INTRODUCTORY.

Ever since America was discovered, the question about it which has excited the most general interest has been, Whence came its inhabitants? The inquiry, Who are the American Indians? has been the theme of many a ponderous folio and labored dissertation, with answers nearly as various as the number of debaters.

Few or none of them have reflected on the unphilosophical character of the inquiry as thus crudely put. Take a precisely analogous question, and this will be apparent—Whence came the African Negroes? All will reply—From Africa, of course. Originally? Yes, originally; they constitute the African or Negro sub-species of Man.

The answer in the case of the American Indians is entirely parallel—their origin is American; the racial type was created and fixed on the American continent; they constitute as true and distinct a sub-species as do the African or the White Race.

Each of the great continental areas moulded the plastic, primitive man into a conformation of body and mind peculiar to itself, in some special harmony with its own geographic features, thus producing a race or sub-species, subtly correlated in a thousand ways to its environment, but never forfeiting its claim to humanity, never failing in its parallel and progressive development with all other varieties of the species.

America was no exception to this rule, and it is time to dismiss as trivial all attempts to connect the American race genealogically with any other, or to trace the typical culture of this continent to the historic forms of the Old World. My early studies inclined me to these opinions, and they have been constantly strengthened by further research. Yet they are not popularly accepted; the very latest writer of competence on the pre-history of America says, “It is now generally held that the earliest population (of the continent) was intruded upon by other races, coming either from Asia or from the Pacific Islands, from whom were descended the various tribes which have occupied the soil down to the present time.”^[1]

It is true that this opinion is that generally held, and for this reason I have selected for reprinting some articles intended to show that it is utterly fallacious—devoid of any respectable foundation.

The first two papers treat of the archÆologic material, and its value for ascertaining the pre-historic life of the American race; the third, on its pretended affinities to Asiatic peoples. These are followed by two papers respectively on the Toltecs and Mound-Builders, setting aright, I hope, the position of these semi-mythical shapes in the culture-history of North America, maintaining that for neither do we have to call in as explanation migrations from Asia, Europe, Oceanica or Africa, as has so often been attempted.

Early in this century the doubt was expressed by Alexander von Humboldt^[3] whether it is philosophical to inquire into the origin of any of the human races or sub-species. Although he expressed this doubt with particular reference to the American race, I believe I am right in assuming that the hesitancy he felt in pushing inquiry so far should now diminish in view of new methods of research and a wider range of observations. We may not, in fact we shall not, be able to trace the American or any other sub-species directly back to its origin in place or time; but by reviewing all the data which have been offered in solution of such a problem, we may perceptibly narrow the question, and also estimate the relative value of the means proposed. It is to such a review, applied to the American race, that I now invite your attention.

The data upon which theories of the antiquity, the genealogy and the affinities of this race have been constructed are varied. For convenience of treatment I shall class them under six heads. They are:

I. Legendary, including the traditions of the native tribes and their own statements of their history.

II. Monumental, where we have to do with those structures whose age or character seems to throw light on the question.

III. Industrial, under which heading we may inquire as to the origin of both the useful and the decorative arts in the New World.

IV. Linguistic, broaching the immense and important questions as to the diversity and affinities of languages.

V. Physical, which takes into consideration the anatomic and morphologic peculiarities of the American race; and finally,

VI. Geologic, where its position in the geologic horizons is to be determined, and the influence upon it of the physical geography of the continent.

Legendary. Turning to the first of these, the legendary data, I confess to a feeling of surprise that learned scholars should still hold to the opinion that the native tribes, even some of the most savage of them, retain to this day traditions which they had brought from their supposed Asiatic homes. Thus the missionaries, Bishop Henry Faraud and the AbbÉ Emile Petitot, both entirely familiar with the Cree and the Athapaskan languages and lore, insist that the myths and legends of these tribes bear such strong resemblances to the Semitic traditions that both must have had a common origin.^[4] No one can deny the resemblance; but the scientific student of mythology discovers such identities too frequently, and at points too remote, to ask any other explanation for them than the common nature of the human mind.

The question has been often raised how long a savage tribe, ignorant of writing, is likely to retain the memory of past deeds. From a great many examples in America and elsewhere, it is probable that the lapse of five generations, or say two centuries, completely obliterates all recollection of historic occurrences. Of course, there are certain events of continuous influence which may be retained in memory longer—for example, the federation of prominent tribes; and perhaps a genealogy may run back farther. My friend, Dr. Franz Boas, informs me that some tribes on Vancouver’s Island pretend to preserve their genealogies for twelve or fifteen generations back; but he adds that the remoter names are clearly of mythical purport.

It appears obvious that all efforts to establish a pre-historic chronology by means of the legends of savage tribes, are and must be vain.

The case is not much better with those semi-civilized American nations, the Mayas and Nahuas, who possessed a partially phonetic alphabet, or with the Quichuas, who preserved their records by the ingenious device of the quipu. Manco Capac, the alleged founder of the Peruvian state, floats before us as a vague and mythical figure, though he is placed in time not earlier than the date when Leif, the son of Erik, anchored his war-ship on the Nova Scotian coast.^[5] Historians are agreed that the long lists of Incas in the pages of Montesinos, extending about two thousand years anterior to the Conquest, are spurious, due to the imagination or the easy credulity of that writer.

The annals of Mexico fare no better before the fire of criticism. It is extremely doubtful that their earliest reminiscences refer to any event outside the narrow valley parcelled out between the petty states of Tenochtitlan, Tezcuco, and Tlacopan.^[6] The only fact that bears out the long and mysterious journey from the land of the Seven Caves, Chicomoztoc, in the distant northwest, by the great water, is that the learned and indefatigable Buschmann has conclusively shown that the four languages of Sonora and all the dialects of the Shoshonian family reveal marks of continued and deep impressions of the Nahuatl tongue.^[7] But the chronicles of Mexico proper contain no fixed date prior to that of the founding of the city of Tenochtitlan, in the year 1325 of our era.

I am aware that there are still some writers who maintain that both the Mexican and the Maya astronomic cycles assume a commencement for their records centuries, even thousands of years, before the beginning of our era. These opinions, however, have not obtained the assent of other students. We are too ignorant both of the astronomy and the methods of writing of these nations to admit such claims; and the facts advanced are capable of quite other interpretation.

It is, on the whole, rare for the American tribes to declare themselves autochthonous. The Mayas, on the peninsula of Yucatan, stated that their earliest ancestors came there from beyond the seas, some from the far east, others from the west. So the Toltecs, under Quetzalcoatl, were fabled to have entered Mexico from beyond the Eastern Ocean. The Creeks and Choctaws pointed to the west, the Algonkins generally to the east, as their primal home.^[8] These legends are chiefly mythical, not much truer than those of other tribes who claimed to have climbed up from some under-world. Sifting them all, we shall find in them little to enlighten us as to the pre-historic chronology of the tribes, though they may furnish interesting vistas in comparative mythology.

That in which we may expect the legends of tribes to be of most avail is their later history, the record of their wars, migrations and social development within a few generations. The spirit of the uncivilized man is, however, very careless of the past. We have means of testing the exactness of such traditions in some instances, and the result is rarely such as to inspire confidence in verbal records. Those of you who were present at the last meeting will remember how diversely two able students of Iroquois tradition estimated its value. Even when remarkable events are not forgotten, the dates of their occurrence are generally vague. The inference, therefore, is that very few data, dependent on legendary evidence alone, can be accepted.

Monumental. When we turn to the monumental data, to the architecture and structural relics of the ancient Americans, we naturally think first of the imposing stone-built fortresses of Peru, the massive pyramids and temples of Yucatan and Mexico, and the vast brick-piles of the Pueblo Indians.

It is doubtful if any of these notable monuments supply pre-historic dates of excessive antiquity. The pueblos, both those now occupied and the vastly greater number whose ruins lie scattered over the valleys and mesas of New Mexico and Arizona, were constructed by the ancestors of the tribes who still inhabit that region, and this at no distant day. Though we cannot assign exact dates to the development of this peculiar civilization, there are abundant reasons, drawn from language, physical geography and the character of the architecture, to include all these structures within the period since the commencement of our era.^[9]

There is every reason to suppose that the same is true of all the stone and brick edifices of Mexico and Central America. The majority of them were occupied at the period of the Conquest; others were in process of building; and of others the record of the date of their construction was clearly in memory and was not distant. Thus, the famous temple of Huitzilopochtli at Tenochtitlan, and the spacious palace—or, if you prefer the word, “communal house”—of the ruler of Tezcuco, had been completed within the lifetime of many who met the Spaniards. To be sure, even then there were once famous cities fallen to ruin and sunk to oblivion in the tropical forests. Such was Palenque, which could not have failed to attract the attention of Cortes had it been inhabited. Such also was T’Ho, on the site of the present city of MÉrida, Yucatan, where the earliest explorers found lofty stone mounds and temples covered with a forest as heavy as the primitive growth around it.^[10] But tradition and the present condition of such of these old cities as have been examined, unite in the probability that they do not antedate the Conquest more than a few centuries.

In the opinion of some observers, the enigmatical ruins on the plain of Tiahuanaco, a few leagues from the shore of Lake Titicaca, in Peru, carry us far, very far, beyond any such modern date. “Even the memory of their builders,” says one of the more recent visitors to these marvellous relics, General BartolomÉ Mitre, “even their memory was lost thousands of years before the discovery of America.”^[11]

Such a statement is neither more nor less than a confession of ignorance. We have not discovered the period nor the people concerned in the ruins of Tiahuanaco. It must be remembered that they are not the remains of a populous city, but merely the foundations and beginnings of some vast religious edifice which was left incomplete, probably owing to the death of the projector or to unforeseen difficulties. If this is borne in mind, much of the obscurity about the origin, the purpose and the position of these structures will be removed. They do not justify a claim to an age of thousands of years before the Conquest; hundreds will suffice. Nor is it necessary to assent to the opinion advanced by General Mitre, and supported by some other archÆologists, that the most ancient monuments in America are those of most perfect construction, and, therefore, that in this continent there has been, in civilization, not progress but failure, not advance but retrogression.

The uncertainty which rests over the age of the structures at Tiahuanaco is scarcely greater than that which still shrouds the origin of the mounds and earthworks of the Ohio and Upper Mississippi valleys. Yet I venture to say that the opinion is steadily gaining ground that these interesting memorials of vanished nations are not older than the mediÆval period of European history. The condition of the arts which they reveal indicates a date that we must place among the more recent in American chronology. The simple fact that tobacco and maize were cultivated plants is evidence enough for this.^[12]

There is, however, a class of monuments of much greater antiquity than any I have mentioned. These are the artificial shell-heaps which are found along the shores of both oceans and of many rivers in both North and South America. They correspond to the kitchen-middens of European archÆology.

In several parts of the continent they have been examined by competent observers and the question of their date approximately ascertained. I need not say this differs widely, for these refuse heaps of ancient villages or stations were of course begun at wide intervals.

Long ago I called attention to the singular size and antiquity of those I found in Florida and along the Tennessee River;^[13] and the later researches of Professor Jeffries Wyman would, in his opinion, measure the age of some of the former by tens of thousands of years.^[14]

Further to the south, in Costa Rica, Dr. Earl Flint has examined the extensive artificial shell deposits which are found along the shores of that republic. They are many feet in height, covered by a dense forest of primeval appearance, and are undoubtedly of human origin.

In Brazil such shell-heaps are called sambaquis, and they are of frequent occurrence along the bays and inlets of the coast. Some of them are of extraordinary dimensions, rising occasionally to more than a hundred feet in height. The lower layers have been consolidated into a firm, stony breccia of shells and bones, while the surface stratum, from six to ten feet thick, is composed of sand and vegetable loam supporting a growth of the largest trees. Yet even the lowest layers of this breccia, or shell-conglomerate, yield tokens of human industry, as stone axes, flint arrow-heads, chisels, and fragments of very rude pottery, as well as human bones, sometimes split to extract the marrow. The shells are by no means all of modern type. Many are of species now wholly extinct, or extinct in the locality. This fact alone carries us back to an antiquity which probably should be counted by thousands of years before our era.

At that remote period not only did a fishing and hunting race dwell along the Brazilian coast, but this race was fairly advanced on the path to culture; it was acquainted with pottery, with compound implements, and with the polishing of stone. We further know that this race was not that which occupied the land when the whites discovered it; for the human skulls disinterred from the sambaquis are, craniologically, almost diametrically opposite those of the Botocudos and the Tupis. Yet if we can trust the researches of Dr. Lund in the caverns of Brazil, the oldest skulls in these deposits, found in immediate connection with the bones of extinct mammalia, belonged to the ancestors of these tribes. Markedly dolichocephalic, they present an entire contrast to the brachycephalic type from the sambaquis.^[15]

This class of monuments, therefore, supply us data which prove man’s existence in America in what some call the “diluvial,” others the “quaternary,” and others again the “pleistocene” epoch—that characterized by the presence of some extinct species.

Industrial. Let us now turn to the industrial activity of the American race, and see whether it will furnish us other data concerning the pre-historic life of the New World. We may reasonably look in this direction for aid, since it is now universally conceded that at no time did man spring into being fully armed and equipped for the struggle for existence, but everywhere followed the same path of painful effort from absolute ignorance and utter feebleness to knowledge and power. At first, his only weapons or tools were such as he possessed in common with the anthropoid apes: to wit, an unshapen stone and a broken stick. Little by little, he learned to fit his stone to his hand and to chip it to an edge, and with this he could sharpen the end of his stick, thus providing himself with a spear and an axe.

It was long before he learned to shape and adjust the stone to the end of the stick, and to hurl this by means of a cord attached to a second and elastic stick—in other words, a bow; still longer before he discovered the art of fashioning clay into vessels and of polishing and boring stones. These simple arts are landmarks in the progress of the race: the latter divides the history of culture into the palÆolithic or rough stone period, and the neolithic or polished stone period; while the shaping of a stone for attachment to a handle or shaft marks the difference between the epoch of compound implements and the earlier epoch of simple implements, both included in the older or palÆolithic age.^[16] With these principles as guides, we may ask how far back on this scale do the industrial relics in America carry us?

I have spoken of the great antiquity of some of the American shell-heaps, how they carry us back to the diluvial epoch, and that of numerous extinct species. Yet it is generally true that in the oldest hitherto examined in Brazil, Guiana, Costa Rica and Florida, fragments of pottery, of polished stone, and compound implements, occur even in the lowest strata.^[17] Venerable though they are, they supply no date older than what in Europe we should call the neolithic period. The arrow-heads which have been exhumed from the loess of the ancient lake-beds of Nebraska, the net-sinkers and celts which have been recovered from the auriferous gravels of California, prove by their form and finish that the tribes who fashioned them had already taken long strides beyond the culture of the earlier palÆolithic age. The same is true, though in a less degree, of the chipped stones and bones which Ameghino exhumed from the lacrustine deposits of the Pampas, although he proves that these relics were the products of tribes contemporary with the extinct glyptodon and mylodon, as well as the fossil horse and dog. In the very oldest station which he examined, there appears to have been found a quartz arrow-head; yet he argues that this station dated from the pliocene division of the tertiary, long anterior to the austral glacial epoch.^[18] This leaves another such open conflict between geology and the history of culture, as Professor Rau has already pointed out as existing in Californian archÆology.

There is, however, one station in America which has furnished an ample line of specimens, and among them not one, so far as I know, indicating a knowledge of compound implements. This is that of the “Trenton gravels,” New Jersey. There we appear to be in face of a stage of culture as primitive as that of the stations of Chelles and St. Acheul in France, absolutely without pottery, without polished stone, without compound implements.^[19]

Assuming that these post-glacial gravels about Trenton supply one of the earliest authentic starting points in the history of culture on this continent, the later developments of industry will furnish a number of other data. This first date was long before the extinction of the native American horse, the elephant, the mammoth, and other animals important to early man. There is nothing unlikely therefore in the reported discoveries of his pointed flints or his bones in place along with the remains of these quadrupeds.

Not only the form but the material of implements supplies us data. If man in his earliest stage was, as some maintain, quite migratory, it is certain that he did not carry his stone implements with him, nor did he obtain by barter or capture those of other tribes. All the oldest implements are manufactured from the rocks of the locality. When, therefore, we find a weapon of a material not obtainable in the vicinity, we have a sure indication that it belongs to a period of development considerably later than the earliest. When the obsidian of the Yellowstone Park is found in Ohio, when the black slate of Vancouver’s Island is exhumed in Delaware, it is obvious we must assume for such extensive transits a very noticeable Æsthetic and commercial development.

I can but touch in the lightest manner on the data offered by the vast realm of industrial activity. The return it offers is abundant, but the harvesting delicate. In the dissemination of certain kinds of arts, certain inventions, certain decorative designs and Æsthetic conceptions from one tribe to another, we have a most valuable means of tracing the pre-historic intercourse of nations: but we must sedulously discriminate such borrowing from the synchronous and similar development of independent culture under like conditions.

In one department of industry we shall be largely free from this danger, that is, in the extension of agriculture. One of America’s ablest ethnologists, Dr. Charles Pickering, as the result of a lifetime devoted to his science, finally settled upon the extension of cultivated plants as the safest guide in the labyrinth of pre-historic migrations. Its value is easily seen in America when we reflect that the two tropical plants, maize and tobacco, extended their area in most remote times from their limited local habitat about the Isthmus of Tehuantepec to the north as far as the St. Lawrence river and to the south quite to the Archipelago of Chiloe. Their presence is easily traced by the stone or earthen-ware implements required for their use. How many ages it must have required for these plants to have thus extended their domain, amid hostile and savage tribes, through five thousand miles of space! The squash, the bean, the potato and the mandioca, are native food-plants offering in a less degree similar material for tracing ancient commerce and migration. Humboldt and others have claimed as much for the banana (Musa paradisiaca), but the recent researches of Dr. Karl von den Steinen have removed that valued fruit from the list of native American plants. Both species of banana (M. paradisiaca and M. sapientium) were undoubtedly introduced into the New World after the discovery.^[20] Indeed, summing up the reply to an inquiry which has often been addressed to the industrial evolution of the indigenes of our continent, I should say that they did not borrow a single art or invention nor a single cultivated plant from any part of the Old World previous to the arrival of Columbus. What they had was their own, developed from their own soil, the outgrowth of their own lives and needs.

Linguistic. This individuality of the race is still more strongly expressed in their languages. You are all aware that it is upon linguistic data almost exclusively that American ethnology has been and must be based. The study of the native tongues becomes therefore of transcendent importance in the pre-historic chronology of the Continent. But to obtain its best results, this study must be conducted in a much more thorough manner than has hitherto been the custom.

In America we are confronted with an astonishing multiplicity of linguistic stocks. They have been placed at about eighty in North and one hundred in South America. It is stated that there are that many radically diverse in elements and structure. To appreciate the vista in time that this fact opens to our thoughts, we must recognize the tenacity of life manifested by these tongues. Some of them have scores of dialects, spoken by tribes wandering over the widest areas. Take the Athapascan or TinnÉ, for example, found in its greatest purity amid the tribes who dwell on the Arctic sea, and along the Mackenzie river, in British America, but which is also the tongue of the Apaches who carried it almost to the valley of Mexico. The Algonkin was spoken from Hudson Bay to the Savannah river and from Newfoundland to the Rocky Mountains. The Guarani of the Rio de la Plata underlies dialects which were current as far north as Florida.

How, then, in spite of such tenacity of American languages, have so many stocks come into existence? This was the question which my predecessor in this chair last year undertook to answer. His suggestions appear to me extremely valuable, and only in one point do I widely differ from him, and that is, in the length of time required for these numerous tongues to originate, to sever into dialects and to be carried to distant regions.^[21] According to the able linguist, Dr. Stoll, the difference which is presented between the Cakchiquel and Maya dialects could not have arisen in less than two thousand years;^[22] and any one who has carefully compared the earliest grammars of an American tongue with its present condition will acknowledge that the changes are surprisingly few. To me the exceeding diversity of languages in America and the many dialects into which these have split, are cogent proofs of the vast antiquity of the race, an antiquity stretching back tens of thousands of years. Nothing less can explain these multitudinous forms of speech.

Underlying all these varied forms of expression, however, I think future investigation will demonstrate some curious identities of internal form, traits almost or entirely peculiar to American languages, and never quite absent from any of them.

Such was the opinion of the two earliest philosophical investigators of these tongues, P. S. Duponceau and Wilhelm von Humboldt. They called these traits polysynthesis and incorporation, and it was proposed to apply the term incorporative as a distinguishing adjective to all American languages. Of late years this opinion has been earnestly combatted by M. Lucien Adam and others; but my own studies have led me to adopt the views of the older analysts against these modern critics. I do not think that the student can compare any two stocks on the continent without being impressed with the resemblance of their expression of the relations of Being, through the incorporative plan.

Along with this identity of plan, there coËxists the utmost independence of expression. An American language is usually perfectly transparent. Nothing is easier than to reduce it to its ultimate elements, its fundamental radicals. These are few in numbers and interjectional in character. The Athapascan, the Algonkin, whose wide extension I have referred to, have been reduced to half a dozen particles or sounds expressive of the simplest conceptions.^[23] Upon these, by combination, repetition, imitation and other such processes, the astonishing structure of the tongue has been erected, every portion of it displaying the mechanism of its origin. It is this transparency which renders these tongues so attractive to the philosophic student of human expression, and so valuable to him who would obtain from them the record of the progress of the nation.

A thorough study of such a language would embrace its material, its formal and its psychologic contents. Its material elements include the peculiarities of its vocabulary: for example, its numerals and the system they indicate, its words for weights and measures, for color and direction, for relations of consanguinity and affinity, for articles of use and ornament, for social and domestic conditions, and the like.

Few studies of American languages go beyond this material or lexicographic limit; but in truth these are merely the externalities of a tongue, and have nothing to do with linguistic science proper. This concerns itself with the forms of the language, with the relation of parts of speech to each other and to the sentence, and with the historical development of the grammatical categories. Beyond this, again, is the determination of the psychical character of the tribe through the forms instinctively adopted for the expression of its thoughts, and reciprocally the reaction exerted by these forms on the later intellectual growth of those who were taught them as their only means of articulate expression.

These are data of the highest value in the study of prehistoric time; but so far as America is concerned, I could name very few scholars who have pursued this promising line of research.

Physical. Much more attention has been paid to the physical than the linguistic data of the native Americans, but it may freely be said, with not more satisfactory results. This failure is partly owing to the preconceived notions which still govern the study of ethnology. LinnÆus offered the cautious division of the human species into races named from the five great geographical areas it inhabited; Blumenbach pointed out that this roughly corresponded with the division into five colors, the white, black, yellow, brown and red races, occupying respectively Europe, Africa, Asia, Polynesia and America. Unfortunately, Cuvier chose to simplify this scheme, by merging the brown and red races, the Polynesian or Malayan and the American, into the yellow or Mongolian. The latest writers of the French school, and I am sorry to add various Americans, servilely follow this groundless rejection of the older scheme, and speak of Malayans and Americans alike as Mongolians or Mongoloids. Neither in language nor ethnic anatomy is there any more resemblance than between whites and Mongolians.

It is gratifying to see that the more accurate German investigators decidedly reject the blunder of Cuvier, and declare that the American race is as independent as any other of those named. Thus Dr. Paul Ehrenreich, who has lately published an admirable monograph on the Botocudos of Brazil, a tribe often quoted for its so-called “Mongoloid” aspect, declares that any such assertion must be contradicted in positive terms. Both in osteology and anatomy, in formation of the hair and shape of the skull, the differences are marked, permanent and radical.

What is true of the Botocudos is not less so of the other American tribes which are claimed to present Mongolian traits. Such assertions are based on the superficial observations of travellers, most of whom do not know the first principles of ethnic anatomy. This is sufficiently shown by the importance they attach to the oblique eye, a slight malformation of the skin of scarcely any weight.^[24]

The anatomy and physiology of the various American tribes present, indeed, great diversity, and yet, beneath it all is a really remarkable fixedness of type. We observe this diversity in the shape of the skull, which may be, as among the Botocudos, strictly dolichocephalic, while the Araucanians are brachycephalic; the nasal index varies more than in the extremest members of the white race; the tint of the skin may be a dark brown with an under-color of red, or of so light a hue that a blush is easily perceptible. The beard is usually absent, but D’Orbigny visited a tribe who wore it full and long.^[25] The height varies from an average of six feet four inches for adult males in Patagonia to less than five feet among the Warraus of Guiana; and so it is with all the other traits of the race. There is not one which is not subject to extensive variation.

On the other hand, these variations are not greater than can be adduced in various members of the white or black race. In spite of them all, there is a wonderful family likeness among tribes of American origin. No observer well acquainted with the type would err in taking it for another. Darwin says that the Fuegians so closely resemble the Botocudos that they seem members of the same tribe. I have seen Arawacks from Guiana who in the northwest would have passed for Sioux.

In spite of the total dissimilarity of climate and other physical surroundings, the tribes of the tropics differ no more from those near the Arctic circle than they do among themselves. This is a striking lesson how independent of environment are the essential characteristics of a race, and it is a sweeping refutation of those theories which make such characteristics dependent upon external agencies.

A still more remarkable fact has been demonstrated by Professor J. Kollmann of BÂle: to wit, that the essential physical identity of the American race is as extended in time as it is in space. This accurate student has analyzed the cranioscopic formulas of the most ancient American skulls, those from the alleged tertiary deposits of the Pampas, those from the caverns of Lagoa Santa in Brazil, that obtained from Rock Bluff, Illinois, the celebrated Calaveras skull from California, and one from Pontemelo in Buenos Ayres of geologic antiquity. His results are most interesting. These very ancient remains prove that in all important craniologic indicia the earliest Americans, those who were contemporaries of the fossil horse and other long since extinct quadrupeds, possessed the same racial character as the natives of the present day, with similar skulls and a like physiognomy.^[26] We reach therefore the momentous conclusion that the American race throughout the whole continent, and from its earliest appearance in time, is and has been one, as distinct in type as any other race, and from its isolation probably the purest of all in its racial traits. This is a fact of the first order in establishing its prehistoric chronology.

Geologic. I have left the geologic data to the last, as it is these which carry us with reasonable safety to the remotest periods. No one who examines the evidence will now deny that man lived in both North and South America during and after the glacial epochs, and that he was the contemporary of many species of animals now extinct. As you are aware, the attempt has several times been made to fix the date for the final retrocession of the glaciers of North America. The estimates have varied from about 12,000 years ago up to 50,000, with a majority in favor of about 35,000 years.

There have also been various discoveries which are said to place the human species in America previous to the appearance of the glaciers. Some remains of man’s industry or of his skeleton have been reported from interglacial, others from tertiary deposits.^[27] Unfortunately, these finds have not always been sufficient, or not of a character to convince the archÆologist. I have before adverted to the impossibility, for instance, of an archÆologist accepting the discovery of a finely-polished stone implement in a tertiary gravel, except as an intrusive deposit. It is a violent anachronism, which is without a parallel in other countries. Even the discovery of a compound implement, as a stemmed arrowhead, in strata of tertiary date, is, with our present knowledge, quite out of the question.

Although there are well recognized signs of glacial action in South America, it is not certain that the glacial epoch coincided in time in the two continents. That there was a reasonable approximation is probable from the appearance of later deposits. We may suppose therefore that the habitable area of the New World was notably less at that period, and that the existing tribes were confined to a much narrower space. This would force them into closer relations, and tend powerfully to the production of that uniformity of type to which I have before referred.

We might also expect to discover in the tropical regions of America more frequent evidence of the primitive Americans than in either temperate zone. This has not been the case, probably because the geologic deposits of the tropics have been less investigated. Throughout the West Indies there is an entire absence of palÆolithic remains. Those islands were first peopled by tribes in the polished stone stage of culture. In the valley of Mexico human remains have been disinterred from a volcanic deposit of supposed tertiary age, and you have all heard of those human footprints which Dr. Earl Flint has unearthed in Nicaragua. These are found under layers of compact volcanic tufas, separated by strata of sand and vegetable loam. There can be no doubt of their human origin or of their great antiquity; but no geologist need be informed of the difficulty of assigning an age to volcanic strata, especially in a tropical country, subject to earthquakes, subsidence and floods.^[28]

It would not be in accordance with my present purpose to examine the numerous alleged finds of human remains in the strata of the tertiary and quaternary. All such furnish data for the pre-historic chronology of America, and should be carefully scrutinized by him who would obtain further light upon that chronology. I must hasten to some other considerations which touch the remote events to which I am now alluding.

Since a comparison of the fauna of South America and Africa, and a survey of the sea-bottom between those continents, have dispelled the dream of the ancient Atlantis, and relegated that land connection at least to the eocene period of the tertiary, no one can suppose the American man to have migrated from Africa or southwestern Europe. For other and equally solid reasons, no immigration of Polynesians can be assumed. Yet zoÖlogists, perfectly willing to derive man from an anthropoid, and polygenists to the utmost, hesitate to consider man an autochthon in the New World. There is too wide a gap between the highest monkeys and the human species in this continent.^[29] Discoveries of fossil apes might bridge this, but none such has been reported.

If we accept the theory that man as a species spread from one primal centre, and in the higher plasticity of his early life separated into well defined races, which became unalterably fixed not much later than the close of the glacial epoch—and this theory appears to be that now most agreeable to anthropologists—then the earliest Americans made their advent on this continent as immigrants. This is our first fact in their pre-historic chronology; but before we can assign it an accurate position on the scale of geologic time, we must await more complete discoveries than we now have at our command.

We must also wait until our friends the geologists have come to some better understanding among themselves as to what took place in the pleistocene age. You have heard me talking freely about the glacial epoch and its extension in America; but geologists are by no means of one mind as to this extension, and a respectable minority of them, led by Sir J. William Dawson, deny the existence or even possibility of any continental glacier. What others point out as a terminal moraine they explain to be “nothing but the southern limit of the ice-drift of a period of submergence.”^[30]

It is clear that when we speak about the migration of the Americans at a time when the polar half of each continent was either covered with a glacier thousands of feet thick, or submerged to that depth beneath an arctic sea, we have to do with geographical conditions totally unlike those of to-day. I call attention to this obvious fact because it has not been obvious to all writers.

In your archÆological reading you will rarely come across a prettier piece of theoretical history than Mr. Lewis A. Morgan’s description of the gradual peopling of the two Americas by tracing the lines of easiest subsistence. He begins at the fishy rivers of the northwest coast, and follows the original colony which he assumes landed at that point, all the way to Patagonia and Florida.^[31] But how baseless becomes this vision when we consider the geography of America as it is shown by geology to have been at a period contemporary with the earliest remains of man! We know to a certainty that the human race had already spread far and wide over both its continental areas before Mr. Morgan’s lines of easiest nutrition had come into existence.

Properly employed, a study of those geologic features of a country which determine its geography will prove of vast advantage in ascertaining the events of pre-historic time. These features undoubtedly fixed the lines of migration and of early commerce. Man in his wanderings has always been guided by the course of rivers, the trend of mountain chains, the direction of ocean currents, the position of deserts, passes and swamps. The railroad of to-day follows the trail of the primitive man, and the rivers have ever been the natural highways of nations. The theories of Morgan therefore remain true as theories; only in their application he fell into an error which was natural enough to the science of twenty years ago. Perhaps when twenty years more shall have elapsed, the post-tertiary geology of our continent will have been so clearly defined that the geography of its different epochs will be known sufficiently to trace these lines of migration at the various epochs of man’s residence in the western world, from his first arrival.

I have now set before you, in a superficial manner it is true, the various sources from which we may derive aid in establishing the pre-historic chronology of America. I have also endeavored, to a limited extent, to express myself as to the relative value of these sources. None of them can be neglected, and it will be only from an exhaustive study of them all that we can expect to solve the numerous knotty problems, and lift the veil which hangs so darkly on all that concerns the existence of the American race before the sixteenth century.

We are merely beginning the enormous labor which is before us; we have yet to discover the methods by which we can analyze fruitfully the facts we already know. But I look forward with the utmost confidence to a rich return from such investigations. The day is coming, and that rapidly, when the pre-historic life of man in both the New and the Old World will be revealed to us in a thousand unexpected details. We have but to turn backward about thirty years to reach a time when the science of pre-historic archÆology was unknown, and its early gropings were jeered at as absurdities. Already it has established for itself a position in the first rank of the sciences which have to do with the highest of problems. It has cast a light upon the pathway of the human race from the time that man first deserved his name down to the commencement of recorded history. Its conquests are but beginning. Year by year masses of new facts are brought to knowledge from unexpected quarters, current errors are corrected, and novel methods of exploration devised.

As Americans by adoption, it should be our first interest and duty to study the Americans by race, in both their present and past development. The task is long and the opportunity is fleeting. A century more, and the anthropologist will scarcely find a native of pure blood; the tribes and languages of to-day will have been extinguished or corrupted. Nor will the archÆologist be in better case. Every day the progress of civilization, ruthless of the monuments of barbarism, is destroying the feeble vestiges of the ancient race; mounds are levelled, embankments disappear, the stones of temples are built into factories, the holy places desecrated. We have assembled here to aid in recovering something from this wreck of a race and its monuments: let me urge upon you all the need of prompt action and earnest work, inasmuch as the opportunities we enjoy will never again present themselves in such fulness.

There has been much talk in scientific circles lately about PalÆoliths, and much misunderstanding about them. Let me try to explain in a few words what they are, what they tell, and what mistakes people make about them.

Since man first appeared on this planet, his history has been a slow progress from the most rudimentary arts up to those which he now possesses. We know this, because in a given locality those remains of his art which are found undisturbed in strata geologically the oldest are always the rudest. The exceptions to this rule are in appearance only, as for instance when a given locality was not occupied by men until they had already acquired considerable knowledge of arts, or when a cultivated nation was overrun by a barbarous one.

The general line of advance I have indicated shows, wherever we can trace it, many similarities—similarities not necessarily dependent on an ancient intercourse, but simply because primitive man felt everywhere the same wants, and satisfied them in pretty much the same manner. He felt the need of defence and attack, and everywhere a stick and a stone offered themselves as the handiest and most effective weapons; he used both wherever he was, and adapted them to like shapes.

In casting about for some standard wherewith to measure the long progress from this simple beginning to the present day, antiquaries have hit upon a very excellent one—the choice of a material employed at any given epoch for obtaining a cutting edge—for manufacturing l’instrument tranchant. Man conquers nature as he does his enemy—by cutting her down. The world at present uses iron, or its next product steel, for that purpose; before it came into vogue many nations employed bronze; but in the earliest periods of man’s history, and to-day in some savage tribes, stone was the substance almost exclusively wrought for this purpose. These distinctions divide the progress of man into the three great periods; the Age of Iron, the Age of Bronze, and the Age of Stone.

Do not make the mistake of supposing that the remains of human art reveal this sequence in every locality; I have already hinted that this is not the case. And do not make that other mistake of supposing that all three are found in chronologic sequence over the whole world. On the contrary, they are synchronous even to-day, as there are now tribes in Brazil in the Age of Stone and nations in Asia in the Age of Bronze. The word “Age” in this connection does not mean a definite period of time, but a recognized condition of art.

In Western Europe, however, where these terms originated, the three Ages were chronologic. Previous to about two thousand years before the Christian era, all the nations in that region employed stone exclusively to manufacture their cutting implements; later, bronze was preferred for the same purpose; and still later, iron. I say “preferred,” for do not imagine that the implement of stone or of bronze was straightway discarded when the better material was learned. We know that stone battle-axes were used in Ireland and Germany down to the tenth century, and bronze was employed by Romans and Egyptians long after they became acquainted with iron.

Each of these three Ages has various subdivisions. Those of the Age of Stone are particularly important. They are two, based upon the manner in which the stone was brought to an edge. All the specimens in geologically the oldest deposits have been brought to an edge by a process of chipping off small pieces, so as to produce a sharp line or crest on a part or the whole of the border of the stone. This artificial process leaves such peculiar traces that a practiced eye cannot confound it with any accidental chipping which natural means effect.

The later deposits of the Age of Stone show that the early workmen had acquired another manner of dressing their material; they rubbed one stone against another, thus grinding it down to a sharp polished edge.

These two methods give the names to the two periods of the Age of Stone, the Period of Chipped Stone and the Period of Polished Stone. Do not suppose, however, that the workmen in polished stone forgot the art of chipping stone. On the contrary, they continued it side by side with their new learning, and you will find on the sites of their workshops plenty of stone implements in form and technical production like the chipped implements of the older period.

We know that the polished or ground-stone implements came into use later than the earliest chipped implements, for in the oldest beds the latter are found exclusively. Hence the time when they were used exclusively is called the older stone implement period or the PalÆolithic period; while, the time when both chipped and polished stones were used, metals were yet unknown, is named the newer stone implement period, or the Neolithic period. A true “PalÆolith” is a typical chipped stone implement, the position of which when found leads us to believe that it was manufactured in the older of these periods.

We are not entirely dependent on its position to decide its antiquity. The kind of stone it is, the amount of weather-wearing or patine it shows, certain characteristics of shape and size, the indication that the chipping was done in a peculiar manner, all these aid the skilled observer in pronouncing definitely as to whether it is a true PalÆolith.

Nor is position always a guarantee of antiquity. A genuine PalÆolith may have been washed into newer strata, or be exposed by natural agencies on the surface of the ground, and in such cases it may not be possible to distinguish it from the products of Neolithic industry. A recent product of art may have sunk or been buried in an ancient stratum, and thus become what is termed an “intrusive deposit.”

The PalÆolithic period itself is advantageously subdivided further into two Epochs, an earlier one in which men made “simple” implements only, and a later one in which they manufactured “compound” implements as well. I was the first to point out this distinction, and as I have found it really useful, and as others have also expressed to me the value which it has been to them in this line of research, I will explain it further.^[33] A “compound” implement is one composed of several parts adapted to each other, as the bow and the arrow, the spear with its shaft and blade, or the axe with its head and helve and the means of fastening the one to the other. These were not early acquisitions. During long ages man contented himself with such tools or weapons as he could frame of a single piece of wood or stone, simply holding it in his hand. When he found he could increase its effectiveness by fitting it to a handle, the discovery marked an era in his culture.

He may indeed in his rudest ages have lashed a stone to the end of his club, or have inserted a spall of flint in the split end of a stick; but these are not compound implements in the proper sense of the term. The expression means an art-product which clearly shows that it was but one part of a mechanical apparatus. The arrow-head with its stem, barbs and body, the stone axe with its grooves or drilled perforation for the handle, are incomplete in themselves, they disclose a preconceived plan for the adjustment of parts which man in his earliest and rudest condition does not seem to have possessed. The most ancient strata in which the remains of human art have been found, either in Europe or America, yield “simple” implements only; “compound” implements are a conquest of his inventive faculty at a later date.

So far as America is concerned it is probable that the oldest remains of man yet discovered on the northern continent have been those exhumed in the valley of the Delaware River, in the states of Pennsylvania, New Jersey and Delaware. According to the most careful geological observers that large deposit of gravel covering about five thousand acres on both banks of the river below Trenton is a post-glacial deposit not less than twelve or fifteen thousand years old. Imbedded in this at various depths a large number of true palÆoliths have been discovered by Dr. C. C. Abbott, Professor F. A. Putnam, myself and others. Every one of them so far as I am aware belongs to the class of “simple” implements, not an arrowhead nor grooved axe nor stemmed scraper having been reported.

Another deposit of gravel further down the Delaware River is much older: The best authorities in such matters believe that it was deposited, not after the recession of the great glacier which once covered Canada and the northern portion of the United States, but while that tremendous phenomena was at its height, and when all the streams of the central United States were periodically choked with vast masses of ice and snow. In this, which is called the Columbian gravel, chipped stone implements have been found by Mr. Cresson, all of the “simple” variety, and at such depths as to preclude the theory of an intrusive deposit. These discoveries carry the age of the appearance of man in the Delaware valley back to a date which is possibly over a hundred thousand years ago.

The great glacier left its mass of boulders, pebbles and broken stone, which it pushed before it, or carried with it, in a long line of so-called “moraines,” extending, roughly speaking, from New York to St. Louis. In this mass, at its edges where the great wash from the melting ice poured down, palÆoliths have been found in undisturbed position, proving that also there man had struggled with the inclemency of the ice-age, and, poorly provided as he was, had come out victorious. Here too all the implements he left are of the “simple” type, indicating at once the vast antiquity of the period and the presence of a race substantially the same as that to the east at the same date.

No tribe has been known to history which was confined to the knowledge of “simple” implements, or which manufactured stone implements exclusively in the PalÆolithic forms. Wherever, therefore, these are found without the admixture of artificially ground or polished stones we may be sure we face the remains of a time whose antiquity cannot be measured by any chronology applied to the historic records of humanity.

This enables us in a measure to define the limits of the region known to the human race at this, its earliest epoch; with our present deficient knowledge we can do so only partially and by exclusion. It is safe to state that in Europe PalÆolithic man did not occupy the central alpine area of Switzerland and its surroundings, nor the plains of Russia, nor any part of the Scandinavian peninsula, Scotland, Ireland, nor Iceland. In North America he had no habitations north of the forty-first parallel of latitude except perhaps close to the shores of the two great oceans;^[34] it is not probable that his foot pressed the soil of any of the West Indian Islands; but when the great Austral Glacier was in its recession depositing the fertile loam of the pampas of Buenos Ayres human beings with their rude PalÆoliths were following up the retreating line of ice, as in the Northern Hemisphere. Ages uncounted and uncountable have passed since then, but man has left indestructible evidences that even in that early morn of his existence he had explored and conquered that continent which a late generation has chosen to call “the New World.”

Were the question I am about to discuss one of merely theoretical bearings, I should not approach it; but the widespread belief that the American tribes are genealogically connected with the Mongolians is constantly directing and coloring the studies of many Americanists, very much as did at one time the belief that the red men are the present representatives of the ten lost tribes of Israel. It is practically worth while, therefore, to examine the grounds on which the American race is classed by these anthropologists as a branch of the Mongolian, and to inquire whether the ancient culture of America betrayed any positive signs of Mongolian influence.

You will permit me to avoid the discussion as to what constitutes races in anthropology. To me they are zoÖlogical sub-species, marked by fixed and correlated characteristics, impressed so firmly that they have suffered no appreciable alteration within the historic period either through time or environment. In this sense, Blumenbach, in the last century, recognized five races, corresponding to the five great land-areas of the globe, and to their characteristic faunal and floral centres. This division was an eminently scientific one, and still remains the most in accord with anatomical and linguistic research. About twenty years after the appearance of Blumenbach’s work, however, the eminent naturalist Cuvier published his great work on “The Animal Kingdom,” in which he rejected Blumenbach’s classification, and proposed one dividing the human species into three races,—the white or Caucasian, the black or Ethiopian, and the yellow or Mongolian. In the latter he included the Malays and the American Indians.

This triple division has been very popular in France, and to some extent in other countries. It is not, and it was not in its inception, a scientific deduction from observed facts, but was a sort of a priori hypothesis based on the physiological theories of Bichat, and at a later day derived support from the philosophic dreams of Auguste Comte. Bichat, for instance, had recognized three fundamental physiological systems in man—the vegetative or visceral, the osso-muscular, and the cerebro-spinal. The anthropologists, in turn, considered it a happy thought to divide the human species into three races, each of which should show the predominance of one or other of these systems. Thus the black race was to show the predominance of the vegetative system; the yellow race, the osso-muscular system; the white race, the nervous system.^[36] As Bichat had not discovered any more physiological systems, so there could be no more human races on the earth: and thus the sacred triplets of the Comtian philosophy could be vindicated.

How little value attaches to any such generalizations you will readily perceive, and you will be prepared, with me, to dismiss them all, and to turn to the facts of the case, inquiring whether there are any traits of the red race which justify their being called “Mongolian” or “Mongoloid.”

Such affinities have been asserted to exist in language, in culture, and in physical peculiarities, and I shall take these up seriatim for examination.

First, as to language.

The great Mongolian stock is divided into the southern branch, speaking monosyllabic, isolating languages, and the northern branch, whose dialects are polysyllabic and agglutinating. The latter are sometimes called Turanian or Ural-Altaic; and as they are geographically contiguous to the Eskimo, and almost to the Athabascans, we might reasonably expect the linguistic kinship, if any exists, to be shown in this branch of Mongol speech. Is such the case? Not in the least. To prove it, I think it enough to quote the positive statement of the best European authority on the Ural-Altaic languages, Dr. Heinrich Winkler. He emphatically says, that, in the present state of linguistic science, not only is there no connection apparent between any Ural-Altaic and any American language, but that such connection is shown to be highly improbable. The evidence is all the other way.^[37]

I need not, therefore, delay over this part of my subject, but will proceed to inquire whether there are any American affinities to the monosyllabic, isolating languages of Asia.

There is one prominent example, which has often been put forward, of a supposed monosyllabic American language; and its relationship to the Chinese has frequently been asserted—a relationship, it has been said, extending both to its vocabulary and its grammar. This is the Otomi, spoken in and near the valley of Mexico. It requires, however, but a brief analysis of the Otomi to see that it is not a monosyllabic language in the linguistic sense, and that in its sentence-building it is incorporative and polysynthetic, like the great majority of American tongues, and totally unlike the Chinese. I may refer to my own published study of the Otomi, and to that of the Count de Charencey, as proving what I say.^[38]

Some have thought that the Maya of Yucatan has in its vocabulary a certain number of Chinese elements; but all these can readily be explained on the doctrine of coincidences. The Mexican antiquary Mendoza has marshalled far more coincidences of like character and equal worth to show that the Nahuatl is an Aryan dialect descended from the Sanscrit.^[39] In fine, any, even the remotest, linguistic connection between American and Mongolian languages has yet to be shown; and any linguist who considers the radically diverse genius of the two groups of tongues will not expect to find such relationship.

I shall not detain you long with arguments touching supposed Mongolian elements of culture in ancient America. Any one at all intimately conversant with the progress of American archÆology in the last twenty years must see how rapidly has grown the conviction that American culture was homebred, to the manor born: that it was wholly indigenous and had borrowed nothing—nothing, from either Europe, Asia, or Africa. The peculiarities of native American culture are typical, and extend throughout the continent. Mr. Lewis Morgan was perfectly right in the general outline of his theory to this effect, though, like all persons enamored of a theory, he carried it too far.

This typical, racial American culture is as far as possible, in spirit and form, from the Mongolian. Compare the rich theology of Mexico or Peru with the barren myths of China. The theory of governments, the method of house-construction, the position of woman, the art of war,^[40] are all equally diverse, equally un-Mongolian. It is useless to bring up single art-products or devices, such as the calendar, and lay stress on certain similarities. The doctrine of the parallelism of human development explains far more satisfactorily all these coincidences. The sooner that Americanists generally, and especially those in Europe, recognize the absolute autochthony of native American culture, the more valuable will their studies become.

It is no longer in season to quote the opinions of Alexander von Humboldt and his contemporaries on this subject, as I see is done in some recent works. The science of archÆology has virtually come into being since they wrote, and we now know that the development of human culture is governed by laws with which they were unacquainted. Civilization sprang up in certain centres in both continents, widely remote from each other; but, as the conditions of its origin were everywhere the same, its early products were much alike.

It is evident from what I have said, that the asserted Mongolian or Mongoloid connection of the American race finds no support either from linguistics or the history of culture. If anywhere, it must be in physical resemblances. In fact, it has been mainly from these that the arguments have been drawn. Let us examine them.

Cuvier, who, as I have said, is responsible for the confusion of the American with the Mongolian race, based his racial scheme on the color of the skin, and included the American within the limits of the yellow race. Cuvier had seen very few pure Mongolians, and perhaps no pure-blooded Americans; otherwise he would not have maintained that the hue of the latter is yellow. Certainly it is not. You may call it reddish, or coppery, or cinnamon, or burnt sugar, but you cannot call it yellow. Some individuals or small tribes may approach the peculiar dusky olive of the Chinaman, but so do some of the European peoples of Aryan descent; and there are not wanting anthropologists who maintain that the Aryans are also Mongoloid. The one position is just as defensible as the other on the ground of color.

Several of the most prominent classifications of mankind are based upon the character of the hair; the three great divisions being, as you know, into the straight, the curly, and the woolly haired varieties. These external features of the hair depend upon the form of the individual hairs as seen in cross-section. The nearer this approaches a circle, the straighter is the hair. It is true that both Mongolians and Americans belong to the straight haired varieties; but of the two, the American has the straighter hair, that whose cross-section comes nearer to a perfect circle. So that by all the rules of terminology and logic, if we are to call either branch a variation from the other, we should say that the Mongol is a variety of the American race, and call it “Americanoid,” instead of vice versa.

The color of the hair of the two races is, moreover, distinctly different. Although superficially both seem black, yet, observed carefully by reflected light, it is seen that the ground-tone of the Mongolian is bluish, while that of the American is reddish.

Of positive cranial characteristics of the red race, I call attention to the interparietal bone (the os IncÆ), which is found in its extreme development in the American, in its greatest rarity among the Mongolians; also to the form of the glabella, found most prominent in American crania, least prominent in Altaic or northern Mongoloid crania; and the peculiar American characteristics of the occipital bone, flattened externally, and internally presenting in nearly forty per cent. of cases the “Aymarian depression,” as it has been termed, instead of the internal occipital protuberance.^[41]

The shape of the skull has been made another ground of race-distinction; and, although we have learned of late years that its value was greatly over-estimated by the earlier craniologists, we have also learned that in the average, and throughout large numbers of peoples, it is a very persistent characteristic, and one potently indicative of descent or relationship. Now, of all the peoples of the world, the Mongols, especially the Turanian branch, are the most brachycephalic; they have the roundest heads; and it is in a high degree noteworthy that precisely the American nation dwelling nearest to these, having undoubted contact with them for unnumbered generations, are long-headed, or dolichocephalic, in a marked degree. I mean the Eskimo, and I cannot but be surprised that such an eminent anthropologist as Virchow,^[42] in spite of this anatomical fact, and in defiance of the linguistic evidence, should have repeated the assertion that the Eskimo are of Mongolian descent.

Throughout the American continent generally, the natives were not markedly brachycephalic. This was abundantly illustrated more than twenty years ago by the late Prof. James Aitkins Meigs, in his “Observations on the Cranial Forms of the American Aborigines.” They certainly, in this respect, show no greater Mongoloid affinities than do their white successors on the soil of the United States.

If color, hair, and crania are thus shown to present such feeble similarities, what is it that has given rise to a notion of the Mongoloid origin of the American Indian? Is it the so-called Mongolian eye, the oblique eye, with a seeming droop at its inner canthus? Yes, a good deal has been made of this by certain writers, especially by travellers who are not anatomists. The distinguished ethnologist Topinard says the Chinese are very often found without it, and I can confirm this opinion by those I have seen in this country. It is, indeed, a slight deformity, affecting the skin of the eyebrow only, and is not at all infrequent in the white race. Surgeons know it under the name epicanthus, and, as with us it is considered a disfigurement, it is usually removed in infancy by a slight operation. In a few American tribes it is rather prevalent, but in most of the pure Indians I have seen, no trace of it was visible. It certainly does not rank as a racial characteristic.^[43]

The nasal index has been recommended by some anatomists as one of the most persistent and trustworthy of racial indications. The Mongolian origin of the red race derives faint support from this quarter. From the measurements given in the last edition of Topinard’s work^[44] the Mongolian index is 80, while that of the Eskimo and tribes of the United States and Canada, as far as observed, is 70, that of the average Parisian of to-day being 69 (omitting fractions). According to this test, the American is much closer to the white than to the yellow race.

Most of the writers (for instance, AvÉ-Lallemant, St. Hilaire, Peschel, and Virchow) who have argued for the Mongoloid character of the Americans, have quoted some one tribe which, it is asserted, shows marked Chinese traits. This has especially been said of the natives of three localities,—the Eskimo, the tribes of the North Pacific coast, and the Botocudos of Brazil. So far as the last-mentioned are concerned, the Botocudos, any such similarity has been categorically denied by the latest and most scientific traveller who has visited them, Dr. Paul Ehrenreich. It is enough if I refer you to his paper in the Zeitschrift fÜr Ethnologie for 1887, where he dismisses, I should say once for all, the notion of any such resemblance existing. I have already pointed out that the Eskimo are totally un-Mongolian in cranial shape, in nasal index, and in linguistic character. They do possess in some instances a general physiognomical similarity, and this is all; and this is not worth much, as against the dissimilarities mentioned. The same is true of the differences and similarities of some tribes of the north-west coast. In estimating the value of resemblances observed in this part of our continent, we should remember that we have sufficient evidence to believe that for many generations some slight intercourse has been going on between the adjacent mainlands and islands of the two continents in the regions of their nearest proximity. The same train of events led to a blending of the negro and the white races along the shores of the Red Sea; but any one who recognizes the distinction of races at all—and I am aware that certain eccentric anthropologists do not—will not, on that account, claim that the white race is negroid. With just as little reason, it seems to me, has it been argued that the native Americans as a race are Mongoloid.^[45]

An acute philosophical writer has stated that the superficial observer is apt to be impressed with the similarities of objects; while the profounder student finds his attention more profitably attracted to their differences. By this maxim we may explain this theory of the affinities of the American race as well as many another which has been broached.

[The following Essay is reprinted without alteration. It appeared in the American Antiquarian for October, 1881, and has a certain degree of historic value as illustrating the progress of archÆologic study in the United States. It is, I believe, the first reasoned argument that the constructors of the mounds of the Ohio Valley were the ancestors of tribes known and resident not remote from the sites of these ancient works. Though this opinion has not yet been fully accepted, the tendency of later studies is unquestionably in its favor.]

The question, Who were the Mound-builders? is one that still remains open in American archÆology. Among the most recent expressions of opinion I may quote Prof. John T. Short, who thinks that one or two thousand years may have elapsed since they deserted the Ohio valley, and probably eight hundred since they finally retired from the Gulf coast.^[46] Mr. J. P. MacLean continues to believe them to have been somehow related to the “Toltecs.”^[47] Dr. J. W. Foster, making a tremendous leap, connects them with a tribe “who, in times far remote, flourished in Brazil,” and adds: “a broad chasm is to be spanned before we can link the Mound-builders to the North American Indians. They were essentially different in their form of government, their habits and their daily pursuits. The latter were never known to erect structures which should survive the lapse of a generation.”^[48]

On the other hand, we have the recent utterance of so able an ethnologist as Major J. W. Powell to the effect that, “With regard to the mounds so widely scattered between the two oceans, it may be said that mound-building tribes were known in the early history of discovery of this continent, and that the vestiges of art discovered do not excel in any respect the arts of the Indian tribes known to history. There is, therefore, no reason for us to search for an extralimital origin through lost tribes for the arts discovered in the mounds of North America.”^[49]

Between opinions so discrepant the student in archÆology may well be at a loss, and it will therefore be worth while to inquire just how far the tribes who inhabited the Mississippi valley and the Atlantic slope at the time of the discovery were accustomed to heap up mounds, excavate trenches, or in other ways leave upon the soil permanent marks of their occupancy.

Beginning with the warlike northern invaders, the Iroquois, it clearly appears that they were accustomed to construct burial mounds. Colden states that the corpse was placed in a large round hole and that “they then raise the Earth in a round Hill over it.”^[50] Further particulars are given by Lafitau: the grave was lined with bark, and the body roofed in with bark and branches in the shape of an arch, which was then covered with earth and stones so as to form an agger or tumulus.^[51] In these instances the mound was erected over a single corpse; but it was also the custom among the Hurons and Iroquois, as we are informed by Charlevoix, to collect the bones of their dead every ten years, and inter them in one mass together.^[52] The slain in a battle were also collected into one place and a large mound heaped over them, as is stated by Mr. Paul Kane,^[53] and that such was an ancient custom of the Iroquois tribes, is further shown by a tradition handed down from the last century, according to which the Iroquois believed that the Ohio mounds were the memorials of a war which in ancient times they waged with the Cherokees.^[54] Mr. E. G. Squier, who carefully examined many of the earthworks in the country of the ancient Iroquois, was inclined at first to suppose the remains he found there were parts of “a system of defence extending from the source of the Allegheny and Susquehanna in New York, diagonally across the country through central and northern Ohio to the Wabash,” and hence drew the inference that “the pressure of hostilities [upon the mound-builders] was from the north-east.”^[55] This opinion has been repeated by some recent writers; but Mr. Squier himself substantially retracted it in a later work, and reached the conviction that whatever ancient remains there are in Western New York and Pennsylvania are to be attributed to the later Indian tribes and not to the Mound-builders.^[56]

The neighbors of the Iroquois, the various Algonkin tribes, were occasionally constructors of mounds. In comparatively recent times we have a description of a “victory mound” raised by the Chippeways after a successful encounter with the Sioux. The women and children threw up the adjacent surface soil into a heap about five feet high and eight or ten feet in diameter, upon which a pole was erected, and to it tufts of grass were hung, one for each scalp taken.^[57]

Robert Beverly, in his History of Virginia, first published in 1705, describes some curious constructions by the tribes there located. He tells us that they erected “pyramids and columns” of stone, which they painted and decorated with wampum, and paid them a sort of worship. They also constructed stone altars on which to offer sacrifices.^[58] This adoration of stones and masses of rocks—or rather of the genius which was supposed to reside in them—prevailed also in Massachusetts and other Algonkin localities, and easily led to erecting such piles.^[59]

Another occasion for mound-building among the Virginian Indians was to celebrate or make a memorial of a solemn treaty. On such an occasion they performed the time honored ceremony of “burying the hatchet,” a tomahawk being literally put in the ground, “and they raise a pile of stones over it, as the Jews did over the body of Absalom.”^[60]

I am not aware of any evidence that the Cherokees were mound-builders: but they appreciated the conveniences of such structures, and in one of their villages William Bartram found their council house situated on a large mound. He adds: “But it may be proper to observe that this mount on which the rotunda stands is of a much ancienter date than the building, and perhaps was raised for another purpose.”^[61] Lieutenant Timberlake is about our best early authority on the Cherokees, and I believe he nowhere mentions that they built upon mounds of artificial construction. Adair, however, states that they were accustomed to heap up and add to piles of loose stones in memory of a departed chief, or as monuments of important events.^[62]

The tribes who inhabited what we now call the Gulf States, embracing the region between the eastern border of Texas and the Atlantic Ocean south of the Savannah River, belonged, with few and small exceptions, to the great Chahta-Muskokee family, embracing the tribes known as Choctaws, Chikasaws, Muskokees or Creeks, Seminoles, Allibamons, Natchez and others. The languages of all these have numerous and unmistakable affinities, the Choctaw or Chahta presenting probably the most archaic form. It is among them, if anywhere within our limits, that we must look for the descendants of the mysterious “Mound-builders.” No other tribes can approach them in claims for this distinction. Their own traditions, it is true, do not point to a migration from the north, but from the west; nor do they contain any reference to the construction of the great works in question; but these people seem to have been a building race, and to have reared tumuli not contemptible in comparison even with the mightiest of the Ohio Valley.

The first explorer who has left us an account of his journey in this region was Cabeza de Vaca, who accompanied the exposition of Pamfilo de Narvaez in 1527. He, however, kept close to the coast for fear of losing his way, and saw for the most part only the inferior fishing tribes. These he describes as generally in a miserable condition. Their huts were of mats erected on piles of oyster shells (the shell heaps now so frequent along the southern coast). Yet he mentions that in one part, which I judge to be somewhere in Louisiana, the natives were accustomed to erect their dwellings on steep hills and around their base to dig a ditch, as a means of defence.^[63]

Our next authorities are very important. They are the narrators of Captain Hernando de Soto’s famous and ill starred expedition. Of this we have the brief account of Biedma, the longer story of “the gentleman of Elvas,” a Portuguese soldier of fortune, intelligent and clear-headed, and the poetical and brilliant composition of Garcilasso de la Vega. In all of these we find the southern tribes described as constructing artificial mounds, using earthworks for defence, excavating ditches and canals, etc. I quote the following passage in illustration:

“The town and the house of the Cacique Ossachile are like those of the other caciques in Florida. *** The Indians try to place their villages on elevated sites; but inasmuch as in Florida there are not many sites of this kind where they can conveniently build, they erect elevations themselves in the following manner: They select the spot and carry there a quantity of earth which they form into a kind of platform two or three pikes in height, the summit of which is large enough to give room for twelve, fifteen or twenty houses, to lodge the cacique and his attendants. At the foot of this elevation they mark out a square place according to the size of the village, around which the leading men have their houses. *** To ascend the elevation they have a straight passage way from bottom to top, fifteen or twenty feet wide. Here steps are made by massive beams, and others are planted firmly in the ground to serve as walls. On all other sides of the platform, the sides are cut steep.”^[64]

Later on La Vega describes the village of Capaha:

“This village is situated on a small hill, and it has about five hundred good houses, surrounded with a ditch ten or twelve cubits (brazas) deep, and a width of fifty paces in most places, in others forty. The ditch is filled with water from a canal which has been cut from the town to Chicagua. The canal is three leagues in length, at least a pike in depth, and so wide that two large boats could easily ascend or descend it, side by side. The ditch which is filled with water from this canal surrounds the town except in one spot, which is closed by heavy beams planted in the earth.”^[65]

Biedma remarks in one passage, speaking of the provinces of Ycasqui and Pacaha: “The caciques of this region were accustomed to erect near the house where they lived very high mounds (tertres trÈs-elevÈes), and there were some who placed their houses on the top of these mounds.”^[66]

I cannot state precisely where these provinces and towns were situated; the successful tracing of De Soto’s journey has never yet been accomplished, but remains as an interesting problem for future antiquaries to solve. One thing I think is certain; that until he crossed the Mississippi he at no time was outside the limits of the wide spread Chahta-Muskokee tribes. The proper names preserved, and the courses and distances given, both confirm this opinion. We find them therefore in his time accustomed to erect lofty mounds, terraces and platforms, and to protect their villages by extensive circumvallations. I shall proceed to inquire whether such statements are supported by later writers.

Our next authorities in point of time are the French Huguenots, who undertook to make a settlement on the St. John River near where St. Augustine now stands in Florida. The short and sad history of this colony is familiar to all. The colonists have, however, left us some interesting descriptions of the aborigines. In the neighborhood of St. Augustine these belonged to the Timuquana tribe, specimens of whose language have been preserved to us, but which, according to the careful analysis recently published by Mr. A. S. Gatschet,^[67] has no relationship with the Chahta-Muskokee, nor, for that matter, with any other known tongue. Throughout the rest of the peninsula a Muskokee dialect probably prevailed.

The “Portuguese gentleman” tells us that at the very spot where De Soto landed, generally supposed to be somewhere about Tampa Bay, at a town called Ucita, the house of the chief “stood near the shore upon a very high mound made by hand for strength.” Such mounds are also spoken of by the Huguenot explorers. They served as the site of the chieftain’s house in the villages, and from them led a broad, smooth road through the village to the water.^[68] These descriptions correspond closely to those of the remains which the botanists, John and William Bartram, discovered and reported about a century ago.

It would also appear that the natives of the peninsula erected mounds over their dead, as memorials. Thus the artist Le Moyne de Morgues, writes: “Defuncto aliquo rege ejus proviciÆ, magna solemnitate sepelitur, et ejus tumulo crater, e quo bibere solebat, imponitur, defixis circum ipsum tumulum multis sagittis.”^[69] The picture he gives of the “tumulus” does not represent it as more than three or four feet in height; so that if this was intended as an accurate representation, the structure scarcely rises to the dignity of a mound.

After the destruction of the Huguenot colony in 1565, the Spanish priests at once went to work to plant their missions. The Jesuit fathers established themselves at various points south of the Savannah River, but their narratives, which have been preserved in full in a historic work of great rarity, describe the natives as broken up into small clans, waging constant wars, leading vagrant lives, and without fixed habitations.^[70] Of these same tribes, however, Richard Blomes, an English traveler, who visited them about a century later, says that they erected piles or pyramids of stones, on the occasion of a successful conflict, or when they founded a new village, for the purpose of keeping the fact in long remembrance.^[71] About the same time another English traveler, by name Bristock, claimed to have visited the interior of the country and to have found in “Apalacha” a half-civilized nation, who constructed stone walls and had a developed sun worship; but in a discussion of the authenticity of his alleged narrative I have elsewhere shown that it cannot be relied upon, and is largely a fabrication.^[72] A correct estimate of the constructive powers of the Creeks is given by the botanist, William Bartram, who visited them twice in the latter half of the last century. He found they had “chunk yards” surrounded by low walls of earth, at one end of which, sometimes on a moderate artificial elevation, was the chief’s dwelling and at the other end the public council house.^[73] His descriptions resemble so closely those in La Vega that evidently the latter was describing the same objects on a larger scale—or from magnified reports.

Within the present century the Seminoles of Florida are said to have retained the custom of collecting the slain after a battle and interring them in one large mound. The writer on whose authority I state this, adds that he “observed on the road from St. Augustine to Tomaka, one mound which must have covered two acres of ground,”^[74] but this must surely have been a communal burial mound.

Passing to the tribes nearer the Mississippi, most of them of Choctaw affiliation, we find considerable testimony in the French writers to their use of mounds. Thus M. de la Harpe says: “The cabins of the Yasous, Courous, Offogoula and Ouspie are dispersed over the country on mounds of earth made with their own hands.”^[75] The Natchez were mostly of Choctaw lineage. In one of their villages Dumont notes that the cabin of the chief was elevated on a mound.^[76] Father Le Petit, a missionary who labored among them, gives the particulars that the residence of the great chief or “Brother of the Sun,” as he was called, was erected on a mound (butte) of earth carried for that purpose. When the chief died, the house was destroyed, and the same mound was not used as the site of the mansion of his successor, but was left vacant and a new one was constructed.^[77] This interesting fact goes to explain the great number of mounds in some localities; and it also teaches us the important truth that we cannot form any correct estimate of the date when a mound-building tribe left a locality by counting the rings in trees, etc., because long before they departed, certain tumuli or earthworks may have been deserted and tabooed from superstitious notions, just as many were among the Natchez.

We have the size of the Natchez mounds given approximately by M. Le Page du Pratz. He observes that the one on which was the house of the Great Sun was “about eight feet high and twenty feet over on the surface.”^[78] He adds that their temple, in which the perpetual fire was kept burning, was on a mound about the same height.

The custom of communal burial has been adverted to. At the time of the discovery it appears to have prevailed in most of the tribes from the Great Lakes to the Gulf. The bones of each phratry or gens—the former, probably—were collected every eight or ten years and conveyed to the spot where they were to be finally interred. A mound was raised over them which gradually increased in size with each additional interment. The particulars of this method of burial have often been described, and it is enough that I refer to a few authorities in the note.^[79] Indeed it has not been pretended that such mounds necessarily date back to a race anterior to that which occupied the soil at the advent of the white man.

I have not included in the above survey the important Dakota stock who once occupied an extended territory on the upper Mississippi and its affluents, and scattered clans of whom were resident on the Atlantic Coast in Virginia and Carolina. But, in fact, I have nowhere found that they erected earthworks of any pretentions whatever.

From what I have collected, therefore, it would appear that the only resident Indians at the time of the discovery who showed any evidence of mound-building comparable to that found in the Ohio valley were the Chahta-Muskokees. I believe that the evidence is sufficient to justify us in accepting this race as the constructors of all those extensive mounds, terraces, platforms, artificial lakes and circumvallations which are scattered over the Gulf States, Georgia and Florida. The earliest explorers distinctly state that such were used and constructed by these nations in the sixteenth century, and probably had been for many generations. Such too, is the opinion arrived at by Col. C. C. Jones, than whom no one is more competent to speak with authority on this point. Referring to the earthworks found in Georgia he writes: “We do not concur in the opinion so often expressed that the mound-builders were a race distinct from and superior in art, government, and religion, to the Southern Indians of the fifteenth and sixteenth centuries.”

It is a Baconian rule which holds good in every department of science that the simplest explanation of a given fact or series of facts should always be accepted; therefore if we can point out a well known race of Indians who, at the time of the discovery, raised mounds and other earthworks, not wholly dissimilar in character and not much inferior in size to those in the Ohio valley, and who resided not very far away from that region and directly in the line which the Mound-Builders are believed by all to have followed in their emigration, then this rule constrains us to accept for the present this race as the most probable descendants of the Mound Tribes, and seek no further for Toltecs, Asiatics or Brazilians. All these conditions are filled by the Chahta tribes.^[80]

It is true, as I have already said, that the traditions of their own origin do not point to the north but rather to the west or northwest; but in one of these traditions it is noticeable that they claim their origin to have been from a large artificial mound, the celebrated Nanih Waiya, the Sloping Hill, an immense pile in the valley of the Big Black River;^[81] and it may be that this is a vague reminiscence of their remote migration from their majestic works in the north.

The size of the southern mounds is often worthy of the descendants of those who raised the vast piles in the northern valleys. Thus one in the Etowah Valley, Georgia, has a cubical capacity of 1,000,000, cubic feet.^[82] The Messier Mound, near the Chatahoochee River, contains about 700,000 cubic feet.^[83] Wholly artificial mounds 50 to 70 feet in height, with base areas of about 200 by 400 feet, are by no means unusual in the river valleys of the Gulf States.

With these figures we may compare the dimensions of the northern mounds. The massive one near Miamisburg, Ohio, 68 feet high, has been calculated to contain 311,350 cubic feet—about half the size of the Messier Mound. At Clark’s Works, Ohio, the embankments and mounds together contain about 3,000,000 cubic feet;^[84] but as the embankment is three miles long, most of this is not in the mounds themselves. Greater than any of these is the truncated pyramid at Cahokia, Illinois, which has an altitude of 90 feet and a base area of 700 by 500 feet. It is, however, doubtful whether this is wholly an artificial construction. Professor Spencer Smith has shown that the once famous “big mound” of St. Louis was largely a natural formation; and he expresses the opinion that many of the mounds in Missouri and Illinois, popularly supposed to be artificial constructions, are wholly, or in great part, of geologic origin.^[85] There is apparently therefore no such great difference between the earth structures of the Chahta tribes, and those left us by the more northern mound-builders, that we need suppose for the latter any material superiority in culture over the former when first they became known to the whites; nor is there any improbability in assuming that the Mound-builders of the Ohio were in fact the progenitors of the Chahta tribes, and were driven south probably about three or four hundred years before the discovery. Such is the conviction to which the above reasoning leads us.

In the course of it, I have said nothing about the condition of the arts of the Mound-builders compared with that of the early southern Indians; nor have I spoken of their supposed peculiar religious beliefs which a recent writer thinks to point to “Toltec” connections;^[86] nor have I discussed the comparative craniology of the Mound-builders, upon which some very remarkable hypotheses have been erected; nor do I think it worth while to do so, for in the present state of anthropologic science, all the facts of these kinds relating to the Mound-builders which we have as yet learned, can have no appreciable weight to the investigator.

[Investigations conducted since the above Essay was printed require some modifications in its statements. The researches of Professor Cyrus Thomas render it likely that the Cherokees were also Mound-builders, and that they occupied portions of Western Pennsylvania and Western Virginia less than two centuries ago. (See also my work The LenÂpÈ and their Legends, pp. 16–18. Philadelphia, 1885.) Probably the Ohio Valley Mound-builders were the ancestors of some of the Cherokees as well as of the Chahta-Muskoki tribes. Craniologic data from the Ohio mounds are still too vague to permit inferences from them.]

In the first addition of my Myths of the New World^[87] published in 1868, I asserted that the story of the city of Tula and its inhabitants, the Toltecs, as currently related in ancient Mexican history, is a myth, and not history. This opinion I have since repeated in various publications,^[88] but writers on pre-Columbian American civilization have been very unwilling to give up their Toltecs, and lately M. Charnay has composed a laborious monograph to defend them.^[89]

Let me state the question squarely.

The orthodox opinion is that the Toltecs, coming from the north (-west or -east), founded the city of Tula (about forty miles north of the present city of Mexico) in the sixth century, A. D.; that their State flourished for about five hundred years, until it numbered nearly four millions of inhabitants, and extended its sway from ocean to ocean over the whole of central Mexico;^[90] that it reached a remarkably high stage of culture in the arts; that in the tenth or eleventh century it was almost totally destroyed by war and famine;^[91] and that its fragments, escaping in separate colonies, carried the civilization of Tula to the south, to Tabasco (Palenque), Yucatan, Guatemala and Nicaragua. Quetzalcoatl, the last ruler of Tula, himself went to the south-east, and reappears in Yucatan as the culture-hero Cukulkan, the traditional founder of the Maya civilization.

This, I say, is the current opinion about the Toltecs. It is found in the works of Ixtlilxochitl, Veitia, Clavigero, Prescott, Brasseur de Bourbourg, Orozco y Berra, and scores of other reputable writers. The dispersion of the Toltecs has been offered as the easy solution of the origin of the civilization not only of Central America, but of New Mexico and the Mississippi valley.^[92]

The opinion that I oppose to this, and which I hope to establish in this article, is as follows:

Tula was merely one of the towns built and occupied by that tribe of the Nahuas known as Azteca or Mexica, whose tribal god was Huitzilopochtli, and who finally settled at Mexico-Tenochtitlan (the present city of Mexico); its inhabitants were called Toltecs, but there was never any such distinct tribe or nationality; they were merely the ancestors of this branch of the Azteca, and when Tula was destroyed by civil and foreign wars, these survivors removed to the valley of Mexico and became merged with their kindred; they enjoyed no supremacy, either in power or in the arts; and the Toltec “empire” is a baseless fable. What gave them their singular fame in later legend was partly the tendency of the human mind to glorify the “good old times” and to merge ancestors into divinities, and especially the significance of the name Tula, “the Place of the Sun,” leading to the confounding and identification of a half-forgotten legend with the ever-living light-and-darkness myth of the gods Quetzalcoatl and Tezcatlipoca.

To support this view, let us inquire what we know about Tula as an historic site.

Its location is on one of the great ancient trails leading from the north into the Valley of Mexico.^[93] The ruins of the old town are upon an elevation about 100 feet in height, whose summit presents a level surface in the shape of an irregular triangle some 800 yards long, with a central width of 300 yards, the apex to the south-east, where the face of the hill is fortified by a rough stone wall.^[94] It is a natural hill, overlooking a small muddy creek, called the Rio de Tula.^[95] Yet this unpretending mound is the celebrated Coatepetl, Serpent-Mount, or Snake-Hill, famous in Nahuatl legend, and the central figure in all the wonderful stories about the Toltecs.^[96] The remains of the artificial tumuli and walls, which are abundantly scattered over the summit, show that, like the pueblos of New Mexico, they were built of large sun-baked bricks mingled with stones, rough or trimmed, and both walls and floors were laid in a firm cement, which was usually painted of different colors. Hence probably the name Palpan, “amid the colors,” which tradition says was applied to these structures on the Coatepetl.^[97] The stone-work, represented by a few broken fragments, appears equal, but not superior, to that of the Valley of Mexico. Both the free and the attached column occur, and figure-carving was known, as a few weather-beaten relics testify. The houses contained many rooms, on different levels, and the roofs were flat. They were no doubt mostly communal structures. At the foot of the Serpent-Hill is a level plain, but little above the river, on which is the modern village with its corn-fields.

These geographical particulars are necessary to understand the ancient legend, and with them in mind its real purport is evident.^[98]

That legend is as follows: When the Azteca or Mexica—for these names were applied to the same tribe^[99]—left their early home in Aztlan—which Ramirez locates in Lake Chalco in the Valley of Mexico, and Orozco y Berra in Lake Chapallan in Michoacan^[100]—they pursued their course for some generations in harmony; but at a certain time, somewhere between the eighth and the eleventh century of our era, they fell out and separated. The legend refers to this as a dispute between the followers of the tribal god Huitzilopochtli and those of his sister Malinalxochitl. We may understand it to have been the separation of two “totems.” The latter entered at once the Valley of Mexico, while the followers of Huitzilopochtli passed on to the plain of Tula and settled on the Coatepetl. Here, says the narrative, they constructed houses of stones and of rushes, built a temple for the worship of Huitzilopochtli, set up his image and those of the fifteen divinities (gentes?) who were subject to him, and erected a large altar of sculptured stone and a court for their ball play.^[101] The level ground at the foot of the hill they partly flooded by damming the river, and used the remainder for planting their crops. After an indeterminate time they abandoned Tula and the Coatepetl, driven out by civil strife and warlike neighbors, and journeyed southward into the Valley of Mexico, there to found the famous city of that name.

This is the simple narrative of Tulan, stripped of its contradictions, metaphors and confusion, as handed down by those highest authorities the Codex Ramirez, Tezozomoc and Father Duran.^[102] It is a plain statement that Tula and its Snake-Hill were merely one of the stations of the Azteca in their migrations—an important station, indeed, with natural strength, and one that they fortified with care, where for some generations, probably, they maintained an independent existence, and which the story-tellers of the tribe recalled with pride and exaggeration.

How long they occupied the site is uncertain.^[103] Ixtlilxochitl gives a list of eight successive rulers of the “Toltecs,” each of whom was computed to reign at least fifty-two years, or one cycle; but it is noteworthy that he states these rulers were not of “Toltec” blood, but imposed upon them by the “Chichimecs.” This does not reflect creditably on the supposed singular cultivation of the Toltecs. Probably the warrior Aztecs subjected a number of neighboring tribes and imposed upon them rulers.^[104]

If we accept the date given by the Codex Ramirez for the departure of the Aztecs from the Coatepetl—A. D. 1168—then it is quite possible that they might have controlled the site for a couple of centuries or longer, and that the number of successive chieftains named by Ixtlilxochitl should not be far wrong. The destructive battles of which he speaks as preceding their departure—battles resulting in the slaughter of more than five million souls—we may regard as the grossly overstated account of some really desperate conflicts.

That the warriors of the Azteca, on leaving Tula, scattered over Mexico, Yucatan and Central America, is directly contrary to the assertion of the high authorities I have quoted, and also to most of the mythical descriptions of the event, which declare they were all, or nearly all, massacred.^[105]

The above I claim to be the real history of Tula and its Serpent-Hill, of the Toltecs and their dynasty. Now comes the question, if we accept this view, how did this ancient town and its inhabitants come to have so wide a celebrity, not merely in the myths of the Nahuas of Mexico, but in the sacred stories of Yucatan and Guatemala as well—which was unquestionably the case?

To explain this, I must have recourse to some of those curious principles of language which have had such influence in building the fabric of mythology. In such inquiries we have more to do with words than with things, with names than with persons, with phrases than with facts.

First about these names, Tula, Tollan, Toltec—what do they mean? They are evidently from the same root. What idea did it convey?

We are first struck with the fact that the Tula I have been describing was not the only one in the Nahuatl district of Mexico. There are other Tulas and Tollans, one near Ococingo, another, now San Pedro Tula, in the State of Mexico, one in Guerrero, San Antonio Tula in Potosi,^[106] etc. The name must have been one of common import. Herrera, who spells it Tulo, by an error, is just as erroneous in his suggestion of a meaning. He says it means “place of the tuna,” this being a term used for the prickly pear.^[107] But tuna was not a Nahuatl word; it belongs to the dialect of Haiti, and was introduced into Mexico by the Spaniards. Therefore Herrera’s derivation must be ruled out. Ixtlilxochitl pretends that the name Tollan was that of the first chieftain of the Toltecs, and that they were named after him; but elsewhere himself contradicts this assertion.^[108] Most writers follow the Codex Ramirez, and maintain that Tollan—of which Tula is but an abbreviation—is from tolin, the Nahuatl word for rush, the kind of which they made mats, and means “the place of rushes,” or where they grow.

The respectable authority of Buschmann is in favor of this derivation; but according to the analogy of the Nahuatl language, the “place of rushes” should be Toltitlan or Tolinan, and there are localities with these names.^[109]

Without doubt, I think, we must accept the derivation of Tollan given by Tezozomoc, in his Cronica Mexicana. This writer, thoroughly familiar with his native tongue, conveys to us its ancient form and real sense. Speaking of the early Aztecs, he says: “They arrived at the spot called Coatepec, on the borders of Tonalan, the place of the sun.”^[110]

This name, Tonallan, is still not unusual in Mexico. Buschmann enumerates four villages so called, besides a mining town, Tonatlan.^[111] “Place of the sun” is a literal rendering, and it would be equally accurate to translate it “sunny-spot,” or “warm place,” or “summer-place.” There is nothing very peculiar or distinctive about these meanings. The warm, sunny plain at the foot of the Snake-Hill was called, naturally enough, Tonallan, syncopated to Tollan, and thus to Tula.^[112]

But the literal meaning of Tollan—“Place of the Sun”—brought it in later days into intimate connection with many a myth of light and of solar divinities, until this ancient Aztec pueblo became apotheosized, its inhabitants transformed into magicians and demigods, and the corn-fields of Tula stand forth as fruitful plains of Paradise.

In the historic fragments to which I have alluded there is scant reference to miraculous events, and the gods play no part in the sober chronicle. But in the mythical cyclus we are at once translated into the sphere of the supernal. The Snake-Hill Coatepetl becomes the Aztec Olympus. On it dwells the great goddess “Our Mother amid the Serpents,” Coatlan Tonan,^[113] otherwise called “The Serpent-skirted,” Coatlicue, with her children, The Myriad Sages, the Centzon Huitznahua.^[114] It was her duty to sweep the Snake-Hill every day, that it might be kept clean for her children. One day while thus engaged, a little bunch of feathers fell upon her, and she hid it under her robe. It was the descent of the spirit, the divine Annunciation. When the Myriad Sages saw that their mother was pregnant, they were enraged, and set about to kill her. But the unborn babe spake from her womb, and provided for her safety, until in due time he came forth armed with a blue javelin, his flesh painted blue, and with a blue shield. His left leg was thin and covered with the plumage of the humming-bird. Hence the name was given to him “On the left, a humming-bird,” Huitzilopochtli.^[115] Four times around the Serpent-Mountain did he drive the Myriad Sages, until nearly all had fallen dead before his dart, and the remainder fled far to the south. Then all the Mexica chose Huitzilopochtli for their god, and paid honors to the Serpent-Hill by Tula as his birthplace.^[116]

An equally ancient and authentic myth makes Huitzilopochtli one of four brothers, born at one time of the uncreated, bi-sexual divinity, the God of our Life, Tonacatecutli, who looms dimly at the head of the Aztec Pantheon. His brothers were the black and white Tezcatlipoca and the fair-skinned, bearded Quetzalcoatl. Yet a third myth places the birthplace of Quetzalcoatl directly in Tula, and names his mother, Chimalman, a virgin, divinely impregnated, like Coatlicue, by the descending spirit of the Father of All.^[117]

Tula was not only the birthplace, but the scene of the highest activity of all these greatest divinities of the ancient Nahuas. Around the Coatepetl and on the shores of the Tollanatl—“the Water of Tula”—as the stream is called which laves the base of the hill, the mighty struggles of the gods took place which form the themes of almost all Aztec mythology. Tulan itself is no longer the hamlet of rush houses at the foot of the Coatepec, surmounted by its pueblo of rough stone and baked brick; it is a glorious city, founded and governed by Quetzalcoatl himself, in his first avatar as Hueman, the strong-handed. “All its structures were stately and gracious, abounding in ornaments. The walls within were incrusted with precious stones or finished in beautiful stucco, presenting the appearance of a rich mosaic. Most wonderful of all was the temple of Quetzalcoatl. It had four chambers, one toward the east finished in pure gold, another toward the west lined with turquoise and emeralds, a third toward the south decorated with all manner of delicate sea-shells, and a fourth toward the north resplendent with red jasper and shells.”^[118] The descriptions of other buildings, equally wondrous, have been lovingly preserved by the ancient songs.^[119] What a grief that our worthy friend, M. Charnay, digging away in 1880 on the Coatepec, at the head of a gang of forty-five men, as he tells us,^[120] unearthed no sign of these ancient glories, in which, for one, he fully believed! But, alas! I fear that they are to be sought nowhere out of the golden realm of fancy and mythical dreaming.

Nor, in that happy age, was the land unworthy such a glorious city. Where now the neglected corn-patches surround the shabby huts of Tula, in the good old time “the crops of maize never failed, and each ear was as long as a man’s arm; the cotton burst its pods, not white only, but spontaneously ready dyed to the hand in brilliant scarlet, green, blue and yellow; the gourds were so large that they could not be clasped in the arms; and birds of brilliant plumage nested on every tree!”

The subjects of Quetzalcoatl, the Toltecs, were not less marvelously qualified. They knew the virtues of plants and could read the forecast of the stars; they could trace the veins of metals in the mountains, and discern the deposits of precious stones by the fine vapor which they emit; they were orators, poets and magicians; so swift were they that they could at once be in the place they wished to reach; as artisans their skill was unmatched, and they were not subject to the attacks of disease.

The failure and end of all this goodly time came about by a battle of the gods, by a contest between Tezcatlipoca and Huitzilopochtli on the one hand, and Quetzalcoatl on the other. Quetzalcoatl refused to make the sacrifices of human beings as required by Huitzilopochtli, and the latter, with Tezcatlipoca, set about the destruction of Tula and its people. This was the chosen theme of the later Aztec bards. What the siege of Troy was to the Grecian poets, the fall of Tula was to the singers and story-tellers of Anahuac—an inexhaustible field for imagination, for glorification, for lamentation. It was placed in the remote past—according to Sahagun, perhaps the best authority, about the year 319 before Christ.^[121] All arts and sciences, all knowledge and culture, were ascribed to this wonderful mythical people; and wherever the natives were asked concerning the origin of ancient and unknown structures, they would reply; “The Toltecs built them.”^[122]

They fixedly believed that some day the immortal Quetzalcoatl would appear in another avatar, and would bring again to the fields of Mexico the exuberant fertility of Tula, the peace and happiness of his former reign, and that the departed glories of the past should surround anew the homes of his votaries.^[123]

What I wish to point out in all this is the contrast between the dry and scanty historic narrative which shows Tula with its Snake-Hill to have been an early station of the Azteca, occupied in the eleventh and twelfth century by one of their clans, and the monstrous myth of the later priests and poets, which makes of it a birthplace and abode of the gods, and its inhabitants the semi-divine conquerors and civilizers of Mexico and Central America. For this latter fable there is not a vestige of solid foundation. The references to Tula and the Toltecs in the Chronicles of the Mayas and the Annals of the Kakchiquels are loans from the later mythology of the Nahuas. It is high time for this talk about the Toltecs as a mighty people, precursors of the Azteca, and their instructors in the arts of civilization, to disappear from the pages of history. The residents of ancient Tula, the Tolteca, were nothing more than a sept of the Nahuas themselves, the ancestors of those Mexica who built Tenochtitlan in 1325. This is stated as plainly as can be in the Aztec records, and should now be conceded by all. The mythical Tula, and all its rulers and inhabitants, are the baseless dreams of poetic fancy, which we principally owe to the Tezcucan poets.^[124]