^^iti^ti^^BsSSi SSagaivSs ^^-^liilili Sgnv! snt SfiK-.VAi :^^5IHr:^iwtwSSSivSjn§«Si^^ INTRODUCTION Study of North American Archaeology PROF. CYRUS THOMAS, Author of "Report on Mound Explorations" (Twelfth Annual Report of the Bureau of American Ethnology); "Catalogue of Prehistoric Works East of the Rocky Mountains"; "A Study of the Manuscript Troano"; "Burial Mounds of the Northern Sections of the United States"; "Aids to the Study of the Maya Codices"; "Notes on Certain Maya and Mexican Manuscripts"; "Problem of the Ohio Mounds"; "The Cherokees in Pre-Columbian Times"; etc. -^- CINCINNATI: THE ROBICRT CLARKE COMPANY. 1898, Copyright, 189.", By Thk Robert Ci.arkk C'ompany. MAJOR JOHN WE.SLEY POWELL, To wliose efficient work as Director of the Bureau of American Ethnology students of ethnology are so largely indebted for las recent additions made to the data relating to North America ; and to whose aid and encouragement is mainly due whatever success the vjriter may have achieved in his special line, Cbta work is i*cspcctful!y dedicated By the Author. PREFACE. The little volume herewith presented to the public is a brief resume of the progress which has been made, up to the present time, in the investigation and study of North American archaeologj^ The in creased activity among students devoting attention to the subject, the numerous explorations made, the rapid accumulation of data and the flood of light thrown on the questions relating to prehistoric North America since the publication of the last general work relating thereto, call for a new summary. Whether tlie work now offered meets this demand must be left for the readers to decide. That some parts of the broad field have been left unnoticed is admitted, the attention being confined chiefly to the more important characteristic features, as those best calculated jto form an Introduction to the subject ; and as best calculated to interest the reader and younger students. With such an object in view, pages broken or interrupted by foot-notes are not only out of place, but often serve to break the thread the reader is following, or prove an interruption to his line of thought ; reference notes have therefore been entirely omitted. The opinion held by Maj. J. W. Powell that the (v) vi Preface. Indians found inhabiting the Atlantic division of North America and their ancestors were the builders of the mounds in that region, which the explorations of the Bureau of American Ethnology under his charge have done much to confirm, has been adopted. And, in general, the conclusions reached by the Bureau of American Ethnology in reference to questions re lating to language and archaeology, so far as these ex tend, have been accepted and used as a basis for further steps in the investigation. But the author alone must be held responsible for any views advanced herein which have not been generally accepted, or in regard to which there are different opinions. I take pleasure in acknowledging here the favors I have received from Maj. J. W. Powell, Director of the Bureau of American Ethnology, and Prof. W. J. McGee, Ethnologist in Charge, in the use cf books, pamphlets and other literary aids needed in my work, and the privilege of obtaining numerous electrotypes of the illustrations herein used, favors, however, which have always been willingly extended to all co workers. I also wish to acknowledge the favors re ceived from Prof. W. H. Holmes, in the privilege of copying illustrations of and profuse borrowing from his late work on the cities of Mexico, published by the Field Columbian Museum ; also to Mr. F. W. Hodge, of the Bureau of American Ethnology, for information communicated and papers furnished relating to the Preface. vii Pueblo region. In addition to the illustrations ob tained from the Bureau publications, others have been copied from figures in the U.S. Geological Sur vey, National Museum, etc. Acknowledgment to the various authors from whose works information has been drawn will be found in the text, the authors' names from whose works and papers illustrations have been obtained either directly or indirectly, are added after the numbers in the list of illustrations, the original being referred to where it is possible. The numbers in the list of illustrations not followed by the author's name are either original figures, modifications of other figures, or theoretical restora tions by the present writer. CONTENTS. Preliminary Observations — page Tlie object, scope and plan of the work 1 Materials for Study and Classification 8 Methods of Study 22 Arctic Division — Monuments and local antiquities 35 Implements, ornajnents, etc 40 Culture home bf the Eskimo 43 Atlantic Division 48 Monuments and local antiquities 50 Mounds 51 Burial mounds 61 , Vessels, implements and ornaments 79 Pottery 87 Long-necked bottles 94 The gulf province 97 Pipes 98 Articles of shell 103 Textile fabrics 108 Copper articles 109 Articles of stone 113 Inclosures and pyramidal mounds 117 Prehistoric movements of population 121 Hut-rings and house-sites 132 Antiquity and authors of the mounds 138 Duration of the mound-building age 147 Inclosures and other mural works 152 (ix) X Contents. PAGE. Pacific Division 169 North Pacific section — Athapascan region 170 North Pacific coast 176 California section 187 Prehistoric movements of population 200 Intormontane or pueblo section. 203 Cave-dwellings , 205 Cliff-dwellings , -208 Ruins on the plateaus and in the valleys 215 Gila valley and Chihuahua 221 Builders of tho cliff-houses 229 Mexican section — civilization 233 Monuments of southern Mexico 252 Monuments of southern Mexico — continued 264 Monuments of Central America 276 Chichen-Itza, Tikal and Copan '. 296 Migrations of the Mexican and Central American tribes, ... 312 Migrations of certain Mayan tribes 328 Origin and development of Central American civilization. . . 339 Priests, hieroglyphs and calendar .356 Conclusion 368 LIST OF ILLUSTRATIONS. PAGE 1, Bone spear head, Eskimo, (Dall,) 37 2. Stone lamp, Eskimo. (Dall.) 37 3. Labret, Eskimo. (Dall.) 38 4, Remains of an ancient Eskimo house. (Boas.) 39 5, L^lu, or woman's knife, Eskimo, (Mason.) 41 6. Soapstone pot, Eskimo. (Murdoch.),., 42 7, Hafted jade adze, Eskimo, (Murdoch.) 42 8. Skin scraper, Eskimo. (Murdoch.) 43 9. Flint flaker, Eskimo. (Murdoch,) 43 10. Plat of mound group, Wisconsin. (Thomas.) 52 11. Terraced mound, Arkansas, (Thomas.) 54 12, Elephant mound, Wisconsin. (Thomas.) 56 13. Group of chain mounds, Wisconsin, (Thomas.) 57 14, Section of Mississippi mound, (Thomas,) 63 15, Section of mound in eastern Tennessee, (Thomas.) 66 16. Earthern pot, eastern Tennessee. (Thomas.) 67 17, Shell ear ornament or hair pin. North Carolina, (Thomas.), 67 18, Engraved shell, North Carolina. (Thomas.) 68 19. Soapstone pipe, east Tennessee. (Thomas.) 68 20. Beehive vaults, North Carolina. (Thoraas.) 69 21. So«pstone pipe, North Carolina, (Thomas,) 70 22, Stone-grave cemetery, Illinois, (Thomas,) 72 23, Triangular pit, North Carolina, (Thomas,) 80 24, Position of skeletons in an east Tennessee mound, (Thomas.) 85 25. Copper hawk's-bell, east Tennessee. (Thomas.) 86 26, Clay vessel, Canada, (Boyle,) 88 (xi) xii List of Illustrations. PAGK. 27. Outline figures of bowls. (Holmes.) 89 28. Ornamental bowl, Tennessee. (Thruston.) 89 29. Animal-shaped bowl, Arkansas. (Holmes.) 90 30. Bird-shaped bowl, Arkansas. (Holmes.) 90 31. Pot-shaped vessel, Arkansas, (Holmes.) 91 32. Pot-shaped vessel, west Tennessee. (Holmes.) 91 33, Wide-mouthed bottles. (Holmes.) 92 34. (a) Oppossum vase, Arkansas. (Holmes.) 92 (b) Sunfish vase, Arkansas. (Holmes.) 93 35. Bowl representing the human head, Arkansas, (Holmes,). 93 .36. Winged and crested rattlesnake design, Ark. (Holmes.). 94 37. Outline figures of long-necked bottles. (Holmes.) 94 38, Eccentric shapes in' long-necked bottles. (Holmes.) 94 39. Owl-shaped bottle, east Tennessee. (Thomas.) 95 40. Burial urn, Georgia, (Jones.) 97 41. Vessel with four legs, Georgia. (Jones.) 97 42. (a and b) Stemlesspipes. (Boyle.) 99 43, Image pipe, Georgia. (Thomas.) 99 44. Image pipe, Arkansas, (Thomas,) 100 45, (a, b and c) Short-necked pipes. (Thomas.),... 100 46. " Monitor " pipe 101 47, Engraved shell, Arkansas, (Thomas.) 104 48, Shell gorget, Tennessee, (Jones.) 106 49. Shell gorget, Georgia. (Thomas.) 106 50. Figured copper plate, Georgia. (Thomas.) Ill 51. Figured copper plate, Illinois. (Thomas,) 112 52. Figured copper plate, Illinois. (Thomas,) 112 53. Stone image, Tennessee. (Thomas.) 114 54. Stone image, Tennessee. (Thomas.) 114 55. Banner stones 115 56. Arrow heads. (Mercer.) Hg 57. Mound with graded way, Georgia, (Thomas,) 118 58, Double terraced mound, Arkansas. (Thomas.) 119 59. Selsertown mound, Mississippi. (Thomas.) 120 60, Newark works, Ohio. (Thomas.) 122 List of Illustrations. xiii PAGE. 61, " Hill Fort," Ohio. (Thomas.) 126 62. Linu.works, Illinois. (Thomas.) 128 63. "Angel mounds," Indiana. (Thomas.) 130 64. House site, Arkansas, (Thomas.) .' 134 65. Supposed method of lathing houses, (Thomas.) 135 66, Surface effigy, South Dakota. (Thomas.) 149 67. Dene stone war club, British America, (Morice,) 173 68. Stone war club, Colorado, (Wickersham.) 173 69. North-west coast pictograph. (Niblack.) 178 70. Ceremonial dress of Chilkat chief, north-west coast, (Niblack,) 178 71. Totem posts of north-west coast. (Niblack,) 179 72. Relics from southern California 188 73, Manner of walling up the front of a cave-dwelling. (Holmes.) 207 74, Cliff-dwelling on the Rio Mancos. (Holmes.) 210 75. Ruins at Aztec Springs. (Holmes.) 216 76. Village group, Arizona. (Mindeleff.) 218 77, Ground plan of the Pueblo Bonito, (Jackson.) 219 78. Ground plan of Casas Grandes. (Bandelier.) 224 79. Ground plan of a building at Casas Grandes. (Bartlett.) . . 228 80. Mayan day symbols ' 242 81. Mexican day symbols 242 82. Part of the inscription of the Tablet of the Cross, Palen- que. (Photograph.) 246 83, Pyramid at Los Edificios 254 84. Sculptured column, Tula 257 85, Ruins of Teotihuacan. (Holmes,) 260 86. Ground plan of Teotihuacan building, (Holmes.) 261 87. Euins of the temple of Xochicalco. (Nadaillac: "L'Am. Prehist.") 265 88. Ruins at Mitla, (Holmes,) 269 89. Room with columns, Mitla. (Holmes.) 271 90. Fretwork in the grand palace, Mitla. (Holmes.) 272 91. Painted designs, Mitla. (Holmes.) 273 xiv List of Illustrations. PAGE. 92. Plan of the ruins at Palenque. (Holmes.) 278 93. Ground plan of the palace, Palenque. (Holmes.) 279 94. Cross-section of palace, Palenque. (Holmes.) 280 95. Part of Palenque palace restored 281 96. Ground plans of Temples at Palenque. (Holmes.) 282 97. Sculptured lintel, LorillardI City, (Photograph.) 286 98, Plan of ruins at Uxmal. (Holmes,) 289 99, Ornamentation on the governor's palace, Uxmal. (Holmes,) 290 100, Section of the Casa del Gobernador, (Holmes,) 292 101. Ornamentation on the nun's palace, Uxmal. (Holmes.).. 293 102. Ground plan of nunnery, Chichen-Itza. (Holmes.) 298 103. Elephant trunk figure, Yucatan 299 104. Vertical section of the Caracol, or tower, Chichen-Itza. (Holmes.) 299 105. Ground plan of the tower, Chichen-Itza. (Holmes.) 300 106, Atlantean figure, Chichen-Itza, (Holmes,) 301 107. Ground plan of ruins, Copan, (Maudslay.) 308 108. Vertical section, main group, Copan. (Maudslay,) 309 INTRODUCTION Study of North American Archaeology, CHAPTER I. PRELIMINARY OBSERVATIONS. Archaeology in its widest sense and by derivation includes the investigation of the origin, language, be liefs, customs, arts — every thing, in a word, that can be learned of the ancient life of a people. It is in this sense that it is used in the title of this work, and not in the more limited scope to which its modern use has a tendency to restrict it, notwithstanding the ef forts of lexicographers to retain its original meaning. The work is intended really as an introduction to the study of prehistoric North America, of the people as well as the monuments. To gather and describe an tiquities, although thoroughly and intelligently done, is by no means all of archaeology. True, these are to archaeology what the unfashioned and unadjusted materials of which the house is to be built are to the house, but they are not the house. The monuments are the tombs of past ages ; the work of archaeology in its broad sense is to revivify the dead, to put life into the past, and, so far as possible, to bring before 1 2 Stud-y of North American Archaeology. the mind the ancient people with their activities, characteristics and customs. In other words, the chief object in view in the study of archaeology is the man of bygone ages. It is with this idea in view that this Introduction to the Study of North American Archaeology has been written, but only as an intro duction, for the field is too broad to be covered in one small volume. The problems which confront the student of Ameri can archaeology are exceedingly difficult, and some of them seemingly beyond the possibility of solution be cause of the sheer break between the historic and prehistoric eras. Omitting the discovery of Greenland and possibly the north-east coast of the continent by the Northmen in the tenth or eleventh century, which left no impress, the history of the western world be gins with the discovery of the West Indies by Colum bus in 1492 ; all that lie back of that date belong to the prehistoric era, a gloom, so to speak, unlight- ened by a single deciphered page of history. In the Old World there are few regions in regard to whose past there are no recorded hints which can be used as stepping stones in the backward march ; in other words, history and prehistory are dovetailed, so to speak, one with another, but not so in the New World. These difficulties have possibly caused more than one student to feel as Palgrave, who, in apparent de spair' over the unsatisfactory results of the efforts, up to his day, to lift the veil which shuts out the past, ex claims : "We must give it up, that speechless past ; whether fact or chronology, doctrine or mythology ; whether in Europe, Asia, Africa or America ; at Preliminary Observations. 3 Thebes or Palenque, on Lycian shore or Salisburj^ Plains; lost is lost, gone is gone forever." But a different spirit animates the students of the present day, the very difficulties in the way are themselves so many incentives to attacks. What seemed beyond human reach to the London antiquary and historian sixty years ago, is deemed by scholars of the closing years of the nineteenth century to be, in a large de gree, attainable. The veil, which Palgrave looked upon as fixed and immovable, has been lifted at nu merous points and rays of light let in upon the past. Some of the problems which were, fifty years ago, yea, but twenty-five, deemed impossible of sohition, have been satisfactorily solved, and have now become foundation stones in the archaeological structure. It is true, as Sir Johii Lubbock remarks : "In at tempting to reconstruct the story of the past, students have too often allowed imagination to usurp the place of research, and have written in the spirit of the nov elist rather than in that of the philosopher." The hundreds of dust-covered works on the subject of pre- Columbian America, and the origin of its people, which now lie, well nigh forgotten, on the shelves of libraries, bear testimony to the truth of this remark, as do also the numerous discarded theories relating thereto. Nevertheless theories will continue to be ad vanced, indeed must be if progress is made in the study of the past, especially where so many links of the chain are still wanting as in American archae ology. The investigating spirit of the age will not brook delay ; wherever there is an unbroken field some scientific plow is sure to enter, though but poorly equipped for the work. However, of late years more 4 Study of North American Archaeology . strict methods of approaching the many problems in volved have been introduced, and now, instead of at tempting by imagination or theorizing to reach con clusions at once, slow and patient investigation is the process pursued. The spade has to a large extent re placed the pen, and instead of building theories chiefly by imagination, there is a careful sifting of all the evidence which appears to have any bearing on the subject. The fragments of data are fitted to gether and tentative theories deduced simply as a plan of further progress, often however to be cast aside or modified, as new material, which will not readily drop into place, is discovered. As no intelligent student will continue his investi gations of the ancient monuments for any consider able length of time without forming theories in regard to the uses, age and authors of the works examined, it is all important to his progress to know which of the questions that arise have been completely or par tially answered, and to ascertain which of the numer ous theories advanced in regard to the various ques tions have been definitely eliminated by universal consent from the class possessing elements of possi bility. This knowledge will clear from his pathway much of the rubbish which would otherwise encumber it. Another important point is to know the lines along which the opposing views are being pushed by their respective advocates. Although it is undoubtedly the part of wisdom to hold in restraint the disposition to theorize, this knowledge directs the careful student's attention to numerous points which might otherwise be overlooked in his investigations. It is for this reason that the author of this little work has ventured Preliminary Observations. 5 to briefly outline the theories relating to some of the more important problems which must confront the student of American archaeology. The chief object, however, will be to present the data, and to arrange them so as to afford the student some means of bring ing into harmony and utilizing his facts and materials. But as it is manifestly impossible to present in a single small volume a full account of the archaeologic remains of the continent, and discuss all the questions which arise in connection therewith, only those con sidered the best representatives of the leading types and those which best illustrate the art, customs and culture status of the former inhabitants will be re ferred to. The writer, as those who peruse this work will ob serve, has not entered into a discussion of the question of the so-called paleolithic age, or glacial man in America, for the reason that he does not believe the evidence on which the theory is based, as yet sufficient to justify its acceptance. The results of the more re cent investigations in America, or at least North America, all tend in the other direction. One by one the strongholds of the advocates are being overturned, and the evidence on which the theory is based dis counted. The author feels constrained to the belief that peopled America though old in years bears no where such marks of antiquity as are to be found in some parts of the Eastern Continent. To accept the antiquity which has been assigned by the advocates of this theory to the early inhabitants would, as the Avriter thinks, require in order to be consistent an en tire recasting of all the more stable theories which have been propounded. "Paleolithic" as a descriptive 6 Study of North American Archaeology . term is of minor importance, but as a theory which would carry back the presence of man in America to that immensely distant era which has been assigned is a very diff'erent thing, Mr. Keary remarks in the opening paragraph of the second chapter of his "Dawn of History" that "Between the earlier and later stone age, between man of the drift period and man of the neolithic era occurs a vast blank which we can not fill in. We bid adieu to the primitive inhabitants of our earth Avhile they are still the contemporaries of the mam moth and woolly rhinoceros, or of the cave lion and the cave bear, and while the very surface of the earth wears a different aspect from what it now wears. With a changed condition of things, with a race of animals which differed not essentially from those known to us, and with a settled conformation of lands and seas not again to be departed from, comes before us the second race of man — man of the polished stone age." It is true that it is claimed by some European authors that this hiatus is not so real as it at first appears to be, and that it has been partially bridged over by some recent finds. But the effort to bridge the chasm shows too clearly to be misunder stood that it is there, and so long as it remains un closed is a weak point, if not fatal flaw, in the theory. We accept as correct the idea advanced by Mr. A. H. Keane in his "Ethnology," that appeal to tradi tional movements and other traditional data will have no bearing upon the question of the origin of the people of America unless paleolithic man in America is abandoned. So believing, though we do not pro pose to discuss this question of the original peopling Preliminary Observations. 7 of the continent, we put aside glacial or i)aleolitliic man of America as yet wanting in the credentials which entitle him to a place in scientific circles. The history of the western continent is supposed to begin with its discovery by Columbus at the close of the fifteenth century, all that antedate that event being considered prehistoric. While this is true in the broad and general sense in which it is used, yet, strictly speaking, the history of the different sections begins with the first knowledge of them obtained by Europeans. Hence the border line between the his toric and prehistoric eras varies in date according to the section referred to. The Ohio valley, for example, was terra incognita to the civilized world for a century after Cortez entered the capital of Anahuac. That which lies back of this border line belongs to the pre historic era, and the student who would penetrate the mystery of that past must examine and carefully study the monuments ; listen to the traditions which have floated down the ages ; gather the folk-lore tales ; and compare the customs, arts, and beliefs of the tribes as first seen and learned. He must study the native form and lineaments, and trace by linguistic evidence the relationship of tribes and groups ; for in America there is no scaffolding of history to assist him as in the Old World. The transition from the prehistoric to the historic was, from the very nature of the case, sudden, there being no true proto-historic period. Study of North American Archaeology. CHAPTER II, MATERIALS FOR STUDY CLASSIFICATION. When Columbus sailed among the Antilles, Cortez landed on the coast of Mexico, when Jacques Cartier sailed up the St. Lawrence and De Soto traversed the Gulf States, each and all found the regions they visited inhabited by people of a race different from any known to the eastern continent. The discoveries which followed brought to light the fact that the lake region and the Mississippi valley were inhabited by people of the same race. Whence they came, and how long they had inhabited these regions — in other words what was their history — could not be ascer tained, as they possessed no historical records save a few symbolic rolls and inscriptions which are as yet- sealed books to scientists. The dim and shadowy traditions which they related to the European dis coverers were so confused and, in most cases, so fabulous as to throw but little light on these ques tions. And what was found to be true of the regions mentioned was found to be true in a general sense of the entire continent. The most important variation discovered was the evidence of more advanced culture in certain areas, as Mexico, Central America, and Peru. The people, however, though split into numer ous stocks and tribes, and differing in minor respects, belonged apparently to the same race, its members being popularly known as "Indians" or "American Indians." Materials for Study — Classification. 9 In these facts we have one fundamental point with which no correct conclusion in regard to the pre historic times of the continent can be at variance. The natives were here a.nd must be recognized by every theory, must be a factor in every general con clusion. The chief fundamental factor in the study of ar chaeology is found in the monuments. "The teach ings of material relics," truly remarks one author, "so far as they go, are irrefutable. Real in them selves, they impart an air of reality to the study of the past." These are indisputable products of human activity, and have imprinted upon them, as it were, the ideas and conceptions of a bygone age. They are records in which we may read not only the culture- status of that past age, but also much in regard to the customs and beliefs of the people. For these reasons attention is directed to them as the chief foundation stones on which our archaeological struc ture must be built. Although the monuments furnish the chief and most reliable data to the archaeologist, and throw more light on the customs, arts and beliefs of the people, and reveal more in regard to the life of the individual and family than any other aids, they are not the only helps he finds in his endeavor to pene trate the unwritten past. Language, which is also reliable, enables him to determine the affinity of tribes and peoples. By this means he can often say with positive certainty that widely separated tribes or groups have, in the past, sundered relations with the great body of their kindred and sought distant homes. He has ascertained by this means that the 10 Study of North American Archaeology. Apaches and Navajos of New Mexico, Arizona, and Northern Mexico are offshoots from the great Athapas can family of Northern British America, and that the Arapahos and Blackfeet Indians of the western plains are members of the Algonquian stock which spread over North America from the Atlantic coast to the Rocky Mountains. Thus he is enabled to trace with more or less accuracy the lines of prehistoric migra tion, and outline the general trend in ancient move ments of population. Traditions, although less reliable than the monu ments and language, furnish some data to the archae ologist which frequently serve to explain otherwise uncertain evidence, and lead to satisfactory conclu sions. Folk-lore, mythology and customs sometimes indicate former contact or relationship not otherwise revealed, and explain many otherwise puzzling monu ments and relics. Craniology is strongly appealed to by European ethnologists" as an important factor in this study, but the results so far obtained, except in cases of artificial pressure, are too unsatisfactory to justify its use except in broad generalizations, and then only as cumulative evidence. This, the writer is well aware, is in conflict with the views of a num ber of leading ethnologists ; nevertheless he feels jus tified in making this statement deliberately to the younger students of American archaeology. The wide differences in many respects between the monumental remains of the Old World and those of the New, and also between the data relating thereto, call for a widely different method of study. Even the classification and nomenclature of the former are not adapted to the latter. The arrangement into four Materials for Study — Classification. 11 classes or ages — the Paleolithic, Neolithic, Bronze and Iron — is conceded to be inapplicable to America. Evidence of the two stone ages may possibly be found, though still denied by a number of our leading archaeologists, and a copper age may be substituted for the bronze, but the similarity will extend no further. The use of iron as a metal was unknown in America previous to the discovery by Columbus. Copper was used to a limited extent, but it is ex tremely doubtful whether the method of manufactur ing bronze had been discovered at any point on the continent. Stone was the chief reliance until the in troduction of European implements. The archae ologic remains of the former, taken as a whole, are so widely different from those of America, that the nomenclature of the one, except as applied to some of the ruder objects, is totally inapplicable to those of the other. It has therefore been found necessary, in studying the archaeology of America, to proceed upon an independent line and to adopt an original basis and a new nomenclature. Although this limits the range of any classificatory system which may be attempted, it falls far short of do ing away with the difficulties the American archaeolo gist is compelled to encounter. Not only is he con fronted by the fact, as apparent in the Old World as in the New, that archaeology, even where it has been longest studied, has not reached that stage where it may be termed a true science, the general principles of which by modification may apply to any sec tion or country, but also by a multiplicity of objects so variant in form and character, and usually in such a fragmentary condition as, without a knowledge of 12 Study of North American Archaeology . their uses, to baffle his attempts at a systematic classi fication. Nadaillac, alluding to the various forms of American antiquities, remarks that "these facts will show how very difficult, not to say impossible, is any classification," a statement which any one who at tempts a systematic arrangement will be disposed to accept as true. When dealing with a limited area where the types are somewhat similar, classification to some extent is possible and advantageous, but the attempt to apply it to the entire continent will prove abortive. However, as some grouping is necessary in order to facilitate reference and comparison, in the absence of a scientific arrangement we must have re course to an arbitrary scheme. As the author has as yet seen no better arrangement of primary groups than that suggested in his "Report on the Mound Explorations of the Bureau of American Ethnology," published in the 12th Annual Report of the Bureau, it is adopted here. By this the objects are divided, in a broad and com prehensive sense, into three classes. 1. Monuments (in the limited sense) , or local antiqui ties. This division or class includes all those antiqui ties that are fixed or stationary, which necessarily pertain to a particular locality or place. 2. Relics and Remains, or movable antiquities. Those not fixed and which have no necessary connection with a particular locality. 3. Paleographic Objects. Inscriptions, picture writ ings, etc., whether on fixed or transportable objects. This is, of course, an arbitrary arrangement, the third group being unnecessary except as a matter of convenience ; however it appears to be a practical Materials for Study — Classification. 13 working system by which the lines of distinction are somewhat rigidly drawn. Moreover it is adapted to the two methods of investigation and study, viz., in the field and in the museum, and is in line with Dr. Moriz Hoernes' suggestion that, in studying archaeological objects, attention should be given to the "Typographic and Museographic order." The first class does not appear, to be susceptible of arrangement into satisfactory primary divisions. The only plan which as yet seems possible is to arrange them by types, chiefly according to form, where the object and use are not apparent, or known. The objects of the second class may be grouped into two divisions : 1. Remains, including human and animal remains ; 2. Relics, including all other mova ble antiquities. The further division of the second group is largely typological, reference being made to use so far as this is evident. Partly because of the difficulties in the way of a satisfactory and useful classification, and partly be cause the chief object of archaeologic investigations is to learn what is possible in regard to the life, char acter, activities and racial affinity of the former in habitants of given sections, it has been found most advantageous to study the monuments according to the culture areas, so far as these can be determined approximately from the data which have been ob tained. As it is practically impossible to make any satisfac tory classification of the antiquities of the whole con tinent, further than into the primary classes men tioned above, the order followed herein will, as above intimated, be geographical rather than typological. 14 Study of North American Archaeology , The divisions will be made to correspond, so far as the data enable us to judge, to the culture areas. But the attempt to mark the culture areas, except as to the three primary divisions mentioned below, can only be partially carried out, hence the subdivisions must be considered as chiefly geographical and intended more as a matter of convenience and comparison than as archaeological. Nevertheless that there are several culture areas, both in the Atlantic and Pacific divis ions, which will be ultimately determined and out lined, is undoubtedly true. It is also true that, even with the data which have been obtained, some of these areas are quite clearly indicated, though they can not be mapped with boundary lines. The tendency of the present day is to base the efforts to arrange the native population into ethnic groups on the linguistic evidence alone, leaving out of view the important aid in tracing the development of these groups to be derived from a careful study of the archaeological data, or referring to them only when they can be used to confirm the theories based on the linguistic evidence. This arises in part from the fact that, while the archaeological data relating to a large portion of the continent are few, and that archaeology can not, as yet, be considered a true science ; ou the other hand the linguistic material, although not complete, is much more abundant, and the treatment thereof reduced to true scientific meth ods. As the latter field affords greater promise of reaching positive conclusions, it is more attractive to methodical students. As the discussion of this subject from the linguistic standpoint is necessarily based upon the study of the Materials for Study — Classification. 15 various linguistic stocks and families of the entire continent, and, to some extent, upon the migrations therein, so the discussion of the same questions from the archaeological standpoint must be based upon the study of the various types and their distribution over the continent. And the same necessity for grouping in some manner arises here as in the linguistic field. Although the materials with which the antiquarian has to deal are not so well defined and distinctly classed as those with which the philologist is con cerned, yet careful study and comparison will enable him to note the differences, geographical and to some extent ethnical. The indications of comprehensive archaeological sections as marked by differences in type are too apparent to be denied, and there are also indications of minor districts. The chief drawback in attempting to use these as evidences of ethnic dis tinctions arises from several causes — uncertainty as to what types are Avholly due to physical conditions and what are tribal or ethnic ; also from a lack of material for comparison ; the overlapping and intermingling of types in consequence of the shifting of position by tribes ; and lastly the fact that types of art are not governed strictly by ethnic lines. Nevertheless race characteristics and tribal customs imjaress themselves to a certain extent under all variations in location and condition, upon the works and art of people in a sav age or semi-civilized state. For instance, although the Mexicans and Mayas lived side by side, and used the same calendar system and the same method of enumeration, yet we notice marked differences be tween their symbolic writings and their types of art. We also notice in the mound section the wide differ- 16 Study of North American Archaeology. ence between the mound types of Wisconsin and the other portions of the Mississippi valley. However, it is difficult with the data so far obtained to fix cor rectly the boundaries of the different culture districts. Although we meet with this difficulty in defining geographically the boundaries of the districts and more comprehensive sections, it does not prevent us from drawing correct conclusions from their general positions and peculiar types. That all the distin guishing types of a district or section can not be at tributed to the peculiar physical features of such dis tricts or sections must be admitted. Will any one claim that the vast difference between the archaeologic types of Mexico and Wisconsin have resulted vt^holly from the physical differences of the two areas? If not, it follows, though physical environment is a po tent influence in the formation of types, that so much as has not resulted from physical peculiarities must be attributed to racial or tribal customs. Yet the powerful influence of physical conditions must not be overlooked. A careful examination of what has been ascertained in regard to North American archaeology, with special reference to the question of archaeologic sections, leads in the flrst place to the conclusion that the ancient remains belong in a broad and comprehensive sense to two general classes. One of these classes is limited geographically to the Atlantic slope, the other chiefly to the Pacific slope, the eastern or Rocky mountain range of the great continental mountain belt to the Rio Grande forming approximately the dividing line between the two areas. According to this division, the Atlantic section includes that part Materials for Study — Classification. 17 of the continent east of the Rocky Mountains and north of the Gulf of Mexico except the Dene or north ern Athapascan region, and the Pacific section 'the re mainder from Alaska to the Isthmus of Panama, in cluding the Athapascan territory. The arctic region, or Eskimo area, which is not taken into consideration bere, forming a separate division. While there are manifest and marked differences in the types and characters of the ancient works and re mains of different areas within each of these two comprehensive sections, yet when those of the Pacific slope, as a whole, are compared Avith those of the Atlantic slope, there is a dissimilarity which marks them as the products of different ethnic groups, or as the result of different influences. If this division into two great archaeologic sections is based on sufficiently reliable data to justify its adoption, it will form a very important landmark in the discussion of the chief problems of the prehistoric times of our continent. Reference to some only of the evidences bearing on this point is made here to show their character, as it would not be possible to present them in detail in a short chapter. One of the first impressions made upon the mind of the student of North American ethnology is the re semblance in a broad and general sense of the features, customs, arts, and archaeological remains of the west coast to those of the islands in and countries border ing on the Pacific ocean, while on the other hand there is no such resemblance between them and those of the Atlantic slope. In other words, the types when classified in the broadest sense appear to ar- 2 18 Study of North American Archaeology. range themselves in two general divisions — those be longing to the Pacific slope and those confined to the Atlantic slope. Although this classification Avas not made in express Avords until it Avas done by the author of this work (see Report on Mound Explorations, irt the 12th Annual Report of the Bureau of American Eth nology, 1894) , yet there is a A^ery evident tendency in the Avorks relating to the west coast ethnology toward such a classification, and a disposition to separate and mark out Avliat may be termed the Pacific types. "If nations of the eastern shores of the Atlantic," says Prof. Dall, "were responsible [for the introduction of the above mentioned types] , Ave should expect the At lantic shores of America to show the results of the in fluence most clearly. This is not the case, but the very reverse of the case." As indicative of this difference a few of the types may be noticed, as foUoAvs : The singular form of carving, representing a figure with the tongue hang ing out, and usually communicating with a frog, otter, bird, snake, or fish, observed on the north-west coast from Oregon to Prince William sound and also in Mexico and Nicaragua. We may add that this feature is found in numerous instances in statues and bas-reliefs from Mexico to the Isthmus, also in the codices of Mexico and Central America, but seldom if ever appears in the antiquities of the Atlantic division. The prominent Tlaloc nose of Mexican and Central American figures, of which the supposed elephant proboscis is but one form and the bird bill (thun der bird) of the north-west coast another, both of which are but different methods of representing the same idea, is a characteristic of the Pacific side. Materials for Study — Classification. 19 The method of superimposing, in totem posts and statues, one figure upon another, usually combining human and animal, is found, except in California, from Alaska to the isthmus, and is a true Pacific type, being almost unknoAvn in the Atlantic division. The angular designs on the pottery and basketry are another marked feature of the west coast division. And thus Ave might, if this Ave re the proper place to enter into details, go on enumerating marked distinc tions betAveen these two primary ethnological sections. As evidence of the fact stated, let any one compare the figures in Ensign Albert P. Niblack 's excellent work on "The Coast Indians of Southern Alaska and Northern British Columbia," Avith the Mexican and Central American monuments and figures, and then compare them Avith the types of the Atlantic slope. While the first comparison shoAVS such a marked re semblance as to lead to the inference that they were derived from some common source, or the result of some common influence, the second comparison shows no such similarity. The spread of types of custom and art Avere governed in part by several influences, as ethnic lines, migrations, contact and physical con ditions. Where we find those of a character which do not depend upon physical conditions, but upon super stitious notions, following a given line without spread ing out indefinitely, Ave may assume, until satisfactory evidence of another cause is given, that they mark a line of migration and are largely ethnic. It is in this light Ave are inclined to view the coast-line ex tension of the types peculiar to the Pacific slope. Dr. Brinton ("American Race"), notAvithstanding his view in regard to the origin and homogeneity of 20 Study of North American Archaeology. the American race, arranges his linguistic groups geographically by the same dividing lines as those Ave here indicate as separating the primary archaeological divisions. His "North Atlantic Group," omitting the Eskimos, corresponds with our Atlantic division, and his "North Pacific" and "Central" groups combined with our Pacific division. This arrangement, as he admits, is not one of convenience only, as he attaches certain ethnographic importance to it. "There is," he continues, "a distinct resemblance between the two Atlantic groups, and an equally distinct contrast be- tAveen them and the Pacific groups, extending to tem perament, culture and physical traits. Each of the groups has mingled extensively within its own limits and but slightly outside of them." Elsewhere he remarks that "a few of the eastern stocks, the Atha pascan and the Shoshonian, have sent out colonies who have settled on the banks of the Pacific ; but as a rule the tribes of the western coast are not connected with any east of the mountains. What is more singu lar, though they differ surprisingly among themseh^es in language, they have marked anthropological sim ilarities, physical and psychical. VirchoAV has em phasized the fact that the skulls from the northern point of Vancouver Island reveal an unmistakable analogy to those of southern California. . . . There are many other physical similarities which mark the Pacific Indians and contrast them with those east of the mountains." In his "Races and Peoples" this division between the eastern and western slopes is ex pressed still more pointedly : "All the higher civiliza tions are contained in the Pacific group, the Mexican really belonging to it by derivation and original loca- Materials far Study — Classification. 21 tion. Between the members of the Pacific and At lantic groups there Avas very little communication at any period, the high sierras Availing them apart." As the arctic section, especially those portions oc cupied by the Eskimo stock, present marked pecu liarities, the whole of North America may be con sidered in three divisions Avhich may, for convenience, be termed : I. The Arctic Division. II. The Atlantic Division. III. The Pacific Division. 9-> Study of North Avierican Archaeology. CHAPTER III, METHODS OF STUDY. Although the method of studying American archae ology has been touched upon to some extent in the preceding chapters, it may be well to add something more on this subject before entering upon a discussion of the antiquities of the divisions outlined above. Most of the Avriters dealing generally Avith this sub ject begin their works Avith the primitive, or supposed primitive inhabitants — paleolithic men, men of the mastodon age, caA^e men, etc. It is probably the cor rect and scientific method in an extended treatise on American archaeology to begin Avith the earliest traces of man on the continent, thence folloAving him down the ages, marking his advance in culture, but it is very questionable whether this is the best method of studying North American archaeology. It is the be lief of the author of this work that the most satis factory plan is to begin AAdth the knoAvn and work back toward the unknown ; to begin Avith the aborig ines and monuments and trace them back step by step into the past. The evidence so far ascertained leads to the con clusion that, as a general rule, the monuments of the various sections are attributable as a Avhole, or in part, to the ancestors of the people found inhabiting those sections at the. incoming of the whites. This has been found true in regard to Mexico and Central America, Methods of Study. 23 and is noAV generally accepted as true in regard to the regions of the Mound-builders and Cliff-dAvellers. It is therefore advisable to proceed upon this supposition in regard to other sections until evidence incompatible Avith this conclusion has been brought to light. Pre historic migrations, of Avhich frequent mention A\'ill be made herein, have undoubtedly taken place, for, Avithout this, population could not have spread over the continent, but this was a slow process AA'hich re quired ages for its accomplishment. Moreover, as numbers increased and cultivation of the soil began, the tribes necessarily became more and more sedentary in habits. This had progressed to that extent when Europeans made their appearance that most of the groups had long been permanent residents of the sec tions they Avere found inhabiting ; in fact, as will here after be seen, there are good reasons for believing that most of the larger stocks had developed into tribes substantially in the respective regions they were found occupying. As this development must haA^e required a long time, the presumption is justified, ex cept where shown by the evidence furnished by the monuments or language to be incorrect, that these remains are attributable in a general sense to the an- -cestors of the inhabitants of the respective sections. That there was still more or less shifting of tribes and to some extent of stocks through the fortunes of Avar and here and there the breaking away of one or more tribes from the parent hiA'e, is no doubt true, but that there Avas a greater degree of permanency than has generally been supposed, is also true, a fact Avhich is becoming more and more evident through the iiiA'esti- ^¦ations of late years. Therefore the nati\'es as Avell 24 Study I if North American Archaeology. as the monuments must be studied, and the language,. physical traits, customs, traditions, mythology and folk-lore of the natiA^es are important factors which the student must bring to his aid. Another fact AA'hich should be borne in mind by the student is the danger of basing conclusions on ab normal objects, or on one or tAvo unusual types. Take for example the supposed elephant mound of Wiscon sin which has played such an important role in most of the AA'orks relating to the mound-builders of the Mississippi valley, but is noAV generally conceded to be the effigy of a bear, the snout, the elephantine feature, resulting from drifting sand. Stones bearing inscriptions in Hebrew or other Old World characters have at last been banished from the list of prehistoric relics. It is wise therefore to refrain from basine theories on one or tAvo specimens of an unusual or abnormal type, unless their claim to a place among genuine prehistoric relics can be established beyond dispute. It is unfortunate that many of the important arti cles found in the best museums of our country are Avithout a history that will justify their acceptance, without doubt, as genuine antiquities. It is safe therefore to base important conclusions only on monu ments in reference to which there is no doubt, and ort articles whose history, as regards the finding, is fully knoAvn, except Avhere the type is well established from genuine antiquities. One of the best recent Avorks on ancient America is marred to some extent by want of this precaution. Mounds and ancient works are de scribed and figured Avhich do not and never did exist ;, 2Icthods of Study. 25 and articles are represented which are modern pro ductions. The method of study to be pursued depends very largely upon the extent to which it is to be carried and the lines to be followed. For the general reader and the individual Avho desires to obtain only a gen eral knowledge of the subject, and for the student who studies the subject merely as a collateral branch, the writer trusts that this work will suffice. But for him who wishes to enter more into details, it can only be what it purports to be, an introduction to the study. For the latter class, a general knowledge of what has been accomplished is necessary in order to avoid wast ing time and energy in going over beaten paths. The student devoting attention to local archaeology, that is, to the monuments and remains of a particular district, will, of course, acquaint himself first with the investigations which have been previously made in that district. However, this does not end with merely ascertaining what monuments have been discovered and located, which of them have been explored and what relics and remains have been obtained, but in cludes a careful study of the types and their relation to the types of the immediately surrounding regions, as archaeology, as a science, if it can be so called, is based largely on analogy. In this Avay he determines what are the prevailing types of the district and what are peculiar to it if there be any ; but this investiga tion in reference to a limited district or to particular classes of antiquities must descend to more minute details than A\"ill be necessary in making a general survey of the antiquities of a more extended area. It may, perhaps, be truly said that we are just en- 26 Study of North American Archaeology. tering upon this stage of archaeologic progress, and yet upon the result of such investigations must de-' pend the answers to some of the important problems relating to the prehistoric times of the various sections of our continent. Among the most abundant and generally distributed classes of prehistoric artefacts are arrow- and spear-points ; and though the varieties seemingly baffle attempts at classification, it will probably be possible to determine all the types of a limited district and thus obtain one means of com parison with the archaeology of surrounding areas. Celts will afford another means of comparison, and so on through the entire list both of monuments and relics. However, in order to study the monuments properly and their bearing on the questions relating to the pre historic times of the given locality, mapping is an im portant step. A local Avorker should have a map of his district Avith the localities of the antiquities marked thereon with symbols indicating the types. Maps and diagrams of the groups of works are of course necessary to intelligent study. In other words, the geographical relations of ancient works in a district as well as the relations of the indiA'idual works to each other in the groups are important. Although the mounds in the groups of the mound area of the United States appear to be usually placed without re spect to order or plan, yet in the southern states they are so arranged in many of the groups as to leave a central, open space or pla.'a, while in Wisconsin the arrangement in lines is an archaeological character istic of the region. The geographical distribution of types forms the chief aid in outlining culture areas. Methods of Study. 27 It is important in studying the types of the monu ments and of the artefacts to determine the essential features of each type. It is often true, especially in the case of imitative objects, that the type is con ventionalized to such an extent as to lose apparently every feature of the object of Avhich it was intended to be a representative ; yet the careful student, by tracing the variations and eliminations, will usually be able to determine the essential features and reach a correct conclusion. Without this study unessential characteristics may be given an undue prominence. There appears to have been a strong desire on the part of the aboriginal artists to introduce the eye and other face features into the Central American hiero glyphics, yet ill many of these they are non-essentials, being simply ornamental ; and the same thing is true in regard to many other antiquities. Nevertheless, these unessential features as to the type are important in comparisons, as they assist in ascertaining affinities and derivation where the type is widely distributed. The oUa or globular bowl has been and is yet a com mon type of pottery vessel among the Pueblo Indians of NcAV Mexico and Arizona, yet the Indians can, in most cases, readily decide from what Pueblo a par ticular vessel came by the ornamentation or other features unessential to the type. The student investigating the archaeology of a given district should, as above indicated, make himself ac quainted, so far as the data AA'ill permit, with the his tory, customs, beliefs, traditions, etc., of the tribes Avhich have inhabited that district. Of course it does not necessarily follow because it is known that the ancestors of the people found inhabiting a certain ex- 28 Study of North American Archaeology. tensive section, as Central America, Mexico, the Pueblo region or the mound area, were the authors of a large portion of the monuments of that section, that the ancestors of the people found in more restricted localities were the authors of the monuments of those particular localities. There are unquestionably some monuments in southern Arizona and northern Mex ico which can not be attributed to the ancestors of the tribes inhabiting or known to have inhabited the par ticular localities where these ruins are found. The same thing is true also of certain ancient works in the mound section of the United States. Although the works as a whole are attributable to the ancestors of the Indians of the section, some tribes who were mound builders may have become extinct through Avars or epidemics, others may have been forced to shift position, and still other tribes may never have adopted the custom of building mounds, yet the propo sition in its general application remains true. One object, therefore, of the local worker should be to de termine, if possible, what tribes or people were the authors of the Avorks of the district he is studying, Avhether those knoAvn to have inhabited the district, or others removed in prehistoric times. The first step in this investigation is to learn the customs, arts, etc., of the people who formerly inhabited that dis trict, as he may thus be enabled to determine the probability that they w6re the authors, or to eliminate them from the investigation. The a priori presump tion is that the local natives were the builders. Every elimination of a factor from the discussion of a prob lem is one step toward the true solution. In the study of types the method must, of necessity, Methods of Study. 29 be very largely geographical with reference to vari eties, if the object in view be to ascertain the distri bution of the different varieties. If the object be simply to trace the development of the type, the geo graphical distribution is of less importance. Mr. A. E. Douglass, of Ncav York, who has a large private collection, suggests in regard to museum collections a double arrangement for these purposes : First, an arrangement of specimens according to geographical distribution ; and second by varieties. As the study of types necessitates the examination of specimens, this plan, where practical, would undoubtedly be ad vantageous. One difficulty in these lines, which has not yet been overcome, is the want of a uniform and acceptable nomenclature ; but nomenclature seems impossible without classification, Avhich has not been accomplished except in regard to limited districts. This is a desideratum to which the attention of cura tors of museums is now being directed, and it is to be hoped, notAvithstanding the difficulties in the way, that they will find some means of classifying collec tions sufficiently to form a basis for names of types. In studying the monuments it Avill be found, as yet, advantageous to limit attempts at grouping or classi ficatory arrangements to districts or sections. Com parison can then be made with the works of other sections or districts, group with group, or class with class. By this type generalization or aggregation the contrasts or similarities are not only more apparent than by single comparisons, but are of much more im portance. By such comparison of the works of the mound-builders with those of the Pueblo region or Central America, the contrast is, so to speak, intensi- 30 Study of North American Archaeology. fied. Within the section or district some grouping, even though it be arbitrary, is absolutely necessary to progress, and without it discussion is impossible and general description of little value. In other words, the student can make but little progress in archaeology until he advances to what may be termed the generic stage. Mr. Holmes has adopted a most excellent method, both in his studies of the monuments and of the minor vestiges of art. He learns by a comparison of specimens or of individual monuments the essen tial characteristics of the different types under inves tigation ; then by means of outline figures or sketches brings the types pertaining to the same general class in their simplest form into comparison. See, for ex ample, his comparison of types of pottery vessels of a certain class shown in our Fig. 37, and his comparison of temple plans in our Fig. 96. Although the idea is not new, his application of it to the antiquities of North America which he has examined is clear, and serves to illustrate a plan which may well be fol lowed. Study may be in the field, in the museum or in the books. In the first case there are numerous practical questions which can be answered only by experience ; the student must therefore learn by practice or by ref erence to the experienced field worker. The Bureau of American Ethnology receives many letters inquiring as to the best method of exploring (opening) and in vestigating mounds, etc. Although the general direc tion, to note every thing so carefully as the exploration proceeds that a complete restoration in every particu lar could be made from these notes, would perhaps Methods of Study. 31 answer the inquiry, the following suggestions are added for the benefit of the young beginner : If the mound to be explored be one of a group, the first step is to make a full and complete description of the group, with diagram as heretofore suggested, noting carefully the topography of the area covered by the group, and of the immediately surrounding country. The plan should show the correct positions of the mounds, and their form and size (diameter and height) should be noted. In addition to the measure ment of the mound to be explored, a horizontal sec tion showing an outline of the base as seen from the summit, and a vertical section showing the contour of the longest diameter, should be drawn on paper, and of sufficient size to note spaces thereon, of a foot measured on the ground. The north and south points should be indicated on the horizontal section. These plans are for the purpose of inserting marks indicat ing the positions, horizontally and vertically, of the articles found as the exploration proceeds. These, with the notes naming the articles by corresponding numbers and giving the measurement as to depth and side, will be sufficient to locate the article in the mound, should its exact position ever become a ques tion of any importance. Such a question occasionally becomes important when the article is found to indi cate contact with Europeans, or is abnormal. In order to note the stratification it is best to dig a trench from side to side through the highest point, or center, and where the mound is of considerable size it will be well to run another at right angles to this. These should commence and end at the extreme outer margins of the mound and be carried down to the 32 Study of North American Archaeology. natural soil or subsoil as the case may be. When a skeleton or relic is found it should not ,be removed until it is well exposed and its character and position noted down. If a vault, tomb, wall or any thing of large size is encountered, the trench should be carried around this until it is fully exposed before being dis turbed. When the trenches are completed, the re maining portions of the mound can be removed, the same care being taken. Where the mound is of large size, sinking shafts and tunneling may have to be re sorted to. Care must be taken to mark all articles found, with numbers corresponding with those in the notes and on the sections. Of course the character and thickness of the strata and every other particular deemed worthy of remembrance should be noted doAvn. Photography will of course be advantageous where clear and distinct pictures can be obtained, but AA'ill not supply the place of sketches. As it would re quire too much space to notice all the variations from these suggestions and add additional ones necessary to meet the numerous peculiarities the explorer may en counter, we can only repeat what is stated above : Note every particular with such care that it Avill be possible from the description to completely restore the mound in every particular. As the author is familiar by personal investigation with the antiquities of the mound region alone, his suggestions in regard to those of other sections must be drawn from the Avorks of other explorers. Profit able suggestions in reference to the method of study ing the ruins of Central America and Mexico may be draAvn from Mr. Holmes' account of the celebrated Palenque group given in his ' 'Archaeological Studies Methods of Study. 33 among the Ancient Cities of Mexico." lie starts out by giving a sketch map of the locality. Then follow in order a "Panoramic View" of the group; the "Orientation and Assemblage," which results in shoAv- ing that the placement of the buildings Avould seem to be due to the natural features of the ground rather than to a regard for the points of the compass ; "Ma terials and Masonry" ; "Construction" ; "Substruc tures," or pyramidal basements; "Superstructures," or buildings ; under the latter he outlines the ground plans of the types, following with the profiles of con struction or elevation accompanied by outlines of ver tical sections, illustrating the mode of construction. This is followed with descriptions of the roofing, of the types of doorways, of pillars, stairways and other essential features of the buildings, the ornamentation being considered last. In his description of Monte Alban and Mitla, in addition to the description of the ruins and mode of construction, he goes back to the quarry in order to study the method of preparing the material from the initial stroke until the blocks of stone are ready for removal to the building site, and to learn what man ner of tools were employed and how used. This might be followed up from the work of others, as that of explorers of the Bureau of American Eth nology and of the Hemenway Expedition among the ruins of New Mexico and Arizona, but what has been mentioned will suffice to indicate the method these field workers have followed. It is something of an art to grasp readily the chief idea or plan of a group of ruins. When this is caught, the lines and parts 3 34 Study of North American Archaeology . are usually easily traced, though hidden from view until uncovered. The study in the museum, that is of articles in col lections, has been alluded to incidentally. The study of the literature, where not in aid of the study of the monuments and remains, is chiefly for the purpose of investigating certain problems. In this case the scope of inquiry is widened and the data furnished by the monuments and remains constitute but one of the factors ; language, physical traits, customs, traditions, mythology and folk-lore must all be brought into the investigation. This involves also an examination of the early histories, the accounts of navigators and ex plorers and of more recent discussions on the same topics. The student must bear in mind the fact that archaeology is based on particulars, on innumerable fragments, and that conclusions and theories to be correct, must, so to speak, be the figures formed when the fragments are rightly placed. This brief and far from complete outline of the method of study will, Avitli the present work, furnish some aid to the student Avho wishes to devote attention to North American archaeology, but the critical investigator is expected to open up new lines and bring to bear new argu ments on the questions Avhich arise. Arctic Division. 35 CHAPTER IV. ARCTIC DIVISION. As the archaeological data of this division are few, and their direct connection with the Eskimo and allied tribes is not questioned, the division is purely an ethnological one. However as the people at their entrance into the domain of history were in the stone age, the implements, utensils and other artefacts in use among them afford a means of comparison which can not wisely be overlooked even in this brief sur vey. Moreover this area furnishes the best field on the continent for the study of the culture of a primi tive people as indicated by their arts. Stone and bone implements found in the graves, mounds and refuse heaps of other sections are often serious puzzles to the archaeologist, because their use was discon tinued before the historical era and is not easily de termined. But in the Eskimo area few have been dis covered of which the use is unknown, almost every form having been continued in use until visited by European navigators. The knowledge thus obtained furnishes a key by which many an archaeological riddle may be solved. Monuments or Local Antiquities. — These consist al most AvhoUy of shell or refuse heaps, the remains of old iglus or Eskimo houses which were constructed in part of stone, and an occasional pile of stones heaped over a grave to protect it from wild beasts. 36 Study of North American Archaeology. No true mounds, inclosures or fortifications of a per manent character, have been discovered in the entire area. This statement will also probably apply to a considerable extent of country lying south of the northern Eskimo belt, as we are informed by Rev, A, G. Morice, who has resided for many years among the north-western Athapascans (or Dene) , that "throughout the whole extent of their territory, no mounds, inclosures, fortifications of a permanent character, or any earthen-works suggesting human agency are to be found." Numerous shell-heaps have been discovered in the Aleutian Islands. Such of these as have been exca vated are found to consist of two or three distinct strata, indicating, it is supposed, successive periods of occupancy. Prof. W. H. Dall describes the typical form as consisting of the following layers : First, or lowest stratum, composed almost exclusively of the broken tests or spines of Echinus, a few shells of dif ferent species of edible mollusks being intermixed ; the next layer above, composed chiefly of fish bones and shells, with an occasional bird bone ; above this was a layer characterized by numerous mammalian bones, of marine species, intermixed with bones of sea birds ; this was covered by modern deposits and vegetable mold. The following articles found in this refuse heap fur nish some indications, Prof. Dall thinks, of the ad vance in culture during the time it was being formed, though this has been questioned. In the loAver stratum a small hammer stone was discovered which had an indentation on each side for the finger and thumb, and bruises on the ends, indicative of use, probably Arctic Division. 37 for breaking Echinus tests. In the second Avere rude net-sinkers, stone knives, and spear-heads both of stone and bone, the latter distinctly barbed (Fig. 1) . Fig, 1. Bone Spear-hejid, Eskimo. These appeared in still greater abundance and varied forms in the mammalian stratum, from which were also obtained stone, bone, and horn skin-dressers, bone awls, stone adzes and lamps ; also carved articles, such as masks, and a single face-form carved on bone. One of the lamps is shown in Fig. 2. Bone and stone Fig. 2. Stone lamp, Eskimo. labrets were found in the upper layer of one of these shell-heaps and also in a cave deposit of corresponding age. One of the labrets is shown in Fig. 3, Prof. Dall, to whom we are indebted for the foregoing de scription of Aleutian shell heaps, discovered also in the same region the marks and remains of ancient villages. The method of building among the ancient 38 Study of North American Archaeology . inhabitants, who are presumed to have been Aleuts, was to excavate slightly, build a wall of flat stones rjk or of the bones of the larger whales, and bank this on the outside with turf and stones. The roof appears to have been form- ed usually of whales' ribs, cov ered with wisps of grass tied together Fig. 3. Labret, Eskimo. and laid on the rafters, then turfed over. The remains of ancient stone houses are found scat tered over the greater part of Arctic America, espe cially the eastern portion, even in sections no longer inhabited by Eskimo, as the Parry Archipelago and the northern part of East Greenland. These are ap propriated by the Eskimo of the present day for tem porary dwellings when they stop in the region where they are found. A figure of the remaining founda tion of one of these ancient structures is given in Fig. 4, from Kumlien. The purpose of the long kayak-like building figured in connection with the stone house is not known. Dr. Boas says he found a similar one twenty feet long, scarcely one foot high, consisting of two rows of stones, at Pangnirtung, Cum berland Sound, but nobody could explain its use. The remains of a number of these ancient stone houses, or iglus, have been found in the American Archipelago and about Cumberland Sound. Those in good condition have a long stone entrance, sometimes Arctic Division. 39 from fifteen to tAventy feet long. This is made by ¦cutting an excaA^ation into the slope of a hill. Its walls are covered with large slabs of stone, about two Fig, 4. Remains of an ancient Eskimo house, and a half feet high and three feet wide, the space between the stone and the sides of the excavation being afterward filled with the earth. The floor of the passage slopes upward toward the hut. The last four feet of the entrance are covered with a very large slab, and are a little higher than the other parts of the roof of the passageway. The slab is at the same height as the benches of the dwelling room, which is also dug out, the walls being formed of stones or whale ribs. These houses are supposed to have been covered in the same way as those already described. Dr. Boas states that he has found at Ukiadliving, among other remains, some very remarkable "store houses." "These structures," he says, "consist of heavy granite pillars, on the top of whicli flat slabs of stone are piled to a height of nine or ten feet. In 40 Study of North American Archaeology. winter, blubber and meat are put away upon these pillars, which are sufficiently high to keep them from the dogs; skin boats were also placed on them." This was doubtless the object in view in building these rude structures, but why the covering should be so thick and heavy is not apparent if this were the only object. Implements, Ornaments, etc. As all the monuments and minor vestiges of art of this division are attributable, as already stated, to the Eskimo, the earliest forms that are known differing but slightly from those of modern times, it is only necessary here to notice a few of the more important types for the purpose of comparison. As agriculture is impracticable in the rigorous cli mate of the Eskimo region, and the means of sub sistence limited to animal food, the variety of imple ments is not large. They consist chiefly of such as are used in killing and capturing the food animals — of which the larger portion are marine mammals ; the implements and vessels used in preparing and cooking food, and in preparing the skins for the va rious uses to which they are applied. The simplicity in the Eskimo manner of life, the necessary uni formity in their method of procuring subsistence, and the manner of clothing themselves, have convention alized to a great extent their implements and arts. As the struggle for existence has been a difficult one AA'ith them, and the clothes and dAvellings necessary to protect them against the cold are ill adapted to the use of ornaments, the variety of such articles is quite limited. Arctic Division. 41 The articles of stone and bone, which are the only ones requiring notice here, consist chiefly of arrow, spear and harpoon heads, skin scrapers, ulus or women's knives, adzes, lamps, cooking pots or kettles, flakers and labrets. The chipped flint heads of arrows and spears are usually well made, finely finished and symmetrically formed, differing in size and slightly in form accord ing to the particular purpose for which they were in tended. Some of the older specimens are somewhat ruder, but would undoubtedly be classed as neolithic. One of the most useful and necessary implements belonging to an Eskimo household was the Ulu or Woman's knife, which, with them, performed all that is done in enlightened communities with the various cutting implements of the butcher-shop and the household kitchen. The simplest form was a flake of flint with a cutting edge, but Avith the Eskimo they were usually made in a particular form, and, Avith the handle, resembled the ordinary kitchen chop- ping-knife, which, in I fact, has to a large ex- \^ tent replaced the stone implement. The blade Avas of horns tone, chert, or flint material and slate, especially the latter. (Fig. 5.) pjg. 5. Ulu, or woman's knife, Eskimo. Another indispensable household article Avas the lamp, Avhich furnished both heat and light. These were usually of soapstone, though a few of other stone have been discovered. The 42 Study of North American Archaeology. form of this vessel was not so strictly conventionalized as that of the kettle or cooking pot, though generally dish-shaped and shallow. (Fig. 2.) A semicircular form was also common, the length varying from six inches to nearly three feet. Before the introduction of European vessels the cooking was usually done in soapstone pots or kettles Fig. G, Soapstone pot, Eskimo. by placing them over the lamps or putting heated stones in the water. They were comparatively small, varying in capacity from a pint to a gallon, rectangu lar in outline with the sides perpendicular or slightly flaring. (Fig. 6.) Fig. 7, Hafted jade adze, Eskimo. Arctic Division. 43 Even at the present day, according to Mr. Murdoch, the Eskimo of Point Barrow use no tool for shaping large pieces of woodwork except a short-handled adze, hafted in the same manner as the old stone tools which were employed before the introduction of iron. (Fig. 7.) The skin scraper usually consisted of a blunt stone blade mounted in a short thick haft of wood or ivory, fitting ex actly to the inside of the hand and having holes or depressions to receive M^ k the fingers and thumb. (^Fio- 8 "1 ¦^^^" ^- ^^^° scraper, Eskimo. The art of making flint arrow and spear heads has not been entirely lost by the Eskimo. Flint pebbles are splintered by percussion into fragments of suitable size, and the sharp-edged spalls are flaked into shape with an implement consisting of a short straight rod of flint or bone mounted in a short curved haft grooved for its reception, (Fig. 9.) Fig. 9. Flint flaker, Eskimo. Culture-home of the Eskimo. The origin of the Eskimo or Innuit is a question which has been much discussed, but as yet remains undecided. The generally-accepted theory has been that they migrated from north-eastern Asia by way of Behring Strait. Recently, however, several writers, among whom are two or three who have made a 44 Study of North American Archaeology. special study of them, have reached the conclusion that they were originally an inland people of North America, and that their migrations were toward the north and Avest. This conclusion is based to a con siderable extent upon the evidence, now generally ac cepted, that the Asiatic Eskimo (the Yu.it) , dAvelling around East Cape and to the south of it, migrated in late prehistoric times from America, and that the Aleuts inhabiting the islands moved in the same di rection. As any opinion Avhich may be advanced on this question is at best but conjecture, the subject does not come properly within the scope of the present work. There is, however, a closely cognate problem which offers greater probability of final solution, and which is of importance in the study of the prehistoric times of our continent. As well stated by Dr. Rink, who has made this arctic people well nigh a life study, "In regard to the cradle of the Eskimo race, we have before all to discern between their original home and the country in which they developed their present culture, which is characterized by their capability of procuring means of subsistence in arctic regions, AA'here no other nation can live." He then points out some "necessary conditions for guessing the site" of this culture-home. Alluding to the vast shore line which Avas, so far as known, occupied by the Eskimo as its only inhabitants before their modern contact with the European race, he divides them into Eastern and Western, separated by Cape Bathurst. He assumes as a basis, Avhich is admitted to be correct by those who differ from him, first, that only one such culture-home can have existed, Arctic Division. 45 and second, that even this one must have been of relatively small extent. The extraordinary uniformity of the utensils, instruments and weapons common to all the widely-spread tribes or groups, and the com paratively slight variation in language, is suggestive of a common origin. He then shows from the vocabu laries of the different sections the identity of the names given by the Eastern and Western groups to the animals used as food, boats, vessels, implements, etc., giving a list which excludes the possibility of accidental likeness. To this is added the similarity in form and use of the vessels and implements re ferred to. The direction of the migration is assumed from the following facts : The gradual completion of the kayak with its im plements, and the art of using them. The gradual change of several customs in proceeding from the south and west to the north and east, namely, the use of labrets or lip ornaments ceasing at the Mac- Kenzie River, the use of masks at festivals ceasing in Baffin's Land, and the women's hair dressing gradu ally changing between Point Barrow and Baffin's Bay, and the change in the houses in certain par ticulars. These indicate that the movement was from the ex treme west, or Alaska, toward the east, and this Dr. Rink believes is the true solution of the problem. On the other hand, Mr. Murdoch and Dr. Boas, who have personally studied the race on opposite sides of the continent, believe the culture-home Avas in the interior about the south end of Hudson's Bay, whence they separated into three principal divisions, 46 Study of North American Archaeology. one going north-east, another north, and the other north-west. This opinion is based chiefly on the primitive art of the central region, the form of the sinew bow, and the westAvard movement above re ferred to. It would seem difficult, however, to ac count upon this theory for the adoption of the kayak and its accompaniments, and the application of the same terms throughout the extended region where they are found, often in widely separated groups, between which intercourse is exceedingly rare. The settlement of this question, which appears possible Avith the accumulation of data, is important to the study of ethnology. If the latter theory be correct, it will have a material bearing on the theories in re gard to the course of migration of the Indian popula tion south and west of this assumed inland culture- home, for it is not probable that any people who have acquired their habits in an interior area, and com paratively moderate climate, would leave it, except under strong pressure, to take up their abode in such inhospitable regions as they now occupy. All the implements and works of the Eskimo appear to be adapted to their peculiar conditions and their only means of subsistence and preservation of life. They are very largely those of a littoral and arctic people, developed through the necessity of procuring, to a large extent, subsistence from the sea and defend ing themselves from the cold without material derived from the forest. Many of the articles, it is true, are adapted to savage life in any section, whether in the interior or on the coast, whether in an arctic or temperate climate, but on the other hand many others are suited only to the conditions under which they Arctic Division. 47 live. Hence it must be assumed, unless valid reasons for a different conclusion are shown, that those pe culiarly adapted to the situation were developed in the area where they are found, or one similar in its conditions. Mr. Murdoch's suggestion that the use of labrets is a habit which has Avorked its AA'ay along the western coast of America from the south is Avorthy of con sideration, though it does not appear to strengthen his theory, but tends rather to support the opposite conclusion. Nevertheless it is not without support, and opens up a new line for thought and investiga tion, and furnishes an additional pointer to a par ticular region of the western coast Avhich possibly may have played an important part in the peopling of the continent. 48 Study of North American Archaeology. CHAPTER V. ATLANTIC DIVISION. This division includes geographically, as heretofore indicated, all that part of North America east of the Rocky Mountains north of the Rio Grande and Gulf of Mexico, except that portion embraced in the Arctic division and except also the area occupied by the northern Athapascan or Dene tribes. At the time Europeans began to plant colonies in this region it Avas occupied by Indians belonging chiefly to some four or five linguistic stocks. The northern portion from Labrador to the Rocky Mount ains, the central area east of the Mississippi from the lakes south to Tennessee, and a strip along the At lantic coast from the Gulf of St. Lawrence to Pamlico Sound, Avas occupied by the great Algonquian stock. Gathered about lakes Erie and Ontario, both north and south, stretching down both sides of the St. Lawrence to Quebec, and extending over Ncav York and most of Eastern Pennsylvania, was the Iroquoian family, be longing to which were outlying groups along the south-eastern border of Virginia, and about the head- Avaters of the Tennessee and Savannah rivers. The Muskhogean family occupied most of the area embraced in the southern states east of the Mississippi. Ex tending westward from the Mississippi river — from its headwaters to the Arkansas — across the broad plains of the west, and occupying most of the drainage area Atlantic Division. 49 of the Missouri and Arkansas rivers, was the Siouan stock, the Bedouin of North America, Belonging to this group were some scattered fragments, one along the piedmont region of Virginia and the Carolinas, and one of small size on the southern coast of Missis sippi and another in Arkansas. Besides these there were the Caddoan stock, chiefly in western Louisiana and eastern Texas ; the Timuquanan occupying the Florida peninsula, and some, few in numbers, covering small areas chiefly about the mouth of the Mississippi. The archaeological conditions we encounter in this area are so widely distinct from those of the Arctic division as to require but little thought or study to mark the differences. It is true we find here flint arrow- and lance-heads in abundance, some of them bearing a close resemblance to and scarcely distin guishable from those of the Eskimo. Chipped stones of a certain form, which are supposed to be skin- scrapers or skinning implements, are also found in great numbers, and though many of them may be compared with the flint points of the Eskimo scrapers, yet the manner in which they were hafted, or whether hafted at all, is in most instances only a surmise. It is noticeable that of the fifty-six American scrapers figured in Prof. O. T. Mason's "Aboriginal Skin Dressing" (Rept. Nat. Museum, 1888-9), all except five are Eskimo, and the five are adze shaped and have iron or steel points. The elbow-shaped handle may be a survival from the stone age, nevertheless it is possible that the advent of iron may have worked some change in form. Local monuments, as we have seen, except, refuse beaps, foundations of old iglus 3 50 Study of North American Archaeology. and some ancient graves, are unknown to the arctic section. On the contrary, in the area we are now en tering upon, the Mississippi valley, from the head waters in Minnesota to the Red River of Louisiana, and from the sources of the Ohio to the border of the western plains, is dotted over with earthen mounds, clustered into groups or scattered singly ; here and there hills and bluffs are crowned with defensive works, indicating tribal AVarfare ; throughout southern Illinois, Kentucky and Tennessee the rude stone sep- ulchers of the ancient inhabitants are found in great numbers ; and other evidences of prehistoric occupancy abound. Thus it will be seen that the difference archaeologically between the two divisions is a wide one.Monuments, or local antiquities. The antiquities of this class found in this division consist chiefly of earthworks, stoneworks, graves, cave deposits and mines and quarries, and might be classed under these heads but for the fact that some belong partly to one class and partly to another ; then there are certain other local antiquities which can not pos sibly be classed under either of these headings. If it were possible to decide positively as to the use of each type, this would afford one means of classification, but unfortunately here our knowledge is sadly at fault. However, as some arrangement for the con venience of reference is necessary, they will be grouped here by leading types under the following heads : Mounds, Refuse Heaps, Inclosures, Hut-rings, Excavations, Graves and Cemeteries, Garden Beds, Hearths or Camp Sites, and Ancient Trails. Besides Atlantic Division. 51 these there are Mines and Quarries, Cave Deposits and Petroglyphs. That the particular sense in which some of these terms are used in this work may be clearly understood, the following explanation is given : Mounds. The tumuli or true mounds, to which the term will be limited in this work, are the most common and most numerous of the fixed antiquities, being found in the valley of the Red River of the North from its source to its mouth, and here and there an isolated one in Canada ; throughout the Mississippi valley and ihe region south of the great lakes to the gulf they constitute the larger portion of the numerous groups, it being exceedingly rare to find a group in which they do not occur. Although the forms are various, they may be classed as conical tumuli, elongate or wall mounds, pyramidal mounds, and effigy mounds. The conical tumuli are artificial hillocks cast up with some special object in view, and not mere accu mulations of debris. The form is usually that of a low, broad, round-topped cone, but as at present found is, in consequence of wear and tear by the ploAV and the elements, often that of an irregular lieap, distinguished from the refuse heap only by in ternal evidence. They vary in size from a scarcely perceptible swell in the ground to elevations of eighty or ninety feet, and from six or eight to three hundred feet in diameter. The outline is generally approxi mately circular where they retain their original shape, though many are oblong or oval and some pear- shaped. Most of the Burial Mounds are of this type. The works to which the name "Elongate or Wall 52 Study of North American Archaeology. MfxUm> Cam^dery \ I Fig. 10. Plat of mound group, Wisconsin. Atlantic Division. 53 Mounds" is applied are certain linear earthen struc tures which seem to be confined almost exclusively to the effigy-mound region mentioned below. The only oxternal characteristic which distinguishes them from the oblong mound of the conical type is their wall- like appearance ; in truth the longer ones may be properly called walls, if we judge by the form alone. This characteristic is apparent even when the length is not great as compared with the width. Usually the length is from one hundred and fifty to three hundred ieet, though some are found as short as fifty while others extend to nine hundred ; the width varies from twenty to forty feet, and the height seldom exceeds four feet. They appear to be simple lines of earth cast up from the adjoining surface, but with what ob ject in view is unknown ; however, they are seldom used as burial-places, and even where so used it is ap parently an after-thought. (Fig. 10.) The typical form of the Pyramidal Mounds is the truncated, quadrangular pyramid ; some, however, are circular or oval and a few pentangular, but are distinguished from the conical type chiefly by the flat top or truncated form. In some instances, as in the Marietta group, Ohio, they are so reduced in height, compared with extent, as to assume the appearance of earthen platforms ; others have terraces extending outward from one or two sides, and others a ramp or roadway leading up to the level surface. In conse quence of wearing by the plow and elements the sharp outlines have, in many instances, been obliterated to such an extent as to render it difficult to determine the original form. With the exception of a few in Ohio, Indiana and northern Illinois, works of this 54 Study of North American Archaeology . 3 ¦ar ft' -~1 V .li|/', a3o fill I ; ¦' Atlantic Division. 55 type are limited almost exclusively to southern Illi nois, south-eastern Missouri, Arkansas, Kentucky, Tennessee, South Carolina and the Gulf States. The two most extensive groups in the division, consisting chiefly of mounds of this form, are widely separated ; one is located in Illinois a few miles east of St. Louis, which includes the giant Cahokia mound, and the other near Carthage, Alabama. The best examples of terraced mounds are found in eastern Arkansas, one of Avhich is shoAvn in the annexed flgure. (Fig. 11.) There is a somewhat different form from either of those mentioned which is intermediate betAveen the conical and pyramidal types, though classed here Avith the latter, as a personal examination by the Avriter of examples Avidely separated geographically, has convinced him that they are slight modifications of the pyramidal type Avith ramps. Examples of both forms are seen together in the "Rich Woods" group, south-eastern Missouri. In this class the main tumulus is really conical or oval, usually with a ramp extending outward on one side in the form of a ridge ; or oval in form and the whole upper surface slightly rounded and sloping toward one end. The most singular earthen structures found on the continent are those representing animals, and usually known as "Effigy Mounds." They are limited geo graphically, almost exclusively, to Wisconsin and the immediately adjoining portions of Illinois and Iowa; some two or three are found in Ohio and two in Georgia ; it is reported that some examples have been discovered in the "Bad Lands" of Dakota ; this, how ever, has not been confirmed. The animals whicli 56 Study of North American. Archaeology. are represented, so far as they can be determined, are those known to the modern fauna of the region occu pied, the supposed elephant mound being in all probability intended for a bear, as the proboscis ap pears to have been an accidental addition of shifting sand, varying in shape at different times, which had entirely disappeared when the survey under the author's direction was made in 1884. (Fig. 12.) Fig. 12. Elephant mound, Wisconsin. Examples of this type are seen in Fig. 10, The author may be excused for expressing his surprise at the truly imitative curving and rounding of the body of the animal in some of the examples which have come under his observation. Standing at the ex tremity of one which has suffered but little weather ing (as the bear in Fig. 10) , he Avas almost persuaded that the builders had the animal lying before them as a model. The greater number, however, are but rude representations, yet there is never any difficulty in assigning them to the proper classes. They vary Atlantic Division. 57 in length from fifty to four hundred feet, and in height from a few inches to four or five feet. Where placed near streams the heads usually point down stream. As a general rule, no special order appears to have been observed in the arrangement of mounds in groups, these being scattered irregularly over the area occupied, the position being governed to some extent by the topography. There are, however, some excep tions to this rule. A somewhat remarkable one oc curs in the region where the effigy mounds prevail. Here we frequently find the conical tumuli of a group arranged in one or two lines, usually straight or nearly so, and somewhat evenly spaced. This may be attributable in some cases to the topography, yet there are a number of instances where this arrange ment has been adopted on level areas of ample extent, and where no special reason therefor is apparent. What renders this the more interesting is the fact that in the same section lines of similar mounds frequently occur, where they are connected with one another by low embankments. An example of this kind is seen in Fig. 13. The surrounding walls of Fig. 13. Group of chain mounds, Wisconsin. the noted group in Wisconsin, known as "Aztalan," and an extensive group in Vanderburg county, In- 58 Study of North American Archaeology. diana, appear to be but slight modifications of the chain-mound type. As the elongate mounds are found in the same section, it is possible that the three types — lines of conical tumuli, chain mounds, as the connected series are named, and the wall mounds — are steps in an evolutionary process, probably from the solid to the separated. So far as mounds of these series have been exam ined, no evidence has been found to justify the belief that they were intended as burial-places. On the contrary, as they are usually Ioav and flattened, and frequently contain indications of fire, they are be lieved to be house or Avigwam sites. One of the groups containing mound series of these types is in the precise locality Winnebago Indians are known to have occupied. Although rich, dry alluvial areas in the vicinity of a stream or a lake were favorite localities with the mound-builders, the necessity for guarding against the approach of enemies, of being in the vicinity of a food and water supply, and other reasons which gov erned the location of native villages, varied this rule. Hence we find numerous ancient works on the re stricted summits of hills and bluffs, on the islets and hummocks in the midst of swamps and marshes, and along the narroAV valleys and even' defiles of mountain regions. Nor are they wanting on the bottom lands of large rivers where the area is subject to occasional overflow. From these facts may be legitimately drawn the inference that the ancient inhabitants who con structed these works were split into numerous hostile tribes, the stronger occupying the level and choice localities, Avhile the weaker were forced to seek refuge Atlantic Division. 59 in the rugged regions or amid the pwamps and marshes. Some of the effigies of Wisconsin occur on quite steep hillsides, and others on crested spurs AAdiere the summit is so narrow as to necessitate lapping over from one side to the other ; and some of the long mounds are found running directly or obliquely doAvn quite steep slopes. In some instances, as in Calhoun county, Illinois, and north-eastern Missouri, long lines of conical tumuli, usually showing evidences of burial, occur on the sharp crests of ridges so narroAv as to barely afford space for their construction. Oc casionally they are placed immediately on the margin of a precipitous bluff. Hundreds of groups, some of which are quite extensive, are located on the low ridges and hummocks in the swampy regions of south eastern Missouri and north-eastern Arkansas ; in fact, one of the richest archaeologic flelds of the Atlantic division is found in this section : it is pre-eminently the region of ancient pottery. The general distribution of the mounds and other ancient works of that portion of the division in the United States may be seen by reference to the map compiled under the direction of the author and pub lished by the Bureau of Ethnology in the 12th Annual Report. It is seen by examining this, that the areas Avhere these prehistoric works are most abundant are central and western New York ; eastern and southern Michigan ; the banks of the Mississippi from La Crosse, Wisconsin, to Natchez, Mississippi ; the central and south-western part of Ohio and adjoining portion of Indiana ; central and western Kentucky ; middle and eastern Tennessee ; and the south-west corner of 60 Study of North American Archaeology. North Carolina and north-east corner of Georgia. The east side of Florida is well dotted with shell- heaps. It would be interesting to refer to the sug gestions Avhich a study of this map brings before the mind, but this must be left chiefly to the reader. There are however one or two inferences which appear legitimate that may be properly mentioned here. One is that the greater numbers on some areas com pared with others is owing in part to the more thor ough exploration of these areas, yet it is not probable that future explorations will materially change the map in this respect. Another is that the statement frequently made by authors that the mound distribu tion continues through Texas is incorrect. It would also appear to be a fair inference, judging by the map, that there were no important movements of popula tion to or from the south-west. The almost total ab sence of mounds east of the Alleghany Mountains is also a marked feature. Burial Mounds. 61 CHAPTER VI. BURIAL MOUNDS. Having studied the form and external appearance of these silent monuments of the past, let us remove the sod with which the growth of centuries has covered them and examine the interior to see what it has to reveal, what it has to tell us of the past. Tombs are often the treasure houses of savages and semi- civilized people. Guarded by superstition the treas ures remain untouched until rifled by people of an other race Avho have no fear of the deity invoked for their protection. However, before seeking for the hidden treasures, we will try to ansAver the question, How did the ancient people do the work required in building these earthen structures ? Though a mound seems to be but a simple heap of earth that called for no skill, yet the question is a pertinent one. The mound-builders had neither iron nor steel of which to form spades and shovels, nor had they beasts of burden to assist in the trans portation of material. Stone hoes, wooden spades and bivalve shells were probably the chief implements they used for digging up the soil ; and baskets, mats and skins borne by individuals were most likely the means they employed for transporting the material. Nor is this wholly conjecture, as stone implements well adapted to this purpose, especially if hafted, are found in almost every section. .All these implements, 62 Study of North American Archaeology. as we are informed by the early explorers, were used by the Indians in their agricultural pursuits. The large, roughly-chipped, leaf -shaped stone implements so abundant in some sections, scores of which were found by the agents of the Bureau of American Ethnol ogy at a single point in southern Illinois, were doubt less used for this purpose. The thin-bladed, so-called grooved axes are supposed to have been used, when transversely hafted, partly as digging implements. It is often the case when a mound is carefully ex cavated and closely scanned as the work proceeds, especially where the material is clay or muck, that the individual loads can readily be discerned. As the earth of which the mounds are composed is usually gathered up from the surrounding surface, the interior will vary in color and character only as the soil so gathered up varies. This may be illustrated by a partial section of a Mississippi mound shown in Fig. 14. Here the lower stratum (No. 5) is black soil in lumps, or small masses, presumably the top soil of the surrounding surface ; No. 4 red earth in small masses ; No. 3 (the grey streak not numbered in the figure) red clay; No. 2 grey clay ; and No. 1 the top cover ing accumulated since the mound was built. How ever, very many of the mounds are stratified in such a way as to shoAv that this has been done intentionally, even where it was necessary to bring the material for one or more layers from a distance of a fourth, or a half a mile, or more. The places from whence material was taken to build the small or moderate sized mounds are seldom discernible at the present day, but depressions plainly mark the points about the larger works, as the Burial Mounds. 63 Cahokia and Etowah mounds and some of the in closures of Ohio and elscAvhere. In some cases the one act has been made to serve tAvo purposes, that is to say, the earth used to construct the mound or other work has been taken from one or two points so as to leave a basin-shaped excavation for holding water, or Fig. 14. Section of a Mississippi mound. to form a trench to serve as a protective moat, or for drainage or other purposes. In some cases the earth has been taken from a trench immediately around the mound. The latter are interesting, as it Avould seem therefrom that the comparative size of the mound had been determined before beginning the Avork. Mr. Gerard Fowke, who has had considerable ex perience in excavating mounds in various sections of 64 Study of North American Archaeology. the country and of almost every form known to the division, has expressed the opinion that a mound one hundred feet in diameter at base and twenty feet high, could have been thrown up by a hundred men, with the means the mound-builders had at hand, in forty-two days. Marquis de Nadaillac objects to this as sertion as one negatived by all the data obtained. How ever it is rather a question of practical mathematics than of archaeology. A simple calculation is all that is necessary to show that twenty-five loads, each con taining half a cubic foot of earth, carried per day by each man, would complete the mound in forty-two days. As the usual distance the loads had to be car ried was from fifty to a hundred yards, and the loose top soil was selected, twenty-five loads of half a cubic foot each is not an unreasonable allowance. The single loads, as plainly indicated by the little biscuit, or pone-shaped masses in many of the mounds, cer tainly exceed in size this estimate. It would appear, therefore, that Mr. Fowke was warranted in his con clusion. The internal arrangements or modifications relating to or having connection with burials are so various that only the more common and important can be re ferred to here. A type quite common in the north western portion of the division, is that, Avhere a slight excavation has been made in the original surface of the ground to receive the body or bodies, or more likely skeletons, as in many, if not a majority of cases of this type, the flesh has been removed before burial, the loAver limbs drawn up, or the bones disarticulated and bundled, or stretched out horizontally and the Burial Mounds. 65 mound heaped over them. It Avas not unusual to form the first or lower layer thrown over them of tough clay, which must have been, in some instances, in a plastic state when deposited, as may be judged by the way it has worked itself into the cavities of the skull. Sometimes the entire mound consists of this hard clay layer. In mounds of this class in trusive burials are readily distinguished from the original ones. The simplest method of burial, of which examples are found in most of the sections, was to lay the out stretched body or bodies on the surface of the ground and heap the earth over them. In Ohio and West Virginia some examples occur where the surface of the ground was first smoothed and packed : over this was spread a floor of bark, on which was sprinkled a layer of ashes a few inches thick. The body was then laid on the ashes and covered with bark, and over this the mound was heaped. In some cases the bodies are found in a sitting posture, and where there are several they are sometimes facing one another. It is probable, however, that some of the cases re corded, especially in the north-western section, were really bundled skeletons, the fact that the bones were in a heap, with the head on top, being taken as proof that they were originally in a sitting posture. In a majority of cases, no rule in regard to the po sition of the bodies relative to the cardinal points was observed. Fig. 15 shows the stratification of a mound in eastern Tennessee containing a large number of skeletons all in the lower layer {g, g,) . The explana tion of this figure is as follows : 5 66 Study of North American Archaeology. «o "a a, a, Dark layer of sandy soil, Ii feet thick. b, b, Thin layer of burnt clay, 3 to 4 inches. c, c, Dark sandy soil, 2i feet. d, d, Second layer of burnt clay, 3 inches. e, e, Dark sandy soil, Ii feet. /, /, Third layer of burnt clay, 3 I inches. i g., g, Dark mucky soil (about 4 feet) resting on the original surface of the 1 ground. I /i, Central shaft' of alternate dish- i shaped layers of burnt clay and ashes. 1 i, i, Remains of upright cedar posts. i Although all the skeletons were in i the bottom layer, they were not all, nor ' even the greater part, resting on the , original surface of the ground, but at : different depths. All were stretched 01 out horizontally except two ; one of these was in a sitting posture, and the other folded up and lying on its right side, and was probably buried after the flesh had been removed. It was judged from the indications that some, at least, of the burials were made in this way : the body, after being deposited, was covered with a layer of cane or brush ; over this was spread clay or muck in a plastic state, and upon this a fire was built. Among the relics found in this tumulus were earthen pots and basins, generally at the heads of the skele- Burial Mounds. 67 tons (Fig. 16); shell beads, shell ear ornaments (Fig. 17) and hair-pins ( ?) ; engraved shells similar to that shown at Fig. 18 ; soapstone pipes (Fig. 19) ; flint Fig. 16. Earthen pot, east Tennessee. arrow and spear heads ; polished celts ; discoidal stones ; bone implements ; and one iron chisel, which was by a skeleton. As the skeleton and iron chisel lying with it Avere in the layer, g, g, they must have been placed before the unbroken stra tum, /, /, and the other undisturbed strata above were deposited, and hence can not possibly be attributed to an in trusive, or even after burial . It is e vident that burials in the mound ceased when layers, /, /, and e, e, were deposited, unless these layers Avere cut, of which there was ^.^_ ^^^ gj^^^j ^^^ ornament or no evidence. hairpin. 68 Study of North American Archaeology. In another large mound in the same valley and be longing to the same series, the plan appears to have been exactly reversed : the bottom layer, which was level and not rounded on top, Avas not used for burial Fig. 18. Engraved shell, North Carolina. purposes, the heavy single layer above it containing all of the ninety skeletons unearthed. This valley of the Little Tennessee was occupied, from prehistoric Fig. 19, Soapstone pipe, east Tennessee. times until their removal, by the Overhill Cherokees, whose villages were located on the precise spots where the mound groups are found. Another form of burial has been observed in west- Burial Mounds. 69 70 Study of North American Archaeology . ern North Carolina. Here a circular or triangular excavation to the depth of two or three feet was made : the bodies (or skeletons) were placed on the bottom, usually in a sitting posture, and most of them covered with beehive-shaped vaults of cobble-stones (Fig. 20) . In one instance, in Eastern Tennessee, instead of an excavation, a wall was built of cobble stones on the surface of the ground, and the vaults arranged within it. Similarly shaped burial vaults, of hardened clay, have been discovered in West Vir ginia mounds. Many important relics were obtained from the North Carolina mounds ; among other things some of the flnest specimens of engraved shells which have been found in the United States (Fig. 18) ; also soapstone pipes with stems, bearing a close resem- Fig. 21. Soapstone pine, North Carolina. blance to the old-fashioned clay pipes of the whites (Fig. 21) , It is somewhat singular that although James Adair, in his "History of the American In dians," describes the soapstone pipes made by the Cherokees as precisely of the form of what is known as the "Monitor Pipe," mentioned below (Fig. 46) ; none of those discovered in North Carolina or east Tennessee mounds are of precisely that form, though probably modifications of it. Another important mode of burial, both in mounds and in cemeteries, was in box-shaped stone sepulchers, Burial Mounds. 71 These appear to have been constructed as folloAvs : In a pit some two or three feet deep and of the desired dimensions, dug for the purpose, a number of flat stones are placed to form the floor ; next, similar pieces are set on edge to form the sides and ends, over which other slabs are laid flat, forming the covering ; the whole, Avhen finished, making a rude, box-shaped coffin or sarcophagus. Sometimes the bottom layer Avas omitted. Graves of this kind occur often in great numbers in southern Illinois, Kentucky, middle and east Tennessee, north-eastern Georgia, and at certain points in Ohio, though the sections of greatest abundance are southern Illinois and middle Ten nessee. Mounds in these last-named sections are frequently made up almost entirely of sepulchers of this type, generally placed without regard to system and sometimes in two or more tiers. One or two, however, have been found in middle Tennessee, in Avhich the graves were arranged like the spokes of a wheel, the heads being toward . the center. In the center of the mound, the point from which the sar cophagi radiated, was a large clay vase or basin- shaped vessel. There were two rows of coffins, one outside of the other. Although the skeleton is usually stretched at full length on the back, in some cases the bones of adults have been disarticulated before burial, and packed into stone graves of this type not exceeding two feet in length and nine inches in width ; and occasionally two and even three skeletons are found in a single grave. A cemetery in Tennessee composed chiefly of small graves of this type was, for a time, supposed to be the burial-place of a race of pigmies, but a more thorough examination showed 72 Study of North American Archaeology. the graves to be the depositories of disarticulated skeletons and children. There is usually no special order in which these graves are arranged ; a cemetery exhibiting the greatest regularity of any yet dis covered is shown in Fig. 22. It is proper, however, •'¦'--¦ 'V. •¦• -Tf .-¦"'¦" .¦-¦¦ss|#:ffia-B%i§:««.5,. . 7 ^ ¦'¦•'¦^'^yfl/iiiiiivc^'' X^ £X ^-^^K ular-shaped bowl, or ^^^ — ^^s-^C ^C ^\=y oUa to the true bottle Fig. 33. Wide-mouthed bottles. shape. The typical forms of the body, as given by Mr. Holmes, are shown in Fig. 33. The eccentricities are usually in the shape Fig, 34(1, Opossum vase, Arkansas, given to the body, as the neck, which is sometimes merely a slightly raised rim, is generally plain. Animal forms are those usually adopted in these variations. (Figs. 34, a and b.) A remarkable vessel, representing the human head, is shoAvn in Fig. 35, a type, of Avhich some two or three specimens have been discovered. There is cer tainly nothing strongly suggestive of the Indian physiognomy in either of these ; on the contrary, the features will probably be taken at first glance, by Vessels, Implements and Ornam.ents. 93 Fig. 'iib. Sunflsh vase, Arkansas. Fig. 35. Bowl representing the human head, Arkansas. 94 Study of North American Archaeology. most persons, for Africans. Nevertheless, the more Ave study them the more doubtful does this con clusion become. The nose is small and the nostrils narrow. The Fig. 36. Winged and crested rattle- general appearance is snake design, Arkansas. that of a female . An in teresting design representing a winged and crested rattlesnake taken from an Arkansas bottle is shown in Fig, 36, Long-necked Bottles. — These form the chief feature of the pottery of the region now under consideration, due perhaps in part to <5 66 Fig. 37. Outline figures of long- necked bottles. the endless variation of which the type is susceptible. Both neck and body in this class are modeled ap parently according to individual fancy rather than after conventional forms. Although animal figures are not uncommon, the human form, especially that of the female, is most frequently rep resented. Outline representations of some of the simple forms are given in Fig. 37, and of some of the eccentric forms in Fig. 38. Some of these vessels were furnished Fig. 38. Eccentric shapes in long-necked . , , . . , bottles. """"'^^ *®®*' eit^ier three knob -like, cylindrical or terraced feet, or a single solid or per- Vessels, Implements and Ornaments. 95 forated foot. The specimen shoAvn in Fig. 39, repre senting an owl, is interesting, as the same pattern Fig. 39. Owl-shaped bottle, east Tennessee. and decoration are found in New Mexico, eastern Ar kansas and eastern Tennessee. 96 Study of North American Archaeology. There is no apparent reason why we may not assume that the making of pottery began in this division coeval with the commencement of mound- building. Is there any reason to believe the manu facture had been discontinued when Europeans appeared on the scene? It is well known that the Indians of the Gulf States and south Atlantic coast were inaking pottery when visited by the early ex plorers. The Gentleman of Elvas, one of the chroni clers of De Soto's expedition, declares that the vessels of earthenware used by the natives (apparently alluding to the region of eastern Arkansas) , differed little in quality from the Spanish ware. DuPratz, alluding to the same region, says : "The women make pots of an extraordinary size, jars with medium-sized openings, bowls, two-pint bottles with long necks, pots or jugs containing bear's oil, Avhich hold as much as forty pints, and finally plates and dishes in the French fashion." And other writers speak of the Indians of the south making pottery down to compari- tively modern times. DuPratz also speaks of their coloring vessels red. Now, it is apparent that we have in these notices mention of the same kinds of vessels as are found in southern mounds, even to the coloring, for this is often present on pottery from Arkansas and south-eastern Missouri. If vessels were made in great numbers within the historic period in the same region as those found in the mounds, "it is reasonable to suppose," as Mr. Holmes says, "that they belonged to the great group of those under discussion. If not, it will be necessary to seek the cause of their total disappearance, since, Vessels, Lmplements and Ornaments. 97 Fig, 40. Burial urn, Georgia. as I have said, the pottery of this district, as shown by the relics, is practically a unit." The Gulf Province. — As the pottery of this province bears a strong resemblance in form to that of the middle section, we Avill notice here only two or three types, which in form or decoration present different features from those described. The most remarkable of these is the so-called "burial urn," found in some mounds of Georgia and South Carolina, one of which is shown in Fig 40. Some of these are the largest vessels made by the mound-builders, unless the supposed salt-pans, of which no complete specimen has been found, exceed them in size. The moccasin-shaped pot, of which one or two specimens have been found, is a very rare form. The vessel shown in Fig. 41 bears a close resemblance to the modern iron pot, being furnished with four legs, which is unusual in mound pottery. Notwithstanding the frequent mention, in works re lating to prehistoric America, of the vessels found in Ohio mounds by Messrs. Squier and Davis, ancient pottery is rare in that state. Mr. Moorehead, who has done much exploring in the state, remarks that "pottery is very rare in the tumuli of any section of Ohio ;" however, the area about Madisonville appears to be an exception to this rule. 7 Fig, 41. Vessel with four legs, Georgia. 98 Study of North American Archaeology. Pipes. Judging by the number of pipes which have been found in mounds and graves, the ancient inhabitants of this division must have been sturdy smokers. However, the distribution of these articles is by no means uniform, as they are comparatively rare in some sections and abundant in others. This distribu tion, if thoroughly worked out, even with the mate rials so far obtained, would furnish valuable hints as to culture areas and ethnic relations. It is noticeable that their distribution does not correspond with that of the pottery ; on the contrary, they are usually more abundant in the regions where earthenware vessels are comparatively rare, and of less frequent occurrence in the great pottery section. They consti tute a marked feature of the archaeological collections of Canada and some other portions of the northern area ; also of eastern loAA'a, northern Illinois, Ohio, and what may be termed the Cherokee, or Appalachian district. They are rare in Arkansas and south-eastern Missouri ; somewhat more common, though by no means abundant, in middle Tennessee and the Gulf States. Articles of this class were made of clay or carved out of slate, soapstone, marble, or other stone. The variety of forms which individual fancy has intro duced is almost endless, yet it is possible that they may, omitting from consideration the ornamentation and fanciful figures, be classed in a general way under the following types : The stemless pipe, consisting simply of a bowl with an opening for the stem. Some of the simpler forms of this type are shown in outline in Fig. 42 (a, b,) . Vessels, ImpJeinents and Ornaments. 99 Of these, a and b are found almost exclusively in the northern section, and are known to have been in use among the Indians. Usually they are simple boAvls, cylindrical, ovoid, or flaring at the top, or curved in the form shown in a, and more or less ornamented ; others represent the human form, or animal figure, or Fig. 42. Stemless pipes. Fig. 43. 'Image pipe, Georgia. some grotesque shape. Although pipes of this type are rare in the middle and southern districts, that shown at Fig. 43 appears to have been a somewhat favorite form Avith the Georgia mound-builders, and has also been found in middle Tennessee. There is another stemless type, AA'hich, though of rare occurrence, should not be omitted from this brief summary. Specimens have been found, so far as known, only in Arkansas, Alabama, and Mississippi. They are of large size, varying from three to five inches in length, and from two to four in height. These usually represent a crouching, panther-like ani mal, or a man in the same position (Fig. 44) . It is possible they were only used as ceremonial objects, and hence considered public property. If this Avere 100 Study of North American Archaeology . Fig. 44. Image pipe, Arkansas. Fig. 45. Short-necked pipes. Vessels, Implements and Or7iaments. 101 so, it is not likely they would have been buried except on some unusual and memorable occasion. They are carved out of stone. Another type or class is the short-necked pipe. The primary or typical forms are seen in the figure (Fig, 45, a, b, c,). These appear to present three va rieties : the upright square bowl, the upright round, and the slanting bowl. The eccentric forms of this type are not numerous, consisting chiefly of a modifi cation of the bowl to represent a human or animal head. They are found both of stone and clay. A fourth type is the so-called "monitor pipe," in which the peculiar feature is the broad, flat, and usually slightly curved base or stem, which projects beyond the boAvl generally to an extent equal to the perforated end (Fig. 46). They are va ried indefinitely by the addition of ani mal and other fea tures, these modifi- tions being confined „. .. ,, ,. .^iwii ^ ^ p^g, ^g Monitor pipe. chiefly to the bowl. The typical forms are confined chiefly to Ohio and the region of eastern Iowa and the adjoining portion of Illinois. A slightly modified form has been found in Canada, New York, Massachusetts, West Virginia, and middle Tennessee. They are found only of stone, usually slate or steatite. The long stem pipe, or rather pipe with a distinct stem, forms a fifth type. This type has been found very rarely, except in the northern and Appalachian dis tricts ; and the forms in these two sections are quite 102 Study of North American Archaeology. distinct. Those of the northern section are of clay, those of the southern always of stone, usually soap- stone. Mr. Boyle says the method of forming the Canada pipes of this class was to model the clay round a flexible twig or thong, one end of which en tered the base of the bowl, and which, being allowed to remain there, disappeared during the burning process. A sixth class is that embracing the elongate animal flgures with the bowl on the back. The animal is sometimes a bird, sometimes a wolf, fox, or other quadruped, with legs drawn up against the sides of the body. Some are of very large size, and many of doubtful antiquity. They are all of stone, and aro conflned chiefly to Tennessee and the Gulf States. A few specimens of the tubular pipe have been found in the Atlantic division, but these have been apparently introduced, or they are simply tubes hav ing somewhat the form of the Pacific or California type. There is no reason for believing that a pipe of this form was ever in use in the Atlantic division. Articles of Shell, Copper, etc. 103 CHAPTER VIII. ARTICLES OP SHELL, COPPER, AND OTHER MATERIALS, Shells appear to have been used quite extensively among the mound-builders as implements and orna ments and probably as a medium of exchange. We have already noticed the fact that pulverized shells Avere used in tempering clay which was to be manufactured into pottery. It is probable that bivalve shells were used as scrapers in dressing hides, and to a limited extent as agricultural implements, as it is known that such use was made of them by the Indians along the southern coast. As they were not carved, it is probable they were not considered of sufficient value to accompany the dead. Never theless, it is not a very unusual thing to find unwrought shells in mounds. The use of certain large univalves, especially the Busy con perversum, as drinking cups, probably on ceremonial occasions, seems to have been somewhat general in the southern section, and not entirely un known further north. Specimens of the species named have been found as far north as the head waters of the Mississippi river. The specimen shown in the figure is from an Arkansas mound. (Fig. 47.) Mr. Holmes, in his paper entitled "Art in Shell," Second Annual Report of the Bureau of American Ethnology, figures a number of shells which he 104 Study of North American Archaeology. thinks were used as spoons or ladles. Most of these were made from the left valves of Unios. Fig. 47. Engraved shell, Arkansas. Much the larger portion of the articles of shell found in mounds and ancient graves consists of those which have been used as ornaments. Articles of Shell, Copper, etc. 105 Mention has been made of the shell pin, and an ex ample shown in Fig. 17 ; another form with a smaller head is frequently found. Various suggestions as to the use of these articles have been presented, but it is probable the following quotation from Dumont's "Memoires Historique Louisiana" Avill give the cor rect explanation : "There are still to be seen on the seashore beautiful shells made by snails (or limapon) , which are called burgaux ; they are very useful for making handsome tobacco boxes, for they bear their mother-of-pearl AA'ith them. It is of these burgaux that the native women make their ear-rings. For this purpose they take the end of it which they rub a long time on hard stones, and thus give it the form of a nail furnished Avith a head, in order that when they place them in their ears, they will be held by this kind of pivot. For these savages have much larger holes in their ears than our Frenchmen ; the thumb could be passed through them, however large it might be. The sav ages also wear around the neck plates made of pieces of these shells, which are shaped in the same manner on stones, and which they form into round or oval pieces of about three or four inches in diameter. They are then pierced near the edge by means of fire and used as ornaments." It is evident from this that they were worn in the ears ; whether used in any other way is a mere surmise. In the same quotation mention is made of the shell gorget, the most elegant of the shell ornaments. One type has already been noticed and figured (Fig. 18) . This has engraved on it what appears, from the num ber of specimens bearing the same figure found in 106 Study of North American Archaeology. different sections, to be a conventionalized representa tion of the rattlesnake, to which was probably at tached some sacred, talismanic or superstitious sig nification. Shell gorgets are found Avith various other designs engraved upon them. Another somewhat common form is shown in Fig. 48. This form appears to have been confined to what is now middle Tennessee. A few have been discovered bearing designs which are strongly suggestive of Mexican or Central Ameri- „, , „ can origin ; one of these from Fig. 48. Shell gorget. Ten- , ^ ° ' ., ^ . . nessee. *^^ Etowah mound, Georgia, is shown in Fig. 49. Another in- Fig. 49. Shell gorget, Georgia. Articles of Shell, Copper, etc. 107 teresting variety is that bearing the figure of a spider. The few specimens of the latter Avhich have been dis covered are mostly from southern Illinois and south eastern Missouri, one coming from eastern Tennessee. Another class of shell ornaments represents more or less distinctly the human face. They are supposed to have been used as masks. These have been dis covered in greatest abundance in the mounds of Ten nessee, but their range is quite Avide, examples having been reported from Kentucky, Virginia, Illinois, Mis souri and. Arkansas, and a somewhat different type from Alabama, Georgia and New York. The class of shell articles found in greatest abun dance is that including the various types of beads. The simplest form is the perforated small univalve, the species most commonly used being the Marginella, Oliva, and Cyprea. One of the most common varieties is the discoidal or button-shaped bead with a hole through the middle. The cylindrical form is also of frequent occurrence. Articles of this class appear to have been in use among the mound-builders of almost every part of the Atlantic division, however they are of most frequent occurrence in the middle and south ern provinces. The extensive use of shell beads or "wampum" as currency among the Indians of the Atlantic coast is a well-knoAvn historic fact that re quires no proof here. That this custom should have been brought about by Europeans at the early date it is known to have been in vogue, is simply impossible, as it is spoken of as a native custom by the first navi gators who visited the continent. It is therefore a reasonable presumption that it had come down from prehistoric times. 108 Study of North American Archaeology. Perforated fresh-water pearls have been found in large numbers in a few Ohio mounds, and specimens have been occasionally unearthed in other sections. Textile fabrics. That cloth was manufactured to a considerable ex tent by the mound-builders has now been demon strated by incontestible evidence. It has been found in several instances attached to copper articles, around which it had been .wrapped and by which it Avas preserved. Examples of this kind have been discovered in eastern Iowa, Illinois, Ohio and Georgia, and probably elsewhere. Cloth has also been found in caves of Kentucky, in some instances Avith mummified or desiccated bodies. A fine exam ple was obtained by the Bureau of American Ethnol ogy from a cave deposit of eastern Tennessee. Ac companying this AA'as an almost complete mat, with the submarginal stripe quite distinct. The burial in this case was apparently comparatively recent, but the tissue of the cloth and the accompanying bone needles are precisely of the type of some of the mound articles. Remains of charred cloth have also been discovered in mounds. Mr. W. H. Holmes has clearly demonstrated that many of the designs on mound pottery have resulted from the pressure of cloth on the surface Avhile the vessel was yet compara tively soft. Probably the vessels had been wrapped in cloth to keep them in shape. By taking impres sions in clay from these he has been enabled to give various patterns of meshes and cord. Matting was probably in common use among the mound-builders, but, like cloth, being subject to early Articles of Shell, Copper, etc. 109 decay Avhen buried in the soil, comparatively few specimens have been discovered. Reed matting was found in connection with the copper articles of the Etowah mound of Georgia, hereafter mentioned. A somewhat remarkable discovery was made by the agent of the Bureau of American Ethnology in an Arkansas mound. This was a layer of burnt matting three inches thick, lying immediately under a layer of burnt clay some six inches thick. This layer, for a considerable space, consisted entirely of burnt mat ting, through which were scattered parched or burnt grains of corn. The mound in which this discovery was made is a large one. Copper articles. Copper appears to have been used to a limited ex tent in almost every part of the mound division. Although it is probable that more articles of this class have been collected in Wisconsin than in any other district, it is doubtful whether this statement will ap ply to specimens obtained from mounds, excluding those pertaining to intrusive burials. Numbers have been discovered in mounds of Iowa, Illinois, and West Virginia ; also in mounds and stone graves of Kentucky, Tennessee, North Carolina, and northern Georgia. Implements of this class appear in considerable numbers among the archaeological collections of Can ada, consisting chiefly of arrow-points, spear-heads, adzes, celts, and knife-blades. But copper ornaments appear to be comparatively rare in this section, and consist chiefly of beads. Some of the knife-blades bear such a strong resemblance to those of European 110 Study of North American Archaeology. make as to lead to the conclusion that they are post- Columbian. One of the most important finds of copper articles in an Ohio mound Avas that made by Mr. W. K. Moorehead in the Hopewell mound, Ross county. At the head of what appeared to be the principal person age buried here, were imitation elk-horns, neatly made of wood and covered with sheet copper rolled into cylindrical form over the prongs. These meas ured twenty-two inches in length, and were fitted to a copper cap or covering over the skull. This is cer tainly a unique specimen, as no other similar article has been found. However, wooden ear-ornaments overlaid with copper have been discovered in a stone grave in a mound of southern Illinois and in an Ohio mound, and Mr. Clarence B. Moore makes mention of them in his splendid work on the sand mounds of Florida. Copper-covered wooden spools were ob tained by the Bureau of American Ethnology from the HoUoway mound, eastern Georgia. In addition to the gorgeous copper head-gear of the chief personage of the Hopewell tumulus, there were copper plates on the breast and stomach, also at the back. The copper had preserved the bones and a feAV of the sinews, also traces of cloth similar to coffee- sacking in texture. Copper spool-shaped objects and other articles were also found with this skeleton. It is Avorthy of notice in passing, that the skulls of this and of some other mounds of the group were of two types, the long-head (dolichocephalic) and short-head (brachycephalic) ; a fact also true of one of the tumuli of Caldwell county, North Carolina. One of the most important and puzzling series of Articles of Shell, Copper, etc. Ill copper articles found in the entire division is that ob tained by the Bureau of American Ethnology from the Etowah group of Bartow county, Georgia.* These consist of thin, even plates of copper, with impressed figures, some of which remind us of Mexican designs (Fig. 50) . Another was that of a bird of the same Fig. 50. Figured copper plate, Georgia * This group is frequently referred to by writers as " on the farm of Col. Tumlin." 112 Study of North American Archaeology. type as one discovered by Maj. J. W. Powell near Peoria, Illinois (Fig. 51) . Other plates of the former type were discovered in a mound of Richmond county, Georgia ; another, with dancing figures, in a stone grave of southern Illinois (Fig. 52) . The skeleton of Fig. 51. Figured copper plate, Illinois. Fig. 52. Figured copper plate, Illinois. the Etowah group, with which the plates were found, was in a box-shaped stone sepulcher, and indicated a man of almost giant stature, being seven feet long. The head rested on the copper plate shown in Fig. 50. This copper had been wrapped in buckskin, and around this had been Avound cane matting. Other types of copper articles found in mounds and ancient graves are celts, so-called axes, beads, disks, gorgets, spindles, ear-pendants, rings, bracelets, etc. The celts, axes, gorgets, spindles, some of the beads, and most of the bracelets, show very clearly that they have been beaten out of malleable copper with the Articles of Shell, Copper, etc. 113 rude implements of the natives. But some of the beads, ear-pendants, and bracelets, and also some of the plates, are formed from sheet copper, as smooth and even as that of European manufacture. There is little doubt that most of this copper, AA-hich was not introduced by Europeans, was obtained from the Lake Superior mines. Some was possibly ob tained from drift copper. Nor is there any reason for attributing the mining or working of this copper to any other people than the Indians. Prof. R. L. Packard, an expert mining chemist, who is personally familiar with the mines of the Lake Superior region, and who has thoroughly studied all the facts bearing on the history of these mines, comes to the conclusion that the ancient work done in them was done by the native Indians. In fact, he shows by positive histor ical evidence that precisely such work as was done by the ancient workmen in the mines of Keweenaw Point, Ontonagon and Isle Royale has been done by Indians, but he warns us that the word " mining," as used in this connection, is not to be taken in its true and technical sense, as there is no evidence that more was done in prehistoric times than simply to expose the copper masses and beat off from these such par ticles as they could with their rude implements. In some cases they heated the mass by building a fire on it, and then suddenly cooled it by throwing cold water over it. Articles of stone. As it is not possible in this brief survey to mention all the types of the minor art products found in con- 114 Study of North American Archaeology. nection Avith the antiquities of the mound-builder, only the typical and more important of this class will be alluded to, it being assumed that the reader is more or less familiar Avith the forms of the chipped stones. The most important of the class are the human images, of which, however, but few have been dis covered, and these are confined, geographically, to Georgia, Tennessee, and southern Illinois. One found by the side of a skeleton lying in a boat-shaped vessel of clay in a mound of eastern Tennessee is shown in Fig. 53. Stone image, Ten nessee. Fig. 54. Stone image, Tennes see. Fig. 53. Another was plowed up near the Etowah mounds representing a sitting female : it is of peculiar Articles of Shell, Copper, etc. 115 interest as showing the same type and apparently one method of arranging the hair. Another similar, but male, image has recently been discovered in Tennes see, the lower parts of which are more perfectly AVorked out (Fig. 54) . With the exception of one found in Union county, Illinois, and fragments of others obtained from the Etowah group, Georgia, the other examples are from middle Tennessee. In most of these, the oblique or sloping face indi cates a modeling after the artificially flattened head. Another noticeable fact is the strong similarity in facial type of a majority of the specimens. This is true of one found at Etowah, Georgia, four from mid dle Tennessee, one from Kentucky, and one found in southern Illinois. It is Avorthy of notice that, Avith one exception, the images found in the mound section have been obtained from the stone grave area. Another class of stone articles supposed to have been used for ceremonial purposes, and to which the name "banner stones" has been applied, is repre sented by the types shown in Fig. 55. A series of Fig. 55. Banner stones. outline forms of the various types of arrow-heads of America, as prepared by Mr. Henry C. Mercer for the Columbian Exposition at Madrid, is shown in Fig. 56, Of these, Nos. 1-29 are from the United States ; 30-34, 116 Study of North American Archaeology . Nicaragua; 35-42, Uruguay; 43-48, Argentine Con federation ; 49-50, Alaska ; 51-53, Costa Rica ; 54-58, Greenland ; 69-64, U, S, of Colombia ; 65-67, Alaska ; a|)^-^> .n3KBfiij^ 1 "¦¦ ^^aiaaajig; D' "1 ! Fig. 98. Plan of ruins at Uxmal, mounds, Avhose superstructures, if any ever existed, have disappeared. The area covered by the main group of ruins is not large, probably not more than half a mile 19 290 Study of North American Archaeology. square, but scattered remains are found beyond this limit. "The place, when inhabited," remarks a re cent visitor, "must have been extensive and impor tant, and no doubt presented a brilliant and imposing effect. Though the buildings are now much dismah- Fig. 99, Ornamentation on the governor's palace, Uxmal. tied and buried in a deep forest, save where recent clearings have been made, they are still impressive in the extreme, and it is difficult to realize that the huge pyramidal masses, rising like hills above the general level, are really wholly artificial." Monuments of Cerdral America. 291 The Casa del Gobernador, or Governor's House, reared on the uppermost of three successive colossal terraces, and forming the large central ground plan in Fig. 98, is the most extensive, best known and most magnificent monument of Central America. The sec ond of these terraces forms a broad esplanade in front of the building ; the third, set back somewhat toward the rear, is long and narrow, so as to leave a promenade of thirty feet around the house. The latter is ex cessively long in proportion to the Avidth, the length being about 325 feet, while the width is only forty feet. The height to the level top is twenty-six feet, nearly one-half of this height on the exterior face be ing occupied by an immense, profusely ornamented frieze, ten feet wide, running entirely around the four Avails of the building, a distance of about 725 feet. This elaborate ornamentation (Fig. 99) , Avhicli is all in wrought stone, consists of a checkered or lattice background ; Greek frets, series of bars terminating with serpent heads, the interspaces being covered with hieroglyphs ; human figures with immense head-dresses over the doorways (the human figures have all been broken away) ; and an upper line of great stone masks, with long, curved, proboscis-like noses. The other facing of the walls without and Avithin is of the gray limestone of the region, in large, squarish blocks, generally cut and laid with great precision, and, with few exceptions, plain. This long and narrow building is divided length- Avise into two series of rooms by a middle wall, the entrance being from the front ; the rear wall is nine feet thick and without opening, except at the recesses near the ends. The width of the rooms is limited by 292 Study of North American Archaeology. the span of the wedge-shaped arch of the ceiling which rarely exceeds ten or twelve feet. (Fig. 100.) Fig. 100. Section of the Casa del Gobernador. The so-called Nunnery is a great quadrangle, con sisting of four rectangular structures surrounding an open court, which stand on terraces, leaving open spaces at the four corners, the south building alone having an entry way through it. The inner fagades facing the court have, on the upper part, like that of the Governor's House, a broad, richly-ornamented frieze reaching from the top of the doorways to the flat roof. These, in variety of designs and delicacy of finish, probably exceed those of the Governor's House. Among these designs the great snouted mask is prominent, being found on all the fronts, and on the north side is placed in vertical tiers of five or six at the corners and over alternate doorways. These masks are formed by the arrangement of minor feat ures and are rectangular in outline. They are prob ably intended to represent the Tlaloc or rain god of Monuments of Central America. 293 the Mayas ; whether Itzamna or Cukulcan, is uncer tain ;- the writer is of the opinion that the former is intended. (Fig, 101.) A similar arrangement of some what similar faces is seen in some of the designs of the north-west coast (Fig. 70) ; even the great nose is some times present, especially on their totem posts, but in the form of a bird's bill instead of the elephant or tapir snout. The upward curve in the mask snouts of Central America give the idea that it was derived in some way from the elephant form instead of from the tapir snout, which has a slightly downward curve. However, figures in the Dresden Codex evidently in tended to represent tapirs have the snout curved up- Avard fully three-quarters of an entire circuit. Next to the masks the most im portant feature in the ornamentation of these fagades is the serpent, the arrangement of which along the face and around the panels is con sidered a masterpiece of decorative sculpture, and brings to mind the introduction of the serpent in the sculpturing of some of the temples of Cambodia. As an entire chapter would not suffice to describe all the ruins of this ancient city, brief reference to. but two more can be The Temple of the Dwarf or Magician palace, Uxmal. is noted as a prominent object of the group because of the very steep pyramid on which the building stands. The temple itself is small and in significant ; the unusual feature of the pile is a temple , , Fig. 101. Ornamen- made here, tation of the Nun's 294 Study of North American Archaeology. built against and into the north side of the pyramid, its roof being on a level with the top of the pyramid. The front of this temple is about twenty-two feet square and is entirely covered with ornamental work. The large doorway is occupied by a colossal snouted face or mask twelve feet square, made up of striking and unusual details. Among these was probably a life-size statue (now lost) standing on the snout and resting against the forehead ; others are a pair of tigers. The corner decorations comprise smaller masks, seven in each tier. The House of Pigeons (Casa de Palomas) , as will be seen by reference to the plan sketch, is a quadran gular structure placed against the terrace of a pyramid. Its prominent features are an arched opening of un usual size through the front building and the immense serrated, perforated comb which rises above the front wall. The facing of the buildings is of cut stone ; the fill ing of the walls and mounds is, as usual, of broken stone set in a liberal matrix of whitish mortar made of lime. "The facings and ornaments," says Mr. Holmes, "are all cut and sculptured with a masterly handling not surpassed where chisels, picks and ham mers of iron and steel are used, and the faces and con tact margins are hewn with perfect precision. Though the finish of the surfaces was often secured by means of abrasion or grinding, picking or pecking Avere the main agencies employed, and the indents of the tool are often apparent and AvonderfuUy fresh looking." This city is supposed to have been built or enlarged by the Tutul-Xiu — of whom further mention will be made in a future chapter — and their reign, with Monuments of Central America. 295 Uxmal as their capital, was the most glorious period of Mayan history, probably extending from the early part of the twelfth century until after the fall of Mayapan. Why this city, the capital of the most ad vanced native culture, was abandoned by the Xiu, and Mani selected as their seat — which they occupied at the coming of the Spaniards — is unknown. That Uxmal was inhabited, at least to some extent, at the arrival of the Spaniards, appears to be proved beyond any reasonable doubt, and is generally conceded. 296 Study of North American Archaeology. CHAPTER XX. CHICHEN-ITZA, TIKAL AND COPAN. Chichen-Itza, to which the reader's attention is now called, vies in the grandeur and extent of its remains with Uxmal. Like the latter, it is situated in the midst of a forest-covered plain, whose monotony is broken only by minor irregularities of the rocky sur face. Its name, which signifies "The Mouth of the Well of the Itzas," is supposed to have been given because of the presence of two great natural wells or cenotes Avithin its area. The principal ruins are in cluded in an area considerably less than a mile square,. and consist of half a dozen important piles, with remains of numerous inferior structures scattered about, which have not been explored. The pyramid- temple is the prevailing type, though some of the buildings are on the natural surface ; the ground plans are mostly simple arrangements of corridors, vesti bules and chambers ; the walls, are mostly vertical, the upper zone of the outer face ornamented, the lower portion plain ; the roofs are level and covered with cement and the floors are mostly of cement. Ordinary surface masonry is often irregular, imper fectly hewn stones laid up with little skill ; but im portant wall surfaces are generally faced with accu rately hewn blocks, neatly laid, but Avith little mortar except at the back. Here, as elsewhere in Yucatan, Chichen-Ltza, Tikal and Copan. 297 the wooden lintel was the Aveak feature of the con struction. "The study of even a single example of the great fagades," says Mr. Holmes, from whom our notes on the architectural features are chiefly drawn, "is suffi cient to impress upon one the vast importance of the sculptor's work, but the immense range of his field is appreciated when the heavy rattlesnake columns, the colossal serpent balustrades, the long lines of caryatid-atlantean figures, and the graphic relief sculptures of temple interiors and pillars have been passed in review. The life subjects had perhaps in all cases a mythologic origin and application, being employed in buildings or situations consistent with their symbolism. Purely geometric motives are nu merous, important and highly varied and specialized, indicating on the part of this people a ripe experience in various branches of art in which the esthetic had equal consideration with the symbolic." Although there was but little modeling in stucco here, plaster was universal ; every imperfect surface Avas made even by this means, and then treated with colors which were varied and brilliant. The general plan of dividing the building into rooms with vaulted ceilings, as heretofore described, was followed here. One of the most interesting remains of this group, in some respects, is the so-called Nun's Palace, a ground plan of which is shown in Fig. 102. This is in three stories ; the lower stage, some ten feet high, although vertical, with moldings around the top and a narrow ledge around the bottom, is apparently solid. Ascent to the second and third stories was made by broad stairways in front, as shown in the ground 298 Study of North American Archaeology. plan. The portion of the second story building sup porting the small upper story is also solid, the filling up having apparently been an after-thought for the purpose of supporting the small upper structure. The door jambs, lintels and rounded corners of the building are formed of stones of large size. Mr. Holmes thinks the most striking peculiarity observed here is the retreating profile of the upper Avail, a character occurring rarely in Yucatec buildings, but almost universal in the Usumacinta province. This fact, as will be seen hereafter, becomes important in Fig. 102. Ground plan of Nunnery, Chichen-Itza. tracing the development of art in this region. This building is also unique in regard to its decorations. The lower wall is elaborately embellished with geo metric sculptures arranged in large panels. The large panels in the ends of the building are filled with heavy lattice Avork. The ornamentation of the upper story, as given by Charney, consists chiefly of panels with central rosettes. This building has associated with it, as seen in the ground plan, an L-shaped adjunct, and two small de tached structures, standing on the natural ground surface. The first of these — the adjunct — is about Chichen-Itza, Tikal and Copan. 299 twenty-three feet high. The loAver portion of the outer face of the north and south walls has, alternating with the doorways, the usual mask and lattice decora tions in panels, together Avith some plain spaces. The fagade on the east is filled with two tiers of great snouted masks at the right and left of the doorway. The form of these snouts is shown in Fig. 103. The tipper zone presents one of the most richly decorated spaces of its kind in America. The north side contains six mask panels, with three rosette panels ; the east fagade has a central panel over the door, in which is a sitting figure, ^.^^^^3 Elephant trunk and mask panels at the right and figure, Yucatan, left. Even the flaring coping stones on the south front are embellished with three examples of, what are supposed to be, Tlaloc symbols as they embody the projecting snout and five or six pendant lines or grooves suggesting the rain god. Another interesting ruin of this group is the Tower, or Caracol, so named because of a special stairway extending upward through the columnar, central mass of the building. It is mounted on the second terrace of a broad elevated platform, as shown in the vertical section, Fig. 104, which passes through the center. Pig. 104. Vertical section of the Caracol or Tower, Chichen-Itza. •300 Study of North American Archaeology. The lower terrace is about twenty feet high and the upper one twelve. The tower is a regular circle about thirty-nine or forty feet in diameter, and when com plete was probably about the same height. The ground plan is seen in Fig. 105. >Mjj-MjuiijJu^uJuiHJ4aMxanii Fig. 105. Grouud plan of the tower, Chichen-Itza. The most imposing monument of Chichen-Itza is the so-called Castillo or Castle. This consists of a steep terraced, or stepped pyramid, seventy-five or eighty feet high, and a block-like superstructure. The sides of the pyramid rise at an angle of about fifty degrees, and are divided into nine steps ; a broad stairway of hewn stone ascends the middle of each face. One at least of these stairways, and probably all, were bordered by a kind of balustrade represent ing a serpent, terminating at the base in a great ser pent head with protruded tongue. The plan of the temple is of the usual form ; a front entry extending the whole length of the building, from which a door way leads into an interior dark room, around three sides of which runs a hall with doorways at the sides Chichen-Ltza, Tikal and Copan. 301 and rear leading out of the building, but not connecting with the inner room. The great front opening is in terrupted by two equally spaced columns which support the wooden lintel. These columns, which are circular, are carved to represent the body of a feathered ser pent, almost exactly like those seen at Tula, the head being bent outward at the base. Columns of the same form are seen again in this group at the so-called House of the Tigers. The chief sculptures in the Castle are representations of the human form. These are elaborately costumed and have stern features. Some of the figures seen here are furnished Avith long, full beards. Two Atlantean forms are shown in Fig. 106. The Gymnasium, House of the Tigers and other in teresting monuments must be passed without notice in our necessarily brief account. In studying the ruined cities of Yu catan and attempting restoration, we should bear in mind the following statement by Landa, who was in that country as early as 1540 : "Before the Pig. i06. Sculptur- arrival of the Spaniards, the aborig- ed Atlantean fig- ines lived in common, were ruled by severe laws, and the lands were cultivated and planted with useful trees. The center of their towns was oc cupied by the temples and squares, round which were grouped the palaces of the lords and the priests, and so on in successive order to the outskirts, which were allotted to the lower classes. The wells, necessarily few, were found close to the dwellings of the nobles, who live in close community for fear of their enemies, 302 Study of North American Archaeology. and not until the time of the Spaniards did they take to the woods." The culture hero of Chichen-Itza was, according to tradition, Cukulcan, a name signifying "Feathered Serpent." According to Landa, it was said that he arrived from the west, but whether with or after the Itzaes was not stated, but probably after the city had been founded. According to the same authority, the principal edifice, built no doubt long after his depart ure, was named Cukulcan. It was through him the people obtained their arts, religion and mode of government. After ruling over Chichen for a time he removed to Mayapan and founded that city. At length he disappeared to be known and honored, as the tradi tion asserts, as the god Quetzalcoatl in Mexico. In Mayapan, a temple was built in his honor, as at Chichen, which Landa says was round, with four doorways. The fact that explorers have found at Mayapan and Chichen, and nowhere else in Yucatan, tAVO circular temples entered by four doors, is seem ingly a partial confirmation of this tradition. The date of the founding of Chichen is of course unknown, yet the traditions, as shown by the author in his "Study of the Manuscript Troano," appear to indicate the sixth century a. d. as the probable date. However, as the place was inhabited, at least in part, at the coming of the Spaniards, it is probable that the structures, whose ruins now mark the site, were built long after that date. One destruction of the city is mentioned in the traditions. As will appear in a subsequent chapter, there are some reasons for be lieving that the Itzaes came to this locality from the Peten region further south. Chichen-Itza, Tikal and Copan. 303 As the remains of Palenque, Uxmal and Chichen- Itza indicate the leading architectural types of Cen tral America, brief references only will be made to some of the particular features of the remainder of this district. Extensive ruins have been discovered at Tikal, about twenty miles north-east of Peten, which are in some respects remarkable. One of the pyramids, in cluding its superstructure of three stories, measured, according to Maudslay (probably up the slope) , nearly 300 feet, which, with its fine wood and stone carvings, this explorer thinks "must have taken hundreds of active minds and thousands of skilled hands to have raised and perfected and kept in order." Here, as at Copan, some, at least, of the pyramids are carried up in great steps. Here also has been found the finest native wood-carving of America, so far as known. The chief features of this magnificent sculpture, which the size of our page will not permit us to introduce, are an enormous arched and profusely ornamented serpent, holding between its expanded jaws a human form with lofty head-dress ; and beneath the serpent fold, a standing human figure with shield on the left arm and holding a staff or lance in the right hand. This figure is literally enveloped in ornaments. In the upper right and left-hand corners are several columns of hieroglyphs skillfully and accurately carved, among which can be easily recognized day symbols with numerals attached, showing not only the forms found at Palenque and in the manuscripts, though more ornamental, but indicating also precisely the same order in counting the day series. The ruins at Quirigua, on Rio Motagua, eastern 304 Study of North American Archaeology. Guatemala, which have been explored and described by Mr. Maudslay, may be briefiy summarized as fol lows : Numerous square or oblong mounds and ter races, varying in. height from six to forty feet, some isolated, others clustered in irregular groups, most of which are faced Avith worked stone, and were ascended by flights of stone stepS. Some thirteen or more large carved monoliths, arranged irregularly around what were probably the most important plazas of the city or pueblo. Six of these monuments are tall stones measuring three to five feet square and standing four teen to twenty feet out of the ground. Five are oblong or rounded blocks of stone, shaped so as to repre sent huge turtles or armadillos or some such animals. All these monoliths are covered with elaborate carvings ; usually, on the front and back of the taller ones, there is carved a huge human figure standing full-face in a stiff and conventional attitude. The sides of these monuments are covered Avith tables of hieroglyphs, and in addition to these tables of hieroglyphs there are series of squares or cartouches of what appears to be actual picture Avriting, each division measuring about eighteen inches square, and containing usually two or three grotesque figures of men and animals. Some of the figures in these monoliths appear to be females. The hieroglyphs are of the same type as those already mentioned, the day symbols and numer als, so far as determinable, being similar to those at Tikal. Hundreds of ruins, many of them interesting, are scattered over Chiapas, Guatemala and Honduras. Quite a number of these have recently been examined by Dr. Carl Sapper, whose drawings, shoAving the Chichen-Itza, Tikal and Copan. 305 ground plans of the structures, with brief explanatory notes, make it evident that they are, in arrangement and general character, substantially of the Copan type. For this reason, a brief account of this impor tant type is given before closing the descriptive por tion of our work. These ruins are situated within the boundary of Honduras, some twenty-five or thirty miles directly south of Quirigua, a portion of the area being bounded by Copan river. Fortunately for students, a descrip tion of these ruins, written as early as 1576 by Diego de Palacio, has been preserved, Avhich Mr. Maudslay, who has devoted much time in exploring the ruins, considers more than ordinarily trustworthy for the time it was written, as he remarks : "This description is such a one as might have been written by any in telligent visitor within even the last few years." It is as follows : "Near here, on the road to the city of San Pedro, in the first town within the province of Honduras, called Copan, are certain ruins and vestiges of a great population and of superb edifices, of such skill and splendor that it appears they could never have been built by the natives of that province. They are found on the banks of a beautiful river, in an exten sive and well-chosen plain, temperate in climate, fer tile and abounding in fish and game. Amongst the ruins are mounds which appear to have been made by the hand of man, as well as many other remarka ble things, "Before arriving at them, we find the remains of thick walls, and a great eagle in stone, having on its 20 306 Study of North American Archaeology . breast a tablet a yard square, and on it certain char-. acters which are not understood. On- arriving at the ruins, we find another stone in the form of a giant, which the elders among the Indians aver was the guardian of the sanctuary. Entering the ruins, we find a cross of stone, three palms in height, with one of the arms broken off. Further on, we come to ruins, and among them, stones sculptured with much skill ; also a great statue, more than four yards in height, Avhich resembles a bishop in his pontifical robes with a well-wrought miter (on his head) and rings on his fingers. "Near this is a well-built plaza (or square), with steps, such as writers tell us are in the Coliseum at Rome. In some places there are eighty steps, in part at least, of fine stone, finished and laid with much skill. "In this square are six great statues, three repre senting men, covered with mosaic work, and with garters round their legs', their Aveapons covered with ornaments ; and the other Iavo of women, Avith long robes and head-dress in the Roman style. The re maining statue is of a bishop, who appears to hold in his hand a box or small coffer. They seem to have been idols, for in front of each of them is a large stone, Avith a small basin and a channel cut in it, where they executed the victim and blood flowed off. We found, also, small altars used for burning incense. In the center of the square is a large basin of stone which appears to have been used for baptism, and in which, also, sacrifices may have been made in com mon. After passing this square, we ascend by a great number of steps to a high place, Avhich appears Chiclien-Itza, Tikal and Copan. 307 to have been devoted to mitotes and other cere monies ; it seems to have been constructed with the greatest care, for through the whole of it there can still be found stone excellently worked. On one side of this structure is a tower or terrace, very high, and overhanging the river which flows at its base. "Here a large piece of the wall has fallen, exposing the entrance' to two caves or passages extending under the structure, very long and narrow and well built. I was not able to discover for what they served or why they were constructed. There is a grand stairway descending by a great number of steps to the river. Besides these things, there are many others which prove that here was formerly the seat of a great power and a great population, civilized and considerably ad vanced in the arts, as is shown in the various figures and building. "I endeavored with all possible care to ascertain from the Indians, through the traditions derived from the ancients, what people lived here, or what they knew or had heard from their ancestors concerning them. But they had no books relating to their an tiquities, nor do I believe that in all this district there is more than one, which I possess. They say that in ancient times there came from Yucatan a great lord, who built these edifices, but that at the end of some years he returned to his native country, leaving them entirely deserted. "And this is what appears most likely, for tradition says the people of Yucatan in time past conquered the provinces of Uyajal, Lacandon, Verapaz, Chiquimula and Copan, and it is certain that the Apay language, which is spoken here, is current and understood in 308 Study of North American Archaeology . Yucatan and the aforesaid provinces. It appears, also, that the design of these edifices is like that of those which the Spaniards first discovered in Yucatan and Tabasco, where there were figures of bishops and Chichen-Itza, Tikal and Copan. 309 armed men and crosses. And as such things are found noAvhere except in the aforesaid places, it may well be believed that the builders of all were of the same nation." The chief interest of archaeologists in these ruins has ahvays attached to the sculptured monoliths scat tered amid the crumbling structures, some fallen, but most still standing as silent Avatchers of the scene of former glory in AA'hich they played, perhaps, an impor tant part. These, hoAvever, are exceeded in impor tance as archaeological remains by the other monu ments. A ground plan of the main group of works is shown in Fig. 107. The Avhole of this area is ele vated, the larger inner courts or spaces being on the first general level. On this as a base arise the various pyramids and terraces shown in the figure, most or all of Avhich were crowned with buildings, now but heaps of ruins or fallen away down the slopes. The section (Fig. 108) shows the elevation of this assem- Fig. 108, Vertical section, main group, Copan. blage of pyramids and terraces. There are other groups of less extent not included in the plan. Although but small portions of walls have been dis covered, sufficient examples of ornamentation have been found to shoAV that art had reached here as ad vanced stage as at any other point in Central America. Among those discovered, the following may be men tioned as indicating the type: The highly orna mented monoliths showing human forms flanked by 310 Study of North American Archaeology. hieroglyphic inscriptions have become well known through Stephens's work. An inner step is men tioned which has carved on it a number of human figures seated cross-legged and covered with elaborate breast-plates and other ornaments. A carved orna ment, made up of several stones let into the wall, rises from this step on each side of the doorway and reaches the top of the Avail. This appears to be the conventionalized form of the serpent head, which is repeated in other parts of the group. Above this ornament, and extending several feet each side of the doorway, runs an elaborate cornice, ornamented Avith seated human figures and hieroglyphs carved in medium relief. Many of the steps bear inscriptions ; one is ornamented Avitli a row of teeth, others with human figures. In one place the space between fiights is covered Avith sculptures, among which are rows of death's heads. One of the altars which stand before the monoliths is a great carved turtle. At points human and grotesque heads are built into the wall ; elsewhere are seen human figures seated on huge skulls, and what Mr. Maudslay, for want of a more definite name, terms serpent men. The interior of the mounds and terraces is chiefly rubble, similar to that found in the Yucatec struct ures, but, seemingly for the purpose of giving addi tional strength, they have here interior supporting Avails buttressed by the mass of the slopes, which are faced with cut stones. It is evident, even from the very brief notice given here, that these ruins mark one of the most important centers of population in Central America, a place where native art had reached, perhaps, its most ad- Chichen-Ltza, Tikal and Copan. 311 vanced stage in North America. The absence in the figures and decorations of armed warriors or war-like scenes indicates a condition of peace, but this fact does not prove, as Mr. Maudslay thinks, that aban donment of this and other cities found in ruins at the coming of the Spaniards was not in consequence of war among the tribes. That the people Avere "in a state of decadence, and that they had almost ceased to be builders," as he contends, is probably true, but this condition was most likely brought about by op pression of rulers and intertribal strife. Passing into Nicaragua, the objects of chief interest to the archaeologist, which have been discovered, are rudely carved monoliths representing human and ani mal forms combined, the animal, usually alligator-like in form, holding the head of the human figure in its great jaws. But the description of these and numer ous other interesting remains of this region must be omitted, as our allotted space will not admit of further additions to this branch of our work. 312 Study of North American Archaeology. CHAPTER XXI. MIGRATIONS OF THE MEXICAN AND CENTRAL AMERICAN TRIBES. The discussion of the prehistoric movements of the Mexican and Central American tribes is introduced at this point because of the necessary and frequent ref erence to the conclusion on this subject in discussing the origin and growth of the native so-called civiliza tion. In fact, the conclusion reached in regard to the origin and development of this advanced culture, and in regard to other questions relating to the past his tory of the tribes of this section, will depend to a very large degree upon the opinion formed respecting the prehistoric movements of these tribes. Reference has already been made to the movements of population in that part of the Pacific division north of Mexico, which were found to be in most cases southward. The conclusion reached as to the exist ence of two great, distinct culture groups, the Atlantic and Pacific, implies that the general or more exten sive movements on the western side Avere parallel with its mountain ranges — that is, north or south — and not to or from the east. It is clear that the course of migration, so far as influenced by the phys ical features, was north or south. Whatever be the theory of the original introduction of population into the continent, .it must be conceded that its spread over it was through growth in num- Mexican and Central American Tribes. 313 bers, expansion and migration. Although it is quite probable there was a greater degree of permanency among the American aborigines than has been gen erally supposed, yet it is evident that the distribution of population could have taken place only through migration, though this was a slow process and may have been merely the gradual extension of the groAV- ing and swelling mass.* However, the wide separa tion of the members of some of the stocks indicate extensive movements in the past. Migration is there fore a necessary factor in the problem, and it must be assumed that every group of population, every stock and tribe, has come to its historic seat from some other point. Not that all such movements Avere necessarily by stocks or tribes, as there were doubtless numerous centers of development in some of which the original germs may have been but feeble bands, or a few families which pushed their way in advance of the tribe or clan and wandered into other sections . Sir John Lubbock re marks that "It is too often supposed that the world was peopled by a series of migrations . But migrations , prop erly so called, are compatible only Avith a comparatively high state of organization. Moreover, it has been ob served that the geographical distribution of the various races of man curiously coincides with that of other races of animals, and there can be no doubt that he originally crept over the earth's surface little by little, year by year — just, for instance, as the weeds of Eu rope are now gradually but surely creeping over the surface of Australia." However, as the only reason * For an explanation of this and other similar expressions see the closing chapter. 314 Study of North American Archaeology. for presenting these thoughts is conceded — to Avit, that migration in some form is a necessary factor in the problem — Ave may proceed on this admitted basis to a discussion of the probable course of migration in the southern portion of the division. The fact already noticed, that the data bearing upon the question indicate that the general move ments north of Mexico were southward, leads to the inference that the general trend in Mexico was in the same direction, which conclusion is justified unless some valid reason can be offered for believing that the order was reversed in this southern region. It must be admitted, however, that the theory of a re verse movement in this region has been advanced by several authors. Hubert Bancroft emphasizes this opinion in his "Native Races of the Pacific States," as follows : "First, as already stated, the Maya and Nahua nations have been within traditionally historic times practically distinct, although coming constantly in contact. Second, this fact is directly opposed to the once accepted theory of a civilized people, coming from the far north, gradually moving southward with frequent halts, constantly increasing in power and culture until the highest point of civilization was reached in Chiapas, Honduras, and Yucatan, or as many believed, in South America, Third, the theory alluded to is rendered altogether untenable by the want of ruins in California and the great north-west ; by the utter want of resemblance between New Mexi can and Mexican monuments ; by the failure to dis cover either Aztec or Maya dialects in the north ; and finally the strong contrasts between the Nahuas and Mexican and Central American Tribes. 315 Mayas, both in language and in monuments of an tiquity. Fourth, the monuments of the south are not only different from but much more ancient than those of Anahuac, and can not possibly have been built by the Toltecs after their migration from Anahuac in the eleventh century, even if such a migration took place. Fifth, these monuments, like those of the north, Avere built by the ancestors of the people found in posses sion of the country at the Conquest, and not by an extinct race or in remote antiquity. Sixth, the cities of Palenque, Ococingo and Copan, at least, were un occupied when the Spaniards came, the natives of the neighboring region knew nothing of their origin, even if they were aware of their existence, and no notice whatever of the existence of such cities appears in the annals of the surrounding civilized nations during the eight or nine centuries preceding the Conquest ; that is, the nation that built Palenque was not one of those found by Europeans in the country, but its greatness had practically departed before the rise of the Quiche, Cakchiquel and Yucatan powers. Seventh, the many resemblances that have been noted between Nahua and Maya beliefs, institutions, arts and relics, may be consistently accounted for by the theory that at some period long preceding the sixth century, the two peoples were practically one so far as their institutions were concerned, although they are of themselves not sufficient to prove the theory. Eighth, the oldest civili zation in America Avhich has left any traces for our consideration, whatever may have been its prehistoric origin, was that in the Usumacinta region represented by the Palenque group of ruins." That several, of the statements made in this quota- 316 Study of North American Archaeology. tion are incorrect and others are not sustained by subsequent investigations will appear from the data presented in this work. J. D. BaldAvin expresses substantially the same opinion in his "Ancient America:" it was also held by J. W. Foster, Squier and some other authors. On the other hand the pre vailing opinion among scholars of the present day, so far as published, appears to be that the Nahuatl group originated in, or at least came from someplace north of the knoAvn localities of the tribes composing the family. In confirmation of the latter opinion, the folloAving con siderations are offered in addition to the incidental notes bearing on the subject in preceding chapters. If Buschmann be correct in uniting the Ute or Shoshone group of dialects with and making them a part of the Nahuatl or Mexican stock, named by Dr. Brinton the "Uto-Aztecan Stock," we have, in the spread of this extensive family, what would seem to be incontrovertible evidence of the tendency in this Avestern section to southern movements. Members of this family are scattered from the vicinity of Columbia river to the Isthmus of Panama : and so far as any evidence has been found in regard to the movements of the tribes, it indicates they were southward. Offshoots from the Aztec group are found southward in Guatemala, Nicaragua and the borders of the Isth mus : and the movement of the Shoshones, so far as known, has been in the same general direction. The Comanches, the most south-eastern Shoshonean branch, and the only one which has come down east of the Rocky Mountains, have a distinct tradition that some two hun dred and thirty or forty winters ago they lived as one people with the Shoshones somewhere to the north of Mexican and Central American Tribes. 317 the headwaters of the Arkansas river. Dr. Gibbs is inclined from his investigations to the belief that this group has moved south and west, and suggests that they came from the eastern side of the Rocky Mount ains by way of the northern passes, in which he is probably correct, as the indications, the closer they are studied, seem to point to the great region west of Hudson's Bay as the point of dispersion from which tAvo streams issued. Parted by the desert plains to the south, one turned south-eastward and poured into the Atlantic division, the other coursing south for a dis tance along the east base of the Rocky Mountains, turned westward, and crossing over the passes fiowed southward along the intermontane valleys. Mr. Gatschet remarks (Appendix to Vol. VII, U. S. Geog. Surv. 100th Meridian) , when speaking of the relation of the Shoshone and Aztec groups: "From all that, it would be, nevertheless, preposterous to con clude that the Aztec civilization came from the north ; but we gain the evidence that the originators of the national Aztec legend, which professes this people to have issued from seven caves in the north, were cogni zant of the affinity of their nation with some tribes living north of Anahuac." It is true that it does not necessarily follow from the relationship of the northern tribes with those of the south, that the civilization of the latter came from the north, or that the movement was southward, as the inception of this advanced cul ture which, in accordance with popular usage, we speak of as " civilization," may have been due to some cause or impetus which did not exist in the northern section. There is, however, in the distribution of this civiliza tion, in respect to the widely scattered members of the 318 Study of North American Archaeology. great family, a strong and apparently conclusive argu ment in favor of the theory of a general southern move ment. As the members sou?th of Anahuac share, to a great degree, this civilization, in its most advanced form, while those north of Mexico do not, it would seem that the only explanation of this fact is that the move ment was from the north toward the south. The uniform tradition of all the tribes of this stock in Sonora and Sinaloa, so far as they were obtained by the early missionaries, was, according to Father Perez de Ribas, who made a careful collection of the legends, to the effect that their ancestors had migrated from localities further to the north. The conclusion reached by Mr. Bandelier, who visited that section and investigated the subject somewhat carefully, was in conformity with this tradition. It is also well known, as mentioned in the quotation from Prescott's Avork, that the Mexican or Nahuatl tradition in regard to their former home is, that their ancestors came from some locality iu the north or north-west, agreeing therein with the linguistic evidence. The general consensus of opinion among scholars of the present day, so far as published, is in conformity with this tradi tion, and may be accepted as probably the correct vicAV. The discussion may therefore be limited to the migra tions of the other stocks of this section, especially those of the Mayas, in regard to which there still appears to be some difference of opinion, though their northern origin is generally conceded. The Zapotecs and Mixtecs, closely affiliated tribes, whose territory is embraced chiefly in the state of Oaxaca, must be classed among the earliest inhabitants of the southern half of Mexico. It is not improbable Mexican and Central American Tribes. 319 that the oldest evidences of Mexican civilization are to be found here. Their early history, however, is almost a blank, as they have neither records nor reli able traditions which refer to their origin, their mi grations or first settlement in this region. Charney says they believed themselves to be autochthones, that they are ignorant of their origin, and have preserved no record of the time when they established them selves in the country. Torquemada ("Monarchia") mentions a story which relates that they were refugees from Cholula. Notwithstanding the radical differ ences in the languages of the two peoples, most of the old writers connect them with the Nahuatl stock. It is stated by some authorities that they formerly in habited the region of Puebla, together with the Olmecs and Xicalancas. Sahagun says they were or claimed to be of Toltec descent. It is even averred that Mitla was founded by the disciples of Quetzalcoatl. The Mixtecs, according to some writers, derived their name from Mixtecatl, one of the seven leaders who tradition says started out from Chicomoztoc, "the seven caves" in the far north. NotAvithstanding this uncertainty in regard to these tribes, there are some grounds for assuming that they came from the north, as all the other tribes of the western part of Mexico. First, the fact that such was the general course of the migration of the tribes of this section would, in the absence of any evidence on the point, lead to this conclusion ; second, the evi dence of long contact with the Mexicans adds support to this belief ; and third, as shown in preceding and following chapters, there is monumental evidence justifying the belief that they were for a time located 320 Study of North American Archaeology. as far north as Zacatecas, and none indicating a southern origin. The small tribes known as the Zoques and the Mixes or Mijes, wedged in between the Zapotecs and tribes of Chiapas, are supposed by some authbrities to have been the earliest inhabitants of the Oaxacan region, and their dim traditions point southward, especially to southern Chiapas, as the place of their former home, from whence they were driven by the Chapanecs. The origin of the Chapanecs is variously given. According to one tradition, they came to Chiapas from Nicaragua, and, having driven out the Zoques, occu pied the country. The generally received opinion, how ever, is that held by Orozco y Berra, Garcia, and other writers, which assumes, as the more authentic tradition, that which says, they moved down from a more northern region, following the Pacific coast until they came to Soconusco, where they divided, one part entering the mountains of Chiapas, the other part going on southward to Nicaragua, where they have been found under the name of Mangues, along the shores of Lake Managua. The various tribes of the Nahuatl or Mexican stock found in Guatemala, Nica ragua, and other parts of Central America, have, ac cording to their traditions so far as given, and as is now generally conceded, migrated from some region as far north at least as central Mexico. It would seem, therefore, from the evidence, that the tribes and stocks of Mexico and Central America, un less the members of the Mayan stock form an excep tion, migrated in prehistoric times from more northern localities than those they were found occupying at the Mexican and Central American Tribes. 321 coming of the Spaniards. We turn, therefore, to the history and traditions of this stock. Positive evidence of minor movements and pre historic shiftings among the tribes of this family is found in their geographic distribution at the time of discovery, a number being scattered over Guatemala, others in Chiapas, on the Rio Lacandon, iu the pe ninsula of Yucatan, and on Panuco river, north of Vera Cruz. Nevertheless, Avith the exception of the Huastecan offshoot on the Rio Panuco, they form a somewhat geographically compact body as com pared with the scattered members of the Nahuatl stock. The rugged range of mountains which sepa rates the eastern and Avestern groups does not appear to have been au effective barrier to prehistoric migra tions. As the characteristics of the languages of a stock af fords, as we have seen, great assistance in tracing the movements, and hence to a certain extent the history of the tribes of a stock, we call attention here to the classification of the languages of the Mayan family so far as we shall have occasion to refer to them. Omitting some of minor importance, the following are perhaps all the tribes of the family AA^e shall have occasion to refer to. Their respective localities are also given : Maya group. 21 Huastecas, on Rio Panuco, north of Vera Cruz, Mayas (proper) in the peninsula of Yucatan, Tzentals, in southern Tabasco and eastern Chiapas, Tzotzils in Chiapas, near the Tzentals, Lacandons, on the upper Usumacinta. 322 Study of North American Archaeology. Mams, in extreme Avestern Guatemala. Quiches, in Guatemala, immediately east of the Mams, Cakchiquels, in Guatemala, immediately south-east of the Quiches. Pokonchi, in central Guatemala, adjoining^ the Quiches on the north. Tzutuhils, a little tribe wedged in between the Quiches and Cakchiquels. Pokomams, in southern Guatemala, directly east of the Cakchiquels. Mam-Quiche group. Otto Stoll, one of the latest and most reliable au thorities on the ethnology of this family, divides it linguistically into three chief branches : the Huaste can, which is placed at the head as the oldest dialect, constitutes the first ; the Maya group, the second branch, standing next in age and relationship to Huasteca ; the Mam-Quiche group, the members of which are located in Guatemala, constituting the third branch. It must be remembered also that it Avas among the people of the Mayan stock pre-Columbian culture in America reached its most advanced position ; and that, in studying the movements of the various tribes and their final settlement in their historic seats, we are in truth studying their history and the history of the monuments of that culture which remain for the examination of explorers. The history of the Mayas and Mexicans has been so overloaded with tradition, idle tales, inventions of imagination by the old authors, and by the embellish ments and theories of modern authorities, that it is difficult to pick one's way through the maze. It is only within the last few years that scholars have un- Mexican and Central American Tribes. 323 dertaken a critical study of the various problems re lating thereto. Our attempt at present will be lim ited to a brief examination of the movements of the branches as arranged' by Stoll, and some suggestions as to the origin of the civilization of this people. The earliest tradition of the Mayas proper, of Yu catan, in regard to their origin, recorded by the Spanish authors, is that given by Lizana. He says that in their ancient language they name the east in another manner from what they do to-day. At the present they call the eastLikin, which is the same as to say the place from which the sun rises upon us. And the west they name Chikin, which is the same as the falling or setting of the sun, or the place where it hides with regard to us. "But in antiquity they said of the east Cenial, 'little descent,' and of the west Nohenial, 'great descent,' of the one side (east), few people, of the other side (west) , the great multitude." The historian CogoUudo, on the other hand, while giving the same tradition, concludes after some con tradictions, that the colony from the east must have been much more numerous and more ancient than the other. Landa and Herrera record a tradition that the oldest inhabitants came from the east, the sea being divided to afford them a passage. Some of their culture heroes come from one direction and some from the other. Itzamna, the chief hero or deity of the Mayas of central Yucatan, comes from the east, dwells long in the land and founds Izamal. The people of Chichen-Itza had also traditions of culture heroes : one regarding three brothers who came from the west and gathered together the people and ruled over them : another, possibly connected with the pre- 324 Study of North American Archaeology. ceding, tells of another and greater hero, Cukulcan, who came also from the west and ruled over the Itzaes and founded Mayapan. Whether the latter came thither with the people, or subsequently, is left uncertain by the tradition. Lastly, we have the myth of Votan, the culture hero of the Tzental group already referred to. To what extent these traditions are to be resolved into light and sun myths is a question we shall not undertake to decide at this point ; that they had in the minds of the inhabitants some relation to their origin and the origin of their civilization can not be denied. Fancourt, in his "History of Yucatan," and Dr. Brinton, in the introduction to "The Maya Chroni cles," reject, no doubt correctly, the idea of an eastern influx, and ascribe the population to immigrants from the west and north-west. As this eastern descent has connected Avith it some miraculous features, and there are no facts to support or give color to it, it is perhaps Avise to reject it in toto., as has been done by most of the recent authorities. But this does not dispose of the Itzamna and Votan myths. It is apparent from the physical features of the country that entrance into the peninsula, if not by the sea, must have been from the south or west, and the most reliable traditions are consistent with this fact. Herrera, who copies Landa, says : "Whilst the Cocomes lived in this regular manner, there came from the southward and the foot of the mountains of Lecando, great numbers of people, looked upon for certain to have been of the province of Chiapa, who traveled forty years about the desert of Yucatan, and at length arrived at the mountains that are almost Mexican and Central American Tribes. 325 opposite to the city of Mayapan, where they settled and raised good structures, and the people of Maya pan, some years after, liking their way of living, sent to invite them to build houses for their lords in the city. The Tutulxiu, so the strangers were called, accepting their courtesy, came into the city and built, and their people spread about the country, submitting themselves to the laws and customs of Mayapan in such peaceable manner that they had no sort of weapons, killing their game with gins and traps." This, Avhich appears to relate to a real occurrence, refers beyond question to the incursion mentioned in the Native Maya Chronicles, which refer to this mi gration as "The departure made from the land, from the house Nonoual, where were the four Tutulxiu from Zuiva at the west : they come from the land Tulapan, having formed a league." It is apparent from Herrera's statement and from the chronicles that the country was already inhabited when the Tutulxiu arrived. A comparison of all the data bearing on the subject, which include not only the traditions, but also the linguistic and historical evidence, the architectural forms, hieroglyphs, etc., lead to the following con clusion : That the Tutulxiu, who were still represented at the coming of the Spaniards, came from the region of the lower or middle Usumacinta, that is to say, some place in Chiapas or Tabasco : that previous to this migration, the Itzaes, who, as veritable history informs us, at a comparatively recent date, moved south to Lake Peten, where they were found by Cor tez, had, at a much earlier date migrated in a body, or sent a large colony northward into the peninsula, 326 Study of North American Archaeology. where they founded the city of Chichen-Itza. This coincides in the main, it is believed, with the opinion of most students of the present day who have devoted attention to the subject. It seems possible to trace the Mayan stock back with reasonable certainty to central Mexico. It is prob able that somewhere in that region the family group was divided into two branches, one of which, with the Huastecas in the lead, went toward the east ; all of the branch, however, except the Huastecas, turn ing southward, entered the valley of the Usumacinta. One offshoot of this branch rnoved on south-eastward to the Peten region, and sent a colony northward which founded Chichen-Itza. Others from the seats in the Usumacinta valley pushed their way northward into the peninsula. The western branch, which in cluded the Pokonchi, Quiche and Mam groups, moved on at a later date toward the south-east, stopping at the "second Tulan," which the author locates, with out any attempt at exact definiteness, in the region of northern Chiapas. This was the final point of dis persion of the western branch, from which the tribes proceeded to their historic seats in Guatemala. It is necessary to bear in mind that, although some of them were migrations in the true sense of that term, others were more likely the slow and gradual results of growth and expansion. It seems quite probable that before the western branch had left the region of the second Tulan, that is to say, the country in and about northern Chiapas, the older colonies located on the Usumacinta had grown in numbers, and gradu ally pushed their settlements into the peninsula. It is also possible that the eastern coast, north of Laguna Mexican and Central American Tribes. 327 ¦de Terminos, was reached and occupied before Chichen- Itza was founded. That the people who founded Chichen-Itza passed from the region of Lorillard City or the upper Usumacinta to the Peten region, and thence northward, appears to be proven by the identity of hieroglyphic types heretofore noted, and similari ties in other respects. 328 Study of North American Archaeology CHAPTER XXII. MIGRATIONS OP CERTAIN MAYAN TRIBES. The Quiche group, which includes the Cakchiquels and Tzutuhils, located in Guatemala, all claim to have come from some distant land which was the home of their ancestors ; and related in their tradi tions their early wanderings, which finally brought them to their historic seats. In the Popol Vuh or Sacred Book of the Quiches, a native work, which is largely mythical, it is stated that they "came from beyond the sea. " After the purely mythological por tion there follows something like tradition in which it is said the people multiplied greatly in a region called the East, and migrated in search of gods to Tulan- Zuiva (the "seven caves"), where four gods were as signed to the four leaders, namely (the gods) : Toliil, Avilix, Hacavitz and Nicahtagah. Here their lan guage was changed or divided, and the division into separate nations was established. Suffering from cold, and endeavors to obtain fire while they were awaiting the sun, are the points most dwelt upon during their stay at Tulan. In connection with these trials they were visited by an envoy from Xibalba, AA'hich is generally supposed to have been located in the region of Palenque, if not purely mythical. They abandoned or were driven from Tulan, and after a tedious journey, including apparently a crossing of an Migrations of Certain Mayan Tribes. 329 arm of the sea, or some water to which this term was applied, reached Mt. Hacavitz. The tradition of the Cakchiquels given in their "Annals," also a native work, is substantially the same as regards the points mentioned. Their ances tors are supposed to have come from the other side of the sea, from the land of Tulan, where they were brought forth. There were four clans. "Four men came from Tulan ; at the sunrise is one Tulan, and one is at Xibalbay, and one is at the sunset ; and we come from this one at the sunset ; and one is where God is. Therefore, there are four Tulans, they say, oh, our sons ; from the setting sun we came, from Tulan, from beyond the sea." It is undoubtedly true that these traditions are largely mixed with myth, and that it is a very difficult task to pick out the real from the mythological. Nev ertheless, there is a certain general trend in all which implies unity of origin ; there are also certain terms which indicate the latitude, the character of the cli mate, the country, etc. Although the incidents related may be largely mythical, they apply only to one re gion of North America, and show the country in Avliich they are supposed to have occurred. The four Tulans referred to in the extract from the Annals of the Cakchiquels heretofore given, are men tioned in the first part of the tradition which relates to the origin of the people, and may on this account be considered chiefly mythical. There are, however, repeated mentions of another Tulan to which they came after crossing the sea, which appears to be considered by the tradition the great and important point in their 330 Study of North American Archaeology. migrations. Numerous extracts might be given show ing this, but the following will suffice : "They say that the seven tribes arrived first at Tulan, and the warriors followed, having taken up the tributes of all the seven tribes when the gate of Tulan was opened." "The Tzutuhils were the first of the seven tribes who finished coming to Tulan, and then we the warriors came, as they say." Here they paid tribute of jade, silver, feather stuffs, of "articles painted, articles sculptured, astrological calendars, reckoning calendars, flute songs, songs hated of you because the seven tribes paid this tribute." Again it is said : "First came the Quiche men ; they acquitted themselves of their tribute in the first month ; then arrived their companions one after another, by their families, their clans, their tribes, their divisions in sequence, and the warriors, until the whole of them had finished arriving in Tulan." When it is added that numerous incidents are men tioned as occurring at and during the departure from Tulan, and that the Quiche tradition, as given in the Popol Vuh or Sacred Book, confirms the statement, that this was also an important point in their migra tions, there would appear to be no good reason for considering it other than some real locality which had an important place in the history of these tribes. It is more than probable that events which properly be long to widely different dates are crowded too closely together, but this does not necessarily invalidate the traditions. In both traditions, the statements relating to this Tulan, which was, as shown thereby, the point of the dispersion of the tribes, indicate a locality correspond- Migrations of Certain Mayan Tribes. 331 ing in features and characteristics to the Central American region. Mention of the tapir limits the territory in North America to that region south of central Mexico. Some of the animals alluded to in the Quiche legend as belonging to this region are found only in this southern section. Among the arti cles given as tribute at Tulan were green feathers worked and scAved together, calendars and cacao. As these references relate to the time they were at Tulan, we are justified in concluding that it was located, at least traditionally, if not really, in this southern section. As the Mams occupy the extreme western portion of Guatemala and the Quiches and Cakchiquels, according to the traditions, passed them in going to their seats in the south central part of Guate mala, they must necessarily have come from some place as far north-west as Chiapas. This agrees with the statement by "Orozco y Berra that they in habited Soconusco from remote times. We thus es tablish the direction of the movement of the western branch, as it is impossible, considering the geograph ical positions of the tribes, to explain a northward movement in harmony Avith the statements of the tra dition. As the only probable and acceptable theory in regard to the movements of the peninsular Mayas is that they came from the west or south-west, we thus obtain two lines pointing toward the same gen eral locality, southern or central Mexico ; either Oaxaca, Tabasco, Puebla, or the region about the •city of Mexico. This view is strengthened by some additional data, which are well worth considering in this connection. It is not impossible that the key to the puzzle is to 332 Study of North American Archaeology. be found in the relations with and prehistory of the Totonacas, a well-knoAvn tribe which Cortez first en countered on landing in Mexico, and from whose ter ritory he began his celebrated march to the imperial city, yet withal a mysterious people, whose ethnic position among the aboriginal nations of this region has not, as yet, been definitely settled. At this time they occupied the country known as Totonicapan, in cluded in the present state of Vera Cruz, and adjoin ing the territory of the Huastecas, According to their traditions, which appear to have some basis of truth, they had resided there for eight hundred years at the coming of the Spaniards, and had been an independent people up to a few generations pre ceding this coming, when they were subjugated by the Mexicans. They had migrated, according to their traditions, from the west and north-west, appa rently from the interior, in the region about the City of Mexico. They claimed to have migrated from the valley of Mexico, and to have lived long near the banks of Lake Tezcuco, where they built the pyra mids of Teotihuacan. Torquemada, who is the chief authority for their traditions, says : "Of their origin, they say that they set out from the place called Chicomoztoc, or 'seven caves,' together with the Xal- panecs ; and that they were twenty divisions, or families, as many of the one as of the other ; and although thus divided into families, they were all of one language and of the same customs. They say that they started from that place, leaving the Chichi mecs still shut up there ; and they directed their journey toward this part of Mexico, and having ar rived at the plains on the lake, they halted at the Migrations of Certain Mayan Tribes. 333 place where Teotihuacan now is ; and they affirm they built there two temples, which were dedicated to the sun and moon. Here they remained for some time, but either not contented with the place, or with a desire to pass to other places, they went to Atenami- tic, where Zacatlan now stands." From there they drifted further eastward, settling on the coast, where they were found by the Spaniards. That they were a primitive — so-called pre-Toltec, at least pre-Aztec — nation in Anahuac, is generally con ceded. That they were a cultured people is also ad mitted. That the temples and pyramids of Teotihua can were built by this people may be doubtful ; nev ertheless, as shown above, they do not appear to be Aztecan. This tribe seems to have been in close relation with the Huastecas, nor is there any valid reason why we may not infer that the two peoples were together in the migrations mentioned. On the other hand, there are some facts which favor this assumption. The position of the Totonac language in relation to the Nahuatl and Maya is yet an unsettled question among linguists. By some it has been joined to Nahuatl, by others to the Maya, but by the latest authorities it is considered as independent. Sahagun says they claimed relationship with the Huastecas, which of course would bring them into the Mayan family. That their language contains numerous words from Maya roots, and presents other resemblances, is known. But at the same time it is claimed, on the other side, that there are fully as great resemblances to the Na huatl. Charencey says the Totonac language presents striking analogies with the Mam-Huastec. 334 Study of North American Archaeology . The tradition recorded by Sahagun, so often quoted by writers, which refers to the landing of Mexican tribes and the Huastecas at the mouth of the Panuco river, is valuable chiefly, and in fact only, as being confirmatory of the tradition given in the earlier and more correct form. That part, hoAvever, which refers to immigrants disembarking from vessels, and imply ing a passage across the sea, must be rejected as an addition and not a part of the native legend. Mr. Bandelier, who is familiar with the early Spanish au thorities, says that neither of the two earliest sources from which this tradition is drawn speaks positively as to a "landing," but on the contrary only of the Xicalancas reaching the coast from the interior. This appears, also, to agree with the tradition of the Huas tecas as given by Marcelo Alejandro in his "Cartilla Huasteca," which says that they had their origin in the regions of the north, establishing their first loca tion where Altamira of to-day is located, in the state of Tamaulipas. According to Gomara, "Xicalancatl walked more earth, arrived at the sea of the north, and on the coast raade many towns ; but the two prin cipal ones he called by the same name. One Xical- anco is in the province of Maxcalcingo, which is near Vera Cruz, and the other Xicalanco is near Tabasco." It appears, therefore, that the idea that the tradition referred to emigrants coming over the sea was an er roneous interpretation given to it by Sahagun, simply from the fact that it spoke of people arriving at the mouth of Panuco river. All the evidence is entirely against the theory that any of the Mayan or Mexican tribes came from Florida or the West Indies. Bringing together all these points, and other indi- Migrations of Certain Mayan Tribes. 335 cations which can not be given in detail here, and studying them carefully, their bearing seems to point to some locality in western or central Mexico as the place of development of the Mayan tribes, and to ren der it probable that they are mixed up with the Toltec tradition. If we adopt the opinion that there was first a divi sion into two branches, one from which the southern and western tribes descended, the other that from which the northern and eastern tribes were developed, we may be justified in the following conclusions : This first division having taken place in western or central Mexico, the Mam branch moved on south to ward Chiapas. Having increased in numbers and clans as time passed, through some political convul sion, or being harassed by enemies, the tribes, which by this time had been at least in part formed, moved on toward the south-east. It is to the latter movement that the Quiche and Cakchiquel traditions chiefly re late. By this time, and perhaps previous thereto, the Nahautl tribes had appeared upon the scene, and pos sibly exerted the pressure which forced the Mayan tribes southward, though there were other peoples both on the east and west. It appears more probable, however, judging by the linguistic evidence, geographical position and differ ence in culture, that the Huastecas must have broken away from the main body before any of the other divisions took place. Or else, if they led off one branch going eastward, they must have parted from it before the other tribes were differentiated. How far north, or at what point this occurred, it is, of course, impossible for us to determine with certainty ; 336 Study of North American Archaeology. that it must have been north or north-west of the valley of Mexico may be assumed with some degree of probability. It is also quite likely that at this time they (the Huastecas) joined, or were joined by the Totonacas, with whom they gradually shifted to the gulf shore. The reason for this supposition is that, so far as known, the Huastecas had not adopted the peculiar calendar system which prevailed among all the other tribes of the stock, as also among the Mexicans, Zapotecs and most of the Central American nations. It is somewhat singular that this calendar appears to have been unknoAvn to or not adopted by the Huastecas, though it seems to have been in use among the Totonacas. This, however, we presume is to be explained by the fact that the latter had been brought under the direct control and supervision of the Mexicans, Avho caused its adoption by them. If it be true, as noAv supposed (though the evidence is not entirely satisfactory) , that this calendar was not in use among the Huastecas, this would seem to furnish conclusive proof that they had broken away from the other portion of the Mayan stock before it came into use. Taking all the facts and indications into considera tion, Ave are inclined therefore to the view that the development of the Mam-Quiche branch took place in Oaxaca, or Chiapas. The chief point of dispersion Avas most likely in the mountainous section of the latter district, though the formation of the tribes had taken place, at least in part, before this point Avas reached. It is also possible, and in fact probable, that the branch which settled in the lower half of the Usumacinta valley broke away, as above indicated. Migrations of Certain Mayan Tribes. 337 before reaching this point of final dispersion, Avhich seems to refer chiefly to the southern tribes. The Totonacas claimed, as shown above, that they were the builders of the temples at Teotihuacan. If Ave suppose the Mayas, or the eastern branch of the stock, to have been associated AA'ith them or in this region at this time, the tradition appears to account for some facts which are otherwise difficult to explain. "While the name Teotihuacan," says Mr.- Bandelier, "is Nahuatl, the confused traditions concerning the origin of the ruins ascribe them to an entirely differ ent tribe." It appears that this author, who rejects the view that the Toltecs were a distinct nation or tribe known by that name, is inclined to attribute the Avorks at Teotihuacan to the Mayas, in AA'hich it is probable he is correct. Some of the types of art found in this section and at Tulan present some marked resemblances to certain types discovered in Yucatan. The result of our inquiry therefore is that the Mayan stock moved south-eastward from some region as far north-west as central Mexico, probably preceded by the Zapotecs and some of the other older stocks. That they were somewhat closely followed by and came into contact with the Nahuatl stock before leav ing the Tulan, which we have supposed was on the boundary of Chiapas, is evident from the facts and traditions mentioned. That the Mayas were the first people to occupy Yucatan, and that they had already made considerable advance in civilization, although not established with positive certainty, is inferred 18 338 Study of North American Archaeology. from the results of Mr. Henry C. Mercer's examination of the hill caves of this section mentioned hereafter. Although the Mayas differ widely in language and to some extent in culture from the Nahuatl stock, yet it is possible that the belief of the Aztecs that they were at some remote period connected, though proba bly only by association, is correct. That the two peo ples moved southward substantially along the same- route appears to be the most reasonable conclusion. Ave can reach from the data so far obtained. The^ original home from which the primal germ of the great Uto-Aztecan stock started on its journey south ward was most probably the region now occupied by the Dene or northern Athapascans. Possibly from the same original germ, or from germs issuing from the same hive, were developed the Mayan, Zapotecan and the other small stocks of southern Mexico and Central America. That slight additions may have been received through occasional stranded vessels on the Pacific coast is possible, but this was not sufficient to leave any marked effect, unless it can be found in some particular customs or arts. Origin of Central American Civilization. 339 CHAPTER XXIII. THE ORIGIN AND DEVELOPMENT OF CENTRAL AMERICAN CIVILIZATION.* One of the most difficult problems of North Ameri can archaeology is that relating to the origin and pe culiarities of Mexican and Central American civiliza tion. That it was indigenous is now the prevailing opinion among antiquaries and ethnologists. While this conclusion may be accepted as beyond any rea sonable doubt true in a general sense, it is possible that in some of its features it may have felt the im press of extraneous or foreign influence. Neverthe less it is treated here as indigenous, that is to say, as being of native origin and growth. This being ad mitted, does the evidence show that its development was entirely within the geographical limits heretofore mentioned? In other words, can we trace by the monuments and traditions the growth of this civiliza tion, from its commencement to the highest stage it reached, wholly within this area, or do we have to look beyond these limits for the evidence? If wo take the architecture and attempt to trace the development of the different types by the monu ments which are found within the respective areas of the different stocks, we shall find our efforts to a large degree baffled. Although Yucatan is dotted over with ruins of Mayan structures, yet none of the many * As no native tribe can be called " civilized " in the true sense of the term, it is used here merely as indicative of the more advanced culture of these nations. 340 Study of North American Archaeology. explorers who have studied them have pointed out the various steps of the builder's art as shown by these ruins. Nor is this to be wondered at, for the very good reason that there do not appear to be in these areas examples Avhich can be pointed to as the begin nings of this art, the first rude efforts. Historically and otherwise we learn that there were dwellings of stone and dwellings of wood, from the simple thatched hut to the stately mansion. Possibly the temples may have been developed from the longitudinally di vided houses described by Landa, but no modification of these would have produced the temple-crowned pyramids. Violet le Due has tried to show how the stone structures, Avith their pointed arches and heavy frieze, are but stone copies of wooden buildings, or that the one has been modeled after the other. Though Fergusson may have traced successfully this kind of transition in the history of Oriental archi tecture, it is evident to any one who will examine with care the Frenchman's figures, that Avith the ex ception of the mode of dividing the body, there never were Mayan structures of wood of the pattern given. In Egypt, where the beginnings are less apparent than in some other centers of Eastern civilization, re mains shoAving primitive efforts are still found. Ex amples of the simple, plain tombs from which, ac cording to Rawlinson, the pyramids were ultimately developed, remain to the present day ; but, so far as the author is aware, nothing has been found in the Maya territory to mark the commencement of that art which designed and constructed the temples and pyramids of Uxmal, Chichen-Itza and Kabah. So far as architectural skill is concerned, there is but little difference observable in the structures of Yuca- Origin of Central American Civilization. 341 tan and the adjoining sections attributable to the Mayas. The principal variation is in size and degree of ornamentation; even the plans, as shown by Mr. W. H. Holmes, in his excellent paper on "Archaeo logical Studies Among the Ancient Cities of Mexico," are, to a large extent, conventionalized. But the fact remains that among the hundreds of examples Avhicli exist, there are none showing the first rude efforts of the builders. As stated by Mr. Holmes in the quota tion given in a previous chapter, while it is true that some of the buildings are composite and show success ive accretions, there are others which stand as perfect units of design. But he speaks nowhere of the rude beginnings found in that section from which they Avorked up to the more perfect form ; his only intima tion of progress is that in the quotation given below. Nor has any author, so far as the writer is aware, given us this information. How are we to account for this absence of earlier forms except upon the theory that when the tribes en tered their historic seats they had already become proficient in the builder's art? The tradition of the Tutul-Xiu, as given by Herrera, indicates that they brought this art with them. That wood was used chiefly for ordinary dwellings is true, and that the Maya architects may have to some extent modeled their stone structures after those of wood may be true, but certainly not to the extent claimed by Violet le Due ; nevertheless it seems improbable that this skill should have been attained without comparatively rude beginnings. The only indication that the art, in this type, was still in an undeveloped stage, is that mentioned by Mr. Holmes : 342 Study of North American Archaeology. "Notwithstanding the success of these Maya masons in erecting buildings capable of standing for hundreds of years, they were yet ignorant of some of the most essential principles of stone construction, and are thus to be regarded as hardly more than novices in the art. They made use of various minor expedients, as any clever nation of builders would, but depended largely on mortar and inertia to hold their buildings to gether." Mr. Henry C. Mercer, who has explored a number of caves in Yucatan for the purpose of searching for indications of the early inhabitants, comes to the con clusion that the Mayas were the first inhabitants, and that they had acquired their civilization before enter ing that territory. These facts, apparently at least, justify us in searching for the remains of their pri mary efforts along the route of their migrations. Although our information is meager in regard to the antiquities of northern and north-western Mexico, yet enough is known, as appears from the descriptions given, to state positively, notwithstanding all that has been written and said to the contrary, that some of the ruins found in that section may have been the primitive efforts of the civilized tribes of southern Mexico and Central America. Whatever opinion we may hold on this point, two things which have an important bearing on the question must be admitted : First, as has already been stated, that in the central region, or Maya sites, there are no evidences of the primitive architectural efforts ; and, second, that it is only to the ruins north of this central region Ave can look for these extralimital primitive forms. There we do find what may possibly be the remains of the prim- Origin of Central American Civilization. 343 itive types ; but southward until we reach Nicaragua the structures and inscriptions, although showing variations and the introduction of additional forms, are too apparently the outgrowths of what we may term the central types for this fact to be overlooked or disputed. It is a somewhat strange fact, if we judge only by the hieroglyphic inscriptions and manuscripts, in the Mayan characters, that we would be compelled to con clude that they were brought to comparative perfection at the time they were invented. A difference, it is true, in the forms and ornamentation, and, to a certain de gree, an advance toward a more perfect type, can be traced, but no examples, so far as the writer is aware, of the first rude beginnings, or the original forms, have been found. Some, comparatively rude, are found painted on pottery, scratched on shells or other soft ma terial, but these belong to what maybe termed demotic writing and are not primitive forms. Comparing the characters of the various inscriptions which have been discovered and those found in the few remaining pre- Columbian manuscripts, the result is as follows : First, it is apparent that the characters in the manuscripts have been adopted from those of the inscriptions. In other words, inscriptions preceded the manuscripts ; hence Ave must look to the former for the older forms. What appear to the writer to be the oldest forms of the glyphs yet discovered are seen in those at Palenque and some of the inscriptions found by Charney at Menche (Lorillard City) , though others discovered by him at this same place belong to the later and more ornamental type discovered in the Peten region, that is, those carved in wood discovered by Bernoulli at Tikal, a type also found at Copan and Chichen-Itza, 344 Study of North American Archaeology. but in none of the inscriptions at Palenque. In all cases the same method of indicating numbers is fol lowed, and the same calendar system as that of the Tzentals and the Dresden Codex is also followed. These facts form part of the evidence on which we base the conclusion given in a previous chapter, that the Itzaes passed from the region of the upper Usuma cinta eastward to their seat about Lake Peten. Another item found in this connection bearing on the migrations of some of the Mayan tribes is worthy of notice. During his studies of the Palenque in scriptions, the author has been surprised to find among the various glyphs one in which the chief character is the figure of a person lying on his back, his knees drawn up, his head partially raised up, and his hands placed on his stomach. A sit ting figure Avith and without a head is found both in the manuscripts and inscriptions, but, unless shown in some of the inscriptions on the statues of Co- pan, the introduction of full length figures in the glyphs as a part thereof is unusual. This unusual hieroglyph is found twice in the inscription on the tablet of the cross knoAvn as No. 2 ; and once in one of the inscriptions in the Temple of the Three Tablets. If we turn noAV to the pages of Charney's "Ancient Cities of the New World," where the two prone statues are figured, we will find, seemingly beyond a reasonable doubt, to what our unusual glyph refers. These statues, which, according to Hamy and Char ney, denote Tlaloc, the god of rain and fertility, were found at widely different points, one at Tlaxcala and the other by Dr. Leplongeon at Chichen-Itza. As has been correctly stated by Dr. Brinton, "a statue of a Origin of Central American Civilization. 345 sleeping god holding a vase was disinterred by Dr, Le plongeon at Chichen-Itza, and it is too entirely similar to others found at Tlaxcala and near the City of Mexico for us to doubt but that they represented the same divinity, and that, the god of rains, fertility and har vest." Mr. Bandelier mentions a fourth one found in the state of Puebla. Shall we attribute the statue found in Yucatan to the Mayas and the others to the Aztecs? Or shall we ascribe both to the Mayas, thus assuming that some one or more of the tribes of the latter stock, at some period before their entrance into their historic seats, dAvelt for a time in the vicinity of Tlaxcala? If the writer be correct in his suggestion that the glyphs re ferred to represent this diety, the latter supposition would seem to be the correct one. Reasons have already been given for believing that the pyramids or temples of Teotihuacan date from a period anterior to the occupation of this region by the Aztecs, and hence can not be ascribed to them. It is true that indications of Aztec culture, which has, to a certain degree, been impressed upon them, are found, but some, in fact most, of the types differ from any thing that is known to belong to this culture, and the ruins are declared by the best authorities to be non-Aztecan. It must also be borne in mind that it is stated by the early writers that the name Tula (Thu- la, Tulan, ToUan, ToUam) , was also applied to this place, a name which, although signifying "place of reeds," was explained, whether correctly or not, as meaning "place of Toltecs." It is apparent from these facts that some of the works of this place are attributable to an older people than the Aztecs, or a 346 Study of North American Archaeology. people antedating their advent. It is also true, as argued by Charney, that some of the types found at Tula, as for example, the ornamental sculpturing on some of the columns, and the serpent-like form of others, bear a strong resemblance to some of the types found at Chichen-Itza. [After this chapter had been written the writer re ceived from Dr. Antonio Penafiel, of Mexico, a letter announcing the discovery in a plowed field at Tula, on the site of the "Temple of the Caryatides" (that of the double column with feet shown by Charney) of a shell on which are engraved hieroglyphic characters supposed to be of the Mayan type. An inspection of the photographs, which accompany the letter, prove this supposition to be correct beyond doubt. It is a puzzling fact, however, that these characters bear a closer resemblance to the southern variety, especially those on the shell found in Belize, of which a figure is given in the author's paper on "Day Symbols of the Maya Year," in the Sixteenth Annual Report of the Bureau of American Ethnology, than to those of Pa lenque or Yucatan. A similar type is seen in Fig. 84, page 140, of Dr. Brinton 's "Primer of Mayan Hiero glyphics," which represents an inscription on a vase from a Quiche tomb, Guatemala. The difference is, however, due in part to the material on which the characters are inscribed and the way in which they are written, as may be seen by reference to Fig. 78 of Dr. Brinton's "Primer," which shows an inscription at Kabah, painted on stucco. The discovery of this inscription, if corroborated by other finds, may have a tendency to show that the sepa ration of the Mayan tribes did not take place until the Origin of Central American Civilization. 347 region of Tula' was reached in the migration south ward, or that the Mam-Quiche group followed the track of the eastern group to this point.] It is proper also to bear in mind in this connection, that the ruins of Comalcalco, as has been stated, belong unquestionably to the type found at Palenque, and Avith more profuse ornamentation at Uxmal and other places in the peninsula. As Comalcalco lies in Ta basco on the road northward, it would be exactly on the route of migration from Mexico or Puebla to the region of Palenque, by the eastern or shore line. While these facts are not sufficient of themselves to prove the presence of the Mayas in this central Mexican region, yet when we take into consideration the traditions which have been mentioned showing the probability that the Huastecas and the Totonacas at one time occupied this section, and that the works both here and at Cholula are certainly pre-Aztecan, we are perhaps justified in attributing some of them to that branch of the Mayas which went off with the Huastecas toward the east. To this must also be added the fact that the evidence so far adduced shows that the Mayas must have come from some point as far northward as central Mexico. The indications also bear us out in the supposition that the builders of the temples and pyramids of Palenque came to this locality, partly, at least, by way of the shore, or Tabasco route. This evidence not only indicates that the Mayas occupied, for a time, portions of the Mexi can valley, in advance of the Aztecs, but it also serves to give some support to the theory that the Toltecs were the Mayas. At Quemada, in Zacatecas, are the ruins of temples. 348 Study of North American Archaeology. pyramids, walled terraces, stairways, etc. There also is seen a row of stone pillars running through a hall, as at Mitla. Bancroft does not think that these ruins show any marked analogies to the structures either north or south, though he admits that the pyramids are similar to those at the south. However, they do show that composite character which indicates transitional forms which must be found in alL ad vances from lower to higher grades. An intermediate type between that of Casas Grandes and those further south, if constructed by a people migrating south ward, not yet formed into tribes, would, of course, differ somewhat from the types both north and south, and would in all probability be, to a certain extent, composite in features. There are, however, some features which lead to the impression that these works should be attributed in part to the Zapotecs, as for example, the rows of columns through the middle of halls, and the increase in the size of the rooms. The former is a peculiarity found elsewhere only in Zapotec ruins and at Teoti huacan, and indicates a different mode of construction from that found either in the Nahuatl or Maya archi tecture. The increase in the size of the rooms is a feature repeatedly mentioned by Mr. Bandelier as he proceeded south Avard in his journey through north western Mexico. The change, according to this writer, begins at Casa Grande. "The rooms," he remarks, "are higher and much more spacious than in the northern ruins ; the doorways are higher and wider." Although Charney's zealousness in advocacy of the Toltecs has a tendency to bias his judgment in regard to any thing which bears on this subject, we quote his Origin of Central American Civilization. 349 opinion on the similarity mentioned, as it shows the impression left on his mind from a personal examina tion of the ruins in Oaxaca. "Las Casas Grandes, the settlements in the Sierra Madre, the ruins of Zape, of Quemada, recalling the monuments at Mitla, others in Queretaro, together with certain features in the building of temples and altars which remind one of the Mexican manuscripts, from which the Toltec, Aztec and Yucatec temple was built, make it clear that the civilized races came from the north-west." It is possible that the Mayas and Zapotecs were in close relation with one another during their migration south ward ; or supposing the latter to have been the earlier emigrants, that the former learned the art from them, subsequently developing their particular types in the more southern sections. It may also be added that the truncated pyramid at Mitla built of adobes, and the nuinerous adobe structures in the Zapotec region, are at least suggestive. The absence of the triangu lar arch is a marked distinction, though the flat roof was not entirely unknown in Yucatan. At Casas Grandes, which is in Chihuahua, we see proofs of the initial steps of mound building. In fact, the evidence of gradual advance toward a higher grade m the architectural art is seen beyond question as we advance southward from Arizona to Quemada, be our opinion in regard to the authors of these works what it may. We must confess that, so far as we are able to judge from all that has been written in regard to the ruins of the south-west, there seems to be no other reason for denying this advance in type than a fixed purpose to maintain a theory. These raised platforms or low pyramids filled in solidly with 850 Study of North American Archaeology. gravel, and the terraces supported by bordering walls, may justly be considered the beginnings of that art which culminated in the temples, terraces and pyra mids of Yucatan and Chiapas. Walled terraces, plat form mounds, an inner filling or hearting of gravel, mortar and stone fragments, are characteristic fea tures of this southern region, and are found also in Oaxaca. It is true that in the southern locality the works are upon a larger scale, the types more perfect, and the ornamentation more profuse than at Casas Grandes. It may also be said that the southern types are more specialized, but this is precisely what would be expected upon the theory that the southern forms were developed from, and grew out of, the northern. The conclusion which Mr. Bandelier, who has studied the regions both of north-western and southern Mex ico, has reached on this question, is so exactly in point iu this connection that we quote it somewhat fully : "Although the communal pueblo houses of the north seem to be different from the structures on the Gila and at Casas Grandes, they still show the same leading characteristics of being intended for abodes and at the same time for defense. In the northern villages, however, both features are intimately con nected ; whereas further south the military purpose is represented by a separate edifice, the central house or stronghold, of which Casa Grande is a good speci men. In this the ancient village of the south-west approaches the ancient settlements of Yucatan and of central Mexico, which consisted of at least three dif ferent kinds of edifices, each distinct from the others in the purposes to Avhich it was destined. It seems, therefore, that between the thirty-fourth and the Origin of Central American Civilization. 351 twenty-ninth parallels of latitude the aboriginal archi tecture of the south-west had begun to change in a manner that brought some of its elements that were of northern origin into disuse, and substituted others derived from southern influences ; in other words, that there was a gradual transformation going on in ancient aboriginal architecture in the direction from north to south. ' ' I have alluded only to the most striking examples of south-western aboriginal architecture, the large houses. In regard to another kind, the small de tached buildings, it must be observed that the small house is probably the germ from which the larger structures were evolved, and that the small houses also undergo modiflcations, especially from north to south, in the size of the rooms. I repeat here what I said in my preliminary report to' the Institute of Au gust 11, 1883 : 'There is a gradual increase in the .size of the rooms in detached buildings, in a direction from north to south, which increase is most distinctly marked over the area where the detached house alone prevails.' "There are regions, like central Sonora, where the small house is the only architectural type now re maining from ancient times. It will be noticed that the square or rectangular dwellings of the Opatas of the Sonora river confirm the impressions above re corded. If we compare them with the dimensions of the huts now inhabited by tribes living still further south, we find their size increases as we advance from a colder to a Avarmer one. "Large halls are not found in the ruins of the north. They appear to be almost the rule at Mitla 352 Study of North American Archaeology. and in Yucatan ; and they are met with on the Gila, under a climate which is semi-tropical. "Equally noteworthy is the increase in dimensions of the doorways and windows. In the lofty struc tures of Arizona and Chihuahua there is considerable resemblance to the doorAvays of ancient edifices in Yucatan and other southern states of Mexico. "The outer coating of the walls is of course differ ent in the arid northern countries from that in the moist regions of the tropics. Elsewhere I have men tioned the plating, with polished slabs, of the walls of Mitla, which was applied, I suspect, not merely for ornamental purposes, but with a practical object. Where summer rains are as violent as under the tropics, a coating of adobe or gypsum would be un able to resist them for any length of time. In the south-west a thinner coat was sufficient ; still there is improvement in such coating, from the northern sec tions to the southern, shown by the finish of the wash applied to the surface. "I have alluded to the appearance of artificial mounds and artificial platforms or terraces on the Gila, and perhaps also in the Casas Grandes region. It is well known that both of these structures are conspicuous in the ruins of southern and central Mex ico. The estufa, however, is a specifically northern feature, and therefore disappears as soon as the cli mate becomes more equable and finally tropical." It may be added that, according to Mr. Bartlett, who examined the Casas Grandes in 1850 and illus trated his description with figures, the rooms began to assume here almost the exact form and arrange ment in series seen in some of the Yucatec structures. Origin of Central American Civilization. 353 We also notice the fact that the aggregation, so to speak, of cells as in the pueblo structures is gradually lost as we proceed southward. Marked evidence of this change is seen before we leave southern Arizona, and at Casas Grandes we see the last faint intimation of it, unless we assume that the room series in Yuca tan are resultants. The cliff-dwellings, which are a characteristic fea ture of the San Juan region of north-eastern Arizona, continue side by side with the other features as we proceed south into Mexico, and are not lost until after we have passed Casas Grandes. That the series of types is unbroken from the region of the pueblos and cliff-dwellers of Arizona to Casas Grandes is unquestionable. True, there are variations in the features and characteristics, but that they are de rived from the same original type is a fair and reas onable presumption. It may also be asserted with a considerable degree of confidence that southAvard from Quemada the series is continuous to Guatemala. It is admitted that there is a greater variation in this series than in the northern one. This, however, ac cords precisely Avith the facts as known and what might have been predicated upon these facts. In other words, the body of emigrants, so to speak, after leaving the region of Quemada, was gradually split into branches and tribes, each moving off into the seat it was found occupying at the coming of the Spaniards. Although each of these divisions would take with it the original or stem pattern, each would gradually modify it and change it according to fancy. Hence, the variations as we move southward, would 23 354 Study of North American Archaeology. be greater than those seen north of that point. There would be no entirely new or different cultures : all were developed from the one original. Had this cul ture spread northward from the southern area instead of being a gradual development during the movement southward, the examples would not show the rude primary types, but imperfect attempts at a higher and more advanced type, the culture would not spread from the central point until it had become advanced. No valid reason has been given, nor is it believed can be given, for rejecting the theory that these northern structures are attributable to the ancestors of the civilized tribes of the regions further south, supposing the theory now generally accepted that these tribes migrated from this northern section, be correct. It is admitted that the Nahuatl and Sho shone tribes have been connected socially if not lin guistically. It is also admitted that the latter did not move northward from the southern region. Hence, the Nahuatl tribes must have moved from the north southward. It is also generally conceded, or at least intimated, and apparently in accordance with the most reliable data, that the Mayas and Zapotecs, if not de rived in the far distant past from the same original stem as the Nahuatl tribes, had long been in intimate association with the latter. Notwithstanding the advance made by the time the latitude of Quemada was reached, the cause or impetus, or whatever we may term it, which gave rise to the more advanced civilization of the southern section, had not yet been exerted or brought fully into play. The beginning of that architectural art which was to result in the splendid structures of Yucatan and Chiapas had been made, but it is probable that it Origin of Central American Civilization. 355 was not until the priestly hierarchy was more thor oughly organized and brought into full sway, that the onward march toward that higher culture began in real earnest. Why there has been such persistent refusal on the part of scholars to accept, as at least pos sible, the theory that the tradition of the "Seven Caves," or "Seven Ravines" (Chicomoztoc and Tulan Zuiva) , refers to the cliff-dwellings or cave-dwellings of north-western Mexico and Arizona, is difficult to account for. There is nothing in this supposition con trary to the traditions, nor to the generally-accepted theory of the course of migrations. The number seven does not necessarily play any particular role in the solution of this problem. Numbers were deter mined from some incident or circumstance which may or may not be known. Seven may have been selected because of some superstition or because it was understood that seven was the number of tribes belonging to a certain group or stock, or it may have arisen in many other ways. It is therefore immaterial in this relation. The reference therefore in the Nahuatl and Maya traditions to seven caves, although largely mixed with myth, may be interpreted as possibly refer ring to the cliff- or cave-dwellings, or to this mode of liv ing while in the north. This would be appropriate as explaining the frequent reference in these traditions to darkness, gloom and a sunless condition. It is well known that caves were often resorted to in the south ern regions as places for holding religious ceremonies and other purposes. The inner cells of many of the Central American temples are virtually caverns so far as light is concerned. Torchlight was essential to some of their religious ceremonies. 356 Study of North American Archaeology. CHAPTER XXIV. PRIESTS HIEROGLYPHS CALENDAR. No step in advance in art or any other branch of culture is likely to be taken, especially among a com paratively rude people, until some need for that step is felt. Burial originated from the necessity of dis posing of the dead, and the stone and wooden cover ings, tombs and cells, from the desire to secure the bodies from injury by wild beasts; dwellings, from the need of shelter and security, etc. So, in the more advanced stages of society, forward steps in culture are generally the result of a felt or imagined need therefor. Sometimes accidental discoveries bring into use more efficient implements and processes than those previously employed, but the adoption of these is an evidence of the felt need therefor. Although it is true, as remarked by Dr. Tylor, that priests are not favorable to changes or novelties in the religious formulas or ceremonies, yet they are ever ready to accept an increase in power or influence, to add to the impressiveness and display of their pub lic ceremonies, and are ever desirous of increasing the size and magnificence of their religious structures. It is to this desire and the poAver and influence of the priests of Mexico and Central America, the author believes, are to be attributed the rise of civilization and its progress, in these regions, to a higher stage than that attained any-where else in North America. Priests — Hierog lyphs — Calendar . 357 The influence of the religious sentiment, especially when directed by a strong and well-organized priest hood, is too well known to need any proof here. The power and influence of the priests in Mexico and Cen tral America are also facts too well known to require verification. In Mexico, according to one writer, "the prelacy was as systematic, and its rule as well defined, as in the Church of Rome," The contest which was waged for a time between the warrior class and priests was ultimately decided in favor of the latter ; in other words, the prelacy became the ruling power. Among the Mayas and Zapotecs the priestly power was, if possible, more complete than among the Aztecs, the authority of the chieftain or ruler being almost completely dimmed by it. We therefore at tribute the rise and progress of the Mexican and Cen tral American civilization chiefly to this priesthood. With their opposition or indifference it could have made no progress. Moreover, the chief advance made is along lines intimately connected with religious ceremonies and priestly needs. The hieroglyphs were doubtless the work of the priests, few of the people understanding, or being able to interpret or use them. To the priests, there fore, must we ascribe the inscriptions and manu scripts, which, so far as we are able to judge from the little that has been ascertained in regard to them, relate, in a large degree, to religious observances, cere monies, etc. The calendar system appears also to be based upon, or at least adapted to the religious systems in vogue among the nations which adopted it. We are justified, therefore, in believing that the advance in culture along the lines of art, especially those 358 Study of North American Archaeology. of architecture and sculpture, the calendar system and symbolic writing, was due to the priesthood. It is probable that no decided impulse toward a higher culture was given until a start had been made toward a thorough organization of the priesthood. That a sloAV and gradual advance was being made in archi tecture is doubtless true, and it is also doubtless' true that, after this beginning, there Avould have been a further slow growth, even without a decided impulse from the priesthood ; but the remarkable advance in dicated by the structures and sculptures of the south ern section, as compared with the northern monu ments, indicates some more potent influence than was exerted in the northern regions. It must be admitted, although we may be unable to entirely solve the mystery, that there was some cause, some impetus, something which gave rise to the civ ilization of Mexico and Central America, which did not exist or occur in California or any other part of North America. Mr. Bancroft is doubtless correct in his declaration that this can not be attributed wholly to differences in the physical conditions. It is more likely that the real direct cause is to be found in the complete organization, and strong influence of the priesthood. This, it is true, only serves to throw the investigation one step further back ; for the next in quiry will be, why this advance in priestly organization and influence over that of other sections ? Although Ave will not undertake to give any answer to this question, it is possible that a thorough and careful study of what we may, in a broad and compre hensive sense, term the priesthoods or religious organ izations of the more northern tribes, beginning with Priests — Hieroglyphs — Calendar . 359 those of the Pueblos, Avould furnish a partial explana tion. The cult societies of some of these Pueblos described by Mrs. Matilda C, Stevenson in her paper on "The Sia,"in the "Eleventh Annual Report of the Bureau of American Ethnology," and those of other Pueblos described in several papers by Dr. Walter FeAvkes, may possibly be the beginnings from Avhich the more elaborate organizations of the southern tribes Avere developed. The latter author mentions a number of somewhat striking similarities between the religious ceremonies of the tribes he studied, and those of the Mexicans and Mayas. As the people among whom he prosecuted his studies were chiefly those of the Hopi or Moki pueblos, people belonging to the great Uto-Aztecan stock, these similarities become impor tant, especially when the caution of this author in ad vancing theories is considered. Where the priesthood in the course of the migration southward attained that perfection in its organization, and obtained that power as a distinct body which en abled them to direct the several lines of improvement leading up to the civilization finally reached, can of course never be definitely determined. If the sugges tion that the southern system was a development from the northern, should by further study be found prob able, this development must of course have been gradual. Although the evidence shows that the great, and apparently rapid advance in civilization after it was once clearly started, was due chiefly to the priest hood, the attempt to determine where this start took place would be in vain. If the supposition that the Mayas, or at least the eastern division thereof, occupied for a time the re- 360 Study of North American Archaeology. gion of Tula (Mexico) and Teotihuacan, and were the authors of some of the works of those localities be correct, Ave may feel assured that the priesthood had by this time come largely into power as a body ; that there was here an organization of some kind. If, as the writer is inclined to believe is true, the Mayas who occupied this region probably with the To tonacas, consisted of the eastern branch which had followed the Huastecas as heretofore suggested, their priesthood must have been organized, and the Palen que and Votan problems become less difficult to solve. The traditions of the other (western) branch indicate that their priesthood did not come fully into power until they reached that Tulan which we suggest was somewhere in or near Chiapas. Here it is said they received their gods. The statement in the Quiche tradition which says: "Truly Tohil is the name of the god of the Yaqui nation who was called Yolcuat Quetzalcoatl," indicates beyond question contact with the Aztecs, though the identification of Tohil with Quetzalcoatl is probably incorrect. The contact of priesthoods is probably implied. It appears certain, however, that Mayan hieroglyphs did not come into use until the valley of the Usuma cinta was reached. The origin of these is difficult to explain, and Avill probably remain an unsolved prob lem until explorations shall bring to light some rude beginnings from Avhich they were, or may have been developed. The Mayan glyphs are of a widely different type from the Mexican, though 'they may have passed through some of the same stages of growth, but the general consensus of opinion is that the Mayan is the older of Priests — Hieroglyphs — Calendar . 361 the two classes, and that these' two classes have de veloped independently. As it seems apparent that they Avere not brought into use until some of the Mayas had reached the Usumacinta valley, it seems possible that they are not older than the Mexican symbols, though the opposite view is generally enter tained. It has been found possible to determine with rea sonable certainty the objects from which some few of the Mayan symbols were draAA'n. These derivations indicate that the symbols had already been brought into use when they were adopted to represent the days, and that considerable advance had been made in hieroglyphic writing. We are compelled, there fore, to admit that the origin of this writing is a mystery we are unable to fully penetrate. The va rious steps through which several of the characters used had passed before reaching their final forms may be theoretically traced, but monumental evidence on this point is wanting, at least so far as the author is aware. There can be but little if any doubt that their development and use was due to the priests. It also appears that this form of writing Avas confined to the Mayan tribes, a fact which, considering the in timate relations of these tribes with other stocks, in dicates that it had advanced to that stage which pre vented its adaptation to other languages. More especially does this appear to be a proper conclusion when we take into consideration the fact that the so- called "Native Calendar" had been adopted by some half a dozen different stocks. How are we to account for the spread of the com plicated calendar system through so many different 362 Study of North American Archaeology. and even hostile stocks, as stated in a preceding chap ter, when each had necessarily to, or at least did, assign different names to the days and months, and adopted different symbols, so far as these are known to have been brought into use ? There seems to be but one an swer : it was through the influence of the priests. Nor does this appear to solve the problem, except upon the further supposition that there was some kind of rela tionship, understanding or intercourse between the priesthoods of the different stocks or tribes, not lim ited by ethnic lines. It is only by the supposition of "a powerful secret organization," as Dr. Brinton has shown, "extending over a wide area, including mem bers of different languages and varying culture," that the spread of Nagualism can be explained. That the various stocks and tribes, some of which were at constant war with one another, should have adopted the same calendar system, which was espe cially adapted to the work and office of the priests, except through their influence, seems impossible. Moreover, that influence must have been specially exerted for this purpose and to this end. Is it not possible that this wilT serve in part to explain the numerous traditions relating to the sudden appear ance of priestly civilizers in various sections? The traditions of the Mexicans and Central Ameri cans regarding the sudden appearance in their midst and as sudden departure of great reformers and civil izers, who were afterward regarded as culture heroes, have long been and still continue to be puzzles, in re gard to which students have, as yet, been unable to offer any generally accepted solution. It is Quetzal coatl among the Mexicans, Votan among the tribes of Priests — Hieroglyphs — Calendar. 363 the Usumacinta, Cukulcan and Itzamna among the Mayas of Yucatan, and Gucumatz with the tribes of Guatemala. Some authors attempt to explain these by the solar myth, resolving all these traditional founders into personified Dawn, Lightning, and the like. But as Hutson justly remarks : "We have had abundant evi dence that this method of explanation can be pushed too far, and its results have always been too vague to add any thing to our real knowledge of early ethnic life." Is it not possible that these traditional person ages were priestly messengers traveling from tribe to tribe to weld together a common brotherhood? Such a supposition would not be more extravagant than that theory which makes of them sun and light myths. If the statement made by some writers, that war was often Avaged between Mexican nations by agree ment or understanding for the purpose of obtaining captives for sacrifice, be true, it is certain that this was instigated by the priests, and, moreover, indi cates some secret relation between the different priest hoods. It is proper, however, to remark that Bandelier questions this. The division of the year into eighteen months of twenty days each is the most radical change of the system from the usual lunar count, or count by moons. It is generally admitted by scholars who have referred to the subject, that the original Mayan calendar must have been based on the usual lunar count. There is even positive evidence that there was in vogue at the advent of the Spaniards, at least in some sections, a secular month of thirty days, giv- 364 Study of North American Archaeology. ing twelve months to the year. In the "Report on the City of Valladolid, written by the Corporation of the city by order of His Majesty and the very illus trious Senor Don Guillen de las Casas, Governor and Captain General, April, 1529," we find the following statement : "They [the Indians] divided the time by months of thirty days, and on the first day of the year, before dawn, every one, including the Alquin [priest] , watched for the rising sun and held a great feast on that day." That this change could have come about by any gradual process seems impossible. We are therefore justified in believing that it was arbitra rily made by the priests, or that it Avas brought into vogue through some foreign influence. Mention of the lunar count and the year of twelve months is also made by other early writers. Where and in what stock did this calendar have its origin, are questions to which students will probably never be able to give satisfactory answers. If we could decide positively which came into vogue first, the calendar or the symbolic writing, we might give approximately correct answers. That the Mayan hiero glyphs, as has already been stated, did not come into use until the people of this stock had reached the valley of the Usumacinta, may perhaps be confidently asserted. It may also be confidently asserted that the Mexican method of representing days and numbers did not come into use until after the Aztecs had settled in the valley of Anahuac. It is stated in the Cakchiquel Annals, according to Dr. Brinton's translation, that among the tributes they paid at Tulan were "astrological calen dars and reckoning calendars." If this refers to the "Native Calendar," it follows that both calendar and Priests — Hieroglyphs — Calendar. 365 symbols were then in use. However, as we have en deavored to show, the advanced Mayan tribes were already located in the Usumacinta region when the Cakchiquels reached this Tulan. It Avill perhaps be safe to assume that the various stocks using the calendar had reached substantially their historic seats at the time of its adoption. That such a time system could have come into use independently in some half-dozen different stocks, relatively at the same time, seems improbable. As its origin is most naturally ascribed to one stock or people, it must have been forced upon other peoples by some strong pressure instigated by the priests, or brought into use through the influence of the priesthood. By the earlier authors it is attributed to the Toltecs, which according to the view herein adopted Avould be indefl- nite, but most likely would refer to the Mayas while in central Mexico, or at least before reaching their historic localities. Orozco y Berra, who is followed in this respect by Dr. Seler, expressed the opinion with confldence that it had its origin among the Za potecs. Dr. Brinton says he has been unable to reach any definite decision on the question, but is inclined to the opinion "that it was the invention of that an cient branch of the Mayan stock who inhabited the present states of Chiapas and Tabasco." It is probable that the former opinion is the correct one. The evidences of an impress of culture by the Zapotecs upon the Mexicans, or the reverse, have been referred to. The figures recently copied by Prof. Frederick Starr from the Mitla ruins, and published in his "Notes on Mexican Archaeology," shoAv types 366 Study of North American Archaeology. corresponding beyond question with those of the Mexicans. One prominent feature of these figures is the Tlaloc nose — or elephant-like probocis. It may be that the Zapotecs were the pioneers in Central American civilization. It is certain, as we have seen, that some of the peculiar types found at Mitla appear also at Quemada, a fact which has been noticed by several writers. The conclusions which seem to be most in accord with the data are : That the Mayan stock came from the north-west, substantially along the same route followed subse quently by the Nahuatl tribes, and that they had been, at some early day, previous to or soon after en tering the area of Mexico, in comparatively close relations with the Nahuatl stock. That somewhere in central Mexico, one branch, or part of the family led by the Huastecas, broke away from the other part and pushing in advance, in com pany with the Totonacas, occupied for a time the re gion embracing the Mexican Tula and Teotihuacan. The Huastecas and Totonacas moved on to the gulf shore, Avhile the remaining portion of this eastern branch passed on southward to the valley of the Usumacinta, a part at least going by way of Tabasco, another portion going directly to the upper portion of the valley. From these colonies the Peten region and the peninsula of Yucatan were peopled. The other branch, after dwelling for a time in the region of Chiapas, scattered in tribes south and south-east to their historic seats. That the advance in the architectural art began Priests — Hieroglyphs — Calendar. 367 while they were yet in central Mexico, its rapid pro gress being due to the influence of the priesthood. That the use of hieroglyphs among the Mayas be gan Avith the colonies who settled on the Usumacinta. That the calendar probably had its origin Avith the Zapotecs, and that it spread to the different stocks through the direct influence of the priests. 368 Study of North American Archaeology . CHAPTER XXV. CONCLUSION. Although it is true, as stated in the opening chap ter, that the monuments and relics must be our chief reliance in studying the customs, arts and activities of prehistoric peoples, language must form the chief basis here, as has been found true in the old AA'orld, of the theories relating to the more remote periods, when the identification of race or stock is the object in view. The character and types of the monuments and arte facts and to some extent of customs and superstitions, as shown by what has been presented in the preceding chapters, are largely the result of physical environ ment ; hence, in attempts to trace relationship of tribes and peoples and to follow the movements of pop ulation in prehishoric times, language must be our chief reliance. But language and monuments, so far as the latter are to be found, should tell the same tale. The coincident testimony of these two classes of ev idence, fortified by other data bearing on the subject, lead to and seem to justify the geographical division of North America, in reference to its archaeology, into the three primary regions which have been outlined. If this conclusion be accepted, at least so far as it re lates to the Atlantic and Pacific divisions, it will (leaving out of view the question of origin or race) relieve the study of prehistoric North America, of some Conclusion. 369 hitherto troublesome questions and eliminate some theories still advocated. It will form one important step in the arrangement of the archaeological data of the northern continent. Led by the dim rays of light the data afford, rely ing chiefly on language, geographical features and tra dition, together with the feAV historic facts relating to the more recent migrations, the conclusion reached is that the great movements of population in pre historic times in North America have been southward. Tracing back the streams toAvard their sources, we have found that the converging point appears to be the inhospitable region stretching from the western shore of Hudson's Bay to the Rocky Mountains. Thence tAvo streams floAved southward : parted by the great treeless plains stretching from the Saskatchewan to the Rio Grande, one moved south along the moun tain skirt and passed to the Pacific side, the other going to the Atlantic side. From straying bands seeking localities of more abundant food, which be came permanently separated from the parent group, were developed, in all probability, partly by growth and partly by combination, one with another, many, possibly most, of the so-called stocks ; others had possibly been differentiated before reaching the continent, if, as is not improbable, the peopling was by the incoming of suc cessive parties along substantially the same route. However, the reader must keep in mind the fact that there are two theories in regard to the general movements of population in the Atlantic division in prehistoric times, each of which is maintained by strong advocates. One of these is that which has 24 370 Study of North American Archaeology . been followed in this work, the other that which holds that the spread of population has been from the At lantic border. If the latter theory be adopted, and it be assumed that the Pacific slope was populated from this eastern group, it would seem necessary, from the evidence herein presented in regard to the movements on the Pacific side, to assume that one important stream from the Atlantic side floAved north-westward, at least as far north as the headwaters of the Missouri (and even farther, judging by the movements of the Athapascans) , and then turned southward, moving down the western side. For it seems certain that the great plains formed a barrier seldom traversed in prehistoric times except toward their northern extremity. It might perhaps be consistent Avith this theory to assume that the Mayan group moved southward along the Gulf coast. Although this theory of a dispersion from the Atlantic coast is held by a number of able advocates, the author of this work has been led by his study of the evidence to believe that the Atlantic division was pop ulated from the north-west, and that the Pacific di vision received its population chiefly from the same region. The reader, however, is left to accept either theory which appears to him to be most in accord with the data Avhich we have presented. In our attempts to trace back the development of tribes and peoples to more distant eras, we have, when the monuments and other evidence failed us, appealed to language as indicating former relationship, but it is necessary here also to remind the reader that on this point different views are entertained by linguists. On the one side, it is held by some authors that affinity of languages implies racial identity or unity of origin ; on the other, it is contended that the theory that the Conclusion. 371 affinity of languages necessarily implies identity of race is not warranted. It may be stated as evi dent that where a tribe or people has incorporated into itself elements from another race or stock, as the negroes in the United States, and these elements have adopted the language of that tribe or people, the lan guage will not be an evidence of race. It is prob able that the correct theory will be found interme diate between these tAvo, and this position, we no tice, has been taken by Mr. A. H. Keane in his recent work on ethnology. Another view, also intermediate, but differing somewhat from any yet presented, and which promises apparently to be more comprehensive in its scope, appears to be foreshadowed by Major J. W. Powell, director of the Bureau of American Eth nology, in his remarks on the Indian linguistic fam ilies in his 7th Annual Report. He says : "The opinion that the differentiation of languages Avithin a single stock is mainly due to the absorption of materials from other stocks, often to the extin guishment of the latter, has grown from year to year as the investigation has proceeded. Wherever the material has been sufficient to warrant a conclusion on this subject, no language has been found to be sim ple in its origin, but every language has been found to be composed of diverse elements. The processes of borrowing knoAvn in historic times are those which have been at work in prehistoric times, and it is not probable that any simple language derived from some single pristine group of roots can be discovered." Our allusions, therefore, to the inference which may be drawn from affinity in language as evidence of migrations must be taken only as indicative of 372 Study of North American Archaeology. former intimate relationship of some kind, not neces sarily racial, for affinity in languages necessarily im plies former intimate relationship of some kind. That the chief stream on the Pacific side had its source east of the Rocky Mountains in the region mentioned seems likely from AA'hat has been shown in regard to the general course of migration so far as it can be traced. One puzzling feature of the California and Oregon coast appears to receive at least a partial explanation by this theory of movements on the west ern side of the continent. We allude to the large number of diminutive stocks, as heretofore mentioned, crowded into this narrow shore strip. The explanation of the problem appears to be found in the position of the minor Athapascan offshoots lo cated along the coast of Oregon and California. Al though the later movements of these branches, which brought them to their historic homes, have taken place within modern times, the date when some of them re moved from their priscan habitats probably belongs to the prehistoric era, though comparatively recent. The other minor stocks are probably fragments of other families from Avhich they separated at a much earlier date, Avhich, under the infiuence of changed surround ings and through intermarriages and combinations Avith other bands, have formed new groups and new languages. The following remarks by Dr. Gibbs, who is considered an authority on the Indian tribes of this region, may be appropriately quoted here : "If I may hazard a conjecture at present, it is that the Tah-kali [Dene] and Selish families, with perhaps the Shoshone and some others, originated east of the Rocky Mountains ; that the country between that Conclusion. 373 chain and the great lakes has been a center from Avhich population has diverged ; that these two tribes crossed by the northern passes of the mountains ; and that their branches have since been pushing westward and southward. Whether the southern branches of the Tah-kali have been separated and driven on by the subsequent irruption of the Selish, or whether they have passed over their heads, can, perhaps, be ascertained on a severe comparison of the different dialects into which each has become divided ; it being reasonable to infer that those which differ most from the present are the oldest in date and emigration. "The route of the Selish has obviously been along the courses of the tAvo great rivers, the Frazer and the Columbia. By the former, they seem to have penetrated to the sea, while on the latter they were stopped by the Sahaptin and the Tsinuk. Some branches undoubtedly crossed the Cascade range, at different points, to the Sound, and the country inter mediate between that and the Columbia. And the Tilamuk have overstepped that boundary and fixed themselves on the coast of Oregon. The southern limit of the T:ili-kali is not yet ascertained. Mr. Hale identified the Umkwa as an offshoot. Lieuten ant Kautz has lately shown the Tu-tu-ten to be another, and it is possible that some of the California languages may also be assimilated. Dr. NeAvell states that, since he was first in the Indian country, all the great tribes have been gradually breaking up into bands. Whenever two chiefs attain about an equality of power and infiuence, jealousies arise, which lead to a separation of the tribe. These are formed by many causes, the chattering of the women, of course, among 374 Study of North American Archaeology. others. Before the introduction of firearms, the range of the different tribes was more limited than now. They did not travel so far from their own country. This last is less applicable to the coast tribes than to those of the interior. The former are, however, more split up, and those of the Sound coun try, perhaps, most of all. The influence possessed even by those claiming to be head chiefs has become almost nothing ; and in case of any disagreement in a band, the dissatisfied party move off to a little dis tance and take the name of the ground they occupy, or any one desirous of establishing a band on his own account induces a party of his immediate followers to accompany him, and start, as it were, a new colony. It is to this separation, and to the petty hostilities which often grew out of it, that we must mainly at tribute the diversity of dialects prevailing." The southward movement was a slow process, re quiring possibly thousands of years, during which streams were Split into .branches, Avhile others per haps coalesced. However, in studying the history and migrations of the aborigines in prehistoric times, we must disabuse our minds of the idea of a dense population. We have spoken in the preceding chap ters of the "growth and swelling of the mass" in the terms used by writers in relation to the movements of peoples in the old world, yet the idea obtained is likely to prove erroneous without explanation, as the usual estimate of our native population is erroneous. The native population of the Atlantic division was prob ably at no time as great in numbers as has been sup posed by those who have not carefully investigated the subject. Detached villages scattered over a large Conclusion. 375 area Avere necessary to a people depending on the chase alone for subsistence ; and Avhen the country Avas clear before them, bands would continue to Avander farther and farther from each other, thus by long isola tion and the influence of changed physical conditions giving rise to new tribes and stocks and various cus toms. It is therefore in this sense the growth and expansion of the native population, while depending on the chase, is to be understood. The Indians Avould have considered Illinois crowded had it contained one- twentieth the number of souls now living in the city of Chicago ; and Pennsylvania overstocked Avith a population of fifty thousand or even forty thousand. As more congenial climes were reached in their movement southward, and cultivation of the soil be gan, the tribes became more and more sedentary and the arts developed. The discovery of maize and its use as a food plant was probably one of the most potent agencies in bringing about this sedentary con dition. Its use crept sloAvly northward against the tide, changing nomads and hunters, at least partially, into settled agricultural tribes. The strong families occupied the interior choice districts, pressing the older, weaker and broken stocks to the shores. Moving soutliAvard from the cold and inhospitable sections of the north, it is not until Ave reach the more favored districts, Avhere agriculture was resorted to as a partial means of subsistence, that monuments indi cating an advanced culture appear. It is in Ohio, in the lower Mississippi valley and the Gulf states of the mound section — regions of richest soil and best adapted to agriculture — that the most stately mounds and imposing earthworks are found. Moving south- 376 Study of North American Archaeology. Avard on the Pacific side it is in the mild climate of Mexico and Central America, the land of maize and tropical fruits, we find evidences of the most advanced civilization. Wisconsin or the effigy-mound region possibly forms an exception to this rule, yet some of the tribes of the southern half of that state, where the chief works are found, may have entered the agri cultural stage at the time they were constructed, nevertheless the objects imitated indicate that they were still in the hunting state, although they must have been to a large extent sedentary. Development of culture appears to have been in some degree retroactive, that is to say, although the unfolding seems to have been from north toward the south, yet the more advanced culture seems to have moved backward, to some extent, on its pathway. What appear to be evidences of this are found in the mound area where a few large pyramidal mounds of the true southern type are found north of the Ohio in Illinois and Indiana, and strange to say, in one in stance at least, combined with a northern type which seems to have been developed in and was confined almost exclusively to the limits of Wisconsin. Thfe same thing appears to have been true to some extent in the Pacific division, especially in regard to certain customs and religious ceremonies. As the cultivation of maize extended northward against the stream, it carried with it some new arts and customs which orig inated in more southern climes. It is evident, how ever, that arts and customs and even superstitions, re sulted to a considerable extent from physical condi tions. This is sliOAvn by the fact that the architecture, designs and customs of particular districts, which are Conclusion. 377 and have long been inhabited by widely different stocks, often present a general similarity. This broad resemblance, or, as it were, uniform coloring or shading of various forms or types, is due, as a general rule, to the physical conditions ; as, for example, in the Pueblo section, and California. No evidence of buildings of stone, adobe or burned brick has been found in the Atlantic division. Wood, which was abundant in most sections, supplied ma terial for all the structures their customs required. It is true that the early mound explorers speak of burned brick found in southern tumuli, but more re cent examinations have shown the material to be but the burnt plastering of dwellings consumed by fire. On the Pacific side, adobe was used as a building ma terial from Arizona to Oaxaca, and stone was the chief material of structures from New Mexico to the Isth mus. This difference is of course to be attributed in a large degree to the differences in the physical con ditions of the two sections. That stone, though it might be laid up rudely and undressed, and adobe would be used in the construction of dwellings in the arid and treeless areas of New Mexico and Arizona, might reasonably be surmised in advance of evidence. Judging by the difference in the advance made to ward civilized life, the character of the monuments and the deeper covering of earth and rubbish over many of the ruins of the southern sections of the Pa cific division, the reasonable conclusion seems to be that the latter was peopled in advance of the Atlantic division. Although intercourse between the two di visions appears to have been very limited, there is some evidence that a few designs in art found their 378 Study of North American Archaeology . Avay from the Pacific into the Atlantic division, but none, so far as the author is aware, that any traveled in the opposite direction. As this work is limited to the archaeology of North America, of which only a brief and, in some respects, incomplete notice has been presented, no allusion to South American antiquities has been made. How ever, as the conclusion reached in regard to the possi bility that the latter continent was peopled from the former will have an important bearing on some of the problems discussed, the following suggestions are presented : Although we have not discussed the origin of man in America, it is apparent from the conclusion reached in regard to migrations that the evidence adduced points to the extreme north-Avest Pacific coast as a probable point of entry into North America. Should this conclusion, which corresponds with a widely pre vailing, though it must be admitted by no means universal, opinion of the present day, be accepted, it does not follow as a necessary inference that South America was peopled from the same source, or by the same route. On the contrary, a careful consideration of the data points, or appears to point, rather to the opposite conclusion. The sup position that the southern continent was peopled from the northern demands an immense stretch of time that seems incompatible with the archaeolog ical features of the latter. The comparative age of the monuments of Peru and those of Central America, as estimated by the more conservative stu dents, will not justify the assumption that the civil ization of the former region was carried southward Conclusion. 379 from the latter. Moreover, the antiquities and lan guages of the Isthmian region north to the borders of Nicaragua indicate a northward movement to this point, where it seems to have been met by the move ment from the north. It appears also to be generally conceded that the West Indies were peopled from the southern continent. On the other hand, there is no evidence, archaeologic or linguistic, of a northern ele ment in the southern continent. Peruvian architec ture was peculiar and imposing, and shows no trace of influence from Central America or Mexico. Slight indications of South American or West Indian influ ence are claimed to have been discovered in southern Florida, but a study of the data so far as made leads apparently to the conclusion that this was the result of mere contact, and not from a northward movement of population, the people even of the Keys, as claimed by some authorities, pertaining to the northern races. Although the attempt to determine the length of time man has lived in North America will be a mere guess and nothing more, there is no apparent reason for carrying it back beyond a date necessary for the de velopment of the various tribes, and possibly most of the stocks. It is possible, though the proof is not yet sufficient to gain general acceptance, that there has been pre historic contact on the western coast of Mexico with people from the Pacific islands or south-eastern Asia. While these pages are being written, news comes of the discovery in the state of Guerrero, on the south west coast of Mexico, of several remarkable groups of ruins showing an advanced stage of culture. Among the ruins are remains of large temples, a tab- 380 Study of North American Archaeology. let with hieroglyphs, the character, however, not stated, a lofty arch, etc. Should these prove to be of the character indicated, they may have some bearing on the question of the origin of this civilization. But the utmost that can be anticipated in this direc tion is proof of foreign influence on the native civil ization already in the process of development. Although our treatise can claim to be nothing more than an outline of the subject of which it treats, yet enough has been presented to show that North Amer ica offers an archaeological field which is yet to yield a rich harvest to antiquarian research, a field which has been as yet but little worked except in some few districts, but well deserves to be thoroughly cultivated. This becomes manifest when it is known that the most extensive group of pyramidal mounds in the Gulf States, a group supposed to mark the site of one of De Soto's halting places, remains undisturbed except by the plow, and that the largest group on the western bank of the Mississippi is yet unexplored ; and is emphasized when is added the further statement, confirmed by abun dant evidence, that there are hundreds of undisturbed groups of ruins in Mexico and Central America, the thorough exploration of which would enable the archaeologist to solve more than one of the unsettled problems relating to that region. It is but recently that the writer of this work was informed that there are ruins along the Rio Panuco which bear the indi cations of advanced Mayan art, and which, if ex plored, might settle the doubtful questions regarding the Huastecan offshoot from the great Mayan stock. II^DEX. Acolhuans, 236. Adair, Jas., 70, 144, 145. Agriculture, Mexican and Central American, 243. Aldrich, J. B., 178. Alejandre, M., on origin of Huas tecas, 334. Aleuts, their ancient houses, 38; their migrations, 44. Algonquian family, their territory, 48 ; their migrations, 158-9. Altar mounds, 77 ; Squier and Da vis on, 78. American Antiquarian, 157, 158 A merican Archipelago, ancient houses at, 38. Angel mounds, Vanderberg Co., Ind., 130. Apaches, 173, 174, 229. Archaeological divisions, 16-21 , Archaeology, defined, 1 ; Ameri can, problems of, difficult, 2; lo cal, 25-8. Architecture, Central American, 249-50, 339-42 ; beginning of the art, 354. Arctic division, 17, 21, 35-47; its monuments, 35-9; its imple ments and ornaments, 40-3. Athapascans, 139 ; area of, 171 ; their divisions, 174-5 ; their mi grations, 174-6. Atlantic division, defined, 16, 48; its archaeological features, 49 ; its monuments, 50; its tribes and stocks, 139 ; difierentiation of tribes in, 147 ; entry of popula tion into, 148 ; prehistoric migra tions in, 159-68; its buildings chiefly wooden, 377. Aztalan, Wisconsin, ancient works at, 129. Aztec Springs, Ariz., ruins at, 215- 17. Aztecs, 236, 248, 273, 345. Baldwin, J. D., cited, 316. Bancroft, H. H,, cited, 199, 233, 240, 284, 358; quoted, 189; defines interior section, 203; on ruins at Quemada, 254; on Mexican migrations, 314-15. Bandelier, A. F,, cited, 240 ; quoted, '337 ; on archaeology of North western Mexico, 222-9 ; on ruins at Teotihaucan, 261 ; on pyramid of Cholula, 268 ; on Mitlan pot tery, 273; on Mexican migra tions, 318 ; on growth of Mexican architectural art, 348, 350-2. Banner stones, 115. Bartlett (J. R.), on ruins at Casas Grandes, 224, 352. Bartram (Wm.), cited, 144. Bees, domesticated in Central America, 244. Bessels (Emil), 182. Boas (Franz), on Eskimo stone (381) 382 Index structures, 38-9 ; on origin and movements of Eskimo, 45. Bossu (M.) cited, 144. Bottles from mounds, wide-mouth ed, 92; long-necked, 94; with winged-snake design, 94. Bowls from mounds, 89-90 ; repre senting human head, 93. Boyle (David), quoted, 88, 102. Brinton, D. G., cited, 324; quoted, 158, 344-5; on differences be tween Atlantic and Pacific types, 19-21; on the Toltecs, 238; on spread of nagualism, 362 ; on origin Central American calen dar, 365. Bronze, 11. Brownell, on Iroquois traditions, 161. Bureau of American Ethnology, vii, 30, 33, 62, 84, 103, 110, 119, 139, 172, 180, 230, 346, 359, 371. Burial, methods, 63-77 ; order as to position not usually observed, 65 ; iii eastern Tennessee, 65-70 ; in North Carolina, 70; in box- shaped stone sepulchers, 70-2 ; in wooden vaults, 73 ; in Gulf states, 73 ; in ossuaries, 74 ; inhuma tion, 74; in Canada, 74; in Ar kansas, 74-5 ; use of fire at, 75-7, 144 ; peculiar method in north eastern Missouri, 77 ; in triangu lar pit, 80 ; in urns, 97 ; beneath dwellings, 144 ; in sitting posture, 144; removal of fiesh before, 144; body wrapped in matting, 145 ; on north-western coast, 182. Burial mounds, 61-8 ; stratification of, 62 ; north-western type, 64. Buschmann (J. C. E.), cited, 234 ; on relation of Shoshone and Nahuatl stocks, 316, Caddoan stock, 49, Cahokia mound (Illinois), 63, 127. Cakchiquels, their migrations, 328, 365 ; traditions, 329 ; annals, 364. Calaveras skull, 190. Calendar, Mexican and Central American, 241-3; not used by Huastecas, 336 ; spread of, 361 ; based on lunar count, 363 ; origin of, 367. California (section), 170, 187-202. Canon de Chelly, 230. Caracol (see Tower). Casa del Gobernador, Uxmal, 291. Casa Grande, 221,232. Casas Grandes, 223-9, 232, 349. "Castle" (Castillo), Chichen-Itza, 300. Cave dwellings, 205-8 ; method of forming, 205; along the Rio Mancos, 207. Central America, defined, 233; its monuments, 276-311 ; architec ture, 339-42 ; civilization, 339 etseq. Chaco cafion, ruins in, 219-20. Chain mounds, 57. Chapanecs, 234 ; migrations of, 320. Charencey (H. de), cited, 333. Charney, Desirfi, on the Toltecs, 240 ; his discoveries at Tula, 257; on ruins of Teotihuacan, 262-3 ; on ruins at Mitla, 272, 274 ; on ruins at Comalcalco, 276-7; on Palenque, 285; on ruins at Lo rillard City, 286-7 ; on origin civ ilized races, 348-9. Chase, A. W., on shell mounds of Oregon, 185, 198. Cherokees, in valley of Little Ten nessee, 68, 76 ; mound-builders, 82, 153-4. Chichen-Itza, ruins, 296-302; or namentation of buildings, 297 ; Index. 383 nuns' palace, 297-8 ; tower, 299 ; castle (castillo), 300; gymna sium, 301 ; house of tigers, 301 ; age of, 302, Chichimecas, 236, 240. Chicomoztoc, 319, 355, Chihuahua, its archaeology, 221-3, 232, Cliiriqui types, 180, 244, Cholula, pyramid of, 267 ; not Na- hautlan, 268. Civilization, of Mexican section, 240-51 ; not limited by tribal or stock boundaries, 240-1 ; Central American, 339 et seq.; Central American, rise and growth of, 356 el seq.; begins with cultivation of the soil, 375 ; retroactive, 376. Classification, 10-21 ; difficulties in the way of, 11 ; division fnto four ages not applicable in America, 11; primary classes, 12; chiefly geographical, 13-21 (see Archaeo logical Divisions) ; as to districts, 29. Clavigero (F. S.), cited, 239. Cocomes, 324. CogoUudo (D. L.), 323. Columns (see Pillars). Comalcalco, ruins at, 276-7 ; men tion, 285, 347. Comanches, 316, Comox (B. A.), mounds near, 184. Copan, ruins at, 305-10; types, 309-10. Copper, among the Mexicans, 244 ; of mound-builders chiefly from Lake Superior, 113 ; distribution in Atlantic division, 166. Copper articles from mounds, 109- 13 ; unusual type from Hopewell mound (Ohio), 110; with fig ured designs (Georgia and Il linois), 111-12. Coronado (F. V. de), 231. Craniology, of doubtful value, 10. Crees, 160. Cremation, on north-west coast, 184. Cukulcan (Maya deity), 257, 293, 302, 324, 363, Culture, development of, 376 (see Civilization). Gushing (F. H.), cited, 224, Dakota, effigy mounds in, 55. Dall (W. H.), quoted, 18; on Aleu tian shell heaps, 36-7 ; ou Cala veras skull, 191 . Den^ (tribe), 172, 174, 175, 338 (see Athapascans). De Soto (Hernando), expedition . of, 140. Dixon, on Yakutat burial, 182-3, Dog-ribs (Indian tribe), 174. Dorsey (J. O.), cited, 162. Douglass, A. E,, suggestion as to collections, 29. Dumont (G. M.), on shell orna ments, 105. Du Pratz (Le P.), on manufacture of pottery by Indians, 96 ; on Indian houses, 135-7. Effigies, surface, 149. Effigy mounds, described, 55-6; location, 59 ; belong to same era as other mounds, 148. Elephant mound (Wisconsin), 24, 56, Elephant trunk ornament, 299, 366; masks with, 293, Elongate mounds (see Wall mounds). Engraved shells, 67, 81, 83, 104-6 (see Shell articles, etc.). Eskimo, area, 35 ; stone lamp, 37 ; labret, 38; their ancient houses, 384 Lndex. 39 ; ulu or woman's knife, 41 soapstone pots, 42; adze, 42 skin scraper, 43 ; fiint flaker, 43 their culture home, 43-7; their migrations, 45; their southern extension, 146. Estufa, 231, Etowah mounds (Georgia), 63, 106, 111, 114, 127. Evans, R. B., discovers double ter raced mound (Arkansas), 118, Fancourt (C. St. J.), cited, 324, Fergusson, cited, 340, Fewkes, W. J., on pueblo cults, 359. Fire, used at burials, 75. Florida, shell heaps in, 60, Fort Ancient, Ohio, 125. Foster, J. W., cited, 316. Fowke, Gerard, on mound-build ing, 63-4. Fretwork at Mitla, 272. Gallatin, Albert, on original Al gonquian stock, 158; on migra tions of Siouan tribes, 162. Garcia, on origin of Chapanecs, 320. Gatschet, A. S., quoted, 163, 164-5, 317 ; on migrations of .^Lthapas- can tribes, 175. Georgia, effigy mounds in, 55. Gibbs (Geo.), quoted, 317 ; on mi grations in Washington and Ore gon, 202; on migrations on Pa cific side, 372-4. Gila valley and Chihuahua, 221-9. Glacial man (see Paleolithic age). Gold, 244. Gomara (F. L. de), on migrations of Xicalancas, 334. Government, Aztec and Tezcucan, 248. GoA'ernor's house, Uxmal, 290-2. Graves with stone sepulchers, 71 ; seldom found in Gulf states, 73. Greenland, eastern, ancient stone houses of, 38. Grinnel, G. R., on migrations of the Blackfeet, 160. Gucumatz (deity of Guatemalan tribes), 363. Guerrero (Mexico), recent discov eries in, 379. Haidah Indians, 177, 182. Hale, Horatio, on original home of Lenape, 158 ; on relations of the Winnebagos, 162. Hardin county,. Tennessee, inclos ure in, 130. Hayden, F. V., on migrations of the Blackfeet, 160. Hawks' bells in mounds, 86, 142, Haywood (John), cited, 76. Hemenway expedition, 33. Herrera (A. de), cited, 288; on Mayan traditions, 323-4, 341. Hieroglyphs, 246-8, 304 ; primitive forms wanting, 343; the work of priests, 357; Mayan and Mexi can diflTer, 360; origin of the Mayan, 360-1, 364, 367. Hill fort. Perry county, Ohio, 127. History of western continent, com mencement of, 7. Hodge, F. W., on advent of Nava jos in Arizona, 175; on ancient irrigation in Arizona, 221; on origin of clifi'-dwellings, 230. Hoernes, Moriz, cited, 13. Holmes, W. H., cited, 88, 103, 108, 180, 207, 279; quoted, 34.1-2; Index. 385 method of study followed by, 30-2 ; on pottery, 96 ; his classi fication of ruins in pueblo sec tion, 204; on clifi'-dwellings, 210-11 ; on Aztec Spring ruins, 215-17 ; on ruins at Teotihuacan, 261-2; on Mitlan art, 269-70; on Central American structures, 279 ; on ruins at Chichen-Itza, 297. House sites, 133-4. Houses of mound-builders, 133-7 ; material and construction, 135. Huastecas, 234, 321-2, 326, 847, 366 ; location and migrations of, 332-6. Hupas, 174. Huron-Iroquois stock, 154, 159, Hut rings (mounds), 132 ; probably remains of wigwams, 133. Illinois, effigy mounds in, 55. Inclosures, 121-32; their forms, 121; their area, 122; the object for which constructed, 131, Indiana, mounds of, 59. Indians, first known inhabitants, 8-9; were mound-builders, 138- 45 ; no proof of race preceding, 146, 150. Inhumation, 74. Iowa, effigy mounds in, 55. Iron, unknown as a metal, 11 ; in mounds, 67, 83, 142. Iroquoian family, territory of, 48; location, 157; migrations, 160-1, Itzaes, 302 ; their migrations, 325, Itzamna (Maya deity j, 293, 323, 363, Irrigation, ancient, in Arizona, 221. Izancanac, 284. Jones, C. C, quoted, 76. 25 Keane, A, H,, on paleolithic man, 6; on relation of language and race, 371, Keary, quoted, 6, Kentucky, mounds in, 59. Kera (stock), 229, Kiva (see Estufa). Labrets, 181; Eskimo, 38; use of, spread north-east, 47. La Hontan, cited, 144, Lamps, Eskimo, 37, Landa (Diego de), quoted, 301-2; on Mayan tradition, 323-4, Language, impoiunice of, in ar chaeology, 9 ; aa racial test, 370-1, Lapham, I. A., on age of mounds, 148. Lawson (John), cited, 145. Lead, 244, Le Due, Violet, cited, 340, 341, Lenape migrations, author's tlii'- ory, 159, Le Plongeon (Dr. A.), discovery by, 345, Lewis, T, H., on distribution of effigy mounds, 148. Linguistic stocks, of Atlantic di vision, 48 ; in California and Or egon, 187, 200, 372 ; of Mexican section, 233-4. Linn Works, Union county, Illi nois, 129. Lipans (tribe), 174. Lizana (Bernardo), on early Mayan tradition, 323. Lorillard City, 286-7. Los Edificios, ruins known as, 251-6 ; pyramid at, 253 ; arch not used at, 255. Lubbock, Sir John, quoted, 3 ; on migrations, 313, 386 Index. Lumholtz ( ), on ruins of Sierra Madre, 222. Maize, 243; discovery of, 375. Mams (Maya tribe), 322, 331, 335, Mackenzie, Alex,, cited, 176, Mancos Canon, ruins in, 210, 219. Mangues (Chapanec tribe), 320, Manuscripts, pre-Columbian, 245, Maps, necessary, 26, Mascoutens (tribe), 159, M.asks, 107, 181 ; snouted, 293, Mason, O, T,, on aboriginal skin dressing, 49, Materials for study, 8-10, Matlazinoas 241. Maudslay, A., P., quoted, 305; on ruins at Tikal, 303 ; on ruins at Quirigua, 304. Mayan days and symbols, 242. Mayan stock, 234 ; tribes of, 241, 321-2 ; migrations of, 326, 328-38 ; origin and route of, 366, Mayapan, 288, 302, 324, Mayas, 239 ; their earliest tra dition, 323; origin, 354; priestly power among, 357 ; in Mexico, 359-60, McGee, W. J., his notice of " en trenched mountains," 222, Mechanical arts of Nahuatl tribes, 244, Menche (see Lorillard City). Mercer, H. C, cited, 115, 146; cave explorations by, 150; on caves of Yucatan, 342, Mesa Verde, ruins of, 208-10, 213, 220, Metates, 226, Methods of study (see Study), Mexican, days and symbols, 242; codices, 272, Mexican section, 170, 233 ; its boundaries, 233; stocks, 233-4. Mexicans, their mechanical arts, 244. Mexico, southern, monuments of, 251; north-western, antiquities of, 342, Migrations, 23. Prehistoric in At lantic division, 152-68; theories regarding, 156-7 ; direction south, south-east and east, 157. Ou Pa cific side, 200-2; of Mexican and Central American tribes, 312 el seq.; Bancroft on, 314-15. In North America, chiefly south ward, 369 ; starting point of, 369; two chief streams, 369 ; two the ories in regard to, 369-70 ; a slow process, 374; checked by adop tion of agriculture, 375 ; of cer tain Mayan tribes, 328 et seq. Mindelefl', Cosmos, on origin of clifi'-dwellings, 230-1. Mindeleg', Victor, plan of Moki village by, 218, Mississippi, mounds in, 59, 62. Mitla, 33; its ancient structures, 268-73 ; general character, 269- 70 ; buildings of one story, 271 ; roofs flat and supported by col umns, 271 ; stairways not in use at, 271 ; fretwork ornamenta tion, 272; painted designs, 272; its pottery, 273; its age and de struction, 273-4 ; founders of, 319. Mixes, their migrations, 320. Mixtecs, 241 ; their culture and relations, 268 ; migrations, 318-19. Moki (tribe), 222, 229, 359; village type, 218. Montagnais (tribe), 160. Monte Alban, 33 ; ruins at, 268. Index. 387 Month, lunar, formerly in use among th« Mayas, 363-4. Monuments, important in study of archaeology, 9 ; term defined, 12 ; authors of, 22-3; of Atlantic di vision, 50; of southern Mexico, 251 et seq.; of Central America, 276 et seq.; North American com pared with South American, 378. Moon, pyramid of the (Teotihua can), 259, 260. Moore, C. B,, cited, 110; his ex plorations, 155. Moorehead, W. K,, explorations of, ' 78, 83 ; on pottery of Ohio, 97 ; dis covery of copper articles by, 110. Morgan, L. H., on inclosures, 131 ; on relation of Muskhogean and Siouan tribes, 164. Morice, A. G., quoted, 36; on cop per among Den4, 166, 174; on archaeology of Athapascan re gion, 171-2. Motolinia (T. de B.), quoted, 266; on Mitla, 274. Mound-builders, were Indians, 138-45 ; customs similar to those of the Indians, 143-45 ; had no settlements east of northern Al- leghanies, 154. Mound-building age, its duration, 147-52; one and unbroken, 149. Mounds, exploration of, 31 ; va rious classes, 51 ; conical, 51 ; elongate or wall mounds, 51-3; pyramidal, 53; effigy, 55-6; ar rangement in groups, 57 ; in lines or chains, 57; location, 58; dis tribution in Atlantic division, 59-60; how built, 61-4; stratifi cation, 62; age of, 82; distribu tion of articles and skeletons in, 84-6; antiquity and authors, 138- 46; Indians the authors of, 138- 45 ; notice of, by De Soto's chron iclers, 140-1 ; European articles found in, 142; building contin ued ii|ito post-European times, 143-52; location of the oldest, 153; Mexican mounds of wor ship, 266. Murdoch (John), on Eskimo adze, 43, 172 ; on culture home and movements of Eskimo, 45 ; on use of labrets, 47. Museums, many articles in, with out satisfactory history, 24, Muskhogean family, 48; migra tions of, 163-4. Nachan (" City of Serpents "), 284. Nadaillac (Marquis de), quoted, 64 ; on classification, 12 ; on age of mounds, 151 ; his " Prehistoric America," 191. Nagualism, 362. Nahuatl (and Nahuas), stock, 233- 4, 241, 268, 316; tribes, 244, 250; movements, 354. Nascapee (tribe), 160. Navajos, 174; at one time clifi'- dwellers, 229. Newark works, Ohio, 122-5, 129, New York, mounds in, 59. Niblack, A. P., on north-west coast Indians, 19; on north-west coast figures, 177-8. Nicaragua, monuments of, 311. Nomenclature, 29. Nordenskiold (G.), cited, 215, 219; his classification of ruins in pueblo section, 204 ; discovery by, 212-213. North America, probably peopled from Pacific coast, 378. North Carolina, mounds in, 60 ; triangular pit in, 80, 388 Lndex. Northern (Pacific) section, 170, Northmen left no impress, 2. North Pacific, coast, 176-86; flg ures, 177-81 ; totem posts, 179 ; mode of burial, 182 ; cremation, 184. Oaxaca, ruins in, 268, Ohio, effigy mounds in, 55. Ojibwas, 159-60. Olmecs, 319. Ontario, pots from, 88. Ordinez, cited, 284. Oregon, shell mounds of, 185. Orozco y Berra, cited, 331 ; on de struction of Mitla, 273 ; on origin of Chapanecs, 320 ; on origin of Central American calendar, 365. Otomies, 234. Ottawas, 159. Pacific division, 169 et seq.; defined, 17 ; types of, difi'er from those of the Atlantic division, 17-20 ; eastern boundary, 169; sections of, 170; its structures were of adobe and stone, 377 ; probably peopled before the Atlantic sec tion, 377. Packard, R. L,, cited, 113. Painting, Mayan, 245; Mexican, 245; Mitlan, 272-3. Palacio, Diego de, on ruins of Co- pan, 305-9. Palenque, Holmes's study of, 32-3 ; ruins at, 278-85; the palace, 279-82; other structures, 282; roof comb a singular feature, 282; history, 284; a religious center, 285. Paleographic objects, 12, Paleolithic age in North America not yet proven, 5-7 ; not dis cussed, 5, Palgrave, F,, quoted, 2, Parry Archipelago, ancient stone houses in, 38. " Pathway of the Dead," Teotihua can, 259-60. Penafiel, Antonio, notes recent dis covery at Tula, 346. Perez de Ribas, on Sonoran tradi tions, 318. Petitot, E. F,, on copper among theDen^, 166, 174, Pigeons, House of, Uxmal, 294, Pigmies, supposed burial place of, 71. Pillars, at Quemada, 252, 255 ; at Teotihuacan, 261 ; at Xochicalco, 265 ; at Mitla, 271 ; at Palenque, 282. Pipes from mounds and graves, 98- 102; stemless type, 98-100; im age form, 99; short-necked type, 101 ; " Monitor," 101 ; long-stem type, 101 ; tubular form, 102. Pirindas, 241. Popol Vuh, 328, 330, Population, Indian, 374-5. Pots from mounds, 91. Pottawotomis (tribe), 159. Pottery, 87-97; in Canada and northern section, 87 ; in Missis sippi province, 87-8 ; method of tempering, 88; of the Gulf prov ince, 97 ; among Mayas and Mex icans, 245 ; of Mitla, 273. Powell, Major J. W., cited, 143; on authors of the mounds, v-vi ; discovery by, 112; linguistic map of, 169, 200 ; his description of Colorado Canon referred to, 209; on climatic conditions of pueblo region, 230; on Uto-Az tecan family, 234; on formation, of languages, 371. Index. 389 Powers, Stephen, on tribes of Cali fornia, 187 ; on California relics, 197-9 ; on migrations in Oregon, 201-2. Poynter (R. H.), on Indian burial, 76. Prescott, W. H,, on Mexican his tory, 234-8. Priesthood, power of, in Mexico and Central America, 249, 357; higher culture due to, 358 ; among the pueblos, 359 ; organ ization of, 358-60 ; of different stocks co-operated, 362. Priests, influence in advancing civ ilization, 356-66 ; inventors of the hieroglyphs, 357. Pueblo Bonito, ruins of, 219. Pueblo, ruins, 215-20; where found, 215 ; builders, 229. Pueblo section, 28, 170, 203-20; physical character of, 203. Pyramid, of the Sun, 259 ; of the Moon, 259-60. Pyramidal mounds, 53, 117-20 ; with terraces, 118-19; with graded way, 118; locations of, 121. Quapaws (tribe), 131. Quemada, 354 ; ruins near, 251-5 ; show analogies to those of Cen tral America and southern Mex ico, 253-4; temples, 347; possi bly partly Zapotecan, 348. Quetzalcoatl (Mexican deity), 257, 284, 302, 319, 360, 362. Quiches, their migrations, 326-31. Quirigua, 303-4. Ramsey (J. G. M.), cited, 76, Rawlinson (Geo.), cited, 340. Relics and remains defined, 12; classified, 13, Ribas (Fr, Perez de), on traditions of north-western Mexico, 318. Rink (H,), on Eskimo culture home, 44-5, Rio Chico valley, 217. Rio Mancos, 207-8. Roseborough, Judge, on migra tions in Oregon, 201. Sacs and Foxes, 159, Sahagun (B. de), cited, 256-7, 319, 333-4. Salado valley, ruins in, 221-2. Sapper, Carl, cited, 304. Schoolcraft, H. R,, on Winnebago tradition, 163. Schumacher, Paul, on California types, 187. Sculpture, Mayan and Mexican, 245 ; at Lorillard City, 286. Seler, Ed,, on origin of Central American calendar, 365, Selsertown mound (Mississippi), 119. Serpent, figure 94, columns 256, 301. Seven caves, tradition of, 355. Shell, articles of, from mounds, 103-7; engraved, 105-6; gorgets, 106; masks, 107; beads, 107; wampum, 107. Shell-heaps, Aleutian, 36-38, Shoshones, 139, 229, 234, 316. Sierra Madre, ruins in, 222. Silver, 244. Siouan stock and tribes, location of, 49, 157; migrations, 162-3, 167-8; not cultivators of the soil, 167. Sonora, ruins in north-eastern 390 Lndex. part, 222 ; tradition of tribes in, 318. South America not peopled from North America, 378 ; no relation between the civilizations of the two, 379. Squier (E. G.), cited, 316. Squier and Davis, cited, 97, 131, 133. Stairways, 277. Starr, Frederick, Mitlan flgures by, 365. Stephens (J. L.), on palace at Pa lenque, 282. Stevenson, Mrs. M. C, cited on pueblo cults, 359. Stoll, Otto, cited, 322. Stone, articles, 113-16; images, 114; arrow-heads, types of, 116. Stone graves, Indians buried in, 72 (see Burial); area of, 154. Study (of Archaeology), methods of, 22-34, Sun, pyramid of the (Teotihuacan), 259-60, Swallow (G. C), ou mound-build ers' houses, 135. Tallegwi (Talega), identified with the Cherokees, 154. Tarascos, 234, 241. Tehua (stock), 229. Tello, Fr,, on ruins at Quemada, 251, Temple of the Dwarf, -Uxmal, 293. Temples, Yucatec, 282. Tennessee, mounds in, 59. Tenochtitlan (Mexico City), no ruins of structures remain, 264. Teotihuacan, 269, 360 ; antiquities at, 259-63, 270; its great pyra mids, 259 ; pathway of the dead, 259; citadel, 260; hewn stone but little used at, 260-1 ; not Nahuatlan, 261-2; pre-Aztecan, 345. Tepanecs, 236. Textile fabrics from mounds, 108. Tezcucans, 236 ; their form of gov ernment, 248. Tikal, ruins at, 303 ; wood carving at, 303. Timuquanan stock, 49. Tlaloc (rain god), symbols, 299; statues of, 344; represented in inscriptions, 344. Tohil (Quiche deity), 360. Toltecs, 235, 236, 238-40; Brinton on, 238 ; Clavigero on, 239 ; prob ably Mayas, 240; Charney on, 240, Torquemada (Juan de), on Teoti huacan, 262; on the Zapotecs, 319; on the Totonacas, 332. Totem posts, 179. Totonacas, 234, 241, 332-3, 336-7, 347, 366. Tower, Chichen-Itza, 299. Traditions, use of, 10; Mexican and Central American, 362. Tribes, permanency as to habitat, 23. Tula, 235, 345-6; antiquities of, 255-7 ; various forms of the name, 255; serpent columns at, 256; sculptured types, 346; hiero glyphs found at, 346. Tulan, 326, 32&-331, 337, 360,364-5. Tulan Zuiva, 355. Tuolomne county, California, re markable discoveries in, 189. Tutul-Xiu, 294, 325, 341. Tylor (E. B.), cited, 356. Types, geographical distribution of, 26, 29 ; essential features of, 27. Tzutuhils, migrations of, 328-30. Index. 391 Ukiadliving (British America), sin gular stone structures at, 39, Ulu or woman's knife (Eskimo), 41. Uto-Aztecan stock, 233-4, 316, 338, 359. Uxmal, ruins at, 288-95; its gov ernor's house, 291-2 ; nunnery, 292-3; temple of the dwarf, 293; house of pigeons, 294, Valencia. Fray Martin, cited, 274, Vancouver (G.), on north-west coast villages, 184. Veytia (M.), quoted on Tula, 257. Violet Le Due, on Mayan build ings, 340-1. Votan (Tzental deity), tradition regarding, 284, 360, 362. Walam Olum, 160, Wall mounds, 51-3, West Indies, peopled from South America, 379. Wheeler, Lieutenant, on ancient California burials, 191-6. Whitney (Prof,), cited, 190. Wickersham (Judge), on stone warclubs, 173. Wilson, Daniel, on traditional home of the Iroquois, 161. Winnebago Indians, 58; relations of, 162. Xibalba (Xibalbay), 284, 328-9. Xicalancas, 319, 334. Xochicalco, ruins at, 264-6; exca vations at, 264-5 ; pyramid of, 265. Yakutat, mode of burial, 182, Yellow Knives (Indian tribe), 174. Yucatan, people of, split into groups or tribes, 248; monu ments of, 287, 339. Yuit, migration of, 44. Zacatecas, ruins in, 251. Zapotec-Mixtec (stock), 234. Zapotecs, 241, 273, 337, 348 ; their culture and affinities, 268 ; cal endar, 268; migrations, 318-19; origin, 354; priestly power among, 357 ; influence of, on Mexicans, 365 ; pioneers in Cen tral American civilization, 366; authors of the Central American calendar, 367. Zoques, migrations of, 320. Zunis, 222, 229. YALE UNIVERSITY a39002 .002tt'4^298b *iiJiJ5'i^l^i''^;^ i^^,it"'i^-i^'l''-*''i'^4t'r'*^TTt'S.clSi.f^'MK^ ¦'"'''¦^"i- 'rj: "t-.''''" ¦''"'''^ ^'.^ *..*'-''' ' »^j ^ t . . ¦*,*.'"