Transcriber’s Note:

The cover image was created by the transcriber and is placed in the public domain.

THE PLEISTOCENE OF NORTH AMERICA AND ITS VERTEBRATED ANIMALS FROM THE STATES EAST OF THE MISSISSIPPI RIVER AND FROM THE CANADIAN PROVINCES EAST OF LONGITUDE 95°.

BY

OLIVER P. HAY

Associate of the Carnegie Institution of Washington

Published by the Carnegie Institution of Washington, February, 1923

CARNEGIE INSTITUTION OF WASHINGTON

Publication No. 322

TECHNICAL PRESS

WASHINGTON, D. C.

CONTENTS

ILLUSTRATIONS

PREFACE.

The writer has been engaged for several years on an investigation of the Pleistocene geology of North America and of the Vertebrata which have been discovered in the deposits of this epoch. It had been his expectation to publish the results of all his studies at the same date. However, on consultation with Dr. John C. Merriam, it was agreed that it would be better to publish immediately that part which pertains to the region lying east of the Mississippi River and, as to the country further north, that east of longitude 95°.

At the outset the writer was convinced that, before just conclusions could be reached, it was necessary to know what fossil materials had been collected and under what geological and geographical conditions. He therefore made as thorough a search as possible of the literature for reports of discoveries of fossil vertebrates. Also, when in scientific journals or in newspapers the finding of fossils was recorded, recourse was had to correspondence, thus securing much exact information as to locality, kind of matrix, depth, and other important data. Often photographs have been obtained and even the materials themselves. The writer has also visited many museums and colleges throughout the country and examined their collections. Even in the smaller institutions, where perhaps only a few objects have been secured and preserved, some of these have furnished important information. Regret may be expressed that in the larger museums and colleges, as well as the smaller ones, too often there have been preserved only meager or no records regarding the history of what would otherwise be valuable specimens.

In order to show the geographical distribution of the most important species that occur in considerable numbers, a series of maps has been prepared, pertaining to the following:

• Whales and porpoises.
• Seals and walruses.
• The edentates.
• Elephas primigenius.
• E. columbi.
• E. imperator.
• E. species undetermined.
• Mastodons, mostly Mammut.
• Horses, mostly Equus.
• Tapirs.
• Peccaries.
• Camels.
• Odocoileus.
• Cervus.
• Rangifer.
• Musk-oxen.
• Bisons, extinct.
• Bison bison.
• Giant beavers.

Where the map of a State has become too crowded with numerals, a special map of that State for that species or genus has been prepared. There are maps of the edentates in Florida; mastodons of Indiana, of New York, of Ohio, of Michigan, of Florida; Elephas columbi in Florida; Elephas imperator in Florida; horses in Florida.

Other maps and figures for illustration of the Pleistocene geology will be found in their proper places.

The first part of the present volume is occupied by a consideration of the specimens recorded on the maps. Such information is noted as could be secured, often satisfactory, little enough sometimes; but it has been found that one can not foresee what important information a given fossil may furnish. At least, the presence of the fossil at a locality indicates the existence there of Pleistocene deposits of some kind. In cases where other species have been associated with the one mapped and described, these are noted.

When the consideration of these mapped species and genera is completed, the Pleistocene geology of the various States and provinces is taken up, so far as it is related to the vertebrate palæontology. This involved an examination of much of the literature of the Glacial period; and here one soon finds himself in face of huge tomes and endless articles and detailed maps. Only somewhat less in amount is the literature of the States beyond the glaciated area. The opportunity to misunderstand and to commit errors is unlimited, and the writer can only hope for lenient criticism.

An attempt has been made in the case of all vertebrate fossils to determine their geological relations and to derive some general conclusions regarding the history of our Pleistocene vertebrates and their relation to the divisions of the Pleistocene epoch. The conclusions reached are embodied in the immediately succeeding pages.

Not much attention has been given to the fossil invertebrates and plants. It is evident that neither the mollusks nor the plants have undergone any considerable change during Pleistocene times and are therefore not available as indicators of geological stages, though often useful for determining local climatic conditions. Their value can be better utilized by the palæomalacologists and palæobotanists.

To the officers of museums and colleges and to the private individuals who have so freely offered the use of their materials and in other ways aided the writer, he takes pleasure in expressing his sincere thanks. Most of all, however, he is indebted to the Carnegie Institution of Washington for the generous support extended during the years of this investigation.

June 1, 1922.

Oliver P. Hay.

THE PLEISTOCENE OF NORTH AMERICA AND ITS VERTEBRATED ANIMALS.

CONCLUSIONS REGARDING THE DIVISIONS OF THE PLEISTOCENE.

I. Limits of the Pleistocene.

The Pleistocene is regarded as being equivalent to what is known as the Glacial period. It began with the deployment of the ice-sheets which, proceeding from their centers of accumulation in British America, laid down in the East the Jerseyan drift and in the West the Nebraskan. The more the Glacial period is studied the more one becomes impressed with the significance of its physical effects on the northern hemisphere and with its influence on the vertebrate life. Doubtless its effects on the world in general are only beginning to be comprehended. The writer knows of no other phenomena, geological or biological, which so well characterize the Pleistocene period as do those comprehended under the term Glacial. They constitute the key to the determination of the subdivisions of the epoch and of their succession and to the history of the vertebrates which during this time occupied the continent.

II. The Blanco Pliocene.

The Blanco is held to belong to the upper, or to the uppermost, Pliocene. It is at present assigned to the Middle Pliocene (Osborn, Bull. U. S. Geol. Surv. No. 361, p. 81; Matthew, ibid., p. 120). Until recently the oldest known Pleistocene vertebrates appeared to be represented by the collections which long ago were made at Fossil Lake, Oregon, and at Grayson (Hay Springs), Nebraska. These assemblages had formerly been referred to the Pliocene, and the belief that they belong there is not yet wholly without supporters. It seemed, therefore, proper to retire the Blanco somewhat. The discovery that the Fossil Lake and Grayson faunas were represented in the Aftonian deposits of Iowa, and belonged probably to the first interglacial stage, reveals the fact that the geological interval between the Blanco and the Aftonian is at least partly filled by the first glacial stage, the Nebraskan. Naturally, it is to be expected that the breach between the earlier and the later faunas will be occupied, in part at least, by the vertebrates of the Nebraskan. What these are is not yet well determined; but the writer believes that as the Blanco and its equivalent and closely related formations and faunas become better known, they will be attracted close to the Pleistocene.

Aside from the facts just mentioned, the Blanco fauna seems to the writer to be more closely related to the Aftonian than has been supposed. The genera which occur in the Blanco are the following:

• Megalonyx.
• Mylodon.
• Glyptotherium.
• Hipparion.
• Pliohippus.
• Protohippus.
• Platygonus.
• Pliauchenia.
• Anancus.
• Gomphotherium.
• Stegomastodon.
• Felis.
• Amphicyon?
• Borophagus.
• Canimartes.

Of these, Megalonyx, Mylodon, Hipparion, Platygonus, Anancus, Gomphotherium?, Stegomastodon, and Felis are known from the first interglacial stage. Anancus includes mastodons with short, tuskless lower jaws and trefoiled molars. Gomphotherium, having long lower jaws with tusks, upper tusks with enamel band, and with trefoiled molars, may be represented by some of the early Pleistocene species. The same species of Stegomastodon appears to be present in the Blanco as in the Pleistocene, S. mirificus. The edentate Glyptotherium is not far removed from Glyptodon of the early Pleistocene. The Blanco genera of horses are so close to Equus that Cope regarded them as belonging to this genus.

The matter may be looked at from another point of view. If Mylodon, Megalonyx, and Glyptotherium are referred to the Middle Pliocene, we shall probably have them recorded as living in Texas before they existed in South America. It is true that Santiago Roth (Neues Jahrb., Min. Beil., Bd., vol. XXVI, table opposite p. 144) states that Glyptodon occurs in the Lower Pampas beds, and these he refers to the Upper Miocene; but the writer believes that Wilckens (Neues Jahrb, Min. Beil., Bd., vol. XXI, p. 193) is more nearly correct in placing them in the Pliocene. While the opinion may be correct that, when no obstacles intervene, the time required for mammals to spread over even a continent constitutes but a small part of a geological age, yet in making their way from South America, especially from Argentina, along the narrow bridge that appears to have been offered them, probably over mountain ranges, and across rivers and gorges, and in the face of the competing fauna advancing from the north, some of which were wolves and saber-tooth tigers, the slowly plodding and inoffensive edentates would have encountered too many hindrances to be able to make the journey in a short time.

The writer, therefore, ventures to range the Blanco immediately below the Pleistocene. On about the same level may be placed the Tulare-Etchegoin and the Thousand Creek formations of Merriam (Bull. Dept. Geol. Univ. Calif., vol. X, pp. 425, 429).

III. The Historical Divisions of the Pleistocene.

The writer accepts the divisions of the Pleistocene which the geologists appear to have established. Formerly it was believed that North America had been subjected to a single glacial epoch; now it seems to be proved that there have occurred five such glacial epochs, or stages, and that there have intervened four interglacial stages of mild climate. The interglacial stages are italicized. The Nebraskan stage is the earliest, the Wisconsin the latest: Wisconsin, Peorian, Iowan, Sangamon, Illinoian, Yarmouth, Kansan, Aftonian, Nebraskan.

The characteristics of the various stages will be briefly discussed. The stages are not equally well understood and at present do not seem to be of equal importance in their relation to vertebrate paleontology.

IV. Elevations of the Continent Immediately Preceding or Accompanying the Opening of the Pleistocene.

In pursuing the study of the Pleistocene, one soon realizes that this period was one of great geological activity. Ranges of mountains, if not begun anew, were at least raised to greater altitudes. The Cascade Range appears to have begun to rear its head at the beginning of the epoch, or even a little later. Here and there the crust of the earth was ruptured and great sheets of lava were poured out over the land. Ice caps repeatedly accumulated over large areas in North America and Europe, and in their movements southward transported vast amounts of earthy débris and turned the courses of great streams. Apparently at times the rainfall was greatly increased. The rivers, quickened by greater slope and the increased volume of water, cut their channels deeper and in the mountains excavated profound gorges. Through elevation of the land North America was, late in the Pliocene or early in the Pleistocene, put into easy communication with Asia and South America, so that vertebrated animals passed freely to and fro. A part of these activities probably belonged to the latter part of the Pliocene. In the more elevated regions of the eastern United States, through the chemical, rupturing, and transporting properties of water, rocks were dissolved and their disintegrated materials produced what has been designated the Lafayette formation; but it is possible that this belongs to the early Pleistocene.

V. Connections with Asia and South America.

Mention has just been made of a land connection with Asia at some time about the beginning of the Pleistocene. The evidence for this may be called circumstantial rather than direct. The geological evidence has not been developed. If any deposits containing marine fossils had been laid down along the Asiatic and Alaskan coasts during a time of elevation, they would now be covered by the sea. Our evidence for the connection is derived from the distribution of the vertebrate animals. During the early Pleistocene our country was invaded by a host of mammals whose home was originally in Asia. These included elephants, bisons, elk, goats, bears, wolves, and foxes, besides many mammals of smaller size. It is the presence in America of the smaller animals, many genera of rodents of Asiatic origin, that shows that there must have been a land connection. These could not have made their passage across Bering Strait on the ice, as it might be imagined the larger animals did.

The way between the two continents had more than once before been open, but it was during the early Pleistocene that modern Asiatic genera entered North America in great numbers. Exactly where the land bridge between the two countries was situated is not certain; it may be that a large part of the area now occupied by Bering Sea was then dry land. Arldt (Entwicklung der Kontinente, plate 21) represents a connection extending from the northern border of Alaska southward to include the Aleutian Islands. Where narrowest, this bridge, as represented by the author named, extended from latitude 60° to 70°, a distance of about 700 miles. In such case the cold currents from the Arctic Ocean would have been prevented from entering the Pacific, while the Japan Current would have warmed up the southern side of the bridge. The route was then open on the north for the boreal animals of Asia to enter Alaska; while on the south the genera inhabiting the more temperate part of eastern Asia would have had free access to the American shore. Once on the continent, the boreal mammals might have spread along the shores of the Arctic Ocean and those of the temperate parts of Asia have made their way up the Yukon Valley, or possibly along the Pacific coast, to the warmer regions toward the south. We do not need to suppose that even during the first glacial, or Nebraskan, stage the climate of that part of North America was as inclement as now.

At the other end of our continent a train of events not wholly dissimilar was in motion. Even in the latter part of the Pliocene some South American edentates, such as Megalonyx, Mylodon, and Glyptotherium, had reached Texas. Probably a little later the bridge had become widened so that other edentates and a few genera of South American hystricine rodents swarmed into our southern borders. At the same time a host of carnivores, tapirs, horses, camels, peccaries, deer, and cricetine and sciurine rodents made their way into South America. It is now certain that the land bridge over which the interchange took place did not include the West Indies. Possibly there yet remained along the western coast of Central America some of the border, now submerged, which Schuchert (Bull. Geol. Soc. Amer., vol. XX, plates 96 to 100) represents as being present during the Tertiary.

VI. The Sources of the Vertebrates of the Pleistocene.

The Pleistocene vertebrate fauna of North America has been derived from three sources. One component had descended from the animals which occupied the continent during the late Tertiary, but even these were of mixed derivation. A few appear to have filtered in from South America during the Pliocene; others had come from Asia during Tertiary invasions; but a large element was native to the country. As such may be taken the camels, the peccaries, the three-toed horses, the prong-horn antelope, the deer of the genus Odocoileus.

Upon a continent of vast extent and great fertility, possessing unbounded variety of climate and habitat, all these animals were thrown together to struggle for their existence. We must depend upon the imagination to picture what the result would have been if nature had pursued a course which might have been predicted. What the result in reality was, we shall see.

VII. The Richness of the Pleistocene Vertebrate Life.

It will be profitable to consider briefly the character of the Pleistocene vertebrate fauna. The writer has compiled a list of the species which have, so far as he knows, been collected and described up to this time. There are in all 637 species; of these, 387 belong to the mammals, 154 to the birds, 26 only to the reptiles, 7 to the amphibians, 56 to the bony fishes, and 7 to the group of sharks and rays. Certainly these form only a part of the species which existed. At present there are known in our existing fauna north of Mexico 693 species of mammals, excluding the cetaceans—somewhat more than twice the number of known Pleistocene species. It is, however, rather in the great variety of forms that the Pleistocene excelled. Following Gerrit S. Miller’s Land Mammals of North America, 1912, we find in our present fauna north of Mexico 29 families; in the Pleistocene there are now known 37 families, not including the cetaceans. In our existing mammalian fauna there are recognized 111 genera; in the Pleistocene, with hardly half as many species recorded, 138 genera are counted. In order to realize more vividly the variety of Pleistocene forms, we have only to recall the animals then present, now absent, namely, the great ground-sloths, the glyptodons, the numerous species of horses, tapirs, numerous peccaries, camels, the extinct relatives of the musk-oxen, extinct bisons, elephants, mastodons of three or four genera, the giant beaver, and the saber-tooth tigers. Among the birds, reptiles, batrachians, and fishes, there were few striking forms, and these were mostly among the birds and the tortoises.

The above account shows the great richness of the vertebrate life during the Pleistocene; furthermore, this abundance evidently existed during the early stages of the epoch. It constituted the materials on which that combination of conditions which we call environment had to work during Pleistocene times. The comparison shows that the result was an impoverishment of the vertebrate fauna. Genera and families, even orders, were wiped out of existence, and these included some of the noblest animals that have graced the face of the earth, the elephants, the mastodons, tapirs, many species of bison, horses, saber-tooth cats, huge tigers, and gigantic wolves. The following nine or ten families became either wholly extinct or continued to exist only in other more hospitable lands: the Megatheriidæ, including several genera of ground-sloths; the Hoplophoridæ or glyptodons; the Caviidæ, which embraced one or more species of huge capybaras; the Elephantidæ, under which are arranged three or four species of elephants and three genera of mastodons; the Equidæ, represented by a dozen or more species of horses; the Camelidæ, of which there were several Pleistocene species and probably three or four genera; the Hyænidæ, of which there appears to have been at least one genus, with one species; the Tapiridæ, including three or four species; and probably the Rhinocerotidæ. Besides these, the subfamily of Felidæ known as Machairodontinæ, embracing those wonderful carnivores the saber-tooth tigers, was suppressed. The Dasypodidæ, which included some armadillos 5 or 6 feet long, are now represented by only one small species in Texas. Of the Tagassuidæ, to which belonged several genera and stately species of peccaries, there exists now in North America north of Mexico but one species, an animal of only moderate size.

VIII. On Evolution During the Pleistocene.

We have seen that the Pleistocene fauna was very different from that which existed when white men first entered the country; also that the difference has in large part been due to the destruction of species, genera, and families. We may now inquire whether or not the loss has been to any considerable extent compensated by the development of new forms. Many of our existing genera and species have been found in the collections that represent the earliest Pleistocene known to us. The writer believes it would be unsafe to say that any living species that one might select may not hereafter be discovered in early Pleistocene collections. It is probably true, however, that some of those small changes by which we distinguish one species from another have been produced. Some small but persistent differences might, for example, have arisen in the teeth or in the form of the skull of a group of muskrats which would justify us in regarding it as forming a new species. It is extremely doubtful that any new genus of vertebrates has been developed since the first interglacial stage. Matthew has concluded (Science, n. s., vol. XL, pp. 232–235) that the evolution of the mammals during the Pleistocene amounts to about one-tenth of that achieved during the Pliocene. The present writer regards this as a liberal estimate.

This failure to evolve new genera and species is not necessarily to be attributed to the shortness of the Pleistocene period; it may have been due rather to the unfavorable conditions. In what direction could an animal make progress when, after being subjected for some thousands of years to one set of conditions, it was compelled for some other thousands to endure just the opposite conditions? If life in front of a glacier for some centuries led to the development of a coat of hair on an elephant, that coat would probably disappear during the succeeding interglacial stage, and in the end, if the elephant had not perished, he would be where he began.

Too much stress must not, however, be placed on this suggestion. It may yet be possible to show that nowhere in the world was any considerable progress made by mammals during the Pleistocene, in the modification of their forms and structure. On the other hand, it is also possible that all over the world climatic conditions were at intervals unfavorably affected by the development of the great glaciers and that all life was retarded in its evolution. The writer believes, therefore, that it can not be shown with certainty that new forms of living things, especially vertebrates, were developed in North America during the Pleistocene. It may be quite as difficult to prove that any genera or species of importance entered from other lands after the first invasion. Under these conditions there appears to be no means for determining successive faunas other than through recording the time of the disappearance of genera and species.

IX. Did the Extinction of Species Take Place Mostly at the End of the Pleistocene?

At the beginning of the Pleistocene there existed, as has been shown, an abundant and highly varied mammalian fauna; at the close of the epoch this fauna had become relatively impoverished. Did all those families and genera and species, that in the end were missing, perish during or after the last glacial stage, the Wisconsin? This opinion has been expressed by some. The writer believes that this view is wholly improbable.

A glacial sheet, stretched across the continent or a large part of it, was not local in its effects; it was not a cap of ice merely concealing a part of the land and covered possibly by forests and allowing occupation by certain hardy animals, while beyond, up to its foot, the country was pleasantly cool, wooded, and abounding with animated creatures. In the Sierra Nevada Mountains of California (Lindgren, Folio 66, U. S. Geol. Surv.) and of Nevada (Knopf, Prof. Pap., U. S. Geol. Surv., 110, pp. 92–105) and in the San Juan Mountains of Colorado (Atwood and Mather, Jour. Geol., vol. XX, p. 385), at distances of approximately 600 or 700 miles from the glacial front, there existed, during more than one stage, extensive local glaciers. Along the Atlantic coast during at least one glacial stage the walrus was driven as far south as Charleston, South Carolina. One can hardly doubt that the whole continent was chilled during each of the glacial stages.

To mammals, which for perhaps various reasons had been with difficulty enduring the stress of existence, the glacial climates would give the final stroke; perhaps to others the interglacial climates would have been quite as fatal. We can not doubt that each glacial and each interglacial stage swept away a few of the less hardy genera and species. Nevertheless, several remarkable animals passed through the vicissitudes of all the glacial and interglacial times and left their bones in the deposits overlying the last, or Wisconsin, drift. Such are two species of elephants, the American mastodon, the giant beaver, and one or more species of peccaries. Why they succumbed at last is difficult to say. Possibly the return of a fifth warm era proved too much for their endurance.

A reason for believing that the genera and species missing from the fauna found here when white men arrived, called sometimes the Columbian fauna, were exterminated gradually and not at one epoch is that certain ones are found in deposits overlying the earlier glacial drift-sheets, but are not found in deposits on later drifts. Camels occur in Aftonian beds overlying the Nebraskan drift, but have not been collected in later interglacial deposits. Horses grow scarcer as the Pleistocene advances. They are known from deposits overlying the Illinoian drift, but do not appear after the Wisconsin.

X. The Stratigraphical and Time Limits of the Earliest Pleistocene.

It is necessary to determine, if possible, where the boundary line shall be drawn between the Pliocene and the Pleistocene. Room must be made for the first interglacial, the Nebraskan, and its fauna. How long this first glacial stage continued we do not know. Chamberlin and Salisbury have indicated (Geology, vol. III, p. 420) that in a rough way the dates from the present of the culmination of the various glacial stages, except the Nebraskan, taken in order backward, may be represented by the geometrical series 1, 2, 4, 8, 16. That is, if the Illinoian stage had its culmination 150,000 years ago, that of the Kansan occurred 300,000 years ago; if the Nebraskan should fall in the same series, it culminated 600,000 years ago; and it and the succeeding Aftonian interglacial held sway as long as all the rest of the Pleistocene put together. It would be rash to assert that this first glacial did last so long; but we see the possibilities. In a personal communication Professor Frank Leverett writes that he estimates that the Kansan culmination took place at not less than 400,000 years ago and the Nebraskan at 500,000. This, as the present writer estimates, would leave for the Nebraskan itself somewhere near 40,000 or 50,000 years. Some changes in the life of the Pleistocene must have been wrought during those years.

The glacial deposits of the Nebraskan stage are not as well known as one might wish. They appear to be in general overlain by the later drifts and are observed mostly where streams have cut through both the overlying drift and the Nebraskan. The old drift found in New Jersey is thin and of no great extent. Moreover, we can hardly expect to find fossil vertebrates in the drift itself. We must therefore depend on studies of supposed Nebraskan fossils found mostly outside of the glaciated area and make comparison of them with earlier and later faunas. If we shall discover collections of Nebraskan vertebrate animals, we may be sure that they will differ from those of the first interglacial, the Aftonian. We may be pretty certain that they will include autochthonous genera of the late Tertiary, which may be missing from the Aftonian, together with at least a few genera from South America and others from Asia.

Now, have any formations and included fossil vertebrates been found which may be fitted into the Nebraskan interval?

In this stage the writer places the beds which Cope designated the Idaho formation (Cope, Proc. Acad. Nat. Sci. Phil., 1883, p. 135). Since Cope’s time several new species have been added to his list from this formation. In 1917 (Bull. Dept. Geol. Univ. Calif., vol. X, p. 432), Dr. J. C. Merriam published a list of the fossils, except fishes, which had been secured up to that time. The list of species referred to the Idaho formation is as follows:

• Equus idahoensis.
• E. excelsus?
• Protohippus?
• Rhinoceros, probably Aphelops (Teleoceras) fossiger.
• Mastodon mirificus.
• Cervus, possibly new. Smaller and more slender than C. canadensis.
• Procamelus, size of P. major.
• Tragocerus? horn-core of antelope.
• Ischyrosmilus n. sp.
• Morotherium leptonyx.
• Castor, possibly n. sp.
• Olor, size of O. paloregonus.
• Graculus idahoensis.

In this collection the presence of horses of the genus Equus, of Cervus, Morotherium, and Castor, is strongly suggestive of the Pleistocene. The type of Mastodon mirificus was found in Pleistocene deposits of probably Aftonian age. Although rhinoceroses are supposed to have become extinct before the end of the Pliocene, this supposition may be an error. The list of Blanco vertebrates is a short one, and the absence of a genus from it is not decisive. One drawing of a seine in the sea-waters of Florida would furnish inadequate materials for conclusions about the fish fauna of that coast.

The Thousand Creek fauna (Merriam, Bull. Dept. Geol. Univ. Calif., vol. X, p. 429), which to the present writer appears of about the same age as the Blanco, contains a species of Teleoceras. The genera Protohippus and Procamelus might be supposed to have continued their existence and evolution until interrupted by an age of ice and by competitors from Asia.

In 1917 (Bull. cit., vol. X, pp. 255–266) Merriam and Buwalda published a short list of fossils which they had collected along the Columbia River in Washington State. A horse was found which was referred to Equus or Pliohippus; also two camelids, one of which was thought to be near Pliauchenia. Merriam concluded that the evidence on the whole favored the Pleistocene. The list will fit into the Nebraskan without difficulty.

In 1889 (Amer. Naturalist, vol. XXIII, p. 253), Professor E. D. Cope published a list of fossil mammals collected in the “Oregon desert,” apparently somewhere in the region of Silver Lake or Summer Lake. The list is as follows:

• Canis sp. indet.
• Elephas or Mastodon.
• Holomeniscus or Auchenia.
• Aphelops sp. indet.
• Hippotherium relictum.
• Equus sp. indet.

Cope looked upon this collection as remarkable in that it showed the presence of true horses and camels associated with a rhinoceros. He concluded that the fossils belonged to his Idaho formation. Dr. W. D. Matthew thought that the collection was a mixture of fossils from two formations (Bull. Amer. Mus. Nat. Hist., vol. XVI, p. 321). It may, however, have been made in Nebraskan deposits.

In 1921 (Proc. U. S. Nat. Mus., vol. LIX, pp. 617–638), the writer described a collection of vertebrate remains from Anita, Coconino County, Arizona. These remains were found in a cave in making explorations for copper ore. The list follows:

• Equus occidentalis.
• E. giganteus?
• Mylohyus? sp. indet.
• Procamelus coconinensis.
• P. longurio.
• Antilocapra americana?
• Marmota arizonæ.
• Citellus tuitus.
• Neotoma cinerea.
• Lepus benjamini.
• Brachylagus browni.
• Taxidea robusta.
• Canis nubilus?
• C. latrans?
• Chasmaporthetes ossifragus.

The writer believes that this assemblage of mammals must be referred to the Pleistocene. It will be noted, however, that there are two species of the genus Procamelus. These resemble so much two species, P. major and P. minimus, described by Leidy and Lucas (Trans. Wagner Free Inst., vol. IV, pp. I-XIV, 15–61) from the Alachua clays of Florida, that it seemed at first necessary to identify them as such. The genus Procamelus seems, therefore, to be brought definitely into the early Pleistocene, probably the Nebraskan.

The collections made in the Alachua clays in Florida were obtained in Alachua and Levy counties. On pages [195] and [375] will be found an account of the geological conditions under which the fossils were found, and lists of the species. The essential features are that such supposed Miocene or Pliocene genera as Gomphotherium, Procamelus, Teleoceras, and Aphelops were found associated with the Pleistocene genera Odocoileus, Tapirus, Megatherium, and Equus. This has been explained on the theory that the clays are of Tertiary age and that the Pleistocene species had become mingled with those of an earlier time. At a number of places in Florida where phosphate rock has been mined there have been secured similar associations of early camels, rhinoceroses, horses (Hipparion, Parahippus) with genera belonging undoubtedly to the Pleistocene. This has occurred so often that the writer doubts the correctness of the explanation given. He ventures, therefore, to include in the Pleistocene of the Nebraskan stage the various deposits that have received the names Alachua clays, the Dunnellon formation, and Bone Valley formation. The latter, called also the land-pebble phosphates, is believed by Sellards to be contemporaneous in age with the Dunnellon or hard phosphates, but to have accumulated under different conditions. Both the Alachuan and the Bone Valley formations were referred by Sellards to either the late Miocene or the early Pliocene, with an evident preference for the latter. It seems to have been the presence of the rhinoceroses that most influenced him in his assignment of the deposits; but there were naturally other considerations. He wrote:

The presence of rhinoceroses in the formation is believed to establish definitely the fact that the beds can not be later than the early Pliocene, since rhinoceroses in America apparently did not survive beyond that time (Fla. Geol. Surv., vol. VII, p. 73).

According to Sellards the hard phosphate, belonging to the Alachua (Dunnellon) formation (Fla. Geol. Surv., vol. V, p. 37) resulted from a disintegration of underlying Upper Oligocene deposits and probably the Vicksburg limestone. Through chemical action these rocks were partly dissolved and the residual materials were mixed by local subsidence and by action of streams and later modified by chemical changes.

The land-pebble phosphate of the Bone Valley formation had, Sellards concluded (Fla. Geol. Surv., vol. VII, p. 55), resulted from underlying phosphate marls of Upper Oligocene age. This occurred during a time of general subsidence of sufficient extent to permit marine waters to reach the area covered by the Bone Valley phosphates. The presence of sea-water is indicated by the occurrence of bones of cetaceans.

With regard to the effects of streams and of the chemical action of the water on the rocks, which contributed to the formation of the hard rock phosphate and the production of sinks and caves, it may be remarked that we know of no time when rocks were dissolved and caves formed to the extent that they were during the Pleistocene.

As shown on page [15], various deposits of marine marls along the Atlantic coast are referred by the writer to the Nebraskan. Among these marls are the coquina rock found at St. Augustine and the marine marl underlying the bed at Vero, which contained early Pleistocene vertebrate fossils. These marls are known to extend well inland, being found at Kissimmee, 50 miles from the coast. In some places they are met with at depths of 70 feet (Sellards, Fla. Geol. Surv., vol. VIII, pp. 105–106). Marls of probably the same age occur on the western coast of Florida (Dall, Bull. 84, U. S. Geol. Surv., p. 152). The writer believes that some of these marls may yet be connected with the phosphate beds of the Bone Valley formation.

A figure taken from Sellards (Geol. Surv. Fla., vol. VII, opp. p. 53) may be found on page [377]. This illustrates the relation of the Dunnellon and Bone Valley formations to the underlying deposits.

XI. The First Interglacial, or Aftonian, Stage.

Mention has been made of collections of fossil vertebrates which long ago were secured at Fossil Lake, Oregon, and of others along Niobrara River, near Grayson, Nebraska. Lists of the species found at each locality were given by Dr. W. D. Matthew in 1902 (Bull. Amer. Mus. Nat. Hist., vol. XVI, pp. 317–320). These deposits and animals were regarded by Cope and Marsh as belonging to the Pliocene, until G. K. Gilbert, in his work on Lake Bonneville (Monogr. I, U. S. Geol. Surv., pp. 393–402) showed that the Oregon fossils must belong to the Glacial epoch, but he referred them to a late time in this epoch, that of the last glaciation. It thus became quite impossible to determine the age of any collection of fossil vertebrates.

In 1887 (Univ. Geol. Surv. Kansas vol. II, pp. 299–308), Williston wrote:

Every fact furnished from Kansas seems to substantiate Cope’s conclusion that the Megalonyx fauna of the East and the Equus fauna of the West were contemporaneous and that both occurred during the period of depression; that is, during late Pleistocene time.

This paragraph was quoted by Osborn in 1910 (“Age of Mammals,” p. 453), who appears to agree in part with Williston, although he expressed the opinion that some of the deposits were earlier than the others. Osborn supported the view of the existence of two faunas, that of the “Equus zone” and that of the “Megalonyx zone.” The former fauna was regarded as the older, but overlapping somewhat during the “mid-Pleistocene” the Megalonyx fauna. He presented a catalogue of deposits belonging to his Equus zone (his page 453) and another of those of the Megalonyx zone (p. [467]). In the latter he included deposits that he would now doubtless refer to the earliest Pleistocene, for example, the Ashley River beds.

It was necessary for the geologists to come again to the rescue of the palæontologists. They established the fact that there had passed, not a single glacial stage, but a series, and that these had been separated by corresponding interglacial stages. They were able to show also that between the drift-sheets there were to be found remnants of old gravels and fossil-bearing soils. In Iowa, through the careful researches of Calvin and Shimek, numerous remains of fossil mammals were discovered in gravels lying between the earliest drift, the Nebraskan, and the second drift, the Kansan. Among these mammals were identified horses, camels, elephants (E. columbi, E. imperator), Mylodon, Megalonyx, and a large ruminant which is certainly a species of bison. This fauna, known as the Aftonian, was correlated by Calvin with that of the Sheridan beds of Nebraska (Bull. Geol. Soc. Amer., vol. XX, p. 354). The writer has had the opportunity to study this Aftonian material (Iowa Geol. Surv., vol. XXIII), and, although it is not as abundant as might be desired, he agrees with Calvin’s correlation.

Making due allowances for environment and the hazards attending preservation and collection, the Aftonian and Sheridan fauna is practically the same as that found at Fossil Lake, Oregon. Furthermore, it may be traced along the plains into Texas and to the shores of the Gulf. Here, at or near tide-level, or not far away, may be found horses, camels, elephants (E. columbi and E. imperator), Mammut americanum, and mastodons with teeth presenting trefoils. In Texas, within a mile of the Louisiana line, Elephas imperator has been collected. The fauna reappears on the west coast of Florida; also on Peace Creek; on the east coast at Vero; at Brunswick and Savannah, Georgia; along Ashley River, near Charleston; probably also on the banks of Neuse River, 16 miles below New Bern, North Carolina; and again probably at Long Branch, New Jersey, where Megatherium has been found; and finally at Port Kennedy, on Schuylkill River, about 25 miles above Philadelphia. All along the coast, apparently from the Rio Grande to Long Branch, the localities which furnish Aftonian fossils are within a few feet of sea-level.

XII. The Yarmouth Interglacial Stage.

Up to the present time the interglacial soils found in a few localities between the Kansan and the Illinoian drifts have furnished only scanty remains of vertebrate fossils—a rabbit and a skunk at the type locality in Iowa. Certainly, however, the same animals were living then that were found at later stages.

XIII. The Illinoian Glacial Stage.

To the Illinoian glacial stage the writer refers the collection of fossil vertebrates which was described in 1908 by Barnum Brown (Mem. Amer. Mus. Nat. Hist., vol. IX, pp. 157–208) and which had been obtained in the Conard fissure near Willcockson, Newton County, Arkansas. It is placed here rather than in the Sangamon stage, because of the number of species present which suggest a rather cold climate. A list of these species will be found on pages 31–32 of volume XXIII, of the Iowa Geological Survey.

XIV. The Sangamon Interglacial Stage.

This was the warm stage which succeeded the glacial Illinoian. Between the Illinoian and the Wisconsin there passed a long period of time. It is now believed that it was interrupted by the Iowan ice-sheet, but this appears not to have lasted long nor to have occupied any considerable area. Associated with it in some way was the accumulation of much loess. This was formerly supposed to have been deposited to a large extent at least during the Sangamon; but, as Leverett informs me, it appears to have been laid down at a time nearer the Wisconsin than the Illinoian. This Iowan drift and the loess has been the subject of a special investigation by Alden and Leighton (Iowa Geol. Surv., vol. XXVI, pp. 49–212). Few vertebrate fossils have been found in the loess. Their bones may have been dissolved out by the percolating rain-water, and yet the delicate shells of land mollusks are abundant. A collection which the writer regards as belonging rightfully to the Sangamon was made at Alton, Illinois, many years ago, by William McAdams. A list of the species and an account of the geological conditions connected with it are presented on page [339]. The remains appear to have accumulated in a pond on the Illinoian drift and to have been covered by loess. The horse was yet in existence, as well as the deer Sangamona and the antelope Taurotragus americanus. Two-thirds of the 15 species are extinct. A smaller number of species have been collected near Kimmswick, just below St. Louis, Missouri. The remains found in a cave in Bexar County, Texas, are believed to belong here (Hay, Proc. U. S. Nat. Mus., vol. LVIII, p. 129). It is, however, in the Alleghany Mountains that most of the vertebrates have been collected which the writer refers to the Sangamon stage. These have been found in caves and fissures from northern Pennsylvania to northern Alabama. Unfortunately, although mostly discovered several years ago, some of these collections have not yet been well studied and have not been accessible to the writer. They contain two or three species of horses, two or three genera of peccaries, tapirs, the deer Sangamona, the antelope Taurotragus, and one or more species of saber-tooth tigers. Half or more of the species are extinct. To the writer these assemblages seem to fit into the history nowhere so well as into the Sangamon stage.

Another assemblage that probably belongs here is that made at Toronto (p. [282]). This indicates a warm climate, since the pawpaw and the osage orange grew there.

XV. The Peorian Interglacial Stage.

This is the interglacial interval between the Iowan glacial and the Wisconsin. It was probably not of long continuance and is chiefly remarkable for the deposition of loess. This has not furnished any important collections of vertebrate fossils. The type locality for the Peorian stage is a locality east of Peoria, Illinois. Leverett (Monogr. XXXVIII, U. S. Geol. Surv.) mentions several cases in which old soils believed to belong to the Peorian were observed in Illinois. None of these has furnished vertebrate fossils. It is usually difficult to distinguish the Sangamon from the Peorian soils.

XVI. The Wisconsin Glacial Stage and the Wabash Beds.

The next stage which furnishes abundant vertebrate fossils is the Wisconsin. These remains are found most abundantly in the old soils and mucks which accumulated in the swamps, ponds, and lakes left on the uneven surface of the Wisconsin drift as the ice retired. To such deposits the writer has given the name Wabash beds. They are often called post-glacial deposits; but that term ought in strictness to be applied only to deposits of the present epoch. They may be called Late Glacial, but that expression has been used for the drift and moraines produced by the second half of the Wisconsin glaciation. It might be better to use for the divisions of the Wisconsin the terms Lower and Upper.

In the late Wisconsin, or the Wabash, deposits there may be found remains of any of the existing animals of the region; also often the bones and teeth of mammals now living in more northern regions. Besides these, there may occur the relics of animals which were able to endure the rigors, changes, and competitions of the Glacial period, but succumbed at its end. These are, especially, two species of elephants, one or two species of mastodons, four or more species of musk-oxen, the moose Cervalces, one or more species of peccary, and the giant beaver.

XVII. On the Theory of the Pleistocene Terraces of the Coastal Plain.

The writer will discuss briefly the widely accepted theory that along the sea-coast from New Jersey to southwestern Texas there occurs a series of terraces and corresponding escarpments, three or more in number, representing successive emergences of the borders of the continent from the sea. The theory was first proposed by Dr. W. J. McGee (Amer. Jour. Sci., ser. 3, vol. XXV, 1888, p. 367; 12th Ann. Rep. U. S. Geol. Surv., pt. I, 1891, pp. 353–521). He included in the initial submergence not only the area occupied by the supposed Pleistocene terraces, but also the borders of the coasts to an elevation corresponding to the Lafayette (Appomattox) formation, which he referred provisionally to the late Pliocene. This submergence required a depression of the eastern half of the continent amounting to 500 feet or more. The theory was accepted especially by the geologists of Maryland in their excellent reports (Shattuck, Maryland Geol. Surv., Pliocene and Pleistocene volume, pp. 62–137, with maps). It has likewise been applied to the geology of Virginia (Clark and Miller, Va. Geol. Surv. Bull. No. IV, pp. 48–56, 179–189), North Carolina (Stephenson, N. C. Geol. Econom. Surv., vol. III, 1912, pp. 266–290), Georgia (Veatch, Geol. Surv. Ga., Bull. No. 26, 1911, pp. 35–50), as Okefenokee and Satilla; (Stephenson, ibid., pp. 425–445), Florida (Matson and Clapp, Fla. Geol. Surv., vol. II, 1909), and to Texas (Deussen, Water Supply Pap. U. S. Geol. Surv. 335, pp. 78–83).

In Maryland and the District of Columbia there have been recognized three Pleistocene terraces (Shattuck, as cited above). The uppermost is the Sunderland, the next the Wicomico, the lowest the Talbot. These are not correlated by Shattuck definitely with glacial divisions of the Pleistocene, but the Sunderland is the oldest, while the Talbot is regarded the most recent, probably about the age of the last glacial stage, the Wisconsin.

When the writer began his study of the Pleistocene he accepted the theory proposed by McGee and the Maryland geologists, and traces of this acceptance may be found in this work; but he is now convinced of its falsity. It is hardly to be believed that the coastal region could have been occupied, even at intervals, since the late Pliocene, when the depression is supposed to have been at least 500 feet, and 200 feet during the Sunderland, down to the end of the Wicomico and even the Talbot, without its having left other traces of marine occupation than the supposed terraces and escarpments. There ought to appear somewhere in the long border from New Jersey to Mexico abundant and extensive deposits of stratified materials, clays, sands, and gravels. Such deposits appear to be relatively rare.

A still more serious objection to the theory of submergence beneath marine waters is the absence of marine fossils. In the materials forming these terraces one might with confidence expect to find at least marine mollusks, mussels, clams, and beds of oysters; probably also remains of fishes, of porpoises, and of whales. Leaving out of consideration the Talbot terrace, which is near sea-level (Shattuck, op. cit., p. 10), the supporters of the theory under consideration admit that not in the Lafayette, nor the Sunderland, nor the Wicomico, have any traces of such fossils been met with. On the other hand, all over these terraces are found remains of land animals and plants. Mastodons, elephants, and horses are by no means rare. Conditions favorable for the preservation of teeth of proboscideans must have been quite as well adapted to preserve shells of oysters. In the Sunderland and Wicomico a few land plants have been secured, an abundance of them in the Talbot. Map No. 39 shows the distribution of Pleistocene mammals, mollusks, and plants on the Coastal Plain of North Carolina.

It seems evident, therefore, that the sea has had nothing to do with the formation of the Lafayette, the Sunderland, and the Wicomico terraces, and little with that of the Talbot. It was natural that the advocates of this theory of the formation of these terraces during the Pleistocene should distribute them somewhat impartially over the time of this epoch, assigning the Talbot to a late interval. On page [11] the writer has called attention to the fact that in many places along the coast from southeastern Texas to New Jersey, at or near sea-level, there are beds which contain a vertebrate fauna of the Aftonian or first interglacial stage. Probably nowhere do these beds have any large amount of later materials overlying them; it is often extremely little. So far as the writer can judge, this means that all the terraces and escarpments were produced before the time of the first interglacial; not since that distant time has there occurred along the Gulf or Atlantic coasts south of New Jersey any considerable elevation or depression of the Coastal Plain.

FINDS OF PLEISTOCENE CETACEANS IN EASTERN NORTH AMERICA.

(Map [1].)

ONTARIO.

1. Nepean Township, Carleton County.—In 1914, Mr. L. M. Lambe, of the Canadian Geological Survey, stated (Summ. Rep. for 1913, p. 299) that Walter Billings, of Ottawa, had presented to the Survey a caudal vertebra of Delphinapterus leucas, found in Pleistocene gravel on lot 15, concession 5, of Nepean township. The locality is near Jock River, a stream which flows northeasterly and enters Rideau River about 11 miles south of Ottawa. With it was sent the lower end of a femur, supposed to belong to the bison.

2. Ottawa East, Carleton County.—In 1910, Mr. L. M. Lambe reported (Summ. Rep. Geol. Surv. Can. for 1909, p. 273) that Mr. A. Penfold had presented to the Survey a caudal vertebra of Delphinapterus leucas, which he had found at Ottawa East, at a depth of 25 feet, while digging a well.

3. Smith’s Falls, Lanark County.—In 1883 (Amer. Jour. Sci., ser. 3, vol. XXV, p. 200) Dr. J. W. Dawson announced the finding of two vertebræ, a part of another, and a fragment of a rib of a large whale, in a ballast pit at Welshe’s, 3 miles north of Smith’s Falls. This whale he identified as Megaptera longimana (M. boöps). The bones were found in gravel at a depth of 30 feet and about 50 feet from the original face of the pit. The elevation of the place is given as about 440 feet above sea-level. Dawson stated that this corresponds exactly with the height of one of the sea-terraces on Royal Mountain at Montreal. He added that this animal might have sailed past that mountain, then only a rocky islet, when a wide sea, 400 feet above the lower levels of Montreal, covered all the plain of the lower St. Lawrence. Inasmuch as the highest terrace containing marine fossils at Montreal stands at a height of about 625 feet (Stansfield, Mem. 73, Canad. Geol. Surv., 1915) above sea-level, the region had apparently risen about 160 feet at least above its lowest submergence when the whale was buried. The discovery of this whale is mentioned by Dawson in his “Canadian Ice Age,” 1894, page 268; also by Professor G. H. Perkins in his Report of the State Geologist of Vermont, 1907–8, page 83.

4. Pakenham, Lanark County.—This locality is about 42 miles north-northwest from Welshe’s, where the whale remains just discussed were found. At Pakenham, in 1906, there were discovered bones, including a nearly perfect skull, of a white whale. The discovery was reported in 1906 and 1907 by Dr. J. F. Whiteaves (Summ. Rep. Geol. Surv. Can. for 1908, p. 171; Ottawa Naturalist, vol. XX, pp. 214–216). The remains were found by a well-digger on a farm (lot 21, 11th concession), and were embedded in blue clay at a depth of 14 feet. Immediately about the bones was a mixture of clay and shells. The animal has been referred to Delphinapterus leucas. As one of the ear-bones was secured, the determination of the species would appear to be possible. According to Perkins, the ear-bone in the type of D. vermontanus differs from that of the existing white whale, D. leucas. The writer is unable to say more than that the whale found at Pakenham belongs to the Late Wisconsin.

5. Cornwall, Stormont County.—In 1870 (Canad. Naturalist and Quart. Jour. Sci., vol. V, pp. 438–439), E. Billings gave an account of the discovery of remains of a white whale at Cornwall. Considerable parts of the skull were secured, including the lower jaws. Besides many vertebræ and some other parts, 8 teeth were saved, but the ear-bones were missing. The animal had been about 15 feet long. Whether it belonged to Delphinapterus leucas or D. vermontanus may be regarded as doubtful. Extracts from Billings’s description are to be found in Professor Perkins’s paper (Rep. State Geologist Vermont, 1907–8, pp. 81–82).

6. Williamstown, Glengarry County.—This place is about 10 miles northeast of Cornwall. In Professor Perkins’s paper just cited it is stated that Edward Ardley, assistant curator at Redpath Museum, McGill University, Montreal, had found here a few bones of a white whale, the hyoid, a few phalanges, and rib fragments. It is impossible from such limited materials to determine whether the animal was Delphinapterus vermontanus or D. leucas. From Mr. Ardley, through Mr. Arthur Willey, curator of Redpath Museum, the present writer has learned that these bones were dug up from a depth of 14 feet, in a well sunken in the Leda clay. Under the surface soil was a band of sandy clay containing shells of Saxicava and Mya. Beneath this was a stiff blue clay showing stratification and containing shells of Leda.

QUEBEC.

7. Montreal.—In 1863 (Rep. Geol. Surv. Canada, p. 919), W. E. Logan announced the finding of some bones of a whale at the Mile-End quarries, Montreal, on a slight ridge, “where are found stratified sand and gravel holding boulders and shells in the lower part.” In corresponding clays in a neighboring brickyard was found a pelvis of a seal, Phoca grœndlandica. In 1895 (Canad. Rec. Sci., vol. VI, p. 351), Dr. J. W. Dawson reported the discovery of a nearly complete skeleton of another white whale at Montreal. This was found in brick clay, near Papineau Road. The locality is said by Dawson to be about 100 feet above the St. Lawrence; the bones were in the clay at a depth of 22 feet. The clay itself was probably deposited at a depth of 50 to 80 fathoms. This is said by Dawson to correspond approximately with a well-marked shore-line at Montreal, found at a height of about 470 feet above the sea and with the old sea-beach at Smith’s Falls as related on page [17]. In 1916, Mr. Edward Ardley, assistant curator of Redpath Museum, reported (Canad. Rec. Sci., vol. IX, pp. 490–493) the discovery of a large part of the skeleton of a white whale, supposed to belong to Delphinapterus leucas, at Montreal East. The skeleton was buried in Leda clay about 15 feet above St. Lawrence River. It was 10.5 feet long. The cranium and lower jaw were secured, besides parts of the trunk and limbs.

8. Rivière du Loup, Temiscouata County.—In his work, “Canadian Ice Age,” 1894, on page 268, Dr. J. W. Dawson reported that bones of Beluga catodon (Delphinapterus leucas) had been found at the place mentioned. It is not probable that parts sufficient for making a definite determination were secured, nor did Dawson give any details regarding the geological conditions connected with the discovery. Doubtless the remains were found in marine deposits of one of the terraces.

9. Metis, Rimouski County.—In the work just cited (p. [269]), Dawson stated that in the summer of 1891 he secured a large jawbone of a whale which had been found in digging a cellar in the shelly marl of the lower terrace at Metis. He did not identify the species, but appears to imply that it belonged to either the “humpback” (Megaptera boöps) or to one of the finner whales (Balænoptera).

NEW BRUNSWICK.

10. Jaquet River, Restigouche County.—In 1874 (Trans. Nova Scotia Inst. Sci., vol. III, pp. 400–404), Dr. J. B. Gilpin gave an account of the discovery of some cetacean bones in a railroad cut at the place named, but did not identify the bones otherwise than as those of a small cetacean. In the same year (Amer. Jour. Sci., ser. 3, vol. VII, p. 597), in a short, unsigned communication, this discovery was mentioned and the whale was identified as Beluga vermontana. In volume VIII of the same journal (1874, p. 219), Dr. D. Honeyman described the deposit and gave a list of the shells found in it. Dawson (Canad. Ice Age, p. 268) refers the bones to Beluga catodon. The locality is a cut of the International Railway, on the north side of the Jaquet River, about 0.25 mile from the sea. Gilpin gives the elevation as 40 feet above the sea; the writer of the unsigned communication just mentioned gives it as 25 feet.

Professor G. H. Perkins (Rep. State Geologist Vermont, 1907–8, pp. 102–112) studied the bones described by Gilpin. They consisted of 18 vertebræ, some fragments of the skull, one of the ear-bones, a part of the lower jaw, some fragments of ribs, and some arm-bones. He identified the animal as belonging to the genus Monodon and probably M. monoceros, the existing narwhal.

11. Mace’s Bay, Charlotte County.—In 1879 (Geol. Survey of Canada, 1877–78, EE, p. 23), Mr. G. F. Matthew reported the discovery of a ramus of the lower jaw of a whale, possibly a species of Delphinapterus, at the mouth of the Popologan (or Pocologan) River. It is now in the Mechanics’ Institute at St. John. It had fallen from a bank of Leda clay. It probably belongs to the late Pleistocene.

VERMONT.

12. Charlotte, Chittenden County.—At this place were discovered considerable parts of a whale, described in 1850 (Amer. Jour. Sci., ser. 2, vol. IX, pp. 256–263) by Zadock Thompson, under the name Beluga vermontana. The animal has by many been regarded as identical with the white whale, Delphinapterus leucas, now appearing sometimes as far up as Montreal. A more extended description of it was given in 1853 (Hist. Vermont, Append., p. 15, figs. 1–13). This was reproduced in Edward Hitchcock’s Report on the Geology of Vermont, 1861, page 164, and was followed by remarks on the specimen by Edward Hitchcock jr. In the second volume of the work just cited (p. 938) Hager furnished a figure of the skeleton as mounted. In 1908 (Rep. State Geologist Vermont, 1907–8, pp. 76–112, plates X-XIX), Professor G. H. Perkins gave an extended description of the remains and reached the conclusion that D. vermontanus is distinct from D. leucas. Since Perkins’s article gives a full history of the discovery and the literature pertaining to the specimen, this account will be much abridged. The bones were found in making a cut for the Rutland and Burlington Railroad, at the town of Charlotte, about a mile east of the shore of Lake Champlain. The bones were 8 or 9 feet below the surface and “were very completely bedded in fine adhesive blue clay.” The locality is 60 feet above the mean level of the lake and 150 feet above the sea. The deposits were laid down in the marine waters which took possession of Lake Champlain and the St. Lawrence Valley when the Wisconsin glacial ice had withdrawn north of St. Lawrence River. The geological age of the animal is therefore late Pleistocene.

NORTH CAROLINA.

13. Below New Bern.—In 1842 (Amer. Jour. Sci., vol. XLIII, p. 143), Richard Harlan reported regarding the species of fossil vertebrates found 16 miles below New Bern. His list, which was long and consisted mainly of vernacular names, included “cetaceans.”

SOUTH CAROLINA.

14. Charleston.—In 1860 (Holmes’s Post-Pliocene Foss. South Carolina, p. 117, plate XXIV, fig. 9), Leidy described a cetacean tooth which he called Physeter antiquus. Later the specific name was changed to vetus. At the same time he figured a tooth (fig. 8) found in the Ashley River deposits. He further stated that teeth apparently of the same species had been taken from the Miocene formations of Virginia, but found no characters by which they could be distinguished from those of the recent sperm whale.

GEORGIA.

15. Brunswick.—In 1911 (Bull. No. 26, Geol. Surv. Georgia, p. 436), Gidley reported from here, among other vertebrates, some teeth which he regarded as those of Physeter vetus; but this may not be correct and they may not belong to the Pleistocene.

FLORIDA.

16. Daytona, Volusia County.—In 1916 (Florida Geol. Surv., vol. VIII, p. 105), Doctor Sellards stated that he had obtained from marl-pits worked at this place for road materials a proboscidean tusk and a rib of a whale, probably of the genus Balænoptera. At the same place had been found a tooth of Elephas columbi.

17. De Land, Volusia County.—At this place was obtained the dolphin skull which Sellards described as Globicephalus bæreckeii (Florida Geol. Surv., vol. VIII, p. 107, plate XIV). It was found embedded in sand at a depth of 10 feet. This sand overlies marls which are regarded as Pliocene or Miocene. Sellards believed that the sands belonged to the Pleistocene. It is not improbable that the marls pertain to the Pleistocene of the first glacial time.

FINDS OF PLEISTOCENE PINNIPEDIA IN EASTERN NORTH AMERICA.

(Map [2].)

GRINNELL LAND.

Dumbbell Harbor.—In 1877 (Ann. Mag. Nat. Hist., ser. 4, vol. XX, p. 488), Fielden published a paper on the Post-Tertiary beds of Grinnell Land and north Greenland. Fielden and De Rance reported on the same subject in 1878 (Quart. Jour. Geol. Soc. Lond., vol. XXXIV, p. 566). In beds having an elevation of 400 feet, in latitude 82° 30′, there were obtained meager remains of Phoca hispida and Ovibos moschatus. In latitude 82° 25′ were secured remains of Rangifer tarandus, Ovibos moschatus, and Phoca barbata. The invertebrate fauna was found to be identical with that existing there to-day. If the beds are of Pleistocene age, as the elevation appears to indicate, they may be referred to the Late Wisconsin.

NOVA SCOTIA.

1. Sable Island.—In the collection of the Philadelphia Academy there is a walrus skull which was sent to the Academy from Sable Island about 1871. According to Rhoads (Proc. Phila. Acad., 1898, p. 197), Leidy regarded this skull as that of a recent individual; but Rhoads states that “the specimen is of precisely the same nature in color, texture, and specific gravity as the larger fossil specimen which Leidy described and figured in the Philosophical Transactions and which came from the beach at Long Branch, New Jersey.” He thinks that it had been derived from an ancient raised sea-beach. This does not appear to be at all improbable.

NEW BRUNSWICK.

2. Fairville, Charlotte County.—In 1879 (Geol. Surv. Canada, Rep. for 1877–8, EE, p. 23), Dr. G. F. Matthew reported the discovery of a skeleton of Phoca grœnlandica near Fairville, at the mouth of St. John River, New Brunswick. The fore limbs and several vertebræ were missing. The skeleton was afterwards destroyed in a fire at St. John. The bones were found at a depth of about 25 feet, in the lower Leda clay.

QUEBEC.

3. Bic, Rimouski County.—In Le Naturaliste Canadien (vol. XXXVI, 1908, p. 51), the editor, V. A. Huard, in commenting on a letter written to him and announcing the capture of a walrus somewhere on the northern coast of the Gulf of St. Lawrence, recalled an article contributed in 1869 by the former editor, a priest named Provancher (Le Naturaliste Canad., vol. II, p. 19). This writer stated that some workmen employed in the construction of the International Railway had discovered at Bic, Rimouski County, Quebec, on the southern shore of the St. Lawrence, a complete skeleton of a walrus. This skeleton had a length of 13 feet. It was found at a depth of 14 feet, in a compact clay, and at a height of more than 100 feet above sea-level. The skeleton was deposited in the museum of the Rimouski Seminary, but was destroyed in a fire in 1881.

It is evident that when that animal died and was buried in the clay the land in that region stood at a level at least 100 feet lower than at present.

Through the late Mr. L. M. Lambe, of the Canada Geological Survey, the writer has received from Mr. W. A. Johnston, who made a special study of the Pleistocene, information regarding the age of the clays at Bic. He says that little can be said definitely regarding the age of the clays in which the walrus skeleton was found. Clays belonging to the Champlain submergence stand now at an elevation of 311 feet in that vicinity; and marine shells occur in clays, supposed to belong to the Champlain, at an altitude of 120 feet. There is a possibility that some of the clays in that region are earlier than the time of the Wisconsin. Mr. Johnston cites Guide Book No. 1, part I, pp. 77–78, of the Canada Survey, and Dawson’s Ice Age, 1893, pp. 186–195. The first article was written by J. W. Goldthwait. On page 921 of Logan’s Geology of Canada, 1863, it is stated that bones of whales and of the morse have been found partially embedded in the Leda clay in several places between Bic and Matanne, about 60 miles farther down the river.

4. Montreal, Quebec.—In 1863, Logan (Geol. Surv. Canada, p. 920) told of the discovery of a skeleton of Phoca grœndlandica near Montreal. The exact locality appears to be about 0.75 mile east of what was then known as the Mile-end quarries. These quarries were about 100 feet above sea-level, and the spot where the skeleton was found was about 40 feet lower down. At a nearby brickyard some bones of a young seal were discovered which belonged probably to the same species. One of the pelvic bones of a seal was found also at the Mile-end quarries. Dr. J. W. Dawson (“Canadian Ice Age,” 1844, p. 267) stated that the skeleton was found in the Leda clay; that it is in the collection of the Geological Survey, at Ottawa; and that detached bones are in the Peter Redpath Museum of McGill University at Montreal. The Leda clay, at least that of the upper portion of the St. Lawrence Valley, is now referred to the Champlain epoch, a time when the sea had invaded this valley and even Lake Ontario.

5. Tétreauville, Ottawa County.—In 1897 H. M. Ami (Ottawa Naturalist, vol. XI, p. 24) announced that he and Ruggles Wright had found some bones which were probably those of a young harbor seal, Phoca vitulina. They were collected in 1888, in a sandy layer about 30 feet below the surface, on a hillside, at Wright’s brick clay pits, on Aylmer Road, Tétreauville, Quebec. This place is about 5 miles west of Hull, and within 10 miles of Ottawa. These bones are in the Victoria Museum at Ottawa. Besides the left half of the lower jaw with teeth, there are both ear-bones, one exoccipital, the greater portion of the backbone, scapula, part of the pelvis, and some of the larger limb-bones. This species is abundant in the Gulf of St. Lawrence, and also ascends the larger rivers to a great distance. Doubtless great numbers inhabited the inland sea which, during Champlain times, is believed to have occupied the valley of the St. Lawrence, Lake Ontario, and the valley of the Ottawa River nearly as far up as the city of Ottawa.

ONTARIO.

6. Ottawa.—Remains believed to belong to Phoca grœnlandica have been found near Ottawa, Ontario. In 1856 (Proc. Phila. Acad. Nat. Sci., vol. VIII, p. 90, plate III), Doctor Leidy described and figured the hinder limbs of a young aquatic animal which he regarded as a seal, but did not more exactly identify. He expressed the opinion that its descendants were yet sporting in the sea-borders of Canada. This specimen was found in Gloucester Township, Carleton County, about 9 miles east of Ottawa. The locality is on Green’s Creek, a tributary of the Ottawa River, the bank of the creek being about 30 feet high and composed of clay. This is regarded as being of Champlain age, the close of the Wisconsin stage. Out of this clay were washed numerous nodules of hardened clay, many of which contained organic remains, such as marine shells and fishes. Among the latter are two species, the capelin (Mallotus villosus) and the lump-sucker (Cyclopterus lumpus).

Later, at the same locality, a lower jawbone of a young seal was found, which was identified as the harp seal; and it was even thought that it might have belonged with the hinder limbs figured by Leidy. A figure of this jaw, with some of the teeth, was published by Dawson in his “Canadian Ice Age.”

MAINE.

7. Addison Point, Washington County.—From the curator of the Portland Society of Natural History, Arthur H. Norton, the information is received that some portions of the skeleton of a walrus, several ribs, parts of two limbs, and a phalanx of a digit, had been found at Reef Point, near Addison Point, Maine. These remains are now in the collection of the society just named. They had been collected in 1881 by C. H. Boyd, who published an account of them (Proc. U. S. Nat. Mus., vol. IV, p. 234). They had washed out of the bank on the eastern side of Pleasant River, about 3 miles below Addison. They had been buried in a stiff blue clay, about 2 feet above high-water. Above them there was 6 feet of the clay, and above this gravel and soil. Mr. Boyd stated that he had seen a tusk, with a part of the socket, which had been washed out at the same place.

8. Andrews Island, Knox County.—The American Museum Journal for 1912 (vol. XII, pp. 269–270) contains an article which calls attention to a walrus skull preserved in the American Museum of Natural History in New York. It is reported as having been found by Sidney Norton, in December 1912, in 50 fathoms of water, near Andrews Island, off Owl’s Head, Penobscot Bay. One of the tusks is complete, the other is gone; also the occiput and zygomatic arches are missing. The bone is said to be quite well petrified, which shows that the skull is not a recent one.

9. Gardiner, Kennebec County.—In 1845 Charles Lyell visited (“Second Visit to the United States,” vol. I, p. 44) Gardiner, Maine, and examined a collection of fossil shells and crustacea which had been made by Mrs. Frederic Allen from the glacial deposits of that vicinity. He recognized the tooth of a walrus, which he stated was similar to the one procured by him on Martha’s Vineyard. This tooth is said by Packard (Mem. Bost. Soc. Nat. Hist., vol. I, 1867, p. 246) to have been a tusk; and he was informed that it had been taken by Lyell to London and had been identified by Professor Richard Owen. Inasmuch as Owen regarded the specimen found on Martha’s Vineyard as a species distinct from the one now living on the Atlantic coast, it is to be supposed that the Gardiner specimen also was thought to be different from the latter. Packard, in another communication (Amer. Naturalist, vol. I, 1868, p. 268), states that the tooth of the walrus and some teeth of a supposed bison were discovered in the clay-beds at Gardiner by Lyell, or at least during his visit, but it is evident that they had been collected before his arrival.

In his discussion of the supposed bison teeth found in clay at Gardiner, Dr. J. A. Allen (The American Bisons, 1876, pp. 89, 91) gives us some information about the fate of Mrs. Frederic Allen’s collection. At her death it passed into the possession of her daughter, by whom the greater part of it was presented to Bowdoin College, Brunswick, Maine. Professor Manton Copeland, of this college, informs the writer that the walrus tusk is in their collection and bears the number FM20. It is badly shattered. The length is about 75 mm.

The important matter concerning the remains of the walrus found at Gardiner is to determine when the animal lived there. It is to be assumed that the tusk had been buried in the Pleistocene clay at that locality. This appears to belong to the closing period of the Wisconsin stage, but there has been some dispute about its age.

Packard (Mem. Bost. Soc. Nat. Hist., vol. I, pp. 245–246) gives a list of the species which had been found in the clay at Gardiner. These are nearly all invertebrates and indicate a climate somewhat colder than that now existing there. Whether the time when the walrus lived at Gardiner was before or after the culmination of the Wisconsin ice period, it was so long ago that those deposits of clay, made in sea-water of considerable depth, have since been lifted above sea-level to a height of perhaps 200 feet.

10. Portland, Cumberland County.—In the American Naturalist, volume XII, 1878, page 633, it is recorded that the larger part of the skeleton of a walrus, including the skull, with tusks over 5 inches long, had lately been found in the Quaternary clays at Portland. It had been discovered by workmen excavating for the foundation of the transfer station of the Boston and Maine Railroad. The remains were partially embedded in a layer of blue clay a foot thick, itself overlain by 2 feet 2 inches of a lighter clay. The latter contained casts and shells of 11 species of mollusks. J. A. Allen, in his work already quoted, states that the skeleton was found at a depth of 7 feet. It was placed in the museum of the Portland Society of Natural History, and is still there, as reported by the curator, Arthur H. Norton.

Mr. Norton has sent the writer an extract from the report of the committee which investigated this discovery. The bed of blue clay in which the greater part of the skeleton was buried contained the following species of mollusks: Mya arenaria, Macoma sabulosa (calcarea), Mytilus edulis, Cardium (Serripes) grœndlandicum, Saxicava distorta, Nucula antiqua, Leda tenuisulcata, L. truncata (Yoldia glacialis), Natica clausa, N. pusilla, and Astarte striata. The lighter-colored clay above the blue clay was more sandy and adhered strongly to the bones. This clay contained Mya arenaria, Mytilus edulis, Serripes grœndlandicus, Astarte striata, Macoma calcarea, Nucula antiqua, Natica, and Balanus.

Above the lighter-colored clay just mentioned was a foot of a clay which contained wood and roots, the unused portion of the brick clay that once existed there, but which had been removed for the manufacture of bricks.

Inasmuch as the clay overlying the bed in which the remains were found contains marine shells, it is certain that since their deposition the land has been considerably elevated.

George N. Stone (Monogr. XXXIV, U. S. Geol. Surv., pp. 286–291) has discussed the age of the glacial deposits at Portland. Professor M. L. Fuller has written to the author that on the Maine coast the chief clay is known as the Leda and is found at Portland and Gardiner, and that it probably antedates the Wisconsin. This is not to be correlated with the Leda clay of the St. Lawrence Valley. It corresponds rather to Clapp’s “high-level clays” (Bull. Geol. Soc. Amer., vol. XVIII, p. 505, seq.).

NEW HAMPSHIRE.

11. Jeffries Reef, off Portsmouth.—The specimen from this place consists of the greater part of the left side of the skull of a large individual. The occipital and the exoccipitals are missing. The bone and especially the tusk have suffered some decay. The fragment is labeled as having been dredged from a depth of 50 to 75 fathoms on Eastern Jeffries Reef. The bottom was hard. Jeffries Reef lies 5 or more miles off the southernmost part of the Maine coast and extends from the Isle of Shoals to Boon Island. The skull belonged to an old individual. The length from the rear of the mastoid process to the front of the premaxilla is 360 mm. The exserted part of the tusk measures 225 mm. in length. At its base the diameters are 65 mm. and 42 mm. There are 4 large grinding teeth. There is no reason for supposing that the species represented is not O. rosmarus.

MASSACHUSETTS.

12. Gay Head, Martha’s Vineyard.—In his “Travels in North America,” volume I, 1845, page 257, plate V, figure 1, Lyell announced the finding of a part of a skull of a walrus at Gay Head. This he had purchased from a fisherman who lived there and who said it had fallen out of a conglomerate found at that place and which contains bones of cetaceans. The skull retained but a small portion of its animal matter. Richard Owen, to whom the skull was shown, regarded it as belonging to a species distinct from O. rosmarus. The upper jaw contained the base of one tusk, the socket for the other, and 3 molar teeth on each side. The reduced number of molars furnishes no distinctive character, for existing individuals sometimes present this number. The base of the tusk has its transverse diameter greater than usual relatively to the fore-and-aft diameter. According to Lyell’s illustration of the specimen, the greater diameter was 70 mm., the shorter 53 mm. The writer has seen no tusk of O. rosmarus as thick as this; but the thickness is variable and may possibly attain to two-thirds of the greater diameter.

Inasmuch as the Tertiary deposits at Gay Head, rising above the sea to a height of about 150 feet, are capped by a sheet of glacial drift and clays, it is probable that the skull in question had fallen from some of these drift deposits. According to Professor J. B. Woodworth (17th Ann. Rept. U. S. Geol. Surv., pt. I, p. 982), there are at Gay Head deposits of drift which represent some of the older glacial stages as well as the last one, the Wisconsin. It is possible, therefore, that this walrus lived there as far back as the middle of the glacial epoch or even earlier. For additional information on the geology of that island consult Woodworth’s paper, in which the literature is cited; also the important paper by N. S. Shaler (7th Ann. Rept. U. S. Geol. Surv., 1888, pp. 303–363.)

The hooded seal, Cystophora cristata, has probably been found fossil at Gay Head. The only reason for this supposition is found in a statement made by Charles Lyell (Proc. Geol. Soc. London, vol. IV, p. 32; Amer. Jour. Sci., vol. XLVI, 1844, p. 319). He says that with other remains on Martha’s Vineyard he found a tooth having the crown fractured. Lyell submitted the tooth to Richard Owen, who pronounced it to be that of a seal which seemed to be nearly allied to the modern Cystophora proboscidea (C. cristata). It seems quite probable that this species lived there at the time when the walrus haunted the region. It is of course possible that the remains reported belonged to an animal that lived in that region as far back as the Miocene. The tooth was not described or figured.

NEW JERSEY.

13. Long Branch.—Portions of several walrus skulls have been found on the beach at Long Branch. Two of these were described and figured by Leidy in 1867 (Trans. Amer. Philos. Soc., vol. XI, p. 83, plate IV, figs. 1, 2, plate V, fig. 1). One skull, lacking the lower jaw, some of the right hinder part of the cranium, and the exserted portion of one tusk, was discovered in 1853. The other specimen, discovered about 1856, furnished the front of the skull as far back as the middle of the palate. Both belonged to old individuals. Leidy concluded that the animals which had possessed these skulls belonged to the existing species Odobenus rosmarus. He surmised that they had been floated to the New Jersey coast on fields of ice or perhaps had lived there during the Glacial period. The skull which was found in 1853 is now in the collection of the Philadelphia Academy; the other is in the collection of the New Jersey Geological Survey. Recently, Mr. Samuel N. Rhoads has studied these skulls. He had also for examination the skull from Sable Island, which has been mentioned. He concluded that these skulls belonged to a species distinct from O. rosmarus and which might bear DeKay’s name, O. virginianus.

It does not appear to the present writer that Rhoads has successfully maintained his proposition. He did not have at hand a sufficient number of skulls of the existing Atlantic walrus to present all the variations that occur in that species. Of course, the number of fossil specimens was very limited. In discussing Rhoads’s conclusion, it will be of advantage to consider a part of a skull which belongs to the Marsh collection in Yale University. This specimen consists of the anterior half of the skull, without the tusks and without the other teeth. It was found at Kitty Hawk, at the mouth of Albemarle Sound, just north of latitude 36°. It is thoroughly fossilized; and, having been found so far south, it may be safely regarded as having belonged to the species which inhabited the New Jersey coast during the Pleistocene.

For purposes of comparison, such measurements are here given as can be obtained from the skull; likewise the corresponding measurements of a specimen from Sable Island, No. 199528 of the U. S. National Museum, and of another, No. 22014 of the National Museum, brought from Ungava Bay. Unfortunately, the basilar length of the fossil can not be determined, nor the width of the mastoids.

The nasal bones of the fossil are so thoroughly consolidated with each other and with the adjoining bones that their dimensions can not be determined. There is no reason, however, for supposing that the length was greater than 70 mm.

The grinding teeth of the fossil do not show the larger size that we might expect from Rhoads’s determinations and from comparison with Leidy’s illustrations. The second socket was almost exactly the diameter of the same socket in the Sable Island specimen measured. The third socket is larger than that of the skull from Sable Island. The sockets for the first molars are very small and shallow; the socket for the left incisor is still smaller, while that for the right incisor is wholly effaced. The diameter of the socket for the second molar is much shorter than that of the corresponding socket in the Ungava Bay specimen. In the latter, the left incisor is present and large, but the other is missing and the socket is nearly filled up. It is evident that the teeth are extremely variable in both size and the number present.

Rhoads has found that the incisive foramina of the fossil skulls in his hands are placed high above the alveolar borders. In the North Carolina specimen this height is 32 mm.; in the Sable Island specimen in the U. S. National Museum, 30 mm.; in the Ungava Bay specimen, about 22 mm. Nor does the distance between the sockets for the incisors in the fossil from North Carolina agree with that dimension in the two specimens from Long Branch.

Despite the differences shown in the measurements in the table given above, the writer must conclude that there are not as yet sufficient reasons for regarding the Pleistocene walrus of the Atlantic coast as specifically different from the existing form.

Dr. Albert Reid Ledoux, mining engineer, of New York City, when a young man bathing at low tide at Long Branch, found a skull of a walrus. This was given to Professor John S. Newberry and is now probably at either Columbia University or the American Museum of Natural History. At the same time and at the same spot was a heel-bone of Megatherium, now in the American Museum (p. [31]). It is very improbable that these two animals lived there at the same time.

According to recent publications of the Geological Survey of New Jersey (Salisbury, Report for 1897, p. 19, pl. I; Lewis and Kümmel, Bull. No. 14, p. 120, with Geologic Map of New Jersey, 1910–1912), Long Branch is situated on the Cape May formation. This is regarded by the geologists just quoted as corresponding in age, in great part at least, to the Wisconsin stage. When this deposit was laid down, the New Jersey coast was depressed from 35 to 50 feet below its present level. It seems very probable that at that time the walrus was living there and that the skulls found have been washed out of this deposit by the waves during storms. Nevertheless, the finding of Megatherium at Long Branch shows that there are deposits present which belong probably to early Pleistocene.

Dr. H. B. Kümmel, State Geologist of New Jersey, has informed the writer that a strip 0.25 to 0.75 mile back from the ocean in the region about Long Branch probably belongs to the Recent time. He states that one would be safe in concluding that the skulls of the walrus were found in deposits not older than the Cape May and that they may have occurred in more recent beds. Against the view that the walruses found along this coast lived there during the Recent period is their well-fossilized condition.

14. Ocean Grove, Monmouth County.—In 1910, after a storm, a part of a skull of a walrus was found on the beach at Ocean Grove, New Jersey. This is still in the possession of the finder, Mr. W. S. Hidden, who furnished the writer with photographs of the specimen. It consists of the front of the skull extending back to the bases of the zygomatic arches, and containing portions of both tusks and most of the teeth. There is no likelihood that this specimen belonged to any other species than Odobenus rosmarus, and it was probably washed out of the same deposits as those which furnished the specimen found at Long Branch.

VIRGINIA.

15. Accomac County.—In the Annals of the Lyceum of Natural History of New York, volume II, 1828, page 271, Messrs. Mitchill, Smith, and Cooper made a report on a fossil walrus skull found along the Virginia coast somewhere in Accomac County. Only the anterior half of the skull was secured. According to this report, portions of the tusks were preserved, but were much mutilated. There were present also 4 of the grinding teeth. The skull was described as being remarkably hard and heavy and the tusks were almost agatized. The sutures of the skull had mostly closed up; hence the animal was evidently an old one. The specimen bore the marks of having been in salt water, and was said to have been found on the beach.

This is the specimen which DeKay, in 1842 (Zool. of N. Y., pt. I, p. 56, plate XIX, fig. 1), made the type of his Trichechus virginianus. Newberry, in 1873 (Proc. Lyc. Nat. Hist. New York, p. 71), identified the specimen as belonging to the existing Atlantic species. Cope (Proc. Amer. Philos. Soc., vol. XIV, 1874, p. 17) does not mention the presence of tusks. He supposed that there was, at that part of the coast, glacial drift, out of which the skull had been washed. There are, however, no such deposits in that region. This specimen was placed in the collection of the Lyceum of Natural History of New York, but according to Rhoads, was afterward destroyed in a fire.

On examination of G. B. Shattuck’s work on the Pleistocene of Maryland (Maryland Geol. Surv., Pliocene and Pleistocene volume, p. 95, plate I), it seems that the coast of Virginia in Accomac County is occupied by the Talbot formation. This, according to his theory, corresponds, at least the part nearest the coast, with the Cape May formation of New Jersey. Hence we might conclude that the walrus skull in question had become buried, probably during the Wisconsin glacial stage. The present writer regards the principal part of the Talbot terrace as being much older.

Messrs. W. B. Clark and B. L. Miller (Virginia Geol. Surv. Bull., No. IV, p. 187) recognize the presence of the Talbot formation in Accomac County, where it seems to reach a thickness of 100 feet; but the authors add that part of this may belong to earlier Pleistocene formations.

NORTH CAROLINA.

16. Kitty Hawk, Currituck County.—In the Marsh collection of fossils belonging to Yale University is a part of a skull found somewhere near Kitty Hawk. No particulars regarding the exact place of discovery accompany the specimen. It has already been described on page [27]; and, while there are some differences between it and the recent skulls used for comparison, it is not believed that a distinct species is indicated.

According to L. W. Stephenson’s map of the Coastal Plain of North Carolina (North Carolina Geol. and Econom. Surv., vol. III, plate XIII), the coast at Kitty Hawk and for about 50 miles back of this is occupied by the Pamlico formation. This corresponds to the upper part of the Talbot of Maryland, and it, or part of it, may have been deposited at the close of the Pleistocene. So far as the present writer knows, there is nothing to show the character of the climate then prevailing. As this Pamlico nowhere rises more than 25 feet above sea-level, and as the thickness is usually only from 15 to 20 feet, it is possible that the walrus skull found at Kitty Hawk had been unearthed by the waves from the Chowan formation or some still earlier deposit.

SOUTH CAROLINA.

17. Charleston.—In 1876 Leidy announced (Proc. Phila. Acad., 1876, p. 80) that a complete tusk of a walrus had been found in the Ashley River, near Charleston. This tusk Leidy described and figured in 1877 (Jour. Phila. Acad., vol. VIII, fig. 6). It had evidently been dredged from the river in collecting phosphate rock, as have been most of the fossils of that region. The tusk was 13 inches long. Near the base it measured 3.62 inches and transversely 1.75 inches. Leidy especially noticed the shortness of the tusk as compared with the diameter, but concluded that the tusk might, during the life of the individual, have been broken off and worn obliquely at the end.

In the collection of the Charleston Museum are some fragments of tusks of a species of walrus, probably O. rosmarus. One of these, No. 1028, furnishes 184 mm. of the distal end. The width at the fracture is 60 mm., the thickness 29 mm. The distal end is worn off somewhat obliquely, but not so much as in the tusk figured by Leidy; also, the tusk appears to have been less curved than the one which he described. The original length can not be determined.

Another fragment, No. 1029, was given to the Charleston Museum by Major E. Willis and was no doubt found in the region about Charleston. This gentleman has sent a fossil horse-tooth and a part of a sirenian to the U. S. National Museum from Wando River. The fragment is short, but belonged to a large tusk, its long diameter being 81 mm., the shorter one 51 mm. It was therefore a larger tusk and one whose thickness was relatively greater than that of the imperfect specimen found at Long Branch and figured by Leidy.

Mr. Earle Sloan’s collection at the Charleston Museum has two other fragments of tusks. One, No. 13497, is 113 mm. long, 60 mm. wide, and 25 mm. thick; the other, No. 13296, is 140 mm. long, 60 mm. wide, and 31 mm. thick.

Considering that all of the remains of the walrus found about Charleston have been picked out of great quantities of phosphate rock collected for commercial purposes, and that no records of the exact locality where obtained have been kept, it is impossible to determine their exact geological age. It is to be supposed that this animal inhabited the region about Charleston at the time it frequented the coasts of North Carolina and New Jersey. This appears to have been during the Wisconsin stage; but it is possible that the walrus extended its range far southward during more than one of the glacial stages. All of the specimens appear to be thoroughly fossilized.

FINDS OF XENARTHRA IN EASTERN NORTH AMERICA.

NEW JERSEY.

(Map [3].)

1. Long Branch, Monmouth County.—In the American Museum of Natural History, New York, there is a large heel-bone which was found at the place named and identified as having belonged to a species of Megatherium, most probably to M. mirabile. It was presented by Dr. A. R. Ledoux, of New York, who wrote that he found it about 40 years ago while bathing at Long Branch. With this bone were found a skull of a walrus and a tooth of a mastodon. The heel-bone is somewhat more than 15 inches long. It was incrusted with barnacles and small oyster shells.

While one can not at present be certain that this animal did not live up to a late stage of the Pleistocene, it is improbable that it did so. It is also quite improbable that the megatherium and walrus lived at Long Branch at the same time. It is more likely that the megatherium had its existence there at the time when horses lived in the same region and when the Port Kennedy fauna existed; that is, at some time during the early Pleistocene about the Aftonian stage.

PENNSYLVANIA.

(Map [3].)

1. Port Kennedy, Montgomery County.—From the noted bone cave at Port Kennedy a number of species of Megalonyx have been described. The presence of this genus was first announced by Wheatley (Amer. Jour. Sci., ser. 3, vol. I, p. 384). Cope, in 1899 (Jour. Acad. Nat. Sci., Phila., vol. XI, pp. 211–219), admitted the occurrence of 4 species, Megalonyx wheatleyi, M. loxodon, M. tortulus, and M. scalper. It must be left to future investigations to determine the status of these species. M. jeffersonii was not recognized by Cope in the materials found in the cave. Of M. loxodon, only a single upper canine molar was found. Of M. wheatleyi, numerous specimens were secured, including considerable parts of crushed and decayed skulls. M. tortulus was represented by a considerable number of teeth; M. scalper by only a single “canine-molar.” On page [312] will be found a list of the species of vertebrates associated with these sloths. Of Mylodon, Wheatley (op. cit., p. 384) had a single ungual phalanx which Cope (op. cit., p. 210) thought belonged probably to M. harlani.

2. Frankstown, Blair County.—Remains of an undetermined species of Megalonyx have been reported from a bone cave at this place by Dr. W. J. Holland (Ann. Carnegie Mus., vol. IV, 1908, p. 231). The associated species are listed on pages [321]–[322].

OHIO.

(Map [3].)

1. North Fairfield, Huron County.—In the Norwalk, Huron County, Museum there are various bones of Megalonyx jeffersonii which were obtained about 7 miles from North Fairfield. The writer learned of the discovery of this skeleton from Mr. Roe Niver, a student of the University of Illinois. Unfortunately Mr. Niver died before the writer could obtain all the desired information. A part of the skeleton was in his possession and is probably in the possession of his family, but the writer has been unable to secure any information from them. The bones were found at a depth of a few feet in a hackberry swamp and were considerably scattered. In the search for these the bones which form the type of Bison sylvestris Hay were found. The locality is within the area of the Wisconsin drift-sheet and evidently the animal lived there after the ice had retired from the region.

2. Millersburg, Holmes County.—In the University of Ohio there is a mounted specimen of Megalonyx jeffersonii containing a considerable part of the skeleton; the missing portions are replaced artificially. The remains were found in the eastern part of Holmes County just north of the terminal moraine of the Wisconsin drift-sheet. This moraine had led to the formation of a marsh, and in this the animal ended his life. The place was said by Orton to be 6 miles east and a mile north of Millersburg. The skeleton lay on shell marl beneath 6 feet of peat. The remains have been described by Claypole (Amer. Geologist, vol. VII, 1891, pp. 122–132, 149–153) and by Hay (Geol. Surv. Indiana, vol. XXXVI, 1913, p. 558; Geol. Surv. Iowa, vol. XXIII, 1914, p. 110).

INDIANA.

(Map [3].)

The only member of the order of Xenarthra that has yet been found in this State is Megalonyx jeffersonii, and this in only one place, viz, Evansville.

1. Evansville, Vanderburg County.—In 1854 (Proc. Acad. Nat. Sci. Phila., vol. VII, pp. 199–200), Leidy described a collection of vertebrate fossils secured by Mr. Francis A. Lincke from the banks of the Ohio River, near the mouth of Pigeon Creek, a short distance below Evansville. At that time and locality bones were usually found sticking out of the bank when the water in the river was low. The bones sent to Leidy were thoroughly impregnated with oxide of iron, which served as a cement to adhering pebbles, sand, and fragments of Unios and shells of other fresh-water mollusks. The remains of the megalonyx consisted of parts of two tibiæ of young individuals, an atlas, a fragment of a heel-bone, a metacarpal and a metatarsal bone, and a claw phalanx. With these were discovered a fragment of a cervical vertebra of a species of bison, various bones of the Virginia deer, a vertebra of a horse, probably Equus complicatus, a tooth of the tapir Tapirus haysii, and a part of the upper jaw of the wolf now known as Ænocyon dirus, but at that time called by Leidy Canis primævus.

The principal interest in these remains is to determine at what time during the Pleistocene the megalonyx lived. Some indications may be obtained from a study of its companions. From a part of a cervical vertebra Leidy could not name the bison, but it belonged probably to one of the extinct species. The deer is yet living, but appears to have existed through most of the Pleistocene. The species of horse represented is extinct, and there is no evidence that it lived after the Wisconsin glacial stage. Its latest representatives probably lived during the Sangamon stage. No tapir is known to have lived after the Wisconsin stage. The wolf, Ænocyon dirus, is believed to be represented in the numerous individuals found in the asphalt beds of Los Angeles, California, probably equivalent in age to the Aftonian.

Mr. Arthur C. Veatch (Jour. Geology, vol. VI, pp. 257–272) has given an account of changes which have occurred along the Ohio River in Spencer County, Indiana, about 25 miles above Evansville, since late Pliocene times. According to his investigations, the valley of the river was deeply excavated into the Carboniferous rocks during the Ozarkian uplift. Since that time, during the Pleistocene epoch, that great valley has been, to a large extent, filled up by alluvial deposits. While the greater part of these deposits were laid down during glacial stages, it is not improbable that some were made during the Aftonian stage and that a part of these yet exist along the borders of the river. It is still more probable that Sangamon beds yet exist there and that the bones Leidy described were found here.

Many bones of the megalonyx were described by Leidy (Smithson. Contrib. Knowl., vol. VII, article V) from a locality 5 or 6 miles below Henderson, Kentucky, not much more than 10 miles in a straight line from the mouth of Pigeon Creek. The bone-bed was said by Dr. D. D. Owen (op. cit., p. 7) to be about 5 feet above ordinary low-water. In the same bed Owen found abundant remains of the deer. He seemed to regard this bone-bed as a continuation of that existing at Pigeon Creek.

Megalonyx has been found at Bigbone Lick, between Cincinnati and Louisville, associated with Equus complicatus, two species of extinct bisons, and the Virginia deer, in deposits overlying Illinoian drift and hence belonging, in part at least, to the Sangamon. These deposits are, however, at a higher level, being now submerged only at times of very high-water in the Ohio River. If these and the Pigeon Creek beds are of the same age, we may suppose that the animals entombed at the latter place were buried low down in the deep valley along the river banks, while those at Bigbone became covered up around salt springs at a higher level.

ILLINOIS.

(Map [3].)

1. Urbana, Champaign County.—In the fall of 1909 a claw phalange of Megalonyx jeffersonii was found near Urbana by Mr. Lindley, of Urbana. An excavation was being made at the eastern end of Crystal Lake, and the tooth, as reported to the writer by Professor C. C. Adams, was discovered in a blue clay. The writer has seen the tooth. The extreme length in a straight line had been close to 145 mm. The greatest thickness was 42 mm. This has been figured by the writer (Iowa. Geol. Surv., vol XXIII, plate III, figs. 5, 6, text-figs. 28–29).

Inasmuch as all this region is covered by Wisconsin drift and this tooth was found in a deposit lying on the top of this drift, there can be no reason for denying that this species lived after, probably long after, the withdrawal of the Wisconsin ice. Two occurrences of the same species in Ohio confirm the conclusion.

2. Alton, Madison County.—The U. S. National Museum contains a fragment of a molar of apparently Megalonyx jeffersonii, from a collection made long ago by William McAdams, at Alton, Illinois. It has on it McAdams’s number 21. This collection, which was long in the hands of Professor O. C. Marsh, as vertebrate palæontologist of the U. S. Geological Survey, is said to have been made in the loess at Alton. Most of the teeth, with occasional bones, are inclosed in nodules of extremely fine sand and carbonate of calcium so hard that the teeth can not be removed without injury. They have been, however, partly exposed by weathering. The nodules which contained the fossils were found between the loess and the underlying Illinoian drift.

The fragment of a megalonyx tooth has the diameters respectively 16 mm. and 24 mm. It is thinner fore-and-aft than other specimens, but this may be an individual variation.

It is believed that this loess belongs to the Sangamon interglacial stage. The geology of the locality and the species found there are discussed on page [339]. Also, the fossils were described by the writer in 1920 (Proc. U. S. Nat. Mus., vol. LVIII, pp. 109–117). The presence of this sloth-like beast appears to indicate that the climate was at that time mild.

3. Galena, Jo Daviess County.—In 1870 (Proc. Acad. Nat. Sci. Phila., 1870, p. 13), Dr. Leidy brought to the notice of the Academy the fossil remains of two species of much interest. These had been presented to the Academy by Henry Green, of Elizabeth, Jo Daviess County, and were reported as having been found in a narrow crevice of the lead-bearing rocks in the vicinity of Galena, at a depth of 130 feet. One fossil was a metacarpal bone of Megalonyx jeffersonii, the other was identified as a last lower molar of Bison antiquus. Leidy mentioned three other species, Platygonus compressus, Procyon priscus, and Anomodon snyderi as having been found about Galena in similar situations. The geological age of the Vertebrata found in the lead crevices about Galena has not been well determined, but the present writer has regarded them as being probably of late Wisconsin time. The Bison tooth may have been that of the yet existing species. However, the possibility is that these fossils are pre-Wisconsin.

VIRGINIA.

(Map [3].)

1. Saltville, Smyth County.—Mr. O. A. Peterson, in 1917 (Ann. Carnegie Mus., vol. XI, p. 472, figs. 4, 5), reported the discovery of the symphyseal portion of the lower jaw of Megalonyx at Saltville. It was referred with some doubt to M. dissimilis Leidy. Further mention of the specimen will be made on page [352].

2. Ivanhoe, Wythe County.—On a page devoted to the consideration of a considerable number of species found by Cope near Ivanhoe, in Wythe County, mention will be made of Megalonyx jeffersonii. Only fragments of teeth were secured by Cope.

WEST VIRGINIA.

(Map [3].)

1. Green Brier County.—In a cave situated somewhere in this county were found the bones described in 1799 by President Thomas Jefferson (Trans. Amer. Philos. Soc., vol. IV, pp. 246–260) under the name Megalonyx. Colonel John Stewart became interested and saved some of the bones from being carried away by curious inhabitants of the region.

The bones, a distal end of a femur, a complete radius, a complete ulna, three claws, and some other foot-bones were secured and presented to the American Philosophical Society of Philadelphia, from which they passed into the possession of the Academy of Natural Sciences, where they are still preserved. Some of these were described by Dr. Caspar Wistar (Trans. Amer. Philos. Soc., vol. IV, 1899, p. 526, plates I, II).

Inasmuch as this species may have existed during a large part of the Pleistocene and certainly after the passing of the Wisconsin epoch, and inasmuch as no other species were found associated with the megalonyx bones, it is impossible to say to what part of the Pleistocene that particular animal is to be assigned.

SOUTH CAROLINA.

(Map [3].)

1. Beaufort, Beaufort County.—In the Charleston Museum the writer has seen a left lower canine tooth of Megalonyx jeffersonii. The fore-and-aft diameter is 34 mm., the transverse 18 mm. It is recorded as found in dredging in Coosaw River. Tuomey (Rep. Geol. South Carolina, 1848, p. 203) found fragments of bones, probably belonging to Megatherium, on Eddings Island, about 10 miles south of Beaufort.

2. Charleston, Charleston County.—In 1855, Doctor Leidy (Smithson. Contrib. Knowl., vol. VII, p. 55) stated that Professor F. S. Holmes, of Charleston, had loaned him fragments of two very small teeth of Megatherium found on the shores of Ashley River. These were figured by Leidy in 1860 (Holmes, Post-Pl. Foss. South Carolina, p. 111, plate XX, figs. 8, 8a). In a collection belonging to Rev. Robert Wilson, in Charleston, the writer has seen a tooth of Megatherium found by the Charleston Mining Company in Ashley River. G. E. Manigault (Proc. Elliott Soc. Nat. Hist., 1886, p. 91) reported the finding of a claw phalanx of Megalonyx at Cainhoy, 12 miles from Charleston, on Wando River.

In the Charleston Museum is a part of the right side of the upper jaw of Megatherium, with the second and third teeth and parts of the sockets of the first and fourth. It is recorded as having been found in the Bolton phosphate mine on or in Stono River. There is in the same museum a fragment of the left side of the lower jaw of the same animal. This jaw contains the second and third molars and parts of the socket of the first and fourth. It is recorded as having been found in the Kiawah phosphate mine, Cooper River.

The Charleston Museum contains considerable parts of the skeleton of a megatherium of which no record has been preserved. In Holmes’s “Post-Pleiocene Fossils of South Carolina,” page 111, plate XX, figures 7 to 7b, Leidy mentioned briefly and figured two small fragments of lower teeth of Mylodon harlani, which had been obtained from the Pleistocene beds of Ashley River. The tooth figured was originally described as Eubradys antiquus. Figures of it are found also in the seventh volume of the Smithsonian Contributions to Knowledge, plate XVI, figures 21a to 21c.

The Pleistocene geology of South Carolina is discussed on pp. [361] to [368].

GEORGIA.

(Map [3].)

1. Brunswick, Glynn County.—In 1842 (Proc. Acad. Nat. Sci. Phila., vol. I, p. 189), Richard Harlan gave to the Academy of Natural Sciences a number of bones which had been collected in the Brunswick Canal by Mr. J. H. Couper and sent to the Academy. Among these was a number of bones of Megatherium. A part of a lower jaw contained 4 teeth. A list of the bones is presented by Couper on page 44 of William B. Hodgson’s memoir on Megatherium published in 1846. There were, besides the part of a mandible, parts of 2 maxillæ without teeth, parts of 6 or 7 femora, a part of an ilium, several dorsal vertebræ, and several teeth. Lyell (Second Visit, ed. 2, 1850, vol. I, p. 347) stated that a part of a skeleton of a Megatherium, dug out in cutting the canal, was so near the surface that it was penetrated by the roots of a pine tree. Most of this material was sent to the Academy of Natural Sciences at Philadelphia (Leidy, Smithson. Contrib. Knowl., vol. VII, art. 5, p. 54).

The accompanying fossils will be named on page [370].

2. Skidaway Island, near Savannah, Chatham County.—The earliest announcement of the discovery of Megatherium in North America was made by Dr. Samuel L. Mitchill in 1824 (Ann. Lyc. Nat. Hist. N. Y., vol. I, pp. 58–61, plate VI). The announcement was based on a number of teeth which had been sent to him from Skidaway Island. In the same volume, on pages 114 to 124, plate VIII, William Cooper described teeth and bones which had been sent to him from the same locality by Joseph E. Habersham. Cooper had some reason to conclude that all the bones and teeth found up to that time had come from the same individual. In 1828 (Annals cited, vol. II, pp. 267–270) Cooper described additional materials which he had received from Skidaway Island.

In 1846 (Hodgson’s Mem. Megath., p. 25), Habersham gave a list of the fossil bones and teeth found at the island mentioned. Lyell (op. cit., p. 313) gave a brief account of a visit to Skidaway Island and stated that Megatherium, Mylodon, Mastodon, Elephas primigenius, and a species of the ox tribe had been found there. In 1855 (Smithson. Contrib. Knowl., vol. VII, art. 5, p. 50) Leidy enumerated the specimens of Megatherium which had been found at Skidaway Island, and he gave an excellent figure (plate xv) of a ramus of the lower jaw containing all its teeth, which had been sent to the National Institute at Washington. These bones ought to be now in the National Museum, but the writer has not been able to find them. They may never have been transferred and may be lost. On the other hand, Leidy did not mention other specimens from Skidaway Island, given by Scriven, and now in the National Museum. One of these is the hinder part of a skull figured in Hodgson’s memoir. Also, the same plate contains what is almost certainly the astragalus; its greatest diameter is 9 inches. Furthermore, there is present the distal end of a right humerus presented by Doctor Scriven. It is probably one of the two mentioned on page 27 of Hodgson’s memoir. As in the one there measured, the distance across the condyles is 14 inches and that across the articular surfaces is 7.75 inches. The Scriven collection also contains several teeth and fragments of others. A piece of the maxilla bears the small hindermost upper molar, no doubt the fragment mentioned by Habersham in his memorandum, page 26. Many of the bones sent from the island show by the presence of barnacles and bryozoa that at one time they lay in salt water; but this was probably not long before they were discovered.

Lyell stated that among other animals which had been found at Skidaway Island was Mylodon. Mylodon was reported by Lyell (“Travels in North America,” vol. I, p. 164) as having been found at Heyner’s Bridge. This is or was situated about 7 miles south of Savannah and about 5 miles northwest from the locality on Skidaway Island where the Megatherium and Mylodon remains were found. The map accompanying Hodgson’s memoir is here reproduced as map [40].

FLORIDA.

(Maps [3], [4].)

1. Archer, Alachua County.—Leidy mentioned (Proc. Acad. Nat. Sci. Phila., 1886, pp. 11, 12) the fact that an astragalus of Megatherium had been found at Archer. Several other species of vertebrates have been found there, among them Teleoceras fossiger, Gomphotherium floridanum, Hipparion plicatile, three species of Procamelus, and a species of Tapirus. The deposits are assigned to the Pliocene, but it is doubtful whether the megatherium and the tapir belonged among the others. The geology of the locality is discussed on page [375]. The megatherium, as an undetermined species, is included in the list of fossils which is recorded by Leidy in Bulletin 84 of the United States Geological Survey, page 129. It may be referred provisionally to Leidy’s Megatherium mirabile.

2. Almero Farm, St. John County.—In the collection of Mr. Fred Allen, at St. Augustine, the writer has seen a right tibia of a mylodon found in the Inland Waterway Canal about 28 miles south of St. Augustine. The bone is complete, except that a sliver has been split off the upper half of the outer border. The total length of the bone is 290 mm.; the greatest width of the upper end 208 mm.; width at middle of length 105 mm.; width of surface for astragalus 130 mm. This appears to be a relatively stouter bone than the larger one described by Harlan (Amer. Jour. Sci., vol. XLIV, 1842, p. 77). It is also larger and relatively stouter than a tibia found at Labelle, Lee County, described on page [40]. It is referred to Mylodon harlani.

11. Williston, Levy County.—In the U. S. National Museum there are some foot-bones of a large ground-sloth, which are labeled as having been collected in 1887 by the U. S. Geological Survey, in the county named. The collector was probably J. B. Hatcher. The astragalus had evidently been studied by Leidy. This bone was described by the writer in 1919 (Proc. U. S. Nat. Mus., vol. LVI, p. 104, plate XXVII) as Thinobadistes segnis. Later, other parts of the foot were found in the museum and described (Proc. U. S. Nat. Mus., vol. LIX, p. 638, plate CXIX, figs. 6–11).

3. Ocala, Marion County.—In 1888, in a fissure in a limestone quarry, probably Phillip’s quarry, near Ocala, Mr. Joseph Willcox discovered some vertebrate remains which were later described by Leidy (Trans. Wagner Free Inst., vol. II, pp. 13–17, plate III, figs. 1, 5, 6 to 9). The species as determined by Leidy were Elephas columbi, Equus fraternus, Auchenia minima, and Machairodus floridanus. They were regarded as belonging to the Quaternary, but in Dall’s paper of 1892 (Bull. 84, U. S. Geol. Surv., p. 129) they are referred to the age of the Alachua clays; that is, to the Pliocene. Sellards, in 1916 (8th Ann. Rep. Florida Geol. Surv., p. 103), regards the fossils as belonging to the Pleistocene, and he adds representatives of 4 genera to the list. These are undetermined species of Bison, Odocoileus, Dasypus, and Sylvilagus. The genus Dasypus is the one to which attention is especially called at this time. A list of the vertebrate animals found at this place is presented on page [378].

4. Dunnellon, Marion County.—In Sellards’s report just referred to, he prints a list of the Pleistocene vertebrates found in Withlacoochee River. Among these is the xenarthrid animal Chlamytherium septentrionale. What parts were secured and exactly at what place the writer does not know.

In the collection of the Florida Geological Survey is a foot-bone, No. 1307, which appears to be the second right metacarpal of Megalonyx. It is smaller than the one figured by Leidy. The extreme length is 60 mm., the greatest diameter of the proximal end 27 mm., that of the distal end 36 mm. It was found in the mine of the Dunnellon Phosphate Company. For a list of the associated species the reader is referred to page 376.

5. Hillsboro River, Hillsboro County.—In 1915 (Amer. Jour. Sci., vol. XL, p. 139), Sellards stated that the Jarman collection at Vanderbilt University, at Nashville, contains several dermal plates of Chlamytherium septentrionale, found in Hillsboro River.

6. Sarasota Bay, Sarasota County.—In 1915, Sellards (op. cit., p. 143) reported that the collection of Wagner Free Institute at Philadelphia contains one dermal plate of Chlamytherium septentrionale found by Joseph Willcox at White Beach, on Sarasota Bay.

The American Museum of Natural History, New York, possesses a dermal plate of a xenarthrid, collected by Barnum Brown 8 miles southeast of Sarasota. This probably belonged to the animal mentioned above.

7. Zolfo, Hardee County.—Dr. W. D. Matthew has informed the writer that there are in the American Museum of Natural History some bones of a very large individual of Megatherium, reported as having been found near Zolfo. An astragalus, the proximal part of a humerus, the distal part of a radius, and the proximal part of a femur were mentioned. These bones may be referred provisionally to Megatherium mirabile Leidy.

8. Vero, St. Lucie County.—At this place there have been found remains representing 4 genera of xenarthrids, as follows: Megalonyx, Mylodon, Chlamytherium, and Dasypus.

Megalonyx jeffersonii is represented by a part of a lower jaw, a right upper canine tooth, a molar tooth, a part of a hyoid bone, an axis, an astragalus, a median phalanx, and a claw (Sellards, 8th Ann. Rep. Florida Geol. Surv., p. 148, plate XXV, fig. 2; plate XXX, fig. 6). These were all found in the stratum denominated No. 2 in the report just cited.

Mylodon harlani? is known from a single claw, but from which stratum it was derived is not known.

Chlamytherium is represented by a part of the right side of the lower jaw, a part of the left side, a foot-bone, and numerous dermal plates (Sellards, op. cit., p. 148, plate XXVIII, figs. 4 to 6; plate XXX, fig. 7). Most of these remains have been taken from stratum No. 2, but some finely preserved dermal plates have been collected from No. 3.

Dasypus remains, consisting of dermal scutes, have been found in both No. 2 and No. 3.

In the collection of the Florida Geological Survey (No. 1795) is a bone, apparently the right parietal of an undetermined xenarthrid. It was found in the canal of the Indian River Farms Company, east of the railway and near Indian River. The length of the bone at the midline is 70 mm. and here the thickness is 22 mm. There appears to have been no median crest and only a feebly indicated occipital crest. There is no rough surface for the temporal muscles, as in Nothrotherium, and the bone is thicker than in that genus.

For complete lists of the fossil vertebrates found at Vero, see page [382].

9. Arcadia, De Soto County.—The Xenarthra are represented in the Pleistocene deposits about Arcadia by the genera Megalonyx, Glyptodon, and Chlamytherium. If these were not found at Arcadia they were collected along Peace Creek, not far from the town. A list of the species found in the vicinity of Arcadia is given on page [380].

Leidy (Trans. Wagner Free Inst., vol. II, p. 27) stated that a first phalanx of Megalonyx jeffersonii was among the fossils collected along Peace Creek. It was probably found on the sand-bar at Arcadia. Among the fossil vertebrates described by Leidy, the paper just cited included some dermal plates which he referred to the genus Glyptodon. Two of these plates were figured (op. cit., plate IV, fig. 9; plate VI, fig. 1) as those of G. petaliferus, a species based on half of a dermal scute described by Cope from southwestern Texas. The dermal scute shown on Leidy’s plate IV appears to be indistinguishable from similar plates which have been referred by the present writer to Cope’s G. petaliferus (Proc. U. S. Nat. Mus., vol. LI, 1916, p. 107, plates III to V). The scute represented by Leidy on his plate VI appears to be far less extensively pitted than any of those of the specimen just referred to. On Leidy’s plate V are two views of a scute which he thought might have belonged on the tail of a glyptodon. It will be observed that this scute has a beak distinctly set off from the body of the scute. Among the few caudal scutes of the specimen which the writer described none presents such a beak, but such may have existed. It seems probable, however, that there was a single species of Glyptodon found on Peace Creek and that it was different from G. petaliferus. Leidy thought that these caudal scutes resembled those on the tail of the South American G. asper; but Burmeister’s figures do not indicate exactly such keeled scutes. It is most probable that the Florida species requires a new name. It is to be called Glyptodon rivipacis Hay.

Leidy referred another dermal scute to some glyptodont animal (op. cit., plate VI, figs. 2, 3), but its nature is doubtful; it may even belong to one of the large species of Testudo. A conical bone (plate III, figs. 10, 11) belonged pretty certainly to Testudo.

In the paper cited Leidy described and figured (p. 24, plate III, figs. 3 to 6) plates of an armadillo-like animal to which he gave the name Glyptodon septentrionalis. It is now known as Chlamytherium septentrionale. Leidy had over 30 of these dermal scutes which had been found at Arcadia. They are now in the Wagner Free Institute at Philadelphia.

Sellards (Amer. Jour. Sci., vol. XL, 1915, p. 143) states that there are 3 dermal plates of this animal in the U. S. National Museum. In 1915 (Florida Geol. Surv., vol. VII, pp. 77, 78, plate on p. 114) he described a lower jaw, a tooth, and 2 dermal plates of the same animal.

10. Labelle, Lee County.—In the Florida Geological Survey is a right tibia of a mylodon, found on the bank of Caloosahatchee River, near Labelle, presented by Capt. F. H. Hendry. The total length is 266 mm.; on the inner border 236 mm. The width across the articulatory surface for the femur is 164 mm. The width at the middle of the length is 84 mm.; fore-and-aft diameter at the same place 38 mm. The side-to-side diameter of the surface for the astragalus is 57 mm. The bone is referred to Mylodon harlani.

11. See page [37].

ALABAMA.

(Map [3].)

1. Tuscumbia, Colbert County.—In his work on the “Extinct Sloth Tribe” in North America (Smithson. Contrib. Knowl., vol. VII, art. V, p. 6, plate XVI, fig. 13), Leidy, in recording the materials belonging to Megalonyx jeffersonii at his disposal, mentioned a supposed third upper molar, said to have come “from Tuscumbia County, Alabama.” This was an error, as the name of the town is Tuscumbia. The tooth had been loaned to him by Dr. Jeffries Wyman. Nothing more is known about its history. Mercer (Proc. Amer. Philos. Soc., vol. XXXVI, p. 38) stated that a well-preserved series of bones of Megalonyx had been sent to the Academy of Natural Sciences at Philadelphia by Mr. Tuomey. They had been obtained in a cave somewhere in northern Alabama. Leidy does not mention this collection in his work just cited.

MISSISSIPPI.

(Map [3].)

1. Natchez, Adams County.—Dr. M. W. Dickeson (Proc. Acad. Nat. Sci. Phila., 1846, p. 106) exhibited before the Academy a large series of fossil bones secured by him near Natchez. Among these were noted especially what was described as an entire head with part of the lower jaw, and many parts of the skeleton of Megalonyx jeffersonii. This skull is still in the collection of the Academy. The lower jaw is missing. It appears that several skeletons were represented in Dickeson’s collection. These, as Dickeson stated, had been found in a tenacious blue clay which underlies what he called diluvial drift, but now regarded as being at least principally loess. Associated with this animal were remains of Ursus, Bos (Bison), Cervus (Odocoileus), Equus, and some other but undetermined genera.

In his “Second Visit to the United States of North America,” edition 2, 1850, volume II, p. 196, Lyell mentions the Megalonyx among other fossils found at Natchez. He states that the fossils found by Doctor Dickeson were obtained in the “Mammoth Ravine” 6 miles from Natchez.

In Southall’s “Recent Origin of Man,” 1875, page 552, is a statement made by Professor C. G. Forshey (as quoted from Foster’s “Prehistoric Races of the United States,” p. 61) in which he says that he visited the locality where the human pelvis was found and that it was situated in Bernard’s Bayou, 2.5 miles from Natchez.

In his memoir of 1853 on “Extinct Species of American Ox” (Smithson. Contrib. Knowl., vol. V, art. III, p. 10), Doctor Leidy included Mylodon among the genera found at Natchez. In his memoir of 1855 on the “Extinct Sloth Tribe of North America” (Smithson. Contrib. Knowl., vol. VII, art. V, p. 48) he gave a list of the bones and brief descriptions of them. They all belonged to one individual, which was about half-grown.

In a list furnished to B. C. L. Wailles by Doctor Leidy (Wailles, Agric. Geol., Mississippi, 1854, p. 286), 4 species of Xenarthra are included among the mammals found fossil in the Pleistocene of Mississippi. These are Megalonyx jeffersonii, M. dissimilis, Mylodon harlani, and Ereptodon priscus. Cope regarded M. dissimilis as the same as M. jeffersonii, and Leidy was disposed to consider his Ereptodon priscus as identical with one of the species of Megalonyx.

A list of the fossil vertebrates found in the vicinity of Natchez will be given on page [392].

TENNESSEE.

(Map [3].)

1. Elroy, Van Buren County.—In 1831 (Jour. Acad. Nat. Sci., Phila., ser. I, vol. VI, pp. 269–286, plates XII to XIV; 1835, Med. Phys. Res., pp. 319–331, plates XII to XV), Richard Harlan described a number of bones of Megalonyx jeffersonii which had been purchased for the Academy of Natural Sciences, Philadelphia, and which he reported had been found in “White Cave,” Kentucky. This was supposed to be situated near Mammoth Cave. It was ascertained later that the bones had been found in Bigbone Cave, Van Buren County, Tennessee.

The bones mentioned by Harlan had belonged to a young animal and consisted of 5 vertebræ, a few fore-limb bones, a few hinder-limb bones, a scapula, a rib, and a part of a molar tooth. Some of the articulating surfaces still retained their cartilage. In the same cave were found bones of “Bos” (Bison), “Cervus” (Odocoileus?), Ursus, and a human metacarpal. These were said to have been found on the surface, while the megalonyx bones were buried at a depth of 2 or 3 feet. The mandible of the bear (Harlan, op. cit., p. 283) was described as displaying appearances of antiquity equal to that of the megalonyx bones. The sloth bones were made the basis of the name Megalonyx laqueatus. In 1855 (Smithson. Contrib. Knowl., vol. VII, art. 5, p. 4), Leidy determined that these bones belonged to M. jeffersonii. He wrote that the collection consisted of one molar tooth, four dorsal vertebræ, one lumbar, a left humerus lacking the upper epiphysis, the proximal two-thirds of the right ulna, the right radius, the left scapula, the distal epiphysis of the right femur, the left tibia, and the distal epiphysis of the right tibia, a right calcaneum, two claws of a hinder foot, and some fragments of ribs. Leidy appears to have concluded that these bones had been those of a young animal, but that other bones in the collection had belonged to adult individuals. He stated that they had come from Bigbone Cave, White County. This adjoins Van Buren on the north and possibly at that time included the latter; or Leidy may have been mistaken. Besides the bones above mentioned, Harlan described from this cave an ilium of Megalonyx (Med. and Phys. Res., p. 334).

In 1892 (Bull. Geol. Soc. Amer., vol. III, pp. 121–123), Professor J. M. Safford reported the discovery of some bones of a megalonyx in Bigbone Cave. They had been met with in the bat manure at a depth of about 3 feet. The parts received by Professor Safford, and which are all probably in Vanderbilt University, were the skull, 17 vertebræ (including 5 sacrals), a fragment of a rib, a right scapula, a right humerus, the two ilia, a part of the right pubis, a part of the right ischium, and a left tibia. Safford concluded that these bones formed a part of the same young animal that Harlan had described.

In 1897 (Proc. Amer. Philos. Soc., vol. XXXVI, pp. 36–70), Dr. H. C. Mercer gave a detailed account of his explorations in this cave. It is situated about a mile from the left bank of Caney Fork River, a mile above the mouth of a confluent called Dry Branch, and at an elevation of about 1,000 feet above sea-level. It is excavated in Carboniferous limestone and opens into what is known as “Beech Cove.” Thomas L. Bailey (“Resources of Tennessee,” vol. VIII, pp. 131–132) described it as being situated 3.5 miles south of Quebeck, near the head of a hollow or cove extending south from McElroy’s store. The latter is probably the locality put down on the topographic sheet of the quadrangle as Elroy. It is further said to be one branch of an extensive cave whose other branch is known as Arch Cave. Bigbone Cave is known to extend a distance of 3 miles. It appears that the cave had been exploited for saltpeter in the wars of 1776, 1812, and 1863 and immense amounts of the nitrous earth had been removed. Mercer found no bones until he had reached a small passage at a distance of 900 feet from the entrance. Here he found an epiphysis of a left humerus, 6 vertebræ, an astragalus, and a calcaneum of a sloth, evidently a young animal; and he concluded that they were probably parts of the same animal that Harlan had described many years before; also a part of a skeleton that had been found there in 1884, which is the one described by Safford. A remarkable feature of the bones of the young animal found in this cave, as noted by Harlan, Leidy, and Mercer, is the presence of some of the cartilage, some shreds of ligaments, and a part of the horny sheath of one claw.

2. Lookout Mountain, Hamilton County.—In 1894 (Amer. Naturalist, vol. XXVIII, pp. 355–357), Dr. H. C. Mercer reported his work, done in 1893, in a cave situated on Lookout Mountain, near Chattanooga, Tennessee. In a brief report made June 4, 1896 (Dept. Amer. and Prehist. Archæol. Univ. Penn.), Mercer stated that this cave is on the left bank of Tennessee River, 0.25 mile below Chattanooga Creek. According to the report last quoted, the cave earth, “with its culture layer,” was removed by him to a distance of 58 feet from the entrance. According to the report of 1894, this was effected by digging 4 trenches, 6 feet 10 inches wide and with a depth of 3 feet, in two cases to rock bottom. Near the bottom of the deposit were found a jaw of Tapirus haysii with teeth, and a jaw of a small Mylodon, identified as such by Professor E. D. Cope. A bone of the extinct peccary appears to have been found higher up in the layer of refuse. In a letter received by the writer in 1919, Doctor Mercer stated that later Cope expressed some doubt regarding the identity of the bone supposed to belong to Mylodon.