species of mammals found there, about 17 in number, is given. Turin, Monona County (3). — In the Elliott gravel pit was found a com- plete right radius of Megalonyx. This appears not to differ from that of M. jeffersonii. It has been described and figured by Calvin (Bull. Geol. Soc. Amer., vol. xxn, p. 215, pi. xxiv) and the writer (Iowa Geol. Surv., vol. xxm, pi. figs. 3, 4). The deposits at Turin in which vertebrate remains were found are regarded as belonging to the Aftonian. For a brief descrip- tion of the pit and a list of the species obtained, reference is made to page 297. Sioux City, Woodbury County (4). — The Aftonian deposits at this place have furnished various remains of Megalonyx. An ungual phalange now at the State University of Iowa was secured by Professor Shimek at the Ander- son pit, at North Eiverside, near Sioux City. This was described and figured by the writer (op. cit., p. 128, pi. v, figs. 1, 2, text-figs. 25-27) as M. jeffersonii. A bone found in this pit and identified as possibly a patella of Megalonyx (Iowa Geol. Surv., vol. xxnr, p. 129) is certainly not a patella. A descrip- tion of the pit and a list of the species of vertebrates collected from it are given on page 298. Akron, Plymouth County (5). — From the Jensen well, near Akron, there was obtained the ankylosed first and second phalanges of the third digit of the hinder foot of a species of ground-sloth, probably Megalonyx. This is the well from which were collected the molars and some other remains of St ego- mastodon mirificus (Mammut mirificum). The well is noticed here on page 299. Dubuque, Dubuque County (6). — In 1862 (Geol. Surv. Wis., vol. I, p. 135), J. D. Whitney reported that some peccary-teeth of apparently the liv- ing species had been found associated with teeth of an undetermined species of Megalonyx, embedded in clay, at a depth of 10 feet from the surface, in a flat crevice in the Galena limestone near Dubuque. Dubuque is situated in the driftless region and hence we have no guidance from the drift-sheets. Whether the animals of the lead crevices belong to the close of the Wisconsin stage or to an earlier one is not yet determined. The association of the pee-' cary and the megalonyx testifies to the prevalence of a mild climate when they lived there. For (7) see page 6. Wayland, Henry County (8).— In the palaeontological collection of the Uni- versity of Iowa, with catalogue No. 398, is a right tibia of Megalonyx jeffer- sonii, which was found somewhere near Wayland. It is reported to have been discovered in the bed of Sugar Creek, north of Wayland. The locality may therefore be in the southern part of Washington County. The stage of the Pleistocene to which it belongs is not known. 8 XENAKTHRA. NEBRASKA. (Map 1.) Tecumseh, Johnson County (1). — The State Museum at the University of Nebraska contains an ungual phalange which is identified as having belonged to some species of the genus Mylodon. The length of the bone is 140 mm. It is not known exactly where this claw-bone was found. The town is on Nemaha River and within the area of the Kansan drift. It is safe to assume that it was found in an Aftonian deposit, especially as from this same county, a tooth of Elephas imperator has been reported. Red Cloud, Webster County (2). — The State Museum at the University of Nebraska, at Lincoln, has a large part of a left humerus of a species of Mylodon, found near Eed Cloud. The more exact location is given as the Rocky Dell farm, belonging to the Amboy Milling Company, situated 3 miles east and 1 mile north of the town named. The bone was secured by Pro- fessor L. T. Frisbee, principal of the public school at University Place, near Lincoln, who presented it to the university. It is reported to have been found at a depth of 5 feet from the surface. A portion of the shaft of this humerus is missing. The head and trochanters form one fragment. The lower fragment has a length of 285 mm. The humerus as restored is 520 mm. long, but this may be more or less inexact. The following measure- ments (in millimeters) are given; the corresponding measurements were ob- tained by courtesy of Dr. J. C. Merriam, from a specimen of Mylodon harlani at the University of California. Mylodon Red Cloud harlani. mylodon. Total length of humerus 430 520 ± Distance through head and trochanter 152 150 Width of bone through epicondyles 250 260 Width of ulno-radial articular surface 127 150 The measurements appear to indicate that the Red Cloud humerus has been made considerably too long in the restoration. If the lengths of the two bones compared above were in the same ratio as the epicondylar widths, the Red Cloud humerus would have a length of about 450 mm. It is possible that the latter-mentioned bone belonged to the species of Mylodon described by Dr. G. M. Allen as M. garniani. In this species the total length of the humerus was about 430 mm. and the width of the lower articular surface 125 mm. The width across the condyles is not given. Red Cloud is situated on Republican River and far outside of the glaciated area. It appears probable that the bone above described should be referred to the Sheridan epoch. Grayson, Sheridan County (3). — Somewhere in the region about Hay Springs, probably near Grayson, Professor Samuel Garman, in 1880, secured, for Harvard University, a well-preserved skull and many parts of the skeleton of a species of Mylodon. This specimen was described in 1913 (Mem. Mus. Comp. Zool., Harvard, vol. xl, pp. 319-346, pis. 1-4) by Dr. Glover M. Allen under the name of Mylodon garmani. Besides the skull and the lower jaw, NEBRASKA — MINNESOTA — SOUTH DAKOTA. 9 there are preserved more or less completely the cervical vertebrae, 16 dorsals, a portion of the pelvis, fragments of ribs, most of the sternum, parts of both scapulas, a humerus, a part of an ulna, a radius, and some foot-bones, includ- ing several ungual phalanges. The American Museum of Natural History at New York (fide Matthew in 1902 Bull. Amer. Mus. Nat. Hist., vol. xvi, p. 317), contains a complete skull and jaw and a large part of the skeleton of a sloth allied to but distinct from M. harlani. This collection was de- scribed in 1903 by Barnum Brown (Bull. Amer. Mus. Nat. Hist., vol. xix, pp. 569-583, pis. l, li) under the name Paramylodon nebrascensis, but the characters supposed to distinguish the genus Paramylodon are sometimes found in Mylodon harlani (Stock, C, Bull. Dept. Geol., vol. vm, p. 319), and Stock regards the species as identical with M . harlani. Boxbutte Creel, Sheridan County (4). — In 1918 (Bull. Amer. Mus. Nat. Hist. vol. xxxvm, p. 229), Dr. W. D. Matthew reported the discovery, at the mouth of Boxbutte Creek, of a few bones which he referred provisionally to Megalonyx leidyi. MINNESOTA. (Maps 1, 29.) Minneapolis, Hennepin County (1). — From Professor Charles E. Johnson, of the University of Minnesota, the writer received, for examination, a caudal vertebra of a xenarthrid which in all probability is Megalonyx. The bone was found in a gravel-bed within the limits of Minneapolis, at a depth of 25 feet. It can not be compared directly with the caudals of Megalonyx, because these are not known. It does not belong to Mylodon or any other ground-sloth whose caudals are known. Found in such a situation and deposit, it certainly lived after the retirement of the Wisconsin glacier; and this fact makes it likely that it was a species of Megalonyx, probably M. jeffersonii. SOUTH DAKOTA. (Map 1.) Philip, Stanley County (1). — In the Morrill collection of the University of Nebraska is a fine complete femur of a Megalonyx, not improbably Mega- lonyx jeffersonii. This was found near Philip in 1908 and sent to the mu- seum of the University of Nebraska in 1910. The number given it in the catalogue is 47-16-6-10. No details are known to the writer regarding the finding of the specimen. Philip is situated on Bad River, in the northern part of township 1 north, range 22 east. FINDS OF MASTODONS IN THE MIDDLE REGION OF NORTH AMERICA. LOUISIANA. (Maps 2, 28.) Head of Bistineau Lake, Webster Parish (1). — An account of the Bistineau salt works has been given by Veatch (Geol. Surv. La., pt. vi, 1902, p. 81). The location is in sections 25, 26, 35, and 36, township 18 north, range 10 west, and therefore near the former head of Bistineau Lake. One of the wells is known as the Frenchman's. Out of this well (Veatch, op. cit., p. 8, pi. xxn ) were obtained at one time "several vertebrae, the leg-bones, and a por- tion of the tusk of some large animal, probably a mastodon." There is noth- ing to show that these bones did not belong to one of the extinct elephants. Castor, Bienville Parish (2). — Xear this place are the salt works known as King's, in section 35, township 15 north, range 8 west. This locality is near Castor Bayou (Harris and Veatch, Geol. Surv. La., pt. v, pp. 63, 123; Veatch, Geol. Surv. La., pt. VI, 1902, p. 79). There is no assurance that any mastodon remains have been found here; but certainly a tusk of some one of the proboscideans was met with. Veatch quotes statements made to him about the discovery of horns and bones of some very large animal. One object was supposed to be a true horn and not a tusk, inasmuch as it was hollow and retained its horny laminations and its odor. The fragment was 3 feet long and 13 inches in diameter. It is far more probable that it was the basal end of a tusk, showing the pulp-cavity, which on drying separated into its layers. Xor is it probable that the sheath of a bison horn would be preserved or that it would have the diameter given. Rayburn's Salt Works, Bienville Parish (3). — These salt works are in sec- tion 31, township 15 north, range 5 west (Harris and Veatch, Geol. Surv. La., pt. v, 1899, pp. 52, 122). The works are about 10 miles south-southeast of the town of Bienville and on Brush Creek. A description of the locality will be found on page 220. These mastodon bones were reported by F. V. Hopkins in 1872 (3d Ann. Rep. Geol. Surv. La., p. 528). They are said to have been found in the laminated clay. Price's Salt Works, Winn Parish (4). — Price's salt works are in section 25, township 13 north, range 5 west, about 12 miles in a direct line south- southeast from Rayburn's works (Harris and Veatch, as cited above, p. 55). The Price works are on a branch of Dugdemona Creek. A brief description of the geology of the locality will be given in the discussion of the Pleistocene geology of Louisiana. Mastodon bones were reported from this place by F. V. Hopkins in his report of 1871 (1872), cited above. Veatch (Geol. Surv. La., pt. vi, 1902, p. 68) stated that "big bones" had been reported from a well just west of Rayburn's furnace, at a depth of from 8 to 10 feet. Un- doubtedly bones of proboscideans have been found here, but there is no cer- tainty that they belonged to the mastodon. Drake's Salt Works, Winn Parish (4). — According to Harris and Veatch (as cited above, pp. 55, 121), the location of these salt works is in section 21, township 12 north, range 5 west, 6 miles southwest of Price's and near 10 LOUISIANA. 11 Saline Bayou. We have the same authority for the occurrence of mastodon remains here as for those at the salt works mentioned above. There is no certainty in any of these cases that the remains belonged to the mastodon and not to one or both of the elephants which might be found there. Opelousas, St. Landry Parish (5). — Goddard (Amer. Nat. Hist., vol. n, p. 248) stated that Doctor Griffin, in 1804, saw mastodon bones exhumed somewhere in the vicinity of Opelousas. W. M. Carpenter, the geologist, in 1839 (Amer. Jour. Sci., vol. xxxv, p. 345), reported finding a skull and other remains of a mastodon about a mile east of this town; only some teeth were saved, their present whereabouts unknown. Petite Arise, Iberia Parish (6). — This place is in the southeastern corner of township 13 south, range 5 east, extending into the next township east. It is distant about 10 miles south-southwest from New Iberia and 3 miles from the north shore of Vermillion Bay. There is here an extensive deposit of salt, which was worked by the Indians even in prehistoric times. Appar- ently the first notice published of the finding of fossils here was by Doctor Leidy, in 1866 (Proc. Acad. Nat. Sci. Phila., 1866, p. 109). He stated, on the authority of T. F. Cleu, that a piece of Indian basket-work had been found lying on the rock salt and 2 feet beneath the remains of an elephant. How it was known that the remains belonged to an elephant rather than to a mastodon was not explained. In 1883 a collection of fossil bones from Petite Anse was sent to the Smitn- sonian Institution by William Crooks, of the company then exploiting the salt mines. These fossils were first described briefly by Leidy (Proc. Acad. Nat. Sci. Phila., 1884, p. 22), but more fully in 1889 (Trans. Wagner Inst., vol. II, pp. 33-39, figs. 1-4). Among other species he recognized Mammut americanum, it being represented by two teeth, fragments of vertebras, and some other bones. The teeth have been examined by the present writer in the collection of the U. S. National Museum catalogue Nos. 213, 214. One is an upper last molar with 5 cross-crests; the other a lower left second true molar. Leidy regarded both these teeth as being lower molars. In the collec- tion of Doctor Joor (Amer. Naturalist, vol. xxix, p. 397), 2 teeth of masto- don were recognized. He referred an atlas and some other bones to the mas- todon with doubt. The other species found here are listed on page 215. Cote Blanche, St. Mary's Parish (7). — About the middle of the north shore of Cote Blanche Bay are some salt deposits and salt works. Veatch (Geol. Surv. La., pt. v, 1899 (1900), p. 230) gives the location of these as being in township 15 south, range 5 east. This would be in Vermillion Bay. The range appears to be really 7 east. In 1872, F. V. Hopkins (Ann. Rep. Geol. Surv. La. for 1871, p. 188) stated that mastodon bones were common in that region. Here again, how- ever, there does not appear to be either description or any specimens pre- served to confirm the identifications. Hopkins reported that these bones were found in deposits with fresh-water shells, Paludina, Melania, Unio, and Cyclas. Hilgard (Smithson. Contrib. Knowl., vol. xxn, No. 248, p. 12) stated that mastodon bones had been found on the island, but he could not ascertain precisely the place, and he probably did not see the bones. 12 MASTODONS. Little Bayou Sara, West Feliciana Parish (8). — \V. M. Carpenter in 1838 (Amer. Jour. Sci., vol. XXXIY, p. 202) reported that mastodon remains had been found along this small stream, which is situated between Bayou Sara and Mississippi River. He stated that the place was about 25 miles from Jackson. It seems, therefore, that the locality must have been near the north- ern border of the parish. One tooth was said to resemble a tooth of Mammui americanum, which is figured by Cuvier. The identification was doubtless correct and the statement that the tooth weighed Qy2 pounds is reasonable. About a mile away was found a part of the jaw of what was supposed to be a young mastodon, with two unworn teeth in it. The whole weighed 8 pounds. About 2 miles from the locality of the first tooth another was found, which weighed 12 pounds. The four cross-crests had been worn down smooth. Cer- tainly no mastodon tooth weighed that much, especially after it had suffered so much wear. It is equally certain that the tooth was from an elephant. In immediate association with this tooth was found one of E quits compile at us, described in its proper place. Carpenter stated that these teeth had been deposited in the College of Louisiana. Port Hudson, East Baton Rouge Parish (9). — Hilgard, in 1872 (Smithson. Contrib. Knowl., vol. xxiii, Xo. 248, p. 5), presented a geological section found along Mississippi River, at Port Hudson. The same section and state- ment were published in 1869 (Amer. Jour. Sci., vol. xlvii, p. 79). In one stratum, composed of miscellaneous fluviatile materials, at a depth of about 35 or 40 feet, bones of a mastodon were found. Why these were thus identi- fied is not explained. They may have been bones of an elephant. Alsworth's, East Feliciana Parish (10). — At this place, 6 miles above Baton Rouge, mastodon remains have been found, according to Hopkins (Ann. Rep. Geol. Surv. La. for 1871 (1872), p. 188) ; but the information is not definite. Baton Rouge, East Baton Rouge Parish (11). — From D. R. Weller, presi- dent of the Standard Oil company of Louisiana, through the kindness of Pro- fessor W. H. Gates, of Louisiana State University, the IT. S. National Museum has received some fragments of bones and teeth of Mammut americanum, ac- companied by a tooth of Equus complicatus and some fossil wood. These re- mains were found about 2 miles north of the city and at a depth of 8 feet. The deposit belongs to the Port Hudson. TEXAS. (Maps 2, 3.) Tascosa, Oldham County (50). — Doctor Francis informed the writer that in 1915 J. D. Cooper, of Tascosa, sent him fragmentary bones of an elephant and that among these were some pieces of mastodon teeth. Palo Duro Canyon, Randall Comity (1). — Cummins (3d Ann. Rep. Geol. Surv. Texas, p. 149) reported that he found teeth of a fossil horse and of a mastodon on the south side of the canyon, opposite the falls. Benjamin, Knox County (2). — Cummins (4th Ann. Rep. Geol. Texas, pp. 182, 183) reported that he found, at Mr. Green's place, 14 miles east of Ben- jamin, a few fragments of mastodon and of Equus, in deposits which he re- TEXAS. 13 ferred to his Seymour beds. A large femur, over 4 feet long, had previously been found at the same place. Knox City, Knox County (3).— C. H. Gordon (Water-supply Paper 317, p. 31) stated that he had seen a mastodon tooth from a well at Knox City at a depth of 40 feet. A femur of a proboscidean was found in a ravine in the vicinity. Gainesville, Cooke County (4). — K. T. Hill, in his report of 1901 (21st Ann. Eep. U. S. Geol. Surv., pt. 7, p. 360), stated that remains of the mas- todon had been found in surficial deposits at Gainesville. Dallas, Dallas County (5).— Doctor Francis states that a tooth of a masto- don was found within the city limits of Dallas some months before May 27, 1913. Probably the same discovery was reported by the Dallas News, Decem- ber 4, 1912. The tooth was found on the site of the building of that news- paper, at a depth of 23 feet. In the autumn of 1923, Professor E. W. Shuler informed the writer that a jaw of a horse and a fine lower jaw of a mastodon had been discovered in a new Lagow gravel-pit, a mile north of the old one and within the city limits. In the old Lagow gravel-pit were found the species recorded on page 241. The writer has seen in the collection of Doctor Francis a palate of Mammut americanum, which was found south of Dallas in the fall of 1920. This palate contains the two last molars of the right side. They are white and thoroughly fossilized, and show little wear; the roots are in fine condition. The second molar is 120 mm. long and 90 mm. wide. The third is 195 mm. long and 105 wide. It has 5 cross-crests and in the first and second valleys there is much cement; only a little in the third valley. The tusk of this specimen was secured by Doctor Francis, and was restored for him by the Ward Establishment at Eochester, New York. Following the convexity of the curve the length is 8.8 feet. The distance from the base to the tip in a straight line is 7.3 feet. The circumference at the base is 21 inches; at the middle of the length, 18 inches. The tusk is somewhat spirally curved. These remains were found in a gravel-and-sand pit about 5 miles south of Dallas, in a terrace situated about 50 feet above the river. The pit is on the farm of Drs. A. E. and G. R. Flowers, and just west of the Missouri, Kansas & Texas Railway. In the last gravel-pit was found, in January, 1923, a tusk 8 feet long and a lower jaw of Elephas columbi (p. 58). Doctor Francis has reported the possession of a mastodon tooth found in February 1923, in Zang's gravel-pit, at the west end of the viaduct in Dallas. Wilmer, Dallas County (6). — Doctor Mark Francis reported that, in 1922, a tooth of a mastodon was found near Wilmer, on the farm of Mrs. M. A. Tresp, under a bridge over a stream. This was possibly Trinity River, which is not far away. The tooth was broken and soon began to crumble. Fort Worth, Tarrant County (7). — Mrs. Jennie Scheuber, of the Carnegie Public Library in Fort Worth, some years ago had an upper left hindmost molar of a mastodon which was found in the vicinity of Fort Worth and which belongs to Anancus gratus. This tooth is worn only on the first crest and there slightly. The length is 200 mm.; the width at the first crest is 95 mm. and it increases slightly to the fourth, 97 mm. There are 4 cross-crests and a rather large heel. On the inner ends of the crests are buttresses which 14 MASTODONS. join and obstruct the first and second valleys. As in the type of the species found by Doctor Mark Francis at Pittbridge, Burleson County, Texas (Bull. No. 71, Univ. Texas, p. 18, pis. in, iv) the third valley is little obstructed. The tooth is somewhat larger than the hindmost molar of the type. Professor W. M. Winton, of the Texas Christian University, at Fort Worth, sent the present writer a photograph of a lower jaw of a mastodon, which is in his institution and which was found near Fort Worth. The exact locality is on the north side of Trinity Eiver, north of the city. The jaw contained no teeth. This jaw and one of Elephas columbi appear to have been discovered in a gravel-pit about 45 feet above high water. In January 1924, the writer received from Dr. Mark Francis a photo- graph of teeth of Mammut americanum, the first and second right lower molars, which had been found 5 miles east of Fort Worth, in the valley of Trinity River, at the junction of this river with Little Fossil Creek. These teeth are now in the Francis collection. Lipan, Hood Comity (8). — From Dr. J. A. Udden, of the University of Texas, the U. S. National Museum received a part of an upper right molar of Mammut americanum, which had been sent to him by W. H. Roach, of Lipan. The tooth was found near that village, which is on the western bound- ary of Hood County and on Kickapoo Creek. It was discovered in a ravine about 6 feet deep. The banks of the ravine appear to be composed of a sandy loam and red; clay. The front crest of the tooth is missing, the second is little worn. The tooth is remarkable for its great width, being 106 mm. at the second crest. It is well fossilized and white in color. Granbury, Hood County (9). — In the collection of the University of Texas, at Austin, is a palate of Mammut americanum from Granbury. Of the first and second molars the former is little worn, the latter not at all. The speci- men is labeled as having been found in the "drift," doubtless the gravels, of Brazos River, and is credited to J. A. Taff, who was on the Texas Geological Survey from 1889 to 1894. Dr. R. T. Hill referred probably to this specimen (2d Ann. Rep. U. S. Geol. Surv., pt. 7, p. 360). Indian Creek, Comanche County (10). — In his report of 1901 (21st Ann. Rep. U. S. Geol. Surv., pt. 7, p. 360) Dr. R. T. Hill stated that remains of Mastodon had been found in surficial deposits at Indian Creek. This creek is a small tributary of Leon River, in the southern part of the county. San Angelo, Tom Green County (11). — Dr. Mark Francis informed the writer, January 10, 1916, that he saw in the store of Fred Schmidt, in San Angelo, a tooth of Mammut americanum, and he inclosed a sketch of it. The tooth was found near the town named. Oatman Creek, near Llano, Llano County (12). — From somewhere along Oatman Creek, Hill (21st Ann. Rep. U. S. Geol. Surv., pt. 7, p. 360) re- ported remains of a mastodon. Whatever was found appears to have been discovered in surficial materials. Oatman Creek is a small tributary opening into Llano River, on the south side. Marble Falls, Burnet County (13). — In July, 1919 there was sent to the U. S. National Museum, for identification, an upper right second molar of a mastodon, found in a small creek near Marble Falls. It was sent by George Harwood, M. D., of the town named. TEXAS. 15 Kendalia, Kendall County (14). — In the San Antonio Express, of Decem- ber 13, 1913, there was published an account of the finding of remains of a mastodon near Kendalia. A half-tone reproduction of a hindmost molar makes it certain that the animal was Mammut americanum. The remains were found on the ranch of Mr. John Kreupper. The writer has been unable to learn the exact location of the farm. Men had been engaged in making drainage-ditches, one of which had been converted by heavy rains into a deep gulley, in which a boy found a jaw with teeth in it. It appears that a skull (or a part of one) and a tusk were discovered. The bones were at a depth of 7 feet in a bed of sand; above the sand was about 5 feet of gravel and over this 2 feet of surface soil. Cibolo Creek, Kendall County (15). — In the collection of Professor At- water, at Houston, the writer saw an upper right hindmost molar of Mammut americanum, which was found on Cibolo Creek, in Kendall County. This creek forms the boundary between Bexar and Kendall counties. Northern part of Medina County (16). — In 1905 (Bull. Sci. Soc. San An- tonio, vol. I, No. 1, pp. 3-10, pis. i-iv, text — fig. 1), Professor Bernard Macken- sen gave an account of the excavation of some mastodon remains at a place about 14 miles north-northeast of Hondo, Medina County. The distance was probably not measured in a direct line. The locality was on the ranch of 0. H. Brucks and on the border of what is called the Edwards Plateau. There is here a flat which is sometimes a pond. This is near Yerde Creek, which empties into Frio Eiver. On the west the locality is bounded by an escarpment of Cretaceous rocks from 200 to 300 feet in height. The bones of the mastodons were found at a depth of about 9 feet, lying in a stratum of grayish clay 2 feet thick. Underneath this was a layer of gravel several inches in thickness, which contained specimens of Exogyra arietina, derived from the neighboring Cretaceous deposits. Over the gray clay which inclosed the mastodon bones is a layer of black muck 7 feet thick. The mastodon remains were much injured in getting them out. Some portions of tusks were secured. Parts of two jaws were exhumed and a third had previously been taken out. A lower jaw was figured by Mackensen, which had in it a tusk. The writer has examined this and found the diameter to be 23 mm. It projects 34 mm. beyond the jaw, with a part broken off. The socket ex- tends backward 130 mm. It seems not improbable that this species is Mammut progcnium. Helotes?, Bexar County (17). — In 1885 (Proc. Amer. Philos. Soc, vol. xxn, p. 8), Cope reported that he had a specimen (what part he did not say) of Mammut americanum, which had been found near San Antonio. This had been secured from G. W. Marnock, who lived at Helotes, about 15 miles northwest of San Antonio; but the specimen may have been found elsewhere. The same part of the animal was probably referred to later by Cope (Amer. Naturalist, vol. xix, p. 1208; vol. xxiii, p. 164, Feb.). Bulverde, Bexar County (18). — From a cave near Bulverde the writer had collected by D. V. Shuhart, of San Antonio, a part of a lower left third milk- molar. The hinder crest is gone. The width is 59 mm. From the same cave have been obtained remains of 18 species of fossil vertebrates which are listed on page 247. 16 MASTODONS. San Antonio, Bexar County (47). — Teeth of Mamrnut aniericanum have been reported from the gravels along San Antonio River at San Antonio, at a depth of 15 feet. At Bethany College, Lindsborg, Kansas, the writer has seen a palate of a mastodon having a right hinder molar 230 mm. long. It is much damaged, so that it is not certain that it belonged to Mamrnut. It seems to have 6 crests. It may belong to S teg o mastodon. The specimen is said to have been found at San Antonio. New Braunfels, Comal County (19). — In the American Museum of Natu- ral History, New York, is a lower right second molar of this species, which is labeled as having been found on the bank of Guadalupe River, 5 miles above New Braunfels, July 1873. It was buried in yellow clay at a depth of 75 feet from the top of the bank. Cedar Creek, Travis County (20). — In the collection of the State Uni- versity of Texas is a much-worn lower left hindmost molar of Mamrnut americanum, labeled as having been found, in 1870, somewhere along Cedar Creek, in this county. On a card accompanying the specimen is the name J. H. Hutchins. Cedar Creek crosses the southern corner of Travis County. Bastrop, Bastrop County (21).— In October 1922, Hugh H. Duval, of Bastrop, sent to the U. S. National Museum an upper hindmost molar of M. americanum, found in Bastrop. The front crest and root were gone. The crests were flat as compared with the usual type of molar. The tooth was discovered in a gravel-pit on the second terrace of Colorado River. From the same pit have been collected Mylodon and Bison. The U. S. National Museum has received from Dr. J. A. Udden, from near Bastrop, an upper left hindmost molar of Mamrnut americanum. This lacks one or two hinder cross-crests. The molar was sent to Doctor Udden by C. E. Lutz, who found it in a sand-bar in Colorado River, 8 miles northwest of Bastrop, on the road from this town to Austin. Speegleville, McLennan County (22). — In the collection of Baylor Univer- sity is a fragment of a tooth of Mamrnut americanum, consisting of 2 cross- crests, which was found on Hog Creek, 3 miles northwest from Speegleville. Here may be recorded an upper left third molar of a mastodon owned by the wife of Dr. H. C. Black, of Waco, and found in a gravel-pit, along Bosque River, about 15 miles from Waco. This would apparently be not far east of Crawford. In the collection of Baylor University is an upper right second molar of a mastodon, which the writer has described (Proc. U. S. Nat. Mus., vol. liii, p. 221, pi. xxvi, fig. 3) as Gomphotherium elegans. The tooth which forms the type of the species was discovered at McPherson, Kansas. Is now re- garded as belonging to Anancus mirificus. The tooth at Baylor University was found on Hog Creek, not far from Speegleville. Inasmuch as this tooth possesses only 3 cross-crests it probably belongs to Anancus brazosius. Sparta, Bell County (23). — In the collection of Baylor University there is a lower right last molar of Mamrnut americanum , which is reported to have been found in the hills near Sparta. This village is situated on Cowhouse River, a tributary of Leon River. The length of the tooth is 155 mm., the width 105 mm. TEXAS. 17 Belton, Bell County (24). — From Miss Lula L. Taylor, of Belton, the writer received a description, sketch, and photograph of a second true molar of Mammut americanum, found near Belton. The tooth has a length of 109 mm. There are no roots preserved ; and as the crown shows no wear, the tooth had not yet come into use. It was discovered in a gravel-pit belonging to I. B. Warner and situated east of Leon River, about half a mile distant. It is also about a mile and a half due east of Belton. The height of the gravel-pit above the river is about 100 feet. Denny, Falls County (25). — Dr. Mark Francis informed the writer of the discovery of remains of a mastodon near Denny. No details are known. The writer has seen this tooth, an upper left last molar, 180 mm. long. It is un- worn and has a large pulp-cavity. Cameron, Milam County (51). — From the region about Cameron, Dr. Mark Francis has secured important remains of mastodons. A tooth of Mammut americanum was discovered on the farm of Frank Schreiber, about 8 miles west of Cameron. It was found in a ditch at a depth not greater than 8 feet, in a soil described as black waxy. Under this number must be mentioned a fragment of a mastodon molar which, in 1924, was found near Buckholts, about 20 miles west of Cameron and not far from Little River. The informa- tion was sent by Dr. Mark Francis. A more important specimen is the horizontal portion of a left ramus of a lower jaw, found in a gravel-pit on Little River, about 2 miles north of Cam- eron. This jaw, found in 1897, apparently by J. D. Heeley, was presented to Dr. Francis by Henry Reaves. The gravel-pit, as the writer is informed by Judge Kemp, of Cameron, is between 40 and 50 feet above high water in the river. The jaw presents the socket for the penultimate molar and most of the crown and roots of the hindmost one. This tooth is worn so that the place of the first crest is occupied by a concavity. Besides this, the hinder crest and the talon have been destroyed and the inner ends of the second and third crests are badly injured. This tooth is 212 mm. long and 90 mm. wide in front. It has 3 crests and a part of a fourth. Evidently there was a large hinder talon. The roots are in the finest condition. The great fang supporting the first crest is curved backward and is 170 mm. long. A smaller fang supports the inner end of the second crest and is 100 mm. long. The great hinder root is 100 mm. long and 110 mm. from front to rear at the crown and 72 mm. at the distal end. In 1886, Lydekker (Cat. Foss. Mamm. Brit. Mus., pt. iv, p. 46) recorded a mastodon tooth in the British Museum, which was found in Texas and which he identified as the third upper true molar of Mastodon cordillerum, other- wise known as M. andium. Fortunately, the U. S. National Museum has a well-made cast of this tooth. An examination shows that it is not an upper molar, but the lower right third molar, there being under the front crest only a single fang. The tooth is slightly worn and shows plainly the structure. A comparison of this cast with the tooth from Cameron shows that the two in all probability belonged to the same species. That this species is Mastodon an- dium (Anancus hyodon) the writer regards as very doubtful. In length, width, and arrangement of the trefoils it resembles more closely M. humboldtii. It 18 MASTODONS. differs from both of the South American species in the greater height of the columns of the crests. Besides, it is improbable that the same species existed in both North America and South America. We know nothing about the upper and lower tusks of the Texas specimen and nothing about its skull. The tooth in the British Museum and the one just received from Cam- eron agree in their characters with the one described by the writer (Pan- Amer. Geologist, vol. xxix, pp. 112-114, pi. VIII, figs. 1, 2) as Anancus brazosius, and to this they are referred. Munson Shoals, Brazos County (26). — In the collection of Dr. Francis is an upper second left molar of Mammut americanum, found on Munson Shoals, in Brazos River, about 2.5 miles below Pittbridge. Pittbridge, Brazos and Burleson Counties (27). — The collection of Dr. Francis contains the left side of the palate of Mammut americanum, which was found at Pittbridge, in Brazos River, August 9, 1899. This specimen has the second and third molars, the latter with 4 crests and a small heel. This may belong with the lower jaw, to be described below. Like the lower teeth, these upper ones are white. The second molar is 112 mm. long and 87 mm. wide; the third, 172 mm. long and 104 mm. wide. A lower jaw collected here during a flood in July 1899 (of which jaw Doctor Francis has sent a photograph) contains a large socket for a tusk. The diameter of the socket is 50 mm. The length of the fragment present is 85 mm. The dis- tance from the angle of the jaw to the front is 830 mm. The ascending ramus has a width of 370 mm. The condyle rises 510 mm. above the lower border of the jaw. The second molar is 118 mm. long and 97 mm. wide; the third molar is 200 mm. long and 97 mm. wide. This jaw is referred to M. progeniurn. Doctor Francis has a mastodon jaw with a 3-ridged tooth, which was found on a gravel-bar 2 miles below Pittbridge ; also a molar found opposite the same place. A beautiful upper last milk-molar, which had just begun to be worn, was found at Pittbridge and is in Doctor Francis's collection. There is a large pulp-cavity and the roots had not yet been formed. The enamel is corrugated and there is a strong cingulum along the inner border. The length is 76 mm., width 65 mm. Here may be mentioned the discovery, at Pittbridge, of a species which the writer described (Bull. No. 71, Univ. Texas 1916 (1917), p. 18, pi. iv, figs. 1, 2) as Gomphotherium gratum. It is now referred to Anancus. The type specimen is owned by Dr. Francis and consists of a maxilla which con- tains the second and third molars. With this comes a third molar of the left side, evidently belonging to the same individual. The remains of Mammut progeniurn and of Anancus grains, above do- scribed, were found in 1899, in Brazos River, a mile below Pittbridge, after a heavy flood, during which a part of the bank fell, exposing the bones. Hidalgo Falls, Washington County (28). — In 1863 (Trans. St. Louis Acad. Sci., vol. ii, p. 141), B. F. Shumard, then on the Geological Survey of Texas, reported that he had found, in gravels overlying Miocene rocks at Hidalgo Falls, remains of Equus fossilis, Mastodon, Elephas, Megalonyx, Crocodilus, and Testudo. TEXAS. 19 In 1874, S. B. Buckley (1st Ann. Eep. Geol. and Agric. Surv. Texas, p. 64), referring probably to the same discoveries, stated that, about 15 years before he wrote, bones of the mastodon and of the mammoth in great abund- ance had been found in the banks of Brazos River. In 1893 (4th Ann. Rep. Geol. Surv. Texas, p. 44) Wm. Kennedy stated that numerous fossil remains had been reported from clays and conglomerates overlying sandstones in the neighborhood of Hidalgo Falls, but he obtained only a mastodon tooth. Doc- tor Francis has shown the writer a much-worn upper left second molar found in Brazos River, opposite Navasota, in 1918, on the Walker-Moore farm. The locality would be very close to Hidalgo Falls and is in Brazos County. A list of the species found here is given on page 245. Chapel Hill, Washington County (29). — The writer has seen a beautifully preserved upper right second molar of Mammut americanum, which was found, in November 1915, by T. B. Lipscomb, in Washington County, oppo- site Hempstead, and probably picked up on Brazos River. The tooth is now owned by Doctor Francis. Its length is 115 mm., width 89 mm. The enamel is rough and wrinkled. The roots were pretty well developed, but are broken off near the bases. There is yet a large pulp-cavity. Scarcely any wear is visible. The enamel is reddish brown, yellow, and whitish. Leota, Madison County (30). — From Doctor Francis the writer received, for examination, a beautifully preserved tooth of Mammut americanum, found in 1915 by B. H. Lander, supposed to be of Madison ville, but a letter addressed to him there was returned, stamped "unknown." The tooth is a lower hindmost molar, with 4 cross-crests and a heel. Such parts of the roots as were developed are broken off. The tooth is 207 mm. long and 102 mm. wide. In the valleys is a little cement. The pulp-cavity contains some pebbles and some greenish sand, cemented together. The tooth is labeled as found in Trinity River, opposite Madisonville. It may therefore have been picked up in Houston County. Clapps Ferry, Trinity County (31). — In 1922, Dr. Mark Francis secured a 3-ridged tooth of Mammut americanum,, which was found in Trinity River, near Clapps Ferry, 10 miles west of Trinity. Drew's Landing, San Jacinto County (49). — In 1920 (Univ. Texas Bull. No. 1869, p. 269) Professor E. T. Dumble reported that he had discovered, a mile below Drew's Landing, on the west bank of the Trinity, remains of mastodons, consisting of limb-bones, lower jaw-bones, and 19 teeth. They were found near the base of a deposit of sandy alluvial clay 30 feet thick. Post, Garza County (32). — Dr. Mark Francis informed the writer that in 1916 a mastodon tooth was found near Post and came into the possession of George Doughty. Denning, San Augustine County (33). — Doctor Francis's collection con- tains an upper left third molar of a mastodon, which he received, in No- vember 1912, from T. J. Miles, of Denning. The roots are not developed and the pulp-cavity is large. The length is 178 mm., the width 100 mm. There is no talon. Port Arthur, Jefferson County (34). — Dr. Mark Francis informed the writer in June 1917, that he had lately received a very fine molar of a mas- 20 MASTODONS. todon, which was found at Port Arthur. It had evidently been washed ashore. The tooth is a lower left hindmost molar, worn on 3 crests. The talon is almost a fifth crest. The tooth has on it patches of marine bryozoa. Port Arthur is situated on the western shore of Sabine Lake and apparently on deposits belonging to the Recent epoch; but it is evident that deposits of Pleistocene age are present at no great depth below the alluvium. Naturally, it is impossible to say to what stage of the Pleistocene the tooth in question belonged. Sour Lake, Hardin County (35). — A fragment of a molar of Mammut americanum was found here in blue clay and sand, beneath a bed of bitumen (Leidy, Ext. Vert. Fauna West Terr., p. 244). For other fossils found at Sour Lake, see page 242. Harrisburg, Harris County (36). — The writer has seen at Houston, in the collection of Professor E. T. Dumble, a fragment of a molar of Mammut americanum, found on the banks of Buffalo Bayou, near Harris Bend. This must be near Harrisburg. "Fifty Miles above the Mouth of Brazos River" (37). — Jeffries Wyman, in 1856 (Proc. Boston Soc. Nat. Hist., vol. vi, p. 53), described 2 molars of a mastodon, an upper last and a lower last molar, which had been found about 50 miles above the mouth of Brazos River. With these were an ulna and part of a humerus supposed to belong to a mastodon; also remains of Megatherium and of an elephant. McDowell, Austin County (38). — Dr. Mark Francis informed the writer that in 1909 P. M. Cuney, of Brookshire, found in Brazos River some fine mastodon and elephant teeth. These are now in the collection of Doctor Francis. The writer has before him a photograph of a tooth from this place. It is an upper left last molar, with fine roots still held in a fragment of the palate. Inasmuch as these materials were found in Brazos River, the locality was probably not far from McDowell. San Felipe, Austin County (39). — This locality has long been noted for its fossil vertebrates. The species found here are recorded on page 246. Re- mains of Mammut americanus were described from here by W. M. Carpenter in 1846 (Amer. Jour. Sci., vol. I, p. 249). Leidy (Proc. Acad. Nat. Sci. Phila., 1870, p. 97) reports on a specimen from this place which is now in the collection at Harvard College. The collection of Doctor Francis has an upper right last molar, 175 mm. long, found in Brazos River at this place; another much like it is 154 mm. long. His collection also has a large lower right last molar of M. americanum from San Felipe, also a palate with upper hindmost molars, right and left, teeth white, length 178 mm., width 107 mm., roots immense, the front root 150 mm. high, the hinder one 160 mm. ; the hinder talon little developed. S. B. Buckley (Prelim. Rep. Geol. Agric. Surv. Texas, 1866, p. 44) stated that, many years before he wrote, almost the entire skeleton of a mastodon had been found on the banks of Brazos River near San Felipe. The bones were taken to New Orleans for exhibition and were reported as destroyed in a fire. At San Felipe has been found a right ramus of a lower jaw with the second and third molars, which the writer described under the name Anancus brazo- TEXAS. 21 sius (Pan-Amer. Geologist, vol. xxxix, p. 112, pi. vm, figs. 1, 2). This jaw is in the Agricultural and Mechanical College at College Station, Texas. From Professor Taylor White, of Sealy, Doctor Francis received a large upper right second molar, which was found opposite Sealy, in the "big bend" of Brazos Eiver. The length is 116 mm.; width, 92 mm.; there are 3 ridges. Near Ezzell, Lavaca County (40). — Dr. Mark Francis informed the writer that he secured in January 1922, a right side of the lower jaw of a mastodon, found on Summers Creek, near Ezzell; also 2 molars of a mastodon, dis- covered on Mixon Creek, somewhere in the same county. Here may be recorded the finding of mastodon teeth at a point 5 miles west of Hallettsville. Some teeth were sent in January 1922 to Doctor Francis by L. P. Davis. With these were sent likewise 2 teeth which are referred to Elephas columbi. Cuero, De Witt County (41). — In the IT. S. National Museum is a lower last molar, No. 2368, which is recorded as sent to the museum by Lee Joseph, of Cuero. Mr. Joseph informed the writer that the tooth was reported to him as found in Clear Creek. In what part of the county this creek is sit- uated the writer has not learned. The tooth is remarkable on account of the flattened condition of the various cusps. It possibly belongs to Mammut progenium or to an unrecognized species. For (47) see page 16; for (49) page 19 ; for (50) page 12 ; for (51) page 17. Coleto Creek, Victoria County (48). — Dr. Alexander Deussen reported (Prof. Pap. IT. S. Geol. Surv. 126, p. 109) that remains of mastodon had been taken from the banks of Coleto Creek about 8 miles west of southwest of Victoria. He implies that the remains were buried in Lissie deposits. Keeran Point, Victoria County (42). — From Dr. Mark Francis the writer learns that in December 1919 he visited this locality and found that an enor- mous mass of the bluff had fallen not long before, and from the fallen ma- terials he secured some fragments of mastodon, of elephant (Elephas co- lumbi), and a vertebra and a part of a metapodial of a camel. Details re- garding the geology of the region are given on page 236. The mastodon was later referred to Anancus brazosius (Pan-Amer. Geologist, vol. xxxix, p. 113). Berclair, Goliad County (43). — The collection of St. Louis University, at St. Louis, Missouri, has a well-preserved lower left third molar of an Ameri- man mastodon, labeled as having come from Berclair, which is on Blanco Creek. Beeville, Bee County (44). — Dr. Mark Francis informed the writer that he learned of the finding of remains of the mastodon at this place. No de- tails have been communicated. Corpus Christi, Nueces County (52). — In October, 1923, A. P. Herndon, editor of the Corpus Christi Caller, sent to the U. S. National Museum a part of a mastodon tooth found at a depth of 16 feet in a clay bank at Flour Bluff, near Corpus Christi. The specimen presents parts of three hinder cross-crests and the small talon of a much-worn right hindmost molar, ap- parently of the lower jaw. The middle part of the grinding surface is deeply 22 MASTODONS. hollowed out. The outer ends of the crests present well-formed trefoils with folded enamel; the inner ends less well developed trefoils. The enamel of the sides and rear of the tooth is strongly wrinkled vertically. It seems to the writer that this tooth represents the species Anancus brazosius. The Corpus Christi tooth has the catalogue number 10888 in the U. S. National Museum. Duval, Jim Wells, or Nueces County (45). — In 1885 (Amer. Naturalist, vol. xix, p. 1208), Cope stated that Mammut americanurn had been found in southwestern Texas. It is quite certain that he referred to the region about San Diego, but whether the locality was in Duval, Jim Wells, or Nueces County is not known. It was probably on Taranchua Creek, Duval County. Nor did Cope anywhere, so far as the writer knows, state what parts of this species he had in his possession. El Paso, El Paso County (46).— In 1907, G. B. Eichardson, of the U. S. Geological Survey, sent to the U. S. National Museum some fossil vertebrate remains collected at El Paso. Among these the writer saw a fragment of a tooth of a mastodon belonging to another genus than Mammut. It was probably a species of Anancus, possibly of Gomphotherium. AEKANSAS. (Map 2.) Helena, Phillips County (1). — The earliest record the writer has found of the occurrence of mastodon remains in Arkansas is of one met with at Helena. Dr. R. E. Call (Ann. Rep. Geol. Surv. Ark., vol. II, p. 39) stated that here bones of a mastodon had, at some time before 1889, been exhumed. They were buried in a blue-black unstratifled carbonaceous clay, at a depth of apparently 50 feet. They were further reported to have been buried in the cliff of loess, about 25 feet back from its face, but not in the loess itself. It is not probable that the remains were seen by Doctor Call; and there is nothing in the account that would exclude the possibility that they belonged to one of the elephants. Near Trumann, Poinsett County (2). — In April 1921, the writer received from Frank Janes, of Trumann, a fragment of a tooth and a dorsal vertebra of Mammut americanurn, which had been turned out of a drainage-canal west of Trumann. The exact locality has been given by Mr. Janes as in the south- east quarter of section 7, township 12 north, range 5 east. The depth was supposed to be about 12 feet. The deposit is said to be a black waxy soil called there "gumbo/'' The digging was being made in the channel of a small stream, the Little Bay. A few other bones, among them a broken leg- bone, were found. Others might have been covered up in the dump. Mississippi River, Chicot County (3). — In 1922, Dr. Clarence M. Pate, of Tutwiler, Mississippi, sent to the U. S. National Museum, for identification and return, an upper second molar of Mammut americanurn. It had been found in a carload of gravel dredged up in Mississippi River. In answer to an inquiry Doctor Pate wrote that the gravel had been taken from the river at Greenville, Mississippi. The tooth is here credited to Arkansas. OKLAHOMA. 23 OKLAHOMA. (Map 2.) Afton, Ottawa County (1).— In the U. S. National Museum are many finely preserved teeth of Mammut americanum, secured near this place by W. H. Holmes in 1901. His report on his discoveries was first published in 1902 (Amer. Anthropologist, vol. iv, pp. 108-129, with 2 pis.) and again in 1903, more fully illustrated (Ann. Eep. Smithson. Inst, for 1901, pp. 237-252, with 26 pis.). The situation of the spring and an account of the palseonto- logical and anthropological materials discovered there will be found on page 254, with a list of the species of mammals secured. From this spring Holmes took at least 100 mastodon teeth. Some of these were illustrated on his plates vi and vn of his report of 1903. Figures of others have been published by the present writer (Geol. Surv. Indiana, vol. xxxvi; Iowa Geol. Surv., vol. xxiii ). All of these teeth are stained brown or black and appear to have lost practically all of their animal matter; but the vascular canals and the canaliculi are not filled with mineral matter; the freshly broken surfaces adhere strongly to the tongue. The report of Holmes says the mastodon teeth were found at a depth of 4 feet and downward. The present writer is of the opinion that most of these belong to the Sheridan stage; naturally some may be somewhat older, others younger. Peno, Le Flore County (2). — Early in 1923, E. V. Lovett, of Peno, sent to the Smithsonian Institution a letter and a photograph of the left ramus of a lower jaw of Mammut americanum, found in Arkansas River, about 2 miles above Peno and 3 miles from Braden. The jaw contained the last and the next to the last molars, little worn and finely preserved. Mr. Lovett has also the skull of a bison, found in the Arkansas River. In the St. Louis Academy of Science is a mastodon tooth, an upper left hindmost molar, which is labeled as found in Oklahoma, 15 miles above Fort Smith. This would be along Arkansas River, probably not far from the town of Braden; the locality may have been on the north side of the river. Pre- served in the same collection and secured in the same region are a part of a skull of an extinct species of bison and a part of a skull of the musk-ox, Symbos cavifrons. Walker, Cleveland County (3). — Mr. Edward Butts, of Kansas City, Mis- souri, informed the writer that he has a tooth of a mastodon, found in Cleve- land County during the excavation of a caisson for a bridge, for the Gulf & Sante Fe Railway, over Canadian River. According to the maps this would be between the stations of Lexington and Noble, in township 7 north, range 2 west. The tooth is stated to have been found at a depth of 30 feet below the water. A sketch of the tooth appears to show that it is a last milk molar. Citra, Hughes County (4). — At Stanford University, California, the writer saw an upper second right molar and unworn parts of probably a third molar, labeled as found 14 miles northeast of Stonewall, Chicasaw Nation, now Pontotoc County. The locality would apparently be near Citra; it might be in either of the two counties named or in the northern part of Coal County. Mulhall, Logan County (5). — In 1917 (Bull. Geol. Soc. Amer., vol. xxvm, p. 212), Professor Troxell stated that he had collected remains of mastodons 24 MASTODONS. near Mulhall. Nothing since that time has been published on the collection. For the associated genera, see page 255. NEW MEXICO. (Map 2.) Taos, Taos County (1).— In 1875 (Append. LL., Ann. Rep. Chief of Engi- neers, p. 986), Professor E. D. Cope reported that he had been shown a tooth of Mammut americanum, found in the clay along Rancho Creek. Probably this was the same tooth that he mentioned in 1874 (Proc. Acad. Nat. Sci. Phila., 1874, p. 221). The exact position of Rancho Creek is not known to the writer, but there is on the maps a village called Ranchos, a few miles south of Taos. Mastodon, Donna Ana County (2). — In 1907 (Bull. Geol. Soc. Amer., vol. xviii, p. 215), W. T. Lee reported the discovery of bones of Mammut ameri- canum and of a horse at a siding called (from the discovery) Mastodon. The locality is between El Paso and Noria, about 15 miles south of the Afton volcanoes, where other horse remains were discovered (p. 258). The bones of the mastodon and of the horse were identified by J. W. Gidley. The same species of mastodon had been previously found only at Taos. MISSOURI. (Map 2.) Galena, Stone County (1). — At a meeting of the Boston Society of Natural History, December 5, 1849, Mr. Desor referred to a fine skeleton of a mas- todon which he had lately seen taken from a depth of 70 feet at Galena. He remarked that it had been thought that this specimen occurred in true drift; but it was found in a deposit mixed with drift materials and buried subse- quent to the drift. The teeth were fine. The tusks had a diameter of 3 inches; the speaker thought that they must have belonged to a large animal. This county is in the region affected by the Ozark uplift and lies between 300 and 400 miles south of the border of the region covered by glacial drift. Galena itself is on the James Fork of White River. We do not know exactly where the mastodon was met with nor under what conditions ; probably it was buried in alluvial deposits along the James. Under a depth so great the ani- mal may have dated far back in the Pleistocene; of this we can not now be certain. Near Joplin, Jasper County (2). — From Byron A. Ash, of Carthage, Mis- souri, the writer learned of the discovery of some remains of mastodons and of Elephas columbi, at a place described as situated 9 miles southwest of Carthage. This would be not far northeast of Joplin. A shaft was being sunk in search of zinc ore. Mr. Ash writes as follows : "In sinking the shaft we first passed through 12 feet of clay and gravel; then through 45 feet of limestone; then through 30 feet of black loam and into a peat formation. In this the bones were found and also pieces of petrified wood." The bones and teeth were thus at a depth of nearly 90 feet from the sur- face. Besides the species mentioned, Mr. Ash stated that bones of a turtle and of a species of Bison, probably extinct, were discovered. These were exam- MISSOURI. 25 ined and identified at the U. S. National Museum. The supposed petrified wood was probably portions of tusks. It is certain that, in sinking the shaft, one of the caves so abundant in that region had been passed through. In this cave had accumulated soft deposits in which were buried the animal remains. It seems that this cave had been completely hidden from view. While there are no evidences of the age of the remains to enable one to speak with cer- tainty, it appears probable that they date back to early Pleistocene, possibly to the Aftonian. The Joplin Folio (No. 148) of the IT. S. Geological Survey describes minutely the geology of this region and discusses the origin of the numerous caves existing there. Springfield, Greene County (3). — The identity of the tusks here reported is doubtful, as no teeth were found with them. They may belong to one of the elephants. A. M. Stalnaker (Kansas City Eev. Sci. Industry, vol. vn, p. 615) reported the discovery of tusks in digging a well. The digger seems to have passed through solid rock, then into mud that filled what is probably a cave. One tusk was, at the time of the report, in possession of the Academy of Sciences at Kansas City. In 1898, Dr. E. M. Shepard (Geol. Surv. Mo., vol. xn, p. 143) told of the discovery of a well-preserved mastodon tusk about 8 feet long, which lay in a horizontal crevice in the limestone at Springfield. It was met with in excavations made just south of the Kansas City, Fort Scott, and Memphis Eailroad. It may be that this refers to the same tusks as those reported by Mr. Stalnaker. Here again it appears that one must attach considerable age to the tusks, probably early Pleistocene. Papinsville, Bates County (4). — In 1881, G. C. Broadhead, the geologist, reported (Kansas City Eev. Sci. Industry, vol. iv, p. 521) that a mastodon tusk had been found at Papinsville, in the gravel. The town is on Osage Eiver, and the tusk, whether of mastodon or of elephant, had probably been washed out of the bank of the stream. On Pomme de Terre River, Benton County (5). — This locality is famous for the remains of mastodons which it has furnished. Besides these, however, there have been found or reported remains of Elephas columbi, Bison, Odo- coileus, Cervus, Equus, Mylodon, and Megatherium. In 1842, Eichard Owen (Proc. Geol. Soc, London, vol. in, p. 689), in discussing Koch's collection left at the British Museum, stated that in it were remains of Mylodon harlani; also portions of large species of Bos, Cervus, etc. ; but the place of their origin is not mentioned. The occurrence of bones in this region was known to B. S. Barton as long ago as 1806. In the fourth edition of Cuvier's Ossemens Fossiles, 1834, vol. n, page 270, is an extract from a letter written by Barton to Cuvier in 1806, in which Barton stated that an intelligent traveler had seen in a particular locality near the river of the Osage Indians thousands of the bones of this animal and that this traveler had collected there, among other things, 17 tusks, of which some had a length of 6 feet and a diameter of a foot. However, the majority of these bones were in a bad state of decomposition. Barton sent a molar to Cuvier which appears to have been found in this region. In his Archceologia Americana, 1814, page 26, Barton speaks of the immense number of bones in 26 MASTODONS. the country of the Osage Indians; he also stated that lie had a molar from there. It was through the explorations of Dr. Albert C. Koch that the Pomme de Terre locality became famous. This indefatigable searcher after fossil re- mains of mastodons and zeuglodons gave a description of his discovery of mastodon bones and teeth in Benton County, Missouri, in 1840. These re- mains are of interest, because from them was restored, by Professor Richard Owen, a skeleton in the British Museum of Natural History. In his "Short description of fossil remains found in the State of Missouri/' a pamphlet of 8 pages, printed at St. Louis in 1840, Koch gave an account of the discovery of the skeleton on which he bestowed the name Missourium. A fifth edition of the same work, enlarged to 28 pages, was published in the same year at Dublin, Ireland. An account in different words was presented by Koch in his pamphlet of 99 pages and 8 plates, printed in Berlin in 1845 and en- titled Die Riesenthiere, etc. The locality was in the southern part of Benton County, on the Pomme de Terre River, a tributary of Osage River. Broad- head (Kansas City Rev. Sci. Industry, vol. iv, 1881, p. 520) described the place as being on Pomme de Terre River, 10 miles southwest of Warsaw ; but, according to the best maps at hand, a point on Pomme de Terre River 10 miles from Warsaw would be almost due south of this town. Broadhead further stated that the place was on the farm of Alexander Brashears. This information was taken from James H. Lay's History of Benton County, page 14, published in 1876. Here the name is given as Breshears. Lay stated that a Scotchman named Cott exhumed a large and well-preserved skeleton, which he took east and sold for $20,000. This probably refers to Doctor Koch's work. Other persons are mentioned as having unprofitably spent much money at this place in search of bones. From Mr. J. S. Kidwell, county surveyor and highway engineer at Warsaw, Missouri, the information has been received that the locality is in the north- west quarter of section 9, township 38, range 22 west. This would be in Hickory County, not far from the town of Avery. According to Koch, an attempt had been made by some of the earlier set- tlers to construct a mill for grinding flour and corn meal. In making the necessary excavation, some bones of young mastodons were met with; but the interest in these was soon lost. In 1839, another tooth was found and the report of this reached Koch early in 1840. On March 31 of that year (Die Riesenthiere, pp. 69, 70) he was at the crossing of Osage River, 24 miles away from his destination. On page 71 of the work quoted he said that lie spent 5 months in a cabin there. On the other hand, P. R. Hoy (Amer. Natu- ralist, vol. v, 1871, p. 147) wrote that in March 1840 he "visited the spot on the Pomme de Terre, Benton County, Missouri, where Dr. Koch had recently disinterred the large male mastodon now in the British Museum, which the Doctor mounted and named the Missourium tetracaulodon." Hoy found that the excavation made by Koch was about 15 feet in diameter and 6 feet in depth and was half filled with water. A man who had helped Koch told Hoy that Koch did not get out all of the bones ; and Hoy succeeded in finding a molar and some other remains. See also statements made by Dr. Edmund MISSOURI. 27 Andrews (Amer. Jour. Sci., vol. x, 1875, pp. 32-34). It is impossible to reconcile such contradictory statements. In 1843, Harlan (Amer. Jour. Sci., vol. xliv, p. 69), speaking of Koch's large collection on exhibition in Philadelphia, stated that it contained more than 300 mastodon teeth, with numerous jaws, besides remains of the ox (probably bison), deer, elk, megatherium, and mylodon. Harlan (loc. cit., pis. i-m) described some parts of the skeleton and about 24 teeth (his Orycterotherium missouriense) which is now referred to Mylodon harlani. He believed that the collection was made on the Pomme de Terre. Koch described quite minutely the character of the deposits met with in making his excavation. Above all was the surface soil, the depth of which was not stated ; beneath this from 2 to 3 feet of gravel ; beneath this about the same thickness of yellowish clay. Then followed two layers of gravel, each from 9 to 18 inches in depth; below these 3 feet of blue clay. Then came 3 feet of alluvial soil, which contained some of the bones of the Missourium. Under this was the deposit which inclosed most of the skeleton. Besides the bones, this layer contained much other animal material, consisting of the flesh, skin, etc., which had been transformed into a brown fatty mass. Beneath this bone-bearing layer was a deposit of sand. The layer of alluvium contained an abundance of vegetable matter in a wonderful state of preservation. Koch regarded the plants as indicating a warm or tropical climate. He found cypress burrs, and wood which he thought was related to logwood ; also a flower of Strelitzia, a genus of African plants; likewise several stems of pal- metto. Koch concluded that "a revolution of the earth'1' had taken place here, between September 15 and October 20 at some time in the past. In discuss- ing the discovery in 1857, (Trans. St. Louis Acad. Sci., vol. I, pp. 63, 64) Koch affirmed that the layer of vegetable mold was 5 or 6 feet deep and was overlain by 20 feet of alternate layers of sand, clay, and gravel. From the bones in Koch's possession, no doubt obtained partly on Pomme de Terre Eiver, he reconstructed his Missourium theristocaulodon, of which in 1845 he gave a description and a figure (Die Eiesenthiere, pp. 43-53, pi. viii ). The tusks were each 10 feet long. The length of the skeleton, from tip of nose to root of tail, is given as 30 feet, height 15 feet. This skeleton was afterwards sold to the British Museum and reconstructed by Eichard Owen, being at the same time reduced to more reasonable proportions. Even in its reconstructed form it was composed of bones from several individuals (Lydekker, Cat. Foss. Mamm., pt. iv, p. 16, fig. 3). The price received by Koch for this skeleton was 1,000 pounds sterling (Amer. Jour. Sci., vol. it, 1846, p. 132, fig.). Besides the bones of the several animals mentioned, which Koch claimed to have found when excavating for his Missourium, he asserted that he took out with his own hand from among the bones 5 stone arrow-heads. One of these lay under the right femur and had left a light-colored spot on the else- where brown-colored bone. We can not form definite conclusions about the age of the fossils found by Koch along Pomme de Terre Eiver. If we were certain that the things which he reported from there were really found there, we would have some founda- 28 MASTODONS. tion to stand on. If he found there remains of horse, mylodon, and megather- ium, early Pleistocene, probably Aftonian, would be indicated. Tackner, Benton County (6). — At a meeting of the Academy of Natural Sciences, December 3, 1843 (Proc. Acad. Nat. Sci. Phila., vol. I, p. 321 ), Doc- tor Chaloner called the attention of the Society to a collection of bones of mastodons and elephants which had been brought from Benton County, Mis- souri, by S. H. Whipple. The weight of the collection was estimated to be about 2,000 pounds. At a meeting of the American Phiolosophical Society held January 19, 1844 (Proc. Amer. Philos. Soc, vol. iv, p. 35), this col- lection of bones was reported as having been purchased of Whipple at the in- stance of Dr. Isaac Hays. A statement by Whipple accompanied the collec- tion to the effect that it had been made in the county mentioned, at a dis- tance of half a mile from Osage Eiver, and evidently south of it. Broadhead (Kansas City Eev. Sci. Indust., vol. iv, p. 520) stated that the place was on the farm of the heirs of Charles Wickliffe. Whipple reported that the locality was in latitude 38° 10' and in 16° 40' longtitude west from Washington, which would be 93° 41' 34" west from Greenwich. However, the latitude as given would locate the spot about 5 miles south of Osage River, and the longi- tude as given would locate it in the middle of Henry, the next county west. One can not rely on the statement. Mr. J. S. Kidwell, county surveyor of Benton County, has informed the writer that the locality is in the southeast quarter of section 24, township 40, range 23 west. The description of the character of the country is no doubt correct. South of the Osage it is charac- terized by its irregular and broken appearance, running up into extensive ridges, rocky cliffs, and flinty knobs. Between these there intervene valleys, through which run streams of water, bordered by alluvial bottoms and lofty forests. In one of these valleys there is a small saline marsh, perhaps an acre in extent. Close to the marsh runs a rocky ridge on which remain columns of limestone, sometimes rising 20 feet above the general level of the ridge. In the soil of the marsh were found the bones, at depths varying from 2 to 12 feet and lying in great disorder. It does not appear that any remains of other animals than mastodons and elephants were secured at this place. The elephant is probably E. columbi, inasmuch as specimens from this county are in the Philadelphia Academy collection. The collections of the American Philosophical Society were many years ago turned over to the Philadelphia Academy of Natural Sciences. In the col- lections of the Academy are found a considerable number of teeth and bones of mastodons, labeled as having been obtained in Benton County, Missouri, but no definite spot is mentioned; most of them probably belong to the Whip- ple collection. In the Academy's collections is the type of Dr. Isaac Hays's Tetracaulodon collinsii, now regarded as belonging to Mammut americanum. This specimen is labeled as having come from Benton County. As it was described in 1834 (Trans. Amer. Philos. Soc, vol. iv, p. 326, pi. xxvin), it appears, if the label is not incorrect, that mastodon materials from Benton County had reached Philadelphia long before Koch and Whipple brought their collections there. Furthermore, the type of Mastodon cuvieri Hays is in the Academy's collections; it, too, is labeled as having been obtained in MISSOURI. 29 Benton County. It was described in 1834 (op. cit., pp. 322, 334, pi. xxiv). Just where these fossils came from in Benton County it is impossible to de- termine. As to the geological age of the fossils of the Whipple collection, one can only indulge in surmises. In James H. Lay's History of Benton County, published in 1876, it is recorded that, at the Wickliff farm, Case and Eedmond took out a part, per- haps the whole, of a large mastodon skeleton and sold it at Cincinnati. Osceola, St. Clair County (7). — From Edward Butts, Kansas City, Mis- souri, the writer received a sketch of a penultimate upper molar of Mammut americanum, which had been dredged up with gravel from the bed of Osage Eiver at Osceola. At the time of writing, it was the possession of Mr. Butts. The little-worn crown had a length of 96 mm. ; the roots were well developed. Nothing can be affirmed regarding the stage of the Pleistocene when the animal lived. Sedalia, Pettis County (8). — The geological collection of the University of Missouri contains a large number of bones of Mammut americanum, which, through the kindness of Professor E. B. Branson, the writer has been allowed to examine. They were obtained near Sedalia by R. A. Blair; they embrace nearly all parts of the skeleton and individuals of different ages. A letter addressed to the writer by N. P. Elmore, county clerk of Pettis County, states that the spring at which these bones were found is in section 25, township 45 north, range 21 west. This would be about 4 miles southeast of Sedalia. The political name of the township is Flat Creek. This collection is obviously the one to which Broadhead referred in 1898 (Geol. Surv. Mo., vol. xn, p. 376). From Broadhead (Kansas City Rev. Sci. Indust., vol. iv, 1881, p. 519) it is learned that this collection was made by R. A. Blair in the winter of 1879-80. The spring where the bones were obtained was about 400 feet dis- tant from the creek, about 15 feet above it and on the opposite side, about 200 feet from the gently rising higher ground. The diameter of the bog sur- rounding the spring was about 20 feet. The mastodon bones were found in "a light spongy, peaty humus''' (and in a space about 15 feet in diameter), together with some pieces of wood resembling cypress. The depth of this humus was about 8 feet, underlain by gravel. It was supposed that remains of about 8 individuals were exhumed. A few flint implements and a stone club were found with the bones; these were probably introduced long after the interment of the mastodons. It does not appear that any remains of other animals were found with these mastodons and the writer did not observe any in the collection at the Uni- versity of Missouri; hence there is nothing to indicate the geological age of the mastodons. Bourbeuse Creek, Gasconade County (9). — In the Philadelphia Presby- terian of January 12, 1839, there appeared an anonymous article concerning the discovery of remains of a mastodon in Gasconade County and of evidences of the existence of man contemporaneously with the mastodon. This article was, on July 1, 1839, reprinted in the American Journal of Science, volume xxxvi, pages 198-200. The discoverer of the remains in question was Dr. Albert Koch, of St. Louis, the finder of the mastodon remains on Pomme de 30 MASTODONS. Terre River, in Benton County. The article was acknowledged by Koch in 1857 (Trans. St. Louis Acad. Sci., vol. I, p. 62), and further description was giveD of the discoveries. He also gave a picturesque account of his work here, in a pamphlet published at St. Louis in 1840, entitled "A short description of fossil remains found in the State of Missouri." The exact locality of the exhumation is not known. Bourbeuse Creek runs northeastward across the extreme southern end of the county. Koch's state- ment as to the longitude and latitude must be disregarded. The investiga- tions were made in 1839. It appears that a farmer, in cleaning out a spring about 400 yards from the bank of the stream, discovered at a depth of 5 feet a part of the back and hip-bone of a large animal. In his excavation for the remainder of the animal, Koch found that it had been mired in a standing position and that its legs were sunken in a layer of clay. Above this clay was a layer of ashes, from 2 to 6 inches thick; in his work of 1840 he stated that the layer of ashes was 6 to 12 inches thick. With these were mingled bones, partly burned, and broken pieces of rock. Koch's conclusion was that human beings had found the great creature helplessly mired and had carried there, from the shore of the stream, rocks to hurl at it. Then these natives had kindled around it, for its destruction, a great fire, which had burned many of the bones and had been especially destructive around the creature's head. The legs sunken in the clay had not been injured. Furthermore, the natives had hurled at the animal many stone arrow-heads, a stone spear-head, and some stone axes. Also between the rocks sunken in the ashes were found large pieces of skin that appeared like freshly tanned leather, strongly impreg- nated with the lye from the ashes ; and a great many of the sinews and arteries were seen, some of which Koch preserved in spirits. The stratum, which was made up of ashes and stones and bones, was over- lain by alluvial deposits of clay, sand, and soil to a depth of 8 to 9 feet. In his first account of this investigation, Koch says that, in case any reader had any doubts about the correctness of the statements, he could be referred to more than twenty witnesses present during the time of digging. Koch in- cluded an account of this discovery in his work of 1845, entitled "Die Riesen- thiere der Urwelt oder das neuentdecJcte Missourium theristocaulodon/' on page 57. In 1857, Dr. A. Wislizenus (Trans. St. Louis Acad. Sci., vol. I, p. 168) en- deavored to show that the conditions under which Koch had found the re- mains of the mastodon and the human implements did not bear out his con- clusions. In 1875, J. D. Dana (Amer. Jour. Sci., vol. ix, pp. 335-346) made a careful review of Koch's discoveries, publications, and pretentions, and showed that, even if Koch's honesty were conceded, he was not competent to form cor- rect judgments on anatomical and geological matters. Dr. Edmund Andrews (Amer. Jour. Sci., vol. x, 1875, p. 33) presented an explanation of the matters supposed to have been observed by Doctor Koch. Doubtless something might yet be learned by making careful excavations on that spot, but it would prob- ably not pay for the trouble. It is impossible to say at what time during the Pleistocene the mastodon lived. MISSOURI. 31 Ste. Genevieve County (10).— In 1873, B. F. Shumard (Geol. Surv. Mo., 1855-1871, p. 291) reported briefly that B. Pratt, of St. Mary's, had obtained in a well, at a depth of 18 feet, a fine molar of a mastodon. This was pre- sented to the Academy of Science of St. Louis ; it should now be in the collec- tion. The place where it was found is said to be in the southwest quarter of section 1, township 36 north, range 9 east. This is in the alluvial plane of Mississippi River. A section found near the well mentioned shows at the top 5 feet of soil, with much humus ; below this, 30 feet of yellow arenaceous clay ; then 15 feet of plastic blue clay. The tooth must have been in the yellow clay. Kimmswick, Jefferson County (11). — A considerable number of mastodons have been collected at Kimmswick at various times and by various persons. Further details regarding" these, the geology of the locality, and the other species obtained there are given on pages 262, 263. St. Louis, St. Louis County (12). — One might readily suppose that in mak- ing innumerable excavations in a great city like St. Louis large numbers of bones and teeth of fossil animals would be brought to light. On the contrary, the writer has found only three accounts of such discoveries ; these were made many years ago. In a report by G. C. Swallow, in 1855 (First and Second Ann. Repts. Geol. Surv. Mo., pt. 2, p. 175), in discussing the Bluff or Loess formation, he stated that fossil remains of the mastodon had been found in these deposits in making an excavation for a sewer in the neighborhood of Geyer Avenue and Eighth Street, in what was then the southern part of the city. They were discovered at a depth of about 16 feet and consisted of 2 nearly perfect teeth and fragments of bones. A musk-ox (Symbos cavifrons) and a horse (Equus complicatus) found here will be mentioned on pages 139, 181, 262. Fertile, Washington County (19). — From Mrs. P. G. Higginbotham, of Fertile, the IT. S. National Museum has received a brief description and a sketch of a mastodon tooth which was found many years ago. It is reported as dug up from sand in or near Big River. The tooth has 3 crests and is about 110 mm. long and 85 mm. wide. No other details are known. Arrowrock, Saline County (13). — George P. Purcell, of Arrowrock, states that several years ago he found a mastodon tooth in the northeast quarter of section 36, township 50 north, range 19 west. Mr. Purcell sent a pencil sketch which shows that the tooth was that of Mammut americanum; there were 4 crests and a large crest-like talon. The tooth was 8 inches long. After- wards another tooth was found near the same place. The locality is very near the right bank of Missouri River. The Kansas drift sheet seems to oc- cupy the immediate vicinity, and probably the teeth were left there at some post-Kansan period. Under this number 13 must be recorded a mention made in 1873, by F. B. Meek (Geol. Surv. Mo., 1855-1871, p. 183), of mastodon remains in Saline County. It seems that the bones and teeth had been found south of Black- water Fork, about 2.5 miles above the mouth of Ferris Creek, hence not far from the south line of the county. Meek saw some of these remains, but was not able to secure them. 32 MASTODONS. Kansas City, Jackson County (14). — In the Daniel B. Dyer Museum, Pub- lic Library building, in Kansas City, the writer saw the lower jaw of a mas- todon, which contains the penultimate and the last molars. With this jaw are an upper last molar, some portions of tusk, and the head of a femur. Mr- Edward Butts, interested in the collection of fossils, wrote that these remains were found at a point about 300 feet west of the intersection of First Street and Lydia Avenue, in the loess formation, and at a depth of about 20 feet be- low the undisturbed surface. The diameter of one tusk is given as 10 inches. About 15 years ago Mr. Butts obtained from a brickyard foreman a mas- todon tooth which had been found in the loess, about 500 feet south of the spot where the teeth and bones first mentioned were met with. The occur- rence of these remains in the loess at the depth given appears to indicate the late Iowan stage or the Peorian stage. H. H. West reported in 1880 (Kansas City Rev. ScL, Indust., vol. I, p. 336) the discovery, at Kansas City, of a portion of a tusk which he regarded as that of a mastodon. This was found in what he regarded as loess, on Camp- bell Street, about 150 feet north of Independence Avenue. He furnished a geological section of the locality. Broadhead (same Eeview, vol. iv, p. 521) mentioned this find and suggested that the tusk might have drifted from de- posits a little older, or, more probably been washed into the loess lake from the adjacent shore. It was then supposed that the loess was laid down in water. In 1880 (Kansas City Rev. Sci., Indust, vol. in, p. 643), Dr. F. A. Bal- lard, of Independence, recorded the finding of a supposed mastodon tusk in a "lick/' in the eastern part of Jackson County. The locality was not more exactly given ; it was said to be "twenty miles east of here," meaning probably east of Kansas City. Brothers by the name of Hulse were mentioned as liv- ing in the neighborhood. The tusk lay across the rut of a wagon road and was much damaged. A section of this 3 feet long and an ulna were secured and presented to Judge H. H. West. It is possible that these fragments are yet preserved in some museum. Line Creek, Piatt County (15). — In 1881, G. C. Broadhead reported (Kan- sas City Rev. Sci., Indust., vol. iv, p. 521) that J. C. Evans had obtained a large tooth of a mastodon from Line Creek. No other information was fur- nished. Broadhead thought that the tooth might have been washed out of glacial clay. The exact position of Line Creek is not known to the writer. Caldwell County (16). — On the page just cited, Broadhead noted that a mastodon tooth had been found in Caldwell County. The information is very vague. St. Joseph, Buchanan County (17). — The collection of the St. Louis Academy of Science contains a lower right hindmost molar, labeled as found near St. Joseph. No details are given. The outer half of the tooth is split off. The hinder root is extremely long, 150 mm. from the base of the crown to the tip. Clarksville, Pike County (18). — At a meeting of the St. Louis Academy of Science, held December 20, 1875, Albert Todd exhibited 2 teeth which he supposed belonged to Mastodon angustidens. They had been found in a creek which empties into Mississippi River at the upper end of Clarksville; and KANSAS. 33 they were brought to St. Louis by Mr. Jameson, of Louisiana, Missouri. These teeth are well preserved, and doubtless belonged to Mammut ameri- canum. For (19) seepage 31. KANSAS. (Map 2.) Little Osage River, Bourbon County (1).— The U. S. National Museum contains 2 mastodon teeth recorded as found along Little Osage River in Kansas, but no more exact locality is given. This would be somewhere near the northern boundary of Bourbon County. One tooth, No. 326 of the cata- logue, is part of a lower right hindmost molar; the other, No. 336, is part of the upper right last molar. The locality is entirely outside of the glaciated area and it is impossible at present to say in what part of the Pleistocene epoch the animal lived. Buffalo, Wilson County (10). — The American Museum of Natural History in New York contains a palate and both rami of the lower jaw, with teeth, be- longing to Mammut americanum, which were secured somewhere near Buffalo by Charles H. Sternberg. In the lower jaw all three molars, in fine condition, are present. The catalogue number is 1464. Emporia, Lyon County (2). — In 1866 (First Ann. Eep. Geol. Surv. Kansas, p. 14), B. F. Muclge stated that a large mastodon tooth, the lower last molar, had been found near Emporia and shown to him. Broadhead in 1880 (Kan- sas City Eev. Sci., Indust., vol. iv, p. 521), in speaking of discoveries of mastodons, made the brief remark that other bones had been found near Emporia. This region is outside of the glaciated area. Franklin County (3). — Professor William Wheeler, in 1878 (Trans. Kan. Acad. Sci., vol. vi, p. 11), reported that a tusk of a mastodon had been found in the bank of Eight Mile Creek and had been sent to the State University. There is no evidence to show that it was mastodon rather than elephant. The same author stated that a mastodon tooth had been found, some years pre- viously, in the same creek, about 5 or 6 miles south of the place where the tusk was discovered. Broadhead, the geologist (Kansas City Review, vol. iv, p. 521), stated that remains of the mastodon had been found in this county. No particulars were furnished. Osawatomie, Miami County (4). — In 1866 (First Ann. Rep. Geol. Surv. Kansas, p. 14), B. F. Mudge stated that a few years previously a part of a large jaw-bone of a mastodon, containing 3 teeth, had been fished up from a stream near this town. It was sent to Ohio and all traces of it lost. Lawrence, Douglas County (5). — J. Savage (Trans. Kan. Acad. Sci., vol. vi, p. 10) told of the discovery of a lower jaw of a mastodon a little below water-level on Wakarusa Creek^ southwest of Lawrence, on land then owned by Judge Lawrence. Near the same place were afterwards found bones of bison, elk, and antelope. This jaw is mentioned by Williston (Univ. Geol. Surv. Kan., vol. ii, p. 301). It is in the collection of the University of Kan- sas and has been examined by the writer. He regards it as belonging to the species Mammut progenium. The following are some of the measurements: 34 MASTODONS. Measurements (in millimeters) of lower jaw of Mastodon. Length from tip of jaw to rear of ascending ramus at level of teeth 812 Width of ascending ramus at level of teeth (outside) 284 From tip of jaw to front of second lower molar 248 Height of jaw at front of second lower molar 192 Length of symphysis 205 Length of second lower molar 110 Width of second lower molar 87 Length of third lower molar 162 Width of third lower molar 96 In the front end of the jaw are two sockets for tusks. The vertical diameter of the right one is 60 mm.; the horizontal, 50 mm. The lower floor of the socket is somewhat convex, showing that the tusk would have been bent down- ward. The jaw had been buried in blue clay and it lay at a level of about 25 feet below the present surface of the ground in the neighborhood. The bones of the other species mentioned were not associated with the jaw. They were found subsequently, evidently at some distance away. They were all in an unfossilized condition and were described as appearing to be bones of bison, elk, and antelope. It was concluded by the finders of the more recent bones that the mastodon jaw in a fossilized state had been washed into the old lake-bed in which the bison and elk bones had been covered up. This is not probable. Professor J. E. Todd (Trans. Kansas Acad. Sci., vols, xxm, xxiv, 1911, pp. 211-218) has given an interesting account of the history of Wakarusa Creek. According to his view, the Kansan ice dammed Kansas Eiver and directed its waters into the Wakarusa. He doubts that the Nebraskan ice reached that region. To the present writer it seems reasonably certain that there are Aftonian deposits in the valley of Kansas Eiver, at Lawrence, as shown by the presence of Equus laurentius there. He believes that an old pre-Kansan valley existed where the Kansas now flows and probably another where is found the Waka- rusa. These became partially filled during the Aftonian stage; and the deposits of the latter contain the early Pleistocene fossils that turn up there now and then. The Wakarusa mastodon jaw could not have been transported far without serious damage. Palermo, Doniphan County (6). — The writer has received letters and a photograph showing the discovery of remains of a mastodon, apparently Mam- mut americanum, near Palermo. The locality was reported as being in the east half of the northeast quarter of section 24, township 4 south, range 21 east. The bones and teeth were found by Albert Kienhoff, of Wathena, in a small creek on his farm, and were sold to George Groneweg, of St. Joseph, Missouri, from whom a photograph of one of the molars, apparently the lower right last one, has been received. A tooth of Elephas columbi and one of Equus niobrarensis have been found on the same farm. A newspaper account of the discovery of the skeleton states that bones and teeth, weighing all to- gether about 200 pounds, had been taken out. The remains were found in a ravine about a mile from the Missouri Eiver and about 50 feet below what was formerly the surface at that point. The bank had been eaten away by KANSAS — COLOKADO. 35 the stream and the bones were thus exposed. It appears evident that the remains discovered here were buried in Aftonian deposits. Manhattan, Riley County (7). — B. F. Mudge, in 1866 (First Ann. Eep. Geol. Surv. Kansas, p. 14), stated that the cabinet of the State Agricultural College contains a femur of a mastodon found at Manhattan. S. C. Mason, in 1883 (Trans. Kan. Acad. Sci., vol. vm, p. 13), mentioned incidentally that remains of elephants and mastodons had been found near Manhattan in the loess and on sand-bars. S. W. Williston, in 1897 (Univ. Kan. Geol. Surv., vol. II, p. 301), stated that he had knowledge of the finding of Mammut americanum at Manhattan. The present writer has seen in the collection of the Agricultural College at Manhattan a finely preserved lower left last molar of M. americanum. It has the catalogue No. 1878 and is labeled as found a mile south of Manhattan on a sand-bar in Kaw (or Kansas) Eiver. The tooth is 185 mm. long, has 5 cross-crests and a large pulp-cavity. The tooth had evidently just come into use. Jewell County (8). — Dr. S. W. Williston, as cited above, stated that he had knowledge that the American mastodon had been found in Jewell County. No particulars were given. McPherson, McPherson County (9). — This paragraph concerns, not the common mastodon, Mammut americanum, but a bunodont mastodon belong- ing to the genus Stegomastodon. The writer described this under the name of Gomphotherium elegans (Proc. U. S. Nat. Mus., vol. Mil, pp. 219-221, pi. xxvi ). The species was based on a single lower last molar found in a sandpit about 2 miles east of McPherson, at a depth of 35 feet. The sands found here belong to the Sheridan, or Equus, beds, and have been described by Haworth and Beede (Univ. Kansas Geol. Surv., vol. n, pp. 285-296, pi. xlv) and by Beede (Trans. Kansas Acad. Sci., vol. xv, pp. 104-110, pis. n, in, vi). A re-study of this tooth convinces the writer that it does not differ specifically from Leidy's species Mastodon mirificus, the type of which was found in the Pleistocene of northwestern Nebraska. The teeth hitherto referred to this species are much worn. The Kansan tooth had not yet come into use and therefore presented an appearance very different from that of the known worn teeth for study. The species is now known as Stegomastodon mirificus or Anancus mirificus. For (10) see page 33. COLORADO. (Map 2.) Fifteen miles northeast of Pueblo, Pueblo County (1). — In the collection of the State Historical and Natural History Society, at Denver, there is a fragment of a mastodon tooth labeled as found about 15 miles northeast of Pueblo. The collector was N. M. Tabor, of Denver, and the year apparently about 1875. The tooth is that of Mammut americanum. Golden, Jefferson County (2).— The City Park Museum, Denver, has an upper right second true molar of Mammut americanum, labeled as having been found in "Quaternary drift" at Golden. It is little worn and presents good large roots. 36 MASTODONS. IOWA. (Maps 2, 4.) 1. Within the Akea Covered by Kansan Drift. Salem, Henry County (1). — In Netta C. Anderson's list of the mastodon and mammoth remains found in Illinois and Iowa (Augustana Lib. Pubs., No. 5), Professor Frank Leverett reported that some mastodon teeth had been found in the vicinity of Salem. He had not himself seen the teeth, and there is the possibility that they belonged to one of the elephants. In the same list, Dr. J. M. Shaffer reported that two mastodon teeth had been dug up near the bank of Skunk River, in this county. In January, 1924, the writer received from Professor A. 0. Thomas, of Iowa University, photographs of an anterior dorsal vertebra of a mastodon, in all probability Mammut americanum, which had been found near Salem. The locality is, more exactly, near the center of section 18, township 70 north, range 7 west. The object was found in a bed of sand on the flood plain and to the north of a little stream, Big Cedar Creek. Professor Thomas cites Savage (Geol. Surv. Iowa, vol. xn, p. 289) for the statement that the deposit is of pre-Kansan age. This would be either Aftonian or Yarmouth. The specimen consists of the dorsal spine, broken at the neural arches and dis- playing the postzygapophysial surfaces. The length of the specimen is close to 18 inches. The writer finds that these articular surfaces are more widely separated in the mastodons than in at least the existing elephants. Mount Pleasant, Henry County (2). — In Netta C. Andersons' list, above mentioned, Professor T. A. Savage reported the finding of some bones and teeth of a mastodon in a well near or in Mount Pleasant. These remains are said to have been met with in or immediately below the Kansan drift. They were placed in the collection of the Iowa Wesleyan University, at Mount Pleasant. It is not probable that the bones of this animal were buried in glacial drift; it is more likely that they were lying in a deposit of Aftonian age. In 1922 (Proc. Iowa Acad. Sci., vol. xxviii, p. 65, fig. 5a), Professor Jaques reported the discovery of a mastodon tooth, an upper hindmost molar, about a mile east of Mount Pleasant. The more exact locality is section 14. township 71 north, range 6 west. It was found in the digging of a shallow well, at what is known as the Roy Spring. The length is 8 inches, the width 4.5 inches. The roots appear to be nearly complete. With it were found a vertebra, a femur lacking the ends, and 2 small tusks. These were 24.8 inches and 26.75 inches long. Wayland, Henry County (34). — In 1923 Professor A. 0. Thomas reported to the writer the discovery of a mastodon tooth in a stream-bed spoken of as Sugar Creek, north of Wayland. The locality is possibly in Washington County. In the same stream-bed was found a tibia of Megalonyx jeffersonii (p. 7). Trenton, Henry County (3). — From Professor H. E. Jaques, of Iowa Wes- leyan College, at Mount Pleasant, the writer received information of the find- ing of a mastodon tooth in a small creek, a tributary of Skunk River, about 10 miles northwest of Mount Pleasant. This would be evidently near Trenton. IOWA. 37 The tooth is said to have a length of about 4^ inches and a width of about 2y2. It is probably the second true molar. As the height is given as 4 or 5 inches, the roots are probably present. The finder reported that a year or two previously a tusk had been found in the same creek. Milton, Van Bur en County (4). — In the palseontological collection at Iowa University is a lower left last molar, recorded as found in the bed of Chequest Creek, near Milton, by W. B. Bell, about the year 1890. According to Gor- don's report on this county (Iowa Geol. Surv., vol. iv, pp. 201-254, map), the point on Chequest Creek nearest to Milton is about 6 miles north. Exactly where the tooth was found is not known. The county is within the area oi the Kansan drift; but as there are two kinds of drift, an upper yellow clay and a lower blue clay, the Nebraskan drift may be represented and between them some remains of the Aftonian. Again, there is a covering of loess in the region and the tooth in question may have been derived from this. Selma, Van Buren County (5). — In the collection of the Iowa Historical Department, at Des Moines, is a right humerus which the writer examined and referred to Mammut americanum, although the side-to-side diameter at the constricted part of the shaft is somewhat greater than it is in either of two mounted specimens of mastodon in the IT. S. National Museum. In Elephas primigenius and the existing elephants this diameter is much less than in the Selma humerus. The remarks regarding the mastodon found north of Milton may be applied to this specimen. Eldon, Wapello County (6). — According to Professor M. F. Arey, of the State Teachers College, at Cedar Falls, there is in the collection of that school a proboscidean femur which was found in the gravels along Des Moines River, near Eldon. From descriptions, the writer regards it as belonging to Mam- mut americanum. No certain conclusion can be reached regarding its geologi- cal age. Ottumwa., Wapello County (7). — In the collection at the Iowa Wesleyan University, at Mount Pleasant, are two proboscidean ribs which the writer has examined and referred to Mammut americanum. They are labeled as found at Ottumwa and as presented by Rev. E. C. Brooks. Portions of the pelvis in the same collection, labeled as found in Des Moines River, in 1859, probably belonged to the same individual mastodon. From some locality about 6 miles south of Ottumwa there was reported, in 1879, (Kansas City Rev. ScL, Industry, vol. m, p. 242) by Mr. Houbler, the discovery of an entire tusk of a mastodon. This, however, may have belonged to one of the ele- phants. Mahaska County (8). — In the collection of the Iowa State Historical De- partment, at Des Moines, is an upper right last molar of Mammut ameri- canum, labeled as found in this county by J. D. Davis, of Des Moines. Un- fortunately no more exact information was left on record. A specimen thus collected is not of great scientific value. New Virginia, Warren County (9). — In Howe's Annals of Iowa, volume n, 1883, page 102, A. R. Fulton described, under the name of Elephas ameri- canus, a tooth which (as the description plainly shows and as appears to have been recognized by Fulton) was that of Mammut americanum. The tooth 38 MASTODONS. had been found in Limestone Creek, 1.5 miles west of the village of New Virginia. The youngest drift in this county is the Kansan. According to Dr. J. L. Tilton's report (Geol. Surv. Iowa, vol. Y, pp. 301-359, map), this drift is usually overlain by loess which may attain a thickness of several feet. It shows indications of being composed of an upper and a lower bed. These beds of loess are of course younger than the Kansan drift and may represent the Sangamon, or the Peorian, or possibly both, interglacial stages. Some geologist examining the locality may yet be able to determine the geological age of the tooth. Clarinda, Page County (10). — In 1870, C. A. White (Iowa Geol. Surv., vol. I, p. 353) wrote: "In the valley of the Nodaway, near Clarinda, some teeth of that huge extinct animal, the mastodon, have been found." It may be mentioned here that Samuel Calvin reported (Iowa Geol. Surv., vol. xi, p. 413) that some large bones which, from the description given, must have belonged to the mammoth or the mastodon were brought up from a depth of 90 or 95 feet, with pieces of bark and wood. This was near or at Blanchard. These things appear to have been buried in an old pre-glacial valley. It is impossible to say to which of the proboscideans these bones be- longed; and it is probable that they were not preserved. The items under this number were omitted in the writer's paper in volume xxiii of the Iowa Geological Survey. Missouri Valley, Harrison County (11). — In the Cox sand-gravel pit, about 2 miles southeast of Missouri Valley, were found remains of two species of Mammut, M. americanum and M. progenium; see Iowa Geological Survey, vol. xxiii, pp. 54, 383, 384). Associated with these were 17 other species of mammals, a list of which is given on page 296. Mount Pisgah, Harrison County (12). — Kemains of the genus Mammut have been found in the Peyton sand-and-gravel pit a mile southwest of Pisgah. The lower jaw occurring there (Calvin, Bull. Geol. Soc. Amer., vol. xx, p. 352, pi. xxv, fig. 2) was referred by the writer to Mammut progenium (Iowa Geol. Surv., vol. xxxni, p. 372). The associated species are named on page 295. Denison, Crawford County (13). — In the palaeontological collection at the University of Iowa is a tooth of Mammut americanum, discovered in a sand- pit near Denison. In 1909, (Bull. Geol. Soc. Amer., vol. xx, p. 352) Calvin reported a tibia and fragments of 2 tusks from Denison. The tusks can only provisionally be referred to the mastodon. It is not known from which of two differently situated pits at Denison these fossils were obtained (p. 298). Turin, Monona County (14). — Various Pleistocene mammalian remains have been collected at Turin, in the Elliott gravel-pit. The locality and its fossils have been described by Shimek (Bull. Geol. Soc. Amer., vol. xx, pp. 344, 345, and vol. xxi, p. 129; Iowa Geol. Surv., vol. xxi, p. 340) and Hay (Iowa Geol. Surv., vol. xxiii, p. 59). The pit is situated in the northeastern part of the village. According to Shimek, Aftonian sands and gravels are ex- posed to a depth of about 12 feet. The superimposed formations are the Kansan drift, the Loveland, and the loess. The bones and teeth were found in the lower part of the exposed Aftonian. A list of the associated species may be seen on page 297. IOWA. 39 Castanea, Monona County (15). — Supposed remains of Mammut have been found in the vicinity of Castanea. The Ordway sand-and-gravel pit is sit- uated on the bluff on the southwest side of Maple River, opposite Castanea, on the southeast quarter of section 13, township 84 north, range 44 west. The following section at this pit is given by Shimek. 5. Loess, abundant on ridge above the pit. 4. Loveland, 5 to 6 feet. 3. Kansan drift, 6 to 18 feet. 2. Aftonian, fine and cross-bedded, 5 to 8 feet. 1. Aftonian, gravel. No vertebrates are recorded from this pit. In the collection of the Iowa. State Historical Department, at Des Moines, is a large tusk labeled as found at Castanea, (Iowa Geol. Surv., vol. xxm, p. 387) ; it is not certain that it belonged to a mastodon. In the collection at the University of Iowa is a part of a scapula found in another pit not far away from the Ordway; it, too, is not generically identifiable. Mapleton, Monona County (16). — In the Wilkenson well, near Mapleton, a large tusk, nearly 8 feet long on the outer curve, a molar, and fragments of cranial bones of Mammut americanum were found. These remains are now in the State University, at Iowa City. The well is situated in the northwest quarter of section 6, township 85 north, range 42 west. The remains were found at a depth of from 35 to 40 feet, in loose sand and gravel. Their age is taken to be Aftonian (Iowa Geol. Surv., vol. xx, p. 343; vol. xxm, pp. 68, 386). Le Mars, Plymouth County (17). — The record from this locality is doubt- ful. The collection at the State University of Iowa contains an imperfect innominate bone which may belong to Mammut americanum, but more prob- ably to one of the elephants. In a gravel-pit near Le Mars, Professor G. F. Kay secured a large fragment of a limb-bone of a proboscidean; it is not identifiable. The age of the interglacial deposit is probably Aftonian (Iowa Geol. Surv., vol. vm, p. 336; vol. xxm, pp. 69, 389). Akron, Plymouth County (18). — The mastodon to be noticed here is the one which bears the specific name, mirificus, referred at various times to the genera Mastodon, Mammut, Rhabdobunus (Hay, Iowa Geol. Surv., vol. xxin, p. 373), and Stegornastodon (Pohlig, Bull. Soc. beige Geol. Paleont., vol. xxvi, P. v, p. 193) and Anancus. In a well 2 miles east of Akron, fragments of tusks, 2 teeth, and some frag- ments of cranial bones were found at a depth of 20 feet. These are now in the collection of the University of Iowa, at Iowa City. This species was origi- nally based on jaws supposed to come from Pliocene deposits in Nebraska, but which are now known to have been collected from the so-called Equus beds. Shimek (Bull. Geol. Soc. Amer., vol. xxi, p. 126) showed that the deposits at Akron belonged to the Aftonian. It thus appears that the genus continued on from the Pliocene (Blanco) into the early Pleistocene. From the same well was obtained a bone composed of the united first and second phalanges of the third digit of the hind-foot of the ground-sloth Megalonyx (p. 7). 40 MASTODONS. Do on, Lyon County (19). — The writer has recorded (Iowa Geol. Surv., vol. xxni, p. 38G) the discovery of 2 large tusks at Doon, on the Great Northern Railroad. They were unearthed by a steam-shovel at a depth of about 25 feet. These tusks may, however, have been those of an elephant. They were much broken up, but the fragments are probably preserved in the Sioux City Academy of Science. Their generic identity may at some time be determined by microscopic examination of their structure. The deposit containing these tusks belongs quite probably to the Aftonian. Clinton, Clinton County (20). — In the collection of the Chicago Academy of Sciences is an upper left last molar of Mammut americanum, said to have been found near Clinton, on the west bank of Mississippi Eiver. It was pre- sented to the Academy by the geologist, J. W. Foster. The area occupied by the town is mostly recent alluvium but there is some Kansan drift. The latter is the most recent drift everywhere immediately back of the town; but the Iowan drift, according to J. A. Udden's map (Iowa Geol. Surv., vol. xv), is not far away. Besides these formations, the drifts are capped by loess, which may reach a thickness of 20 feet. The tooth may therefore have been found in the loess or in an interglacial deposit. The only information accom- panying the tooth is that it was found near Clinton. Bryant, Clinton County (21). — From Louis Eockrohr, living near Bryant, the writer received a photograph of a lower left last molar of Mammut ameri- canum unearthed at a depth of 8 feet below the original surface of the ground. The tooth was described by the writer in his paper on the Pleistocene mam- mals of Iowa (Iowa Geol. Surv., vol. xxm, p. 381). Bryant is within the area of the Kansan drift. Beneath this drift might occur Aftonian interglacial deposits. It seems not unlikely that a competent geologist might determine the age of the gravels in which the tooth was found. It had not yet been cut and shows no signs of abrasion by transportation. It is not probable that it was carried by water for any considerable distance before it was buried. Maquoketa, Jackson County (22). — In J^etta C. Anderson's list, already referred to, Professor W. H. Norton, of Cornell College, Mount Vernon, Iowa, reported that an atlas and two other vertebrae of some proboscidean had been found near Maquoketa and presented to his college. Maquoketa is on the river bearing the same name and is within the strip of Kansan drift which lies between the Iowan drift and Mississippi River. Garber, Clayton County (23). — The writer has received from Professor A. 0. Thomas, of the University of Iowa, a photograph of an upper right second molar of Mammut americanum, found in a gravel-pit 2 miles east of Garber. The exact locality is section 32, township 92 north, range 3 west. According to Professor Thomas the age of the gravels is uncertain, being either Kansan drift or Iowan valley trains. 2. Within the Area Covered by Illinoian Drift. Fort Madison, Lee County (24). — The collection at Iowa Wesleyan Univer- sity, at Mount Pleasant, contains a complete last lower molar (with fine roots) of Mammut americanum. Charles Buetner, of Burlington, who presented this tooth, informed the writer that it was found in a creek 3 miles due west from Fort Madison. This would be very close to the western border of the Uli- IOWA. 41 noian drift, possibly beyond it. One might suspect that some Yarmouth inter- glacial occurs there. Lost Creek, Lee County (25). — In Netta C. Anderson's list, Professor Frank Leverett reported that he had seen a large leg-bone of a mastodon, found in the valley of Lost Creek, section 3 or 4 of Washington township, and which he thought belonged to Justus M. T. Myers, of Fort Madison. Wash- ington township is No. 68 north, range 4 west. On the same page, Mr. Myers himself reports a leg-bone (probably the same as that mentioned by Leverett), together with a piece of tusk, of a mastodon found in Lost Creek. These bones might well have been those of one of the elephants. The locality is about 2 miles southeast of Denmark. This locality is not far from the west- ern border of the area of Illinoian drift in Iowa; and it seems probable that the creek might there have cut into the Yarmouth interglacial. On the other hand, these bones may have belonged to Elephas primigenius and been buried in the Illinoian drift. It is to be hoped that the bones are yet preserved by somebody, so that they may hereafter be identified with some certainty. Burlington, Des Moines County (35). — In the U. S. National Museum is a cast (No. 10939) of a lower right hindmost molar of Mammut americanum found at Burlington many years ago. The cast was given to Professor S. F. Baird by Dr. Charles A. White, once state geologist of Iowa. The tooth was a large one, length 180 mm., width 96 mm., with 5 crests and a small talon and 2 large roots. It was only slightly worn. The number 35 is not on map 4. Wilton, Muscatine County (26). — In 1899 (Iowa Geol. Surv., vol. ix, p. 352), J. A. Udden quoted a report made by Samuel Calvin, in 1874, to the president of the University of Iowa, on some proboscidean remains found near Wilton. These consisted of 12 vertebras, 13 ribs, a segment of the sternum, parts of both innominate bones, a femur, a tibia, a scapula, a part of a humerus, and a number of foot-bones. The writer figured the scapula (Iowa Geol. Surv., vol. xxm, pi. lii, fig. 1) and identified it as belonging to Mam- mut americanum. Calvin, in the report referred to, gave the dimensions of this bone and of some others. His measurement of 35 inches as the length of the tibia must be erroneous. These bones were found in the south bank of Mud Creek, about half a mile south of Wilton, in probably the northwest corner of section 12, township 78 north, range 2 west. Where the bones were found the bank rose about 30 feet above the water. The several pieces of the skeleton were scattered, but all lay at the same level. Calvin did not state how high above the water they were buried. The deposits containing the bones consisted of alternating strata of very fine sand and clay. It was Udden's opinion that this deposit had been laid down in quiet water during the Sangamon stage; he reported, further, that it underlies the Iowan loess. Davenport, Scott County (27). — In 1899 (Iowa Geol. Surv., vol. ix, p. 356), J. A. Udden stated that mastodon remains had been taken from the San- gamon soil resting on Illinoian till in the west part of Davenport. The writer has seen no other reference to this discovery and the word mastodon may have been unintentionally used instead of mammoth. Udden, as cited, presented a long list of diatoms found in the mud at this locality. 42 MASTODON'S. 3. Within the Area Covered by Iowan Drift. Shellsburg, Benton County (28). — In Netta C. Anderson's list, on page 25, Professor T. E. Savage reported that a mastodon tooth had been found by J. Grubb, in 1903, in the alluvium of Bear creek. J. A. Burns, of Shells- burg, a son-in-law of Mr. Grubb, wrote, on February 11, 1911, that the tooth in his possession, is about 6 inches long, and has 4 cross-crests and a talon. His sketch shows well that the tooth belonged to the mastodon. It was found lying in the bed of a small creek, in section 33, township 85 north, range 9 west. This is about 3 miles northwest of Shellsburg. It seems to be unprofitable at present to speculate on the geological age of this tooth. It belongs more probably to a post-Kansan interglacial stage than to any of the glacial stages. An account of the Pleistocene geology of this county has been given by Professor T. E. Savage (Iowa Geol. Surv., vol. xv, pp. 201-217, with map). Springville, Linn County (29). — In Netta C. Anderson's list, Professor W. H. Norton, of Cornell College, reported that a small molar and the crown of a large molar had been found near Springville, in or on Iowan drift. These teeth are now in the collection of Cornell College, at Mount Vernon, Iowa. Judging from our present knowledge of the mastodon, we may conclude that the animal in question lived in that region more probably after than during the Iowan glacial stage. For an account of the Pleisto- cene geology of this county, Professor W. H. Norton's report may be con- sulted (Iowa Geol. Surv., vol. iv, pp. 168-174). In the same list, Professor F. C. Baker reported that there is in the collection of the Chicago Academy of Sciences a part of a tusk found in a gravel-pit at Bertram, Linn County. This tusk might, however, belong to one of the elephants. 4. Within the Area of the Wisconsin Drift. Des Moines, Polk County (30). — The collection of the State Historical Department, at Des Moines, contains an upper last molar of Mammut ameri- canum, apparently found at Des Moines and presented by L. Hamilton in 1883. The number of the tooth is 4519. The record as to locality, kind of deposit, and depth is not sufficiently exact to permit a decision as to its geo- logical age. In the immediate vicinity of Des Moines are found Kansan and Wisconsin drifts and loess. Probably also interglacial deposits may be found. Dr. James H. Lees sent the information that about 2 years ago 2 mastodon teeth were found in South Des Moines, in the gravel of the Eac- coon Valley. They may have been in Wisconsin outwash or in older de- posits. A&el, Dallas County (31). — In 1911 (Bull. Geol. Soc. Amer., vol. xxn, p. 215), Calvin reported that in 1876 a complete skeleton of a mastodon was found at Adel. It was met with in a peat deposit which partly filled a de- pression on the surface of the Wisconsin drift. It is not known what became of this valuable specimen. In this case one can not be in doubt as to the geological age of the animal. It had undoubtedly lived after the passing away of the last glacial ice sheet. Rippey, Greene County (32). — In the collection of the Iowa State His- torical Department, at Des Moines, there is a right scapula which the writer NEBRASKA. 43 identified as that of a mastodon. It is reported to have been found at Rip- pey, but no detailed information is furnished. Inasmuch as the whole of Greene County is covered by Wisconsin drift, it is probable that the bone belonged to an animal which lived after the Wisconsin ice sheet had retired from that region. Boone County (33). — The same collection contains various remains which the writer referred (Iowa Geol. Surv., vol. xxm, p. 380) to Mammut ameri- canum. There is no record to show from what place in the county they were obtained. It may be regarded as certain that they were found in deposits overlying the Wisconsin drift. For (34) see page 36 ; for (35) see page 41. NEBRASKA. (Map 2.) Pender, Thurston County (1). — The C. H. Morrill collection, in the State Museum at Lincoln, formerly contained part of a skull of a mastodon, found at Pender, but this was destroyed in a fire in 1912. The remarkable feature about the skull was the tusks. These had a length of 2,700 mm., somewhat more than 8 feet, and instead of being strongly curved, as they usually are, they were nearly straight. In other respects the skull appeared to resemble that of Mammut americanum. Two feet from the base the diameter of the tusk was 140 mm. The lower jaw was not present. Each side of the upper jaw had all three molars, the first (M1) being much worn; the second (M2) only moderately so; the third (M3) not at all. No information has been obtained regarding the exact place or depth of burial of the animal or character of deposit. The region is occupied by the Kansan drift. It might be supposed that the skull had been found in some Aftonian interglacial de- posit, but there is no assurance that such is the case. The region is covered with a deposit of loess and the skull may have been buried in this and hence may have existed at a later stage of the Pleistocene. Pickrell, Gage County (2). — In the collection of the State Museum, at Lincoln, is an upper left second molar found near Pickrell, on the farm of Wilke Jurgens, 4 miles east and a mile south of the town. The tooth is sup- posed to be that of Mammut americanum. Nothing is known about the na- ture of the deposits in which the tooth was buried ; one can only surmise that the geological age is post-Nebraskan. The county is within the area occupied by the Kansan drift, but the region about this town is covered with loess. Sutton, Clay County (3). — The collection of the State Museum also con- tains a lower jaw of a mastodon supposed to be Mammut americanum and found in a gravel-pit at Sutton. The symphysis of the jaw and the greater part of the ascending rami are restored in plaster. The hindmost milk-tooth is present, with good roots. The first molar, 95 mm. long and 64 mm. wide, had just begun to wear; the second molar is deep down in the jaw. This re- gion is occupied by the Equus, or Sheridan, beds. The animal represented by the jaw probably lived during the Sheridan stage. From the same region has been obtained Elephas imperator. Buffalo County (4).— In 1866 (Proc. Acad. Nat. Sci. Phila., p. 316), Dr. F. V. Hayden announced the finding of a tooth of a mastodon in the "post- 14 MASTODONS. Pliocene drif t,?J> near Fort Kearney. The exact locality is not recorded and probably never will be known. It may have been in Kearney County, south of Platte River. Valentine, Cherry County (5). — The Minneapolis Journal of June 4, 1911, announced that during the previous year the head and some cervical vertebrae of a mastodon had been sent from Valentine to the State University of Min- nesota. The upper jaw contained teeth and the tusks; lower jaw missing Professor F. W. Sardeson informed the writer that the animal is Mammui arnericanum and that a compact gray sandstone adhered to the teeth. The animal probably lived during the Sheridan stage. Seneca, Thomas County (6). — In 1858 (Proc. Acad. Nat. Sci. Phila., vol. x, p. 10), Leidy described briefly his Mastodon mirificus. In 1869 (Jour. Acad. Sci. Phila., ser. 2, vol. VII, pp. 249, 396, pi. xxv) it was more fully characterized and illustrated. The remains appear to have been found on Middle Loup Eiver, probably not far from Seneca. One part of the type specimen is in the U. S. National Museum, the other in the Academy at Phila- delphia. Other species collected in the same region are listed on page 304. The species now bears the name Anancus mirificus or Stegomastodon mirificus. MINNESOTA. (Maps 2, 29.) Fairhaven, Stearns County (1). — In 1910 (Bull. Minn. Acad. Sci., vol. iv, p. 416), Professor N. H. Winchell recorded the finding of what he regarded as a large elephant's tooth in a marsh near Fairhaven. The statement made by Winchell that the tooth had a hard, shining surface awakens the suspicion that the tooth was that of a mastodon. Evidently Winchell had not seen the tooth, but relied on the statements of E. E. Woodworth, of Minneapolis, from whom the writer received a letter, dated September 13, 1912. He was a boy 8 years old at the time of the discovery. His father, a studious and observing man, yet living in 1912, had examined the remains, which included many bones. He was of the opinion that they belonged to a mastodon. The re- mains were not preserved by the finder and at this day it is impossible to settle the question. The statement made regarding the surface of the tooth does not apply well to the tooth of an elephant, but does apply to the enamel of a tooth of a mastodon. Stillwater, Washington County (2). — In the Sixth Annual Report of the Minnesota Geological Survey, 1877, page 61, N. H. Winchell gave an account of the finding, at this place, of a tusk supposed to be that of a mastodon. A later discussion of the discovery was presented in 1888 (Winchell Final Rep. Geol. Nat. Hist. Surv. Minn., vol. n, pp. 397-398). This tusk was found near or in the town of Stillwater, in the valley of Brown's Creek. The de- posit is referred to by Winchell as "drift deposits and river gravels." They were probably not deposited directly by the glacial ice, but by a stream which flowed from it. Winchell evidently regarded it as belonging to the close of the Wisconsin stage. The tusk may have belonged to an elephant. It was preserved in the St. Paul Academy of Sciences, but later was destroyed in a fire. In 1910 (Bull. Minn. Acad. Sci., vol. iv, p. 416), Winchell again re- ferred to this tusk and offered the opinion that it might have been that of MINNESOTA — SOUTH DAKOTA. 45 an elephant. The discoverer of the tusk affirmed that in the same deposits were found some specimens of pottery. Winchell did not doubt the correct- ness of the finder's account and believed that the pottery had been included in the drift gravel as native and constituent parts of it ; he conceded that pos- sibly they had fallen down from more recent deposits. Northfield, Rice County (3).— In 1884 (Final Eep. Geol. Nat. Hist. Surv. Minn., vol. i, p. 670) N. H. Winchell published a letter relating to the dis- covery, in 1879, at Northfield, of a tusk supposed to be that of a mastodon. It was exposed by some workmen while digging in a drift deposit. The tusk was then in Carleton College. It may have been that of an elephant, and the supposed drift was probably a valley gravel. The town is on Cannon Eiver; and on WinchelPs map (pi. xxxi) that valley is represented as being occupied by gravel, sand, and valley drift. In 1910 (Bull. Minn. Acad. Sci., vol. iv, p. 416), Winchell again mentions the tusk and says that it may have belonged to an elephant. Mankato, Blue Earth County (4). — From D. L. Bose,, of Mankato, the writer received an account of a head of a mastodon, found at Mankato and deposited in the high school of that town. In January 1914 the writer visited the school and found the specimen. It consisted of a piece of a lower jaw containing part of a badly broken tooth. The jaw was found at a depth of 15 feet in digging a sewer, at the intersection of State and Byron streets. The valley of Minnesota Eiver is filled with a modified drift (Final Eep. Geol. Nat. Hist. Surv. Minn., vol. I, p. 444). The animal had probably lived there after the recession of the Wisconsin drift sheet. This is probably the mas- todon tooth mentioned by Winchell in 1910 (Bull. Minn. Acad. Sci., vol. iv, p. 417). Albert Lea, Freeborn County (5). — N. H. Winchell, in 1884 (Final Eep. Geol. Surv. Minn., vol. i, p. 386), reported that according to William Morin the jaw of a mastodon had been found here, some years before that time, in a gravel-bank. In 1881 it was lost in a fire in the capitol at St. Paul. Ac- cording to Upham (Final Eep. Minn. Geol. Surv., vol. I, p. 526), the whole county is covered with drift deposits of Wisconsin age. Minnesota City, Winona County (6). — Professor Winchell (op. cit., p. 264) stated that at this place the remains of a mastodon had been taken out of the alluvium. This (his p. 261) appears to have been on a terrace elevated about 58 feet above the flood-plain of the Mississippi. Winchell, in 1910, re- ferred to this find, and stated that the remains consisted of a tusk and that this might have belonged to an elephant (Bull. Minn. Acad. Sci., vol. iv, p. 446). SOUTH DAKOTA. (Map 2.) Pine Ridge, Shannon County (1). — In the U. S. National Museum is an upper right hindmost molar of Mammut americanum, from near Pine Eidge. It was discovered about a mile south of the town, in a gravel ridge. It is in a good state of preservation and is credited to Major William H. Clapp. No details are recorded as to conditions of burial. Perkins County (2). — From Doane Eobinson, superintendent of the de- partment of history, South Dakota, the writer has received a photograph of 46 MASTODONS. 3 teeth of Mammut americanum which were found in the southeast corner of Perkins County, on Moreau River, 4 or 5 miles below the mouth of Rabbit Creek. These appear to be upper grinders, one a hindmost molar; the other two are second molars. The number 2 is not on the map. MONTANA. (Map 2.) Pryor Creek, Crow Reservation (1). — The writer has seen a tooth of Mam- mut americanum, found at some place along Pryor Creek. It was the lower left penultimate molar, moderately worn and furnished with complete roots. It had been met with in a ditch, made for reclamation purposes, at a depth of 18 feet. Exact details regarding the locality are wanting. Pryor Creek is a few miles east of the east line of Carbon County. Diamond City, Broadwater County (2). — In St. Louis University, St. Louis, Missouri, is an upper right hindmost molar of a mastodon labeled as found at Diamond City and as presented by Rev. J. G. Venneman ; also, from the same place, a tooth of Elephas columbi. FINDS OF ELEPHAS BOEEUS IN THE MIDDLE REGION OF NORTH AMERICA. This is the species of Elephant which in the writer's work of 1923, on the Pleistocene of North America east of Mississippi River, etc., is called Elephas primigenius. TEXAS. (Map 5.) Temple, Bell County (1). — From near this place Mark Francis sent the writer for examination a lower left antepenultimate milk-molar of Elephas boreus, found by W. S. McGregor, in his gravel-pit, on Leon River, 5 miles west of Temple and a mile north of Shallow Ford. In this same pit, 40 feet above the river, were remains of a mylodon (p. 2), a tapir (p. 155), a camel (p. 161), and of Equus semiplicatus (p. 129). The elephant tooth (Proc. U. S. Nat. Mus., vol. lviii, p. 138, pi. x, figs. 3, 4) is only slightly damaged; it presents 7 plates, with front and rear talons ; length 62 mm., width 32 mm. Bulverde, Bexar County (2). — In 1920 (Proc. U. S. Nat. Mus., vol. lviii, p. 136, pi. ix, figs. 1-7; pi. figs. 3-7; pi. xi, figs. 1-6), the writer described remains of elephants found near Bulverde, which he referred to Elephas primigenius but which are now regarded as belonging to Elephas boreus. The animals described were young, one of them retaining the first milk-molar. The second milk-molars, both upper and lower, are beautifully preserved. These remains were found in a large cave about 35 feet under the surface, entered only by a perpendicular shaft. All together, 18 species of vertebrates were discovered. See list of these on page 247. MISSOURI. (Map 5.) Osage River, Benton County (1). — In the collection of the Academy of Natural Sciences of Philadelphia is an upper right last molar which the writer identifies as that of Elephas boreus. It is worn down in front to the base and backward to about the sixteenth plate. The length along the base is 235 mm. ; height of the sixteenth plate, 120 mm. How this tooth came into the col- lection is not known. It may have been a part of the collection made by Al- bert Koch, in which case it was probably found at the Pomme de Terre lo- cality (p. 25) ; more probably it belonged to a collection made about 1843 by S. H. Whipple, along Osage River (p. 28). As to the geological age of the tooth, one can only indulge in surmises. It seems probable, from what is known regarding the habits of the species, that the individual lived in that region during one of the glacial stages. The Kansan drift is not far away. There is another tooth in the collection of the Academy from this county and probably from the same locality. It is a lower right last molar, in fine condition. Probably one or two plates are missing from the rear. The length along the base is 270 mm.; height of the eleventh plate (the first unworn one), 170 mm.; greatest width, 105 mm. On a lateral face there are 8 plates in a line 100 mm. long. The collection contains still another molar which appears to belong here. There are present 14 plates, but some of the rear ones are missing. The length of the part present, measured at right angles 47 48 KLKPHAS BOKEUS. with the edges of the plates, is 190 mm. The sixth plate is 200 mm. high. Another fragment of a tooth, with 11 plates, seems likewise to belong to Elephas boreus. KANSAS. (Map 5.) St. George, Pottawatomie County (1). — In the collection of the Kansas Agricultural College, at Manhattan, the writer has examined a part of an upper right molar found, probably many years ago, at St. George. It is labeled "Drift (Loess), on bank of Black Jack Creek, St. George, Potta- watomie Co., Kan. Wm. Dalton, collector." The tooth has 12 ridge-plates; a few in front are missing. It is probably a second molar. The plates are flat and the hinder ones do not make an angle with those farther in front. There are 10 plates in a 100 mm. line. The enamel is somewhat thicker than usual. The writer identifies the animal as Elephas boreus. Exactly where Black Jack Creek is the writer can not determine. The region north of Kan- sas River is within the area of the Kansas drift sheet; but the deposits along a stream might belong to a later stage. Pendennis, Lane County (2). — Among the teeth of elephants, collected by C. H. Sternberg, 7 miles northeast of Pendennis and therefore probably near the line between Lane and Ness Counties, are some which the writer is com- pelled to refer to Elephas boreus. Among these are two now in the American Museum of Natural History, New York. These appear to be upper second molars, right and left, of the same individual. All the ridge-plates are present and each tooth is worn back to the eleventh plate. The length of one tooth, from the base in front to the base behind, is 250 mm. ; that of the other tooth is slightly less; thickness 85 mm.; height of eleventh plate 155 mm. There are 18 or 19 plates, besides a front and one or two rear talons; 8 plates are crossed by a 100 mm. line. The plates are not warped, but rise from the base with gentle undulations. The hinder plates make hardly any angle with the anterior ones. The size and structure of this tooth are very much as in the tooth of E. roosevelti, found in Milwaukee, Wisconsin, and figured by the writer in 1914 (Iowa Geol. Surv., vol. xxm, p. 409, pi. lix) as E. primigenius. Possibly it belongs to E. roosevelti rather than to E. boreus. At the American Museum are other teeth which must be referred to E. boreus. Three are lower milk-molars, having 11 plates and anterior and posterior talons. The length is 108 mm.; width, about 45 mm. In the Uni- versity of Kansas are similar teeth. One of the best-preserved appears to be a lower last milk -molar. Its length along the base is 132 mm. ; there are 12 ridge-plates. In Ward's Natural History Establishment at Rochester, New York, the writer has seen 2 teeth, a last upper molar, and a lower penultimate, which are labeled as coming from Lane County. They may have been a part of Sternberg*^ collection. They have thin plates, 10 in a line 100 mm. long, and are otherwise not to be distinguished from teeth of E. boreus. Remarks on the geological aspects of this locality are to be found on page 270, where the remains of E. columbi from the same place are considered. IOWA. 49 IOWA. (Maps 5, 6.) 1. Within the Area of the Kansan Drift. Ottumwa, Wapello County (1). — In the collection of the Iowa Wesleyan College, at Mount Pleasant, is an upper first molar of Elephas boreus, found, as reported by Professor H. E. Jaques, about 1900, in Des Moines River, at Ottumwa. The tooth has 12 or 13 plates and front and rear talons. Ac- cording to A. G. Leonard (Iowa Geol. Surv., vol. xn, pp. 472-475), the only drift materials seen in this county belong to the Kansan; but there are rea- sons for suspecting that the Nebraskan underlies this. In the vicinity of Ottumwa the Kansan averages a thickness of about 25 feet. This is overlain almost everywhere by a deposit of loess, which in the neighborhood of Ot- tumwa has a thickness of about 8 feet. Before this was laid down the upper surface of the Kansan had suffered much weathering and probably erosion. It can hardly be supposed that the tooth in question got into the river from the body of the Kansan drift. It is possible that it was derived from ma- terials laid down in some unobserved late Aftonian or early Kansan deposit. It is, however, impossible to determine the geological age of this elephant. Brighton Township, Washington County (2). — In 1883 (Proc. Davenport Acad. Sci., vol. m, p. 177), J. Gass and W. H. Pratt gave an account of the discovery of some teeth and bones of Elephas primigenius, on the farm of Jerry Hopping (misspelled Hoppin), in Washington County. The location was given as section 14, township 22, range 3, but this was an error; for this would locate the place near the southern border of Missouri. The writer has learned that the farm occupies the northwest quarter of section 14, township 74 north, range 8 west of the fifth principal meridian. It is in the southern part of the county and north of Skunk River. William Hesseltine, a son-in- law of Mr. Hopping, has informed the writer that the spot where the bones were found is near the center of the quarter section and near the forks of the east branch of Walnut Creek. Gass and Pratt stated that the scapula and some other bones were found in the bed of the creek ; others were secured by digging into the bank; all the bones were found within an area of 15 feet each way, in the black mud (sedimentary deposit, chiefly of vegetable mold with some clay) and about 6 feet below the surface of the level ground. Mr. Hesseltine states that they were found in the creek-bank under 8 feet of loose earth and sand. The stream takes its rise about 2 miles away. The principal portions of the skeleton recovered were 2 upper molars, a considerable frag- ment of a tusk, a complete atlas and 3 other vertebrae, a scapula somewhat injured, a portion of a humerus, a fibula, and some fragments of ribs. From Mr. Hopping^s hands these remains passed into the possession of James M. Wier, of Muscatine, Iowa, who was gathering curiosities of various kinds. On his death his collection was placed in the Muscatine Public Li- brary. The writer had the opportunity to examine the collection and was able to identify most of the bones of the Hopping elephant. There were no records accompanying either these bones or others acquired by Mr. Wier. The writer gave a brief account of the Hopping elephant in his paper on the 50 ELEPHA8 BOREUS. Pleistocene Mammalia of [owa (Iowa Geol. Surv., vol. xxnr, p. 118). but he omitted to state that lie regarded the animal as belonging to Elephas primi- genius, now called E. boreus. Tt might he thought that the rather loose, sandy, clayey, and muck-like materials forming the banks of Walnut Creek consisted of recent alluvium and that the elephant remains had been redeposited from older beds. This is not probable, since a skeleton would on redeposition inevitably have been widely scattered. Those superficial deposits were probably laid down during one of the glacial stages, when the climate was sufficiently cold to attract this species into that region. The Illinoian stage has strong claims. The locality appears to lie very close to, or within the course taken by, one branch of Mississippi River when the Illinoian drift had forced it out of its old bed. The elephant might have been buried in deposits along this river or near the mouth of a stream open- ing into it. On the other hand, the possibility exists that the elephant lived there during the Iowan or Wisconsin stages. Union Township, Louisa County (3). — In the writer's paper on the Pleis- tocene Mammalia of Iowa (Iowa Geol. Surv., vol. xxm, p. 438) are described teeth of E. primigenius (E. boreus), found in a gulley in the northern part of Louisa County, about 6 miles north of Columbus Junction. The exact locality, as given the writer by E. B. Tucker, of Columbus Junction, is the northeast quarter of the southwest quarter of section 34, township 76 north, range 5 west. The Pleistocene geology of the region has been described by J. A. Udden (Iowa Geol. Surv., vol. xi, p. 102). According to his determina- tions there are exposed in that part of the county Nebraskan drift, Kansan drift, a leached Sangamon soil, and over this a covering of loess. With such exact data as to locality, a competent geologist on the spot might determine from what deposit those teeth were derived. Grinnell, Poweshiek County (4). — In Science (vol. iv, 1884, p. 46), Pro- fessor H. W. Parker reported the discovery of elephant remains in Grinnell. A tusk was found at a depth of 5 feet and 3 molars at a depth of 8 feet, in yellow clayey loam. The place was at the corner of Main and Fourth Streets. The writer examined the molars and found that two, certainly belonging to this individual, are those of Elephas boreus (Iowa Geol. Surv., vol. xxin, p. 445). Another tooth, apparently a second lower molar, labeled as belongino- to the same specimen, appeared to the writer to belong rather to E. columbi and to be probably a tooth described by Dr. E. H. Barbour. For an account of this, see page 293. From Parker's account it appears quite evident that this elephant had been buried in the loess. Dr. G. F. Kay, director of the Iowa Geo- logical Survey, informed the writer that he was at Grinnell in 1915, when much work was being done on the city sewer system and he obtained good sections to a depth of nearly 50 feet from the surface. He found that the loess has a thick- ness of about 18 feet. The upper 12 feet is yellowish and highly calcareous; the lower 6 feet, gray and also calcareous. The evidence appears, therefore, to place this animal in either the late Iowan or the early Peorian. Marengo, Iowa County (5). — In 1912 (Iowa Geol. Surv., vol. xxm, p. 436, pi. lv, fig. 2), the writer reported the finding, at Marengo, of a well- preserved lower first true molar of E. boreus (E. primigenius) . This tooth IOWA. 51 was met with in alluvial gravel along Bear Creek, in the northwest quarter of section 25, township 81 north, range 11 west. Near the same town, in al- luvial materials along Iowa River, was found the lower jaw, with the first true molar of each side, which was described by Calvin (Bull. Geol. kSoc. Amer., vol. xx, pi. xxv, fig. 3) as E. columbi. The writer regards the jaw as that of E. boreus. One can not be certain as to the origin of the tooth found in Bear Creek and of the jaw found along Iowa Eiver. It is most probable that they had been washed out of the loess, which occurs of considerable depth along those streams. They may belong to the late Iowan, the early Peorian, or even to a later stage. Neither the tooth nor the jaw could have been transported far, for they show no abrasion. West Union, Fayette County (6). — A part of an imperfect upper molar, probably the second, No. 42 of the collection at the Iowa State University, was found somewhere about West Union. It belongs to Elephas boreus (Iowa Geol. Surv., vol. xxm, p. 433). Possibly this tooth was found within the border of the Iowan drift, but the exact place is not recorded. Clermont, Fayette County (7). — On the page just quoted is an account of another molar, regarded as belonging to Elephas boreus and reported to be in the possession of C. E. Allen, of Clermont. Professor T. E. Savage in- formed the writer that it was found in materials which fill the valley of Tur- key River. Mr. Allen sends the information that it was found in the gravel- pit of the Rock Island Railroad, between Clermont and Elgin, at a depth of about 20 feet. The geology of this county has been described by Professor T. E. Savage (Iowa Geol. Surv., vol. xv, pp. 435-546, with figs, and 2 maps). His conclusions as to the Pleistocene are recorded on pages 439-454; 521-532. He expressed the opinion that the high terrace along Turkey River, where the Allen molar was found, is composed in part of gravels of Iowan age. In this connection it may be noted that a skull of Ovibos moschatus was found just east of Clermont, for an account of which see page 183. Dr. Alden and Dr. Leighton (Iowa Geol. Surv., vol. xxvi, pp. 133-134) discuss the gravel terraces along Turkey River and mention the fossils found there. They wrote that they regard the gravel as of Iowan rather than Wisconsin age. Missouri Valley, Harrison County (8). — In the Claude Cox gravel-pit at this place was obtained, by Professor B. Shimek, an upper right second molar which the writer refers to Elephas boreus (Iowa Geol. Surv., vol. xxm, p. 435). This important gravel-pit is described on page 296, where a list is given of the species which have been secured in the pit. These species belong to the Aftonian interglacial. Most of them indicate a climate probably not colder than that of the present day; but this mammoth tooth belonged to a species which, so far as known, preferred a cool or cold climate. Denison, Crawford County (9). — The geological situation at Denison has been briefly set forth on page 298. Calvin described from Denison (Bull. Geol. Soc. Amer., vol. xx, pi. xxiii; vol. xxn, p. 212, pi. xx) 2 teeth which are to be referred to Elephas boreus. These have been described and figured by the writer also (Iowa Geol. Surv., vol. xxm, p. 431, pi. lvi, fig. 2, pi. lxix, fig. 1). The first-mentioned tooth was obtained in the upper pit at Denison; the other tooth in a fragment of jaw was found in this pit about a 52 ELEPHAS BOREUS. mile below the town. It is impossible to determine with certainty the geolog- ical age of the species found at this locality. Correctionville, Woodbury County (10). — In the collection at the Iowa State University is a much-worn tooth of Elephas boreus, possibly the last milk-molar. It was found in the Gilleas pit and has the catalogue No. 355. From the Welch pit, in the same neighborhood, have been taken remains re- ferred to Bison occidentalis and Rangifer. The age of the sands and gravels of these pits is not well determined. They are without doubt post-Kansan. The presence of Elephas boreus and Rangifer appears to indicate a climate appertaining to a glacial stage. This might be the Wisconsin; and the pres- ence of Bison occidentalis is not opposed to this view. It is probable, how- ever, that the three species here named lived in that region during late Uli- noian or early Sangamon times. The writer has discussed this locality and the species occurring there in his paper on the Pleistocene Mammals of Iowa (Iowa Geol. Surv., vol. xxm, pp. 74, 282, 325, 449). The geology of this county was described by H. F. Bain (Iowa Geol. Surv., vol. v, pp. 243-299, maps) and by Carman (Iowa Geol. Surv., vol. xxvi, pp. 409-414). From Professor H. W. Norris, of Grinnell College, Grinnell, Iowa, the writer learns that that college has recently received from a sand-pit at Cor- rectionville, a tooth believed to belong to Elephas boreus. Dr. J. E. Carman (Iowa Geol. Surv., vol. xxvi, p. 410) reported that a worn tooth of this ele- phant had been found in the Gilleas gravel-pit, a few miles south of Correc- tionville, in the Little Sioux Valley. With it were parts of a bison skull and antlers of a deer. In discussing the age of these river gravels Carman (his page 412) concluded that they were post-Kansan and preloess and deposited long after the Kansan. This appears to confirm the view that they belong to the Sangamon or to the late Illinoian. From the Paul Fleming pit, at the mouth of Pierson Creek Valley, Carman examined a large proboscidean tooth, a horn core, and a horse tooth. The presence of the horse tooth appears to indicate a time not far away from the Sangamon. Le Mars, Plymouth County (11). — The writer has received from Dr. J. Ernest Carman some enamel plates of Elephas boreus, dredged from the deep pit just northwest of Le Mars (Iowa Geol. Surv., vol. xxvi, pp. 394, 395, 409). The thinness of the enamel indicates that the species is E. boreus. Rock Rapids, Lyon County (12). — In the collection at the Iowa State Uni- versity is an atlas of a proboscidean, which the writer has referred to Elephas boreus (Iowa Geol. Surv., vol. xxiii, p. 439). It was found in what was sup- posed to be a train of gravel which had been carried down Rock River from the Wisconsin ice sheet. The geological age may therefore be regarded as Wisconsin. This locality is mentioned by Carman (Iowa Geol. Surv., vol. xxvi, pp. 382, 410). While it is well out on the Kansan drift area, it may, he thought, contain Wisconsin gravel. 2. Within the Aeea of the Illinoian Drift. Burlington, Des Moines County (13). — In the collection of the Iowa State University is a large upper molar of Elephas boreus, No. 22, which is recorded as found at Burlington (Iowa Geol. Surv., vol. xxm, pp. 59, 432). There were, however, no details furnished, and it is impossible to determine the geological age of this tooth. IOWA. 53 Wapello, Louisa County (14). — In the collection just noted (No. 61) is a tooth of Elephas boreus, apparently the lower second molar, found at or near Wapello (Iowa Geol. Surv., vol. xxm, pp. 70, 438). Its geological age is unknown. Muscatine, Muscatine County (15). — At the Iowa State University, at Iowa City, the writer has seen an upper tooth of Elephas boreus, apparently a second molar, reported to have been found in a railroad cut a mile west of Muscatine. From Professor F. M. Van Tuyl, of the geological department of the University of Illinois, and collector of the tooth, it is learned that, according to his recollection, the tooth was found in the cut of the interurban line running from Muscatine to Iowa City. Buffalo, Scott County (16). — In the Davenport Academy of Science is a part of an upper molar of Elephas boreus, found on the farm of Mr. Sullivan, near Buffalo. The fragment presents 12 enamel plates of the hinder half of the tooth. Professor J. H. Paarmann, curator of the collection, has fur- nished the writer with photographs and measurements of the tooth. The plates are flat and rise at right angles with the base. The height of the tooth is 200 mm.; thickness, 92 mm.; length of fragment, 125 mm. Between 9 and 10 plates are in a 100 mm. line. The part preserved had not been sub- jected to use. The tooth is taken to be the second true molar. Davenport, Scott County (17). — In 1876, W. H. Pratt (Davenport Acad. Sci., vol. i, p. 96) gave an account of the discovery of some elephant remains in a cut along the Chicago, Eock Island, & Pacific Eailroad, just west of Davenport. A tusk, some molars, and some bones were found, and are now in the collection of the Davenport Academy, where the writer has examined them (Iowa Geol. Surv., vol. xxm, p. 446). They belong to Elephas boreus. The geological situation at this place is discussed on page 289. The conclu- sion reached is that the Davenport elephant probably lived there during the Iowan stage. 3. Within the Area of the Iowan Drift. Cedar Rapids, Linn County (18). — The writer has reported (Iowa Geol. Surv., vol. xxiii, p. 438) a tooth of Elephas boreus, found within the limits of Cedar Eapids and in the southwest quarter of the southwest quarter of sec- tion 27, township 83 north, range 7 west. It was brought up from the bottom of the river in a suction-pipe. Although the tooth has some plates missing in front, possibly only one or two, it does not seem to have suffered any im- portant abrasion from transportation. Cedar Eapids is within the area of the Iowan drift; but the tooth may have been geologically older or younger than this drift. In a collection of fossils in the City Hall, Portland, Oregon, the writer saw, in 1915, an upper second molar of Elephas boreus, labeled as having been found in Cedar Eiver, at Cedar Eapids. It had been loaned by H. H. Usher. Waterloo, Black Hawk County (19). — The writer has described (Iowa Geol. Surv., vol. xxm, p. 428) some large teeth of an elephant, found in 1897 in a sand-pit at Waterloo, at a depth of 7 feet from the surface. These are large teeth and have some of the characteristics of teeth of Elephas columbi; but the writer regards them as belonging to E. boreus. The region about Waterloo is covered by Kansan drift, overlain in most places by Iowan drift. Over 54 ELEPHA8 BOREUS. this again may occur loess sometimes reaching a depth of 8 feet (Aubrey, Iowa Geol. Surv., vol. xvi). Doubtless the geological age of the sands which contained this tooth will some time be determined. J. W. Wilby, the owner of the sand-pit, states that it is situated on the southeast quarter of the southeast quarter of section 24, township 89 north, range 13 west. It is about a mile east of Cedar Kiver and apparently at the eastern edge of its old valley. Close to the sand-pit is a stone-quarry. Hampton, Franklin County (20.) — From William Brandt, of Hampton, the writer received photographs of an upper second molar of Elephas boreus, found northwest of Hampton, near Beed's lake. The more exact locality is in section 19, township 92 north, range 20 west. The place is evidently on Spring Creek, one of several streams of that county which take their origin in the moraine bounding on the east the Des Moines lobe of the Wis- cosin drift. Williams, in his account of the Pleistocene geology of the county (Iowa Geol. Surv., vol. xvi, p. 469), says that this stream occupies a pre- Wisconsin depression and its valley is marked by the presence of a Wisconsin train of gravel. It seems probable, therefore, that the animal lived there when the Wisconsin glacier was not far away. The possibility exists that the tooth might have been derived from some part of the Iowan drift; it cer- tainly had not been transported any considerable distance. Marble Rock, Floyd County (21). — The writer has described and figured (Iowa Geol. Surv., vol. xxm, p. 434, pi. lix, fig. 2) an upper hindmost molar of Elephas boreus, found near Marble Rock. It was discovered in a pit in a large gravel deposit which is a valley train formed probably during the withdrawal of the Wisconsin ice sheet from the Des Moines lobe. The geologi- cal age of the tooth may be regarded as late Wisconsin. Mason City, Cerro Gordo County (22). — Through Professor A. 0. Thomas, of the Iowa State University, the writer has learned of a tooth of Elephas boreus, a hindmost, probably upper, molar found east of Mason City, in Gab- ler's gravel-pit, in section 11, township 96 north, range 20 west. This gravel was regarded as a part of the same gravel train that is found lower down on Shell Rock River at Marble Rock. Probably the Wisconsin ice sheet was not far distant when the possessor of this molar lived. The tooth was described by the writer in 1912 (Iowa Geol. Surv., vol. xxm, p. 85). 4. Within the Area of the Wisconsin Drift. Des Moines, Polk County (23). — In the writer's paper on the Pleistocene Mammals of Iowa (Iowa Geol. Surv., vol. xxiii, pp. 31, 443), descriptions are given of some teeth found at Des Moines and referred to Elephas boreus. The Pleistocene geology of this county has been described by Calvin (Iowa Geol. Surv., vol. vn, pp. 335-352, map). Inasmuch as Des Moines is situated on the border of the Wisconsin drift it is reasonable to suppose that the teeth were derived from animals which lived there while the front of the Des Moines lobe was not very distant. Clear Lake, Cerro Gordo County (24). — In the Iowa State Historical De- partment is a part of a lower jaw of an elephant (E. boreus), found in 1898 by H. I. Smith, in the vicinity of Clear Lake. In this jaw is inclosed a tooth which may be either the last or the next to the last molar. It is further de- scribed in the writer's paper on the Pleistocene Mammals of Iowa (Iowa Geol. IOWA — NEBRASKA. 55 Surv., vol. xxiii, p. 429). Clear Lake is situated on the eastern border of the Des Moines lobe of the Wisconsin drift sheet. It is probable that the elephant in question lived there shortly after the ice began to withdraw.. 5. Within the Driftless Area Center Grove, Dubuque County (25). — The writer has seen, in the pos- session of Dr. H. G. Knapp, of Dubuque, a small but characteristic fragment of a tooth of Elephas boreus, found in making a drift toward a lead crevice, on land owned by William Brunskill, about half a mile from Center Grove. In the collection of Eichard Herrmann, of Dubuque, the writer has seen an upper last molar found in 1896 by David Dawson, at the 11-mile post of the Illinois Central Railroad, west of Dubuque. This was probably not far from Center Grove. The tooth is that of Elephas boreus. Center Grove is in or near to, section 27, township 89 north, range 3 east, and near Catfish Creek. The border of the Kansan drift is, however, not far away; at a somewhat greater distance are fingers of the Iowan drift. The Pleistocene geology of this county has been described by Calvin and Bain (Iowa Geol. Surv., vol. x, pp. 463-478, map). Chamberlin and Salisbury (6th Ann. Rep. U. S. Geol. Surv., 1885, pp. 199-322) have discussed thoroughly the whole of the drift- less region. By consulting their plate xxvi it will be seen that the western part of this driftless region, especially that west of Mississippi River, is cov- ered by a mantle of loess. While, therefore, no certain conclusion can be reached about the geological age of the mammoth teeth here reported, it is probable that they were buried beneath or in the loess and would be, at earliest, post-Kansan in age. One might assign them provisionally to the Iowan stage. NEBRASKA. (Map 5.) York, York County (1). — In the collection in the State Museum at Lin- coln is a lower right hindmost milk molar, referred to Elephas boreus. There are present 10 plates, but apparently one or more are missing in front. The length along the worn face is 105 mm.; width, 52 mm. The enamel is thin. The specimen is designated in the collection by the numbers 3-10-11. The region in which this tooth was found is occupied by the Sheridan beds ; these are covered by loess. The tooth might have been found in either deposit. Dannebrog, Howard County (2). — In the collection of the State Museum at Lincoln is a large upper left last molar of an elephant, found at or near Dannebrog. It is one of a pair, the other of which is or was in the possession of Dr. P. Harold Salter, of Norfolk, Nebraska. The one in the museum was deposited by Dr. Franklin Schauffelberger, of Hastings. This tooth was de- scribed and figured by Dr. E. H. Barbour (Neb. Geol. Surv., vol. iv, p. 63, pi. i) as Elephas columbi. The present writer now believes that this tooth belongs to Elephas boreus. He wishes to assume the principal part of the responsibility for the assignment of it to E. columbi. The tooth is a large one and as a result the plates are thicker than usual in E. boreus. How- ever, the form of the tooth is that usually assumed by the last-named species; the plates are little warped and run parallel with one another. The writer also made a note to the effect that the enamel is thin, a feature which indicates E. boreus. Doctor Barbour records the fact that these teeth were dug out 56 ELEPHAS BOREUS. of the bank of Middle Loup River, in 1895, by H. L. Underwood. The farm belonged to Louis Gognemans, and is 4 miles northeast of Dannebrog. The animal may have lived then during the time of the Kansan drift sheet. Hayes Center, Hayes County (3). — In Field Natural History Museum, Chicago, is an upper last molar of Elephas boreus, recorded as No. 6767 and as found about halfway between Hayes Center and the town of Palisade. The latter town is about 12 miles south of southwest of Hayes. There are in the tooth 19 plates, besides front and rear talons. The extreme length is 247 mm.; height of the seventh plate, 145 mm. Nothing is known about the his- tory of this tooth. It appears to have been found in the region between Red Willow and Stinking Water Creeks and therefore probably on high land. It seems probable that it was buried in Sheridan deposits ; but we have no proof. Agate, Sioux County (4). — In 1914 (Nebraska Geol. Surv., vol. vn, pi. i), Harold Cook described and figured a lower tooth found near Agate, which he identified as probably E. columbi. It was supposed to be a milk-molar. An examination of the description and figure makes the present writer think that the tooth is that of Elephas boreus and that it is the second true molar. There is too high a number of plates (10) in a decimeter line for the tooth to belong to E. columbi. There are too many plates in the tooth (15) for it to be a milk-molar. The tooth had only just begun to wear ; the plates of the hinder end had not become united at their bases. It is probable that one or two are missing. Mr. Cook stated that the tooth had been washed out by a flood and that it had come from the gravels which underlie the surface deposits in the bottom of Niobrara Valley, at Agate. It may be that these gravels be- long to the Sheridan stage, but the writer does not know that this is true. MINNESOTA. (Map 5.) Minneapolis, Hennepin County (1). — In the collection at Yale University is a tooth of this species (Cat. No. 11679), labeled as found at Minneapolis. No details are given. The tooth is an upper left hindmost molar. There are present 20 enamel plates; a few in front had probably been worn off. The length of the fragment is 220 mm. Ten plates occupy a line 100 mm. long. This is probably the elephant tooth to which Professor Lull referred as being from Minnesota (Amer. Jour. Sci., ser. 4, vol. xxv, p. 198). Other supposed elephant remains from the region about Minneapolis were reported by N. H. Winchell in 1910 (Bull. Minn. Acad. Sci., vol. IV, pp. 416-417) ; but'there is doubt regarding the species, or even that they are not parts of the mastodon. Wabasha, Wabasha County (2). — In Professor Winchell's paper of 1910, on page 417, with a plate, he recorded the finding of a small elephant tooth in a bed of gravel forming a part of the terrace along Mississippi River. It belongs evidently to Elephas boreus. Winchell regarded it as having lived during the Wisconsin epoch. Mankato, Blue Earth County (3). — The writer saw, in January 1914, a part of a tooth of Elephas boreus, found on a sand-bar in Blue Earth River, near its junction with the Minnesota. In the high school of Mankato there is a part of another tooth of this species, found near the town in excavating for sand. FINDS OF ELEPHAS COLUMBI IN THE MIDDLE REGION OF NORTH AMERICA. LOUISIANA. (Map 7.) Petite Anse, Iberia Parish (1). — In the U. S. National Museum is a lower molar of an elephant, presented, in 1866, by J. F. Clew, of Petite Anse. At the same time he contributed some rock salt, parts of tusks of mastodon or elephant, and some matting, or part of a basket, which had, as reported, been found on the surface of the rock salt. The elephant tooth is a lower molar, probably the penultimate, very much worn, 9 plates are represented. The length of the grinding-face is 150 mm.; width, 83 mm.; 5 plates in a 100 mm. line. Each plate, as shown on the grinding-surface, is much constricted in one or two places and nearly all have an expansion near the middle. The enamel is very thick and much corrugated. The rear shows no indications of there having been another tooth behind it; yet it appears to be quite narrow for a hindmost molar. The tooth was mentioned by Leidy (Ext. Mamm. Dak., Neb., p. 254), who referred it to Elephas columbi. In the de- termination of the species it must be taken into account that the plates, espe- cially those of the rear of the tooth, may be thicker than the anterior plates; also, that the tooth curves somewhat, so that the length would be less than estimated above. Dr. Joseph F. Joor, in 1895 (Amer. Naturalist, xxix, p. 397), stated that he had been shown a tooth of an elephant which he was told had been found in a shaft of a salt mine at Petite Anse. This probably belonged to Elephas columbi. It may well be that some of the remains which have been found in northwestern Louisiana and reported as mastodon were really elephant. TEXAS. (Maps 7, 8.) Paloduro, Armstrong County (1). — In the American Museum of Natural History, New York, are some remains of one or more individuals of Elephas columbi, collected near Paloduro, in 1911, by Mr. Barnum Brown. Among these is a nearly complete lower right third molar. It lacks a few of the hindmost plates. There are 17 preserved ; 6 of these occupy a 100 mm. line. There are present a part of the corresponding left molar and a part of the jaw. The catalogue number is 15539. Rock Creek, Briscoe County (2). — In 1884 (4th Ann. Rep. Geol. Surv. Texas, p. 79), under the name Elephas primigenius, Cope reported the pres- ence of elephant remains at this place. He stated that the teeth from Rock Creek were neither thick-plated nor thin-plated, but were much as in the typical variety of Elephas primigenius (E. boreus), and that some plates were slightly crenate, while others were plain. Such teeth could not have be- longed to E. imperator; and there is no sufficient reason for believing that Elaphas boreus occurred there. The teeth must be referred to Elephas columbi. Foard City, Foard County (3).— In May 1921, J. W. Beverly, of Crowell, Foard County, sent to the U. S. National Museum the rear part, about 8 57 58 ELEPHAS COLUMBI. plates, of a large upper hindmost molar of Elephas columbi. This tooth was found near a small stream, Salt Creek, about a mile southwest of Foard City. Vernon, Wilbarger County (4). — From R. L. More, of Vernon, the U. S. National Museum received a lower left hindmost molar of Elephas columbi, catalogue No. 8873, which was found in a gravel-pit about 2 miles southeast of Vernon. It is considerably worn in front and some plates are missing. Sweetwater, Nolan County (5). — On November 13, 1913, there were re- ceived at the U. S. National Museum, from Thomas L. Hughes, parts of two large elephant teeth found near this place. The fragments had suffered little through wear, the circles of the worn digitations remaining on the surfaces. The ridge-plates are from 14 to 16 mm. thick and about 155 mm. high. There are two plates in each fragment and these belong to corresponding molars of the right and left sides of the jaw. In a letter dated December 3, 1913, Mr. Hughes informed the writer that the teeth had been found in the northwest quarter of section 59, block 22, Nolan County, on a small creek called Icem- good, on land owned by Mr. Grogan. The fragments of teeth were in a bluff at a depth of about 8 or 10 feet from the surface of the ground and 4 or 5 feet above the bed of the creek. They were buried in a stratum of strongly cemented gravel. Besides these fragments of teeth, the jaw-bone and a part of a tusk were dug out. Tulip, Fannin County (6). — About 1910, near the town of Tulip, not far from Red River, what appears to have been a nearly complete skeleton of an elephant was found. This at length came into the possession of R. E. Medford, of Bonham. At last accounts it was being exhibited in Texas towns. Doctor Francis, who saw the remains in Dallas, states that they are now in very bad condition. He thought the skeleton belonged to Elephas columbi. Rock Hill, Collin County (7). — In the collection of the Texas State Uni- versity, is a part, 10 ridge-plates, of an elephant upper left molar, found on Panther Creek, 2 miles south of Rock Hill. It is credited, to J. A. Taff. While referred to E. columbi provisionally, it may! belong to E. imperator, inasmuch as there are only 5 ridge-plates in a 100 mm. line. McKinney, Collin County (42). — From Dr. Francis the writer received in 1924 a photograph of a molar, perhaps the lower second, of Elephas columbi, which had been found in a gravel pit near McKinney. The tooth is close to 180 mm. long and 75 mm. wide. There are 7 of the plates in a 100 mm. line. There appear to be 14 of them in the tooth as preserved and these cross the grinding surface obliquely. This number is not on the maps. Fort Worth, Tarrant County (8). — At the Public Library in Fort Worth the writer saw a lower apparently third molar of Elephas columbi in a part of the jaw found somewhere about the city, probably in the gravel deposits along Trinity River. From Professor W. M. Winton, of Texas Christian University, Fort Worth, the writer received a photograph of a lower jaw of Elephas columbi, preserved in the collection of the institution named. This was found on the north side of Trinity River, directly north of the city. The teeth appear to be the hindmost molars and there are 7 of the ridge-plates in a 100 mm. line. The jaw was found in a gravel-pit at an elevation of about 45 feet above high water. At the same place was found a jaw of a mastodon. Dallas, Dallas County (9).— In the collection of the Southern Methodist University, at Dallas, are remains of various elephants, but unfortunately it TEXAS. 59 is not known, in all cases, where the remains were found; but all or most of them evidently came from gravel-pits east of the city, along or near White Eock Creek. One piece, which consists of the bases of both tusks and the bones which contained them and of both upper hindmost molars yet in the maxilla, was reported in the Dallas News of May 2, 1914, to have been found at a depth of 10 feet in the Lagow gravel-pit, just east of the Fair Grounds. The remains are those of a very old Elephas columbi. The hindmost teeth are worn so that only 12 or 13 ridge-plates remain. The tusks are large, the diameters near the base being 200 mm. and 190 mm. The distance across the sheaths, where this is least, is 400 mm. The tusks diverge at the base at an angle of about 45 degrees. For a list of the species found in this gravel-pit, see page 241; also Lull (Amer. Jour. Sci., vol. u, 1921, p. 160). In a small collection at the Central High School in Dallas is an upper right last molar of E. columbi, found in the gravel along Trinity River. The tooth is nearly white. It measures 190 mm. in length and 90 mm. in width. An elephant found at Dallas and referred (Amer. Naturalist, vol. xxm, p. 208) to E. columbi is regarded as belonging to E. imperator. Here may be recorded elephant remains found 5 miles south of Dallas, along the Missouri, Kansas, and Texas Railroad. Just west of the railroad is an extensive gravel-pit on the farm of Dr. A. E. and Dr. G. R. Flowers. A fine tusk and some teeth of Mammut americanum had previously been found in this pit. In January 1923 a very large tusk and two molar teeth, upper hindmost molars, were found in the pit. The remains were secured by Doctor Francis and he has sent to the writer photographs of the teeth. These are much worn. There are left only 11 or 12 plates and the plane of wear strikes them obliquely. Five plates are counted in a 100 mm. line. There would apparently be more plates in such a line at right angles to the plates. The writer believes that the teeth and tusk belong to Elephas columbi. Waxaliatchie, Ellis County (10). — In Field Museum of Natural History, Chicago (No. 5930), the writer saw a well-preserved left molar of Elephas columbi, found somewhere in this region. There are present 14 plates. Corsicana, Navarro County (11). — In the collection of the University of Texas is part of an upper molar, referred to E. columbi and labeled as having been found near Corsicana. No other information is given regarding the history of the specimen. Trinity, Trinity County (38). — In the collection of Dr. Mark Francis is a part of a lower molar, probably the hindmost one, consisting of 14 plates, which was found near Trinity. A few plates are missing from the front end and a few from the rear end. The hindmost of the plates present is 160 mm. high and 97 mm. wide. The plates are curved in their ascent. The hinder plates are thick, 5 in a 100 mm. line. The enamel is thinner than one expects to find in E. imperator. It possibly belongs to the species last mentioned. The tooth was sent to Doctor Francis by Lewis Runnels, of Trinity. This town is near the western end of the county and near Trinity River. Another lower molar seen by Doctor Francis in the Trinity National Bank had been found in Trinity River, near Trinity. It had about 6 ridge- plates in 100 mm. Onalaska, Polk County (39).— From F. M. Burks, of Onalaska, the U. S. National Museum received in 1923 an upper left hindmost molar of Elephas 60 ELEPHAS COLUMB1. columbi. He sent likewise a photograph of a lot of bones, consisting of a jaw- bone with good teeth, portions of other teeth, fragments of apparently the pelvis, and parts of various limb-bones. It is evident that the jaw-bone con- taining a large hindmost molar belongs to Elephas imperator; it will be recorded in its proper place. All of these teeth and bones were found in a gravel-pit in a terrace 1.5 miles from Trinity Eiver rising from the flood plain and at an elevation of about 35 feet above the river. The tooth re- ceived presents only about 13 plates; some gone from in front, a few from the rear. There are 6 plates in a 100 mm. line. Whether the bones shown in the photograph belong to E. columbi or E. imperator, it is impossible to say. Paluxy Creek, Hood County (12). — In the American Museum of Natural History, New York, is a remarkable tooth, referred to Elephas columbi. The catalogue number is 8071. The tooth was collected by W. S. Nehms. It ap- pears to be the upper hindmost molar of the right side. There are 14 plates present; several of the front ones, probably about 10, are worn out. The hinder plates are 230 mm. high. The notable feature of the teeth is the extreme thinness of the cement between the enamel plates. The enamel is thick. The plates converge from the base toward the grinding surface. At the middle of the height there are 7 plates in 100 mm. Were the cement of the usual thickness there would be fewer plates in this distance. It is pos- sible that the tooth is one of E. imperator. Clifton, Bosque County (13). — In 1913, Jacob Olson, of Norse, in Bosque County, sent to the U. S. National Museum a fragment of an upper molar of Elephas columbi, found, as the present writer understands, at Clifton, on Bosque River. The tooth was finely preserved and white in color. It was returned to Mr. Olson. Waco, McLennan County (14). — In Baylor University, Waco, Texas, are two teeth, the last milk-molars of both sides of the upper jaw, which were found in the White Eock sand-and-gravel pit, 5 miles above Waco, on the left bank and in the third terrace. They are worn on only 4 plates. Of these plates, there are 9 in each tooth. From the base in front to the rear talon the length is 117 mm.; width, 58 mm. There are a little more than 4 plates in a 100 mm. line. These teeth are referred to Elephas columbi. From the same pits was obtained a part of the lower jaw of a young E. columbi, con- taining a finely preserved lower hindmost milk-molar. This tooth presents 11 ridge-plates. It is worn on all the plates except the hindmost, but not to the base in front. The length is 143 mm.; width, 63 mm. This tooth was pre- sented to the National Museum by Richard M. Cooper, Waco. It has the catalogue number 8207. In the collection of Baylor University are two finely preserved upper hind- most molars of Elephas columbi, found in a gravel-pit at Third Street in South Waco. This is near the river and in the second terrace. These teeth are worn to the base in front and the front roots are missing. Probably 2 or 3 ridge-plates are missing in front. Dr. Mark Francis reports that he re- ceived, in 1923, a tooth of E. columbi, found along Brazos River, about 6 miles northwest of Waco. TEXAS. 61 Temple, Bell County (40). — A lower hindmost molar found between Tem- ple and Belton is of such a nature that it is doubtful whether it belongs to Elephas columbi or E. imperator. A description of it is given on page 104 as Elephas sp. indet. Marlin, Falls County (15). — In the collection of Dr. Mark Francis at College Station is part of a right molar tooth of Elephas columbi from some- where about Marlin; it consists of 13 plates, the highest 190 mm. On the convex side of the tooth are only 5 plates in a 100 mm. line; on the concave side are about 6 or 7. From Doctor Francis the writer received a sketch of a tooth of Elephas columbi, obtained at Marlin, in 1922, in the gravel-pits of I. H. Parten. The tooth had lost some plates at the rear. About 12 plates were worn, but not to the base in front. The extreme length of the specimen was 12 inches. Eight plates occupy a 100 mm. line. With this tooth were what Doctor Francis identified as a radius of a horse and a tooth of Elephas imperator. J. M. Kennedy, editor of The Marlin Democrat, reports the finding of a tooth of an elephant in the same gravel-pit in which was found the type skull of Elephas francisi. On one of the faces he counted 8 ridges in a line 4 inches long. This indicates that the tooth belonged to Elephas columbi. Satin, Falls County (16). — The writer has seen photographs of some teeth of an elephant, found in the bed of Brazos Eiver by W. P. Moore, of Satin. There are 4 teeth and a vertebra. In 4 inches, along the grinding-surface, there are 6 plates. The animal was certainly Elephas columbi. Pittbridge and Munson's Shoals, Brazos County (17). — From Doctor Fran- cis the writer received a photogragh of a much-worn molar of Elephas columbi, obtained in a store in Bryan, but which was found at Munson's Shoals, 2 or 3 miles below Pittbridge. The tooth is worn to the base in fronii and no doubt some plates are missing. The length as preserved is 142 mm.; the width is about 70 mm. The tooth is probably the lower first molar. The plates cross the grinding-face in a very irregular way. There are somewhat more than 6 plates in a 100 mm. line. In the same collection is a molar found 3 miles below Pittbridge. Near the same place was found, by E. J. Foun- tain, of Bryan, a lower second molar of E. columbi. It belongs in the Francis collection. Here may be mentioned a fine upper left hindmost molar referred to E. columbi, which was found at Pittbridge in 1912 and is owned by Doctor Francis. The front end, with its root, is worn off and some plates with it; 19 or 20 plates remain. Of these, there are 5 in a 100 mm. line. The length from the base in front to the rear is 370 mm.; the width 105 mm. The enamel is thick and much folded. Were it not that the number of plates ex- ceed 18, the tooth might be referred to E. imperator. Doctor Francis has an upper second molar of E. columbi, which he reports was found at Pittbridge. There are 16 plates and a hinder talon. This tooth is worn on about 10 plates, but not to the base in front. It shows the base of the front root with a space behind it. The length from the front of the tooth to the rear of the hinder talon is 215 mm.; height of the tenth plate, 205 mm.; width of the grinding-surface, 80 mm. In the rear is a depression for the front of the hindmost molar. In 1924 Doctor Francis received an 62 ELEPHAS COLUMBI. elephant molar, recently found at Pittbridge. It had 9 plates in a LOO rnm. line, and is referred to Elephas columbi. Navasota, Grimes County (18). — In the eolleetion of Doctor Francis is an upper right third molar, found in Brazos River, near Xavasota. This would be practically the same as Hidalgo Falls. The front of the tooth is missing. The width of the grinding-surface is 110 mm. There are 6 plates in a 100 mm. line. On the inner side of the tooth, opposite the edges of some of the median plates, is a row of apparently distinct columns of dentine surrounded by enamel, and some intercalated smaller columns. These appear to form a sort of cingulum. The tooth is referred to Elephas columbi. With this tooth were found 2 large cervical vertebrae, one of them the axis, but there is no assurance that they belonged to the same animal. The axis has a total height from bottom of centrum to top of spine of 290 mm. The width of anterior articular surface is 240 mm. These fossils probably belong to E. columbi, but there is some doubt. Big Spring, Washington County (19). — In the National Museum is a part of a tooth of Elephas columbi, collected on Big Spring Creek, 6 miles south- east of Burton, in 1910, by August Stelter. It was sent to the museum by Dr. Alexander Deussen. With it was found a part of a femur. Xothing else is known about conditions connected with the discovery. San Felipe, Austin County (20). — On page 246 is described the geology of the region of San Felipe, and a list is given of the species of vertebrates found there. In 1846 (Amer. Jour. Sci., ser. 2, vol. I, pp. 244-247), Carpen- ter described a collection made here by a Mr. Huff (or Hough). This went to the British Museum in London. Among the materials of this collection, Falconer found teeth of Elephas columbi. Two of these are described in de- tail, a part of a lower last molar and a part of a penultimate lower molar. On page 228 the same author* describes a fragment of a cranium which formed part of the Huff collection. In this fragment was an enormous last true molar, of which there remained 20 plates, some anterior ones having been worn away. The distance in a straight line from the base of the hinder talon to the front of the socket for the anterior root was 18.25 inches. The four- teenth plate from the front was 10.7 inches high. The plates were vertical and averaged only half an inch in thickness. Such a tooth reminds one of the great upper molar figured by the author (Iowa Oeol. Surv., vol. xxin, pi. lxiv, fig. 2) and found near Mount Angel, Oregon. Falconer referred the tooth to E. primigenius, but it is possible that it and the Oregon tooth be- longed to a yet unnamed species. According to Deussen's latest map (Prof. Pap. 126, U. S. Geol. Surv.), this locality lies somewhat outside of the Lissie area. In the collection of Doctor Francis is a fine lower right second molar, found at San Felipe in 1908. It is attached to a fragment of the lower jaw; 6 of the front plates abraded; no roots yet developed. Some plates, appar- ently one or two, are missing from the rear ; 14 plates and the front talon are present. Most of these are concave on the hinder face. There are 6 plates in a 100 mm. line. The length of the tooth as preserved is 242 mm. ; thickness, 70 mm.; height perpendicular to the base, about 150 mm. * Falconer, Palfleontological Memoirs, volume it. TEXAS. 63 McDowell, Austin County (21). — This town is on the west side of Brazos Biver about 10 miles west of Brookshire. A number of fossils are credited to the place, but some may have been found nearer Brookshire. In the col- lection of Doctor Francis is an upper left molar, probably the second of EUphas columbi, which presents 15 plates, some being gone from the front. There are 6 plates in a 100 mm. line. The molar was found in Brazos River. Another tooth in this collection, found at Brookshire, is a lower right second molar presenting 14 plates, some from the rear being lost. There are 6 plates in a 100 mm. line. From the front of the tooth to the base of the last plate present, the distance is 240 mm. There is doubt whether some other fine teeth in Francis's collection were found at Brookshire or Pittbridge. San Leon, Galveston County (22). — On the shore of Galveston Bay, at San Leon, 9 miles east of Dickinson, was found, in April 1921, a skull of an elephant, described on page 242 as Elephas imperator. In examining this collection the writer recognized one or two teeth of Elephas columbi and a few bones of a species of horse. The collection has become a part of that made by Dr. Mark Francis, at College Station, Texas. Houston, Harris County (41). — From Doctor Francis the writer received in 1924 a photograph of a fragment of an upper hindmost molar which was found at a depth of 18 feet in the excavation for the new Masonic Building in Houston. There are about 8 plates preserved. The height of these is about 150 mm.; 5 or 6 of these are crossed by a line 60 mm. long. The tooth ap- pears to belong to Elephas columbi. Alvin, Brazoria County (23). — In the collection of Dr. Mark Francis are 5 fragments of teeth of an Elephas columbi, sent to him from Alvin by J. A. Froberg, of the same town. Mr. Froberg informed the writer that these teeth were picked up by W. R. Bunch, of Alvin, from materials thrown out of a freshly dug ditch, about 3.5 miles southeast from the town. They had been buried at a depth of from 4 to 5 feet. The earth in which they were inclosed consisted of a whitish sand. There appeared to be some fragments of large bones in the same place, but these were not exhumed. The ditch is one which penetrated Mustang Bayou in several places. The locality is well down toward the southern border of the Beaumont clays. All the fragments belong to 2 lower teeth, which seem to be the right and the left second true molars. The most complete tooth is that of the left side. Perhaps one plate and the talon only are missing in front. Then come 9 plates in two frag- ments which form accurate contact. A rear fragment presents 4 plates and a part of another. Between this fragment and the one in front, one or two plates are missing. We seem thus to account for at least 16 plates. This number appears to indicate the second true molar, but the width of the tooth is only 78 mm., and the height of the supposed ninth or tenth plate from the front is only 135 mm. The tooth probably belonged to a small individual. The plates are thick, there being 6 in a 100 mm. line. They are much bent as they ascend from the base. Wharton, Wharton County (24). — The collection of Doctor Francis from Wharton, contains a very complete lower jaw of an elephant, which is referred to E. columbi. It was found in Colorado River. Ten plates of each hind- most molar are present, these teeth having been worn considerably. There 64 ELEPHAS COLUMBI. are 6 plates in a 100 mm. line. The enamel is very thick and is crimped From front to rear the part of the tooth remaining measures 220 mm. From outside to outside of the condyles the distance is 496 mm. From the middle of the line joining the condyles to the top of the chin is 612 mm. The width of the lingual groove at the middle of its length is 68 mm.; height of jaw at middle of tooth, 160 mm.; width, 140 mm. From the front of the ascend- ing ramus, at the level of the teeth, to the rear of the jaw is 250 mm. Keeran Point, Victoria County (37). — In September 1919 a violent hurri- cane led to the falling of a large mass of earth along the shore of Lavaca Bay, at Keeran Point. In the fallen mass were discovered bones of Elephas columbi, a mastodon later described by the writer as Anancus brazosius, and a bone of a camel. These specimens were acquired by Dr. Mark Francis. Among the elephant bones was a moderately curved tusk, of which about 6 feet were preserved, in as many pieces; at the base it had a diameter of about 215 mm. Two upper hindmost teeth are present, that of the left side com- plete, except that the roots are mostly broken. The anterior root is partly retained in the bone of the jaw. It appears to have supported 4 plates. Be- tween it and the small tooth belonging to hinder plates there appears only a short interval. In this tooth are counted 24 plates. Only 8 or 9 plates are worn; the eighth has a height of 260 mm. There are 6 plates in a 100 mm. line on the inner face. Of the left lower tooth are present 11 front plates, of which 10 are affected by wear. The great front root is represented by its base and it supported 3 plates. A notch separates it from the hinder rootlets, which were not yet fully developed. It is remarkable that the plates are thick, there being hardly more than 4 in a 100 mm. line. Were it not for the upper teeth, the animal would be referred to Elephas imperator. These plates are much bent forward. The tooth is 100 mm. thick. The lower right third molar is almost wholly buried in the bone. At the front of this tooth the jaw is 243 mm. high and 174 mm. thick. Its greatest thickness, far- ther back, is 205 mm. The tooth itself is close to 410 mm. long. A frag- ment of the right maxilla contains a considerable part of the second molar, 12 plates. It is 135 mm. long. Nearly 9 plates are spanned by a 100 mm. line. This fragment alone would suggest Elephas boreus. Bastrop, Bastrop County (25). — In July 1921 there was sent to the U. S. National Museum, by Hugh H. Duval, a tooth of Elephas columbi. It was found at Bastrop, in a gravel-pit, 50 feet above Colorado Eiver. The tooth is white and beautifully preserved. It is an upper left second molar with 11 plates, but some, probably 2, have been broken from the front end; wear had about reached the base in front and the last plate behind. The plates are flat and the enamel thick. The length is about 225 mm.; width of grinding- surface, 80 mm. There are 7 plates in a 100 mm. line. Marble Falls, Burnet County (26). — In March 1922, Doctor Francis in- formed the writer that he had visited Marble Falls, where the skeleton of an elephant was being washed out of a gully. He identified it as E. columbi. The lower jaws were in fair condition, but the other bones were badly broken and were not exhumed. Cherry Spring, Gillespie County (27). — In the collection of the State Uni- versity, at Austin, is a part of a lower molar, consisting of 6 or 7 plates, TEXAS. 65 which is labeled as having been found at Cherry Spring by G. Jermy. Noth- ing more is known about the specimen. Cherry Spring is situated near the head of Hickory Creek, a tributary of Llano River, which empties into Colo- rado River. It appears to be on the highest land in that immediate region. Luling, Caldwell County (28). — In the Francis collection at College Sta- tion are 2 molars, a part of a right innominatum, and 2 pieces of tusks, each about 3 feet long, which were found at or near Luling by R. E. Faires. These parts appear to belong to Elephas columbi. A part of the maxilla is at- tached to the left tooth. The teeth are the upper hindmost molars and are well worn down in front; 16 plates are present; about 8 must have been worn out. The hindmost plates had begun to wear. The present length of the crown is 200 mm. The width is 110 mm., excluding the cement. A 100 mm. line extends across 6 plates. The enamel is thick and somewhat crimped. The acetabulum has a diameter of 215 mm. Ottine, Gonzales County (29). — In the University of Texas, at Austin, is an upper left last molar of Elephas columbi, found in a spring near Ottine by Professor F. L. Whitney. Ottine is situated on Marcos River. There are present 20 plates and from 6 to 7 of these in a 100 mm. line. The height of the unworn sixth plate from the rear is 230 mm. Marion, Guadalupe County (30). — In 1922 the writer saw 2 fragments, of 2 plates each, of a molar of Elephas columbi, which Dr. L. W. Stephenson and E. B. Stiles found near Marion. The fragments were discovered in al- luvial materials along one of the headwater arroyos of Santa Clara Creek, about 1.5 miles west by south of Marion and south of the Galveston and San Antonio Railroad. They were collected August 28, 1921. There is nothing to indicate the stage of the Pleistocene to which they belonged. San Antonio, Bexar County (31). — In the U. S. National Museum (cata- logue Nos. 4840 and 4841) are 3 fragments of large molars, which were sent there by Frank Grice and which were found at San Antonio. The San An- tonio Daily Express of March 25, 1900, gave an account of the discovery of these fragments. They were found in the great gravel-pits of the San An- tonio Street Railway Company, near Riverside Park, about 15 feet from the surface. A fragment of a lower tooth (No. 4840) presents 7 ridge-plates which belonged near the front end of the tooth. There are 6 plates in a 100 mm. line and these are strongly flexed. From this newspaper account it is learned that in 1853 C. F. Beitel found in a well, at a depth of 16 feet, some large bones, among them a jaw in which were several teeth. These appear to have been preserved ; and on comparison with the teeth found near Riverside Park they were regarded as being the same. The U. S. National Museum (No. 8867) has received from Dr. Charles L. Baker a lower right molar, probably a penultimate, which is referred to this species. It was found, according to Doctor Baker's label, in 1917, in Pleistocene gravels and silts, on the east bank of Salado Creek, south of the crossing of the Suther- land Springs road, probably about 5 miles east of San Antonio. Counting from the rear, there are 9 ridge-plates remaining, about 6 of which occupy a line 100 mm. long. The large hinder root, 120 mm. long, is present, behind which there is a depression for a succeeding tooth. Probably as many as 7 ridge-plates were worn away. Were not the plates so thick, the tooth might 66 ELEPHAS COLUMBI. be taken as that of Elephas boreus. Sellards (Univ. Texas Bull. 1932, p. 73) refers to this specimen as occurring on the lower plain along Sal ado Creek. Leon, Bexar County (32). — The American Museum of Natural History, in New York, has some remains of an elephant, mostly bones of the skull, found somewhere along Leon River, southwest of San Antonio. Five ridge-plates of a lower molar indicate that the animal belonged probably to Elephas columbi. Beeville, Bee County (33). — In the collection of Dr. Mark Francis is a part of an upper molar of Elephas columbi, found somewhere about Beeville. It was evidently a very large tooth. There are 7 plates in a 100 mm. line. There is also present a part of a lower molar of the same species. Under the same number may be mentioned an elephant tooth received by Doctor Francis from V. E. Kessler, of Skidmore, and regarded by him as belonging to Elephas columbi. From Mr. Kessler the writer learns that this tooth was found 3 miles northwest from Skidmore, in the bank of a creek, under about 5 feet of ground. Alfred, Jim Wells County (34). — Dr. Mark Francis received from Alfred 3 ridge-plates of a molar, probably the upper hindmost, of Elephas columbi. The height is 195 mm.; width, 81 mm. The three occupy a thickness of 42 mm. No details are known about the geological relations of the specimen. Two short pieces of tusk were with the piece of tooth; all were found on the Pease ranch and sent by A. A. Wright. San Diego, Duval County (35). — In 1889 (Amer. Naturalist, vol. xxm, p. 209), Cope reported a fragment of a tusk of an elephant from southwestern Texas (meaning doubtless the San Diego region). This had a diameter of 210 mm. at the base. There is no evidence that this was not a tusk of E. imperator. The American Museum of Natural History, New York, has part of a molar and a leg-bone, which is referred by the writer to E. columbi. It was collected by William Taylor, of San Diego. Dumble (Trans. Amer. Inst. Min. Engineers, vol. xxxiii, p. 984) stated that most of the fossils sent to Cope by Taylor were found on Tarancahaus Creek. El Paso, El Paso County (36). — In 1907, George B. Richardson (Science, vol. xxv, p. 32) reported the finding of Elephas columbi in a gravel-pit at El Paso. With it were Equus complicatus and a tapir. The identifications were made by James W. Gidley. In Richardson's materials the writer saw a fragment of a tooth of a species of a mastodon other than Mammut, probably Anancus. For (37) see page 64; for (38) and (39) page 59; for (40) page 61; for (41) page 59; for (42) page 58. OKLAHOMA. (Map 7.) Afton, Ottawa County (1). — In the department of palaeontology in the U. S. National Museum are about a dozen teeth of E. columbi, which were ob- tained in a spring near Afton. These, with remains of 18 other species of mammals and many flint implements, were secured in 1901 by Professor W. H. Holmes. An account of the spring and its contents will be found on page 254. The teeth of this species include upper and lower ones of both milk and permanent series. Some of these are finely preserved. They are usually OKLAHOMA — NEW MEXICO. 67 stained black and show the presence of but little animal matter. Some of them are so well mineralized that they ring on being struck. One of these teeth was figured by Holmes (1903, pi. ix, lower figure) and by Lucas (Mary- land Geol. Surv., Pleistocene vol., pi. xxxviii). Others have been illustrated by the present writer (Indiana Geol. Surv., vol. xxvi; Iowa Geol. Surv., vol. xxiii ). Judged by the condition of fossilization of these teeth, they are to be referred to the Sheridan or Aftonian stage. Avery, Lincoln County (2). — On December 9, 1912, IT. S. National Mu- seum received from E. S. Eice, connected with the Atchison and Santa Fe Railway, an upper molar found near Avery; it has 11 ridge-plates; some are missing from the front end. The present writer has received a letter from F. M. Rice, who states that the elephant tooth was dug up in the northwest quarter of section 11, township 16 north, range 5 east, on the right-of-way of the railway mentioned above, at a depth of 8 or 9 feet; it was buried in what is described as soft, sandy clay. The ditch being made was 8 to 10 feet wide, and bones were seen and left undisturbed on both sides of it. A femur dug up is described as being 5 feet or more in length. Hennessey, Kingfisher County (3). — In 1894 (Proc. Acad. Nat. Sci. Phila., p. 68), Cope reported that he had visited Hennessey and had found some ele- phant remains in the possession of a rancher by the name of Painter, who lived 8 miles west of the town. There were 4 molar teeth, and these Cope referred to an adult Elephas primigenius, of the coarse-plated variety; they evidently belonged to E. columbi, unless, indeed, they were teeth of E. im- perator. Cope did not say that he secured these teeth; but he probably did and they may now be at the Academy in Philadelphia. With these mammoth- teeth, Mr. Painter found some teeth and bones of a saber-tooth tiger; these were later described by Cope under the name of Dinobastis serus. These bones and teeth, as well as those of the elephant, were stained red by the iron of the Permian clay and were covered by a thin layer of it. This deposit of clay, derived from the Permian, had probably been laid down during the earlier part of the Pleistocene. In the Journal of the Philadelphia Academy (vol. ix, p. 453), Cope refers to these elephant teeth and says that they were found 50 miles west of Hennessey. This statement of distance is evidently an error. On April 21, 1913, the author received a letter from R. A. Hamil- ton, Hennessey R. F. D. No. 2, and there was inclosed a photograph of a tooth of Elephas columbi, which had been found somewhere in that region. Norman, Cleveland County (4). — In the Oklahoma State University is a lower right last molar of Elephas columbi, found in loose sand in Canadian River, just south of Norman; there are 9 plates, occupying a length of 6 inches; the plates are much curved. NEW MEXICO. (Map 7.) Sacramento Mountains, Otero County (1). — In 1919, 0. L. Tinklepaugh, of El Paso, Texas, sent to the U. S. National Museum a large lower left hind- most molar of an elephant, which had been found somewhere in the Sacra- mento Mountains. On the label "The Rincon" is mentioned, and this may be a creek. The tooth is nearly complete, the anterior roots and plates being 68 I I > K.I 'HAS COLUMBI. represented. There are 20 ridge-plates present and apparently 2 or 3 are missing from the rear. The length from the base in front to rear is 412 mm.; width, 102 mm. The plates are strongly bent, so that the hinder faces are very concave. There are 5 plates on a 100 mm. line. It is possible that this tooth belongs to E. imperator, but the writer believes that it is an unus- ually large molar of E. columbi. Capilan, Lincoln County (2). — In the U. S. National Museum are 2 teeth of Elephas columbi, which came from the vicinity of Fort Stanton. One of these has the catalogue No. 2370 and was sent in 1903 by Dr. P. M. Car- rington, of Fort Stanton. It was found about 5 miles north of Capitan by a Mexican in digging a well. It appears to be a first true molar. About 10 feet of a tusk was found, as well as some bones, but these were kept by the finder. Apparently the locality was in township 8 south, range 13 east. In the Museum is also another tooth of the same species, sent apparently in 1871, by Governor Arny, from the vicinity of Fort Stanton; but he had re- ceived it from W. F. Holmes. No details were recorded. Albuquerque, Bernalillo County (3). — Through Professor E. W. Ellis, of the State University of New Mexico, the writer received notes and photo- graphs of a tooth of E. columbi, which belongs to the university collection. The tooth was found about 30 feet below the surface in a gravel terrace, in the Rio Grande Valley, near Albuquerque. The top of the terrace is about 75 feet above the flood plain. The tooth is the upper left hindmost molar and had been worn to the base in front. About 14 plates are represented. There are 7 plates in a 100 mm. line, and these are little curved. The enamel is thick and considerably crimped. Pena Blanca, Sandoval County (8). — In 1903 (Amer. Geologist, vol. xxxi, p. 89), Reagan reported that teeth and bones of Elephas primigenius (doubt- less E. columbi) and other Pleistocene animals had been found in Pleistocene marls at Pena Blanca. This locality is on the Rio Grande, about 40 miles above Albuquerque. It is in township 16 north, range 6 east. Placita, Sandoval County (4).— In 1877 (Geo.gr. Surv. West of 100th Merid., vol. iv, p. 25, pi. lxx), Cope mentioned and figured a molar of Elephas columbi, which he found along Placita Creek, or possibly one of its tributaries. The village of Placita is about 20 miles east of north of Albuquerque. The hills consist of Cretaceous materials; a deposit of indurated clay about 40 feet thick was found in the valleys. In such a bed, Cope found the tooth re- ferred to. It is now in the U. S. National Museum and has the catalogue No. 4180. Other elephant bones were found in place in the banks of the arroyo. Shells of Planorbis and Physa were observed. Cope named these deposits the Placita marls. This tooth is evidently the one mentioned by Cope in 1874 (Proc. Phila. Acad., p. 221) as having been found at the base of the Sandia Mountains. It is possible that the tooth was found somewhat farther south, in Bernalillo County. Las Vegas, San Miguel County (5). — In the American Museum of Natural History, New York, there is a fragment of an upper hindmost molar, which the writer refers to E. columbi. The catalogue No. is 15544 ; the tooth was presented by H. P. Frambly in 1911. It consists of the front 10 plates and retains the base of the large anterior double root. There are 7 plates in a 100 mm. line. MISSOURI — KANSAS. (59 Santa Fe, Santa Fe County (6).— In 1873 (Contrib. Ext. Vert. Fauna Terrs., p. 238), Leidy reported a tooth of E. columbi from near Santa Fe. This tooth is in the IT. S. National Museum and has the catalogue No. 286. Black Rocks, McKinley County (7). — In making excavations at Black Bocks, about 5 miles east of Zuni, in 1908, some fossil bones were found and sent to the U. S. National Museum. Those to whom credit is due for this are F. E. Leupp, Commissioner of Indian Affairs, John B. Harper, Superintendent of Irrigation, and Frank Mead, Supervisor. No details, however, have been secured as to the depth and kind of deposit in which the remains were found. The species represented are Elephas columbi, Equus sp. indet., camel, sp. indet., and the musk-ox, Gidleya zuniensis. Of Elephas columbi, there are in the U. S. National Museum an upper left hindmost molar and a lower right hindmost molar. MISSOURI. (Map 7.) Near Joplin, Jasper County (1). — The writer received from Byron A. Ash, of Carthage, Missouri, a photograph of a lower molar, probably the second or third, which was found about 9 miles southwest of Carthage. For an ac- count of the situation, reference may be made to page 24. The tooth is well worn down, with probably some plates gone in front; 11 are yet present. The tooth is Elephas columbi. In the same situation, probably an old cave, were found a tooth of Mammut americanum, a bone of a species of bison, and some remains of a turtle otherwise unidentified. The age of the remains cannot be determined from the species present. Tackner, Benton County (2). — In the collection of the Academy of Natural Sciences of Philadelphia are various remains of Elephas columbi, labeled as having been found in Benton County. These are probably a part of a collec- tion made about 1843 by S. H. Whipple. This collection is discussed on page 28. Among the specimens at Philadelphia is a part of a lower jaw of a young elephant, with the third and fourth milk-molars. Rockport, Atchison County (3). — In 1911 (Bull. Geol. Soc. Amer., vol. xxn, pp. 211, 212), Professor Calvin stated that a molar of Elephas columbi had been found near Rockport, in section 22, township 64 north, range 41 west. In the same pit were a foot-bone of a horse (Equus), a tooth of a species of Hipparion, and a premolar of a species of large camel, probably Camelops. Calvin regarded the deposits as belonging to the Aftonian stage. Jackson County (4). — In Popular Science (vol. xxxiv, on pages 198, 199), Dr. T. R. Thornton reported that a tooth of a mammoth had been found about 30 miles southeast of Kansas City. A number of bones were found along a gully in an abandoned public road, in yellowish clay. The bones on drying went to pieces. Three teeth were preserved, one of which Doctor Thornton owned. Inasmuch as there appear to have been about 5 plates in a 100 mm. line, the teeth were probably those of Elephas columbi. KANSAS. (Map 7.) Palermo, Doniphan County (1). — In 1912, Albert Kienhoff, living near Wathena, found on his farm a mastodon tooth, some teeth of Equus niobra- 70 ELEPHAS COLUMBI. rensis, and a part of a tooth of Elephas columbi. The exact locality and con- clusions about the age of the remains are given on page 268. Hammond, Bourbon County (2). — At Kansas City, Missouri, the writer saw, in the collection of Rev. John Bennett, the front half, presenting 7 ridge- plates, of a molar of Elephas columbi, which had been found near the Kansas- Missouri line, somewhere between Fort Scott and Fulton, probably in the vicinity of Hammond; but no more exact record had been preserved. Ottawa, Franklin County (9). — In 1904, as the writer learns from Pro- fessor A. E. Bell, some elephant remains were found in the bank of Marais des Cygnes River, about 4 miles west of Ottawa. Some of these remains passed from the hands of J. U. Gill, of Ottawa, into those of C. H. Sternberg, and then into the possession of George Langford, of Joilet, Illinois. From the latter the writer has received measurements and photographs of a tooth in the right ramus of the lower jaw, and another, lacking some hinder plates, free from all bone. Both are hindmost molars and well preserved. They are undoubtedly teeth of Elephas columbi. Rantoul, Franklin County (10). — In 1890 (Trans. Kan. Acad. Sci., vol. xn, p. 74), Professor 0. C. Charlton gave an interesting account of finding re- mains of elephants in Franklin County. It appears that at intervals during many years bones and teeth had been washed out of the bank of Osage (Marais des Cygnes) River, just above Rantoul. The exact locality was given as being in the northeast quarter of the northeast quarter of section 17, township 17 south, range 21 east. One of the teeth weighed 9 pounds ; the diameter of the glenoid cavity of a shoulder-blade measured 11 inches. At this point the river-bed is formed by shale, on which rests a foot of gravel. Then come 6 feet of a whitish, friable clay and above it 12 feet of brittle clay and the surface soil. The elephant remains appear to have been at the bottom of the whitish clay. Professor W. B. Wilson, of Ottawa University, Ottawa, Kansas, has sent to the writer a photograph and measurements of a large lower molar, one of those found at Rantoul, and these show that the animal was Elephas columbi. The molar is in fine condition and had just begun to wear. Burlington, Coffey County (3). — In the collection of the Philadelphia Academy of Sciences is a single ridge-plate of a tooth of Elephas columbi, which is recorded as having been found near Burlington, at a depth of 30 feet. This fragment was mentioned by Leidy in 1870 (Proc. Acad. Nat. Sci. Phila., 1870, p. 69). It had been presented to the Academy by Dr. W. F. McAllister, of Burlington. With it, and also found in gravel at a depth of 30 feet in digging a well, was a part of the lower jaw of a bison, which Leidy thought might belong to the existing species. It is more probable that the species is an extinct one. Arkansas City, Cowley County (4). — The writer received from Louis S. Crouse, of Arkansas City, a letter and a small photograph of an upper hind- most molar which appears to belong to Elephas columbi. It was discovered by Mr. Crouse in a gravelly deposit, under 3 feet of water, in Walnut River, about 1.5 miles east of Arkansas City. Evidently the tooth had recently been washed out of a bank of Pleistocene materials. The tooth has attached to it a fragment of the jaw. There are apparently about 16 rid