4 -f, ■■ J^^>. •^9 ^ 4. /- ^* \ *>, » . 1 ' . •< " f > ,0 «. j% . ",. A-i- (J I 1 "* \Q '-^^ " 1 IM O H I I fi *, ■? f J '^ 3 N " ^-i, °°'^ .^^ .. N c , <::^ '-"^ -^ ^''""^ 'o ^ * -J N \V , v^;-;. ^"<\" ' ■A V • ■ '^ xO°<. v> .^^•% "^-^ V^ .;xV )^ O .i. -X = Digitized by the Internet Archive in 2010 with funding from The Library of Congress http://www.archive.org/details/inscriptionsatcoOOmorl THE INSCRIPTIONS AT COPAN BY SYLVANUS GRISWOLD MORLEY ASSOCIATE OF THE CARNEGIE INSTITUTION OF WASHINGTON PUBLISHED BY THE CARNEGIE INSTITUTION OF WASHINGTON Washington, 1920 ^ THE INSCRIPTIONS AT COPAN fcaO .S < M C '-3 o O 3 -a c o O O c ■3 et ibid., 1919, pp. 320, 321. ■■The origin of the Maya civilization has been a fruitful field of inquiry since the days of the Spanish Conquest, when the conquistadores first beheld the great aboriginal cities of Middle America; and there is scarcely a country on the face of the globe which has not at one time or another been identified as the original home of the Maya race. From Lord Kingsborough's nine large folio volumes, which sought to prove that the Maya were descended from the Ten Lost Tribes of Israel, down to the present time, innumerable earnest but misguided attempts have been made to derive the Maya directly from some nation or peopleof the Old World. Some indeed have even gone so tar as to reverse this at least natural procedure, and have contended that Egypt was colonized from .America, claiming for the Maya an antiquity of more than 11,000 years (Le Plongeon, 1886). Unfortunately, such aberrations as these have not entirely disappeared before the advance of modern research, since we find within the past decade earnest searchers after the truth solemnly connecting the Maya civilization with the comparatively recent cultures of Java and Southern India (Arnold and Frost, 1909), or, by more or less circuitous routes of migration, with the equally remote civilization of ancient Egypt. (Smith, 1916-1917, pp. 190-195, 241-246.) The last even contends that the elephant's head occurs as a decorative element on Stela B at Copan, (1915-1916, pp. 340, 341, 425, 593-595)- He has been ably refuted by Tozzer (1916, p. 592), Spinden (1916, pp. 592, 593), Goldenweiser (1916, pp. 531-533). and Means (1916, pp. 533, 534). Such extravagant hypotheses would scarcely merit even passing notice were it not for the fact that their very spectacularity renders them peculiarly attractive to the general public. It is, therefore, perhaps not superfluous to repeat that the Maya civilization was a native American product, developed in its entirety in the New World, and probably not far from the region where its extensive remains are now to be found. MORUEY PLATE 4 a. General view of the Main Structure looking east, showing the plain covered with the remains of stone buildings. b. General view of the Main Structure looking south from the hill near the quarries. CouTttiy af the Pcalifidy Mitsatm HISTORY OF DECIPHERMENT OF MAYA HIEROGLYPHIC WRITING. 29 the Royal Academy of History at Madrid in 1863. In the winter of that year the eminent French antiquarian. Abbe Brasseur de Bourbourg, while on a scientific mission to Spain, found in the archives of that establishment a manuscript entitled Relacion de las cosas de Yucatan, sacada de la que escribio el padre Fray Diego de Landa de la orden de San Francisco, and bearing the date MDLXvi.^ The author of this manuscript was no less a person than the Padre Diego de Landa, who later became the fourth bishop of the Diocese of Yucatan and Cozumel with seat at Merida, and who filled that see from 1573 until the time of his death in 1579.- He was one of the first Franciscans to enter Yucatan (1549) and one of the most zealous in converting the Indians from their idolatry. His zeal in this latter direction eventually brought him into difiiculties with his ecclesiastical superior. Bishop Toral, who accused him of usurping higher functions than were his right; and he later returned to Spain and appeared before the Council of the Indies to answer the charges growing out of this controversy. He was tried before a body of "seven learned persons of the Kingdom ot Toledo," in 1565, and was finall}^ exon- erated of all the charges against him in 1569. On the retirement of Bishop Toral, four years later, he was named to succeed to the vacant see, which he occupied until his death in 1579, as noted above.-'' The especial value of Landa's "relacion," a veritable Maya Rosetta Stone indeed, lies in the fact that it was composed during the first generation after the Spanish Conquest, 1 541-1566; and the information which it contains was obtained by him directly from natives who had reached mental maturity under their own social, political, and religious institutions before the shock of European conquest had forever shattered the native regime. Indeed, Landa actually states that one of his informants was Don Juan Cocom, a lineal descendant of the last ruler of Mayapan, and a man particularly well versed in the ancient learning of his people.* And with such information as this 'This valuable manuscript has been thrice published, as follows: (i) By the Abbe Brasseur de Bourbourg with a French translation, Paris, 1864. (2) By Juan de Dios de la Raday Delgado, Madrid, i8Si,as an Appendix to his Ensayo sobre la inttrpreiacion de la escriiura hierdtica de la America Central, a translation into Spanish, of Leon de Rosny's work of the same name. (3) By the Royal Academy of History, Madrid, in the Coleccion de Documentos Ineditos relativos al descubrimienio, conquisia y organhacion de las antiguas posesiones espaiiolas de ultramar, Segunda serie, torn. 13, Relaciones de Yucatan, II, Madrid, 1900. Of these, the first is by far the poorest; Brinton, in a critical comparison of the first two editions (1887), points out its shortcomings. Brasseur de Bourbourg omits, without a word of explanation, fully one-sixth of the original (the last part of the manuscript) and inserts a num- ber of chapter headings not found in the manuscript at all, in addition to making many errors of translation. The second edition is the only complete one of the three, i. e., having all the text and drawings of the original manu- script. Unfortunately it was limited to 200 copies, all large folios, and is very rare. All references to Landa in the present work, however, are to this second and only complete edition. The third edition is not only comparatively rare but is also incomplete, lacking all the part describing the calendar and annual feasts, including the drawings of the glyphs, the katun-wheel, and Landa's famous "alphabet." There is an unpublished English translation by Bowditch in the Peabody Museum library. ^Molina Solis (1897, pp. 13-78) reviews the evidence bearing upon the creation of the Diocese of Yucatan and Cozumel, which he concludes was established after 1547 and before 1553, Fray Juan de San Francisco being named the first bishop. The latter resigned before 1552, never having gone to Yucatan, in which year Fray Juan de La Puerta was named second bishop of the diocese. The latter died at Sevilla on the eve of his departure for New Spain with twenty frailes, and it was not until 1562 that the third bishop. Fray F'rancisco de Toral, reached Merida. ^Landa himself describes the afl^air at some length (18S1, pp. 79, 80), and the whole controversy has been made the subject of a special papet by Medina (1913, pp. 4S4-496). *"The successor of the Cocomes, named Don Juan Cocom, who has since become a Christian, was a man of gteat reputation, learned in their affairs, and of remarkable sagacity and intelligence in native matters. He was very intimate with the author of this book. Fray Diego de Landa, and told him many facts concerning the antiqui- ties." (Landa, 1881, p. 76.) 30 THE INSCRIPTIONS AT COPAN. at his disposal, he speaks with an authority not equaled by that of any other early European writer. Landa describes the manners and customs of the Maya of Yucatan at considerable length, giving even a few tantalizing glimpses of their history. By far his most important contribution, however, is his description of the native calendar, in which he gives drawings of the hieroglyphs for the 20 days of the Maya month and of 18 out of the 19 divisions of the Maya year,^ and also a series of signs, which he claimed were the letters of the Maya alphabet. The last was at once hailed with acclaim by Americanists as the long-sought key to the hieroglyphic writing, and a number of investigators, more credulous than critical, hastened into print with so-called "interlinear translations" of the texts. In every case, however, these have broken down under "higher criticism," until it has finally become necessary to abandon all hope of translating the Maya inscriptions by means of the Landa alphabet, some even going so far as to brand it as a "Spanish fabrication."- But if we must reject Landa's alphabet in its entirety, as a phonetic key to the inscriptions, as now seems necessary, we are still deeply indebted to him for his illuminating observations on the Maya calendar, by means of which only we have been able partially to decipher the Maya writing. The first real advance in interpreting the Maya writing was made by Professor Ernst Forstemann, of the Royal Library of Dresden, who in the decade from 1880-1890, while Maudslay was doing valuable service in the field, published a number of studies on a Maya hieroglyphic manuscript in the Royal Library at Dresden. Using Landa's values for the day and month signs, Forstemann finally worked out the basic principles of Maya chro- nology, and in 1887 he announced the fundamental discovery that the long numbers of the Dresden manuscript designate particular days in Maya history and are all counted from the same starting-point, a sort of Maya birth of Christ, to borrow an analogy from our own chronology.'* Curiously enough, an American scholar, Mr. J. T. Goodman, of Alameda, California, working independently upon difi^erent subject-matter, i. e., the inscriptions on the monuments, and without knowledge of Forste- mann's researches, duplicated the latter's remarkable discovery a little later, 1883-95. Forstemann seems to have made his discovery as early as April 'Landa gives no sign for Uayeb (iSSi, pp. 96, 97). ^Valentini (1880). Many have been the attempts to decipher the manuscripts by means of the Landa alphabet. Brasseurde Bourbourg was the first in this field ( 1 869-1 870); and he was quickly followed by others, de Rosny (1876), Le Plongeon (1885), La Rochefoucauld (1888), Thomas (1893), and Cresson (1892, 1892a, and 1892^). These various attempts to explain the Maya writing on a purely phonetic basis have entirely broken down, and at present this "school" has few followers. See Brinton, 1S95, pp. 14-17, for an able summary of these studies. ^Forstemann, 1887. An English translation of this important paper was published by the Bureau of American Ethnology (see Forstemann, 1904, pp. 393-472). Tozzer (1907, pp. 153-159) gives a brief sketch of Forstemann's life and a fairly complete bibliography. Only a few of the more important titles bearing directly on the present investigation, however, have been included in the bibliography at the end of this memoir. The results of his studies on the manuscripts are given in most complete form in his commentaries on the three codices, Forstemann, 190!, 19020, and 1903, all three of which have been translated into English, and the first and most important, published as volume iv. No. 2, of the papers of the Peabody Museum (1906). He published two facsimile repro- ductions of the Dresden Codex (1880 and 1892), thus making this important manuscript accessible for general study, and also wrote a large number of shorter articles, only a few of which deal with the inscriptions and only one with those at Copan (1904a). His especial province, however, was the manuscripts, and here he easily ranks first. HISTORY OF DECIPHERMENT OF MAYA HIEROGLYPHIC WRITING. 3 1 1885;^ while Goodman, on the other hand, did not announce his results until 1895, though his preface states that he had been at work on the inscriptions for 12 years prior to that date.' Perhaps the fairest solution is to recognize the priority of Forstemann in the field of the manuscripts and that of Good- man in the field of the monuments. Goodman's preeminence in the field of the inscriptions, moreover, is beyond question. In addition to working out the Maya calendar system as used therein, he was the first to make known the existence of the head-variant numerals, the so-called Maya Arabic notation, and to identify their different values; and finally he devised his justly famous chronological tables, the logarithmic short-cuts of Maya arithmetic.^ As early as 1890 Maudslay had realized the importance of the first seven glyphs in each text, and had given them the name Initial Series, by which they have since been known.* In fact, in one of the earliest plates of the section on archaeology of the Biologia Centrali- Americana, he figured nine of these Initial Series together, an arrangement which clearly brought out their similarity.^ Because it made accessible, for the first time, accurate copies of the originals, the publication of Maudslay's work gave tremendous impetus to the study of the Maya inscriptions. Goodman testifies in this connection that it alone made possible his results; and other students were not slow to devote themselves to this problem, investigating not only Goodman's conclusions, but also the drawings upon which they were based. Goodman's report appeared in 1897, and two years later Professor Eduard Seler, o Berlin, published a long discussion of the Copan, Quirigua, and Palenque texts, based upon Maudslay's reproductions.*^ This contains little new material, however, Goodman having covered the ground rather completely. Simultaneously with the publication of Seler's studies, Cjtus Thomas, of the Bureau of American Ethnology, published an extended commentary on Goodman's work.'' This is a critical examination of the latter's con- clusions, with which it agrees in the main. Some new points are brought out, but in general Goodman's theories are sustained. The most valuable '■"It was a source of special satisfaction to me that in April 1885 I was able to determine the sign for zero and soon afterward to discover the way in which the Mayas expressed the higher numbers, so that they can now be read from zero up to millions. Upon this discovery is based the largest part of my later researches." (Forstemann, 1894, p. 78.) And agam: "In the year 1S85 the reading of all Maya numbers up to millions svas found here [in the Dresden Codex], 1887, the origin of the historical Maya reckoning was found, and also the form of the calendar date composed of two numbers and two hieroglyphs was recognized [i.e., 4 Ahau 8 Cumhuj," (Forstemann, 1902^, P- ISO.) ^Goodman, 1897, p. iii. Goodman's results were published as the Appendix to the section on archaeology, of the Biologia Centrali- A7nericana, m 1897. 'For a brief review of the scientific and literary achievements of Goodman, see Morley, 1919a, pp. 441-445. ^Maudslay, 1889-1902, vol. I of text, p. 40. ''Ibid., vol. I, plate 31; also in vol. 11, plate 65, and vol. iv, plate 92. 'Seler 1899, pp. 670-738, and 1900, pp. 188-227. These were published in the Ferhandlungen der Berliner anthropologischen Gesellschajt for November 18, 1899, and for March 17, 1900, and later in Seler's collected works (1902-1908, vol. I, pp. 712-836). There is an unpublished English translation of these articles in the Peabody Museum by Bowditch, who has had practically all of Seler's works translated. 'Thomas, 1900, pp. 693-819; also 1904, pp. 197-305. 32 THE INSCRIPTIONS AT COPAN. contributions of Thomas are rather in the field of the manuscripts than in that of the inscriptions. His commentaries on the codices are particularly valuable and in some lines parallel the work of Forstemann.^ Although most of his writing preceded the period of precise decipher- ment. Dr. D. G. Brinton, of the University of Pennsylvania, also made valuable contributions to the general field of Middle American archaeology, and particularly to its aboriginal history. He published many original sources, the Maya Chronicles, the A^inals of the Cakchiquels, etc., with highly illuminative notes and comments by himself, and with justice he may be called the "Father of Maya History," in which field for many years he stood practically alone.- At one time it was the fashion to discredit these native chronicles and to regard them as untrustworthy and of little consequence, but now the whole trend of Maya research is tending toward their complete authentication as reliable sources for the reconstruction of ancient Maya history, a position taken and ably defended by Brinton 37 years ago.^ The work of Mr. C. P. Bowditch deserves especial mention in this section. It was chiefly through his instrumentality that the Peabody Mu- seum first entered the field of Maya archaeology 30 years ago, and it has been largely due to his continued interest and support that investigations have been maintained there ever since. In addition to these activities, he has made important original contributions to the study of the hieroglyphic writing. His first paper on this subject was published in 1900, and during the next decade he brought out a number of articles containing much new information. The more important results of his researches are to be found in his Nu7neration, Calendar Systeius, a nd Astro noviical Knozvledge of the Mayas, published in 1910, a scholarly presentation of the subject, embodying not only all the previous knowledge but also the fruit of his own extensive labors.'' 'Thomas's writings are fairly voluminous. Most of them are to be found in various governmental pubhca- tions, chiefly those of the Bureau of American Ethnology, with which he was connected for many years. His more important contributions will be found in the bibliography under the following heads: Thomas, 1882, 1888, 1894, 1897, 1900, and 1904. ^Stephens was the first to publish one of these chronicles (1843, vol. 11, pp. 465-469). This particular manuscript, the Book of Chilan Balani of Mani, had been discovered at Ticul, Yucatan, by the Yucatecan antiquarian, Pio Perez, just before Stephens's visit to the country. Realizing its importance, Stephens prevailed upon Perez to allow him to publish the original Maya text and an English translation. A more scholarly treatise on the same manuscript is that by Valentini, The Katunes of Maya History (see Valentini, 1879), which contains a critical analysis of the chronological data, and an able defense of its historical accuracy. Brinton republished the Maya text with an English translation in his Maya Chronicles (Brinton, 1882, pp. 89-135). ^Brinton, 1882. Brinton's bibliography is extensive and varied. Chamberlain (1899, pp. 215-225) states he wrote over 150 books and articles from 1859 to 1898. His contributions to the study of the Maya hieroglyphic writing alone are too voluminous for complete citation here. They cover almost every phase of the subject, and in the field of history, as noted above, are without equal. The more important titles will be found in the bibliography under the following heads; Brinton, 1882, 1882a, 1882^, 1885, 1886, 1886a, 1887, 1890, 1894, 1895, and 1896. See also Culin, 1900. 'Bowditch, 1910. The hieroglyphs figured in this book were drawn by Mrs. A. J. Tretheway, whose work is both painstaking and accurate, and compares favorably with that of Miss Hunter. Mrs. Tretheway uses a heavier line than Miss Hunter, however, and is less prone to overdraw. While this characteristic gives her work a less finished appearance than that of Maudslay's illustrator, it probably makes it correspondingly more accurate, particularly in the delineation of texts from the Early and Middle Periods. For Bowditch's other publications, see the bibliography. SCOPE OF THE PRESENT INVESTIGATION. 33 The great expansion of interest in the Maya field following these spec- tacular advances in the decipherment of the Maya hieroglyphic writing, and corresponding advances in the whole general field of Middle American archaeology, have led to the establishment of courses of instruction in these subjects at some of the larger American universities, notably at Harvard, where this work is now in its second decade under Dr. A. M. Tozzer. The writer's previous publications on the Maya inscriptions have been confined to brief articles on special phases of the subject, and to a text-book entitled An Introduction to the Study of the Maya Hieroglyphs, published as Bulletin 57 of the Bureau of American Ethnology in 191 5, a work especially designed to meet the requirements of the beginner.^ SCOPE OF THE PRESENT INVESTIGATION. The present investigation is limited to a consideration of the chrono- logical data found in the Copan inscriptions. In the present state of knowl- edge it has appeared inadvisable to extend the research beyond this point into the realm of the undeciphered glyphs, since too little is yet known about them even to approximate their meanings. So far as they have been deciphered, however — and it is now possible to read about one-half of the characters — the Maya inscriptions have been found to deal exclusively with the counting of time. Brinton, with his usual acumen in such matters, clearly perceived this important truth 25 years ago, and in his Primer of Mayan Hieroglyphics gave it precise expression: "The frequency and prominence of these elementary numerals in nearly every relic of Mayan writing, whether on paper, stone, or pottery, constitute a striking feature of such remains, and forcibly suggest that by far the majority of them have one and the same purpose, that is, counting; and when we find with almost equal frequency the signs for days and months associated with these numerals, we become certain that in these records we have before us time-counts, some sort of ephemer- ides or almanacs. This is true of all the Codices and of nine out of ten of the inscriptions. Here, therefore, is a first and most important step gained toward the solution of the puzzle before us."^ Unlike the inscriptions of every other people ot antiquity, the Maya records on stone do not appear to have been concerned — at least primarily — with the exploits of man, such as the achievements of rulers, priests, or warriors — in short, with the purely personal phenomena of life; on the con- trary, time in its many manifestations was their chief content. The Maya priesthood, in whose hands exclusively rested the knowledge of the hieroglyphic writing, conceived time more elaborately than any other people the world has ever known at a corresponding stage of general culture. They observed and recorded its more obvious phenomena, the apparent revolutions of the Sun, Moon, Venus, and possibly other planets, solar eclipses, planetary configurations; and, most important of all, they accom- plished its exact measure: the accurate toll of the passing days. Of first 'Morley, 1915. For the writer's shorter articles see the bibliography. ^Brinton, 1895, p. 18. 24 THE INSCRIPTIONS AT COPAN. importance in this latter connection were the contemporaneous dates of the different monuments. These are usually found at the beginnings of the inscriptions, hence Maudslay's name "Initial Series," by which they are known, and they are of such amazing accuracy as to be fixed within a period of 374,000 years, a tremendous achievement for any time-count, European or otherwise.^ This truly remarkable chronological system was counted from a fixed starting-point, and it furnishes an exact measure of time, by means of which, it already appears probable, all related or connected cultures will eventually be datable. This point is of such importance as to warrant further elaboration. By means of the Maya chronological yardstick it is already possible to measure roughly the time element in the cultures immediately adjoining the Maya on the north: the Zapotec, the Miztec, the Totonac, the Toltec, and the Aztec; indeed, even to date them with considerable accuracy.^ But in addition to this, it is eventually hoped, by means of overlapping culture horizons, to extend this dating as far south as the great cultures of Peru and Ecuador,^ and as far north as the intensive cultures of the southwestern sec- tion of the United States.^ In short, it now appears as within the range of probability that ultimately all cultures of Middle America and of the con- tiguous parts of North and South America will be referable, by means of such ijf, as the writer believes, the eight time-periods recorded in sequence on Stela 10 at Tikal all belong to one and the same Initial Series, the starting-point of Maya chronology itself was fixed in a much grander chronological scheme, a chronology which at the time this monument was erected had already reached more than 5,000,000 years and might even have been expanded to 64,000,000 years without breaking down, a truly geologic conception of measured time. See Motley, 1915, pp. 114-127. 2By means of this accurate chronological control, Spinden (1917) has been able to work out provisional chronol- ogies for the above peoples {ibid., chapter ill), which are probably more accurate than Egyptian chronology in the Old Empire or Babylonian chronology at the time of Hammurabi. 'Means (1917(2, pp. 383-389, and 1918, pp. 152-170) has already attempted this with considerable success for the high cultures of Ecuador and Peru, and now the question would seem to be largely one for the arch.-Eologist, i. e., tracing closer cultural resemblances in pottery types, stone-carving, metal technique, and general esthetic designs between Central and South America. When these have been definitely established, the time element can be measured from the northern, that is to say, the Maya end. ^A beginning has already been made in this field also. The discovery of copper bells of Mexican origin at a number of archsological sites in Arizona and New Mexico clearly points to trade relations between the Pueblo Indians and the tribes of central Mexico in pre-Columbian times. Sites where such bells have been found are Awatobi (see Fewkes, 1898, pp. 628 and 629, and figure 261), Black Falls (ibid., 1904, p. 50), Chaves Pass (ibid., p. 1 1 1 and figure 67), Taylor (ibid., pp. 162 and 163, and figure 108), and Tonto Basin (Hough, 1914, p. 37 and figure 78), all in .A,rizona; and Tularosa, New Mexico (Hough, ibid., p. 38 and figure 79), and more recently at Aztec, New Mexico (Morris, 1919, p. 100 and figure 71a). Copper bells were also found at Pueblo Bonito, New Mexico, by the American Museum Expeditions under Putnam in 1S96 et seq.; and they have been reported as far east as the eastern part of Tennessee. (Thomas, 1894a, pp. 376 and 714, and figure 252.) These bells very closely resemble copper bells found in the Sacred Cenote at Chichen Itza, Yucatan (New Em- pire), which can hardly have been carried thither before 1200 A. D. The same is true of the so-called Mexican cloisonne or encaustic ware, also found both at Pueblo Bonito and at Chichen Itza in the Sacred Cenote, although indigenous to neither site, the region of manufacture being in the northern part of the State of Jalisco, Mexico— Totoate, Hacienda Estanzuela, etc., and the specimens found having had to be carried at least 1,600 kilometers to have reached either place. See Lumholtz, 1902, vol. 11, pp. 460-462 and plates 13-15, Hrdlicka, 1903, p. 385 and plate 39, and Spinden, 1917, pp. 161-164. Finally, the Andover-Pecos Expedition of 1915, under Kidder, found a Mexican spindle-whorl in a pre-Colum- bian stratum of the refuse heap at Pecos, New Mexico. These are sporadic cases to be sure, but they indicate none the less sutely that futther excavation in these areas will undoubtedly develop other points of contact between the Maya and the contiguous cultures of Middle, North, and South America, by means of which alone, in default of local chronologies, will it ever be possible to date the latter accurately. MORLEY PLATE 5 r ^'/ '^m^ ' ' a. The main structure looking southwest, showing the great cross-section exposed by the river. b. The main structure looking northwest, Courtesy af the Peahudy Miticum SCOPE OF THE PRESENT INVESTIGATION. 35 interlocking relationships, to the standard section of ancient American his- tory furnished by the Maya inscriptions. In addition to this purely chronological matter in the Maya inscriptions, there is also a mass of related astronomical data, as noted above. For example, the apparent revolutions of the Moon, Venus, and probably other planets were accurately observed and recorded. Solar eclipses were pre- dicted; and larger time-periods exactly containing different smaller time- periods were evolved.^ Moreover, as we gradually press our way into the meaning of the undeciphered glyphs, it is becoming increasingly apparent that these intricate graphic symbols deal more with the subject-matter of astronomy than with the details of history. With far-reaching vision and profound understanding, Brinton, in another passage written over 25 years ago, long before the results of modern research had justified such a conclusion, closely prognosticated this condition : "A careful examination of Dr. Forstemann's remarkable studies, as well as a number of other considerations drawn from the Codices themselves, has persuaded me that the general purpose of the Codices and the greater inscriptions, as those of Palenque, have been misunderstood and underrated by most writers. In one of his latest papers Professor Cyrus Thomas says of the Codices: 'These records are to a large extent only religious calendars'; and Dr. Seler has expressed his distrust in Dr. Forstemann's opinions as to their astronomic contents. My own conviction is that they will prove to be much more astronomical than even the latter believes; that they are primarily and essentially records of the motions of the heavenly bodies; and that both figures and characters are to be interpreted as referring in the first instance to the sun and moon, the planets, and those constellations which are most prominent in the nightly sky in the latitude of Yucatan. "This conclusion is entirely in accordance with the results of the most recent research in neighboring fields of American culture. The profound studies of the Mexi- can calendar undertaken by Mrs. Zelia Nuttall have vindicated for it a truly surpris- ing accuracy which could have come only from prolonged and accurately registered observations of the relative apparent motions of the celestial bodies. We may be sure that the Mayas were not behind the Nahuas in this, and in the grotesque figures and strange groupings which illustrate the pages of their books we should look for pictorial representations of astronomic events. "Of course, as everywhere else, with this serious astronomic lore were associated notions of astrology, dates for fixing rites and ceremonies, mythical narratives, cosmogonical traditions and liturgies, incantations, and prescriptions for religious functions. But through this maze of superstition I believe we can thread our way if we hold onto the clue which astronomy can furnish us."" ^The best example of this last is the period of 2,920 days, containing 8 solar years of 365 days each, and 5 Venus years of 584 days each, so elaborately set forth in pages 24 and 46-50 of the Dresden Codex. SeeBowditch, 19 10, pp. 63-68, 229; Forstemann, 1906, pp. 110-120, 182-196; Motley, 1915, pp. 31-32, 276-278; Spinden, 1917, pp. 109-112; and Seler, 1902-1908, vol. I, pp. 618-667. For an English translation of the last, see Seler, 1904a, pp. 355-391. ^Brinton, 1895, pp. 32, 33. This prediction is all the more remarkable in view of the state of knowledge on the subject when it was made. Goodman's important publication had not yet appeared, and the Maya chronological system as recorded in the inscriptions was but imperfectly comprehended. In Maya arch.-eology Brinton's contri- butions are for the most part fundamental and permanent. Thus his Maya Chronicles (18S2), published a genera- tion ago, in spite of many inaccuracies, contains the essentials for the entire reconstruction of Maya history. To an extraordinary degree, doubtless due as much to an innately philosophical mind as to his breadth of learn- ing, he seems to have possessed a faculty of prevision, of seeing in advance of exact proof, the broad general out- lines of the subjects with which he was dealing. This is certainly true of his work in the Maya field, and probably also of his work in the related field of Mexican linguistics. 36 THE INSCRIPTIONS AT COPAN. Indeed, decipherment has now proceeded to the point where it may possibly be doubted whether history as such was ever recorded on the monu- ments. This question, of course, must necessarily remain open until the last hieroglyph shall have been deciphered, but certainly the experience of the past and present indications of the future tend to strengthen rather than dispel such doubts. On the other hand, it seems incredible that any people could have erected so many monuments and inscribed them so laboriously without recording something of their history. Such a condition would be so unique in the annals of mankind that, if for no other reason than its isolation, it might well be doubted here. In fact, the writer still believes that some residuum of history will yet be found after all the calendric glyphs have been deciphered. If this should prove to be true, however, it is certain such historical data will be confined to very brief allusions to the most important events, perhaps by means of highly specialized signs standing for such general ideas as victory, defeat, conquest, war, pestilence, famine, birth, death, fruition of crops, trade, religious festivals, foundation of cities, and the accession of rulers, and more specifically for the names of particular towns, tribes, and individuals. Such historical data as these are clearly found in the allied field of the Aztec hieroglyphic writing, particularly as it was used in the manuscripts or codices. In these texts, by means of specialized characters having just such general meanings as the above, much aboriginal history is accurately recorded. To begin with, the signs for a large number of Aztec personal and place names have been identified. For example, the name-glyphs of the nine Aztec tlahtoiiani,^ or rulers, from Acamapichtli, 1376-1395 a. d, down to Moctezuma II, 1502-1520 a. d., are known, as well as the name-glyphs of many towns and cities in central Mexico.- By means of these characters and the signs of general meaning just noted, and with tlie help of a fairly accurate chronological system, the Aztec were able to record the principal events of their history with considerable precision." In figure i is shown the death of the eighth tlahtouani, Ahuitzotl, in the year loTochtli (1502 a. d.) and the succession of his nephew, Moctezuma II, to the throne, as recorded in several different Aztec manuscripts. Figure i, a, shows this event as set forth in the Codex de Tepechpan.'' In the center of the circle at the top is seen the head of a rabbit, tochtli, with 10 small dots around it. This stands for the year 10 Tochtli,^ or 1502 of the Christian Era. 'Literally, "he who speaks" or "the one who speaks" (Seler, 1904, pp. 146, 156). This was the Aztec name for ruler or king. The corresponding position among the Maya was called halach vinic, "real or true man" (Brinton, 1882, p. 128), both words implying within themselves supreme authority. ^The second part of the Codex Mendoza sets forth the tribute paid by different cities to the Aztec rulers, the signs for the different cities being followed by the amount of tribute paid by each. Penafiel, in his Nomhres geogrdficos de Mexico has collected the signs for many of these place-names. See Penafiel, 1885. 'The Aztec calendar did not permit of accuracy in dating beyond a period of 52 years, unless an unbroken series of the successive 52-year periods was preserved. In other words, any Aztec date could recur fulfilling all the necessary conditions imposed by their calendar, after a lapse of 52 years. However, if there are no lacunas the record remains accurate indefinitely, indeed until an omission occurs. 'See Codex (Mapa) de Tepechpan, plate 3. *The Aztec named their years after the days with which they began, always one of the following four: Acatl (reed), Tochtli (rabbit), Calli (house), and Tecpatl (flint). SCOPE OF THE PRESENT INVESTIGATION. 37 Attached to this sign are two human figures. The one to the left is swathed in bands and tied with rope to indicate that it is dead, a mummy. In all the Aztec manuscripts this mummy-bundle is the universal sign for death. The fact that the dead man had been the tlahtouani is shown by the xiuhtzontli or turquoise-studded crown that rests on the mummy's head. The xiuhtzontli or Aztec crown resembled somewhat the red crown of Lower Egypt in Pharaonic times, reversed, being high in front and low in back, with a tassel falling behind. It was worn only by the tlahtouani and is used in the manuscripts as a sign for royalty. Fig. I. — Death of the eighth tlahtouani (rulet) Ahuitzotl and succession ot Moctezuma 11 in the year lo Tochtli (;. e., lo Rabbit, 1502 a.d.) as represented in; a, Codex de Tepechpan; b, Codex Aubin 1576; d, tight half, Codex Telleriano-Remensis; e, right half. Codex Vaticanus 3738. Conquest of the town of Tecuantepec in the same year by Ahuitzotl as represented in: c. Codex Mendoza; Landa, 1881, p. 74. 'Ponce, 1872, p. 392. 'Lizana, 1893, p. 3. •Landa, op. cit., same page. SCOPE OF THE PRESENT INVESTIGATION. 43 Bishop CogoUudo, nearly a century later (1688), adds the following evidence: "In a city named Tixhualatun, which signifies 'place where one graven stone is placed upon another,' they say are their archives, where everybody had recourse for events of all kinds, as we do to Simancas."^ Finally, as late as i697,Villagutierre found historical records still being kept in the hieroglyphic writing by an independent Maya tribe, the Itza, of Peten: "Because their king (Canek) had read it in his analtehes [fiber- books or codices] they had knowledge of the provinces of Yucatan, and of the fact that their ancestors had formerly come from them; analtehes or histories being one and the same thing. "- Indeed, so far as the manuscripts are concerned, there can be little doubt, in the face of such direct evidence as the foregoing, that the Maya also recorded their history. When we come to examine the Maya codices extant, however, these statements are not substantiated. The Dresden Codex, for example, deals with the tonalamatl, and astronomical phenomena and calculations. The Codex Tro-Cortesianus is almost entirely given over to the record of tonal- amatls, and the Codex Peresianus partially so. There is a possibility, how- ever, that the last may have some small content of history, as a succession of katuns like the u kahlay katunoh in the Books of Chilan Balam clearly appears on one side of the manuscript.'^ II [Katun] 2 Ahau XIII [Katun] 13 Ahau XI [Katun] II Ahau IX [Katun] 9 Ahau VII [Katun] 7 Ahau V [Katun] S Ahau III [Katun] 3 Ahau I [Katun] I Ahau XII [Katun] 12 Ahau X [Katun] 10 Ahau VIII [Katun] 8 Ahau VI [Katun] 6 Ahau IV [Katun] 4 Ahau [Ilr [Katun 2 Ahau XIII [Katun 13 Ahau XI [Katun] XI Ahau There was fighting in the fortress of IVIayapan because of the seizure of the fortress and the for- tified town by the joint government in the city of Mayapan. The pestilence took place; the general death took place in the fortress. The small-pox broke out. Ahpulha died the sixth year. The count of the years was toward the East, [the month] Pop began on [the day] 4 Kan to the East .... 9 Imix was the day on which Ahpulha Napot Xiu died in the year of the Lord 158. The mighty men came from the East. They brought the sickness. They arrived for the first time in this country we Maya men say in the year 1513.^ 'CogoUudo, 1688, p. 186. ^Villagutierre, 1701, p. 353. 'Morley, 1915, pp. 33, 84, 79-S6 and Appendix VII, p. 576. ■■Matter inclosed in brackets, thus [ ], does not appear in the original. Here the "2 Ahau" preserves the con- tinuity of the sequence, however. 'This extract appears on the back of folio 41 of the Book of Chilan Balam of Chumayel. For a facsimile reproduction see Gordon, 1913, plate 76, and for a translation with notes, Brinton, 1882, pp. 155, 156, i6i, and 162. AA THE INSCRIPTIONS AT COPAN. These u kahlay katunob "or records of the katuns," were written approx- imately in 1575 to 1800 a. d. by native Maya in the Maya language but in the characters of the Spanish script, and at least one of them, the first chronicle of the Book of Chilan Balam of Chumayel,i gj^gg ^^j^ unbroken succession of the katuns, or 7,200-day periods, for more than 1,100 years, carrying back the chronological outline of Maya history to about 450 a. d. The successive katuns appear in a column at the left of each page, with the corresponding events, if any, written after each, only the more important events being recorded, as shown at the bottom of the preceding page. The close resemblance of this extract from the first u kahlay katunob in the Book of Chilan Balam of Chumayel to the Aztec historical codices, such as the Codex Mendoza, the Codex Telleriano-Remensis, and the Codex Aubin, 1576, for example, with their sequences of years, is so apparent as to require no further comment here, and it may be accepted as indicating that the u kahlay katunob in the Books of Chilan Balam were copied by natives in the Spanish script from older pre-Columbian historical manuscripts, which gave the outline and principal events of Maya history — in short, that Maya historical manuscripts formerly existed in spite of the fact that none of them have yet been discovered. Moreover, it should be borne in mind in this connection that many Maya manuscripts are known to have been destroyed by the Spanish priests in their efforts to stamp out the native religion, and the fact that only those of a non-historical char- acter have been preserved is doubtless due to accident rather than to any failure on the part of the Maya to have recorded their history.- Turning next to the Maya inscriptions, it must be admitted that their possible historical content is still an open question, although both Lehmann and Spinden (as well as the writer) are of the opinion that the as yet unde- ciphered glyphs will prove to contain some historical data. Indeed, occa- sionally the subjects portrayed on the monuments are themselves such as to lend color to the idea. Many of the stela? show bound captives with glyphs inscribed on their shoulders and thighs, or somewhere near them, and in such cases the conclusion is almost inevitable that these signs stand for the personal or place names of conquered rulers, tribes, or cities, as in the related Aztec codices. Lehmann says in this connection: "I feel no doubts that a number of the Maya reliefs and inscriptions are intended to commemorate historical events, particularly the scenes wherein a number of men in humble attitude, often loaded with chains, approach the Maya ruler. Each of these figures is apparently the chieftain of a conquered tribe, the name and origin of which are carefully denoted by a number of hieroglyphics."'^ 'See Brinton, 1882, pp. 152-157; and Gordon, 1913, plates 74-77, for a facsimile reproduction of this chronicle. ^Bishop Landa himself, to whom we otherwise owe so much, naively confesses to having burned a number of these manuscripts: "We found among them a great number of books in their letters, and because they had nothmg but supetstitions and lies of the devil, we burned them all, which they felt marvellously and gave them pain." (1881, p. 103.) In this conflagration, according to a letter written by the Yucatecan Jesuit, Domingo Rodri- guez, to a Sefior Estevez from Bologne on March 20, 1805, the following material was destroyed: 5,000 idols of different forms and dimensions; 13 large stones which served as altars; 22 small stones of various forms; 27 rolls of signs and hieroglyphics on deerskin; and 197 vases of all dimensions and shapes. See Molina Solis, 1897, p. 195. 'Lehmann, 1909, pp. 16, 17. SCOPE OF THE PRESENT INVESTIGATION. 45 Spinden has elaborated this hypothesis in a recent paper, arguing for the historical character of some of the reliefs: "Judging by the graven pictures many monuments of the southern Maya are memorials ot conquest. Captives bound with rope or held by the hair are actually represented in several instances. On a still greater number of monuments the principal personage stands upon a crouched or prostrate man devoid of all signs of rank and power. . . . Now it is obvious that the presence of vassals and overlords on the monuments increases the probability that actual historical events are being commemorated and that actual historical persons are being portrayed. . . . All the human figures in this tableau [Piedras Negras, Stela 12], including three victors and nine victims, have short incised inscriptions upon their bodies or near their heads. These inscriptions consist of two or more glyphs, and it seems reasonable to suppose that names of both persons and places are recorded."^ This evidence in itself is very convincing, but when it is coupled with the fact that historical data are known to be recorded in the related Aztec manuscripts, and stated by the early Spanish writers to have been recorded in the Maya manuscripts as well, and actually found in the Books of Chilan Balam, there can be little doubt that some historical data will yet be found in the Maya inscriptions, even though the most recent discoveries are not in this direction. Whether or not the Maya inscriptions contain such an historical re- siduum, however, is of little moment in the present connection, since, from what has already been said, there can be no doubt as to the fundamental part played by chronology in their records. The decipherment of the dates on the monuments at Copan, therefore, is the object of the present investigation. Upon the dates of the monuments hinges the solution of other important and related problems: the development of Maya art, the determination of Maya astronomical learning, the possible discovery of historical material, and, in fact, the very function of the monu- ments themselves. Indeed, it is safe to say in this connection that little or no progress can be made in understanding the true nature of the Maya monuments and their inscriptions until their dates shall have been deciphered. 'Spinden, 1916a, pp. 442,443. Stela 12, at Piedras Negras, mentioned above, is one of the best examples of these possible "historical monuments." See Maler, 1901, plate 21. The subject portrayed is a ruler seated upon a throne with an attendant standing on either side. Seated cross-legged below the throne, and between the two attendants, are nine captives bound with ropes, their ear-plugs removed, no head-dresses or clothing, and an unmis- takable expression of distress on their faces. The scene would appear to be that of a conqueror and his captives. It is interesting to note that of these twelve figures the ruler, his two attendants, and five of the captives have the same glyph either inscribed on them or near them, namely, the familiar bat head or Zotz-sign with a knot prefix. In other words, two-thirds have the same character attached to them. Lehmann (1909, p. 17, note l) suggests this may have something to do with the Maya bat tribe, the Tzotzil (Maya /zo/z,bat). Spinden (/ii'rf.) believes this glyph may have the general meaning "here follows a name." The writer thinks it more likely that this bat glyph is a general sign for "conquest" or "conquered." The other glyph or glyphs would then indicate the name of the figure in each case, the picture itself showing the relation of the figure to the central idea, i. e., whether as conquering or being conquered. Lintel 2 at Piedras Negras has a similar subject. See Maler, 1901, plate 31. Here an elaborately dressed ruler, with spear in hand, faces six kneeling figures, though these are not bound. Behind the ruler stands a single attend- ant. Again, four of the six captives (?) have the Zotz glyph standing above them. Although it is too early to attempt to speak finally as to the meaning of this glyph, the accompanying pictures are such as to indicate that it may well have had the general meaning suggested by the writer above. The strong probability, as Spinden notes above, that portraiture occurs on the monuments, also tends to confirm their historical character, for the reason that if the figures portrayed are particular individuals, the accompanying inscriptions would presumably record their activities. 46 THE INSCRIPTIONS AT COPAN. It is highly probable, moreover, that the exact dating of the monuments will throw light on the meanings of the as yet undeciphered glyphs. This is certainly true of the Supplementary Series, a group of 7 or 8 signs normally standing between the two parts of the Initial Series terminal date, which, as the writer has shown elsewhere,^ depend for the very fact of their existence upon the corresponding Initial Series. Indeed, so far as the Supplementary Series are concerned, it is necessary to know the date recorded by the accom- panying Initial Series before interpretation can even be attempted. Again, in the case of the Secondary Series," if we are to determine the nature of the phenomena governing this important count, which ranges from one day to more than one million years — that is, whether it records astro- nomical, mythological, traditional, or historical data, it will first be neces- sary to know the dates on which the governing phenomena occurred. The record of these three counts, the Initial, Supplementary, and Secondary Series, the first and third solar, the second lunar, comprises approximately one-half of the Maya inscriptions, and enough has already been said concerning them to show their intimate connection with, and dependence upon, the counting of time. To the ultimate solution of these and other related problems, therefore, not only in this archaeological area, but also in the much broader field of contemporary ancient America, an accurate knowledge of Maya chronology is indispensable; and in the present volume this particular phase of the inscriptions at one of the largest Maya cities has been exhaustively reviewed. METHOD OF TREATMENT. The immediate object of this research has determined to a large extent the method of treatment followed in describing the different monuments. The Old Maya Empire, approximately the period covered by the Copan inscriptions (between three and four centuries), has been divided into three smaller periods: the Early, the Middle, and the Great;'* and under these headings the individual monuments have been described according to their relative chronological positions in each subdivision, the inscriptions of the Early Period being found in Chapter II, those of the Middle Period in Chap- ter III, and those of the Great Period in Chapter IV. The general discussion of each monument is preceded by a synoptic outline giving its most essential features: name, provenance, date, reproduc- tions of the text, and references. By this standardized treatment, which is followed throughout the book, the same important points about every monu- ment are given in a synopsis at the head of each, so that it is not necessary to read all the accompanying description in order to arrive at the most essential 'Morley, 1916. The general nature of this count, dealing with the moon, is described briefly in Appendix VI. ^Secondary Series is the name which has been applied to glyphs recording dates other than Initial Series dates. They consist of a number and date and are usually counted from the Initial Series or from some date which may be referred back to the Initial Series. For an explanation of this count, see Morley, 1915, pp. 74-76. 'For the further discussion of this question see pp. 53, 54. MORLEY. PLATE 6 Plan of main structure, showing the location of the associated monuments. METHOD OF TREATMENT. 47 points. It is thought this method of treatment will make the information available here more accessible, and will, at the same time, facilitate use of this book as a chronological concordance of the Copan inscriptions. An example of one of these synoptic headings follows: Stela A. Provenance: In the Great Plaza just north of Mound 4, Main Structure. (See plate 6.) Date: 9.15.0.0.0 4 Ahau 13 Yax. Text, (a) photograph: Maudslay, 1889-1902, vol. i, plates 25, 27-29. Spinden, 1913, plate 20, 7 (part only). (b) drawing: Maudslay, ibid., plates 26, 30. Morley, 1915, plate 7, b. Stephens, 1841, vol. I, 3 plates after p. 158. References: Bowditch, 1910, pp. loi, 126, 127, 182, 183, and tables 29 and 31. Galindo, 1834, Appendix XI, p. 598. Goodman, 1897, p. 129. Gordon, 1896, p. 35. Gordon, 1902, p. 171. Maudslay, 1889-1902, vol. i of text, pp. 36-39. Morley, 1915, pp. 169, 170. Seler, 1902-1908, vol. i, pp. 754,755. Spinden, 1913, pp. 158, 159, 162, and table i. Stephens, 1841, vol. i, p. 158. Thomas, 1900, pp. 776,801. Under the first heading are given the location of the monument and the corresponding map in this volume where this is shown. Under the second heading is given the date of the monument expressed in terms of the Maya chronological system. The method followed here in transcribing Maya dates into the terms of our own Arabic notation is that first used by Bowditch.^ The largest time-period usually present in a Maya date, the cycle, is written first, i. e., to the left; next come the katuns; next the tuns; next the uinals; and last the kins, each being separated from the next by a dot, thus, 9.15 .0.0.0, the whole number being read : 9 cycles, 1 5 katuns, o tuns, o uinals, and o kins. Immediately following this is the terminal date (4 Ahau 13 Yax in the present example), reached by counting this period forward from the starting-point of Maya chronology. The whole record 9. 15.0.0.0 4 Ahau 13 Yax, therefore, means that, if 9.15.0.0.0 be counted forward from the starting-point of Maya chronology, the day reached will be 4 Ahau, the 14th day of the month Yax, written as 13 Yax in the Maya notation.^ Maya dates are therefore simply records of periods of elapsed time. It is doubtful, indeed, whether the current day, as such, ever was recorded. In the present example the 9 cycles, 15 katuns, o tuns, o uinals, and o kins refer to past time, and the closing day of this period, 4 Ahau 13 Yax, was probably already past when it was counted. We have an identical practice in describing the time of day, that is, in counting hours, minutes, and seconds. When we say it is 2 o'clock in the afternoon, in reality the second hour after noon has passed, and the third hour is about to commence. In other 'Bowditch, 1901, p. i; also Bowditch, 1910, p. 38, note i. -Morley, 1915, pp. 46-48. 48 THE INSCRIPTIONS AT COPAN. words, when we count the time of day we refer to elapsed, not current, time. This same method is used in reckoning astronomical time. During the passage of the first hour after midnight the hours are said to be zero, the time being counted by the number of minutes and seconds elapsed. Thus half past 12 is written q*"' 30™"'' o''"-. Indeed, in this method i hour can not be written until the first hour after midnight is completed or until it is I o'clock, namely, i"" o"'" o^^" .^ And so it was with the Maya. The time periods recorded refer to elapsed, not current, time, and since the Maya did not subdivide the day (at least periods smaller than the day have not yet been found in the records which have come down to us), the day is the basic unit of their count. In speaking of a date as "in Cycle 9" in this memoir, what is really meant is that such a date is in the tenth cycle. However, in order to preserve the association of the Maya numerals actually recorded, it will be under- stood that dates thus described occurred in the period following the numeral actually recorded; as Katun 15 for a date in the sixteenth katun or Katun 18 for a date in the nineteenth katun. The basic unit of the Maya Calendar then was the day, and Maya dates were recorded by stating how many days, expressed as so many cycles of 144,000 days each, so many katuns of 7,200 days each, so many tuns of 360 days each, so many uinals of 20 days each, and so many kins (odd number of days under 20) had elapsed since the starting-point of their chronology to reach the date recorded. This method of dating is identical with the use of the Julian day by modern astronomers and chronologists, the corresponding Julian day of any date giving the total number of days which have elapsed from the starting-point of the Julian Period, 4713 b. c, to the given date. In recording dates in our own Christian chronology we follow a similar, though not an identical, practice. In writing 1916 A. d., Sunday, January i, for example, we understand that I period of a thousand years, 9 periods of a hundred years, i period of ten years, and 6 periods of one year have elapsed since the birth of Christ — the starting-point of our chronology — to reach the current day, which is Sunday, the first day of the month of January.. In this latter case, however, the basic unit of the count is the year, not the day, as in the Julian and Mayan Periods. Indeed, in our own method of writing dates there is no direct record of the fractional parts of a year, i. e., the number of days in the new year to reach the date recorded; and this information has to be calculated from the month date given. For this reason, as often claimed, the Maya kept a more convenient record of the total number of elapsed days since the beginning of their chronology than we can by our system, complicated as the latter is by the bissextile element. The Bowditch method of transcribing Maya dates is the only one now in use, having entirely replaced the clumsy system devised by Goodman or 'Morley, 1915, pp. 46-48. MORLEY I I 1 I PLATE 7 I I I F C G,G, B HIE a. The Great Plaza looking east from the northwest corner. I I I 32E L 4B C DHF b. The Great Plaza looking northwest from the summit of Mound 26. Courtfsy of the Peabody MiiivAin METHOD OF TREATMENT. 49 the Still clumsier expedient of Forstemann and Seler of reducing the dates to their corresponding day totals in our own Arabic notation. In comparing Maya dates throughout this work the terminal dates have been omitted; for example, 9.15.0.0.0 is used for 9. 15.0.0.0 4 Ahau 13 Yax, and 9. 16. o. o. o for 9. 16. o. o. o 2 Ahau 13 Tzec. This omission, how- ever, in no way affects the values of the dates thus abbreviated, and has been introduced only to simplify and expedite the comparisons. When monuments of the same date are known at other sites the fact is noted and reference made to Appendix VIII, where their location will be found. It has appeared advisable not to give the corresponding equivalents in Christian chronology in the general text, first because of the widely differ- ent results reached in the several systems of correlation which have been proposed, and second because eventually it is hoped that it will be possible to correlate the two chronologies to a day by means of astronomical data present in the inscriptions. The writer believes, however, that it is now possible to fix Maya dates to their corresponding positions in the Christian Era with a maximum error of less than 5 years and probably of less than i year. However, because of even this slight uncertainty, it has seemed best to reserve the presentation of the entire correlation question, as well as the table of equivalents suggested by the writer, for treatment in an appendix. (See Appendix II.) Under the third heading in each synopsis will be found plate and figure references to specific publications where the monument under discussion has been reproduced; also whether the reproduction is from a photograph or drawing, or from both. Under the fourth and last heading are given page references to the several authorities who have described the monument. Then follows the general discussion of the text. The latter in each case opens with the size of the monument and a brief description of its principal characteristics — how the inscription is presented, whether upon one or more sides. Then comes the detailed consideration of the text, the decipherment of its several dates, and finally a summary of the dates. At the end of each chapter, the chronological data have been briefly analyzed, particularly with reference to the growth and expansion of the tribe or people whose capital was Copan, during the corresponding period. In designating individual glyph-blocks in a text, the method followed is that devised by Bowditch, namely, a set of two coordinates, the vertical rows or columns being given letters from left to right, thus: A, B, C, D, etc., and the horizontal rows, numbers from top to bottom, thus: i, 2, 3, 4, etc. By means of these two sets of coordinates, any glyph-block can be simply and clearly designated. For example, in figure 5, glyph a would be described as ai, glyph /3 as B3, and glyph 7 as b6. This simple method of glyph designation will 50 THE INSCRIPTIONS AT COPAN. — 1 i ! j 1 1 1 P r 1 Fig. 5. — Method of designating individual glyph-blocks. be found to apply in the great majority of cases, but in a few texts where the arrangement is irregular, other methods have to be used. The word hieroglyphic has been abbreviated to glyph throughout the present work in referring to signs ex- pressing single ideas, such as the cycle-sign and its co- efficient, the katun-sign and its coefficient, a day-sign and its coefficient, for example. The term glyph-block, as used hereinafter, refers to what Maudslay calls a glyph — that is, a rectangular block or character composed of one or more basic elements. Thus, in many glyph- blocks there are two ideas expressed, one recorded in the left half of the block, designated here as h\a or Bia, and the other in the right half of the block designated here as K\h or bi^. In some cases there are even four glyphs in a single glyph-block; in such cases the four glyphs are designated as K\a u. h. (upper half), ai^ u. h. (upper half), Aifl 1. h. (lower half), and ai^ 1. h. (lower half), re- spectively. When the presentation is so irregular as to lie without theoperationof these simple rules, Maudslay's numerical designations have been followed. The Maya monuments have been usually classified into two groups according to their shape and size: stelae and altars. As used in Maya archae- ology, the word stela (plural, stelae) refers to a monolith, of columnar, shaft, or slab-like shape, usually over 2 meters in height. As a rule, these are inscribed with glyphs and human or grotesque figures, though either or both may be wanting. When both are wanting — /. e., when a stela is devoid of sculptural decoration — we may probably assume that it was originally painted instead of carved. As used in classical archaeology, the word stele or stela usually has a more restricted application. A stele specifically refers to painted or carved slabs or pillars erected over Greek graves, or to milestones near towns, and more generally to inscribed stones in public places, the last more closely paralleling its use in Maya archaeology. Although Maya stelae vary greatly in size and shape, the highest being over 10 meters and the lowest less than i meter in height, it is necessary to group them all together as opposed to the other general class of Maya mon- uments, the so-called "altars." The latter are much smaller and less uni- form in shape than the stelae, and, indeed, appear to have had some function subordinate to the latter, perhaps as true altars — i. e., "places for offering sacrifice." A fundamental difference in function also probably existed between the two groups at first. As the writer has shown elsewhere,^ the stelae are probably time-markers, erected perhaps primarily to mark the passage of 'See Appendix VII and Morley, igiyi, pp. 195-201. METHOD OF TREATMENT. 5 I time. Throughout the greater part of the Old Empire they were set up at intervals of i,8oo days, in the different cities, and may perhaps be likened to 5-year almanacs, setting forth the principal astronomical or historical events of the preceding 5-year period. The altars, on the other hand, rarely appear to have been thus used, except toward the end of the Great Period. They are usually associated with stelae, standing in front of them, although occasionally independent altars are found. As suggested above, it is possible that they are altars in the truest sense of the word — places where sacrifices were offered in front of and to the stelae. Inscriptions are also found on architectural members, such as stairways, steps, door-jambs, lintels, cornices, wall-panels (both exterior and interior), and columns, the two types of monuments described, however, being suffi- ciently elastic to include all the detached inscribed stones. In selecting the illustrations for this work, the principal object the writer has kept in view has been to use chiefly unpublished material; that is, to figure such texts as are accessible nowhere else. With the illustrations given in this work, and those already published by Maudslay and by the Peabody Museum, reproductions of practically all the Copan inscriptions are now accessible — certainly all the chronological portions — with but one notable exception, namely, the Hieroglyphic Stairway of Mound 26. This lengthy text, the longest known in the Corpus Inscriptionum Mayarum, is unfortunately in too fragmentary a condition to permit anything approaching its complete reassemblage. An analysis of the chronological parts, however, has been attempted in Chapter IV, pages 237-274. The urgency of placing on record reproductions of all the Copan inscrip- tions is pressing. Several of the monuments have been destroyed in the past decade, indeed within the past 4 years, ^ and further irreparable losses may occur at any time. With the publication of the present volume, how- ever, a part at least of every text now known at Copan will be accessible to the student, and in most cases reproductions of all the chronological glyphs will have been published. In many cases photographs only are figured; in more, the reproductions are from drawings of the originals; and in a few, both drawings and photo- graphs of the same text are given. Some of the photographs used have been loaned by the Peabody Museum, for which the writer wishes to express his thanks. The remaining photographs, except that of the painting of Copan by Vierra (plate 33), were taken by the Carnegie Institution Central American Expeditions of 191 5 and 1916. The Vierra painting is reproduced through the courtesy of the Museum of San Diego, San Diego, California, where the original is on exhibition. 'In 1912 Stelae 8 and 9 were broken up for use in the foundations of a wall then being built around the village cemetery, and in 1916 Altars L' and M' were smashed into small pieces for use in the foundations of the new village church. ^2 THE INSCRIPTIONS AT COPAN. All drawings of inscriptions in this memoir were made by the writer, in the great majority of cases directly from the originals, but two or three being drawn from casts or photographs in the Peabody Museum. How far the writer has succeeded in eliminating his personal equation in these drawings is difficult to say, although every glyph was measured and drawn to scale, i. e., not sketched free-hand. This point has already received some attention in connection with the work of Miss Hunter and Mrs. Tretheway, the delineators of the Maudslay and Bowditch publications, respectively, and a closing word may be added here. In such complex compositions as the Maya glyphs, the element of selection, the quality of line employed, the method of rendering the depth of the relief, in some cases as much as 15 cm., and of showing the effaced portions, to say nothing of the debatable ground of restoring partially effaced glyphs, all make for considerable individual variation, even in copies of the same original. Moreover, the texts themselves differ greatly one from another, due in part to the different periods from which they date, and in part to the varying skill of the sculptors by whom they were severally exe- cuted. In view of all these complicating factors, therefore, it is not sur- prising to find slight dissimilarities in different representations of the same glyph. Such differences, however, are usually unimportant. They do not interfere with the accurate representation of the details upon which the glyphs depend for their meanings; and they are of moment only when it becomes necessary to rely upon the stylistic criteria for dating. In such cases, however, it is imperative to have the spirit of the original preserved so far as possible (character of line, detail of decoration, and depth of relief), for it is only by means of such secondary criteria that even approximate dating can then be attempted. In closing this chapter it should be pointed out that all direct quotations from French, German, Spanish, and Mayan authorities, which are especially numerous in the appendices, have been translated into English in order to make them more serviceable to the general reader. In all such cases, how- ever, footnotes indicate where the original passages may be consulted. CHAPTER II. THE INSCRIPTIONS OF THE EARLY PERIOD. Any attempt to divide a series of monuments, arranged according to their relative positions in a stylistic sequence, into chronological periods must necessarily be more or less arbitrary. The more homogeneous and consistent the stylistic development, the more arbitrary (and in a sense the more unsatisfactory) are the resulting chronological subdivisions. This is particularly true of such an art sequence as that at Copan, where sculpture in stone was gradually and consistently developed for more than three centuries. Spinden in his earlier classification of the Copan sculptures^ regarded everything after 9.0.0.0.0 and prior to 9.15.0.0.0 as belonging to the Early Period, and after 9.15.0.0.0 as belonging to the Great Period, eliminating altogether a middle or intermediate period. But such a classifi- cation, while sufficiently accurate for purposes of preliminary investigation, was found to break down upon closer study, especially at other sites. It necessitated placing within the limits of a single period sculptures of very considerable stylistic dissimilarity — sculptures, indeed, as technically and esthetically different as the Apollo of Tenea (circa 550 b. c.) and the Chario- teer of Delphi (circa 470 b. c). Indeed, for purposes of close chronological description it was found necessary to recognize a Middle or intermediate period between the Early and Great Periods, the limits of which are fixed at one end by the first appearance of sustained improvement in technical processes, treatment, carving, depth of relief, and the like, and at the other end by the final disappearance of archaism. Spinden, in another passage of the same work, virtually reached a similar conclusion himself: "The chronology of Copan may be summed up as follows: The earliest mon- uments are very crude and archaic, particularly in regard to the carving of the human face. A steady improvement is noted, extending from the ninth [Katun 9] to the fifteenth katun [Katun 15]. By the beginning of the fifteenth katun almost the last trace of archaic treatment had vanished. The brilliant period lasted until the middle of the sixteenth katun [Katun 16] and possibly somewhat longer."^ The only real difference between this classification and that suggested by the writer is that in the former the Early Period is made to end in 9.9.0.0.0, while in the latter it is extended to 9.10.0.0.0, another 20 years. On this latter date the first half of Cycle 9 came to an end, which, being a round number in Maya chronology, is a more appropriate as well as con- venient point at which to close a general stylistic period than the preceding 'Spinden 1911, table 2. -Ibid., 1913, p. 165. S3 54 THE INSCRIPTIONS AT COPAN. katun. Indeed, in a recent article^ Spinden has accepted this threefold division of the Old Empire as suggested by the writer, not only for chrono- logical purposes, but also for stylistic classification as well. Giving due weight therefore to both lines of evidence, stylistic as well as chronologic, it has seemed best to subdivide the Old Empire, the inscrip- tions of which at Copan form the subject of the present investigation, into three periods as follows: The Early Period. From the earliest times down to 9.10.0.0.0. The Middle Period. From 9.10.0.0.0 to 9.15.0.0.0. The Great Period. From 9.15.0.0.0 to 10.2.0.0.0.- Having defined the limits of the several periods as hereinafter used, let us next ascertain what monuments at Copan may be referred to the Early Period. As already stated, the inscriptions at this city are more numerous than those of any other Maya site, and although they date from every period of its occupation as would naturally be expected, they are especially numerous from the Early Period. Indeed, in this latter respect Copan is surpassed only by Tikal, the great northern metropolis of the Old Empire, and prob- ably her most powerful rival (plate i). So archaic and fragmentary are some of these earliest texts at Copan that it has been impossible to decipher them exactly and assign them to their proper positions in Maya chronology. That they precede the earliest surely dated monuments there, however, is clear from the earlier character of their glyphs. These are carved in very low relief, in some cases being little more than incised. The technique is crude and the style undeveloped. They present, moreover, certain technical similarities with the most archaic inscriptions at Tikal. which are apparently of about the same period, notably in the irregularity of the outlines of the individual glyph-blocks, as shown in Stela 20 for example (plate (),b and figures 9 and 10), and in the omission of ornamental elements in the bar and dot numerals for i, 6, 11, and 16, as shown in Altars X and Y (plate 8, c, ai, and plate 8, b, ai), and Stela 17 (plate 11, a, B3). Finally, pronounced complexity and elaboration in glyphic details are also characteristic of the sculptures of the Early Period, as shown in the extensive use of parallel lines in Stela 24 (figure 13), Stela 15 (plate 12), Stela 7 (plate 13), and Stela E (plate 14, a-c) for example. Toward the close of the Early Period at Copan notable advances were made in the art of sculpture. The portrayal of the human figure was attempted, and although the earliest efi^orts in this direction are somewhat lifeless, as in the case of Stelae 18, 7, E, and P, for example, they clearly forecast the sculptural brilliance which was to follow a hundred years later. 'Spinden, 1917, pp. 130-132. -The several periods of Maya history are given in Appendix II (see p. 50;), and their presentation will not be anticipated here, except to note that there were two general divisions: the Old Empire extending from the earliest times to 10.2.0.0.0 and the New Empire from 9.14.0.0.0 to the Spanish Conquest in 1541. As Copan was probably abandoned before 10.2.0.0.0, only rhe Old Empire and its subdivisions concern us in rhe present connection. INSCRIPTIONS OF THE EARLY PERIOD. 55 There are 22 monuments' now known which may be assigned to the Early Period at Copan; and this number will doubtless be increased by further excavation. In the very nature of the case, the earliest monuments are the most deeply buried and are also those which suffered secondary usage the most extensively. Already in ancient times the practice of reusing earlier monuments in later constructions was prevalent. This seems to have been begun as early as Katun 9, i. e., at the end of the Early Period, when part of Stela 24 was reused in the foundations of Stela 7, and was common throughout the Middle and Great Periods. It is particularly true in the case of the so-called banded altars, the only type of altar yet found, in the Early Period. These were repeatedly reused in the foundations of later monuments, as, for example, Altars J' and K' in the foundations of Stela 10, Altar X in the foundations of Stela 5, Altar Y in the foundations of Stela 4, and Altar A' in the Hiero- glyphic Stairway of Mound 26. This practice, however, was not confined to the smaller monuments alone. Thus, for example, in addition to the case of Stela 24 just noted, Gordon reports'- that Owens found Stela 9 had been reused in the foundations of Stela 8, the former being a very early monument (9.6.10.0.0), and the latter a very late one (9. 17. 12.6. 2). Again, the archaic stela, No. 17, found by Morris in 191 2, had been reshaped into a building- block in such a way that part of the original design has disappeared. (See plate II, a and b.) And in 1916, during the demolition of the mound at the southeastern corner of the village plaza for building material, part of the top of an archaic stela — No. 21 — was found, which had been made over into a building-block in ancient times, all of the inscription being destroyed save only parts of three Initial Series introducing glyphs. (See figure 14.) During the excavation of Mound 9 at the Main Structure, a small slab of stone, Fragment S', which had glyphs on the under side, was uncovered in the pavement on the eastern slope. It had obviously been reshaped for this secondary purpose and part of the inscription was missing. (See figure 21.) Doubtless complete excavation of the site would bring to light other building-blocks showing similar secondary usage. As the city grew, the earliest monuments probably passed out of fashion — became obsolete as it were — and thus, having outlasted the'purposes for which they were originally designed, and being in every case exceptionally good blocks of stone, they were occasionally reused in later constructions. The earliest monuments at Copan are Altars J', K', L', M', P', and Q', and Stels 20, 22, 24, and 25. Unfortunately, none but the next to last (Stela 24) has been surely dated; and two. Altars J' and K', do not have any glyphs at all. Of these earliest altars, all but one, P', have the same design, 'This does not include the altar of Stela E, since, as will appear later (pp. 109-1 14), its inscription is only a continuation of the text presented on the stela with which it is associated, and it is therefore to be considered as an integral part of that monument. The above total also takes no account of the several Fragments V, and Fragment S', since the former almost certainly belong to one or other of the several fragmentary steke and altars, which were foiind in their immediate vicinity, and which are already included in the above totah A list of the monuments which may be referred to the Early Period, with their provenance, is given in Appendix IX. ^Gordon, 1896, p. 38. 56 THE INSCRIPTIONS AT COPAN. a large, grotesque serpent's head, and on stylistic grounds all of these monu- ments, both stelae and altars, may be referred with certainty to the earliest group of sculptures now extant at the site. Altar J'. Provenance: Original position unknown. Found in the foundations of Stela 10 (Group 12), and now fallen several hun- dred meters down the slope to the east of this monu- ment. ( See plate 3.) Date: 9.0.0.0.0 to 9.5.0.0.0. Text, photograph: plate 8, d. Altar J' is 90 cm. long, 42 cm. wide, and 20 to 22 cm. thick. The top, left end, and front are sculptured, and the back is dressed but has no carving. The right end is rough, as if there were a piece broken off here. The bottom is plain. Two bands, crossing each other at right angles in the upper left- hand corner of the top, extend down over the sculptured end and front. The grotesque serpent's head on the top also extends down over the front. The relief is very low, the carving being little more than outlined. The execution is crude and the design is simple. There are no glyphs on the frag- ment preserved and it is therefore impossible to date this monument exactly. Concerning its relative age, however, we are not entirely in the dark; since the date of Stela 10, under which it was found, is surely 9. 10. 19. 13.0; Altar J' must therefore be older than this. On stylistic grounds it may probably be referred to the early part of Cycle 9, perhaps to the first four or five katuns. Altar K'. Provenance: Original position unknown. Found in the foundations of Stela 10 (Group 12), and now fallen several hun- dred meters down the slope east of this monument with Altar J'. (See plate 3.) Date: 9.0.0.0.0 to 9.5.0.0.0. Text, photograph: plate 8, e. Altar K' is in every way the sister piece of Altar J', and may even be a part of the same monument. It is 58 cm. long, 42 cm. wide, and 20 to 22 cm. thick, the last two measurements being identical with the corresponding dimensions of Altar J'. As both J' and K' apparently have their right ends missing, it is not improbable that originally both may have been of the same length. The design is the same in each and is similarly presented. The top, one end, and front are sculptured, the back being dressed but having no carving. The remaining end is rough and shows a broken edge. The bottom is plain. Two bands crossing each other in the upper right-hand corner of the top extend down over one end and the front. There is also a grotesque serpent's head on the top, crudely executed in the same low relief as that on Altar J'. On the sculptured end of Altar K' there is what appears to be the lower part of the day-sign Ahau |l((o))j)- Unfortunately the upper portion is missing and with it the v^^^y coefficient, if such were ever present. Because it was found in the same place and because of its close stylistic INSCRIPTIONS OF THE EARLY PERIOD. 57 similarity, it has been assigned to the same period as Altar J', namely, to the first four or five katuns of Cycle 9. Altar L'. Provenance: Original position unknown, probably from just west of the pyramid at the southeastern corner of the village plaza (Group 9). Found built into the altar of the village church, when it was torn down in March 191 5. Destroyed in 1916. (See plate 3 and figure 22, g and h.) Date: 9.4.0.0.0 to 9.6.0.0.0. Text, photograph: plate 8,/. Altar L' is 1.7 meters long, 89 cm. wide, and 28 cm. thick. Only the top and front are sculptured. The top is divided by two vertical bands, one on the left, the other in the center. At the left of the central band is a panel of 6 glyph-blocks arranged in two vertical columns of 3 each (see plate 8,/); and at the right is a large, grotesque serpent's head, similar to those on Altars ]' and K', except that the one here has a human figure in its widely opened mouth. The front shows a continuation of the central vertical band, the space to the left again being filled with glyphs — 4 double blocks — and that to the right with the familar twisted rope pattern '^^^^^^>., . The back and ends are dressed but not carved. The bottom is plain. Although the glyphs are well preserved, none are decipherable. The first, ai, has the coefiicient 3, but the accompanying sign is indeterminate. Its sufiix appears to be the moon-sign. b2 may be 9 cycles p O Q C ^ ; the coefficient is clearly 9, but the sign to which it is attached is ((^^(?^^^oo indistinct to identify with certainty. That the end of Cycle e ^^UJJg is recorded here, how- ever, seems improbable, since the day on which this period ended, 8 Ahau 13 Ceh, appears nowhere in the text.^ The remaining glyphs of the inscription both on the top and front are of unknown meaning. The style of the carving, although still crude and in low relief, is some- what better than that on Altars J' and K'. It closely resembles Altar Q' both in carving and arrangement of the design; and since the latter has been pretty definitely dated as 9.4.10.0.0, this monument has been referred to the same general period, i. e., Katuns 4 to 6. During the writer's last visit to Copan, additional evidence was secured as to the original provenance of both Altars L' and M'. Of the three oldest inhabitants of the village in June 1919, Christina Ramirez, Pio Garin, and Maria Melendrez, all born between 1840 and 1850, two agreed that these two altars originally came from somewhere south of the church. Maria Melendrez believed they had been in the yard of a rancho which had formerly stood just south of the old church (figure 22, C), while Pio Garin stated that as a child he remembered them as just south of the large plain stela in front of the mound at the southeastern corner of the village plaza, i. e., some 75 meters farther south. (See figure 22, g and i.) 'As will appear later (pp. 8S, 89) this important date is probably recorded on .Stela 15 and certainly as one of the two Initial Series on Sttla 3 (pp. 157, 158.) ^8 THE INSCRIPTIONS AT COPAN. Jacobo Madrid, one of the most intelligent of the middle-aged villagers (born in 1875) is inclined to accept Pio Garin's story rather than that of Maria Melendrez on the grounds that as a child he was a frequent visitor to this rancho south of the church, and he believes he would have remembered these altars had they been there at that time. Christina Ramirez, the oldest inhabitant of the village, has no remembrance of them. The recollections of these three old people go back clearly to a time when the dense bush, which formerly covered the whole valley-floor, came right up to the present village plaza, and when the village itself was still only a small cluster of huts, not more than a dozen scattered through the forest, with small clearings here and there for each rancho. This was before the forest had been felled in the valley by the party of colonists from Guate- mala between i860 and 1870, mentioned in Chapter I. Their testimony on this, as well as on other points to be treated later, may be accepted as correct. In the present case the story of Pio Garin has been followed, being corroborated to a certain extent by that of Jacobo Madrid; but it should be noted that the account of Maria Melendrez differs only by a few meters and both agree as to the part of the village from which these altars originally came. (See figure 22, g and i.) In 1892, when the first church was built, they were removed thither and let into the high altar, where they remained for 23 years. (See figure 22, h and /.) The writer first saw them in March 191 5, after this building had been torn down to make room for the new church then in course of construc- tion, at which time the photographs shown in plate 8, / and g were taken. On returning to the village a year later, he found that a few weeks previous- ly both had been broken into small pieces by a mason from Santa Rosa for use in the walls of the new church and no trace of either was to be found. Altar M'. Provenance: Original position unknown, probably from just west of pyramid at southeast corner of village plaza (Group 9). Found built into altar of village church when it was torn down in March 1915. Destroyed in 1916. (See plate 3 and figure 22, i and ;'.) Date: 9.4.0.0.0 to 9.6.0.0.0 Text, photograph: plate 8, g. Altar M' is 1.3 meters long, 61 cm. wide, and 29 cm. thick, and appears in every way to be the companion piece of Altar L'. The top and front are sculptured with the same design as Altar L', and the top is similarly divided by two vertical bands. To the left of the right band is a panel of six glyph- blocks arranged in two columns of three glyph-blocks each; and to the right is the same grotesque serpent head as in the corresponding position on Altar L', with the same human figure in its mouth. To the left of the vertical band at the left, the stone is broken, though traces of a sculptured design appear at the edge. The vertical bands on top continue down over the front and divide it into three fields; the center has a panel of glyph-blocks, while the right has INSCRIPTIONS OF THE EARLY PERIOD. 59 the same twisted-rope pattern as in the corresponding position on Altar L'. The back is dressed but not carved. Both ends show fractured surfaces, and it is evident from the design on the top that a piece is missing from each, in which last particular Altar M' differs from Altar L'. The glyphs are again well preserved, but unfortunately are undecipher- able. The first, ai, appears to be the cycle-sign preceded by 3 and sur- mounted by II. O j . 5^ -m ^^^ meaning is unknown. The omission of ornamental side el- ©^^^^^ ements in the number 1 1 should be noted. As already mentioned, @vsJi'^=^J this is characteristic of the Early Period;^ A3 may be the kin-sign, although this identification is doubtful. The remaining glyphs are unknown. As already explained, the style of carving and the subject-matter are identical with those of Altar L' and very similar to those of Altar Q', for which reason Altar M' has been assigned to the same general period, i. e., Katuns 4 to 6 of Cycle 9. The left-hand ends of both L' and M' are prob- ably missing, since their original designs would appear to have been like that of Altar Q': a pair of grotesque serpent-heads flanking a central glyph panel. The writer has been unable to find any previous reference either to this altar or to its companion-piece, Altar L'. As noted in the description of the latter, when he was in Copan in March 1915, the village church had just been torn down, and these two monuments had been removed from the high altar, but before he returned the following year both had been destroyed. Although the exact dates of Altars J', K', L', and M' can not be deter- mined, their relative ages may be accepted as established above. Theie is very little doubt on stylistic grounds, for example, that Altars L' and M' are later than J' and K'. The technique of L' and M' is a little more advanced, the style a little more developed, and the subject-matter a little more elaborate. These differences, although slight in themselves, in the aggregate indicate a corresponding advance in sculpture and warrant the relative chronological sequence suggested above. Altar Q'. Provenance: Found on the mound of Stela 7 (Group 9). Now part of a wall behind the house of Domingo Hernandez in the southwestern quarter of the village. (See plate 3 and figure 22, c.) Date: 9.4.10.0.0 12 Ahau 8 Mol (.?).- Text, {a) photograph: plate 24, /. {b) drawing: figure 6. In 191 2 Spinden found four sculptured fragments of archaic monuments behind the house of Domingo Hernandez, in the southwestern quarter of the iJn some provincial cities this practice obtained down to the latest times. For example, at La Q- Honradez, in northern Guatemala (see plate i), as late as 9.17.0.0.0, i.e., at the height of theGreat Period, we find ornamental dots omitted in bar-and-dot numerals, the katun coefficient, A2, on the west side of Stela 7, being a case in point. Here the number 17 is recorded without an ornamental p, central dot. But the omission is due to provincialism rather than to archaistic treatment. 2 For other monuments recording this same hotun-ending, see Appendix VI IL 6o THE INSCRIPTIONS AT COPAN. village. He was the first to call attention to them, and was then of the opinion that all four were parts of the same monument, to which he gave the name Stela i6} When the writer visited Copan in 191 5, he gave these fragments a close examination, and by means of exact measurements, as well as a comparative study of their subject-matter, it was possible to prove that, instead of being parts of one and the same monument, at least three different monuments are represented here, fragments of two stelse, 18 and 20, and one altar, Q', two of the four fragments possibly belonging to the last. Because of the great importance of these monuments, particularly of Stela 20, which is probably the oldest stela now extant at Copan, and of Stela 18, which Spinden believes to be the first attempt to portray the human figure in front presentation at Copan. special eff^orts were made to ascertain their original provenance. The house of Domingo Hernandez was built in 1897 by Jacobo Madrid, who gives the following information about these sculptured pieces. He states that he himself carried all four of them, together with several other large unsculptured blocks and smaller sculptured pieces,- from the mound of Stela 7 (see figure 22) to their present position in order to use them in the walls of the house, as well as in the foundation of a low wall along the back corridor. This house was sold by Madrid in 1901 to Siriaco Ardon, who sold it the same year to Cristobal Melendez. From Melendez's hands it passed to Clementino Lopez in 1903, thence to Manuel Sagastume in 1906, thence to Antonio Guerra in 1909, and finally to Domingo Hernandez, the present owner, in 1917. It has seemed advisable to give the history of this house in detail, so that future students will have no difficulty in tracing the pedigree of these highly important fragments and in establishing their original provenance as the mound of Stela 7. Of the two pieces probably belonging to Altar Q', the first is 91 cm. long and 39 cm. thick. In facing it, the left side presents a broken edge, and it is therefore impossible to give the original width. The present maximum width is 58 cm. The top is very badly mutilated, most of the relief having scaled oflf. Traces of an interlacing of diagonal bands ^ appear in one place. The bottom and preserved side are dressed, jjl^ but not carved. The destroyed side was probably also plain. -t/T' The second fragment (see figure 6) shows this same treatment, i. e., top sculptured, the bottom and back dressed but plain. It is 86 cm. wide and 40 cm. thick. The front and both ends are broken ofi^, the present maximum height being 67 cm. The top of this second fragment is divided into three panels by two vertical bands which pass over the top from front to back, overlying the horizontal bands along the edges. (See top of figure 6.) The two lateral panels are incomplete. Both present the same subject, however, as Altars J', K', L', and M', namely, a large, grotesque serpent ' Spinden, 1913, table I. -One of the two pieces of Stela 25 and Fragment V I. INSCRIPTIONS OF THE EARLY PERIOD. .61 Fig. 6. — Inscription on top of Altar Q'. head. The middle field or glyph-panel originally contained four glyph- blocks; though most of the bottom block is now missing. From the part remaining, however, it is possible to estimate the original height of the top as having been 74 cm. As the first two glyph-blocks and upper horizontal band are 37 cm. high, the whole altar must have been twice this height, or 74 cm., of which the bottom 8 cm. are now missing. The thickness of these two fragments is practically the same, 39 cm. and 40 cm. respectively, and it may well be that both were originally parts of the same monument. If so, it was over 1.78 meters long. The inscription is fairly clear, with the exception of the last glyph-block. Unfortunately the loss of this glyph-block alone is sufficient to prevent exact dating, since it recorded the terminal day of the Period Ending date in Kib. (See plate 24, /, and figure 6.) This latter glyph is clearly the sign for the lahuntun or 3600-day perioa ^j]]-^ and should be interpreted as indicat- ing that Altar Q' dates from UujiJ one of these half-katun periods.' The day closing the particular ^^cP lahuntun here in question was re- corded in A4(3, and although the day-sign itself (Ahau) and possibly part of the coefficient are missing, the date of the altar can be limited to one of four possible readings under our postulate, with the probabilities in favor of one in preference to the other three. From what is left of the day coeffi- cients in A4<3, {?^^^ it would appear to have been either 7 or 12, although 6 or II or even 8 or 13 are not impossible readings. Two numerical dots appear and the possibility of another, now effaced, must be recognized. Inspection of the coefficient, therefore, gives 7 or 12 as the best values, with 6, II, 8, or 13 as remoter possibilities. There were 10 lahuntuns in the first half of Cycle 9, /. e., during the Early Period, as follows: 7 Ahau 3 Yax 5 Ahau 3 Tzec 3 Ahau 8 Cumhu 1 Ahau 8 Mac 12 Ahau 8 Mol 10 Ahau 8 Zip 8 Ahau 13 Pax 6 Ahau 13 Zac 4 Ahau 13 Xul 2 Ahau 13 Pop 9.0.10.0.0 9.1.10.0.0 9.2.10.0.0 9.3.10.0.0 9.4.10.0.0 9.5.10.0.0 9.6.10.0.0 9.7.10.0.0 9.8.10.0.0 9.9.10.0.0 'The term "lahuntun" has been suggested by the writer for the half-katun or lo-tun period, lahun being the Maya word for ten. See Morley, igiyi, p. 197 and plate 2 and Appendix II, pp. 566, 567. 62 THE INSCRIPTIONS AT COPAN. Of these, the results of our preliminary inspection would give preference to the first or fifth, 9.0.10.0.0 7 Ahau 3 Yax or 9.4.10.0.0 12 Ahau 8 Mol, respectively, though between these two it is impossible to choose on the basis of anything now recognizable in the text. Moreover, it is even neces- sary to admit two other values, 9.6.10.0.0 8 Ahau 13 Pax or 9.7.10.0.0 6 Ahau 13 Zac, as remoter possibilities. There are present, however, two other factors which make it extremely likely that 9.4.10.0.0 12 Ahau 8 Mol was the date originally recorded here. In the first place, historical probability as well as the stylistic criteria favors the later rather than earlier reading, and in the second place there is a stela in the immediate vicinity, {. e.. No. 15, recording this same date (see pp. 86-89). Moreover, 9.0.10.0.0 is 40 years older than the earliest surely deciphered date at Copan (Stela 24), and if accepted would cause a lacuna of that length in the sequence of the sculptures. Again, the fact that there is a stela recording the later date, for which no corresponding altar has yet been found, itself renders the later reading the more probable. The case may be summed up as follows: Although exact proof is wanting, it is not unlikely that Altar Q' may have recorded the lahuntun 9.4.10.0.0, the same as Stela 15, and in that case it may have been associated with Stela 15 in ancient times. If this reading is rejected, the next best appears to be 9.7.10.0.0 6 Ahau 13 Zac, on the ground that the three dots of the day coefficient are obviously not of the same size. The style of Altar Q' closely resembles that of Altars L' and M'. Indeed, if we assume that the missing left end of Altar W , and probably of Altar \J as well, presented the same large serpent heads as their respective right ends, only reversed, an assumption the writer favors, we have exactly the same design as on Altar Q', /. e., a panel of glyphs flanked on either side by a large serpent head and divided by the same arrangement of vertical and hori- zontal bands. The only difi^erence would then be in the number of the glyph-blocks, Q' having 4 and L' and M' 6. And further, since Q' is almost certainly referable to the lahuntun 9.4.10.0.0, it is probable that L' and M' date from the same general period. Possibly being a little more complex in subject-matter, they may be slightly later, 9. 5.0.0.0 or 9.5.10.0.0, for example, although it is dangerous to push the stylistic criteria too far when objects and treatment are so similar. In closing the presentation of these five archaic altars, it should be pointed out that they are all tables, flat slabs of stone, which were intended to lie on their broad faces rather than stand on their ends or narrow faces. This is conclusively proved by L' and M', where the front narrow faces are sculptured with glyphs, the backs and bottoms being dressed, but plain. In other words, to have both the top and front designs appear right side up at the same time, it is necessary to have the stone lying on its plain broad surface, and face the narrow sculptured front. In this latter position only will the designs on both the top and front appear right side up. inscriptions of the early period. 63 Altar P'. Provenance: Found on the mound of Stela 7 (Group 9). Destroyed in 1915 or 1916. (See plate 3 and figure 22, d'.) Date: 9.0.0.0.0 to 9.5.0.0,0. During the writer's visit to Copan in April 191 5, he found at the western edge of the mound of Stela 7, in the yard of the house of Clementino Lopez, in the southwestern quarter of the village (see figure 22, d'), a badly effaced fragment of what appeared to be an archaic altar. Although only very faint traces of the original design were preserved, it was possible to distinguish the outlines of three glyph-blocks in a vertical column and one or two scrolls or circles above. The single side preserved was dressed, but not carved. The condition of the stone was such that it was unsafe to attempt to date it even upon stylistic grounds, other than to refer it to the early part of Cycle 9. It appears to have been destroyed about the same time as Altars L' and M', as the writer could not find it when he was at Copan in March 1916. Altar X. Provenance: Original position unknown. Found i kilometer west of the Main Structure in the foundations of Stela 5 (Group 8). (See plate 3.) Date: 9.3. 6. 17. 18 11 Eznab i Kankin (?) or 9. 5. 19. 12. 18 II Eznab i Kankin (.') or 9.8.12. 7.18 II Eznab i Kankin (?). Text, [a) photograph: plate 8, c. (b) drawing: Gordon, 1902a, plate 13. References: Gordon, 1896, pp. 42, 43. Gordon, 1902a, pp. 130-132, 139-143. Maudslay, 1889-1902, vol. I of text, pp. 66, 67. Spinden, 1913, p. 161, 164, and table i. Altar X is 1.22 meters long, 91 cm. wide, and 30 cm. thick. When found, it was serving as the pedestal of Stela 5, about a third of a meter below the level of the pavement of small stones which had surrounded this monument. It is sculptured on its top and four sides, with a design of bands crossing each other at right angles and dividing each sculptured face into four compart- ments of equal size. The top compartments are plain. Those on the long sides show human figures, and those on the ends are inscribed with glyphs. (See plate 8, c.) There are four of these glyph panels, each containing 4 glyphs, a total of 16 for the entire text. The sculpture is in very low, flat relief. The first two glyphs, ai, bi (plate 8, c), record a Calendar Round date,' which reads as follows: 11 Lamat or Eznab, i, 2, or 3 Kankin. Since neither Lamat nor Eznab can occupy the third or fourth position in any month, - it is obvious that the month coefficient recorded here must be that corre- 'Calendar Round dates recurred at intervals of 52 years, and unless additional data are present (j. e., the corresponding Initial Series or Period Endings) tliey can not be assigned to their proper positions in Maya chro- nology. 'The only positions either of these days could ever occupy were the second, seventh, twelfth, or seventeenth divided has a small circle in the center, thus: tirely wanting in ai, we must identify it as only other day-sign possible here. (Compare ^]^ra 64 THE INSCRIPTIONS AT COPAN. sponding to the second position, namely i.' That is, the upper and lower dots in Bi are purely ornamental. An examination of the original, more- over, proves this to be the case, the upper and lower dots being clearly differ- ent from the middle one. (See plate 8, c, bi.) We may therefore restrict this date to one of two possible readings, 11 Lamat i Kankin or 11 Eznab I Kankin. Since both Lamat and Eznab may occupy the second position in any month, the identity of ai can be determined only by a study of the internal characteristics of the sign itself. Whenever the normal form- of Lamat appears in the inscriptions, each of the four quadrants into which it is As such circles are en- the sign for Eznab, the Ai, plate 8, c, with the forms for Eznab in Bowditch, 1910, plate 6.) ai, bi, therefore, reads 11 Eznab i Kankin. This latter date, as stated above, recurred at intervals of every 52 years in the Long Count,-'' and therefore additional data are necessary if we are to determine the exact position in the Long Count which the ancient sculptors had in mind when they carved it. Finally, as such data appear to be lacking in the text itself, we must depend upon the style of the monument and its position in the stylistic sequence to settle this question. Fortunately a consideration of the style of Altar X leaves little room for doubt as to its position in the stylistic sequence at Copan. For example, the omission of the ornamental elements on each side of the dot in the num- ber II in Ai is a fairly reliable indication that it belongs somewhere in the Early Period; and when this point is taken into consideration with other indubitably technical as well as stylistic crudities present, there can be no doubt that it is one of the earliest sculptures found there. Combining the data derived from these two independent lines of evidence, i. e., the chrono- logic and artistic, it will be found that there are only three positions possible for the date 1 1 Eznab i Kankin in the Early Period, namely: 9.3. 6. 17. 18 II Eznab i Kankin 9. 5. 19. 12. 18 II Eznab i Kankin 9.8.12. 7.18 II Eznab i Kankin But when it comes to choosing further between these three, we venture upon uncertain ground. Indeed, each has something that may be urged in ^Owing to the Maya custom of recording only elapsed time, the first position in a Maya month was written zero, viz, o Kankin, the second position, i Kankin, the third, 2 Kankin, the fourth, 3 Kankin, etc. The second position, therefore, i. e., I Kankin, is the only reading possible here. 'Most Maya glyphs have two distinct forms: (l) the normal form and (2) the head variant. The latter, as its name implies, is a human, animal, or grotesque head. In the day and month-signs the head variants are charac- terized by the same essential elements as their corresponding normal forms; but in the period glyphs the two forms usually have little or nothing in common. See Morley, 1915, pp. 24, 25. ^The Long Count is a term that has been applied to the old Maya chronological epoch. Dates are fixed in this period by the record of their corresponding Initial Series, i. ^., their distances from the starting point of Maya chronology. This method of recording dates, as pointed out in Chapter I, was so accurate that a given date could not recur, filling all the given conditions, until after an interval of 374,000 years, and possibly until after 5,000,000 years. See note i, p. 34. MORLEY PLATE 8 a. Altar A', re-uscd in the Hieroglyphic Stairway of Mound 26. b. Altar Y, re-used in the foundations under Stela 4. c. Altar X, re-used in the foundations under Stela 5. d. Altar J', re-used in the foundations under Stela 10. -^'■^^e; ^j^'' e. Altar K', re-used in the foundations under Stela 10. f. Altar L', destroyed in 1916. g. Altar M', destroyed in 1916. INSCRIPTIONS OF THE EARLY PERIOD. 65 its favor. For example, if the first, 9. 3. 6. 17.18, were tiie value originally intended for 11 Eznab i Kankin here, it would have a peculiar fitness, as being just two days before the end of a tun in the Long Count: 9.3.7.0.0. On the other hand, there are strong reasons for believing that Altars X and Y and Stelae 16 and 17 are closely related; possibly X and Y originally having been the altars associated with Stelae 16 and 17. If this is true, the second date, 9.5. 19. 12. 18, was probably the one intended, since Stela 17 is known to have been erected some time in Katun 6 (p. 90). This second value for II Eznab i Kankin, moreover, is only 102 days before the end of Katun 5, i. e., 9.6.0.0.0. The importance of katun, lahuntun, and hotun-endings in the Maya system of counting time can not be overestimated,^ and it is not at all improbable that the stela with which Altar X was originally associated may have recorded the date 9.6.0.0.0 9 Ahau 3 Uayeb. Indeed, Stela 17 itself may be this very monument, since it surely dates from Katun 6 and could hardly have been other than 9.6.0.0.0 or 9.6.10.0.0. In the present state of knowledge it is diflficult, indeed unsafe, to press the evidence available as to the age of Altar X further than to state that this altar almost certainly may be assigned to one of the three dates suggested, with the probabilities in favor of the second. Gordon- suggests the reading 4.6.0.0.0.0 11 Ahau 3 Kankin for this date, based upon an exceedingly ingenious explanation of the decorative elements on the monument, i. e., the bands and human figures. This reading is more than 250,000 years earlier than the earliest contemporaneous date found anywhere else in the Maya inscriptions, and for this reason alone it should be accepted with reservation. Its rejection, however, rests on firmer grounds than historical impossibility, since the text itself does not allow the interpretation he suggests for it. The date actually recorded in ai, bi, as we have seen, is not 11 Ahau 3 Kankin, but 11 Eznab i Kankin. Even admitting that the month Coefficient looks as much like 3 as i, it is impossible to identify the day-sign in ai as Ahau. A study of the forms for Ahau'* elsewhere and also in this same text at hi discloses no other form which bears the slightest resemblance to this, while its resemblance to the sign for Eznab on the other hand is striking. For these two reasons, then, (i) the inherent historical impossibility of such a remote date and (2) the impossibility of the day-sign recorded being Ahau, the writer has rejected Gordon's reading. Some of the remaining glyphs of this text are familiar, but of unknown meaning, as ci, f2, gi, and g2. The last glyph, h2, as already noted, is 3 Ahau. 'This point is fully covered in Appendix VII .Tnd its presentation will not be anticipated here. -Gordon, igo2a, p. 141. ^Bowditch, 1910, plate 6, and in Appendix X. 66 THE INSCRIPTIONS AT COPAN. Altar Y. Provenance: Original position unknown. Found in the Great Plaza at the Main Structure in the foundations of Stela 4. (See plate 6.) Date: 94- 8.12.6 6 Cimi 19 Uo (?) or 9.7. I. 7.6 6 Cimi 19 Uo (?) or 9.9.14. 2.6 6 Cimi 19 Uo (?). Text, (a) photograph: plate 8, fe. Maudslay, 1889-1902, vol. I, plate lOj, c (front only). (b) drawing: ibid., plate 104. Gordon, 19020, plate 14. References: Gordon, 1896, pp. 42, 43. Gordon, 1902a, pp. 130-132, 139-143. Maudslay, 1889-1902, vol. i of text, pp. 66, 67. Spinden, 1913, pp. 161, 164, and table i. Altar Y is 1.22 meters long, 91 cm. wide, and 38 cm. thick. It was found buried in the ground underneath Stela 4, in the support of which it seems to have served. Its decoration is similar in every respect to that of Altar X. There are the same bands dividing the top and four sides into four compartments each, the same arrangement of figures in the four panels on the long sides, and of glyphs in the four panels on the short sides. Each glyph panel similarly has 4 glyph-blocks, making a total of 16 for the entire text. In size, shape, and decoration the two monuments are prac- tically identical. The text on Altar Y (plate 8, b), like that on Altar X, opens with a Calendar Round date, which the writer deciphers as 6 Cimi 19 Uo,' although the month-sign in bi may possibly be Zip instead of Uo. As the main elements in the signs for Uo and Zip are identical r (a pair of bands crossing in the center), one is frequently mis- ^ taken for the other, and vice versa, ^he determ ining charac- teristic in each case is the superfix (or prefix),^ which in Uo has several variants but which in Zip is confined to a 'A serious error should be noted in Gordon's drawing of this text (19020, plate 14, 3, Ei). He shows the month coefficient thus: Maudslav coefficient 19, but makes introduces an interior nation of the original that is as 14, 2 bars and 4 dots, the first bar showing a decorative inner line. ( 1 889-1902, vol. I, plate 104, upper half, glyph 2) shows the correct month the right-hand bar thicker than the middle and left ones, and decoration in the dots, not present in the original. An exami- E showed that Maudslay's drawing of the coefficient is substan- ^ tially correct, except that all three of the bars are practically of the same thickness. Gordon's error p seems to have arisen through mistaking the line between the first and second bars for a decorative ' element of the first bar, thus reducing the number of bars from 3 to 2 and making the coefficient 14 instead of 19. -There are four other month signs Chen, Yax, Zac, and Ceh, which have their main elements alike. These also are to be distinguished from each other only by their super- fixes or prefixes. It should be noted in this connection that the superfixes in the signs for Zip and Ceh are identical, the only difference between thetvvo signs being their main elements. See last two signs above. 'In one text at Copan, Altar Q, E6, the Uo superfix is wanting altogether. In this case the main element takes the unusual form shown herewith. INSCRIPTIONS OF THE EARLY PERIOD. 67 single form.i £J]][^ ^Hfe ^^^ ^^^ superfix in bi (plate 8, b), how- ever, bears a strong resemblance to the first and second variants of the Uo superfix just given, and at the same time is totally dissimilar to any of the known forms of the Zip superfix. The reading 6 Cimi 19 Uo therefore appears reasonably certain. - This date occurred in Cycle 9 within the limits stylistically probable at three^ positions, namely: 9.4. 8.12.6 6 Cimi 19 Uo 9.7. I. 7.6 6 Cimi 19 Uo 9.9.14. 2.6 6 Cimi 19 Uo Whichever of these values is chosen, it will be found that it is within 21 years of one or other of the three values already given for Altar X, viz: 9.4.8.12. 6 9.7. I. 7. 6 9.9.14. 2. 6 9.3.6.17.18 9.5. 19.12. 18 9.8.12. 7.18 1. 1. 12. 8 I. 1. 12. 8 I. 1. 12. 8 In other words, although the dates of Altars X and Y may be 73 or even 125 years apart,^ in view of their very close stylistic similarity, they were probably only 21 years apart, which is as near as their dates will permit them to be. Therefore, if we could establish the date either of Altar X or Altar Y, the corresponding date in the other set would probably be correct for the other monument. Unfortunately, viewed in the light of Altar X, the most probable date for Altar Y would be the second value given above, whereas, judged on its own merits alone, the first is the only one of the three which has anything particular that may be urged in its favor. To begin with, 9.4.8.12.6 is less than two years earlier than the date of Stela 15, which is 9.4.10.0.0. Again, this date is only 104 days earlier than the next tun ending in the Long Count, 9.4.9.0.0. Finally, it is only 454 days earlier than one of the two best readings for Stela 16, 9.4.9.17.0. Even in spite of these rather satis- factory connections with other monuments, it appears unwise in the absence of more definite evidence to accept this reading as final or to reject altogether the other two possibilities. Further consideration of the date of Altar Y will be deferred until after Stelae 16 and 17 have been described. The remaining glyphs of this text are either unfamiliar or of unknown meaning. 'In three texts at Copan, namely, Stela N(east side), A15, Altar L, a2, and th e r eview ing stand o n the south side of Mound 11, vi, another variant for Zip seems to have been used, f '\ />— ^>v ( j— j '\ These three texts date from the same decade (9.16.10.0.0 to 9.17.0.0.0), and the close 1^^ ^^ ? \ similarity between their three forms for Zip may be due to the personal equation 11(^^1 J y^~y (^^^J of a single sculptor. At least, the above variants have not been found elsewhere, and fW/®~J f ^^ ) ?<^ N* ) •"■ Copan they appear only during this particular decade. ^In order that the student may draw his own conclusions, however, Appendix X should be consulted, where all known occurrences of all the day and month-signs in the Copan inscription are listed. 'The very earliest occurrence of 6 Cimi 19 Uo in Cycle 9, ;'. f, 9. 1. 15. 17. 6, is not included above, as it is too early to be either historically or stylistically probable. ^That is I or 2 Calendar Rounds + i.l.12.8. 68 THE INSCRIPTIONS AT COPAN. Altar A'. Provenance: Original position unknown. Found on the under side of the last block of Step P of the Hieroglyphic Stairway on the west side of Mound 26 at the Main Structure. (See plate 6.) Date: 9.3.0.0.0 to 9.7.0.0.0. Text, photograph: plate 8, a. Gordon, 1902, plate 13, U. Spinden, 1913, plate 20, i. References: Gordon, 1902, p. 19. Gordon, 1902a, p. 130. Maudslay, 1889-1902, vol. I of text, pp. 66, 67. Spinden, 1913, p. 161, and table i. Altar A' was found during the excavation of the Hieroglyphic Stairway of Mound 26, on the under side of the last block in Step P (see p. 251, note 2), having been cut down from some larger monument in ancient times. ^ In reshaping it for secondary use here, part of the original design was broken off and is now missing. Judging from the fragments preserved, the original monu- ment must have closely resembled Altars X and Y, though the arrangement of the intersecting bands is slightly different. On Altar A' three vertical bands and one horizontal band divide the field of the single sculptured sur- face preserved into four panels of 4 glyph-blocks each, making a total of 16 glyph-blocks for this one side alone. The difference between this monument and Altars X and Y is that the two latter have no horizontal bands crossing their short sides where the glyphs are presented. Consequently, there are only two glyph panels on each short side of Altars X and Y as compared with four glyph panels on Altar A'. Unfortunately, in reshaping this block for use in the Hieroglyphic Stairway, the upper row of glyphs was broken off clear across the top, and since the date was probably presented in the upper left-hand corner, as on Altars X and Y, it is impossible to fix the position of this monument in the Long Count. The remaining glyphs have escaped interpretation up to the present time, although a few are not unfa- miliar. It is safe, however, to assign Altar A' to the same general period as Altars X and Y on the basis of its close stylistic similarity, apparent in execu- tion, subject-matter, and arrangement. With Altar A' we reach the last of this type of monument in the Early Period, and turn next to a considera- tion of the early stelae, of which there are 13 now known. Stela 22. Provenance: Found on the north side of a small plaza on the south- western outskirts of the village (Group 9). Now in the cabildo. (See plate 3, and figure 22, w.) Date: 9.3.0.0.0 to 9.5. 0.0.0. Text, (a) photograph: plate 28, b. (b) drawing: figure 7. Only a single small fragment of this stela was found. (See figure 7.) The part recovered is 53 cm. long, 47 cm. wide, and 32 cm. thick. From the size of the glyph-blocks preserved, 16 cm. high by 21 cm. wide, it was possible to estimate the original width of the monument to have been 58 cm. 'Gordon, 1902, p. 19, and igoza, p. 130. INSCRIPTIONS OF THE EARLY PERIOD. 69 This fragment was found in a small plaza on the southwestern outskirts of the village on the edge of the bank or terrace which overlooks the flood- plain of the river. (See figure 22, zv' .) It was lying on the north side of this plaza in front of a low platform mound approached by five broad steps on its southern side. Although there was very little debris in this court, excavations were made on the north side with the hope that other fragments might be recovered, but nothing was found. Indeed, it is more than probable that the other pieces have been destroyed. In March 1916, when this fragment was found, the blocks of stone forming the stairways of the mounds surrounding this court had been removed and were about to be broken up for paving material for the streets. The single piece recovered may have been removed from one of these steps, and it was rescued from destruction only by the writer's chance visit at the time. It is presumably part of a stela, but of very archaic character. The front is sculptured with glyphs, the back and left side being plain, having been dressed smooth by pecking and rubbing. The right side was missing, as well as the top and bottom. There is a plain marginal band along the left edge of the front. This presentation is unique, and almost raises the question whether this fragment ever belonged to a stela, possibly having been part of an altar. The arrangement of the glyphs in two vertical columns, however, suggests a stela, and for this reason it has been so identified here. On the basis of the arrangement of the design, which is the simplest possible, i. e.. one surface sculptured, the other three being left plain. Stela 22 has been assigned to Class i. Parts of 6 glyph-blocks are preserved, and although none is of recog- nizable form, they all clearly indicate the archaic character of the inscrip- tion. Note the highly ornamented numerical bars in a2 and the archaic prefix in b2. The latter is identical with the prefix of C6 on Stela 15 (g g 'g and of B5 on Stela 9 ^^^^ and very similar to that of b5 on Stela 24. But these three monuments are very early, 9.4.10.0.0, 9.6.10.0.0, and 9.2.10.0.0 respectively, for which reason Stela 22 also has been assigned to Katuns 2 to 4. Stela 25. Provenance: Found on the mound of Stela 7 (Group 9). Two frag- ments only recovered. Now in the cabildo. (See plate 3, and figure 22, /' and g'. Date: 9.2.10.0.0 3 Ahau 8 Cumhu (?). Text, drawing: figure 8. Only two contiguous fragments of this monument have been recovered up to the present time, both from the southwestern quarter of the village, one having been found behind the house of Clementino Lopez in the mound of Stela 7 in 1918 (figure 22, /'), and the other 40 meters to the northwest at the house of Domingo Hernandez in 1919 (figure 22, g')- These two frag- ments were fitted together in June 1919 and found to be parts of a new monument, to which the name Stela 25 was given. Fortunately the history of the Hernandez house is well known (see p. 60), and it is possible to establish the fact that the fragment found 70 THE INSCRIPTIONS AT COPAN. there in 191 8 had been carried thither from the mound of Stela 7 in 1897 during its construction. (See figure 22,^'.) When fitted together these two fragments make one piece, which originally formed a part of the right half of the front (or back) of a stela, the adjoining surface to the right being dressed but not sculptured, and that to the left being rough, as though a part were broken off here. Fic. 7. — Inscription on front of Srela 22 Fig. 8. — Inscription on front (or back) of Stela 25. The single sculptured surface of the assembled piece (see figure 8) shows parts of two glyph-blocks, 34 cm. wide and at present 21 cm. high. And if, as the writer believes, the two upper signs are parts of the same glyph-block as the two lower ones — and no space shows between the two characters in the upper half of the glyph-block (see figure 8) — the glyph- blocks on this monument originally must have been 30 cm. high. Finally, if the piece recovered represents only the right half of the monument, as seems possible, its total width originally must have been 66 to 68 cm. It is possible, however, that the piece as found, represents the original width of the monument, in which case it was only 33 or 34 cm. wide. The only surely decipherable character is the lower left-hand glyph in the upper glyph-block, which unmistakably records the day 3 Ahau. This is surmounted by a tassel-like ornament practically identical with that just described as occurring in the day-signs on Stelae 24, 15, and 9. This glyph is followed by a sign, the coeflicient of which is surely 8. Parts of three other signs show, but they are all of unknown meaning. In order to decipher this date, it is first necessary to niake the following postulate, namely, that this day 3 Ahau stood either at the end of some katun or lahuntun of the Long Count. This postulate, as will appear at the end of this chapter, is amply substantiated by practically all the stelae of the Early Period at Copan, and if hotuns, i. c, quarter katuns, be included, it is substantiated by 95 per cent of all stelae throughout the Maya area. 7 Ahau 3 Yax ^ Ahau 3 Tzec 3 Ahau 8 Cumhu I Ahau 8 Mac 12 Ahau 8Mol 10 Ahau 8 Zip 8 Ahau 13 Pax 6 Ahau 13 Zac 4 Ahau 13 Xul 2 Ahau 13 Pop INSCRIPTIONS OF THE EARLY PERIOD. 7I As applied to Stela 25, this postulate restricts the number of places where this day 3 Ahau could have occurred in the first half of Cycle 9 to 2 places out of a possible 277, one a katun-ending, A, and the other a lahun- tun-ending, B, as follows : 9.0.0.0.0 8 Ahau 13 Ceh 9.0.10.0.0 9.1.0.0.0 6 Ahau 13 Yaxkin 9.1.10.0.0 9.2.0.0.0 4 Ahau 13 Uo B 9.2.10.0.0 9.3.0.0.0 2 Ahau 18 Muan 9.3.10.0.0 9.4.0.0.0 13 Ahau 18 Yax 9.4.10.0.0 9.5.0.0.0 II Ahau 18 Tzec 9.5.10.0.0 9.6.0.0.0 9 Ahau 3 Uayeb 9.6.10.0.0 9.7.0.0.0 7 Ahau 3 Kankin 9.7.10.0.0 9.8.0.0.0 5 Ahau 3 Chen 9.8.10.0.0 A 9.9.0.0.0 3 Ahau 3 Zotz 9.9.10.0.0 The foregoing tabulation shows that under our postulate there are only two readings possible for this date in the first half of Cycle 9, namely, A, 9.9.0.0.0 3 Ahau 3 Zotz and B, 9.2.10.0.0 3 Ahau 8 Cumhu; and of these it can be shown that the latter is by far the better choice. In the first place, 9.9.0.0.0 is too late on stylistic grounds for the glyphs of Stela 25 to have been executed, which, as we have seen, more closely resemble those of Stelae 24, 15, and 9 (9.2.10.0.0, 9.4.10.0.0, and 9.6.10.0.0 respectively) than those of Stelae 7, E and P (9.9.0.0.0, 9.9.5.0.0, and 9.9.10.0.0 respectively); and in the second place, the earlier reading receives remarkable corroboration from the coefiicient of 8 in the next glyph. For in the event of the latter being the correct reading, the next glyph might then be 8 Cumhu, that is, the corresponding month part of the Initial Series terminal date, 9.2.10.0.0 3 Ahau 8 Cumhu. It is true the human head to which this 8 is attached bears little or no resemblance to any known form for the month Cumhu,^ but it should be borne in mind in this connection that at the early period Stela 25 was carved, many glyphs had not yet developed the characteristics by which they were distinguished in later times, and that the glyph in question may therefore possibly be an early form for this month. The corroboration afforded by finding the appropriate month coefiicient (i. e., 8) in the following glyph more than counterbalances the failure to discover in the sign itself recognizable characteristics of Cumhu, and the writer therefore believes that the date here recorded is probably 9.2.10.0.0 3 Ahau 8 Cumhu. This date appears as the Initial Series on Stela 24 (see pp. 80, 81), and at first it seemed possible that these two fragments might be parts of that monument; but a comparison of their glyph-blocks immedi- ately showed that this never could have been the case, those on Stela 24 being 18 to 19 cm. high and 27 to 28 cm. wide, and those on Stela 25, 30 cm. high and 34 cm. wide, and it was therefore necessary to recognize this piece as part of another stela, to which the number 25 has been given. 'See Bowditch, 19 lo, plate 10, and Appendix X, p. 592. THE INSCRIPTIONS AT COPAN. Stela 20. Provenance: Found on the mound of Stela 7 (Group 9). Now in the cabildo. (See plate 3, and figure 22, v.) Date: 9.1.10.0.0 5 Ahau 3 Tzec (?). Text, (a) photograph: plate g,b. ib) drawing: figures 9, 10, and 11. The largest fragment of Stela 20 (figure 9 and plate 9, b), and in fact the only one which is surely a part ofthis monument, was found by Spinden in 1914, in the house of Domingo Hernan- ^ ^ dez. In 1916, however, the writer discovered two other widely scat- tered fragments (see figure 10), one built into the wall of the house of Pedro Ramirez in the street lead- ing south from the southwestern corner of the village plaza, and the other in the foundations of the now destroyed house of Felix Galvan, just east of the new market on the street leading west from the same corner, and some 100 meters distant from the first. (See figure 22, y and z respectively.) On being assembled, these last two fragments were found to fit together exactly and thus to have been parts of the same monument; and in the discussion which follows they will be referred to as one piece. The important question, however, is whether or not this second piece is a part of the same monument as the fragment found by Spinden in the house of Domingo Hernandez in 1914. Fig. 9.- -Inscription on front and riuin side of Stela 20 (Fragment I). Fig. 10. — Inscription on back and sides of Stela 20 (Fragments 2 and 3). The answer to this question must be ascertained, first from the measure- ments of the two pieces, next from their subject-matter, i. e., the inscription, and finally from the stylistic criteria present. The measurements of the two pieces are compared on page 73. INSCRIPTIONS OF THE EARLY PERIOD. 73 It will be seen from the measurements below that both pieces are of the same width, 52 cm., but since both are only fragments, the present heights have no significance. Further, as the pieces can not be fitted together back to back, it is impossible to determine the original thickness of the monument exactly. Judging from the incomplete glyph-blocks on their respective sides, the latter must have been of unusual width, at least 23 to 24 cm. and 22 cm. respectively, that is, wider than the glyph-blocks on their correspond- ing fronts. This almost certainly indicates that originally there had been only one column of glyph-blocks on the sides of each. And on the basis of this assumption, the former thickness of each piece can be estimated to have been about 40 cm. From house of Hernandez. From houses of Ramirez and Galvan. Dimensions of the fragments : Width CVl. 52 75 30 16 to 18 19 to 20 16 to 19 23 to 24 cm. 52 64 20 16 to 16.5 19 to 20 17 to 19 22 Present height Present thickness Dimensions of the glyph-blocks: Front and back: Width Sides: Height Present width Continuing this comparison, it will be noted that both pieces have double columns of glyph-blocks on one of their two broad faces and that the opposite broad faces are fractured. The measurements of their respective glyph-blocks show even more satisfactory agreements. The widths — the important measurement, if both are parts of the same monument — are identical, 19 to 20 cm., and the heights vary by less than 2 cm. at the most. On the sides it is impossible to secure the original widths, but the heights are the same in both cases. This identity in the essential measurements, not only of the two pieces but also of their respective glyph-blocks, strongly suggests that both are parts of the same monument; but if so, how were they related. All attempts to fit them back to back on the ground failed, and indeed it was apparent that the two fragments recovered never could have fitted together in that way. It will appear presently, in the discussion of the inscription, that the fragment found by Spinden in 1914 was surely a section of the front of the stela, coming from immediately below a top section, 1.14 meters long by calculation, which presented the Initial Series. Since the piece found in 1916 is only 64 cm. long, and could not have been from the front in any case, it seems more probable that it came from a lower section of the same stela (if it belongs to it at all) instead of from the top, as this missing top was 50 cm. longer, if it broke in one piece. The best assemblage of the two pieces is shown in figure 11, where the second piece is placed at the bottom. 74 THE INSCRIPTIONS AT COPAN. I , ,-- __i I --\ I I Spinden supposed the fragment found by him to be a part of Stela i8, but its inscription indicates that this could not have been the case. This fragment, as just noted, was clearly from the middle part of a monument. This fact is established by internal evidence in the text itself on the front (see plate 9, b, and figure 9), which unmistakably records a com- plete Supplementary Series. Let us analyze this evidence further. So far as the writer knows, there is not a single instance in the Maya inscriptions where a Supplementary Series is recorded that an Initial Series does not immediately precede it; but the reverse of this propo- sition does occur, i. e., the record of an Initial Series without an accompanying Supplementary Series, although rarely. Therefore it is safe to conclude that the fragment shown in plate 9, b was originally / preceded by one or more fragments which presented an Initial Series. The last glyph O/'jmilA in plate 9, /^ is the month-sign O of an Initial Series terminal date,^ but unfortunately the fracture runs across its lower half, and beyond the fact that the coefficient is surely 3, it is difficult to determine at first inspec- tion what month-sign was recorded here. Making the same postulate as in the case of Stela 25 (p. 70), namely, that the Initial Series of Stela 20, whatever it was, recorded a katun, lahuntun, or at least a hotun^ ending in the Early Period, it will be p,„ shown presently that the possible readings for this Initial Series, which could have had a month coefficient of 3, under our postulate are limited to ten. Further, the fact that the month coefficient of the Initial Series terminal date on Stela 20 is 3, is at least strong presumptive evidence that we have here either a katun, lahuntun, or hotun ending, since these only could end 'That this glyph could possibly be the day of the Initial Series terminal date would appear highly unlikely from what is left of it. The part preserved looks like a superfix, i. e., it curves in at both ends, and is ornamented with lines unlike any known forms of the day-sign cartouche. This in itself renders such a possibility extremely remote, even if such a position for the day-sign were not contrary to the general practice of recording the day at the end of the Initial Series number, and not at the end of the Supplementary Series. Out of 80 Initial Series, only 7.5 per cent, were found to have the Initial Series terminal date follow the Supplementary Series. See Morley, 1916, p. 368, note I. ^The term "hotun" has been suggested by the writer for the quarter katun or 5-tun period, "ho" being the Maya word for five. This whole question, i. e., what chronological considerations controlled the erection of the stelae, a matter which very closely touches their probable function and significance, is extensively reviewed in Appendix VII, and alsfi in Morley, \()ijb. I, 1 1 -4 L_ , --1 1 1 1 II J 1 If 1 1 1 ' -""1 1 II. — Diagram showing probable assem- blage of Fragments i, 2, and 3 of Stela 20. INSCRIPTIONS OF THE EARLY PERIOD. 75 on days the corresponding month coefficients of which were either 3, 8, 13 or 18.^ It is therefore presumptively probable that the missing Initial Series of this monument recorded the end of some katun, lahuntun, or hotun in the Early Period. Referring to Goodman's tables,- it will be found that there were ten katuns, lahuntuns, and hotuns in the Early Period, that ended on days the month coefficients of which were 3, namely: 9.0.10.0.0 7 Ahau 3 Yax 9.5.15.0.0 3 Ahau 3 Uo 9.I.IO.O.O " '*' " '^ ~ ^ ~ "' ' TT_ 9.2.15.0.0 9.3.15.0.0 9.4.15.0.0 5 Ahau 3 Tzec 9.6. 0.0.0 9 Ahau 3 Uayeb 9 Ahau 3 Kayab 9.7. 0.0.0 7 Ahau 3 Kankin 7 Ahau 3 Ceh 9.8. 0.0.0 5 Ahau 3 Chen 5 Ahau 3 Yaxkin 9.9. 0.0.0 3 Ahau 3 Zotz Because of the extremely archaic style of Stela 20, which is surely earlier than that of Stela 24, the date of which is definitely fixed as 9.2.10.0.0, it is possible to eliminate all but the first two values in the above list. To choose further between these, however, it is necessary to depend upon inter- nal evidence supplied by the month-sign itself, that is, the last glyphs in plate 9, h, and figure 9. Unfortunately the lower half of this glyph is missing. The upper half, however, is fairly clear, and as it bears no resemblance to any of the known forms for Yax,-' the first value above may be eliminated, which leaves the second as the only possible reading for this date under our postulate. Moreover, inspection of what is left of the month-sign shows that its superfix has a series of parallel lines (^^^ which is characteristic of the superfix of the sign for the month Tzec; and on the strength of this corroboratory evidence from the text itself it seems not unlikely that the date of Stela 20 may be 9.1. 10. 0.0 5 Ahau 3 Tzec. ^^I^ The correctness of this reading, it is obvious, rests upon the truth of our two postulates: (i) that Stela 20 dates from the early part of Cycle 9, and (2) that it records an even katun, lahuntun, or hotun ending in the Long Count. Concerning the first, let us examine the stylistic criteria present and see where this monument is to be assigned on stylistic grounds. As already pointed out, the style of Stela 20 is probably more archaic than that of any other monument at Copan. Indeed, in order to find monuments of equal or greater age, from which comparative stylistic criteria may be 'All time periods above the kin or day in the Maya chronological system ended on some one of the thirteen days Ahau. And since Ahau could have only a corresponding month coefficient of 3, 8, 13, or 18, all hotiin-endings were thereby automatically restricted to one of these same four numerals for their month coefficients. ^For deciphering Maya dates, as well as performing other calculations arising therefrom, the use of Good- man's Archaic Annual Calendar and Goodman's Archaic Chronological Calendar is strongly recommended. See (loodman, 1897. Other tables, although somewhat less convenient, are those devised by Gates. See Gates, 1900. A longer method involving reduction of the Maya terms to the decimal system was first used by Forste- mann in his pioneer work on the Maya hieroglyphic writmg. See Forstemann, 1887, p. 36. Aside from any sentimental interest attaching to this method as being the first, it is also probably the most readily comprehended by the European mmd, because it reduces the Maya periods to the terms of our own decimal notation. For an extended treatment of this method, see Motley, 1915, particularly Chapters III and IV. Bowditch (1910, Appen- dix Vll) has devised a method which considerably decreases the amount of purely arithmetical work in performing the different calculations present in the inscriptions, and R. K. Morley (1918) has developed this method even farther. While the greater brevity of these is conceded, theit use by beginners is hardly to be recommended, because of the complexity of the arithmetical principles upon which they depend; and for performing the common calculations present in the inscriptions, Goodman's tables are more convenient and expeditious. ■■■See Bowditch, 1910, plate 10, also Appendix X. 76 THE INSCRIPTIONS AT COPAN. derived, it is necessary to look to Tikal, the only other city now known which can lay claim to an equal antiquity with Copan.^ Tikal is probably the site best suited for the study of the art of the Early Period. Four of the five earliest dated stelae in the Maya area are found here,- and there are also more undated archaic monuments here than at any other site, Copan not excepted.-' The most constant characteristic of all the earliest stelae at Tikal is the pronounced irregularity in the outlines of the individual glyph-blocks. Two examples, one from Stela 3 and the other from Stela 10, will suffice to illustrate this point. On Stela 3 a2 OO is the katun-sign and coeffi- cient. Note how irregular the outline 9o-^ ^^ ^^is glyph /^0^*=^j^ is, and how inadequately it fills the '^^^::J available space. -J^:^^^ On Stela 10 a31 the uinal-sign and coefficient, also show the ^ same characteristic. In later times, both at Copan and elsewhere, the glyph-blocks assumed more or less rigid rectangular outlines with only the corners slightly rounded, but in Stela 20, which represents perhaps one of the first attempts to carve an inscription upon stone at Copan, and probably the earliest monument yet found here, this regularity in glyphic outline had not yet been achieved. The earliest stela yet found in the Maya area. No. 9 at Uaxactun, has this same characteristic. The glyph-blocks are not only irregular in outline, but also unevenly arranged in the two vertical columns; that is, a glyph- block in one column is not always exactly opposite the corresponding glyph-block in the other. (See figure 66.) Two other pieces of sculpture still more archaic, the Leyden Plate and the Tuxtla Statuette, also present this same characteristic. The Leyden Plate is a small celt-like object of nephrite about 21.6 cm. long, 8.5 cm. wide, and 2 to 5 mm. thick, upon the front of which a late Cycle 8 Initial Series, 8. 14. 3. 1. 12, is incised. The point claiming our particular interest in this connection is the fact that on this very early celt, which antedates Stela 24 at Copan and Stela 3 at Tikal by some 170 years, and Stela 9 at Uaxactun by 7 years, the outlines of the glyphs are even more irregular. (See figure 65.) The uinal glyph of the Initial O^^^l-^ Series here, as, well illus- trates this peculiarity. Not only is ^ ^^~f h) the uinal-sign itself very irregular in outline, but its coefficient ^^zlKp^ also is asymmetrically placed with reference to it.^ 'Uaxactun (see plate i) has earlier dates than Tikal, but it is so near by, not more than 25 kilometers distant, that it undoubtedly was tributary to the larger city, and has been considered here as one with it. ^Uaxactun, Stela 9, 8.14.13.10.15, and Stela 3, 9.3.13.0.0; Tikal, Stela 3, 9.2.13.0.0, and Stela 10, 9.3.ll.2.o(?); and Copan, Stela 24, 9.2.10.0.0. In the next to last case (Stela ro at Tikal) there is some doubt as to the value of the kin coefficient However, if ir were ig, the highest value possible, the above reading is correct to within 20 days of the true date. To these should be added the doubtfully dated: Stela 5 at Uaxactun, 8. 15. 10. 3. 12; Stela 8 at Tikal, 9.0.10.0.0; and Stela 9 at Tikal, 9.2.0.0.0. ^The writer is inclined to believe that of the 17 sculptured monuments described by Maler (1911, pp. 61-91) at Tikal, all but 3 or 4, Stels 5, 11, and 16, and possibly 6, belong to the Early Period, and most of them prob- ably to its earliest part. ^For further particulars concerning this interesting specimen, see Leemans 1877, pp. 299-301, Holden i88r, pp. 229-237, Motley, 1915, pp. 196-198 and Chapter V, pp. 411, 412. INSCRIPTIONS OF THE EARLY PERIOD. 77 The Tuxtla Statuette, a small anthropomorphic figure about 16.5 cm. high and 9.5 cm. in diameter at the base, is also of nephrite. It presents a still earlier Cycle 8 Initial Series, namely, 8.6.2.4.17, which antedates Stela 24 at Copan and Stela 3 at Tikal by more than 300 years, and Stela 9 at Uaxactun by about 150 years. Here again the outlines of the glyphs are still more irregular and sketchy. (See figure 63.) The first glyph on this figure, ai, probably the Initial Series intro- ducing glyph, (jy illustrates the extreme of irregularity in glyphic outline; indeed, there y^\ appears to have been only a very casual attempt to make this glyph- Ml i block rectangular at all.^ From the foregoing it is apparent that irregularity of glyphic outline is the most reliable criterion of archaism in glyph delineation — the more archaic the text, the more irregular the outlines of its glyphs. This is perhaps a natural consequence of transferring a graphic system developed upon wood or fiber-paper to stone, since the earliest attempts at glyphic delineation on stone could hardly have had the same rigid rectangular out- lines as those made after the sculptors had had long experience with the new medium. Returning to Stela 20, it is possible that in this monument we have the only one at Copan which presents this extremely archaic characteristic; and so far as its stylistic position is concerned, there can be no doubt that it is the earliest stela, if indeed it is not the earliest monument of any type yet found here, being certainly prior to Stela 24, the earliest surely deciphered stela, and probably prior to Altars J', K', L', M', P', and Q' as well. Concerning the second of the two postulates above upon which the read- ing suggested rests, namely, that Stela 20 records a katun, lahuntun, or hotun-ending, it should be noted that the vast majority of all Maya stelae do record such endings and that antecedent probability therefore is over- whelmingly in its favor.- In conclusion, therefore, it appears that notwithstanding the fact that this inscription is fragmentary, only one glyph out of the original eight in the Initial Series being preserved, the stylistic and textual corroboration is so satisfactory that the date suggested may be accepted as probably correct and the stela itself regarded as the oldest monument so far discovered at Copan. On the basis of the arrangement of the subject-matter, Stela 20 has been assigned to Class 3, i. e., all four faces sculptured with glyphs. Class 2 (not encountered so far) being reserved for monuments sculptured on two opposite faces, the remaining faces being left plain. 'For a complete description of tliis most important object, see Holmes, 1907, Morley, 1915, pp. 194-196. and Chapter V, p. 403. -See also Appendix VII. 78 the inscriptions at copan. Stela 24. Provenance: Original position unknown. Found in the foundations of Stela 7 (Group 9). Now in the cabildo. (See plate 3, and figures 12, i, and 22, q.) Date: 9.2.10.0.0 3 Ahau 8 Cumhu. Text, drawing: figure 13. With Stela 24 we reach at last firm chronological ground, this being the earliest monument at Copan the date of which has been surely deciphered. Only a single piece has been found, probably not more than a quarter of the original monument, although possibly as much as a third of the sculptured part. The fragment recovered is 61 cm. high, 73 cm. wide, and 29 cm. thick. The front and back are sculptured with glyphs, the two sides being plain, though dressed smooth. This arrangement of the design is a step beyond that seen in Stelte 22 and 25 (Class i), where only one face, i. e., the front, is sculptured with glyphs, the remaining three faces being left plain; and somewhat less advanced than Stela 20 (Class 3), where all four faces are sculptured. On the basis of arrangement, therefore. Stela 24 may be assigned to a new class, 2. The circumstances surrounding the discovery of this fragment are of especial interest because of its unusual importance, already noted, no less than that of presenting the earliest date yet deciphered at Copan about which there can be no doubt. In August 1916 the writer received a letter from Copan stating that a " piece of stone inscribed with hieroglyphics" had been found recently during the course of some excavations in the village; but it was not until May of the following year that he had an opportunity of examining this fragment at first hand, and of ascertaining the circumstances which surrounded its discovery. During the early part of the summer of 1916, Clementino Lopez, living near the southwestern corner of the village plaza (see figure 22, F), was dig- ging a well in the yard behind his house, and required some stone with which to line it. In the middle of this yard there is a low mound of earth and stone 70 cm. high, 27 meters long north and south, and 17 meters wide east and west, near the eastern edge of which Maudslay had found Stela 7 lying in 1885.^ (See figure 18, b.) Lopez had dug into this mound for stone to line his well, and at a depth of about half a meter below the surface he found a pila or cylindrical altar with a depression in its top (figure 12, d). This is 46 cm. high, 56 cm. in diameter at the top, and tapers toward the bottom, being 39 cm. in diameter at the base. Just below this altar was found the fragment of Stela 24 shown in figure 13 (for its position, see figure 12, h), which in turn rested directly upon a large, plain rectangular slab of stone 1.62 meters long (north and south), 96 cm. wide (east and west), and 29 cm. thick (figure 12, a). Above this slab and closely packed in around the altar and stela-fragment were many 'See pp. 102, 103. INSCRIPTIONS OF THE EARLY PERIOD. 79 small broken stones laid in lime mortar, the remains of the two pavements of lime-plaster and the strata of small broken rock of which the mound of Stela 7 was composed. (See figure ig, b, c, d, f, and g.) There is little doubt but that this slab is in situ and was the foundation- stone upon which Stela 7 had originally rested. (See pp. 103-105.) All circumstances point to this conclusion. In the first place, Stela 7 was found lying on the surface of the ground only 2 meters from this slab, presumably just where it had fallen. Again, this foundation-slab is of the same general size and shape and was found buried about the same distance below the surface as others used for this same purpose elsewhere; and finally, certain measurements on Stela 7 itself tend to confirm this conclusion. Fig. 12. — Sketch showing position of fragment of Stela 24 found in 1916, resting on foundation-stone of Stela 7: a, founda- tion-stone of Stela 7; h, fragment of Stela 24; c, Stela 7; (/, small altar with depression m top. Just below the sculptured panels on Stela 7 there is a shallow groove which runs clear around the monument, the bottom of which is 90 cm. above the bottom of the monument and 33 cm. below the glyph-panel. The gener- ally accepted explanation of this is that the stone or concrete flooring of the court where this stela stood fitted into this groove and tended to lock the monument more securely to its foundation. If this explanation is correct, the floor-level of the court where Stela 7 originally stood fell somewhere between the bottom of this groove and the bottom of the glyph-panel when the monument was upright, /. f., between 90 cm. and 1.23 meters from the bottom of the monument. Returning once more to the foundation-slab under Stela 24, its top surface was tound to be 1.02 meters below the present level of the ground, that is to say, the top of 8o THE INSCRIPTIONS AT COPAN. this groove would have come just where the flooring of the court touched the stela. (See figure 19 for a cross-section showing these details.) The provenance of Stela 7, the discovery of a foundation-stone just below where it was found, and finally the above measurements, leave little doubt that this monument formerly stood here and, more important still, that a fragment of Stela 24 was reused in its foundations. Before developing this point any further, let us first examine the inscrip- tion on Stela 24. This is presented on the front and back, the former showing parts of 6 glyph-blocks (figure 13, a), and the latter parts of 5 (figure 13, b). The glyphs on the front are as clear as though they had just left the sculptor's hands, and they unmistakably record the date 9.2.10.0.0 3 Ahau 8 Cumhu. (See figure 13, a.) Fig. 13. — Inscription on Stela 24: a, front; b, back (only one fragment recovered). The first glyph-block on the front records 9 cycles, the line of fracture passing through the upper part of the block. The upper dot and the upper end of the bar of the coefficient are missing, but judging from the height of the complete glyph-block just below, 18 to 19 cm., the coefficient originally recorded could have been none other than 9. Since this fragment begins with the cycle-sign and coeflScient, it is evident that the Initial Series introducing glyph occupied the space of four glyph-blocks above it, i. e., ai-B2, and that including the missing parts of the cycle and katun-signs and coefficients and the plain band across the top, the stela originally extended another 49 cm. above the top of the fragment found. The katuns follow the cycles in b3. The coefficient is clearly 2, in spite of some loss of detail and of the upper half of the upper dot. The lower dot has the same interior treatment, i. e., parallel lines, as the dots of the cycle coefficient. The period glyph is also partially effaced. The next glyph-block, a4, is perfect, and unmistakably records 10 tuns. The upper left hand corner of the next block, B4, is eff^aced, but both the period glyph and coefficient are clear as o uinals. MORLEY PLATE 9 Stela 18. (-'^J^ (-'1.,. ,■ -'.4- 'iSry^l^-'^'V^^^'^ Stela 21. (c) Back.? (d) Side, (e) Front.? mm h. Altar O'. (g) Front, (f and h) Sides. INSCRIPTIONS OF THE EARLY PERIOD. 8l The kins, as, are all gone, except the upper right-hand corner of the period glyph; however, the day-sign in the next block, bs, is so clearly Ahau that the missing kin coefficient may be safely restored as o. The day-sign coefficient at first appears confusing. The lower left-hand corner of this glyph-block is missing, and with it has gone part of the coeffi- cient. The same ornamental element seen in the day-signs on Stelae 25, 15, and 9 — of all the other stelae the most closely related to Stela 24 both in point of time as well as in style — is also found in the upper left-hand corner of this glyph-block. The coefficient of the day-sign is below this. One dot and part of another are still preserved, and to fill out the glyph-block properly it is necessary to postulate the former existence of a third of equal size below these two, indicated in dotted line in figure 13, a. The upper dot, moreover, shows the same use of parallel lines in its interior decoration as do the dots in the cycle and katun coefficients above, and it is therefore evident that this coefficient must have been either 2 or 3, i. e., two numerical dots with an ornamental element between or three numerical dots. Referring to Goodman's tables, it will be found that the Initial Series number 9.2.10.0.0 leads to the terminal date 3 Ahau 8 Cumhu, and the day recorded in b5 is therefore 3 and not 2 Ahau. The Supplementary Series and the month 8 Cumhu were recorded on the next piece below, but this is still missing: Fragment i missing a i-B 2 = Initial Series introducing glyph Fragment 2 A3= 9 cycles Fragment 2 83= 2 katuns Fragment 2 A4=iotuns Fragment 2 84= o uinals Fragment 3 missing A5= (o kins) Fragment 2 35= 3 Ahau Fragment 3 missing (8 Cumhu) The inscription on the back (figure 13, b) presents no decipherable glyphs. D3 has a well-known ending prefix and C5 is a beautifully executed death's head, having all the well-known Maya skeletal characteristics — the fieshless lower jaw, the prominent upper teeth, the truncated nose, the large bony eye-socket, here filled with what may be a realistic attempt to represent the cranial sutures,^ and the many small spots, almost invariably associated by the Maya with death. The very early date recorded upon Stela 24 — the earliest surely deciph- ered at Copan, and 3 tuns earlier than the earliest yet found at Tikal, her great northern rival — makes this monument one of the most important of the whole Corpus Inscriptionum Mayarum. Very few indeed are of greater antiquity or even equal age; hence the stress here laid upon the circum- stances attending its discovery. 'The resemblance of this eye to the sign for the day Eznab is so obvious as to suggest that there may have been some connection between the two. The sign for Eznab may be a representation of the cranial sutures, par- ticularly since these lines are wavy. See Bowditch, 191c, plate 6, Motley, 1915, figure 16, and Appendix X. This day is very rarely recorded in the inscriptions and its avoidance may have been due possibly to its resemblance to a death's-head characteristic and the consequent association of the idea of death with it, of which the Maya stood in great fear. 82 THE INSCRIPTIONS AT COPAN. It was erected, as we have seen, on the third lahuntun-ending of Cycle 9, viz, 9.2.10.0.0, but where can not be estabhshed. It would seem probable, however, that it must have stood at or very near the place where it was found, namely, at Group 9, where the modern village now stands, which, as will appear later, was probably the earliest permanent settlement in the valley. Some 128 years later, in 9.9.0.0.0, we find a piece of it, indeed the most important piece of all, since it carries the dedicatory date, was built into the foundations of Stela 7. This practice of reusing earlier monuments in the foundations of later ones was fairly common at Copan in the Middle and Great Periods, but so far as the writer is aware, this is not only the earliest example of this practice but also the only example yet reported from the Early Period. What is the explanation of this custom? Was it in compliance with some religious concept, or had it no other foundation than the purely utili- tarian purpose of making use of old monuments which had outgrown their usefulness? Spinden believes that it may have had a ceremonial significance and is possibly another expression of the custom of placing caches of bowls, vases, etc., in the vaults or chambers under the monuments, as in the cases of Stelae 3, i, I, M, C, and 4, the latter, perhaps, analogous to our own cus- tom of depositing current periodicals, coins, photographs, and the like in corner-stones. It appears to the writer as not improbable that such reuse of their ear- lier monuments by the Maya may have been made for some religious pur- pose; in short, that such a practice was "good medicine." If, for example, the lahuntun ending in 9.2.10.0.0 had been a particularly prosperous one, in which the crops were unusually fruitful, it would not be an unnatural or illogical assumption for the primitive mind that the monument which had been erected to commemorate this particular period had partaken of its beneficent character; and further, that if such a monument were placed in the foundation of a later one, a like period of prosperity and plenty might be made to follow. In the present case, the fact that the piece of the monu- ment used for this purpose was the one upon which the date was recorded might be interpreted as being due to deliberate selection, and thus lend color to the idea; but amid such speculations the purely utilitarian explana- tion should not be overlooked, namely, that by 9.9.0.0.0, Stela 24 may have outlived the purpose for which it was originally designed, and being a selected piece of stone in the first place, large as well as strong, and more- over already shaped and dressed, it was used in the foundations of Stela 7 as being convenient, suitable and available for that purpose.^ Since this fragment of Stela 24 was buried only 128 years after it had been carved, its remarkable preservation is easily explained. The detail, at least on the Initial Series side, seems to have preserved most, if not all, of its 'Other examples of this custom will be found elsewhere at Copan as follows: Altar J', p. 56; Altar K', p. 56; Altar X, p. 63; Altar Y, p. 66; Stela 9, p. 93; Stela 21, p. 95; Stela 22, p. 69; Altar A', p. 68; and Fragment S', p. izi. INSCRIPTIONS OF THE EARLY PERIOD. 83 original clearness and brilliancy, traces of the red paint with which it was originally covered still adhering in some places. The relief is uniformly i cm. deep, and consideiing the extremely early date at which it was executed is fine work. Although low and flat, it possesses strength and character and already indicates an almost perfect control of the medium, much more so, for example, than the work being done at Tikal at the same time, i. e.. Stela 3. The glyphs on Stela 24 have already assumed the rectangular outline which was to characterize Maya glyphic delineation throughout the Old Empire, but which did not appear at Tikal until later. We have already seen that there is only one other monument now known at Copan which has glyphs of an earlier and less rectangular character, namely. Stela 20 (9.1.10.0.0). The fact that this irregularity of outline had disappeared at Copan by the time Stela 24 was erected, probably only 20 years later, therefore tends to authenticate on stylistic grounds the fact that Stela 20 was the earlier of these two. Another early feature of this inscription is the lack of specialization in the essential characteristics of its period glyphs. The cycle-sign lacks the hand on the lower part of the face; the katun-sign, the oval in the upper part of the head; and the tun-sign, the fleshless lower jaw. Indeed, the only period-glyph which would appear to have developed its distinguishing charac- teristics as early as this is the uinal-sign, which is the full figure of a toad. The uinal would seem to have been the first period to have acquired special characteristics. In the Leyden Plate Initial Series the uinal-sign is the only one of the five period-glyphs which has the same essential element as that by which it was recognized in later times. This lack of specialization is in itself an indubitable mark of antiquity, since it indicates that at the early date this inscription was carved the period-glyphs, with the exception of the uinal-sign, had not yet developed the special characteristics by which they were severally known later. A close similarity in a minor detail between Stela 24 at Copan and Stela 3 at Tikal should be noted here, namely, that the day-sign cartouche in the Initial Series terminal date of each has a pair of small tassel-like protub- erances, one from each of the upper corners. In the Copan glyph there is another in the lower right-hand corner as well. The presence of this minor detail — and since it is repeated nowhere else in the whole range of the Maya writing we must believe an adventitious one as well — at two such widely separated cities as Copan and Tikal argues for close intercourse within the area at a very early date, and a correspondingly early homo- geneity of culture. The question as to which of these two cities was the older will be fully presented in Chapter V, and that discussion will not be anticipated here, except to note that although Stela 24 at Copan is earlier than any date yet deciphered at Tikal, and although Stela 20 is probably still older, there are strong grounds for believing that the northern metropolis is probably the older of the two. 84 the inscriptions at copan. Stela i6. Provenance: Original position unknown. Placed by the Second Pea- body Museum Expedition at the head of Owens's grave in the Great Plaza at the Main Structure in 1893. (See plate 6.) Date: 9.4.9.17.0 5 Ahau 8 Yaxkin (.?) or 9.7.2.12.0 5 Ahau 8 Yaxkin (?). Text, {a) photograph: plate 10, a and b. {h) drawing: plate 10, a and b. No reference is made in any of the Peabody Museum publications to this sculptured fragment, although it is the upper part of a very early stela. Neither has it been possible to ascertain just where it was found. When first examined by the writer in 1910, it was standing at the head of Owens's grave in the Great Plaza, though obviously not in situ here. A sister monu- ment, Stela 17 (see p. 90) was found on the terrace just north of the Great Plaza, a few yards west of Mound 2 (see plate 6), which suggests that Stela 16 also may have come from somewhere in this general vicinity. The fragment preserved is i meter long, 63 cm. wide, and 42 cm. thick. It is sculptured with glyphs on both broad faces, the narrow sides remaining plain. On the basis of arrangement of design, therefore, it may be assigned to Class 2. Both inscriptions open with large Initial Series introducing glyphs A1-B2 and CI-D2; the variable element on the side not presenting the Initial Series is the normal form of the kin-sign.^ The text opens with an Initial Series introducing glyph (see plate 10, a) in AI-B2, which is followed by an Initial Series number in a3-b5. Unfor- tunately almost all the carving on this side has scaled off, carrying with it the details of these glyphs. Enough remains to distinguish faint traces of the cycle coefficient in a3<3 and the katun sign in b3^, but not enough to decipher the date. The line of fracture runs across the tun and uinal signs; consequently, the Initial Series terminal date — if it were recorded on this side — is missing. Happily, the back of the monument is in a better state of preservation (see plate 10, b). The text on this side also opens with an Initial Series intro- ducing glyph followed by a Calendar Round date, which, so far as the writer knows, is unique throughout the entire range of the Maya inscriptions in having its month-sign precede its day-sign. As recorded, this date very clearly reads 8 Yaxkin 5 Ahau, which can hardly be other than an inversion of s Ahau 8 Yaxkin. This date occurred four times in the Early Period, as follows: 9.1. 17. 4.0 9.4. 9.17.0 9.7. 2.12.0 9-9-I5- 7-0 'It should be noted in this connection that the variable elements in the introducing glyphs of the Initial Series on the Leyden plate and also on Stela 23, at Santa Rita (Group I, plate 3), and Dates 20 and 28 of the Hiero- glyphic Stairway, are the same. INSCRIPTIONS OF THE EARLY PERIOD. 85 The first and last of these, however, may be elirainated at once, on the ground of historic and styhstic improbabiUty, the first being as much too early as the last is too late. Of the two remaining dates, 9.4.9.17.0 has more in its favor than 9.7.2.12.0 for the following reasons: 1. It is just I uinal or 20 days earlier than the end of a tun, hotun, and lahun- tun in the Long Count, and 2. It is within i uinal or 20 days of the date of another monument here, Stela 15 (9.4.10.0.0). On the other hand, the later value would be nearer the date of Stela 17, which is some time in Katun 6, and which on stylistic grounds this monu- ment more closely resembles than any other. An entirely different explana- tion, however, is possible here, although one less likely to be correct than the preceding, since it necessitates a forced reading of the original before it becomes possible at all. In plate 10, b, the glyph immediately following 5 Ahau in A3 has a co- efficient of 5 and a hand at the right side. Can this possibly signify that the preceding inverted Calendar Round date was a hotun-ending, the head between the 5 and the hand in Bj being some unknown early form of the tun-sign ? Referring to Goodman's tables, it will be found that there are only three hotuns in the Early Period which ended on the day 5 Ahau, as follows: 9.1.10.0.0 5 Ahau 3 Tzec; 9.4.15.0.0 5 Ahau 3 Yaxkin, and 9.8.0.0.0 5 Ahau 3 Chen; and of these, the only one at all possible here is the second, which agrees with the date in A3, except for its month coefficient, which is 3 instead of the 8 recorded. It is barely possible, though not probable, that the bar in the month coefficient of A3 is ornamental and not numerical, and that this date is 5 Ahau 3 Yaxkin and not 5 Ahau 8 Yaxkin, as it first appears to be. If this is true the corresponding Initial Series is almost certainly 9.4.15.0.0. Against this reading, however, must be offset the fact that the month- coefficient certainly looks more like 8 than like 3, and also that using it as 3 gives a third hotun as the resulting date, i. e., 9. 14. 15. 0.0, which would be by nearly a century the earliest example of the record of a quarter katun yet found at Copan. Indeed, as the several monuments of the Early Period are described hereinafter, it will be seen that at first the custom seems to have been to commemorate only the lahuntuns and the katuns by the erec- tion of stelae, and that it was not until the very end of the Early Period that the quarter katuns were similarly commemorated.^ Before attempting to decide, even tentatively, however, upon any of these dates, it is best to describe Stela 17, the sister monument, after which a better choice can be made.' ■See p. 126. -For the discussion of Stela 17, see pp. 89-93. §5 the inscriptions at copan. Stela 15. Provenance: Found on the mound of Stela 7 (Group 9). Now in the cabildo. (See plate 3 and figure 22, s.) Datg. 9.4.10.0.0 12 Ahau 8 Mol Text, (fl) photograph: Morley, 1915, plate 13. Spinden, 1913, plate 23, 2. (b) drawing: plate 12. References: Morley, 1915, pp. 187, 188. Spinden, 1913, pp. 160, 163, 164, and table i. The vicissitudes through which Stela 15 has passed during the last three-quarters of a century well illustrate what has happened to many of the monuments of Group 9, where the modern village now stands. When the writer first visited Copan in 1910 the two larger pieces of this monument (fragments i and 2) were built into the east wall of a house at the northwestern corner of the village plaza. (See figure 22, J, and ti.) Infor- mation received then and in 1912 led to the discovery of a third and smaller piece (fragment 3) in 1915 (see figure 22, r), and during the writer's last visit additional data as to the history of this monument were secured. Mariano Madrid, the father of Jacobo Madrid, first came to Copan in 1891, when he bought the property on the southern side of the plaza at the southwestern corner, from Anita Acevedo. (See figure 22, G.) At that time this property was surrounded by a stone wall, the only one of its kind in the village, which was still only an aldea or hamlet, the municipality not being organized until two years later.^ Jacobo Madrid states that fragments i and 2 were built into the founda- tions of the stone wall along the eastern side of this property at that time. (See figure 22, t.) Three or four years later (1894 or 1895) his father built the house at the northwestern corner of the plaza, and in order to secure large stones for the foundations of its adobe walls he tore down this stone wall and removed these two fragments to this other house, where the writer first saw them in 1910 (see figure 22, u), whence they were subsequently removed to the cabildo in 1913. But Anita Acevedo was not the original owner of the property where these fragments were first described, nor was her husband, Juan Villeda, the builder of the stone wall where they were found. Maria Melendrez, one of the oldest inhabitants, states that Anita Acevedo bought this property from an Ana Carlos Orellano about 1865 (f. e., when the informant was 16, being about 70 now) and that the stone wall was already built when she first remembers this place. Cristina Ramirez, the oldest inhabitant of the village, has recollections of still an earlier period. She states that as a child she was accustomed to go iThemunicipalltybf Copan was organized on January I, 1893, by the following men: Teodoro Destephen, InJalecio Guerra (alcalde), Emilio Cuellar (first regidor), Cristobal Melendez (second regidor), Mariano Madrid (sindico), and J. Manuel Collar (secretario interino), Guadelupe Lopez being the first regular secretary of the municipality. Teodoro Destephen and J. Manuel Collar were not residents of Copan, but signed the organiza- tion papers, the former in his capacity as commandante and the latter as secrelario of Santa Rita, under the juris- diction of which Copan had been heretofore. INSCRIPTIONS OF THE EARLY PERIOD. 87 to this house to buy cheese and remembers distinctly when the stone wall in question was being built by the husband of Ana Carlos Orellano, Domingo de Aguilar. She recalls the laborers carrying the stone that went into it, but not the fragments of Stela 15 specifically, although she remembers them afterwards as having been in the wall. When questioned as to her age at this time, she replied that she was still a little girl. If she is now 75, and her grandchildren believe she can be no less, and if she were about 10 when the wall was being built, it would make the wall date from about 1854 or, allowing a slight leeway either way, roughly from the decade 1 850-1 860. However, even this does not establish the original provenance of Stela 15, and indeed it was only by the chance discovery of a third fragment of this monument in 191 5 that it has been possible to establish its former posi- tion at all. In examining the mound of Stela 7 in that year, the writer discovered a small fragment having only one glyph on it, which looked as though it might be a part of Stela 15. He had this carried over to the cabildo, where frag- ments I and 2 of this monument had already been deposited, and it was found to fit against the lower end of fragment 2. (See figure 22, r.) This discovery was important as establishing beyond much doubt that the original provenance of Stela 15 had been the mound of Stela 7; that fragments i and 2 had been removed from here some time between 1850 and i860 for use in the foundations of the stone wall above described; and finally, that because it was only a small and irregularly shaped piece, frag- ment 3 had been left undisturbed at the mound of Stela 7 practically in situ. When these three fragments were assembled they measured 1.89 meters long, 76 cm. wide, and 42 cm. thick. The inscription covers all four sides, each side opening with an Initial Series introducing glyph. The variable element in all of these except the one above the single Initial Series number is the same, z;^^^^^^:^ the variable element in the other being a grotesque head. On ll^Tvil \f(7^\\ the basis of arrangement of design, therefore. Stela 15 may be assigned to Class 3. Spinden was the first to call attention to the early character of this monu- ment. He suggested as its probable date 9.4.10.0.0 12 Ahau 8 Mol.^ After a close study of fragments i, 2, and 3, the last not seen by Spinden, the writer was able to corroborate this reading.'^ The single Initial Series opens with an introducing glyph in ai-b2 (plate 12). This is followed by 9 cycles in a3-b3 and 4 katuns in a4-B4. Here occurs the break described above, and then follows on fragment 2, 'Spinden, 1913, pp. 160, 163, 164. ^Since Spinden's conclusions regarding the archaic character of Stela 15 were based largely upon stylistic criteria, it is significant that the date actually recorded on this monument indicates a corresponding antiquity. This is but one of the many agreements between the chronological sequence and the stylistic sequence of the monuments throughout the Maya area, an agreement so complete in fact as to prove beyond all doubt the chrono- logical sequence of Maya art. 88 THE INSCRIPTIONS AT COPAN. lo tuns in A5-B5, and o uinals in a6-b6. The kin-sign is recorded at b7 on fragment 3 ; the kin coefficient and terminal date are missing. Arguing from antecedent probabihty in such Initial Series (f. e., where the tun coefficient is either o, 5, 10, or 15, and the uinal coefficient o), the missing kin coefficient can hardly have been other than o; and we may assume that a lahuntun-ending in the Long Count was probably recorded here, 9.4.10.0.0. That this assumption is correct, the writer was able to demonstrate by the discovery of a lahuntun-sign on fragment 2, thus con- firming this reading. An examination of the mound of Stela 7 in 19 12 had disclosed several small fragments, one of which appeared to have the coefficients of the terminal date of Stela 15, as deciphered by Spinden, i.e., 12 (Ahau)^ 8 (Mol), pt^o p Q Q the day-sign, and the greater part of the month-sign having I 1(7=:=^ been broken off. And in 1915, when the writer next visited Copan, he had this piece removed to the cabildo, where it was there found to fit exactly against the bottom of fragment 2, and was, in fact, a part of Stela 15. The front of this new piece shows the kin-sign of the Initial Series number (b7, plate 12) just below the full-figure form of the uinal, i. e., the toad or frog in b6. Curiously enough, what had been mistaken for the day and month coefficients 12 and 8 respectively of the Initial Series terminal date on the back of this fragment in 1912, on closer study in 1915 turned out to be 12 or 7 (?) 6 (.?), or the coefficients of another date, o ^ o pgna Thus the former reading, incorrect as it later proved to be, ' -i ' — ' was the means of identifying this fragment as a part of Stela 15. The last glyph on the back of fragment 2 (f6^), ^^^^ ^- ^-^ on the side opposite the Initial Series, is the well-known m ^^ sign for the lahuntun, which is used only on monuments, which lii^§^ record the ends of second hotuns (z. e., lahuntuns) in the Long Count. Its presence here renders practically certain Spinden's reading of this Initial Series. As already suggested, Altar Q' probably records this same lahuntun-ending and may have been the altar originally associated with this stela. (See p. 62.) One of the narrow faces of Stela 15 has the month-sign 13 Kayab (c3) following the Initial Series introducing glyph (ci-c2). It is impossible, however, to connect this with any other date in the text. A more interesting and perhaps a better understood lyjuji date occurs in d5-d6 on the same side, in which C6, d6 ^^M\ possibly records the Calendar Round date 8 Ahau 13 .?, C^^ the month-sign being a human head. d5 appears to be a tun-ending sign, although this identification is not certain. Assuming that the date 8 Ahau 13 1 ended some even tun of the Early Period, it can be found from Goodman's tables that there are only six 'Matter inclosed in parentheses does not appear in the text as preserved. Such omissions are due to one of two causes: either the missing part may have been effaced or broken ofT, as here, or, as in other cases, it may have been understood, i. e., suppHed mentally, without the necessity of actually recording it. INSCRIPTIONS OF THE EARLY PERIOD. 89 places where the conditions imposed by these two coefficients are fulfilled, namely: 9.0. 0.0.0 8 Ahau 13 Ceh 9.6.10.0.0 8 Ahau 13 Pax 9.3. 5.0.0 8 Ahau 13 Kankin 9.9. 2.0.0 8 Ahau 13 Zip 9.5.17.0.0 8 Ahau 13 Pop 9.9.15.0.0 8 Ahau 13 Cumhu But D6 more closely resembles Ceh than it does any of these other month-signs, actually having the superfix of Ceh as its own superfix here (compare d6 with the forms for Ceh in Appendix X), and this Calendar Round date therefore probably stands for the full Initial Seiies 9.0.0.0.08 Ahau 13 Ceh, i. e., the end of Cycle 9. This is the second example of the month-sign being a human head. (See also the month-glyph, 8 Cumhu on Stela 25, figure 8, and p. 71.) Can it be possible that these heads are those of the deities who presided over these months? Whether so or not, this is an interesting parallel between Stelae 25 and 15 and may foreshadow the discovery of other examples of the same kind. When Stela 15 was erected, less than 90 years had elapsed since this important date, which must have been held in very lively memory still; at all events it appears to have been made a matter of record here. The whole inscription on Stela 15 so far as it has been deciphered reads: Front. Fragment i A1-B3 Initial Series introducing glyph A3, B3 9 cycles A4, B4 4 katuns Fragment 2 A5, B5 10 tuns a6, b6 o uinals Fragment 3 B7 (o) kins Side. Fragment i C3 13 Kayab Fragment 2 d; "End of a tun or a hotun" (?) Fragment 3 C6, d6 8 Ahau 13 Ceh (9.0.0.0.0) Back. Fragment 2 ¥6b "End of a lahuntun," probably referring to the con- temporaneous date of the stela on the other side. Because of the fragmentary condition of Stela 15, it is impossible to more than approximate the number of glyph-blocks which its inscription originally contained, but the three fragments preserved have upward of 40. Stela 15 dates from the fifth lahuntun of Cycle 9, being exactly 2 katuns later than Stela 24. Stela 17 Provenance: Original position unknown. Found just west of Mound 2 on the low terrace north of the Great Plaza at the Main Structure. Now near Owens's grave in the Great Plaza. (See plate 6.) Date: 9.6.0.0.0 9 Ahau 3 Uayeb (.''). Text, (a) photograph: plate 11, a and b. {b) drawing: plate 11, a and b. 90 THE INSCRIPTIONS AT COPAN. Only one fragment of Stela 17, the sister monument of Stela 16, has been found, and even this small piece has been reshaped for some secondary purpose. It was discovered by Morris in May 1912, at the western base of Mound 2, on the low terrace north of the Great Plaza, while he was searching for the missing parts of Stela 16. It is 63 cm. long, 38 cm. wide, and 27 cm. thick. The fragment preserved is the upper right-hand corner of the monu- ment as one faced it (see plate 11, a). In size, style, and treatment it very closely resembles Stela 16. Both are executed in the same low relief, little more than incised lines; and both have the same type of Initial Series introducing glyphs at the tops of their broad faces. The variable central element of the Initial Series introducing glyph on the back of Stela 16 is the kin or sun, the corresponding element on the front of Stela 17 being the moon. In short, a very close similarity exists between the two monuments, and on the basis of the arrangement of its design it has been assigned to Class 2. The inscription on the front opens with an Initial Series introducing glyph in ai-B2, which was originally followed by the corresponding Initial Series number in a3-a5, all of which is now missing except the katun-sign and coefficient in b3 and part of the cycle-sign in A3. The most important single glyph in determining the age of any Maya monument is the katun-sign and coefficient, since by means of this character alone any date can be deciphered to within 20 years of its true position at the outside. Most fortunately, in this text the katun coefficient is unusually clear as 6. (See b3, plate 11, a.) This glyph alone enables us to place the date of Stela 17 as falling somewhere in the katun or 20-year period from g. 6.0.0.0 to 9.7.0.0.0. Unfortunately, the inscription on the back has almost entirely dis- appeared, and except for the Initial Series introducing glyph in ci-d2 no other signs can be deciphered. Although it is impossible to go beyond this point with certainty because the rest of the inscription is missing, there are several factors which make it probable that Stela 17 dates from 9.6.0.0.0. As explained in Appendix VII, the overwhelming majority of Maya stelae were erected, or at least formally dedicated, at the expiration of hotuns, lahuntuns, or katuns in the Long Count. Because of this fact, therefore, we are justified in assuming that the Initial Series of Stela 1 7 recorded one of the four following dates in Katun 6 out of the 7,200 dates, which must be admitted as possibilities without this assumption:^ 9.6. 0.0.0 9 Ahau 3 Uayeb 9.6. 5.0.0 2 Ahau 18 Kayab 9.6.10.0.0 8 Ahau 13 Pax 9.6.15.0.0 I Ahau 8 Muan In the very early period from which this monument dates, the custom of erecting a stela at the expiration of every hotun does not appear to have prevailed. Possibly at this early date the priests, or those in whose hands 'Since there are 7,200 days in any given katun, there are 7,200 possible dates here. INSCRIPTIONS OF THE EARLY PERIOD. 9I the matter rested, may not have felt equal to such a task. At all events, it is not until 9.9.5.0.0 at Copan (45 to 60 years later than Stela 17) that the attempt first appears to have been made to mark the end of every hotun, and even after this the sequence is incomplete. It is also explained in Appendix VII that the second and fourth hotuns were frequently com- memorated by the erection of stelae when the first and third were not, because the former coincided with lahuntun and katun-endings respectively, and were consequently of greater importance than the latter. This seems to have been particularly true of Copan, since none of the four earliest stelae here, which have been surely deciphered. Stelae 24, 15,9, and 7, date from a first or a third hotun. On the contrary. Stelae 24, 15, and 9 date from second hotuns (lahuntuns), and Stela 7 from a fourth hotun (a katun). Moreover, in the cases of the three doubtfully deciphered monu- ments previously encountered. Altar Q' and Stelae 20 and 25, the best readings for these are also lahuntuns, 9.4.10.0.0, 9.1. 10. 0.0, and 9.4.10.0.0 respectively; indeed, the lahuntun-sign actually appears in the first. (See figure 6, A3^.) Therefore, a priori, it is probable, that Stela 17 dates from either 9.6.0.0.0 or 9.6.10.0.0 rather than from the first and third hotuns of Katun 6. But the second one of these two dates, 9.6.10.0.0, is surely recorded elsewhere here at Copan, namely, on Stela 9; and rather than accept the conclusion that this particular hotun was commemorated by the erection of two stelae, when so many of the earlier hotuns are known to have had none at all, the writer prefers to accept the other reading, 9.6.0.0.0, as the date of this monument. Moreover, this latter date has two other minor points in its favor. It is not only a katun-ending, but it is also within 102 days of the best date for Altar X, with which Stela 17 may have been formerly associated. In spite of its fragmentary condition, this monument may be surely assigned to Katun 6; and because of the fact that the great majority of all Maya stelae record hotun-endings, it may probably be assigned to one of the four hotuns given above, with the first as the best possibility. We are now in possession of sufficient data to resume the consideration of the relative ages of Altars X and Y and Stelae 16 and 17, the possible dates for which follow: Altar X Altar Y Stela 16 Stela 17 9.3. 6. 17. 18 9.5.19.12.18 9.8.12. 7.18 9.4. 8.12.6 9.7. I. 7-(> 9.9.14. 2.6 9.4.9.17.0 9.7.2.12.0 9.6.0.0.0 to 9.7.0.0.0 It is unfortunate that the exact date of Stela 17 is unknown, since it may have stood in the same close relation to Altar X as Stela 16 appears to have stood in relation to Altar Y. Indeed, the dates of the two latter monuments, no matter which set of readings be selected, are within two years of each other. 92 THE INSCRIPTIONS AT COPAN. If such a relationship did exist between Stela 17 and Altar X, it is highly probable that the date of Altar X is the second value given above, namely, 9.5. 19. 12. 18, since this value would be within 6 uinals, i. e., 102 days of the best date for Stela 17, and can not be farther off than 20 years, whereas the nearest that the first or third values above can be to the nearest possible readings of Stela 17 is 53 years before or 52 years after respectively. Again, on stylistic grounds, 9. 5. 19. 12. 18 is a better date for Altar X than either 9. 3. 6. 17. 18 or 9. 8. 12. 7. 18. In short, all factors considered, 9. 5. 19. 12. 18 seems to be the best date for Altar X, and 9.6.0.0.0 the best date for Stela 17. And finally, because these two dates are so close together, it is probable that Altar X was formerly associated with Stela 17. The real crux of this question is in regard to the dates of Stela 16 and Altar Y, that is, which of the two sets of values given above is to be asso- ciated with them, 9.4.9.17.0 and 9.4.8.12.6 or 9.7.2.12.0 and 9.7.1.7.6; and since either set appears to be equally possible, chronologically considered, it is necessary to turn to the stylistic criteria present in order to determine the relative sequence of these four monuments. Unfortunately, there is little help coming from this direction either, as between Altars X and Y there appears to be but little choice. The glyphs on Y are perhaps a shade better than those on X, but, to offset this, the human figures on X seem slightly more naturalistic than those on Y. Indeed, on stylistic grounds it is all but impossible to say which is the earlier of the two. As between Stelae 16 and 17 the choice is little better, but here we have another avenue of approach. Since the date of Stela 17 is fixed certainly to Katun 6, and probably to 9.6.0.0.0, the question is to determine whether Stela 16 is earlier or later than this date by means of a comparison of its style with the styles of the next earlier and later monuments in the sequence. The stela next earlier than Stela 17 is Stela 15 (9.4.10.0.0), and the one next later, Stela 9 (9.6.10.0.0); and our next step, therefore, is to compare the style of Stela 16 with the styles of these two monuments. The Initial Series introducing glyph of Stela 16 is more like that of Stela 9 than that of Stela 15, and in general its closest stylistic affinities may be said to be later rather than earlier, {. e., with Stela 9 rather than with Stela 15, and since the date of Stela 17 can be only 10 years earlier than the date of Stela 9 at the outside, and may indeed even be the same, it seems probable, all things considered, that the later date, i. e., 9.7.2.12.0, is the better one of the two for Stela 16. It should be remembered, however, that the earlier date is almost equally asgood,andthesequencegivenbelowisby no means certain: Altar X 9.5. 19. 12. 18 Stela 17 9.6. 0. 0. Altar Y 9.7. I. 2. 6 Stela 16 9.7. 2.12. All these monuments except Altar X were found at the Main Structure, and all but Stela 16 were in positions clearly indicating secondary usage. INSCRIPTIONS OF THE EARLY PERIOD. 93 While the point escapes definite proof at this time, it seems not improbable that Altar X may formerly have been associated with Stela 17, and Altar Y with Stela 16, and finally, that the two former may possibly have ante- dated the two latter. Stela 9. Provenance: Original position unknown. Found in the modern cemetery (Group 10) about i kilometer southwest of the village (Group 9), serving as one of the two stones which had supported Stela 8. Destroyed in 191 2. (See plate 3.) Date: 9.6.10.0.0 8 Ahau 13 Pax. Text, (a) photograph: Maudslay, 1889-1902, vol. i, plate 109. Spinden, 1913, plate 20, 5. (b) drawing: Maudslay, 1 889-1902, vol i, plate no. Morley, 1915, plate 8, B. References: Bowditch, 1901a, pp. 136, 137. Bowditch, 1910, p. loi and table 29. Gordon, 1896, p. 38. Maudslay, 1889-1902, vol i of text, p. 68. Morley, 1910a, pp. 196, 198, 199, 204. Morley, 1915, pp. 171-173. Spinden, 1913, p. 160 and table i. Stela 9 was found lying prostrate in the modern cemetery about i kilo- meter southwest of the village, where, according to Gordon (1896, p. 38), it had been reused in the foundations of Stela 8. Although obviously not in situ in this position, it seems safe to assume that it had originally stood nearby. It was intact in 1910, when the writer first visited Copan, but when Spinden was there two years later he found that, together with Stela 8, it had been broken up to form the foundation of an adobe wall then in course of construction around the cemetery.^ This unpardonable act of vandalism, although greatly to be lamented, is not irreparable, since excellent casts exist of both the destroyed monuments. The writer, however, regards himself as particularly fortunate in having been able to study the original before its destruction. Stela 9 was 2.44 meters long, 6y cm. wide, and 39 cm. thick. Maudslay says that all four sides had originally been sculptured, but that the carving on the fourth side was almost entirely destroyed.- Spinden believes all four sides were sculptured, and that the effaced side, which has never been reproduced, possibly presented a full-length human figure front view, like the one on Stela 18. (See p. 97.) The writer, on the other hand, believes all four sides were originally sculptured, with glyphs like Stela 15, the style of which it very closely resembles, and on the basis of this arrangement of the design it may be assigned to Class 3. The inscription opens with an Initial Series on one of its broad faces. This is unusually clear for such an archaic text, the numerals all being in the 'Spinden, 1913, p. 160, footnote i. "Maudslay, 1889-1902, vol. 1 of text, p. 68. 94 THE INSCRIPTIONS AT COPAN. bar-and-dot notation. It unmistakably records the date 9.6.10.0.0 8 Ahau 13 Pax, as follows: A1B2 Initial Series introducing glypii A3 9 cycles B3 6 katuns A4 10 tuns B4 o uinals AS o kins B5 8 Ahau B8 13 Pax AlO may be the hotun glyph; and the rest of the text shows other famil- iar signs, although of unknown meaning. The tops of both of the narrow faces are destroyed, and it is now difficult to say whether Initial Series intro- ducing glyphs had originally stood here or not. Judging from what is left of the top glyph (02) on one side, they had not stood here.^ Assuming that the effaced side (the broad side opposite the Initial Series) opened with an Initial Series introducing glyph, the number of glyph-blocks originally carved on this monument was 19+19 + 22 + 22 = 82, each introducing glyph occupying the space of four glyph-blocks. Stela 9 is exactly 2 katuns later than Stela 15, which it closely resembled, and commemorated the seventh lahuntun of Cycle 9. The relief, though very low, is elaborately executed, the glyphs showing a wealth of minute detail scarcely equaled, even in the Middle and Great Periods. This complexity appears especialy in the treatment of the numerical bars, all of which present interior decoration and have square corners, thus : [f r mi m In later times the bar is left undecorated, the corners are rounded, and the whole element is made narrower. As already pointed out, Stela 9 is very similar in style to Stela 15. Indeed, both appear to have been products of the same school, possibly indeed of the same hand, although the latter is hardly likely, since an interval of 40 years separated the two monuments. Both are of about the same width and thickness. Stela 9 being 69 cm. wide and 39 cm. thick, while Stela 15 is 76 cm. wide and 42 cm. thick. Stela 9 is about 2.5 to 3 meters long and the fragments of Stela 15 pre- served are 1.89 meters long. To this latter measurement, however, must be added something to compensate for the missing bottom-piece, probably at least half a meter. In style the two are perhaps even more closely connected, glyphic details in particular being very similar. Thus, for example, the proportions of the different elements of the Initial Series introducing glyphs on the back and sides of Stela 15 are the same as those in the one on Stela 9, the comb- like lateral appendages in the former being almost identical with the same elements in the latter. Even the variable elements in the Initial Series 'In Maudslay, 1889-1902, vol. I, plate no, c, enough remains of D2 to show that it had never been an Initial Series introducing glyph. Had it been, it is probable that it would have occupied the same space as the Initial Series introducing glyph on the front, namely, AI-B2. That this was not the case on this side, D2 clearly indicates. INSCRIPTIONS OF THE EARLY PERIOD. 95 introducing glyphs on the two fronts appear to have been the same grotesque head.^ Again, the element preceding the day-sign of the Initial Series terminal date in b5 on Stela 9, ^^^^M is almost exactly the same as the element (^H51 preceding the day-sign of a date in c6, d6 on Stela 15. Finally, in treatment, i. e., having all four sides inscribed with glyphs (Class 3), the two monuments are identical. All these close similarities can not be the result of chance and are to be explained only on the grounds that both monuments date from the same general period and are the work of the same school, and possibly, as suggested, even of the same hand. Stela 21. Provenance: Date: Original position unknown. Found in the center of the mound at the southeastern corner of the village plaza (Group 9). Now in the cabildo. (See plate 3 and figure 22, k). 9.6.0.0.0 9 Ahau 3 Uayeb to 9.7.0.0.0 7 Ahau 3 Kan- kin. Text, (a) photograph: plate 9, c, d, e. (b) drawing: figure 14. While the writer was at Copan in March 1916, the villagers were demol- ishing the mound at the southeastern corner of the village plaza, on top of which Altar S was found (see p. 226), in order to secure paving-stones for the streets. During the course of this work a fragment of an archaic stela was found buried in the hearting of the mound, to which the number 21 was given. (See figure 22, k.) This fragment measures 32 cm. high, 37 cm. wide, and 35 cm. thick. Prom the subject-matter presented on its three sculptured faces it is possible to estimate its original width as having been 74 cm. It is part of the top of an archaic stela which was originally sculp- tured on four sides with glyphs like Stelae 20, 15, and 9 (see plate 9, c, d, e, and figure 14) and on the basis of the arrangement of the design, therefore, it r' Fig. 14. — Inscription on front, back, and one side of Stela 21. may be assigned to Class 3. At some later time it had been broken up into pieces, and the only fragment recovered shows that it had been used again as a building-block; in the reshaping necessary for this purpose, the glyph or glyphs on one side and part of those on the front and back were broken off.^ Three Initial Series introducing glyphs, two of which are not even complete, are all that is now preserved of the inscription. (See figure 14.) 'This element on Stela 15 is partially effaced. 'Since there is an Initial Series introducing glyph on the preserved narrow face as well as upon the front and back, it is practically certain that there must have been one on the destroyed narrow face also, as in the case of Stelse 15 and 9. 96 THE INSCRIPTIONS AT COPAN. In the reshaping process about half of the Initial Series introducing glyphs on both the front and the back were destroyed. That on the preserved side is practically complete. The variable element in the Initial Series introducing glyph on one of the broad faces is unmistakably the head of God C of the Schellhas classifica- tion of Maya deities. ^ In the Initial Series introducing glyph on the narrow face this element appears to be rather more of a filling character, like the ones on the three non-Initial Series sides of Stela 15. As the period glyphs are all missing, it is impossible to date this stela exactly, although approximate dating by comparison of the stylistic criteria with those of other archaic stelae is practicable. In arrangement Stela 21 most closely resembles Stelae 20, 15, and 9, having glyphs on all four of its sides. Stylistically its closest resemblances are with Stelae 9, 16, 17, and 24. All seven of these monuments are of about the same size: Stela2l. Stela 20. Stela 15. Stela 9. Stela 16. Stela 17. Stela 24. Width Thickness. .. . cm. 74 35 cm. S3 40 cm. 76 42 cm. 69 39 cm. 6S 42 cm. S3 27 cm. 73 29 The relief on Stela 21 is low and flat (about 6 mm. high), although well executed. Traces of red paint still adhere in a few places. Since two of its three closest affinities on stylistic grounds are definitely referable to the katun from 9.6.0.0.0 to 9.7.0.0.0, it seems probable that it also may date from about the same period. This clearly established case of secondary usage in ancient times sug- gests that other archaic stelae may have suffered a similar fate, which would satisfactorily explain some of the existing lacunae in the sequence of the dated monuments. (See pp. 123, 125 for a further discussion of this point. Stela 18. Provenance: Date: Found on the mound of Stela 7 (Group 9). Now part of a wall behind the house of Domingo Hernandez in the southwestern quarter of the village. (See plate 3 and figure 22, a'.) 9.7.0.0.0 7 Ahau 3 Kankin (?). Text, {a) photograph: plate 9, a. {b) drawing: figure 15. Reference: Spinden, 1913, p. 159, footnote i, and table i.^ Stela 18 was found by Spinden in 1912, with Stela 20 and Altar Q', in a wall behind the house of Domingo Hernandez, in the southwestern quarter of the village. (See plate 3 and figure 22, a' .) At that time he incorrectly 'Schellhas, 1904, p. 19. 'Spinden refers to Stela 18 as Stela i6 in this work. In the nomenclature followed here, and since adopted by Spinden, this monument is renamed Stela 18. MORLEY. PLATE 10 - , . .-, J, SjP ' P > ' i '^1^' ^r/> 1*3^ 1 L. ' ^ ^ ^' -"■^ ^ Stela i6. (ajFront, (b) back. Drawn from the original. INSCRIPTIONS OF THE EARLY PERIOD. 97 grouped the four fragments together, calhng all parts of the same monu- ment Stela 16. (See p. 60.) As previously stated, a study of these four fragments revealed the fact that instead of being parts of one and the same monument, they are divided among three monuments — two stelae (18 and 20) and one altar (Q'). The single fragment of Stela 18 preserved is 69 cm. long, 67 cm. wide, and 43 cm. thick. The front was sculptured with a human figure — the earliest example of the front presentation of the human form yet found at Copan,^ and the back and sides with glyphs. On the basis of this new arrangement of the design, therefore. Stela 18 has been assigned to a new class, 4. This fragment came from near the top of the monument, although, as will appear presently, the top itself, a piece about half a meter long, is still missing, the upper edge of the piece recovered being about half a meter below the former top of the stela. Spinden, to whom the discovery of this important monument is due, has prepared the following well-considered description of it for use here: "Stela 18 is the designation given to a fragment that now lies under the eaves of an inside porch or workshop of a house in Copan village. The house belongs to Antonio Guerra^ and is situated on the south side of the street that runs west from the southwest corner of the plaza. "The realistic design shows a grotesque head-dress similar to those worn by human figures on other Copan monuments. The sculpture is in very low, flat relief and is so badly battered that all the details can not be made out. The gro- tesque face is evidently intended for a jaguar and is decorated with small circles. The round eyes have the lids half closed and under these lids is a design very like the Venus symbol. The top of the open mouth is at the bottom of the sculptured block and one may see the two large canine teeth between which are four incisor teeth. What may be part of the curve of the usual spiral fang appears at each side of thernouth. The nose is rather unusual m shape, but at the top of it is the charac- teristic divided scroll. The design above this jaguar head may, perhaps, be regarded as its head-dress. In the center is a heart-shaped object, from the sides of which issue more or less flame-shaped scrolls. The design as a whole shows general simi- larity in assemblage of details to the head-dresses on Stelae E, 7, P, and 2, although simpler than any of them. For purposes of comparison the head-dress on Stela 2 may best be used, since it also consists of a jaguar face with a heart-shaped object above. To be sure, this heart-shaped object is modified by a grotesque face at the bottom and an open hand at the top, but the fundamental likeness is there. The lateral ornaments of Stela 2 also recall those on Stela 18. Other details on Stela 2 are in addition on the new monument. "It is certain that we have in this fragment the earliest example with a human figure so far described at Copan. It is a delicate question whether it is earlier or later than Stela 15, which has hieroglyphs on all four sides. The earliest dated 'The small figures presented on the long sides of Altars X and Y are possibly earlier, but they are much smaller and in profile and also the archaic statues found unde-r them in the foundations of Stelae 5 and 4, figure 67, a and b, respectively. ^When Spinden first visited Copan (1912) this house was in the possession of Antonio Guerra. It subsequently passed into the hands of Domingo Hernandez, as already noted, p. 60. 98 THE INSCRIPTIONS AT COPAN. Fic. 15. — Inscription on back of Stela iS. monument previously known with a human figure is Stela 7 (9.9.0.0.0). The evidence of a human figure on Stela 18 makes it practically certain that a human figure once existed on the unmolded side of Stela 9/ and it may be regarded as established that full-length human sculptures in stone were known at Copan in Katun 4, and that they were probably very similar, in details represented, to those carved a hundred years later." Most unfortunately the text on the single fragment recovered is in poor condition. As one faces the monument, the left side has an Initial Series introducing glyph, although only its bottom part is preserved. | \czJ&}±D\ I Owing to the diagonal direction ot the line of fracture, the corresponding glyphs on the back and right side are on the top fragment, now missing. That Initial Series introducing glyphs were formerly present, however, seems certain. Indeed, on the back the Initial Series number itself is partially preserved. (See plate 9, a, and figure 15.) There seems to have been recorded, there- fore, on the back of Stela 18, an Initial Series, the introducing glyph of which is now missing. Two heads appear indistinctly in A3, figure 15, the first a human type, the second grotesque. These doubtless record 9 cycles. The entire destruction of the katun sign and coefficient in b3 is a very serious loss indeed, since it considerably increases the range of the possible readings. The coefficient of the tun sign in A4a is fr^-^^XU] the clearest glyph in the entire inscription; and although its form If^SpT" is irregular, it can hardly be other than o, 5, 10, or 15, on the r^^CcrlJ grounds of antece- dent probability, i. e., corresponding to some hotun-ending. However, this identification appears to rest on a somewhat firmer foundation than that of antecedent probability alone, as the following will tend to show, although in order to make the point clear a somewhat lengthy digression is necessary. In figure 16 is shown part of the inscription from the Temple of the Cross at Palenque.- At d3-c5 is recorded "4 Ahau 8 Cumhu, End of Cycle 13," and immediately following this in d5-c6 is a Secondary Series of 1.9.2. Following after several intermediate and probably non-calendric glyphs (not shown in figure 16) the date 13 Ik ? Mol is recorded at C9, D9, the month coefficient D9a being something like the tun coefficient on Stela 18. Compare Aia, figure 15, with ■D9a, figure 16. Performing the calculations indicated here, it will be found that if 1.9.2 is counted forward from 4 Ahau 8 Cumhu, the date reached will be 13 Ik o Chen, the day of which is recorded in C9. Passing over the month in d9 for the moment, let us continue the inspection of the text. 'The writer's dissent with this view has already been noted. See p. 93. ^For the sake of brevity certain intermediate glyphs — probably of a non-calendric nature — have been omitted. The glyph designations in figure 16, however, correspond with those in the complete text, for a reproduction of which see Maudslay, 1889-1902, vol. iv, plates 73-77. INSCRIPTIONS OF THE EARLY PERIOD. 99 d eesss cm; Following after seven intermediate glyphs, there is recorded in di3-ci5 a distance number of 1. 18.3. 12.0, and after five more glyphs the date 9 Ik 15 Ceh at El, Fi. If 1. 1 8. 3. 1 2.0 be counted forward from 13 Ik o Chen, the terminal date reached in the first calculation, the date reached will be 9 Ik 15 Ceh, which is recorded in ei, fi. It therefore seems certain that with the possible exception of the month- sign of the second date, i. e., ? Mol in d9, this text is correct, the calculations agreeing with the glyphs recorded. Let us next examine d9. Since the day-sign in C9 is very clearly Ik, the month coefficient in D9fl must be either o, 5, 10, or 15, which reduces the possible meaning of 9a to one of these four numerals. Now, although the month-sign re- corded is Mol, and the month-sign as indicated by the calculations should be Chen, the month coeflficient itself can only be o. For if 1.9.2 is counted forward from 4 Ahau 8 Cumhu or 1. 18.3. 12.0 is counted backward from 9 Ik 15 Ceh, the date reached in either case will be 9 Ik o Chen. It therefore seems certain that D9a is to be identified as o, the month to which it is attached being incorrectly given as Mol instead of Chen. Mol was the month immediately pre- ceding Chen, and this may have given rise to the error in the month-sign here. But the resemblance between ma and the tun coefficient of Stela 18 in AAa, figure 15, has already been pointed out, and on the basis of this close similarity we are probably justified in iden- tifying Aia as a sign for o. This identification, moreover, does not rest on this one apparently corroboratory passage alone. In figure 17, a, is shown a Calendar Round date from one of the piers in the western court of the Palace Group at Palenque.^ This is clearly 13 Manik ? Yaxkin, the place of the month co- efficient being occupied by the same sign as those in Aia, figure 15, and V9a in figure 16. Now, Manik can have only four possible month coefficients, o, 5, 10, and 15. Therefore, because of this fact, and because of the resemblance just noted, it seems reasonably safe to conclude that the C9 C OIK >^ S3 3 rmn]^,. C ¥ Fig. 16. — Parts of inscription (col- umns C and D) on left slab of tablet in the Temple of the Cross at Palenque. 'See Maudslay, 1889-1902, vol. iv, plate 29, No. 7, Gl, and vol. iv of text, page 16. lOO THE INSCRIPTIONS AT COPAN. m. month coefficient in figure 17, a, is o, even though calculations to prove it are wanting. The most that can be ofifered in corroboration of this second case is that one of the places where 13 Manik o Yaxkin occurred in the Long Count was 9.1 2.1 1.2.7, ^1"*^ there is another date at Palenque, an Ini- tial Series, which records 9.12.6.5.8^ 3 Lamat 6 Zac, which is within 5 years of 9. 12. II. 2. 7. A third occurrence of this unusual form for o is shown in figure 17, ^, a Calendar Round date from Lintel 9 at Yaxchilan.- This is the date i Eb ? Yaxkin. Now, Eb can only have o, 5, 10, or 15 for its month coefficients, and therefore one of these values must be substituted for the month coeffi- cient in figure 17, b. But this sign is like the other three month 1 "^ M coefficients just described, and we are therefore probably jus- tified in identifying it also as o. Unfortunately there is again no calculation by means of which this reading can be verified. A fourth possible occurrence of this form for o is shown in figure 17, c, which is the Initial Series terminal date on the Ley- den Plate.^ This Initial Series records the date 8. 14. 3. 1. 12 i Eb o Yaxkin, all of which is perfectly clear except the month coefficient. While the bar attached to the month- sign may be the month coefficient 5 (one of the four values possible here, i. e., c, S, 10, or 15), the writer prefers to regard the unusual element as the month coefficient, as the calculations show that o and not 5 is required here. If this character is turned on either end, it looks not unlike 'Maudslay, 1S89-1902, vol. iv, plate 78. This latter date Is recorded on a fragment of a sculptured slab, possibly a stela, which was found on the slope of the pyramid of the Temple of the Foliated Cross at Palenque. Although the Initial Series introducing glyph, the cycle-sign and coefficient, and katun-sign and coefficient are missing, the writer was able to decipher this Initial Series as 9.12.6.5.8 3 Lamat 6 Zac, which agrees with the day of the terminal date as recorded. There is a Secondary Series of 1. 10. 1, which is counted backward from the Initial Series terminal date to reach the date i Manik 10 Pop also recorded. The whole inscription seems to have read: 9.12.6. 5.8 3 Lamat 6 Zac 1. 10. 1 Backward 9. 12.4. 13.7 I Manik 10 Pop This is recorded as follows: Fragment l missing AI-B2 Initial Series intro- Fragment 2 ducing glyph 3 I missing A3 9 cycles 2 1 missing B3 12 katuns 2 2 A4 6 tuns 2 2 B4 5 uinals 2 The only irregular feature of this text is the omission of the month of the Initial Series terminal date, 6 Zac. ^Maler, 1903, plate 53. 'See figure 65 and Holden, 1881, figure 52, and Moriey, 1915, figure 75, A. b c Fig. 17. — Parts of inscriptions on: a, pier in western court of the Palace Group at Palenque; b. Lintel 9 at Ya.\- chilan; c, Leyden Plate. A^ 8 kins BS 3 Lamat (6 Zac) A9 I kin, 10 uinals B9 I tun AIO I Manik BIO 10 Pop MORLEY. PLATE II /.--- I I Stela 17. (a)Front, (b) back. Drawn from the original. INSCRIPTIONS OF THE EARLY PERIOD. Id the form for o here under consideration, and, as pointed out, is the value demanded by the Initial Series number actually recorded.' Even disregarding this last doubtful example, the writer believes suffi- cient evidence has been presented to show that the tun coefficient in K^a, figure 1 5, is probably o; and further, that it is probably the normal form of the tun sign, though why this should have been chosen to stand for o in these particular cases is unknown. Returning once more to the consideration of our text (Stela i8), the next glyph after the tun sign and coefficient, i. e., b4 (see plate 9, a, and figure 15), is badly effaced. Enough remains of the coefficient, however {v,\d), to show that it had almost certainly been o, the outline ^ being that of the usual sign for o ^ The kin coefficient in as is entirely / effaced. The day [ of the Initial Series terminal date ^ is probably re- corded at SBfl. ^ At first sight the coefficient appears to be 6, but a very careful examination of the original revealed the fact that formerly there appears to have been two other dots, one on each side of the central one, and that this coefficient may therefore have been 7 or even 8, all three possibilities having to be reckoned with. Antecedent probability, we have seen, probably justifies the postulate that a katun, lahuntun, or at least a hotun-ending was recorded by this Initial Series. That this was a general practice, moreover, is explained in Appendix VII and has been exemplified already in the cases of Stelae 24, 15, and 9, and probably also in the cases of Stelae 20, 25, and 17, and we are therefore justified in accepting it as our first postulate. Again, as shown above, there are excellent reasons for believing that a4(3, the tun coefficient, is o, which would make the period-ending here recorded a katun-ending, which we will make our second postulate, thereby restricting the possible dates under our first postulate to katun-endings only. That the coefficient of the day-sign in BSa is surely 6, 7, or 8, will be our third postulate; and finally, on stylistic grounds, Spinden has shown that Stela 18 must be earlier than Stela 7 (9.9.0.0.0), which will be our fourth and last postulate. An examination of Goodman's tables discloses that there are only three places in the Early Period where the conditions imposed by these four postulates are fulfilled, namely: 9.0.0.0.0 8 Ahau 13 Cell 9.1.0.0.0 6 Ahau 13 Yaxkin 9.7.0.0.0 7 Ahau 3 Kankin Unfortunately there is nothing further in the text as it has come down to us — the most hopelessly effaced of all the Copan stelae — to aid further in its decipherment, and we are forced to fall back upon the stylistic criteria and historical probability to decipher its date even approximately. 'It is interesting to note in these three examples that the month having this unusual sign for o is Yaxlcin. Since there are some grounds for behevlng that the Maya year may have begun with this month at one time, the three days recorded in figure 17 would then have been Maya New Year Days. I02 THE INSCRIPTIONS AT COPAN. Historically considered, the first two values can probably be eliminated at the outset. The earliest surely deciphered date at Copan is 50 years after the first and 30 years after the second. And even if the reading suggested for Stela 20, 9.1.10.0.0, be accepted as correct, both of these readings still antedate it. Stylistically considered, the last reading is very much preferable to the other two. If 9.7.0.0.0 is the correct reading for this Initial Series, it makes much less of a gap between Stela 18 and the next stela in Class 4, i. e., Stela 7, than if either of the other two are correct, i. e., 40 years as compared with 180 and 160 years respectively. This may seem a minor point, but where the development of art unfolded as rapidly as it did at Copan, 180 or even 160 years is too much time to have separated Stelae 18 and 7. Furthermore, once having developed this new type of stela (Class 4) with a human figure on its front, it seems more than likely that the ancient sculptors would not have allowed 160 or 180 years to go by before attempting it again; indeed, even 40 years would appear to have been a long period to have waited. Finally, chronologically considered. Stela 18 fits in much better with the other stelae of the Early Period at 9.7.0.0.0 than at 9.0.0.0.0 or 9.1.0.0.0, and in conclusion, in want of further evidence, the writer has accepted 9.7.0.0.0 as the date most likely to be recorded here. Since Stela 18 is probably the earliest stela upon which the human form was represented at Copan, it is unfortunate that it is in such a fragmentary condition as to prevent certain dating; but even so, the reading suggested may probably be accepted as approximately correct, and possibly, actually so. Stela 7. Provenance: Found fallen above its own foundations at Group 9. Now in the cabildo. (See plate 3 and figures 18, b, 19, and 22, 0.) Date: 9.9.0.0.0 3 Ahau 3 Zotz. Text, (a) photograph: plate 13. Spinden, 1913, plate 18, i (front only). {b) drawing: plate 13. Maudslay, 1889-1902, vol. i, plate 108. References: Gordon, 1896, p. 38. Maudslay, 1889-1902, vol. i of text, pp. 16, 67. Spinden, 1913, pp. 159, 160, 164, and table I. Stela 7, broken and badly mutilated, now lies in the cabildo, having been removed thither from the yard of the house of Clementino Lopez at the southwest corner of the village plaza in 1917. (See figure 22, 0.) Mauds- lay describes it as "a broken and defaced monolith lying in the scrub about 50 yards to the west of the village";^ and Gordon says he found it lying "in the bush to the west of the modern village, . . . fallen and broken".^ Maudslay says again that it was in " Copan Village, west of Altar U."-'' (See also plate 3.) 'Maudslay, 1889-1902, vol. I of text, p. 16. ^See Gordon, 1896, p. 38. 'See Maudslay, 1889-1903, vol, i of text, p. 67. INSCRIPTIONS OF THE EARLY PERIOD. 103 Maria Melendrez gives the following account of the breaking of this monument. In the spring of 1874 a Colonel Vicente Solis came to the village from Santa Rosa, the capital of the Department of Copan, with troops, pursuing some political malefactors who were fleeing toward the Guatemalan frontier. While at the village he tried to move Stela 7 from where it lay in the bush to the plaza for re-erection, but his men succeeded only in dragging it a few meters before it broke in two and it was left in the bush where it was. Clementino Lopez later acquired this property, and it was under the back portal of his house that the writer first saw this stela in 1910. The original provenance of Stela 7, however, has been established by the discovery of the foundation-stone (and the chamber underneath) upon which this monument rested. This important discovery has already received some attention in connection with the de- scription of Stela 24 (pp. 78-80), and its con- sideration will be resumed here. The mound of Stela 7 is about 50 meters southwest of the southwestern corner of the village plaza (see figure 22, F). By far the greater part of it lies in the property of Clementino Lopez (see figure 22), but the northwest corner is on the lot of Domingo Hernandez and the northeast corner on the lot of Florencio Lemos. It is 27 meters long north and south, 17 meters wide east and west, and 70 cm. high. The southwest corner of the foundation-stone of Stela 7 (figure 18, a) is 13.5 meters from the southern side of the mound and 4.33 meters from the eastern side. Stela 7 itself (figure 18, b), from the best informa- tion available, was found lying on the ground 2 meters west of its foundation-stone, appar- ently just where it had fallen. The accidental discovery of this foundation-stone, together with a frag- ment of Stela 24 and a small round altar, in 1916, was by no means the first of its kind, but the culmination of a long series of similar discoveries, such as archaic stelae and altars traceable to this mound, numerous smaller archaic fragments found in the walls of houses and pavements in the immediate vicinity, etc., all of which made it practically certain that formerly there had been here an important early center of occupation, possibly even the very earliest settlement in the valley. Indeed, this mound had become so important because of these several discoveries that its examination became imperative, and in 1919 the writer excavated the northern half. (See figure 18.) LEGEND. ° ' ^ ^ ? ? ^0' '.r^Z-'r -. presented next. Stela 19 is the most distant of all the monuments from the Main Structure. The inscription is presented on all four sides, and on the basis of this arrangement it may be assigned to Class 3. Only one side has an Initial Series, the introducing glyph of which (ai-b2) is of extraordinary form. (See plate 16.) The trinal superfix and sub- fix are both present, as well as the vari- able central element, which is appar- ently a human figure. The "tun" sign and comb-like lateral appendages, however, are almost entirely obscured by two twining serpents, whose gro- tesque heads with widely extended jaws occupy the greater part of the glyph-block. This latter feature, so far as the writer knows, is unique in the whole Corpus Inscriptionum Mayarum. The cycle-sign and coefficient ap- pear in A3, B3. The latter is clearly 9, there being one large double-outlined, cross-hatched dot on the lower part of the face. The katun coefficient (a4) is 10, the death's head being perfectly clear. The tun coefficient (as) is very clearly 19, and the uinal coefficient (a6), 15. The line of fracture runs through the kin coefficient (a/), which is partially effaced. The lower part of this head, however, shows the fleshless lower jaw, which would ordinarily indicate the number 10. The day-sign of the Initial Series terminal date should be recorded at A8, but even in spite of the effaced condition of this glyph it is clear that it never was a day-sign. Following through the remaining glyphs on the front, the last, Aio, is Glyph B of the Supplementary Series. Glyph A of the Supplementary Series (ci) is found on the top of one of the sides and is shown in figure 25, a, ci, and immediately following this is the day of the Initial Series terminal date (c2) 4 Ahau. Since the day-sign *■»,•. Fig. 24. — Map of cenrral mound of Group 13 (Hacienda Grande), showing location of Stela 19 and associated altars. INSCRIPTIONS OF THE MIDDLE PERIOD. 145 is Ahau, the kin coefficient in a7 must be o and not 10, as first deciphered. This is one of the cases already cited in connection with Stela 2, where the fleshless lower jaw has the value of o instead of 10, as usual. The record of the day-sign of the Initial Series terminal date after Glyph A of the Supplementary Series in this text, and not after the kin-sign, as is usually the case, has a par- allel in Stela 13, the next monu- ment but one in this group to be described, and one moreover which records a date only 60 days later than the Initial Se- ries of Stela 19. The next glyph in figure 25, a (03), should record the month of the Initial Series terminal date, but it is too eft'aced to decipher. Gather- ing together the above data, it will be found that the Initial Series 9. 10. 19. 15.0 4 Ahau 8 Chen is recorded on Stela 19 as follows: Fig. 25.- -a, Part of inscription on side of Stela 19. Altars associated with Stela 19: b, c, fragments of in- scribed altar; d, plain altar. AI-B2 A.;, B,; A4, B4 Initial Series introducing glyph 9 cycles 10 katuns A5,B5 A6, B6 19 tuns 15 uinals A7, B7 C2 C3 kins 4 Ahau (8 Chen) ? Although there are no other decipherable glyphs on Stela 19 itself, its inscription was probably continued on the periphery of the altar originally associated with it. During his visit to Hacienda Grande in April 1915, the writer discovered two fragments of a round altar about 30 meters north of Stela 19 and in the same plaza. (See figures 24 and 25, b and c.) These two fragments, though not contiguous, give sufficient of the cir- cumference to show that the altar was originally i meter in diameter and about 30 cm. in height. Judging from the size of the preserved glyph- blocks, if the glyph band extended clear around the periphery, there were 13 glyph-blocks in all. In view of its extremely fragmentary condition, it must be considered a rare piece of good fortune, therefore, that of the only five glyphs recovered either in whole or in part, about a third of the original inscription, two clearly record the Initial Series terminal date of Stela 19, namely, 4 Ahau 8 Chen. (See figure 25, b, bi, ci.) Although the upper part of this date is missing, it is perfectly clear as deciphered, and since the Initial Series on Stela 19 is 9. 10. 19. 15.0 4 Ahau 8 Chen, this Calendar Round date doubtless had the same position in the Long Count, 146 THE INSCRIPTIONS AT COPAN. Unfortunately the rest of the text is missing. The writer believes, however, that it was probably continued from this date 4 Ahau 8 Chen by a Secondary Series composed of 3 .0 to 9. 11 .0.0.0 1 2 Ahau 8 Ceh, which would bring the text forward to the nearest hotun-ending. A parallel case would be that of Stela E and its altar, already presented, although in this case the hotun-ending date is preserved and the date from which the count starts is effaced. On the altar of Stela 19 the positions are just reversed, the starting-date being preserved and the hotun-ending miss- ing. Another parallel will be found in the case of Stela I and its altar. (See pp. 179,182.) We have, then, on Stela 19, the record of a date 9. 10. 19. 15.0, which was 3.0 (60 days) before the end of the current hotun, 9.1 1. 0.0.0, and 2.0 (40 days) after the Initial Series of Stela 10, /. e., 9. 10.19. 13.0. And on the associated altar we have the same date as the Initial Series terminal date on the stela. This latter date, the writer believes, was brought forward to 9. II. 0.0.0 by glyphs on the pieces now missing. A summary of the combined texts follows: Stela 19 A1-B7, ci 9. 10. 19. 15.0 4 Ahau 8 (Chen) Altar Bi.ci 9.10.19.15.0 4 Ahau 8 Chen (3.0) missing (9. II. o. 0.0 12 Ahau 8 Ceh) missing There are 17 + 10 + 20 + 10 = 57 glyph-blocks on the stela and 13 on the altar, making a total of 70 for the entire text. The Initial Series introducing glyph occupies the space of 4 glyph-blocks. In addition to the round altar just described, Stela 19 seems to have had another of very unusual shape. (See figures 24 and 25, d.) This is also round, but has four deep grooves cut in the periphery 90° apart. There is a depression in the top, perhaps for holding offerings. It is plain and stands about 30 cm. high, being about a meter in diameter. It was found southwest of the inscribed altar. The inscription on Stela ig has one other unique feature, namely, that each of the period-glyph coefficients occupies a glyph-block by itself. Note also the filling element at the right of each except the cycle coefficient, i. e., A4, AS, A6, and A7, plate 16, which look like ending prefixes. Stela 23. Provenance: Original position unknown. Found hui t into the walls of the cabildo at Santa Rita, about 12 kilometers up the valley from the Main Structure, at Group I. (See plate 3.) Date: 9. 11. 0.0.0 12 Ahau 8 Ceh.' Text, drawing: figure 26. References: Mauds'ay, 1889-1902, vol. i of text, p. 16. Motley, 1916a, p 338. 'For other monuments recording this same hotun-ending, see Appendix VIII. INSCRIPTIONS OF THE MIDDLE PERIOD. I47 Three pieces of Stela 23, the most distant of all the outlying monuments, were found by the writer in March 1916, built into the walls of the cabildo at Santa Rita or Cachapa (Group i), 12 kilometers northeast of the Main Structure. It is probably the same monument as that mentioned by Maud- slay at this place: "The small village of Cachapa, which is situated about seven miles up the river-valley to the northeast, also stands on the site of old buildings, and several pieces of stone with remains of hieroglyphic writing on them are still to be seen lying amongst the rubbish of the plaza. "^ From such information as was to be had from the villagers, these appear to have been placed in the walls of the cabildo about 30 years ago. Con- cerning its original provenance little could be learned, other than that it was found somewhere in the village. However, in the village plaza there is a large drum-shaped stone, 1.06 meters in diameter and 37 cm. high, which would seem to have been either the foundation-stone or altar of Stela 23. As it is without decoration, perhaps the former is the better explanation of its use; and to this we may add Maudslay's testimony that there were several hieroglyphic fragments lying in the rubbish of the plaza in 1885. Santa Rita is built on the site of an ancient Maya settlement like the modern village of Copan, and similarly it has grown at the expense of the earlier group. Dressed stones are found in house-walls all over the village, and fragmentary sculptures, heads, torsos, and the like are scattered around the court yards of different houses. Mounds even may still be found within the limits of the village, and the ancient and modern plaza areas would appear to be coincident. In short, there is abundant evidence of the existence in former times of a not inconsiderable settlement here. When Stela 23 was broken it seems to have cracked into five large pieces, of which only three have been recovered — the top, third, and fourth fragments. Of the two missing pieces, only one had sculpture, however; the other, the bottom fragment, being the part that was embedded in the ground. The combined height of the three fragments recovered is 1.70 meters, to which it seems necessary to add another half meter for the missing second piece. The monument must have thus stood at least 2.5 meters above ground and probably 3 meters. It is 61 cm. wide and 46 cm. in thickness. An exhaustive search of the village failed to disclose the present where- abouts of the missing sculptured piece. All the walls in the vicinity were examined, but without success. One old woman said she thought it had been broken up into smaller pieces and used in the foundations of the church, but she was not sure. "It had been a long time ago." An examination of the church walls failed to confirm this story, although it may well be true. What the writer fears most is that it was broken into very small pieces and built into positions where the sculptured faces are either hidden or inconspicuous, as in the case of Altars L' and M' in the walls of the church at Copan village. The front is sculptured with a human figure side-view — the first, and with one other exception. Stela 11, the only example of a side presentation 'Maudslay, 1889-1902, vol. i of text, p. 16. 140 THE INSCRIPTIONS AT COPAN. of the main figure on a stela yet reported at Copan. The back and sides are inscribed with glyphs. On the basis of this arrangement of the design, Stela 23 may be assigned to Class 4. ( ) 1 te J Fig. 26. — Inscription on Stela 23: a, c, sides; h, back. The inscription (see figure 26, b) opens on the back with an Initial Series introducing glyph, di-e2, the variable central element of which is the kin- sign in a shield. It is almost identical with the corresponding element in the Initial Series introducing glyph on the back of Stela 16. Compare figure 26, b, with plate 10, b, and also with Date 28 of the Hieroglyphic Stairway. The cycle-glyph and coefficient, d3, figure 26. b, are almost entirely effaced, and the katun-sign and coefificient, E3, partially so. Fortunately, enough of the latter remains to show that it is 10, the two bars appearing very clearly. The tuns, uinals, kins, and day, d4, e4, ds, and e5 respectively, are on the missing second fragment, as are also the first two glyphs of the Supple- INSCRIPTIONS OF THE MIDDLE PERIOD. I49 mentary Series, D6 and e6. It is difficult to say whether four or six glyph- blocks are missing here, though from what remains of the Supplementary Series on the next piece, it would seem as though there had probably been six, as shown in figure 26, b. If so, the last two, d6 and e6, doubtless were Glyphs G and F, respectively, of the Supplementary Series. The glyphs on the third fragment continue the Supplementary Series, E7 being very clearly Glyph C; d8, Glyph X; and es. Glyph B. The fourth fragment opens with Glyph A, the last glyph of the Supple- mentary Series at d9, followed by the month of the Initial Series terminal date 8 .? at e9. There follow two familiar, though undecipherable, glyphs in Dio, Eio, in which the kin or day-sign appears to be the most important element, and then comes a Secondary Series in dii of 5.1 1. Finally, the last glyph on the back, Eii^, is very clearly 12 Ahau, and the top of the adjoining side on the right facing the front of the stela, figure 26, a, fi, is 8 Ceh. The next glyph, gi (figure 26, a), is entirely effaced, and the next, F2, is a period- glyph, recording probably 11, 12, or 13 katuns. Goodman's tables show that the only place the Calendar Round date 12 Ahau 8 Ceh could have ended an even tun in Cycle 9 was in 9.1 1. 0.0.0, at which time it also ended a katun and hotun as well, so there is little doubt that 12 Ahau 8 Ceh, End of Katun 1 1, is the date recorded by eiiZ'-f2. Other points in support of this reading are: the proximity of the Initial Series date, which we know is in Katun 10 (e3); the proximity of the Calendar Round date on the other side (left facing the front of the stela), i. e., within two years of 9. 11. 0.0.0, as we shall see; the prominence of this date 9. II. 0.0.0 elsewhere at Copan being recorded on no less than six other stelae; and finally the general stylistic criteria. Indeed, the contemporaneous date of this monument may safely be accepted as 9.1 1. 0.0.0 12 Ahau 8 Ceh. If the Secondary Series number in Dii, i. e., 5.1 1, is counted back from 9. II. 0.0.0 12 Ahau 8 Ceh, it will not reach the Initial Series terminal date: 9.11. o. o. o 12 Ahau 8 Ceh S-ii 9.10.19.12. 9 5 Muluc 17 Xul The katun coefficient in e3 agrees, but the month coefficient in e9 does not. As recorded, it is clearly 8, while by the above calculation 17 is reached. If, however, we change the Secondary Series number from 5.1 1 to 5.12, i. e., supposing one dot to have disappeared from the coefficient in una, we reach a date 9. 10. 19. 12.8 4 Lamat 16 Xul, which is exactly i tun later than the best reading of the Calendar Round date on the left side facing the monument (see figure 26, c, h4, 14), a significant coincidence: 9. 10. 18. 12. 8 8 Lamat i Yaxkin I. 0.0 9. 10. 19. 12. 8 4 Lamat 16 Xul Neither of these, however, can be the Initial Series terminal date, and owing to the loss of the second fragment it is impossible to fix this other than as having been some time in Katun 10. ISO THE INSCRIPTIONS AT COPAN. There is another period-ending on the side under discussion at G7-F9, figure 26, <3, which is exceedingly important, no less than 4 Ahau 8 Cumhu End of Cycle 1 3 , the starting-point of the Maya chronological system. So far as the writer knows, this unique date is found on only four other monuments in the Maya field: Stela C at Quirigua, Altar i at Piedras Negras, Stela J at Copan, and in the Temple of the Cross at Palenque. As recorded here, all is very clear but the Cycle 13 in f9, though the left-hand dot and part of the middle dot and topmost bar may still be seen. Cycle 13 is again recorded on the front of this monument at bi, the second glyph in the horizontal band in the upper left-hand corner, arrangement of the glyphs on said later. One other date (£ remains to be de- (£ indicate the reading (£ Concerning the unusual the front, more will be O at H4, 14, figure 26, c, on the opposite side, scribed. The preliminary inspectionwould 8 ? I or 2 Yaxkin. If the month coefficient is 2, the day-sign must be either Kan, Muluc, Ix, or Cauac. A comparison of H4, however, with all the known variants of these days shows that it has not the slightest resemblance to any of them, and we are forced to conclude either that this is an entirely new variant of one of them, or else the month coefficient in i4 is not 2. The latter is the better explanation of the matter. The day-sign, while it bears no resemblance to any of the known forms for Kan, Muluc, Ix, and Cauac, has on the other hand a very strong resem- blance to an unusual variant of Lamat, which, so far as the writer knows, occurs only here at Copan and once at Palenque. (See figure 39.) This variant appears to be the same as the sign for the planet Venus, or rather one-half of that sign,^ which may indicate some ceremonial connection between Lamat and that planet. At all events, half of the Venus-sign appears to have been used also as a sign for Lamat.- But Lamat can only have a corresponding month coefficient of i, 6, 11, or 16, and it is therefore appar- ent that the coefficient of Yaxkin in H4 must be i instead of 2.^ 'The complete sign for the planet Venus cal and horizontal axes of the glyph. Half of stitute, both in the inscriptions and in the where the Venus-solar period is set forth, relations. Indeed, the use of but half of the is bilaterally symmetrical with reference both to the verti- the sign C°)fr^Co) however, is frequently used as a sub- codices. ^^jV^~1 I" PP- 46~SO of the Dresden Codex, the two signs appear side by side, in identical sign for the whole would appear to have been due to the 3- 4- 5- 6. desire for brevity rather than as indicating any corresponding change of meaning. -The other examples of the use of this variant for Lamat are: 1. Stela J, east side. Glyph 32J (Maudslay's numeration), see p. 203, and figure 39, b. 2. Hier. steps south side Mound 2. G2 (plate 26, g), see p. 234 and figure 39, c. Hier. Stairway west side Mound 26. Date 2. sa, see p. 243 and figure 39, /. Hier. Stairway west side Mound 26. Date 3. ob, see p. 244 and figure 39, g. Hier. Stairway west side Mound 26. Date 24. Fragment 6 (plate 27), see p. 259 and figure 39, /;. Hier. Stairway west side Mound 26. ca (Gordon, 1902, plate 13 D), see p. 265 and figure 39, e. 7. Palenque, Palace Group, House C, Stairway. Maudslay (1889-1902, vol. iv, plate 23, b6 I, h.), and also figure 39, a. It will be noted that two-thirds of these examples are from the Hieroglyphic Stairway of Mound 26. Another connection between the Venus-sign and the day Lamat, although not a direct use of the former for the latter, is seen on Altar K (see p. 185), where the variable element in the Initial Series introducing glyph is the Venus-sign and the day of the Initial Series terminal date is Lamat. ^As drawn in figure 26, r, the coefficient of Yaxkin looks as much like 2 as I, but a final examination of the original in 1917 convinced the writer that either i or 2 is equally possible here, the original having one large, plain, TO!(H(/ dot, inadequately drawn in the text above, between two smaller ones, also plain. MORLEY. PLATE 17 .kJfck Stela 12. Part of the inscription on the back and south side. Drawn from the original. INSCRIPTIONS OF THE MIDDLE PERIOD. I51 The best defense of this reading that can be offered is the satisfactory chronological coincidence which its use develops; for if we read h4, u as 8 Lamat i Yaxkin, we will have a date which is within 2 years of the hotun- ending recorded on this stela, and which is exactly i tun earlier than a date reached in calculations on the other side, as already noted : 9. 10.18. 12. 8 8 Lamat i Yaxkin I. o. o (360 days) 9. 10.19. 12. 8 4 Lamat 16 Xul 5.12 (112 days) g.ii. o. o. o 12 Ahaii 8 Cell The month coefficient in 14, as noted in note 3 , page 1 50, looks as much like I as 2, and this, coupled with the fact that the day-sign is probably Lamat, and the satisfactory chronologic proximity arising from such an identifica- tion tends to corroborate the accuracy of the reading suggested. A sum- mary of the text follows : Front Bl Cycle 13 Back DI-ES, E9 9.10..?..?..? .? ? 6, 7, or 8.? Dll S-I2 Back and side Ellb, Fl 9.1 1. 0. 0. 12 Ahau 8 Ceh Side Gl, F2 End of Katun 11 G9, FlO 13. 0. 0. 0. 4 Ahau 8 Cumhu Gio, FlI End of Cycle 13 Side H7, 17 9. 10.18. 12. 8 8 Lamat i Yaxkin There are 6 glyph-blocks on the front of this monument, 19 on the back, and 22 on each side, making a total of 6+19 + 22 + 22 = 69 for the entire text. The portrayal of the principal figure of a stela, in profile, together with the peculiar inverted L-shape arrangement of the glyphs on the front, strikes a unique note of presentation at Copan, one indeed more characteristic of the northern cities of the Old Empire, i. e., those of the Usumacintla Valley and the Peten region of northern Guatemala. (See plate i.) Examples of this presentation in the former region are Lintels 32, 42, and 46, and Stelae II, 19, and 20 at Yaxchilan, where it is particularly noticeable; and in the latter region Stelae 23 and 29 at Naranjo. The only other stela at Copan having the same presentation of the principal figure is the small late stela. No. 11 (p. 369), which, however, lacks the peculiar inverted L-shaped glyph-panel so characteristic of the stelae and lintels at the northern cities. Stela 23, therefore, is particularly important as a connecting-link between the two parts of the Old Empire. It indicates that even at this early date (9. 1 1. 0.0.0), at the beginning of the Middle Period, there must have been close and constant intercourse between the different parts of the Old Empire. Indeed, we can possibly imagine northern sculptors as coming south to the great southern metropolis for new ideas, and the Copan sculptors as going north to draw inspiration and help from the great northern centers. There must have been a free and general exchange of ideas and technical processes and endless borrowing and copying of decorative motives. Possibly the 1C2 THE INSCRIPTIONS AT COPAN. human figure in profile may have been introduced at Copan with this monu- ment, only to have fallen into immediate disfavor. The glyphic presentation on the front of Stela 23 is but a single item in a vast body of evidence now available tending to establish the absolute homogeneity of culture throughout the Old Empire region. Stela 13. Provenance: On a steep bluff overhanging the north bank of the Copan River 6.5 kilometers northeast of the Main Structure at Group 2. (See plate 3.) Date: (Stela) 9. 11. 0.0.0 12 Ahau 8 Ceh.' (Altar) 9.10.17.0.4 2 Kan 7 Mac (??). Text, drawing: plates 15, b, and 22, d. Stela 13 was found by one of the Peabody Museum Expedition, but no description of it has ever been published, not even the fact of its discovery. Through information received from Gordon, the writer was able to re-locate this monument during his visit to Copan in 1912. It lies on the summit of a low hill or bluff overhanging the north bank of the Copan River, 6.5 kilometers up the valley from the Main Structure, and within 5 or 6 kilometers of Santa Rita, where Stela 23 was found. (See plate 3.) It is 3.34 meters long, 71 cm. wide, and 61 cm. thick. The asso- ciated altar stands in front of the north side of the monument, which we may therefore conclude was its front. All four sides are inscribed with glyphs, as in the cases of Stelae 20, 15, 21, 9, 12, 10, and 19, and on the basis of the arrangement of its design, Stela 13 is to be assigned to Class 3. The Initial Series introducing glyph appears at A1-B2. (See plate 15, b.) The cycle coeflficient (a3) is 9 expressed by a bar- and-dot numeral. The katun coefficient (b3) is unusually clear as 11, also expressed by a bar-and-dot numeral. The monument is broken across the tun and uinal glyphs, which are somewhat destroyed in consequence. The two coefficients, however (aw and B\a respectively), are suflficiently preserved to permit the identification of both as o. The kin coeflficient (a5<2) is in better condition and is exactly like the tun and uinal coeflficients, namely, o. The day of the Initial Series terminal date, 12 Ahau, is not recorded at B5 as usual, but is found at a8 after the last glyph of the Supple- mentary Series, which in this inscription is of very unusual form. The ball element in the oval in the upper part is a grotesque head. This is very rare; in fact, the writer recalls but one other instance in the entire range of the Maya inscriptions, also here at Copan, not only on one of the monuments of this same group, namely. Stela 3,- but also upon one recording the same date, namely, 9. 11. 0.0.0. 'For other monuments recording this same hotun-ending, see Appendix VIII. ^A few cases where the moon-glyph occurs in constellation bands are not included here, although even in such cases the head appearing is sometimes that of God D. Thus, for example, in the constellation band on the base of Stela 10 at Piedras Negras, God D appears in the upper part of the moon-glyph. See Maler, 1901, plate 19. INSCRIPTIONS OF THE MIDDLE PERIOD. I53 In this other example the grotesque head is of the same form as the one here on Stela sent God D, the deity with the last 13, and the writer suggests that both may repre- Moon God.^ The association of this particular glyph of the Supplementary Series is very appro- propriate, since this glyph as explained in Appendix VI is undoubtedly the sign for the moon. Moreover, in the codices the moon-glyph is the main part of the sign for God D. This connection between God D and the moon, and hence between God D and the last glyph of the Supplementary Series, which is probably the sign for the moon, is strengthened by the discovery of the head of God D in this glyph on Stela 13 and also in the corresponding glyph on Stela 3 as well. Returning once more to our text, the month-sign, 8 Ceh, will be found in A9. The last glyph in plate 15, b, B9, is very interesting. It records the " End of a tun," the tun-sign being the unusual winged-Cauac variant, the Cauac element here modified into a human head. Note the ending prefix and the hand. Another example is seen in Glyph qi5 from the tablet of the Temple of the Sun at Palenque (Maudslay, 1889-1902, vol. iv, plates 87-89). Here the end of a Tun 10 is recorded, the tun-sign being the usual head- variant, modified by well-known Cauac elements — the wing as a subfix, the "cloud-balls" as a head ornament — and the circle, here shown as a hook of dots, in the eye. These two examples are extremely important as establishing the early synonymity of the winged-Cauac and the tun-signs, and they are the beginnings of what later became almost the exclusive use of the former for the latter in the New Empire. (See also the inscribed peccary-skull in figure 56 and the accompanying discussion, pages 379-381, for a still earlier occur- rence of this same variant.) The whole Initial Series of Stela 13 therefore reads 9.1 1.0. 0.0 12 Ahau 8 Ceh, as follows: AI-B2 Initial Series introducing glyph A3 9 cycles B3 II katuns A4 tuns B4 uinals AS kins B8 12 Ahau A9 8 Ceh B9 End of a tun There are a few other signs which are familiar, although of unknown meaning. The last glyph on the Initial Series side is the day i Ahau. Its record here is diflficult to explain unless it be taken as a reference to the fact that the preceding katun ended on a day i Ahau, viz, 9.10.0.0.0 i Ahau 8 Kayab. The altar associated with Stela 13 is of the round, drum-shaped type, and is 1.49 meters in diameter and 46 cm. high. It had some crude carving on top, and four pairs of glyphs on the sides, the pairs being 90° apart. Only 'See Morley, 1916, p. 370, and Schcllhns, 1904, pp. 22, 23, figures 18-21, and plate I, D. 1^4 THE INSCRIPTIONS AT COPAN. the pair on the north side (see plate 22, d) seems to be of a calendrical nature. It records a Calendar Round date as follows: i or 2 Kan 7 ?, the day-sign surely being Kan and the month coefficient surely 7. The best value of the day coefficient is 2, though i^ is possible. The month-sign is possibly Zotz, though Muan or even Kankin and Mac are other possible values. The best reading of the date therefore seems to be 2 Kan 7 Zotz. This date occurred at intervals of every 52 years throughout Cycle 9. The probabilities in this case, however, are that it was some time near 9. 11. 0.0.0, the date on the stela with which the altar is correlated. It is clear, moreover, that this altar does not record a hotun-ending, since all hotuns ended on some day Ahau, while Kan is the day here recorded. The two occurrences of 2 Kan 7 Zotz in Cycle 9 nearest this Initial Series are, 9.9.8.2.4 and 9. 12. 0.15. 4. The former is 1.11.15.16 (about 31 years) before the Initial Series of Stela 13 and the latter is i. 0.15. 4 (about 20.5 years) after the Initial Series of Stela 13. Both of these readings are unsatisfactory and the writer is loath to accept either as the date recorded on this altar. It is quite possible, moreover, that the terminal date is not 2 Kan 7 Zotz. If the day coefficient is not 2, it must be I. This gives the date i Kan 7 Zotz, which occurred in Cycle 9 nearest before and after 9.1 1. 0.0.0 at the following places: 9.10.0.5.4 and 9. 12. 13. 0.4, both of which are equally unsatisfactory here. Another possible reading, i Kan 7 Muan, occurred nearest before and after the date of Stela 13 at 9.9.8.13.4 and 9. 12. 1.8.4, both of which are again equally unsatisfatory. Another possibility is 9. 10.8. 17.4 i Kan 7 Kankin, although little may be urged in its favor, except that the month-sign might possibly be Kankin. Approaching the question from a different angle, let us next find what are the nearest occurrences of the date i or 2 Kan 7 ?, that is, leaving the month-sign indeterminate for the present, both before and after 9. 11. 0.0.0. These can be shown to be: BEFORE. AFTER. 9. 10. 17. 0.4 2 Kan 7 Mac 9.11.0. 9.4 i Kan 7 Zip 9. 10.16. 16. 4 I Kan 7 Zac 9. 11. o. 11. 4 2 Kan 7 Tzec 9.1 I.I. 4.4 I Kan 7 Pax 9. II. I. 6.4 2 Kan 7 Cumhu Of these, the first is the only one whose month-sign may be a grotesque head, and even this reading has little in its favor. All things considered, however, the writer is inclined to accept it as the least of all evils and to regard the month-sign in bi as an unusual variant for Mac, although it should be understood that this reading is far from satis- factory and is by no means to be accepted as proved. A summary of the entire text follows: Stela A1-A9 9.11. o. o. o 12 Ahau 8 Ceh (2. 17. 16) undeclared Altar Ai-Bi 9.10.17. o. 4 2 Kan 7 Mac.?.? 'It can not be 3, since the three dots are not al! of the same size, and, moreover, are otherwise unhke. MORLEY. PLATE 18 Stela 2. Part of the inscription on the back, and east side. Drawn trom the original. >. CO., W*Sm.,O.C. INSCRIPTIONS OF THE MIDDLE PERIOD. ISS Stela 3. Provenance: In the Middle Court south of Mound 4 at the Main Structure. (See plate 6.) Date: 9. 11. 0.0.0 12 Ahau 8 Ceh.' Text, drawing: plate ig, a and b. Spinden, 1913, plate 19, 3 (front only). References: Gordon, 1896, pp. 35, 36. Maudslay, 1889-1902, vol. i of text, p. 66. Spinden, 1913, pp. 159, 164, and table I. Stephens, 1841, vol. i, p. 149. Stela 3 134.30 meters long, 95 cm. wide, and 1.07 meters thick. It now lies near the center of the Middle Court, some distance south of the south- eastern corner of Mound 4, broken into three pieces.- Originally it seems to have been supported by a roughly worked drum-shaped foundation-stone, which is still in situ. No chamber was found under this foundation-stone, but instead, under the three layers of narrow oblong stones on which it rested, there was a pocket of pigments of different colors; in this were found four nephrite heads, each carved with a representation of the human face, well executed and highly polished. These had been pierced longitudinally, so that they could be sus- pended on a string and used as pendants.^ :::d:^ ^^-^ I b Fig. 27. — Inscription on collar of Stela 3: a, one side complete; b, c, c, /, remains of glyphs on ends; d, glyph in center of side opposite a. At the level of the ground, the base of the monument was surrounded by a sculptured stone collar of four pieces. (See figure 27.) This presents the familiar knotted-band decoration so prevalent in the Early and Middle Periods at Copan. The centers of the front and back are each occupied by a single large glyph (figure 27, a and d), and the two ends of each side by smaller glyphs (figure 27, b, c, c, and/), making a total of 6 for the entire collar. One of the single glyphs, c, looks like Ahau, although this identifica- tion is by no means certain. The ends of the long sides terminate in large, grotesque serpent-heads, probably a development of the flanking serpent- 'For other monmnents recording this same hotun-ending, see Appendix VIII. ^When Stephens first saw this stela in 1839 it was fallen but does not appear to have been broken: "This statue is fallen and the face destroyed" (1841, vol. i, p. 149). He calls it Statue K. See .Appendix III. ^Gordon, 1896, p. 36. 1^6 THE INSCRIPTIONS AT COPAN. heads seen on Altars L', M', and Q', and later to be seen on more complex altars of the Great Period. In Stela 3 for the first time we find both the front and back of the monument carved with representations of the human figure, each side being inscribed with a single column of 10 glyph-blocks, which, with the 6 on the collar, make 26 for the entire text. This is a new development, and on the basis of this arrangement of the design. Stela 3 may be referred to a new class, 5. The inscription on each side commences with an Initial Series introducing glyph; and on this monument, also for the first time, we have 2 Initial Series recorded.^ Gordon says Stela 3 faced north and south, i. e., the surfaces presenting the human figures faced in these directions;- and if this was true, the inscrip- tion is presented on the east and west sides. Since it is now impossible to tell which of the latter faced east and which west, and since there is a rounded inclusion of harder rock embedded in the side on which the inscription begins, this side, in default of the knowledge of the original orientation, will be called the inclusion side. The Initial Series introducing glyph is at ai; the upper part and lower right-hand corner are missing. (See plate 19, b.) In the top of the variable central element, here a grotesque head, a rectangular hole has been cut. This is 76 mm. high and 63 mm. wide, and passes diagonally through the stone, emerging on the top, 51 mm. behind the edge and in line with the variable element. The interior surfaces of this hole are worn smooth, as though they had been subjected to rubbing, perhaps by cords. The Initial Series introducing glyph on the opposite, bi, has the same kind of a hole cut in the same relative position. This hole also emerges at the top behind the variable element on this side. The function of these two openings is unknown; perhaps banners or streamers of feather-work were attached to the stela by means of them; no other stelae at Copan show this feature.^ The order of the glyphs within the individual glyph-blocks is very unusual in this text. Instead of reading from left to right and top to bottom within the glyph-block, they read from top to bottom first and then from left to right. Unusual as this order is, however, it is amply substantiated by the sequence of the known glyphs on both sides of the monument. The cycle-sign and coefficient are found in A2a, the upper half being the coefficient and the lower half the period-glyph. (See plate 19, b.) The former is very clearly 9, and the traces of the clasped hand on the lower part of the 'Although Stelae 15, 18, 16, 17, 21, 7, and P have 4, 3, 2, 2, 4, 3, and 3 Initial Series introducing glyphs respec- tively, none of them has more than one Initial Series number. Indeed Stela 3 is not only the first example of this kind at Copan, but also the only one yet found here. Such stelas are very rare, there being but seven others now known in the whole Maya area: Stela 17 at Tikal; Stela II at Yaxchilan; Stelx i and 3 at Piedras Negras, and Stelffi F, D, and E at Quirigua, less than 4 per cent of all known stel«e. See plate i for the location of these sites. 'See Gordon, 1896, p. 35. This orientation makes Stela 3 face the Middle Court. ^ Stela I at Cancuen on the Rio de la Pasion (plate l) has holes passing through its top, but they are not only larger but are also of different shape, being round. (See Maler, 1908, p. 44 and pi. 13.) Maler here suggests these holes were used to bind sacrificial victims to the stel.ne: "It may be assumed that the victims were bound by means of the perforations to these steUp, the sacrifice probably being usually performed with the victim in an upright position before stel.-e of this kind." This explanation, it is hardly necessary to add, appears scarcely probable. INSCRIPTIONS OF THE MIDDLE PERIOD. 157 latter leave no doubt as to its identity. The left half of the katun coefficient in Aih u. h. is fortunately preserved, although it is like none of the known head-variant numerals. It has the fleshless lower jaw, which usually stands for lo, but which may on occasion mean o. This, we have already seen, was the case in the kin coefficients on Stelae 2 and 19, and it will also be found to be true of the tun coefficient in both this Initial Series and that on the other side of this same monument. Again, the large circle around the eye in this glyph is also characteristic of the kin coefficient of this same nitial Series (A4rt u. h.), and also of the kin coefficient of the Initial Series on the other side {bau u. h.) both of which are o; indeed, the former also has this same fleshless lower jaw. All things considered, therefore, we are probably justified in reading Aib as o katuns. The tun-sign and coefficient (A3fl) are unusually clear. The former (A3a u. h.) again has the fleshless lower jaw, ordinarily 10, but, as in the case of the katun coefficient above, here probably o. The tun-sign in A3a 1. h. is unusually clear, and helps to substantiate the correctness of the above sequence of the period-glyphs and their coef- ficients. The uinal-sign and coefficient in a3^ are gone. From what little is left of the kin coefficient in A4a u. h., it may almost certainly be identified as o, i. e., having the same fleshless jaw and large circle around the eye as the katun coefficient. Finally, since the katun, tun, and kin coefficients are probably o, the missing uinal coefficient in A3b u. h. must almost certainly have been o also, and our Initial Series therefore reads 9.0.0.0.0. Most unfortunately, the day of the Initial Series terminal date, A\b, is entirely eff^aced, and we thereby lose one opportunity of checking the above reading. It can be found from Goodman's tables, however, to have been 8 Ahau. Passing along, we reach the next to last and last glyphs of the Supplementary Series in A6. The large plain circle in the upper half of this glyph-block in plate 19, b, is the rounded inclusion alluded to above. The ancient sculptors were unable to carry the details of their design across its refractory surface and were forced to content themselves with breaking it otif flush with the face of the monument.^ Again most unfortunately, the month-sign of the Initial Series terminal date is suppressed, being filled with the grotesque head and coefficient 9, the occurrence of which will be found elsewhere here at Copan, namely, on Stela 2 and possibly on Altar H' (p. 138). Thus our second and last opportunity of checking the accuracy of the reading suggested above is gone; but in view of the close similarity of Aib u. h. with Aia u. h. and aau u. h. on this side and B3a u. h. on the other side, the last of which is proved to be o by the calculations, there is little doubt but that the correct reading 1 Through the kindness of Dr. F. E. Wright, of the Geophysical Laboratory of the Carnegie Institution, the writer is able to give a description of this inclusion, as well as of the material of the stela proper See Appendix I. 158 THE INSCRIPTIONS AT COPAN. here is 9.0.0.0.0 (8 Ahau 13 Ceh). The remaining glyphs on this side are of unknown meaning. The inscription on the opposite side (plate 19, a) opens with an Initial Series introducing glyph in Bi, the variable element of which, as already noted, has a rectangular hole passing through its upper part and out on top of the monument. The cycle coefficient in Bia u. h. and the katun-sign and coefficient in hib are entirely effaced. The former was doubtless 9. Passing over the latter for the present, it will be found that o tuns are recorded in B3a, the tun coefficient again having the fleshless lower jaw, usually 10, but on this stela always o.^ The uinal-sign and coefficient in 'Bib are entirely effaced. The kin-sign and coefficient in bau are clear; the latter is surely o, the clasped hand being used to denote o instead of the fleshless lower jaw, as elsewhere on this monument. The next glyph, which ordinarily would be the day-sign of the Initial Series terminal date, is entirely effaced. Passing over b5, we reach in B6fl, Glyphs B and A, the next to last and last glyphs of the Supplementary Series respectively. The latter, as already explained in connection with Stela 13, is very unusual in having a grotesque head, probably that of God D, in place of the dot element in the oval in the upper part of the glyph. (See pp. 152, 153.) Following this in b6^ is the day 12 Ahau, the day-sign being the familiar grotesque head variant;- and in bS(3 u. h., the month 8 Ceh, and in b8i3 1. h., Katun 11. The Initial Series recorded upon this monu- ment, therefore, is 9. 11. 0.0.0 12 Ahau 8 Ceh, and we may fill in the missing katun coefficient as 11 and the missing uinal coefficient as o, and finally accept as proven that the fleshless lower jaw in the tun coefficient in this Initial Series and in the katun, tun, and kin coefficients of the Initial Series on the other side of this monument stands for o. In Bib u. h. is very clearly recorded 13 uinals, or exactly i tonalamatl, 260 days. Record of tonalamatls, although very common in the codices, is almost unknown in the inscriptions, the present case being one of the very few cases known. The end of a tun is probably recorded in the next to last glyph-block on this side, B&b 1. h. There are no other decipherable glyphs on this side of the monument. A summary of the inscription follows. 'When the writer examined Stela 3 in April 19 1 5, he was unable to find the fragment which presented the cycle-sign and the tun coefficient, i. /•., Ma 1. h. and Aja u. h. respectively. A protracted search of the immediate vicinity of the monument failed tc disclose the whereabouts of this piece and he was obliged to leave the site without drawing these two glyphs. On his return to this country, however, during a visit to the Peabody Museum, he found this missing fragment miscatalogiied under the name cf "a piece of Stela 4." It had been brought from the ruins to Cambiidgc by the Second Peabody Museum Expedition in 1S93; and when its two glyphs were drawn to scale, they were found to fit exactly in the position shown in plate 19,0, and their fortunate recovery materially aided in the final decipherment of the date cf this monument. = See Bowditch, 1910, plate 6, Nos. 35-38; and Motley, 1915, p. 41 and figure 16,/ and //. Also compare Appendix X. INSCRIPTIONS OF THE MIDDLE PERIOD. I59 Inclusion side, ai Initial Series introducing glyph A2<2 9 cycles i^ib o katuns A3a o tuns A3^ o uinals A413 o kins h\b 8 Ahau (13 Ceh) not recorded Opposite side, Bi Initial Series introducing glyph V,ia 9 cycles ^ib II katuns B3a o tuns B3^ o uinals B4a o kins B6^ 12 Ahau B8a u. h. 8 Ceh B8a 1. h. Katun 11 'Bjb u. h. 13 uinals = 260 kins = i tonalamatl B8i 1. h. End of a tun The Style of the glyphs on Stela 3 is unusual (see plate 19, a and b); indeed, they are the most intricate at Copan. The design is exceedingly complex, and is executed with a delicacy and feeling hardly to be thought possible with tools of stone. Each glyph-block is a beautiful example of harmony and balance, and each, regarded as a unit, conforms to the most rigid canons of pure design. Barring some of the best work of the Great Period, the inscription on this monument is the finest glyph delineation which was done at Copan. In summarizing the inscriptions on the foregoing monuments (i. e., Stelae 12, 2, 10, 19, 23, 13, and 3), it is evident that, quite irrespective of the dates recorded upon them, they present very considerable stylistic diver- gences. Thus, for example, the glyphs on Stela 12 are crude and unskill- fuUy executed as compared with the beautiful ones on Stela 3 just described, and yet both of these monuments have the same closing date, namely, 9. 1 1. 0.0.0. Again, the glyphs on Stelae 10 and 19 appear to be better carved than those on Stela 13, and yet the dates of these two monuments are actually 100 days and 60 days earlier, respectively, than the date of Stela 13, which is also 9. 11. 0.0.0. A close examination of these monuments, however, shows that the observed stylistic divergences are rather those of execution and technique than of subject-matter. A study of their glyphic details, moreover, estab- lishes close similarities between the several monuments of the group, and at the same time differentiates them from all other monuments here at Copan. A case in point is the unique use of the fleshless lower jaw to denote o, found only on Stelae 2, 19, and 3, here at Copan, and nowhere else in the entire Corpus Inscriptionum Mayarum. Another example is the portrayal of a head in place of the dot element in the last glyph of the Supplementary Series, found only on Stelae 13 and 3 ; or again, the substitution of a sign with a grotesque head and a coefficient of 9 in place of the month-sign of the Initial Series terminal date, found only on Stelae 2 and 3 ;' or again, the use ' Altar H' may possibly be another example of this. See plate 23, a, vb 1. h., and p. 138. l6o THE INSCRIPTIONS AT COPAN. of the full-figure variant of the uinal-sign, and of head variants for all the other periods, found only on Stelae 15, 24/ 12, and 2; or again, the use of a very unusual grotesque head with large prominent teeth for the kin-sign found only in Stelae 10 and 3. These stylistic similarities between the several monuments of the group establish a very close connection between them, and probably indicate a single source of inspiration, or group of sources very closely related, and almost certainly contemporary. Doubtless some artisans were more skillful than others; some more clever in reproducing in stone the working drawings on paper or skin, which must have preceded such elaborate compositions as those represented on the Copan stelae. Such differences, however, due to individual variations in the personal equation, are inevitable in any art at any time. For every Maya Phidias or Praxiteles there must have been a score of Alcamenes, Naucydes, or Thrasymedes; for every genius, a host of lesser lights. And the observed stylistic divergences in the several monuments of this group are not greater than would have been due to the varying personal equations of contemporary sculptors either here at Copan, or at Athens, or at Thebes. Indeed, it is almost necessary to postulate a contemporaneous origin for the monuments of this group to explain satisfactorily their glyphic similarities. The foregoing stylistic analysis forecasts, as it were, the chronologic situation as established by the dates actually recorded upon these monu- ments, namely, that all seven date from the same period, five (Stelae 12, 2, 23, 13, and 3) actually recording the date 9. 11. 0.0.0, and the other two, Stelae 10 and 19, dates only 100 days and 60 days earlier respectively; and in the case of the last-mentioned, the inscription on the associated altar doubtless also brought its date down to 9.11.0.0.0. Considering the chronologic record somewhat closer, it will be remem- bered that (i) two of these monuments (Stelae 12 and 2) have 9. 10.15. 0.0, the previous hotun-ending as their Initial Series, but close with Period End- ing dates of 9.1 1. 0.0.0; (2) that another. Stela 23, has a Katun 10 Initial Series, but closes with this same Period Ending date; (3) that two others (Stelae 10 and 19) have as their Initial Series dates less than 6 months earlier, one of which, as just explained, was probably brought down to 9. 11. 0.0.0 by a Secondary Series on the associated altar; and finally (4) that the remaining two. Stelae 13 and 3, actually have 9. 11. 0.0.0 as their Initial Series.''* In short, chronologically considered, these monuments probably date from the same hotun-ending, namely, 9. 11. 0.0.0, a condition corroborated by the stylistic criteria in spite of considerable technical divergences.-'' The situation, therefore, may be summed up as follows: These seven stelae were probably erected, or at least dedicated, at the same time, namely, 'Although Stelae i; and 24 show this same unusual feature, they are 130 and 170 years earlier respectively than Stela 12 and Stela 2 and therefore are in no danger of being confused with them either on chronologic or stylistic grounds. ''Although Stela 3 presents two Initial Series, 9. 0.0. 0.0 and 9. 11. 0.0.0, only the latter of course could have been the contemporaneous date. ^It is possible that Stelse 12 and 2 may date from the previous hotun-ending, 9. 10. 15.0.0, in which case the Katun 1 1 dates on them are prophetic. INSCRIPTIONS OF THE MIDDLE PERIOD. l6l 9.11.0.0.0, but the actual making of them, the carving, was intrusted to different hands. In two cases, Stelae 2 and 3, representations of the human figure were attempted, Stela 3 being the more successful, as well as the more ambitious, since it has two figures. One, Stela 23, shows a side pre- sentation of the human figure, the only example of its kind at Copan. The other four were inscribed with glyphs on all four sides and vary in excellence in the following order, the first being the crudest: Stelae 12, 13, 10, and 19. Finally, although differing considerably in technique, all seven may safely be assigned to the same hotun-ending. We come next to a period of some perplexity, namely, the katun after 9. II. 0.0.0. Following this latter date, there seems to have been a hiatus in the sequence of the monuments at Copan for two hotuns, since there are no sculptures, either stelae or altars, that can be assigned to the period between 9. 1 1 .0.0.0 and 9.11.15 .0.0, on which latter date Stela i and the East Altar of Stela 5 were erected. The next hotun, 9.12.0.0.0, seems to have been marked by the erection of two altars but no stelae — the West Altar of Stela 5 and the Altar of Stela i. At least no stela has yet been found recording this date. The uncertainty alluded to above, in regard to these four monuments, lies first in the possibility that the dates of these three altars may not be correctly deciphered as given, a possibility, however, which the writer believes to be remote, and second, in the fact that the East Altar of Stela 5 apparently should be associated with Stela i and not with Stela 5, the altar now associated with Stela i belonging elsewhere. These points will receive further attention later, and since the dates suggested above are probably correct, these four m.onuments will be presented in the following order: Stela I, the East and West Altars of Stela 5, and the altar of Stela i. Stela i. Provenance: On the second step of the stairway on the western slope of Mound 9 at the southeastern corner of the Mid- dle Court, Main Structure. (See plate 6.) Date: 9.11. 15.0.0 4Ahaui3Mol.^ Text, (a) photograph: Maudslay 1889-1902, vol. i, plate loo. (b) drawing: ibid., plate 100, a, b, and c. Gordon, 1896, figure 6. References: Bowditch, 1910, pp. 100, loi, 135, 196, and table 31. Gordon, 1896, pp. 36, 37. Gordon, 1902, pp. 174-176. Maudslay, 1889-1902, vol. i of text, p. 66. Spinden, 1913, p. 159, and table i. Stela I is broken into two pieces. The upper and larger part lies on the ground, the base being ifi situ on the second step of the stairway ascending the western slope of Mound 9. Underneath the base of this stela was a cruciform chamber like those under Stelae 7, 1, M, and C. This was opened by the Fourth Peabody Museum Expedition in 1895, and was found to con- tain "five rude earthenware vessels, fragments of stalactites, shells of the Spondylus calcifer, a large jadeite bead, and a quantity of cinnabar," all ' For other monuments recording this same hotun-ending, see Appendix VIII. l62 THE INSCRIPTIONS AT COPAN. of which material is now on exhibition at the American Museum of Natural History, New York.^ Below, on the ground-level, is the circular altar described on pages ly^-ijj^ and figured in plate 21. As explained more fully there and as noted briefly above, it is probable that this altar was not originally cor- related with Stela i, but with some other monument now missing. Stela I is about 2.75 meters long, 91 cm. wide and 76 cm. thick. Its front (west face) presents a large human figure; and its back and sides are inscribed with glyphs, on the basis of which arrangement of the design it is to be referred to Class 4. There is but one Initial Series introducing glyph at A1-B2 on the back (east face), and this is followed by the Initial Series 9. II. 1 5. 14.0 II Ahau 8 Zotz, as follows: A IB 2 Initial Series introducing glyph A3 9 cycles B3 II katuns A4 15 tuns B4 14- uinals A 5 o kins ci II Ahau C2 8 Zotz This date is just 14 uinals after a hotun-ending in the Long Count; and in d5^ exactly this number of uinals is recorded. Finally, there follows next in d6 the date 4 Ahau 13 Mol, which is the ending-date of the previous hotun, 9.11.15.0.0 4 Ahau 13 Mol. This distance-number of 14 uinals, therefore, is to be counted backward from the Initial Series instead of forward to reach the hotun-ending recorded. Secondary Series numbers which are to be counted backward are very unusual,^ and only the desire to bring the Initial Series terminal date back to the nearest hotun-ending could have justified this violation of precedent here.* The coefficient of the day-sign in this hotun-ending date is also very unusual. It is neither the usual type of bar-and-dot numeral nor a head- variant. The number recorded is clearly 4 and is represented by 4 dots in a circular cartouche, thus ((^jl- These slight irregularities, however, in no way destroy the accuracy ^<=^ of the above reading, and the date of Stela i may be safely accepted as 9. 1 1. 1 5. 0.0 4 Ahau 13 Mol. This text has 124-9+9 = 30 glyph-blocks, the Initial Series introducing glyph occupying the space of 4 glyph-blocks, and a very curious glyph at the bottom of the Initial Series side, the space of 2 glyph-blocks R^| ^|||tfjl - This is a unique variant of the hotun-sign, which further m W^)f^^^ confirms the above reading as the correct date of this monu- ^j X^ ■ ^-ff^jj^ ment. The prefix, 1 Gordon, 1896, pp. 36, 37. - Maudslay's drawing (1889-1902, vol. I, pi. 100, A, glyph 4) incorrectly shows 12 uinals. An examination of the original, however, provid that the two outside dots had the same interior circles as the two inside ones, and that all four are therefore numerical. Compare the dots in B4 with those in A3, B} and ci, all of which show the same interior circle, for verification of this statement. In Maudslay's drawing also the month coefficient in C2 {ibid., c, glyph 2) appears as 13; the original, however, is clearly 8. ' For a discussion of the diiection of the count in Maya numerical scries, see Morley, 1915, pp. 136-138. ^ Several other cases of this kind occur both hue at Copan (reviewing stand in the Western Court and Altars Q and W) and elsewhere, and the whole question has been reviewed in Chapter IV, p. 333. INSCRIPTIONS OF THE MIDDLE PERIOD. 163 although unusual, is probably an ending-sign, the superfix an ornamented numerical bar, i. e., ^, and the main element, the winged-Cauac variant of the tun-sign. Note the "balls" in the upper right-hand corner and the half circle of dots at the bottom, both characteristics of the tun element inthehotun-sign. The foundations of Stela i, i. e., the manner in which the cruciform chamber underneath is built into the stairway on the western side of Mound 9, are such as to necessitate that both stela and mound were erected at the same time, namely, 9. ii. 15.0.0 (the date of Stela i); and if so, this monu- ment is the earliest surely in situ at the Main Structure. It has already been pointed out (pp. 107, 115, 125) that Stelae EandP, even though the latter is still standing, are neither of them in the positions for which they were originally designed, but that, on the contrary, both had been removed to the Main Structure in ancient times from some other locality, probably Group 9. There remain, however, two other monuments at the Main Structure, Stelae 2 and 3, which are 15 years older than Stela i, and which may possibly be in situ. Of the seven stelae recording the date 9. 11. 0.0.0, these are the only two, except Stela 23 at Group i, which have the human figure carved upon them, all the others being in Class 3. And it is conceivable for this very reason that these two may have had a greater value than the others in the eyes of the ancient inhabitants of the city, sufficient, indeed, to have insured their removal to the Main Structure from some earlier location alter 9. II. 0.0.0, like Stelae E and P. In other words, all four of these monuments were too good to be left standing in outlying groups, once the Main Structure had become the principal settlement in the valley, and they were removed thither sometime after 9. 11. 0.0.0, and the writer believes after 9.1 1. 15. 0.0, when Stela i was erected, and possibly even before the Great Plaza was laid out shortly after 9.12.5.0.0. The earliest stela erected at the Main Struc- ture was probably one of these three, Nos. i, 2, or 3, and of these. Stela i, as the only one of the three indubitably m situ where found, has the best claims to this distinction. In any event, the Main Structure began to become the most important settlement in the valley about 9.1 1. 0.0.0 or a little later, and Mound 9 is probably the oldest construction now extant there in any- thing like its original condition. East Altar of Stela 5. Provenance: Probably not in situ. Originally may have been associ- ated with Stela i at the Main Structure. Now i kilometer west of the Main Structure in front of the east face of Stela 5 at Group 8. (See plate 3.) Date: 9.11. 15.0.0 4 Ahau 13 Mol.' Text, drawing: plate 20, b, and figures 28 and 29. References: Morley, I9i7f, p. 287. Spinden, 1913, p. 161 and table i. ' For other monuments recording this same hotun-ending, see Appendix VIII. i64 THE INSCRIPTIONS AT COPAN. Fic. 2S.- -Top of East Altar of Stela 5 sliowin;; number of fragments into which it is broken. The East Altar of Stela 5 is 1.93 meters in diameter and 51 cm. high. It is broken into 9 pieces, of which 2, Fragments T and R' in figure 28, are still missing. When the writer first visited Copan in 1910, four fragments, S, U, W, and Z, were missing. The first two of these were found in 19 15 in the modern stone wall just across the road from where this altar now stands, in front of the east face of Stela 5. This wall was built some 20 or 30 years ago of faced stone blocks, and even sculptured pieces from ruined buildings in the immediate vicinity; and it had long been suspected that some of the missing pieces of Stela 5 and of its two altars might have been used in its construction. In 191 2 Spin- den found one of the most important pieces of Stela 5, that presenting the upper part of the Initial Series, embedded in this wall; and therefore it seemed advisable to search here for the other missing pieces. In 191 5 a section of 125 meters — 62 or 63 meters on either side of Stela 5 — was taken down to its very foundations, and each stone examined for traces of sculpture; and in the course of this work two pieces of this altar were recovered. Fragment S from the wall itself and Fragment U from just east of Stela 5, half buried in the earth. Two pieces of the West Altar of Stela 5, Fragments Y and Z, figure 32, were also recovered at the same time. Unfortunately the piece presenting the missing part of the Initial Series on Stela 5 itself was not found. In 1917 the second-growth bush north of Stela 5 was felled, and this parcel of land put into tobacco. During the course of the clearing two other fragments of this altar, W and Z, were recovered about 30 meters northeast of Stela 5. Fragment W (see figure 29)- presents Glyphs C, D, and E of the Sup- plementary Series. Fragment Z fitted in between Fragments S and Y. The inscription on this altar is pre- sented upon the periphery, the top and bottom being plain. There is a single band of 15 glyph-blocks, which completely encircles the stone. The order of reading within the individual glyph-blocks is again ' Fragment R originally came from just above Fragment U, and is not shown in figure 28. ^ Fragment W is not shown on plate 20, h, which was made before this piece was discovered in 1917. A drawing of its two glyph-blocks, however, will be found in figure 29 These should occupy the third and fourth dotted squares to the right of the Initial Series in plate 20, b, i. t ., just to the left of Glyph X of the Supplementary Series. 9, — Inscription on Fragment W of East Altar of Stela 5. INSCRIPTIONS OF THE MIDDLE PERIOD. 165 unusual, being like that on Stela 3, i. e., from top to bottom and left to right, and, unlike that on the altar of Stela i, the inscription of which is composed like this, of a single band of glyph-blocks. The text opens on Fragments S and U, and the missing Fragment T, with an Initial Series introducing glyph at a, plate 20, b, shown by the star in figure 28, and the next piece (plate 20, b, and Fragment V, figure 28) pre- sents the cycles, katuns, tuns, and uinals of the corresponding Initial Series number. The cycle coefficient (ba u. h.) is missing, although it doubtless may be restored as 9. The cycle-sign itself is recorded in Ba 1. h. and traces of the clasped hand on the lower part of the face may still be distinguished. The katun coefllicient (b^ u. h.) is perfectly clear, but of unfamiliar form. The tun coefficient (ca u. h.) is unusual; the head-variant numeral is itself very clearly the death's head, but this is preceded by a bar which ordinarily would have the value of 5, presumably S + io, since 5X10, i. e., uniting the two by multiplication, would give an impossible value for the tun coefficient. Leaving this point indeterminate for the present, however, let us continue the inspection of our text. The uinal-sign and coefficient in cb are perfectly clear as o uinals. Note the clasped hand denoting o in the coefficient. Unfortunately the next two glyph-blocks, i. e., d and e, are missing, both being on the missing Fragment T (see figure 28). These, doubtless, recorded the kins and day of the Initial Series and the first two glyphs of the Supplementary Series. To compensate for this rather serious loss, however, the glyphs on Fragments W and X (figure 29 and plate 20, b) are unusually clear, and in a series of 10 consecutive characters they give sufficient data by which this altar may be dated. The first of these, ¥a, is Glyph E of the Supplementary Series, and there follow in ¥b, g, na, ub, and la. Glyphs D, C, X, B, and A of the Supplementary Series, respectively, the last, Glyph A, here having a coefficient of 9. Next, in ib follows the month-sign of the Initial Series terminal date, unusually clear as 13 Mol. Following this in j(7 is a well-known ending-sign, being almost identical with the forms used on Stelae 12 and 2 (compare ja with the next to last glyph in plate 17 and the third to last glyph in plate 18), and after this in ]b, the glyph meaning "End of a hotun." Finally, in Ka and Kb (the latter on Fragment Y) is the day 12 Ahau and another ending-sign. Fragment Z presents no decipherable glyphs, although ua is two katuns and Mb is a familiar grotesque head of unknown meaning. The last glyph-block, o, that is on the left of the Initial Series introducing glyph, is on Fragment S again and has no decipherable signs. Whatever the doubtful coefficients in the Initial Series number may be, there seems little doubt that we have recorded in ib-} the statement that the month of the Initial Series terminal date is "13 Mol," and further, that it fell on a hotun-ending. By referring to Goodman's tables, it will be found that the only hotun in Cycle 9 ending on a month position of 13 Mol within l66 THE INSCRIPTIONS AT COPAN. a range of about 360 years, was 9. 11. 15. 0.0 4 Ahau 13 Mol, the next earlier occurrence being in 8. 13. 10. 0.0, and the next later in lo.io.o.o.o, both impossible dates, so far as Copan is concerned, and both impossible here because the hotun-sign and not the lahuntun-sign is recorded. And finally, the record of the day 12 Ahau just after this in Ka may record the preceding katun-ending, i. e., 9.1 1. 0.0.0 12 Ahau 8 Ceh. Indeed, there is little doubt that 9.1 1. 1 5. 0.0 was the hotun recorded by this Initial Series, in spite of the irregularity of the tun coefficient, and the loss of the kin and day-signs. Let us turn again to our Initial Series, and see how these values for the katun, tun, and uinal coefficients agree with those actually recorded on Fragment V (plate 20, b). The uinal coefficient in cb u. h. is surely o, which agrees with the value for the corresponding coefficient obtained above. The katun coeffi- /y^ ^ cient (b^ u. h.) should be 11; unfortunately only two other occur- /^"^ Yr rences of the head for 1 1 are known, namely, the katun coefficient on ^ ^ Lintel 2 at Piedras Negras i^^ and the katun coefficient on Stela 6 at Yaxchilan ^/'^ ■ Although there (f i appears to be no element common to both of Cj (f| these signs, both B) nevertheless are of the same type, namely, the ^I^ normal human head, and to this extent the head in b^ u. h. may be said to resemble them. The tun coefficient (ca u. h.) is composed of a head variant, clearly the death's head, denoting 10, preceded by a bar, possibly denoting 5, as already suggested. If we may join these two elements by addition, i. e., 10 + 5 = 15, forming a composite numeral, we will have the tun coeffi- cient demanded by the corresponding Initial Series terminal date. But such a combination, i. e., a bar-and-dot numeral joined with a head-variant num- eral, is a new feature in Maya notation, and before it can be accepted, even in the face of the very strong evidence presented by the hotun-ending date in i^j, it is necessary to adduce other examples of its occurrence. Fortun- ately several such may be cited. The customary way to have expressed the number 15 would have been either by 3 bars nnn or, since all the other coefficients in this text are head-variant num- erals, by the head for 10 + the head for 5 ^^®Q. Neither, however, ULIU was employed in the present instance, >^-^S but instead a combination of the two seems to have been used. ^3£D One other instance of this kind has already been noted, namely, the tun coefficient of the Initial Series on Stela 12. (See plate 17, a4, and pp. 135, 136.) Unfortunately this example is doubtful, since the numerical bar inclosed in an oval above the death's head is partially effaced. If it is a bar at all, however, the parallel is complete between these two cases, both being tun coefficients in Initial Series and both expressing the same number, 15, in the same unusual way — a bar joined to a death's head. Other examples, not in Initial Series, however, may be cited. The first and most striking of these is in the Supplementary Series from the Temple of the Cross at Palenque, but in order to bring this point out, it is MORLEY. PLATE I l^ilvSt^K / *i\fe*iS-t'^-iJ2iC.jE2jitiii.ii_ C/^fc w ^^^^^ ...ii ^«s= ^ ; ^:: MORLEY. PLATE 19 V//^7im I fSS o^ ■ :\..td ^ Stela 3. Inscriptions on (a) the east side, and (b) the west side. Drawn from the original. Stela 5. Inscriptions on (c) the south side, and (d) the north side. Drawn from the original. INSCRIPTIONS OF THE MIDDLE PERIOD. 167 necessary to compare this text glyph by glyph with the Supplementary Series from the neighboring Temple of the Foliated Cross. In figure 30 these two Supplementary Series are shown side by side, the upper line being that Fig. 30. — Two Supplementary Series at Palenque, from: a-j, Temple of the Foliated Cross; fl'-/', Temple of the Cross. from the Temple of the Foliated Cross and the lower line that from the Temple of the Cross. Commencing at the right^ and reading from right to left, the last glyph in each line,/ and/', respectively, are the same, namely, Glyph A, the last glyph of the Supplementary Series, shown with a head- variant coefficient of 10 in/ and with a bar-and-dot coefficient of 9 in/'. The next glyphs (to the left), e and e\ are also the same, being in each case the next to the last glyph of the Supplementary Series, Glyph B. The next glyphs, d and d' , are different, as would be expected. As the writer has shown elsewhere,^ the third position from the right in Supplementary Series is occupied by a number of different glyphs, hence the name variable glyph {i. e., Glyph X) by which the signs occupying this position have been designated. The next glyphs, c and c', are the same, being Glyph C of the Supplementary Series, regularly found in the fourth place from the right. Let us pass over the next glyph, b, in the upper line, and the next two, b" , h', in the lower line, and come to the last glyph in each line, a and a' respec- tively, which are the same. Glyph F of the Supplementary Series. Imme- diately preceding these signs in the two texts from which these passages are taken are the month-signs of their respective Initial Series terminal dates. In other words, barring the glyphs passed over, the signs in the two passages are the same and stand, moreover, in the same relative order to each other. Let us next examine the glyphs omitted in our preliminary inspection, b, b' , and b" . The former, b, is a grotesque head standing on its end — a very unusual position for head-variants — with a coefficient of 10, /. c, 2 bars. The right-hand glyph of the pair in the lower line, b" , is exactly the same head as the one in b, and also stands on its end in exactly the same unusual manner, evidently a characteristic of the glyph, whatever it may be. In the lower line, however, at first sight it appears to have no coefficient. 'The regularity in the sequence of the glyphs of the Supplementary Series is more apparent reading from right to left than vice versa, for which reason this order is followed here. See Morlcy, 1916, p. 369. * See Morley, 1916, pp. 374-376. i68 THE INSCRIPTIONS AT COPAN. This leaves but one glyph unexplained, b' , i. e., the one immediately preceding the foregoing. This is composed of a bar, i. e., 5, prefixed to a very clear death's head, which is in every way similar to the death's heads repre- senting the number 10 found elsewhere. We would appear to have in this glyph, then, a composite numeral, a bar-and-dot 5 joined to a head-variant ID exactly as in the tun coefficients on the East Altar of Stela 5 and probably Stela 12. Compare Glyph b' , figure 30, with ca u. h., plate 20, b, and it will be found that the parallel is complete. But we have already seen that in the latter case these are strong — indeed well-nigh irrefutable — reasons for believing that these two numerical elements are joined by addition, and not by multiplication, which, if true, would give for b' the meaning 15, i. c, 5 + 10=15. Moreover, the right-hand glyph, b" , has no coefficient at all, whereas the same glyph in the upper line, b, has a coefiicient of 10. Therefore it seems not unlikely that Glyph b' may be joined to b", making 15 times the gro- tesque head for the lower line instead of 10 times, as it is in the upper line. That these two numerals, one a bar-and-dot and the other a head-variant, are joined to each other by addition and to the glyphs they modify by multiplication, is probably true here. Moreover, such an arrangement completes the parallel between the upper and lower lines, since in this event there would then be only 6 glyphs in each of these Supplementary Series, and the fifth glyph in each, counting from the right, i. e., b and b'b", would then be the same, having a coefliicient of 10 in the upper line and of 15 in the lower line. ~y)J g Fig. 31. — Glyphs showing composite numerals from: n and g, Yaxchilan, Lintel 21; b, Yaxchilan, Lintel in Berlin IVluseum; c, Yaxchilan, Lintel 29; d, Palenque, Temple of the Cross; f, Copan, Stela 12; /, Copan, East Altar of Stela 5. The other examples (see figure 31) are all from Supplementary Series at Yaxchilan, where use of this composite type of numeral seems to have been quite common. The first example (figure 31, a) is from Lintel 21, Struc- ture 22. Here a bar is prefixed to a death's head, this composite numeral being attached to a sign which, the writer has shown elsewhere,^ is a variant of the kin-sign. The whole glyph would appear to record 15 kins. Notice should be taken of the five dots in the death's head, also a characteristic of the death's head in the Palenque text. (See figure 30, b' , and figure 31, g.) The second example (figure 31, b) is from a lintel in the Berlin Museum. Again there is the same bar, here standing above the same death's head; 1 Morley, 191 6, pp. 369, 385-387. INSCRIPTIONS OF THE MIDDLE PERIOD. 169 and again this glyph is attached to the same variant of the kin-sign, the meaning probably being 15 kins. The third example (figure 31, c) is from Lintel 29, Structure 10. In this case a variation has been introduced. The bar, instead of being prefixed to a head, is prefixed to a sign, the principal element of which is five dots. Can this be a substitute for the death's head, which has the same charac- teristic in figure 31, a, a?, and g, a sort of glyphic synecdoche wherein a part, i. e., the five dots, is used for the whole, i. e., the death's head.? At any rate, this glyph is again attached to the same variant of the kin-sign, and the writer has little hesitancy in reading the two characters as 15 kins. The foregoing six examples, i. e., including the one from Palenque and the two from Copan, d, e, and /, figure 31, respectively, are all composite forms for the number 15, that is, single bars prefixed to death's heads. There is one example, however (figure 31, g), of a composite 17, which would seem to indicate that other numbers beside 15 could be formed in this same unusual way. This is also from Lintel 21, Structure 22 at Yaxchilan, and has a bar-and-dot numeral 7 prefixed to a death's head with the same five dots in the upper part of the latter, the resulting number being 17. Reviewing these seven examples, two from Copan, four from Yaxchilan, and one from Palenque, it will be seen that they are fairly well scattered geographically (plate i), and moreover, that insofar as three of them are concerned (figure 31, a, d, and g) the dots in the head appear to be an important characteristic. They constitute, the writer believes, sufficient evidence to prove the existence of these composite numerals, and to demon- strate their use in certain rare instances. Indeed, in the present case the burden of proof would certainly appear to rest upon those who decline to recognize in ca u. h., plate 20, b, the numeral 15, irregularly as it is there expressed, since the record of the month-sign, 13 Mol, and the "End of a hotun" in la-j render any other reading practically impossible. When we attempt to explain why this unusual variant should have been used in this particular text, however, we enter upon uncertain ground. The follow- ing possible explanation, therefore, is advanced only as a suggestion. The sculptor of this monument knew that if he used the head-variant for 15 he would have to carve a death's head with a tun-sign for its head- dress. Inasmuch as all the period glyphs and their coefficients are head- variants, in this text it is probable on artistic grounds that a simple bar-and- dot numeral 15 could not have been contemplated here. On the other hand, to have recorded a head-variant 15 in the space available would have resulted in an unsightly contraction (a flattening) of the death's head, to make room for the tun head-dress, which would have thrown it badly out of line with the heads in the cycle, katun, uinal, and kin coefficients. In short, artistic considerations may have weighed so heavily against such a violation of symmetry that the sculptor took the only other course open to him, namely, that of recording the number 10 as a head variant and then prefixing a bar to it, i. e., 5, making 15, as required by the calculations. lyo THE INSCRIPTIONS AT COPAN. Such a departure from regular practices must necessarily have been due to an unusual circumstance, perhaps such as this, and if this explanation really explains the situation, it is but another example of that broad prin- ciple found applying throughout the Maya hieroglyphic writing, namely, the extraordinarily coercive influence of artistic considerations in glyph delineation.^ But after all, the task of accounting for this unusual departure from regularly established practices is not so important as the recognition of its existence here, from which, in view of the decisive evidence presented, we can hardly escape, and we may therefore accept 9.1 1. 15.0.0 4 Ahau 13 Mol as the date of this altar. A summary of the text follows : A Fragments S, R, and U, Initial Series introducing glyph Bfl Fragment V 9 cycles Bi Fragment V II katuns ca Fragment V 15 tuns cb Fragment V uinals Dfl Fragment T (missing) (0 kins) D& Fragment T (missing) (4 Ahau) lb Fragment X 13 Mol J Fragment X End of a hotun Ka Fragment X 12 Ahau (perhaps 9.1 1. 0.0.0 12 Ahau 8 Ceh) The above date is the same as that on Stela i just described, and this chronologic coincidence at once raises the interesting question as to whether or not this altar may originally have been associated with Stela i instead of Stela 5. It will be shown in the description of the latter that it was not erected until 35 or 40 years later than either of the two altars now asso- ciated with it, which might be taken as indicating that neither was designed originally for use with it. Similarly, the altar now associated with Stela i marks a later hotun-ending, i. e., 9.12.0.0.0 (see p. 176), than the stela itself. It therefore appears not unlikely that the East Altar of Stela 5 itself may have been associated with Stela i originally. The points in favor of this association are summarized below: 1. The same hotun-ending 9. 11. 15. 0.0 is unmistakably recorded on both. 2. The stela with which this altar is now associated (Stela 5) was not erected until 40 years after the date on this altar. 3. The altar now associated with Stela I similarly records a later hotun-ending than Stela i, and one, moreover, which is also recorded on another altar, namely, the West Altar of Stela 5. (See p. 173.) 4. Finally, all the other altars of this type, except the altar of Stela 13 and the altar of Stela 19,^ which is fragmentary, namely, the altars of Stelae E, i, and I and the West Altar of Stela 5, record hotun-endings; and in the case of two, the altars of E and I, they are still associated wiih their original stelae. This whold' question, of course, hinges upon whether Stela i is in situ where it was first found at the Main Structure, or whether it was carried ' Morley, 1915, pp. 23, 24. - As already pointed out (pp. 145, 146), this altar probably brought the Initial Series of the stela with which it is associated, i. e., 9. 10. ig. 15 .0, forward to the next hotun-ending, t. u. h. clearly indicates the former, and the dot on the lower part of the face in Ba u. h. indicates the usual cycle coefficient of 9. The katun-sign and coefficient in b 1. h. are equally clear, the latter, B<3 1. h., being 9. (Note the three dots in the double circle on the lower part of the cheek.) The next glyph-block, c, unfortunately, is badly weathered, though happily both of the coefficients may be distinguished \\ith little difficulty. (See plate 21, ca u. h. and ca 1. h.) Both are alike and prob- ably record the numeral 10. (Note the fleshless lower jaw and prominent upper teeth, both characteristic of the head variant for 10.) Compare these two glyphs with the katun coefficient of Stela 12, for example (plate 17, B}a). The kin-sign and coefficient in d u. h. are entirely effaced, and the day-sign and coefficient in d 1. h. practically so; and with the complete destruction of the next two glyph-blocks, e and f, and the subsequent loss of the month of the Initial Series terminal date, the last means of dating this Initial Series to the exact day are gone. However, even in spite of the loss of the kin coefficient, the other coefficients are sufficiently clear to date it 1 Unfortaiiately the dimensions of this altar were not secured. The drawing in plate 21 shows that ir is 36 cm. high, and the diameter is doubtless between 1.5 and 2 meters. -Maudslay, 1889-1902, vol. i of text, p. 66. 176 THE INSCRIPTIONS AT COPAN. within 19 days, as 9.9.10. 10. ?. Furthermore, since the Ivin coefficient is usually o, we may probably decipher this Initial Series further as 9.9. 10. 10. o 7 Ahau 13 Zac, with 19 days as the maximum error possible. This date is neither a hotun-ending nor even a tun-ending, however, and for that reason, if for no other, it is hardly to be considered as the con- temporaneous date of the altar. Continuing the inspection of our text, the remains of a Secondary Series are clearly to be distinguished in cbji. The uinal-sign shows very clearly in na and part of a tun or katun sign in nb. Unfortunately the coefficients are destroyed. In i is a Calendar Round date, .? ? .? Yaxkin, the second doubtful element probably being the day- sign Ahau. Finally, in j is recorded very clearly and unmistakably "The End of Katun 12." The ending-sign is the familiar hand, with the tassel-like postfix and subfix with curl infix, and the Katun 12 itself has the familiar ending superfix or prefix. By referring to Goodman's tables, it will be found that Katun 12 of Cycle 9 ended on the date 10 Ahau 8 Yaxkin, viz, 9.12.0.0.0 10 Ahau 8 Yaxkin, which agrees exactly with the month-sign recorded in ib, as far as it is preserved; and in spite of the fact that both of the coefficients are missing, we are doubtless justified in restoring them as 10 and 8 respec- tively, and in assigning this altar to the katun-ending 9.12.0.0.0. It seems probable that there was but one Secondary Series on this altar, because a Supplementary Series of the usual length, i. e., 7 or 8 glyphs, would fill most, if not all, of e and f, and there would have been room for no more glyphs between f and g^, which begins the Secondary Series noted above, i. e., cbu. Therefore, if the coefficients of this Secondary Series had been preserved, we could probably have obtained the exact value of the Initial Series terminal date by subtraction. As it is, we may possibly fill in these missing coefficients as follows: 2.9.8.0, viz: 9. 9. 10. 10. o 7 Ahau 13 Zac (?) 2. 9. 8.0 9.12. o. 0.0 12 Ahau 8 Yaxkin The last glyph-block, k, is undecipherable, although ku 1. h. is the head of God C, which also appears on the altar of Stela I at h^. (See plate 22, c.) A summary of this text follows: A-Da 9. 9. 10. 10. o 7 Ahau 13 Zac (?) cbn 2. 9. 8.0 (.?) I 9.12. o. 0.0 10 Ahau 8 Yaxkin J End of Katun 12 We have seen that the East Altar of Stela 5 and Stela i both record the same hotun-ending, namely, 9. 11. 15. 0.0, and therefore originally may have been associated with each other. Furthermore, we have seen that the date of the West Altar of Stela 5 is the same as the date of this altar, namely, 9.12.0.0.0. It therefore seems not improbable that these last two altars also may have been formerly associated with some stela which was erected to commemorate this hotun-ending. It must be admitted, however, that such a stela has yet to be found. MORLEY. PLATE 21 ^^«*=-.- ^ - ^y.^'T'jM ii j) i jj ii w i > ■ ) !■ ^ n[_i ^ ^n .i^g Mtynii un ' gq t ■ L ^i>y n»p»- 'fi ** ' .ne^C i" ilw Altar of Stela i. Inscription on the periphery. Drawn from the original KCBT LITHO, CO., ■ ' Four of the stels in the center of the Great Plaza are arranged m two hnes facmg each other, Stels A and B being directly opposite SteU H and F respectively, the two former facing east, the two latter, west. (See plate 6.) 3 Stephens, 1841, vol. I, map facing p. 133, and p. 158. 222 THE INSCRIPTIONS AT COPAN. 9. 14. 19.8.0 12 Ahau 18 Cumhu 3.0 backward 9. 14. 19.5.0 4 Ahau 18 Muan The latter date, 9. 14. 19. 5.0, is just i tonalamatl (13 uinals) before the closing date on this monument, and is, moreover, a katun anniversary of another date here at Copan, namely, the second on Stela I, which is exactly 3 katuns earlier: 9. II. 19. 5.0 10 Ahau 13 Ceh 3. 0.0.0 9. 14. 19. 5.0 4 Ahau 18 Muan The record of two dates which are exactly i tonalamatl before the ends of katuns in the Long Count might indicate the recurrence of some festival at such times. Perhaps like the last 5 days of the year in northern Yucatan, the closing tonalamatl of each katun was set aside for a particular group of ceremonies.^ If this were true, the dates 9. 11. 19. 5.0 10 Ahau 13 Ceh on Stela I, and 9. 14. 19.5.0 4 Ahau 18 Muan here would mark the beginnings of such festivals. At least, the recurrence of such an unusual date as this arrests the attention and may well have had some such corresponding sig- nificance. The count on Stela A is continued in hub, where a Secondary Series of lo.o is followed by the date 4 Ahau 13 Yax in cii^, BUa. All efforts, however, to reach this by counting lo.o either forward or backward from the preceding date 4 Ahau 18 Muan will prove unavailing, but if lo.o is counted forward from the Initial Series terminal date 12 Ahau 18 Cumhu, the date reached will be found to be 4 Ahau 13 Yax, as recorded: 9. 14.19. 8.0 12 Ahau 18 Cumhu 1 0.0 9.15. o. 0.0 4 Ahau 13 Yax Finally, the record of "End of Katun 15" in Bub, cna proves that the 4 Ahau 13 Yax in cii^, BUa is in fact 9.15.0.0.0 4 Ahau 13 Yax, and the above calculations are therefore correct. There follows in ci2^ the day 12 Ahau and in Biia, 13. ^ the doubtful glyph, probably being a month-sign. There seems to be no reason, however, for the record of such a Calendar Round date here. The remaining side (south) has no glyphs of a decipherable nature except \)ub, which is the lahuntun ending-sign already referred to. (See pp. 61, 88, 116, 183, 188, 195, 200, and Morley, 1917^, p. 197.) ' Says Landa in describing these ceremonies at the end of the year: "It has been said in the preceding chapters that the Indians [of Yucatan] commenced their years from these days without name [the last five days of their year July 11-15 inclusive, Old Style, July 21-25 New Style], pre- paring for them with vigils in order to celebrate the feast of their new year. And besides the preparations which they make for the feast of the demon Uayebab, for which they go out of their houses, the rest of the preparations were to go out of their houses very little in these five days, except to offer the gifts and objects of the general feast to their devils and to the others in their temples. These things which are thus offered, they never take back for their own uses, nor anything which may have been offered to the demon, for whom they buy incense to burn. During these days they do not comb or wash themselves, the men do not remove the vermin from their bodies, and the women do no menial or heavy work, because they fear some evil would result if they did it." (Landa, 1881, pp. 96, 97.) MORLEY PLATE 24 a. Glyphs in the north gallery of Structure 18. b. Altar S, front and adjacent side on the right. a,^i;.:i?^£:ai^' c. Altar W, top. ^^'\^'^ e. Altar P. f. Altar Q'. INSCRIPTIONS OF THE GREAT PERIOD. 223 Attention should be called to the peculiar form of the last glyph of the Supplementary Series in this text at A.9a: ^^ This sign has the coefficient 9, one of the only two possible (9 or I o), ^1^ and stands in the regular position, i. e., immediately preceding the bood month-sign. The main part of the glyph, however, is entirely different from the usual forms. (See Morley, 1916, plates i-io, Glyph A.) So far as the writer knows, this variant occurs as the last glyph of the Supplementary Series in but two other places, on Stela N, also here at Copan, and on Slab 6 of the Hieroglyphic Stairway at Naranjo. On Stela N the coefficient is 10 instead of 9.^ This same variant but with another superfix appears again on Stela A at Cbb. The reason for its use in these three texts and in no other is unknown, but that it is the same as the usual form seems probable from the fact that: (i) This variant in all three cases is the last glyph of the Supplementary Series, and in two immediately precedes the month-sign; and (2) In all three cases it has the coefficient 9 or 10, the only ones possible with the last glyph of the Supplementary Series. A summary of the chronological parts of this text follows: North side, ai-A4^, aqJ 9. 14. 19. 8.0 12 Ahau 18 Cumhu West side, Bia 3.0 backward C2 9. 14. 19. 5.0 4 Ahau 18 Muan Bii^ 10. o from Initial Series ciib, Bi2a 9-iS. o. 0.0 4 Ahau 13 Yax Biii, c\2a End of Katun 15 cizi, Bi3a 12 Ahau 13.' The possible significance of the second date above has already been explained. The next to last date, 9.15.0.0.0, is the katun-ending this monu- ment was erected to commemorate. Stela A is the most elaborate monument yet encountered. The carving of the glyphs is well-rounded, giving a finish and softness of outline to the glyph-blocks not encountered in any of the earlier sculptures. There are 13+26 + 13 = 52 glyph-blocks in the entire text. According to Spinden, there are a few traces of a monolithic altar of undetermined character in front of this monument.'- Stela B. Provenance: About the middle of the Great Plaza. Main Structure. (See plate 6.) Date: 9.15.0.0.0 4 Ahau 13 Yax.' Text, (a) photograph: Maudslay, 1889-1902, vol. i, plates 33, 35, 36, 38, 39,0. (6) drawing: Arnold and Frost, 1909, plate opposite p. 284. Maudslay, ibid, plates 34, 37. Morley, 1915, plate 7, a. Stephens, 1841, vol. i, 3 plates after p. 156. References: Arnold and Frost, 1909, p. 284. Bowditch, 1910, pp. loi, 183, and table 29. 1 See Morley, 1916, plate 10. Nos. 77 and 80, Glyph A, and p. 370. - See Spinden, 1913, table 1. ' For other monuments recording this same hotun-ending, see Appendix VIII. 224 "^^^ INSCRIPTIONS AT COPAN. Stela B — ^continued. References— cont'd: Galindo, 1834, Appendix XI, p. 598. Goodman, 1897, p. 130. Gordon, 1896, p. 35. Gordon, 1902, p. 167. Maudslay, 1889-1902, vol. I of text, pp. 42, 43. Morley, 1915, pp. 167-169. Seler, 1902-1908, vol. I, pp. 751, 752. Spinden, 1913, pp. 157-159, and table I. Stephens, 1841, vol. i, p. 156. Thomas. 1900, pp. 776, 801. Stela B stands near the center of the Great Plaza facing east. It is 3 . 58 meters high and more than a meter wide. The front is sculptured with a human figure of heroic size, whose somewhat Mongoloid cast of counte- nance has given rise to a flood of ill-considered speculations regarding the possible Asiatic origin of the Maya civilization. One of the more recent sup- porters of this extravagant hypothesis, long since relegated to the rubbish- pile of scientific discards, is Arnold, who writes concerning this monument as follows: "Here as pointed out on page 268 are carvings so strikingly Oriental that one can not doubt their origin. The faces of the figures on the stelae are the faces one can see today in Cambodia and Siam. The dress, the ornamentation, the turban- shaped head-dress (found on no other carvings but these) are all purely ancient Indo-Chinese."^ Even as sober and restrained a writer as Stephens, who calls this monu- ment N in his map, was led astray by the apparent resemblance of certain decorative elements on this monument to elephant trunks: "The two elements at the top appear like the trunks of an elephant, an animal unknown in that country."- More recently Elliot Smith has revived this highly improbable identi- fication, finding detailed anatomical similarities between this decorative element on Stela B and the trunk of an elephant. He has been ably answered, however, by Tozzer, Spinden, and Means.^ It is hardly necessary to point out that any attempt which seeks to establish direct cultural connection between the Maya and any old-world civilization, either Egyptian or Mongolian, is quite at variance with the results of modern research in this field. And yet the superficial similarities of the Maya to these civilizations are such as to win for this now-exploded hypothesis new adherents from time to time. 1 Arnold, 1909, p. 284. . , • 1 l- 1 ■ = Stephens, 1841, vol. i, p. 156. Spinden was the first to point out the true nature ot this element, which is the beak ot the blue macaw somewhat exaggerated in length. 3 This controversy arose through the publication by Elliot Smith in Nature for November 25, 1915 (Smith, 1915-1916, pp. 340, 340,of an article entitled "Pre-Columbian Representations of the Elephant in America," and additions in Nature for December 16, 1915 Uhid., p. 425). Tozzer (1916, p. 592), Spinden (1916, pp. 592, 593), and Means (1916, pp. 533, 534) hastened to refute this extraordinary hypothesis, in which the writer believes they were successful; Spinden {op. cit.) shows clearly that the element in question is the beak of the blue macaw. Other Elliot Smith contributions to the discussions are, 1915-1916, pp. S93-S95> and 1916-1917, pp. 190-195, 241-246. INSCRIPTIONS OF THE GREAT PERIOD. 225 Returning to the description of Stela B, the back presents a huge gro- tesque face, the eyes and mouth of which each contain a single glyph-block, making three in all. The center of the forehead is occupied by a human figure (see figure 36) seated cross-legged, which from this time on seems to have been a favorite subject of the Copan sculptors. The one on Stela B is probably the earliest example ofthe type yet found.' Each of the narrow sides of this stela con- tains a single line of glyph-blocks, 14 on the north side and 13 on the south side, which, with the 3 on the back, make a total of 14 + 13 +3 = 30 for the entire text. This arrangement of the design is new. It is most like Class 4, i. e., a human figure on the front and glyphs on the back and two sides, except that in the case of Stela B the back has a large grotesque face with only three glyph-blocks — one in each eye and one in the mouth — in place of the usual two columns of glyph-blocks. Stela B is, in fact, a transitional type between Classes 4 and 5, more closely resembling the former, however, than the latter, to which, therefore, it has been assigned here. The inscription on the north side opens with an Initial Series introducing glyph at Ai and this is followed by the Initial Series 9.15.0.0.0 4 Ahau 13 Yax in A2-A8: Initial Series introducing glyph 9 cycles 15 katuns o tuns o uinals o kins 4 Ahau 13 Yax This reading is perfectly clear and ofters no difficulties. The south side has only two glyphs of known meaning: b5, B6, "The End of Katun 15." Several others, however, are of familiar form as Bja, b7, b8, B12, and bij. It should be noted that there is no Supplementary Series on Stela B, the month of the Initial Series terminal date immediately following the day. A summary of the text follows : Fig. 36. — Seated human figure on back of Stela B. Al A2 A3 hi, A5 a6 A7 A8 North side, ai South side, B5, b6 9.15.0.0.0 4 Ahau 13 Yax End of Katun 15 It is evident from the foregoing that Stelae A and B both record the same hotun-ending. Stylistically considered, however. Stela B is a little later than Stela A. ' Other sculptures showing this figure are: Altars Q, T, S, and L, the step in Temple II, and two sculptures found on the terrace east of the Great Plaza. The last are now in the Museum of the Normal School at Teguci- galpa. Of a closely related type may be mentioned the figures on Altars L, Q, and R at Ouirigua. 226 THE INSCRIPTIONS AT COPAN. Spinden, in describing the latter, states: "In style the principal figure like Stelae A and D .... grotesque face occupies entire back of stela. Head-dress of principal figure a turban .... feet turned at slightly less than i8o°, hence the sculpture later than Stela A."^ Indeed, some of the carving on Stela B amounts to sculpture in the round, the macaw-beaks mentioned above standing quite free from the body of the monument. Stels A and B were doubtless erected at the same time, although it is quite possible, in fact likely, that the latter, because of its technical superior- ity, was carved after the former. Catherwood's drawing of the text on the north side of this monument, made in 1839, is so accurate that it is possible to read the date from it.^ When it is taken into consideration that the Maya hieroglyphic writing was a sealed book at the time he visited Copan, and that he knew nothing about the subject-matter of the glyphs he drew, such accuracy is remarkable. It is also possible to read his drawing of the Initial Series on Stela D.^ Altar S. Provenance: On the summit of the pyramid (Mound 29, Maudslay's numeration) at the southeast corner of the village plaza (Group 9). (See plate 3 and figure 22, /.) Date: 9.15.0.0.0 4 Ahau 13 Yax.^ Text, (fl) photogiaph: plate 24, b. (fe) drawing: Maudslay, 1889-1902, vol. i, plate 94. Morley, 19x5, figure 81. References: Bowditch, 1910, pp. 95-97, 128, 129, 179, and table 29. Goodman, 1897, p. 134. Maudslay, 1889-1902, vol. i of text, p. 60. Morley, 1915, pp. 231-233. Seler, 1902-1908, vol. I, pp. 753, 754. Spinden, 1913, table I. Thomas, igoo, pp. 788, 802. Altar S was found by Maudslay on the summit of the pyramid at the southeast corner of the plaza of the modern village (Group 9). (See figure 22, /.) The Peabody Museum photographs also show that this was its original position. Subsequently it was removed to the cabildo on the west side of the plaza, and it is now in the museum there. It is a rectangular block of stone about I meter long, 76 cm. wide, and 38 cm. high. The top is sculptured with a seated human figure holding some object in its extended right hand, the four sides being covered with glyphs, three glyph-blocks on each of the long sides and two on each of the short sides, a total of 10 for the entire text. The Initial Series commences on one of the long sides and is concluded on the adjacent side to the right. (See plate 24, b.) The date recorded is very clearly 9.15.0.0.0 4 Ahau 13 Yax, as shown at top of next page. 1 See Spinden, 1913, table I. - Stephens, 1S41, vol. i, last plate between pp. 156 and 157. -' Ibid., plate facing p. 153. 1 For other monuments lecotding this same hotun-ending, see Appendix VIII. INSCRIPTIONS OF THE GREAT PERIOD. 22/ A Initial Series introducing glyph B 9 cycles ca 15 katuns cb o tuns Da o uinals T)b o kins Ea 4 Ahau E^ 13 Yax Maudslay's drawinj^^of the katun coefficient (1889-1902, vol. i, plate 94, ^, c<3 u. h.) incorrectly shows 13. The original, however, is perfectly clear, and shows that it is composed of three bars, i. e., 15. (See plate 24., b.) Passing around to the next side, there is found at ca, 5 katuns, and at H the date 7 Ahau 18 Zip. If 5 katuns are counted forward from 9.15.0.0.0 4 Ahau 13 Yax, the date reached will be lo.o. 0.0.0 7 Ahau 18 Zip; and the "End of Cycle 10" is in fact recorded at la, the first glyph on the remaining side. The last three glyphs are unknown, except that }a 1. h. may be the sign for the planet Venus. A summary of this short text is given below: A-E 9.15.0.0.0 4 Ahau 13 Yax ca 5.0.0.0 H 10. 0.0.0.0 7 Ahau 18 Zip la End of Cycle 10 We have here two dates nearly 100 years apart, and the important ques- tion at once arises, which indicates the time this altar was made? Which of the two was its contemporaneous date ? In general, when two or more dates are found on the same monument, the latest corresponds with the date of erection, that is, the latest is the "contemporaneous date." When there is only one hotun-ending on a monument, it is invariably the date upon which the monument was erected or dedicated. The writer knows of no exceptions to this rule. Many monuments of this latter kind, however, begin with other dates which are earlier than the hotun-endings they commemorate. Take, for example. Stela 36, at Piedras Negras, the Initial Series of which is 9.10.6.5.9 8 Muluc 2 Yaxkin, while its final date is 9. 11. 15. 0.0 4 Ahau 13 Mol, a hotun-ending 28 years later. Another case in point is that of Stela E at Quirigua, the Initial Series on the west side of which is 9. 14. 13. 4. 17 12 Caban 5 Kayab, while its final date is 9.17.0.0.0 13 Ahau 18 Cumhu, a hotun-ending 46 years later. In texts of this character, the final date, if a hotun-ending, is invariably the contemporaneous date of the monument. The case is somewhat difi^er- ent, however, when more than one hotun-ending is recorded upon the same monument, as here, and it is necessary to examine this condition further. Other monuments of this kind, besides Altar S, are: Stelae J and 8 here at Copan, Zoomorph G at Quirigua, and Altar i and Stelae 4 and 7 at Piedras Negras. The full discussion of these monuments would carry us too far afield to attempt here, but the conclusion to which they point may be briefly stated as follows: The earlier rather than the later hotun-ending in such cases is invariably the contemporaneous date, and the later one was 228 THE INSCRIPTIONS AT COPAN. usually "future time" wheu it was recorded. In a sense, the closing dates on these seven monuments are prophetic, that is, they probably represent priestly calculations, which dealt with the future, even at the time they were recorded. It is to be noted further that the later hotun-endings in these cases are invariably of unusual importance. For example, the closing dates on Altar S and Stelae 8 and J at Copan, Zoomorph G at Quirigua, and Altar i at Piedras Negras all record "The End of Cycle lo." We shall see later that katun-endings were commemorated more elaborately than other hotun- endings; and the end of Cycle 9, i. e., 1 0.0.0.0.0 7 Ahau 18 Zip, the period which had witnessed the first great development and cultural florescenceof the Maya, must have been a date of unusual importance to which all were looking forward, especially the priestly astronomers in whose hands lay the reckoning of time. For this reason it is not surprising that the " End of Cycle 10" was recorded more than once before it had actually come to pass.^ A particularly appropriate time for doing this was in 9.15.0.0.0, the date of the Initial Series of Altar S, at which moment the current cycle (Cycle 9) was then exactly three-quarters over, and the beginning of Cycle 10 just a quarter of a cycle off. The Initial Series of Stelae 4 and 7 at Piedras Negras record lahuntun- endings. Stela 4 having the date 9.13.10.0.0 7 Ahau 3 Cumhu and Stela 7 the date 9.14.10.0.0 5 Ahau 3 Mac. In both these cases, however, 10 addi- tional tuns are recorded after the Initial Series terminal dates, and these are followed by the dates which end the next katuns, i. e., 6 Ahau 13 Muan End of Katun 14 in the first case, and 4 Ahau 13 Yax End of Katun 15 in the second case. If the final date is to be regarded as the contemporaneous date of each stela, there will then be tzvo monuments at Piedras Negras (Stelae 4 and 3) commemorating the katun-ending 9.14.0.0.0, and two (Stelae 7 and 11) commemorating the katun-ending 9.15.0.0.0, but none commemorating the lahuntun-endings 9.13.10.0.0 and 9.14.10.0.0, the Initial Series dates of these two stelae. The improbability of this is apparent at a glance, and it seems perfectly safe to assume that the earlier hotun-ending on each stela was the contemporaneous one and that the final one was simply a calculation of the priests to show how far off the nearest katun-ending was at that time. The final dates on all these monuments are doubtless prophetic, in that they still lay in the lap of the gods when they were recorded. In the case of Altar S, the particular monument which provoked the above digression, stylistic evidence, moreover, is not lacking to support the conclusions reached on chronologic grounds. Spinden places this altar in the same group with Stela B, the date of which, as we have already seen, is also 9.15.0.0.0. ' Cycle 10 is also recorded as a Period Ending date on Stela li at Seibal, altlioui;li the contemporaneous date of that monument is one katiin later, i.e., lo. i.o.o.o, (See Maler, igoS, plate 9, glyphs E2, ¥2a.) INSCRIPTIONS OF THE GREAT PERIOD. 229 "Altar S bears same date as Stela B — only glyphs — carving of glyphs shows much the same progress as carvmg of human figure — block-like quality becomes less evident — sharp edges rounded, and modelling apparent."^ In view of all the evidence presented, it seems highly probable that Altar S was erected not at the later hotun-ending recorded upon it, but at the earlier one, that is at 9.15.0.0.0 4 Ahau 13 Yax instead of lo.o. 0.0.0 7 Ahau 18 Zip,^ and, indeed, that originally it may have been associated with Stela A or B at the Main Structure. It will be seen from the summary of this inscription on page 227 that out of a total of 18 glyphs-'' the meanings of 11 are completely known (b, ca, cb, Da, Db, Ea, Eb, ca, na, nb, and la) and the function of one other, i. e., the Initial Series introducing glyph (a), is clearly understood. This leaves only six, or 33.3 per cent, of the text undeciphered. Three of the undeter- mined glyphs (fa, Gb, and ]h) have the familiar ending superfix {i^W^, also seen above \a, the glyph recording "End of Cycle 10." It seems probable from these glyphs, therefore, that other and still unknown Maya time-periods also came to an end on one or other of the two dates recorded on this altar. The first glyph showing this ending-sign superfix, Ya, (^^ follows immediately after the Initial Series terminal date; and it is not at ^-^ all un- likely that some unknown time-period, whose sign is expressed W^ by Fa, came to an end on the date 9.15.0.0.0 4 Ahau 13 Yax. The prefix of Ea is also another ending-sign and the element between the superfixial ending- sign and the main glyph is a familiar form tor o. The record of two ending signs and o in the same glyph strongly suggests that such a sign designates the end of some time-period. The next glyph with this ending-sign superfix is g/; (^g) standing between the "5 katuns" and the date "7 Ahau 18 Zip". This f^^ may in- dicate that some time-period expressed by gIj is contained t^ffij an even number of times in five katuns; that is, it came to an end in this period, ob is apparently the day-sign Imix f^^\ with the day-sign Ahau (7^^ ap- pearing in place of its central I (i|jj|j element. These two days ll^p'j are the first and last of all Maya GE© time-periods, Ahau being SZO the closing day and Imix the starting day of all known periods greater than the kin. If 7 Ahau 18 Zip closed Cycle 10, as we have seen, then 8 Imix 19 Zip began the following cycle, g^ may therefore be a general glyph indicating the end of one time period and the beginning of another. The last ending-sign appears at ]b, ^^ » where it is attached to a com- posite glyph, the main element of which T^W] also appears in the next the latter finally reached the altar and the 15 steps in situ at the base of the mound (see plate 25, b, and figure 37, ce), and it then became apparent that the steps fg, which had at first been taken for the original stairway, were not fn situ at all, but had slipped down from some higher position. What had happened appears very clearly in the east-and-west cross-section of Mound 26, through the Hieroglyphic Stairway, shown in figure 37. The ' Gordon (1902, pp. 15^, 154) mentions a landslide on the north side of Mound 26 which scooped out almost the whole of this side, the debris being piled at the bottom of the slope. This side seems to have had only a ter- raced treatment, and its partial destruction could hardly have "much injured the effect of the ruins." Cer- tainly the destruction of the Hieroglyphic Stairway did far more damage. MORLEY PLATE 25 a. The Hieroglyphic Stairway on the west side of Mound 26 before excavation. J-^^ fi^^eTia^^!? b. A nearer view of the same after excavation. Courtesy of the Peabody Museum. INSCRIPTIONS OF THE GREAT PERIOD. 24I line ab represents the western slope of Mound 26 as found by Owens in 1892; cd, the original position of the stairway; and fg the 15 consecutive stepson the surface of the slope, which were mistaken for the oiiginal stairway at the outset of the work. These, however, were not in situ, as the excavation proved, but had slid down from some position nearer the top of the stair- way, designated in figure 37 as xy. There is left, then, of the original stairway, as a result of this landslide, two sections of 15 consecutive steps each,i ce and/g (?. e., xy), figure 37, and two other sections of unknown lengths, ex and yd, which have fallen to the bottom and are hopelessly confused. The fragments of these were carefully removed by the Fourth Peabody Museum Expedition from the debris at the base of the stairway and were laid in the court, where they may be seen to-day in indescribable and, the writer fears, inextricable confusion, the wreckage of America's greatest aboriginal effort in the science of writing. The fifteen consecutive steps not in situ but in sequence (see fg, figure 37) were carefully removed block by block and reassembled in the court below in their relative positions.- These, together with the 12 or 15 in situ, are reproduced in plates 5 and 6 of Gordon's monograph on the Hieroglyphic Stairway. In plates 12 and 13 some of the disconnected fragments are shown, although these are a very small fraction indeed of the rest of this inscription. It is apparent from the foregoing that we have preserved in its original order only about one-third of the inscription, i. ^.,30 out of 90 steps, and even these are not all consecutive, as we have seen, half coming from the bottom of the stairway and half from some unknown position higher up. Since, from the very nature of the case, the lowest steps were built first, and moreover, since they are the only ones now in situ, this section will be described first. Date l.^ The first three steps (Gordon, 1902, plate 6, A, B, and C) show no decipherable glyphs, but the fourth (D) opens with an Initial Series intro- ducing glyph at A.^ (See plate 26, b.) Then follows in b-d, ib the Initial Series 9.5.19.3.0 8 Ahau 3 Zotz: A Bd Bb Initial Series introducing glyph 9 cycles 5 katuns ca cb 19 tuns 3 uinals Da vb lb kins 8 Ahau^ 3 Zotz ' Gordon reproduces only 12 steps in situ (Gordon, 1902, plate 6) as against the 15 of which he speaks on p. 157, making a total of 27 instead of 30 consecutive steps. This is doubtless due to the fact that the three top steps found in situ, the thirteenth, fourteenth, and fifteenth from the bottom, not counting the plain bottom step, were so fragmentary that it was not thought worth while to publish them. ^Gordon, 1902, p. 157. 'Gordon also calls this Date i in his monograph on the stairway. (Gordon, 1902, p. 169.) ■" Since each step is given a different letter, no vertical numerations of the glyph-blocks, as .-M, bi, ci, etc., is necessary, and such numbeis will therefore be omitted. 'The kins and day of this date are here recorded on the same block of stone. At the two ends, i. <•., before the kin-sign and after the day-sign, are two unsculptured bars from 5 to 6 cm. wide. It would almost appear as though these ends had never been finished. 242 THE INSCRIPTIONS AT COPAN. For the most part these glyphs are clear and require no comment. The bar in both the cycle and katun coefficients has an interior decoration, thus, (' ^1 : and care should be taken not to mistake it for two bars instead of one.' The uinal coefficient is either i, 2, or 3, with by far the best reading at 3. The kin coefficient looks like o, and as the day-sign is clearly Ahau, it can be nothing else. Even assuming that the uinal coefficient were either i or 2, this date can only be 20 or 40 days earlier at the outside. Our preliminary inspection of the Initial Series number, therefore, limits us to three possible readings, the extremes within 40 days of each other: 9. 5. 19. 1. o 7 Ahau 3 Uo 9.5.19.2.0 I Ahau 3 Zip 9.5.19.3.0 8 Ahau 3 Zotz An inspection of the terminal date in T>b, \b, however, at once clears up all uncertainty as to the date recorded, and eliminates the first two, leaving the third as the only reading possible here. The day is clearly 8 Ahau, and although the month in \b immediately following Glyph A of the Supplementary Series in \a (not shown in plate 26, b) is partially effaced, enough remains to show that it is 3 Zotz, and that this Initial Series therefore can only be 9.5.19.3.0 8 Ahau 3 Zotz. Gordon, through a misidentification of the uinal coefficient as 12 instead of 3, and of the month as Mac instead of Zotz, reaches a date 9 uinals later, namely, 9. 5. 19. 12.0 6 Ahau 3 Mac. Even admitting that the head in ib looks as much like Mac as Zotz, which it does not, the uinal coefficient can not be 12, as it is clearly i, 2, or 3. Gordon reads it as 12 only by following an error of Goodman, who assigns the value 10 to the element above the month-sign: (50^=7 -^ ^'^^^ element, however, is clearly non-numerical, as can be proved in a number of instances." Indeed, the very passage from which Goodman derived his value of 10 for it, has an entirely different and demonstrable meaning.^ In short, it is quite certain that, whatever it may mean, it in no way affects the numerical value of the coefficients of the glyphs in which it appears, and consequently cb here is 3 and not 13 uinals. Without attempting to explain at this point why such a very early date as 9.5.19.3.0 should be recorded on such a late construction as the Hiero- glyphic Stairway, let us pass to the consideration of the other dates of this text, reserving explanation of its probable meaning until all the evidence has been presented. 1 Goodman, 1S97, p. 130. ,, , ., , r 1 ■ ■ ■ ■ 1 °- An identical case is found in the Initial Series on Stela n at Ya.xchilan. In A3 of tins inscription, where the tuns of the Initial Series are recorded, this same element stands between the period-glyph and its coefficient of i: The context here clearly proves that the tun coefficient is i and not 11, as it would have to be if Goodman's decipherment of this element were correct. This element also appears at D3 in the same text between the month-sign (Tzec) and its coefficient (8), without, however, changing the numerical value of the latter to 18, also proved by the accompanying calculations, s'see Stela C, p. 350, where it will be found that the date which has this element is 5 Ahau 8 Cumhu, probably 9.17.2.0.0 5 Ahau S Cumhu, and therffore, if Goodman were correct in his decipherment of this element as 10, the day here would be 15 Ahau, clearly an impossible value. QP rn II li^ lml-°- INSCRIPTIONS OF THE GREAT PERIOD. 243 r^nrrn Date 2. Passing over the glyph-blocks following ib in Step D, we reach in ob, p a Secondary Series composed of katuns, tuns, uinals, and kins. The katun coefficient, pb, is surely under 6, and an examination of the original disclosed three rather wide stumps, i. c, i, 2, or 3 katuns. The tun coefficient is again surely under 6. The numeral is broken, but 5 appears to be the best reading. Both ends are gone, but there is hardly room for a dot to the right. The kins and uinals are in cb. The uinal coefficient, though badly effaced, was high; 15 seems to be the best reading. Careful study of the coefficient to the left (;'. e., the kin coefficient) showed i bar and 3 dots, the upper one being almost entirely destroyed. The best read- ing of this number would therefore appear to be i, 2, or 3.5.15.8; but before deciding definitely let us ex- amine the corresponding terminal date at s. (See figure 38, a.) The day coefficient is 8, and the day- sign looks like Ahau, but is Lamat. This is not only proved by the cal- culations, but the ear-ornament of ^ the head in sa is the same as yA the ear-ornament in the day- ^ sign in Date 26 on this same stairway (see plate 27, Gb), which is surely Lamat. For other occur- rences of this Venus variant of La- mat see note 2, page 150, and fig- ure 39. The month coefficient is 6, and the month-sign, though very unus- ual, looks more like Chen or Mac than anything else. If the Second- ary Series in ob, p is 1.5. 15. 18, and is added to Date i, 9.5.19.3.0 8 Ahau 3 Zotz, it will be seen that the month-sign here must be Mac. SKDO Fig. 38.- -Dates from the Hieroglyphic Stairway: a. Date 2; h. Date 3; c, Date 7; d. Date 8; c. Date 12; /, Date II; g. Date 13; /;, Date 14; I, Date 21. A-D, \h 9.5.19. 3.0 8 Ahau 3 Zotz ob, p I. q.15.8 s 9.7. 5. 0.8 8 Lamat 6 Mac Since use of the katun coefficient in p/; as i agrees with the best reading of the date in s, the second and third possible values for the former, 2 and 3 respectively, may be disregarded and the reading accepted as given. 244 THE INSCRIPTIONS AT COPAN. Another consideration also supports the selection of i as the katun coefficient in pb. The use of 2 or 3 gives rise to impossible values for the terminal date actually recorded in s, viz: 9.5.19. 3.0 SAhausZotz, 01-9.5.19. 3.0 8 Ahau 3 Zotz 2. 5.15.8 9.8. 5. 0.8 6 Lamat 6 Mol, 3- S-iS-8 9.9. 5. 0.8 4 Lamat 6 Zip In the first case, although the day coefficient might possibly be 6, the month-sign can not be Mol, and in the second case, the day coefficient can not be 4 and the month-sign can not be Zip. More important than either e X g Fig. 39. — Venus variants of the day-sign Lamat from: a, Palenque, Palace, Stairway of House C; b, Copan, Stela J; c. Hieroglyphic Steps of Mound 2; J, Stela 23; e, fragment Hieroglyphic Stairway; /, Hier- oglyphic Stairway, Date 2; g, Hieroglyphic Stairway, Date 3; h, Hieroglyphic Stairway. Date 24. of these considerations, however, is the fact that the next step (E) opens with the Initial Series 9.7.5.0.8 8 Lamat 6 Mac, i. e., Date 3, which is identical with the above reading of Date 2. Date 3.' The fifth step (E) opens with an Initial Series just below the one on Step D. Although this is much damaged, the reading suggested is probably correct. The Initial Series introducing glyph and 9 cycles are clear at a and B(3 respectively. The katun coefficient, ^b, always the most important factor in deciphering dates, is equally clear as 7 pcr^c- • The tun coefficient, ca, is entirely effaced. The uinal coefficient, ^ ■ — c^, is doubtful, c;;:x:z=r:q the best reading being o. The kin coefficient in Da is entirely gone. ■-'' ' ' ' Fortunately, the day-sign in i>b is very clearly Lamat (note the Venus-sign in the ear-ornament in figure 38, b), and therefore the destroyed kin coeffi- cient in Dfl must have been 8. The day-coefficient is either 6, 7, or 8, with the best reading as 8. Finally, cb is Glyph A of the Supplementary Series and Ha, the month, is 6 Mac (see figure 38, b). Collecting these values, we have: 9.7. ?. ?.8 6, 7, or 8 Lamat 6 Mac. Referring to Goodman's tables, it will be found that the only places in Katun 7 where all the necessary conditions imposed ^ Gordon (1902, p. 171) calls this Date 2. INSCRIPTIONS OF THE GREAT PERIOD. 245 by the terminal date are fulfilled are 9.7.5.0.8 8 Lamat 6 Mac and 9.7.17.3.8 7 Lamat 6 Mac. But the uinal coefficient can not be 3, and, moreover, the first reading is exactly the same as Date 2. In view of these facts, the only reading at all probable here is the first suggested above: A Initial Series introducing glyph Bfl 9 cycles Bb 7 katuns ca 5 tuns cb uinals Da 8 kins i>b 8 Lamat Ha 6 Mac The purpose here would appear to have been to have Step E open with the same date, declared as an Initial Series, as the closing date on D, the next step above. How far this practice may have extended to other steps on the stairway it is now impossible to say, owing to the destruction of the greater part of the inscription, but such appeals to have been the case here at the bottom, at least. Gordon suggests two readings for this date: 9.8.8.6.5 7 Chicchan 8 Mac and 9.8.12.7.5 11 Chicchan 8 Mac, favoring the former.^ However, since each demands that the katun coefficient be 8 instead of 7 as actually recorded, both may be rejected. Dates 4 and 5.^ Dates 4 and 5, at the end of Step E and the beginning of Step F, are almost entirely effaced, and exact decipherment is difficult, if indeed not impossible. The interpretation suggested depends upon the postulate just noted, that the closing date on Step E is the same as the opening date on Step F, and connected by the Secondary Series in n-q (Step E) with the Initial Series date at the beginning of Step E. We have already seen that such a relation exists between the beginning and closing dates on Step D and the opening date on Step E, and if this same relation obtains between Dates 4 and 5, it is possible to reach three readings for each of them, one of which must be correct under the postulate. The accuracy of this postulate is by no means established, and since the glyphs are badly effaced, this read- ing, although probably correct, should be accepted with reservation. Passing over h^-m (Step E), we reach in n-q a Secondary Series followed by a date at s^, T(2(?). The former is composed of katuns, tuns, uinals, and kins, of which only the tun coefficient in p is certain. This is 9. The uinal coefficient in o is a head variant @Q and is clearly above 10. The best readings are 14, 16, or 17. The fn katun coefficient is a small head of human aspect. As will appear in ^?j connection with Date 5, if our pos- tulate is correct, this coefficient must be either i, 2, or 3. A day-sign is recorded at s^, the coefficient of which is best as i, 2, or 3, and there may be a month-sign at ra, although this is doubtful. If so, the best reading of its coefficient is also i, 2, or 3. 'Gordon, 1902, pp. 171, 172. ^Gordon (1902. p. 172) calls this Date 3. 246 THE INSCB-IPTIONS AT COPAN. Step F opens with an Initial Series, the only surely decipherable parts of which are the cycle, and katun coefficients and the month. The cycle coefficient Ba, is 9, and the katun coefficient, Bb, is surely above 5 and under 10, I. e., 6, 7, 8, or 9. Of these, 9 is the best reading, as there appear to be traces of four dots above the one bar ( ■ ■" ^ "■ j . The month is surely 18 Kayab, and the day coefficient is 6, 7, or 8. Under the terms of our postulate that Dates 4 and 5 are the same, and that the Secondary Series recorded in N-Q indicates their distance from the Initial Series on Step E, i. e., Date 3, there are only three possible values for Dates 4 and 5, as follows: Date 4. (9. 7. 5- o. 8 8 Lamat 6 Mac) (9.7. 5. o. 8 8 Lamat 6 Mac) I. 9.11.17 2. 9.16.17 Date 5. 9. 8. 14. 12. 5 12 Chicchan 18 Kayab 9.9.14.17. 5 6 Chicchan 18 Kayab Date 4. (9. 7. 5- o. 8 8 Lamat 6 Mac.) 3- 9- 3-12 Date 5. 9.10.14. 4. o 12 Ahau 18 Kayab. Since the uinal coefficient in the Secondary Series at o on Step E can be neither 11 (first possibility above) or 3 (third possibility above), the sec- ond reading, 9.9. 14. 17.5 6 Chicchan 18 Kayab, is the only one possible under the postulate. Several minor points tend to confirm this reading. The uinal coefficient at o, as we have already seen, is possibly either 14, 16, or 17, a tleshless lower jaw and large eye-socket showing clearly. Coming to Date 5, we have seen that the katun coefficient is probably 9. Close study shows that the tun coefficient is above 10 and under 15; this reading gives 14. The uinal coef- ficient shows that it is above 15 and must be under 18; this reading gives 17. Finally, the best reading of the day coefficient is 6, 7, or 8; this read- ing gives 6. The chief objection to this interpretation is that the day coefficient in s^. Step E, apparently looks more like i, 2, or 3 than 6, or even 7 or 8; but sb is partially effaced and the other agreements are so satisfactory that the writer is inclined to accept 9.9. 14. 17. 5 6 Chicchan 18 Kayab as correct. Gordon suggests two readings for this date: 9.9. 14. 17.4 5 Kan 17 Kayab and 9.9. 18. 18.4, 9 Kan 17 Kayab, favoring the former.^ It will be noted that his first reading is only i day before the reading suggested above; but since the month coefficient is surely 18, not 17, these two may be regarded as identical. Gordon's second reading may be eliminated, because the tun coefficient can not be above 14. Date 6. In N-Q on Step F there is a Secondary Series composed of katuns, tuns, uinals, and kins, (Date 6), of which all the coefficients except that of the katun are eftaced, and even that is a head numeral of unknown value. As the corresponding terminal date is missing, no reading can be suggested here. 'Gordon, 1902, p. 172. His transcription of the second date as 9.9. 18. 18.4, instead of 9.9.19.0.4, that is writing 18 uinals instead of its equivalent, I tun, is not only un-Maya in spirit, but also is contrary to the gen- erally accepted method of tran.scribing Maya dates. INSCRIPTIONS OF THE GREAT PERIOD. 247 Date 7. The next step, G, has no decipherable glyphs, and there appear to be no dates recorded on it. The next step, however, H, opens with a day, the day-sign of which is efifaced and the coefficient almost so. Enough remains to show that it was a head numeral of unknown value. This same day may be repeated in the date in o, p on the same step, however, toward the right of the stairway, the month of which is almost certainly 17 Yaxkin. (See figure 38, c.) Unfortunately, the day coefficient and sign are badly weathered. The latter must have been either Kan, Muluc, Ix, or Cauac, since only these days could have a corresponding month coefficient of 17. The loss of the day coefficient, however, prevents further decipherment at this point, but, as will appear later in connection with Dates 8 and g following, it is probably possible to postulate that Dates 7 and 9 on the one hand and Dates 8 and 10 on the other are the same, and thus to reach a reading for Date 7. Dates 8 and 9. There may be a Secondary Series on Step I at b-d but the glyphs here are too badly weathered to distinguish details. At q on Step I is a day 12.?. As 12 Muluc is the last day on the left of the bottom step nearby, this day in Q may possibly be 12 Muluc also. The next step, J, has no dates; the left half is fairly well preserved, and the right half sufficiently so to show the absence of calendary glyphs. It is indeed unfortunate that the next to last step, K, is not in as good condition as the step above it. In general, the glyphs on this step are in an excellent state of preservation, with the exception of the upper edges, which are broken in most places. As the bar-and-dot coefficients were recorded above most of the glyphs on this step, most of them are damaged. The readings suggested for Dates 8 and 9, therefore, again rest upon a necessary postulate, namely, that Dates 7 and 9 on the one hand and Dates 8 and 10 on the other are the same. Date 8. Step K opens with a date (see figure 38, d) which the writer believes is exactly the same as Date 10 just below it, namely, 12 Muluc 7 Muan. (Com- pare figure 38, d, with f and H, plate 26, a, where these resemblances are readily apparent.) A similar relationship has already been definitely established between Dates 2 and 3 on successive steps above, and probably between Dates 4 and 5 on successive steps above, and in view of the striking agree- ments developed by comparison of these two dates, the writer believes that Date 8 on Step K is 12 Muluc 7 Muan. But in the case of Date 10, as we will presently see, an accompanying Initial Series number declares it to have been 9. 13. 18. 17.9 12 Muluc 7 Muan. Therefore the same Initial Series value probably may be assigned to the 12 Muluc 7 Muan of Date 8 on Step K. Following Date 8, there is a Secondary Series in c^, d, and in Eb, probably the day from which this series is counted in order to reach Date 8. The day- 248 THE INSCRIPTIONS AT COPAN. sign coefficient is surely 9 p.-^x') ^"^ ^^ ^^^^ terminal date of this series is 12 Muluc 7 Muan (Date 8), ' * the day-sign in Eb must be Cauac. This is true because the kin coefficient of the Secondary Series in cb is unmistak- ably 10, and 10 backward from Muluc gives Cauac. The kin coefficient in cb, however, is the only coefficient in this Second- ary Series which is certain. The uinal coefficient cb appears to be composed of one bar and four dots, i. e., 9, but this is somewhat doubtful, fr-r-x"; The tun coefficient, da, is the highest in the series. There seems to be ' ~ ' room for a number between 11 and 15, inclusive, that is, a number composed of two bars and one or more dots, or three bars. In support of this higher value for the tun coefficient than those of the uinal and katun, it should be noted that the tun-sign is lower than the katun and uinal-signs on each side of it, as though more space had been necessary for its coefficient. The katun coefficient is above 5 and below 11 and probably below 10. Returning once more to our postulate that Dates 7 and 9 are the same, we can now fill in the destroyed day of Date 7 as 9 Cauac (the day of Date 9), and Date 7 then becomes 9 Cauac 17 Yaxkin. Our problem, then, is to find an Initial Series for 9 Cauac 17 Yaxkin which can be joined to 9. 13. 18. 17. 9 12 Muluc 7 Muan (Date 8), by the Secondary Series in cb, d on Step K. Since the katun coefficient of this Secondary Series is either 6, 7, 8, 9, or 10, Date 7 must be in Katuns 3, 4, 5, 6, or 7. By referring to Goodman's tables, it will be found that 9 Cauac 17 Yaxkin occurred only thrice in these five katuns, as follows: 9. 3. 11. 12. 19, 9.6.4.7.19, and 9.8. 17.2. 19. Subtracting each one of these dates from 9. 13. 18. 17. 9, one of the resulting differences should, if our procedure and postulate be correct, correspond with the Secondary Series in c^, d: 9.13. 18.17. 9 9.13. 18.17. 9 9.13. 18.17. 9 9. 3. II. 12. 19 9. 6. 4. 7.19 9. 8.17. 2.19 10. 7. 4.10 7.14. 9.10 5. 1. 14.10 A comparison of these difi^erences with the several coefficients of the Secondary Series in c^-d shows clearly that the second, 7. 14.9. 10, is the only one possible here. The best value of the uinal coefficient above was seen to be 9, and the tun coefficient was seen to be between 10 and 16 exclusive; finally, the katun coefficient is between 5 and 11 exclusive. Both the first and third differences disagree for each of these values, as determined by inspection, and it therefore may be concluded, if our postulate is correct, that Dates 7 and 9 are 9.6.4.7.19 9 Cauac 17 Yaxkin, and that the Secondary Series in cb-D, Step K, connecting Dates 7 and 9 with Date 8, is 7. 14.9. 10, viz: Dates 7 and 9 9. 6. 4. 7.19 9 Cauac 17 Yaxkin 7.14. 9.10 Date 8 9.13. 18. 17. 9 12 Muluc 7 Muan INSCRIPTIONS OF THE GREAT PERIOD. 249 Date lO.i The last date on the Hieroglyphic Stairway (see plate 26, a) is fortunately decipherable. It is an Initial Series and commences on the next to bottom step (K), just to the right of the large altar at the base.'- The Initial Series introducing glyph at p and the cycles at q are almost entirely effaced. (See plate 26, a.) The coefficient of the latter, however, is 9. Although only partly effaced, the katun coefficient, Ka, is somewhat uncertain, the best reading being 13. The tun coefficient, sa, is a normal-form head of doubtful value. The uinal coefficient, la, is surely either 12 or 17, and the kin coeffi- cient, Th, is surely 9. The Initial Series terminal date is recorded on the bottom step, L, at A, c. It is 12 ? 7, or 17 Muan,but since the kin coefficient is 9, the day-sign must be Muluc, and we have for the Initial Series terminal date 12 Muluc 7 or 17 Muan. It can be found by calculation that the only places in Katun 13 where these two dates could have occurred are: 9. 13. 18. 17.9 12 Aiuluc 7 Muan and 9. 13. 8. 15. 9 12 Muluc 17 Muan. Since the uinal coefficient is 15 in the second reading, it may be eliminated, and we have left as the only possible reading for Date 10, 9. 13. 18. 17.9 12 Muluc 7 Muan, recorded as follows : Step K p Initial Series introducing glyph P 9 cycles R 13 katiins s 18 tuns la 17 uinals lb 9 kins Step La 12 Muluc c 7 Muan Following this in p-r is the corresponding Supplementary Series. Gordon suggests the reading 11. 13. 9. 14.9 12 Muluc 7 Muan for Date ID,-' reaching the same value for the terminal date as the writer, but differ- ing as to its corresponding Initial Series number. His value for Date 10, if correct, would make it the latest Initial Series known anywhere in the Old Empire by more than 600 years, and for this reason alone, if for no other, it should be viewed with suspicion.'^ Moreover, against Gordon's reading there is another fundamental objection, namely, that throughout the entire range of the Maya hieroglyphic writing no certain Cycle 11 Initial Series is known. Indeed, it is highly probable that the Initial Series method of recording dates 1 Gordon (1902, p. 173) calls this Date 4. 2 This date and the next, No. II, might possibly be included with Stela 4 and Altai I' here at Copan and Stela 10 at Tikal, as exceptions to the general rule that Initial Seiies introducing glyphs only occupy the first positions in the several texts where they occur. Strictly speaking, this is true, but the cases are hardly parallel, and the arrangement of this particular text is such, the longest in the Corpus Inscriptionum Mayarum, that the writer has not thought it necessary to regard them as exceptions. 'Gordon, 1902, pp. 173-177. -ikv< 8 in Temple 18, and here the head has a super- ^^^ r. • • 1 1 J • T^ 1 o nri ■ Fig. CC. — Part of inscription on fix wanting m the heads in lemple 18. Ihis AitarX'. superfix, the writer believes, is the same as the one in the great-cycle glyphs on Stela 10 at Tikal, Stela N here at Copan, and on the tablet from the Temple of the Inscriptions at Palenque. Else- where he has suggested that this superfix increases the cycle-sign in a ratio of 20, i. e., 20X144,000 = 2,880,000,^ and if this is true elsewhere, it should have the same value here; but the meaning or value of the main ele- ment here is unknown. The coefficient 9 recalls the cycle coefficient of the historic period, and if we could identify this grotesque head as a katun-sign, this superfix of 20 would raise it to the value of a cycle, and the 9 would then indicate the current cycle of the historic period. We could then explain the whole glyph as a novel synonym for Cycle 9. Bowditch figures a number of head-variant katun-signs (1910, plate 12), all of which show the prominent beak-like nose and a few the filed upper front tooth. None, however, shows the fleshless lower jaw. It does not appear advisable to press this interpretation farther, although it is interesting to note that it gives a plausible explanation for the coefficient 9. The date is of course undeterminable, except that on stylistic grounds it may be referred to the Great Period. Altar U'. Provenance: Found in 1916, in the patio of the house at the south- eastern corner of the village plaza. (Group 9). (See plate 3 and figure 22, n.) Date: The Great Period. Altar U' was found by the writer in 1916 in the patio of the house at the southeastern corner of the village plaza, in which position it is near the base of the pyramid on the summit of which Altar S was found and from the ' Morley, 1915, pp. 120, I2I. INSCRIPTIONS OF THE GREAT PERIOD. 377 interior of which the fragment of Stela 21 was recovered. (See plate 3 and fig. 22, k, I, and n.) It is a small drum-shaped stone like Altar T', 29 cm. high and 48 cm. in diameter. The top and bottom are plain, the periphery being inscribed with three glyphs. These are very badly weathered, but from what is left it is evident that each had a coefficient of 9, and that each was probably the same grotesque head as on Altar T', whatever that may be. The similarity of these two altars in size, shape, and treatment is very close, and both doubtless may be referred to the Great Period. Shrine R'. Provenance : On the hillside south of the Copan River, facing the Main Structure and shghtly east of south from it. (Group 7, see plate 3.) Date: The Great Period (?). Text, photograph: plate 28, (f. Almost due south of the Main Structure, on the side of the hill which rises just beyond the river, there is an outcropping of the native rock in a ledge, perhaps 60 or 70 meters above the bed of the stream. Advantage has been taken of this stone in situ, to carve from it an elaborate shrine over- looking the valley, the central figure of which is a large and realistic toad. (See plate 28, d.) No previous reference to this shrine appears to have been made, and the writer only heard of it on his third visit to Copan in 1915. It can hardly have escaped the indefatigable labors of Gordon or Maudslay, however, although neither mentions it in his writings. From one end to the other it is 6 meters long, and at the widest point about 3 meters wide. The height from the base of the altar at the front to the top of the last platform at the back is about i.S meters. Facing the shrine, i. e., with one's back to the valley, there is an altar on the right, slightly in advance of the other parts. This is the shape of a truncated cone, being perfectly flat on top. It is 41 cm. high at the back, though somewhat higher in front, owing to the slope of the hillside, 2 meters in diameter at the base, and 91 cm. in diameter at the top. Its position with reference to the other figures, as well as its shape, probably indicates its use as an altar for burning incense at the shrine. It is asymmetrically placed with reference to the long axis of the shrine, being at one side. Just behind the altar is a large, rough rock only partially dressed. The back and right sides show plain dressed surfaces merging into the steps or seats carved out of the ledge to the right of the shrine. The front and left sides of this rock are rough. Opposed to this on the other side of the shrine at the front, and in a sense balancing it, there is another large unfinished rock. The center of the front is occupied by the large toad figured in plate 28, d, the most conspicuous feature of all. This is 1.07 meters long, 61 cm. high, and 61 cm. wide, and squats on a bench of dressed stone, lips parted, tongue protruding, eyes half closed, and throat enormously swollen. The ledge of 378 THE INSCRIPTIONS AT COPAN. rock from which it is carved comes up much higher against the left side of the toad (facing it) than the right, and here, on the sharply sloping front, a column of three glyph-blocks is sculptured. These are the only glyphs anywhere about the shrine, and it is indeed unfortunate, therefore, that all three should be effaced, since with the loss of the inscription in such an unusual monument as this approximate dating on stylistic grounds becomes impossible. It is even hazardous to assume this shrine dates from the Great Period at all, since, judging from the stylistic criteria present, it is well within range of the sculptors of the Middle Period. To the right of the toad and slightly in advance is a lower grotesque figure, and another of the same kind behind to the right and also lower. Next comes a pair of rough unsculptured blocks balancing each other; and finally, at the top of the last bench, the highest point of the shrine, and some 20 meters behind, are several mounds. The toad is easily the central figure, and a not inappropriate name for this whole cluster of sculptured rocks is that suggested here, "the Shrine of the Toad." Its function, except as an obvious place of sacrifice, as indi- cated by the altar in front, can only be conjectured. Possibly the toad may have had some connection with the idea of rain, and this shrine may have been the place where sacrifices to the rain deity were made. The treatment of the head of the toad, even to adventitious details, is just like that of the head-variant of the uinal-sign. (Note the three dots in the ear in plate 28, d, and compare with the uinal-signs in plate 26, d, D, and plate 27, E^.) Can this possibly indicate that this shrine was consecrated to the deity presiding over the uinal, namely, a toad-like god .? Making due allowance for Maya psychological processes, the latter would appear to be the most attractive hypothesis now available. Here as the uinals waxed and waned the ancient Maya priests may have made the sacrifices appropriate to each of their 20-day periods; at least from, its unique character and prominent position we can not doubt but that the Shrine of the Toad played a not unimportant part in the religious life of the ancient city. Fragment Z'. Provenance: At Group 4, 2.5 kilometers up the valley from the Main Structure. (See plate 3 .) Date: The Great Period. Text, drawing: figure 51, &. Fragment 7J is a small piece of a glyph mosaic 32 cm. high and 17 cm. wide from the principal temple at Group 4, 2.5 kilometers up the valley, northeast of the Main Structure. (See plate 3.) The first bench of hills above the valley floor at this point has been leveled off and a number of stone stiuctures built facing the valley. Judg- ing from the number of sculptured fragments lying around the principal mound, it formerly must have been an important temple. INSCRIPTIONS OF THE GREAT PERIOD. 379 Fragment TJ would appear to have been part of a hieroglyphic cornice, or at least part of a horizontal architectural member. (Note the plain bands both above and below.) Part of one glyph only appears, which looks some- thing like the head of God C. Excavation here would doubtless bring to light the remaining fragments of this text, which, however, may be assigned to the Great Period on the basis of the stylistic criteria. This bench of hills on the north side of the valley, from the Quebrada Seca, Group 4, westward to Group 9, a distance 4.5 kilometers, has been artificially leveled off from end to end and covered with stone buildings. Practically no excavations have been made along this bench, and it offers one of the most promising parts of the city for future investigation, especially Groups 4 and 6, which, judging from the number of sculptured fragments found lying loose on the surface, must have been sections of con- siderable importance. Before closing this analysis of the Copan inscriptions, it is necessary to describe one last text from this site, namely, an engraved peccary skull dating from the Early Period, the description of which has been deferred to this point in order that it should not interrupt the continuous presentation of the monumental sequence. This skull was found, with another similarly engraved, by the First Peabody Museum Expedition under Saville in 1892 in Tomb i just south of the Main Structure between Mounds 34 and 36. (See plate i.) It lay on the floor of the tomb by the side of the other skull, and was associated with other objects of bone, jade beads and the like, and several skeletons. Spinden gives a drawing of it (1913, fig. 210) and an excellent description of the design {ihid, p. 151); the part of the inscription presenting the date shown in figure 56, a, however, was re- drawn by the writer directly from the original now in the Peabody Museum (catalogue number C. 201). The panel of four glyphs shown here is in the center of the top of the skull, and is the most conspic- uous part of the entire design. It is flanked by two seated human fig- ures facing the panel, and elsewhere on the top there are three running peccary, a jaguar, a monkey, and other human figures and glyphs, the whole beautifully executed in incised lines, except for a few of the glyphs on the sides, which are in low relief, made by cutting the background somewhat deeper. e Fig. 56. — a. Part of inscription on peccary skull from Tomb I. Early forms of day-sign Alian: h. Altar Y; c, Stela 25; d. Stela 15; c, Stela 16. 380 THE INSCRIPTIONS AT COPAN. The text opens with the day i Ahau in ai and then follows in bi a sign which the writer long mistook for 8 Chen, Yax, Zac, or Ceh, followed by the end of a tun in A2. But it will be found from Goodman's tables that no tun of Cycle 9 ended on the day i Ahau 8 Chen, Yax, Zac, or Ceh, and for a long time this date baffled all efforts looking toward its decipherment. Indeed, it was not until the writer's discovery in 1918 that the "winged Cauac" sign is a variant of the tun-glyph that decipherment was made possi- ble, and it then became apparent that bi, which had previously been mis- taken for 8 Chen, Yax, Zac, or Ceh was in reality Tun 8, which in October 19 19 led to the reading given below. This text reads: "i Ahau (ai), Tun 8 (ei). End of a tun (a2), Ahau (?) (b2)"; and referring to Goodman's tables once more, it will be found that the only Tun 8 in Cycle 9 which ended on the day i Ahau was 9.7.8.0.0 I Ahau 3 Ceh, the month being suppressed in the present text, as was not infrequently the case. 9.7.8.0.0 is a fairly early date, but happily a highly unusual feature in this text authenticates the correctness of this reading on stylistic grounds also. It will be noted that the Ahau-sign, both in ai and B2 of figure 56, a, shows an unusual notch on either side of the face (the inner line in ai and the central element in B2). This is a very unusual characteristic, and so far as the writer is aware, is only found in four other texts known: Altar Y, Stela 25, Stela 15, and Stela 16, all at Copan, h, c, d, and e, figure 56, respectively, all of which are earlier than 9.7.8.0.0, viz.: Stela 25 9.2.10. 0.0 Stela 15 9.4.10. 0.0 Altar Y 9.7. I. 2.6 Stela 16 9.7. 2.12.0 Thus on stylistic grounds (its resemblance to other known dated inscrip- tions), the early character of this text is amply substantiated, and is in perfect agreement with the early date which it records. Slight and apparently adventitious details like this are frequently of more importance in establishing general contemporaneity on stylistic grounds between two texts than resemblances between the larger features, another case illustrating this point being the protuberances at the corners of the day-signs on Stela 3 at Tikal and on Stela 24 at Copan, only 3 years apart in date, a feature found in no other inscriptions known. Another interesting feature in connection with this text is that it pre- sents the earhest example of the use of the "winged-Cauac" variant of the tun-sign yet discovered. It shows, moreover, that tun-ending dating was in vogue as early as the Early Period of the Old Empire, and as the writer has shown elsewhere,^ that it continued in practice down to the close of the New Empire, a range of more than a thousand years. Smaller objects such as this peccary skull were doubtless executed whenever fancy dictated, although even here a tun-ending was chosen for the contemporaneous date, but the larger monuments, particularly during 1 Morley, 1919, p. 274, and Appendix II. INSCRIPTIONS OF THE GREAT PERIOD. 381 the Early and Middle Periods, were erected only at the hotun-endings. This date, although it is not on a monument, has been inserted in its proper chronological position between Stela 18 and Stela 7 under the texts from the Early Period in Appendix IX. This concludes the presentation of the inscriptions of Copan, but before proceeding to summarize these data, it appears advisable to describe here three other monuments found at sites which undoubtedly drew their artistic inspiration from Copan, if indeed they were not colonized directly therefrom — LEGEND ARCHAEOLOGICAL SITES O MODEEN TOWMS K I LOM ETERS Fig. 57. — Map showing location of Copan and the principal cities along the southeastern Maya frontier. Rio Amarillo on the east bank of the Copan River, 30 kilometers northeast of Copan, and Los Higos, across the divide in the Chamelecon Valley, 80 kilometers northeast of Copan. (See figure 57 for the location of these sites.) The first was discovered by the Carnegie Institution Central American Expedition of 1917 and the second was "relocated" at the same time, although the original discovery of the latter would appear to have been made by Squier as early as 1850.^ The map in figure 57 shows clearly the chain of sites northeast of Copan, Santa Rita, Rio Amarillo, Paraiso,- Los Higos, ' See Morley, igi/f, pp. 287-289; 1918a, p. 276, note; and Squier, 1883, pp. 468-480. ^ Paraiso is located in a small valley on the divide between the Motagua and Copan Rivers, somewhat nearer Quirigua than Copan. (See fig. 57.) The writer visited this site in 1914, but although typical Maya sculp- tures — heads, decorative elements, etc. — were found, unfortunately no inscriptions were recovered. The site is fairly large and excavations here would doubtless bring to light additional material. The heads examined showed rather closer affinities with the art of Copan than with that of Quirigua. 382 THE INSCRIPTIONS AT COPAN. and Quirigua, all of which drew their artistic inspiration from the great southern metropohs, and all of which were probably colonized therefrom. The last four probably date from the close of the Middle Period or the begin- ning of the Great Period and are but another expression of that great expan- sion which began at Copan in 9.11.0.0.0 and continued in ever-widening waves for the next century and a quarter. Altars i and 2 at Rio Amarillo. Provenance: On a terrace at the foot t)f the hills on the east side of the Copan River, 30 kilometers northeast of Copan. (See figs. 57 and 58.) Date: The Great Period. Text, drawing: figures 59 and 60. Reference: Morley, igiyc, pp. 287, 288. LEGEND g ALTABS Fig. 58. — Plan of the ruins of Rio Amarillo. Rio Amarillo is located 30 kilometers northeast of Copan, on the east bank of the Copan River, at the foot of the hills on the east side of the valley. The road from La Florida to Copan, after emerging from the hills and just before crossing to the west bank of the river (see figs. 57 and 58), passes along the edge of the terrace where both Altars i and 2 lie. Altar I is approximately square, being 86 cm. long, 84 cm. wide, and 30 cm. high. The top and bottom are plain, each of the four sides having 3 INSCRIPTIONS OF THE GREAT PERIOD. i83 glyph-blocks, or 12 for the entire text. (See fig. 59.) Clear as these are on the east, south, and west faces, every sign of which is perfectly legible, no one appears to contain data by means ot which the position of this altar can be fixed in the Long Count; and it is necessary to conclude that the date must have been recorded in the first two glyph-blocks on the north side, a, b, now unfortunately effaced. (See fig. 59, a.) Many individual glyphs. Fig. 59. — Inscription on four sides of Altar I at Rio Amarillo. however, are recognizable. Thus, for example, on the west side (fig. 59, b) Da u. h. contains the day-sign Ahau; T)b u. h. is 3 katuns and T>b 1. h. looks like 7 Ix (.''); Ea 1. h. looks like a head-variant for the day-sign Kan, and va 1. h. is the head of God C. On the south side the first glyph, ca u. h. is 4 Cib and na 1. h. is the winged-Cauac variant of the tun-sign, and on the east side at ]a 1. h. the day-sign Ahau appears, and again at K.a u. h. (here inverted), and again the day I Ahau in i.b u. h. There were 2 or 3 tuns recorded in la u. h., but unfortunately the coefficient here is partially efi^aced. In spite ot the fact ^84 THE INSCRIPTIONS AT COPAN. that it is possible to decipher several of the above glyphs, owing to the loss of the first two glyph-blocks, a and b, this inscription can not be dated other than as falling in the Great Period. Altar 2 is slightly smaller than _^ Altar I and is again approximately square, being 67 cm. long, 63 cm. wide, and 3 5 cm. high. Only half of it was found, and only one of the two faces on Ul "X^ this piece is sufficiently preserved to I ^ ^1 • •„*•„„ fC^ ^^\ nru^ Fig. 60. — Part of inscription on Altar make out the inscription (hg. 60). Ihe AmariUo. first sign, a, is Ahau, but without any coefficient, the two remaining glyphs being undecipherable. The carving is of the same character as that on Altar i, and this altar also may safely be referred to the Great Period on stylistic grounds. at Rio MBSTfLAE • ROUND ALTARS ^ SCUUPTUDt tSia E53 m m PMP G REAT PLAZA '|^l'/^|^^\'%|\^l''^^,^\\ /// /. Fig. 61. — Plan of the ruins of Los Higos. Stela i at Los Higos. Provenance : On the west side of the Chamelecon River, 80 kilometers northeast of Copan and 30 kilometers northeast of La Florida. (See figs. 57 and 61). Date: 9.17.10.0.0 12 Ahau 8 Pax.' Text, drawing: figure 62. References: Morley, 1917c, pp. 288, 289. Morley, 1918(2, p. 276, note. Squier, 1883, pp. 468-480. For other monuments recording this same hotun-ending, see Appendix VIII. INSCRIPTIONS OF THE GREAT PERIOD. 38s This site is located on the west side of the Chamelecon River, just below the Hacienda of Los Higos, from which it takes its name, not more than 200 meters back from the river-bank. The city was built just where the ground begins to rise from the flood-plain, and its long axis runs north and south. Behind the city to the west is a rocky hill and cliff which probably supplied the building material. Approaching from the east or river side over a low platform (W, fig. 61),^ one climbs a terrace 3 meters high, along the eastern edge of which there is a range of low mounds, S, T, U, and V. Behind these and some 7 or 8 meters higher is a second range of fallen buildings, C, D, E, F, and G. North of G this third terrace turns to the west and defines the north end of the site. Behind this second range of buildings is the Great Plaza, dominated by the principal temple, H, which surmounts a pyramid 10 or 12 meters high. This had stairways on all four sides, but the temple itself faced toward the river, i.e., west. In the court on the north side of this temple lies Stela i, the only inscription yet found at this site. (See fig. 62.) A torso and head (in two pieces) and two small plain altars were found at a, b, and c, respectively, figure 61. Interest here centers in the stela, which, although badly broken and part missing, is a fine example of Maya art during the Great Period, from the zenith of which it dates. The present height of the several fragments recovered is 1.6 meters, and probably another quarter to half a meter is still missing. It is 43 cm. wide and 36 cm. thick. The front is sculptured with a human figure which very closely resembles that on Stela B at Copan, both having the same kind of turban head- dress. (See fig. 62, a and c.) The back and sides are inscribed with glyphs, on the basis of which arrangement it is to be assigned to Class 4 of the Copan stelae. The inscription, which is beautifully clear, opens with an Initial Series introducing glyph at A1-B2 (fig. 62, b), followed by the corresponding Initial Series at A3-B5, B9 {:). This records the date 9. 17. 10.7.0 9 Ahau (3 Tzec) as follows. Fig. 62. — Inscription on Stela I at Los Higos: a, c, sides; b. back. 'The map of Los Higos shown in figure 6i was made by Mr. John Held, Jr. 386 THE INSCRIPTIONS AT COPAN. A1-B2 A3 B3 Initial Series introducing glyph 9 cycles 17 katuns A4 10 tuns B4 AS BS B9 7 uinals kins 9 Ahau [3 Tzec] Although the month-sign itself is effaced, the presence of the glyphs of the Supplementary Series in a6-a9, the last, Glyph A, appearing in a9, renders it practically certain that 3 Tzec was recorded in B9.^ Continuing the inspection of this text, there are no decipherable glyphs on the left side facing the stela (fig. 62, c), although C5 may be the moon-sign with a coeffi- cient of 2, which the writer has shown elsewhere may stand for 40 days.- The first three glyph-blocks on the right side, di-d3, figure 62, a, are effaced; the fourth, d4, looks like Imix with a Ben-Ik superfix, and the fifth, D5, is half of the Venus-sign. In D8 is the day-sign Ahau inverted, and in Dii Imix again with the Ben-Ik superfix. The last four glyphs, D12-D15, are exceedingly important, since they record a Secondary Series ot 7 uinals in D12, which if counted backward from the Initial Series will reach the date 12 Ahau 8 Pax in du, di3, a hotun-ending, and therefore, in all probability, the contemporaneous date of this stela: AI-B5, A9 9.17. 10.7.0 9 Ahau 3 Tzec DI2 7.0 backward DI4, DI5 9. 17. 10. 0.0 12 Ahau 8 Pax The decipherment of this text has already been given in connection with the discussion of Altar W (p. 333), where other examples of this overlapping of the current hotun-ending by a few uinals, always less than 18, are given. It was shown there, and may be repeated here, that such overlapping prob- ably in no way interfered with the character of such monuments as hotun- markers, and Stela i at Los Higos therefore may be accepted as dating from 9.17.10.0.0, being about 2 years earlier than the last group of stelae in the Great Plaza at Copan, C, H, F, and 4, and of exactly the same date as Zoo- morph B at Quirigua. The relief on this monument is beautifully executed, particularly the glyphs, which show a delicacy of treatment and a refinement of detail scarcely surpassed by the best work at Copan, with which its date of 9. 1 7. 10. 0.0, that is at the height of the Great Period agrees appropriately.^ ' Attention should be called to the unusual form of Glyph X at a8, in this Supplementary Series. This variant is only known in three other texts in the Corpus Inscriptionum Mayarum, namely. Stela 20 here at Copan (9.1.10.0.0) (?); the tablet in the Temple of the Cross at Palenque (9.i3.o.o.o(,?)); and Stela E at Quirigua (9.17.0.0.0), the last only 10 years earlier than the contemporaneous date of this stela. = That is, 2 X 20. See Morley, 1916, pp. 384-386. ' The writer was enabled to relocate this site, which was discovered by Squier between 60 and 70 years ago and subsequently forgotten, by a fortunate coincidence which brought together in Guatemala City, in May 1917. Professor Marshall Saville, Mr. S. K. Lothrop, and Mr. Basil Booth. Information then received from these three gentlemen enabled him to find this site a few weeks later. INSCRIPTIONS OF THE GREAT PERIOD. 387 The beginning of the Great Period at Copan was characterized by a remarkable outburst of sculptural activity in the Great Plaza, where the first of the Great Period monuments, Stelae A and B, were erected, followed by Stela D at the end of the next hotun. These monuments ushered in an era of elaboration in stone carving, the like of which had never been seen before in the city. Delicate details of relief were actually freed from the plinth, as in the undercutting of the teather-work in the head-dress of Stela B. Indeed, the sculptors of the day were at the Maya esthetic zenith. Skillful control of technical processes, developed through several centuries of continuous experience, coupled with lotty esthetic ideals, the result of increasingly beautiful art productions, had finally brought the Maya to their Golden Age, the accrued benefits of which they were now enjoying. Somewhat earlier than this, a tremendous project had been inaugurated, no less than the construction of the lofty pyramid known as Mound 26, and of the magnificent Hieroglyphic Stairway ascending its western side and the imposing temple on its summit. It is even possible that this construction may have been started as early as 9. 13. 17. 18. 9 in the Middle Period; but what is certain, however, is that its completion did not take place until 20 years after the erection of Stela D in 9.15.5.0.0, and further, that after the dedication of Stela D probably all sculptors and masons in the city were transferred to the more important public works then in course of con- struction at the southern end of the Main Structure, namely, the Acropolis and the associated pyramids and temples. The Hieroglyphic Stairway in particular, with its many glyphs and elaborate decorative elements, human figures of heroic size, birds, serpents, etc., must have required a large number of skilled artisans for its execution; at all events, after Stela D there are no surely dated monuments until after the completion of the Hieroglyphic Stairway, Temple 26, and the dedication of Stela M at its base, in 9. 16.5.0.0. The next hotun, 9.16.10.0.0, was marked by Stela N, and then came the period of maximum architectural, sculptural, and indeed cultural efflores- cence. For the next 20 years there followed a series ot brilliant artistic achievements, intricately carved altars, gorgeously ornate stelae, and mag- nificently embellished temples; the city was now rounded into its final form as found to-day, and became, in truth, the esthetic center of the Maya world. Coming to a more detailed consideration of the monuments, we note at the outset a considerable extension in the use ot the hieroglyphic inscrip- tions. In addition to stelae and altars, the latter greatly diversified in type, we now find texts inscribed upon hieroglyphic steps and stairways, both exterior and interior, door-jambs, and possibly even cornices.' This exten- sion of texts beyond the narrower field of stelae and altars is ot itself an indi- cation of "lateness," and such inscriptions on this ground alone are to be ' Although no examples of hieroglyphic cornices have yet been reported at Copan, with the possible exception of Fragment Z', pp. 378, 379, which is doubtful, one was found by the writer at the neighboring city ol Quirigua in 1912 (see Morley, 1913, pp. 347, 352); and it is highly probable that examples of it may yet be recovered at Copan. 388 THE INSCRIPTIONS AT COPAN. assigned to the Great Period. For the purposes of this investigation we may classify the texts of the Great Period according to the media upon which they are presented into four general groups: (i) stelae, (2) altars, (3) steps, and (4) jambs. The 11 stelae of the Great Period may be divided into four classes, according to the arrangement of their designs, as follows : 2.' Stelae having inscriptions on two alternate faces, the two remaining faces being plain (Stela 8). 4.- Stelae having inscriptions on three faces, the fourth being carved with the representation of the human figure (Stelae A and B).** 5. Stelae having inscriptions on two alternate faces, the remaining two being carved with representations of the human figure (Stelae C and N). 6. Stelae having inscriptions on the back only, the front and sides bemg carved with representations of a single human figure with increased ornamentation (Stelae D, M, H,4 F, 4, and 11). The 5 earlier stelae of the Great Period, A, B, D, M, and N, all have Initial Series and all record hotun-endings. Indeed, all but one, A, have their respective hotun-endings recorded by their respective Initial Series. Of the 6 in the later group, however, 8, C, H, F, 4, and 11, probably none were erected to commemorate hotun-endings, and only one in fact has an Initial Series, i.e., Stela 4. Here is a sharp break in practice as contrasted with the stelae of the Early and Middle Periods and the beginning of the Great Period. The earlier stelae of the group are regular, the later irregular, the break coming after Stela N, in 9. 16. 10. 0.0. We have seen that the lahuntun from 9.1 6.1 0.0.0 to 9.17.0.0.0 was characterized by a tremendous outburst of architectural as well as sculptural activity, and during the busy days which followed the erection of Stela N, Initial Series dating passed out of fashion at Copan. Thus all the final stelae and temples have their contemporaneous dates recorded as Period Endings or simply Calendar Round dates, or not at all. We may therefore formulate the following generalization concerning the stela; of the Great Period, namely: During the latter half of the Great Period, the custom of erecting stelae at the expiration of the hotuns fell into disuse, and with it also went the custom of record- ing Initial Series on any kind of monuments.'' The altars of the Great Period are as complex as the stelae. Of those associated with stelae, moreover, none present inscriptions. These are chiefly of the double-headed-monster type and are exceedingly complex. In some cases, as the altars of Stelae M and F, for example, they are not even monolithic, being built up of several sculptured pieces fitted together. '■ Class I (see p. 125) is not represented in the Great Period. ^ Class 3 (see p. 125) is not represented in the Gieat Period. ' Stela B shows a transition between Classes 4 and 5, in that its back is covered with a large grotesque mask instead of a human figure of- heroic size. ^ Stela H shows a transition between Classes 5 and 6, having a grotesque mask on its back in addition to the glyph-panel. ' Only two exceptions, Stela 4 and Fragment E', need be made. INSCRIPTIONS OF THE GREAT PERIOD. 389 The 22^ independent altars described in this chapter as having in- scriptions may be divided into a number of types. The rectangular altars of the Early and Middle Periods, although represented by a few simple examples (S and V) grow larger and more ornate. Grotesque or death's heads fill first one end (R and Z) and later both ends (B' and C). Still later, as the altars grow larger, more complicated designs are introduced, such as figures seated on glyphs and animal forms (D', W, L, U, Q, and T). A special development of the rectangular altar is the shaft or square column, of which O', F', and N' are examples. The round altars of the Early and Middle Periods become less frequent and very much smaller (W, T', and U'), reaching the form of a table with legs as their latest development (C). The most elaborate altars of all, however, are the double-headed serpents or monsters with glyph-panels in the coils of their bodies (G3, G2, and Gi). The dates of G2 and Gi indicate that they are probably the latest inscriptions at Copan. The same tendency toward flamboyancy noted in the stelae of the period is also found in the decoration of the altars, and as the end drew near made itself increasingly felt. Only one of these 22 altars, the earliest, Altar S, records an Initial Series, though a number have hotun-endings as their contemporaneous dates i.e., Z, G3, Q, W, W, G2, and Gi. This may, indeed probably does, indi- cate the passing of the function of marking the hotun-endings from the stelae to the altars. While none of the later stelae are hotun-markers, most of the later altars are, a distinct reversal of the general practice in earlier times. - But the onward sweep of intellectual and esthetic development was driving the Copan sculptor even further afield than the mere diversification of earlier and familiar forms. The hieroglyphic step came into fashion, both as an exterior and an interior element of architectural decoration.-'' Three examples of the former have been found at Copan, namely, that on Mound 2, the Hieroglyphic Stairway of Mound 26, and the Reviewing- stand in the Western Court, and three of the latter, the interior steps of Temples 11, 22, and 21a. lAItar S Altar V Altar Q Altar Gj Altar G' L R W G, N' B' U T 0' T' C z W F' U' D' Gz 2 Earlier examples of this later practice may be tlie altar of Stela I and the West Altar of Stela 5, commemo- rating the katim-ending 9.12.0.0.0, and Altars H' and I', commemorating the ne.\t katun-ending, 9.13.0.0.0. 'The earliest use of the hieroglyphic step or stairway is probably at Palenque, in House C of the Palace group. (See plate l.) The Initial Series on this stairway is 9.8.9.13.0, but this is brought forward into Katun II, 12, or 13 by a Secondary Series. The Initial Series on the hieroglyphic stairway at Naranjo commences with an early date, 9.10.10.0.0, but Secondary Series present in this inscription may bring this date forward, perhaps to Katun 18 or 19. Maler's glowing accoimt of a hieroglyphic stairway on Structure 5 at Yaxchilan hardly appears justifiable: "This hieroglyphic stairway of structure 5 of Yaxchilan is the most magnificent one I have ever seen" (Maler, 1903, p. 122). In 1914 the writer could hardly satisfy himself that these steps had ever been inscribed with glyphs at all. The next use of the hieroglyphic step would appear to have been here at Copan in the Hieroglyphic Stairway of Mound 26. In 9.16.0.0.0 we find it at Seibal; and the latest occurrence is at Quirigua, in Structure I, w'bere the date 9.19.0.0.0 is recorded, as a Period Ending. The custom of inscribing both the risers and treads of steps (see the stairway of House C of the Palace (Iroup at Palenque, Maudslay, 1889-1902, vol. iv, plate 23) never seems to have come into general use. 3go THE INSCRIPTIONS AT COPAN. Not content with this ornate feature of decoration, in the broad current of flamboyancy now in full swing, door-jambs were next subjected to hiero- glyphic treatment (Temples 26, 11, and 18), and possibly also exterior cornices (Fragment Z' at Group 4). But at last the end was at hand. This costly process of intellectual exhaustion, leading so surely to decadence and ultimate futility, was stopped at its most brilliant moment probably by that same catastrophe which overwhelmed all the cities of the Old Empire about 10.2.0.0.0, but which appeared at Copan three-quarters of a century earlier. About the middle of Katun 16, coincident indeed with the important date 9. 16. 12. 5. 17 6 Caban 10 Mol, the mastery of technical processes be- came so complete that contemporary esthetic ideals were released from all practical limitations of material or treatment, and at once soared to flam- boyant heights; and during the closing half century of the city's history (9. 16. 10. 0.0 to 9.19.0.0.0) this tendency carried sculpture into a variety of new media and led eventually to a condition which was only saved from decline by the abandonment of the region and the migration of its inhabi- tants elsewhere. CHAPTER V. CONCLUSIONS. GENERAL COMPARISONS. The inscriptions of Copan are more numerous than those of any other Maya site, or indeed of any other three combined, constituting possibly as high as 40 per cent of the Corpus Inscriptionum Mayarum, and in their chronologic range covering 335 years, or the greater part of the Old Empire. These two factors — the large amount of material available for study and the long period covered by the dated monuments — thus make Copan the best fitted of all the Maya cities at which to investigate the chronology of the Old Empire. In the important field of the Initial Series alone, for example, as will be seen from the following table, Copan stands preeminent, having about one third of all the Initial Series known — 59^ out of 177 — and as many as those of her three nearest competitors in this respect combined, namely, Piedras Negras, Quirigua, and Naranjo: Copan 59 Piedras Negras. . 25 Quirigua 17 Naranjo 17 Yaxchilan 13 Palenque 9 Uaxactun 7 Tikal 6 La Honradez. ... 5 Itsimte 3 Ixkun 2 Altar de Sac. ... 2 Quen Santo 2 Ocosingo I El Pabellon.... i El Cayo I Yaxha I Los Higos I Chichen Itza. . . . Holactun Tuluum Leyden Plate. . . Tuxtla Statuette. Total 177 All the features noticeable in the inscriptions of the other cities, moreover, are found at Copan, and some of them, indeed, may have been inaugurated here. Take, for example, the custom of erecting the stelae at the ends of even periods of the Long Count, instead of at odd times, a prac- tice which became the controlling factor in the erection of monuments throughout the Maya area, and persisted for more than thirteen centuries. It will be seen in a later section, page 396, that the three earliest Initial Series known, those on the Tuxtla Statuette, the Leyden Plate, and Stela 9 at Uaxactun, do not conform to this practice, but that, on the con- trary, they record dates which do not stand at the ends, even of tuns or uinals of the Long Count. The first certain example of this practice is on Stela 24 at Copan, in 9.2.10.0.0, although, in all probability, Stela 20 at the same site records the next previous lahuntun-ending, 9.1. 10. 0.0. The earliest known date at Tikal, Copan's only great contemporary at this time, 9.2.13.0.0 on Stela 3, 'This number, moreover, does not include those Initial Series on the Hieroglyphic Stairway that were destroyed, of which there must have been at least lo. 391 392 THE INSCRIPTIONS AT COPAN. does not conform to this practice; and although it is possible earlier hotun- endings may yet be found there, ^ the custom does not appear to have been followed so closely as at Copan. This practice, with varying modifications, continued in force down to the close of the New Empire in Yucatan, and is mentioned by several of the early Spanish and native writers.^ It was indeed one of the oldest of all Maya institutions, and, so far as the stelae are concerned, it was the most important factor in determining the dates of their erection. Another feature of Maya chronology, possibly inaugurated at Copan, was the gradual abandonment of Initial Series dating and the coincidental increase of Period Ending dating during the last half of the Great Period. Of the cities of Classes I and II given on page 441, with the exception of Nakum, where Initial Series have not been found at all, and of Palenque, which m.ay have been abandoned before this time, Copan was the first to discard this cumbersome but extremely accurate method of dating, the last Initial Series there being carved in 9. 17. 12. 13.0 (Stela 4) as compared with 9. 18. 3. 1. 5 at Yaxchilan (small altar near Stela i), 9.18.5.0.0 at Piedras Negras (Stela 12), 9. 18. 10. 0.0 at Naranjo (Stelae 8, 12, and 28), 9.19.0.0.0 at Quirigua (Structure i), and 10.2.0.0.0 at Tikal (Stela 1 1). A glance at the fourth column in Appendix IX, which gives the kinds of dates on the different monuments at Copan, i.e., whether Initial Series, Period Ending, or Calendar Round, this condition appears clearly. During the Early and Middle Periods, Initial Series are the almost invariable rule; indeed, in the Middle Period there is not a single stela without one; but after 9. 17. 12. 13.0 (Stela 4) they stop at Copan, the closing hotun-endings, 9.18.0.0.0, 9.18.5.0.0, and 9. 18. 10. 0.0 being marked by Period Ending dates on altars instead. The Initial Series method of dating required eight glyphs to express any single date of the Maya Era, and was correspondingly costly of space and effort. The Secondary Series was probably developed veiy early to escape from such a tedious process when more than one date had to be recorded on the same monument, and as early as 9.4.10.0.0 (Altai Q') Period Ending dating, as applied to lahuntuns, was in use at Copan, and as early as 9.7.8.0.0 as applied to any tun (inscribed peccary skull from Tomb i). By the middle of the Great Period this method of dating was beginning to supersede the Initial Series, in some cities altogether, as for example at ' There are two stelae at Tikal, Nos. 8 and 9, which the writer believes he may possibly have deciphered correctly as 9.0.10.0.0 and 9.2.0.0.0 respectively. On the west side of Stela 8 (see Maler, 191 1, plate 19, l), a6 is an ending- sign in which the hand element is very conspicuous, and following this in A7 appears to be the day 7 Ahau. Assum- ing this is a katun or lahuntun-ending, by no means certain, however, the only two positions where this day could have occurred in the Early Period are at 9.0.10.0.0 7 Ahau 3 Yax and 9.7.0.0.0 7 Ahau 3 Kankin, and since this latter date is too late on stylistic grounds to be possible, the former may be the date of this_^monument. The date suggested for Stela 9 seems more certain. The west side of this stela (see ibid., plate 20, i) opens with a hand ending-sign at Ai; in A2 there appears to be either 4 or 2 katuns, and in A3 the day 4 Ahau. The only katun in the Early Period ending on a day 4 Ahau was 9.2.0.0.0 4 Ahau 13 Uo, and therefore, in spite of the fact that the katun coefficient in a2 looks more like 4 than 2, the latter would appear to be the better reading, and 9.2.0.0.0, the probable date of this stela. These two readings are so dnuhrful, however, that they have been disregarded in the comparisons and conclusions which follow. 2 See Appendix VII. GENERAL COMPARISONS. 393 Seibal and Nakum, where the earhest dates are 9.16.0.0.0 and 9.17.0.0.0 respectively, and at neither of which is there a single Initial Series known; and by the end of the Old Empire the Initial Series had disappeared every- where except at Tikal (Stela 11). In the New Empire, barring the three Initial Series at Chichen Itza (Temple of the Initial Series), Holactun (Temple of the Initial Series), and Tuluum (Stela i). Initial Series dating had disappeared entirely, to be replaced by Period Ending dating and especially by tun-ending dating. The Carnegie Institution Central American Expedition of 191 8 estab- lished this lattei fact from its investigations in Yucatan, and further, that it was the unusual "winged-Cauac" variant of the tun-sign, which was used in these late New Empire tun-ending dates;i ^^d very recently the writer has discovered this same variant on an inscribed peccary skull from Tomb i at Copan (see page 380), mentioned above, which proves that use of this sign in this connection extended back to the Early Period of the Old Empire, or that it was in use for more than a thousand years. The development of the art of stone-carving among the Maya may best be studied at Copan in all its stages, save only the very beginnings. As already noted in Chapter II (see pages 54, 76), and to be described further in a later section, the eariiest style of glyph delineation is found most extensively at Tikal. This is characterized by 1. Very low, flat relief. 2. Irregular, non-rectangular outlines of the individual glyph-blocks. 3. General absence of specialized elements to denote the different signs. These eariy characteristics are best exemplified in the Tuxtla Statuette and the Leyden Plate, but omitting both from consideration on the ground that they are smaller antiquities and not large monolithic monuments, the same characteristics are to be found on the eariiest monuments, as, for example. Stela 9 at Uaxactun, and Stelse 4, 7, 8, 9, and 13 at Tikal, and Stela 20 at Copan. Indeed, the ouUines of the glyphs on Stela 9 at Uaxactun and Stelaj 8 and 9 at Tikal are so irregular as forcibly to suggest the carving of the Leyden Plate, where this characteristic is so pronounced. It is suggested in a later section that the origin of the Maya hieroglyphic writing, or even its transfer from an earlier medium such as wood to stone, need not be looked for at Copan, where the scarcity of inscriptions (indeed only one, Stela 20) which present this most reliable of all the criteria of antiquity, is probably to be interpreted as indicating that the Maya graphic system was developed elsewhere. But except for these very earliest stages of the hieroglyphic writing on stone, Copan is the best place at which its evolution may be traced, because of the greater abundance of material here than anywhere else. The rigid rectangular outlines of the glyph-blocks with corners only slightly rounded seems to have been developed first at Copan (Stela 24), as > Morley, 19180, pp. 272-274. 394 THE INSCRIPTIONS AT COPAN. well as the lavish wealth of detail, double-lining, cross-hatching, and the ornamentation of numerical elements such as the bar and dot, with interior decorations. Gradually, as the Old Empire wore on, these features became more simplified. The bar-and-dot elements lose their interior decoration and become plain, extensive double-lining, cross-hatching, and ornate details disappear, and the whole treatment becomes freer and simpler. Along with these developments went an increasing depth of relief. In the earliest texts the carving is very flat and low, but later it begins to stand out from the body of the monument, the background being cut back farther and farther until in the last group of stelae in the Great Plaza it is from lo to 12 cm. deep. A consistent development is also seen at Copan in the arrangement of the design in the stela type of monument, of which there are 36. While this does not follow a strictly chronologic order in the monumental sequence as a whole, the several "schools" or classes of monuments clearly were of successive origin, as the following table shows: Petiod. Date. Class 1(2). Class 2(4). Class 3 (9). Class 4 (11). Class 5(4). Class 6 (6). Eatly Middle....- Great ■ 9. I. 10. oC") Stela 20 Stela 22. Stela 25. 9. 2.10. 9. 4.10. 0.0. . . Stela 24. Stela 15 Stela 16. Stela 17. 9. 6. 0. 9. 6.10. of'") 0.0. . . Stela 9 Stela 21 9. 7. 0. 9. 9. 0. 9- 9- 5- 9. 9.10. 9. 11. 0. 9.11.15. 9.12. 5. 9.12. 10. 9. 13. 10. 9.14. 0. 9.15. 0. 9. 15. 5- 9.16. 5. 9. 16. 10. 9.17. 5. 9. 17. 12. 9. 17. 12. 9. 17. 12. of") Stela 18... Stela 7. . . . Stela E.... Stela P.... 0.0. . . Stelae I2, lo, 19, 13. Stelae 2, 23. Stela I . . . . Stela 3 Stela I . . . . Stela 6.... Stela J Stela 5.. Stela; A, B. Stela D. Stela M. Stela N.. n d(^) Stela II. Stela H. Stela C. 6 2 Stela 8. . Stelae F, 4. The simplest arrangement possible is where only one face is sculptured, the remaining three faces being plain but dressed (Class i), of which there are only 2 examples. The next is that where two alternate faces aie sculp- tured, the remaining two being plain but dressed (Class 2). This arrange- ment was introduced as early as 9.2.10.0.0, and 4 examples of it have been found. GENERAL COMPARISONS. 395 Possibly as early as 9.1.10.0.0, and certainly not later than 9.4.10.0.0, the all-glyphic stela was introduced, i. e., inscriptions on all four faces (Class 3), which was destined to endure for more than two centuries. This class is represented by 9 monuments. The first lepresentation of the human figure on stelae was possibly as early as 9.7.0.0.0 and certainly not later than 9.9.0.0.0, the back and sides being devoted to the inscriptions (Class 4). This class persisted for about 160 years, disappearing at the beginning of the Great Period. It has 11 examples. Stelae having human figures sculptured on two faces, the remaining two faces being inscribed with glyphs (Class 5), first appear in 9.1 1. 0.0.0, and lasted for about 125 years. This class, however, is limited to 4 examples. Early in the Great Period, the apparel of the human figure became so elaborate that the glyph-panel was crowded around onto the back of the stela, the sides being given over to the sweeping plumes of the head-dress or other details of the clothing (Class 6). This is the latest class of stela at Copan and lasted until the end, i. e., for about 45 years. It has 6 examples. The chronologic sequence of these several classes is very satisfactory and is only broken in two places. Stela 20, the earliest stela, being in Class 3 instead of Class i, the simplest arrangement, and Stela 8, a late stela being in an early class, i.e., a late return to an earlier arrangement, which had gone out of fashion more than two centuries before the date of its erection; but aside from these two exceptions the several classes follow each other in a consistent chronologic order. The same applies to the altar type of monument, although in this type the chronologic sequence of the individual monuments has not been so rigidly adhered to. On the basis of shape, they may be divided into the following classes: 1. Anthropomorphic altars, of which only 2 examples are known, the very crude examples under the foundations of Stelae 5 and 4, which may not be altars at all. (See fig. 67, a and b.) 2. Rectangular altars, 27 examples: (a) the earliest group being banded, J', K', L', M', ?', Q', X, Y, A', F', and the altar of Stela 6; (b) the middle group being without decoration: K, H', I', and V; {c) the late group being decorated with other and more complex designs: L, Q, S, and T with seated human or animal figures, B', C, and R with death-heads, D' and T with toads sprawled across their tops, U and Z with grotesque heads, W with the double-headed monster, and N' with a human figure. 3. Round altars, 12 examples: (a) a plain undecorated group: the altars of Stelae E, 19, 23, 5 (2), and i, and Altars T' and U'; (h) a banded group, the altar of Stela I and Altars 14,^ W, and G'. 4. Irregular-shaped altars, such as Altars Gi, G2, G3, and O, and all other altars not included in one or other of the above classes. Omitting Class i as doubtful, the earliest type of altar at Copan is the rectangular altar, Class 2. At first this is decorated with an arrangement ' Altars 14 and O have no inscription, for which reason neither has been described in Chapters II-IV. The former probably may be assigned to the Middle Period and the latter surely to the Great Period on stylistic grounds. 396 THE INSCRIPTIONS AT COPAN. of intersecting bands, but later these disappear, and during the Middle Period and early in the Great Period the altars of this class are plain. Still later the designs become more elaborate, consisting of seated human and animal figures, death-heads, grotesque heads, toads, double-headed mon- sters, and even the human figure. Class 2 altars are the commonest type of altar at Copan. The round altar (Class 3) is first found at the close of the Early Period in 9.9.5.0.0 and lasted down to 9.18.0.0.0, having its greatest development in the Middle Period. The irregular-shaped altar (Class 4) is a development of the Great Period, 9.17.0.0.0, and persisted down to the end, 9. 18. 10. 0.0. The above classification of the altars at Copan, while not so satisfactory as that of the stelae, shows nevertheless considerable evidenceof the chron- ologic order of the classes, the rectangular altars appearing first in the Early Period, the round altars appearing next at the end of the Early Period, reaching their greatest development in the Middle Period, and the irregular- shaped altars appearing last in the Great Period, a consistent development conforming with the increasing complexity of sculptural art as the Old Empire advanced. In addition to the two foregoing types of monuments, stelae, and altars, the inscriptions at Copan are presented upon steps, both interior and exte- rior, stairways, wall-panels, both interior and exterior, and possibly even cor- nices, but the last-mentioned medium, with the exception of the Hieroglyphic Stairway, is a late development, not appearing until the Great Period, at which time Maya art was at its zenith and the Maya sculptors were literally reaching out in all directions for new fields in which to express their esthetic feelings. PROBABLE FUNCTION OF THE MAYA MONUMENTS. From this study of the Copan inscriptions, it appears possible to indi- cate, in a general way at least, the probable function of the Maya monu- ments. At first, and during the period before Copan was founded, the mon- uments would appear to have been erected to commemorate specific events, such as actual historical happenings 01 current astronomical phenomena, examples of which are the Tuxtla Statuette, the Leyden Plate, and Stela 9 at Uaxactun, dating from 8.6.2.4. 17, 8. 14. 3.1. 12, and 8. 14. 10. 13. 15 respectively. Very early, however, possibly shortly after the beginning of Cycle 9, if the writer's reading of 9.0.10.0.0 for Stela 8 at Tikal is correct, and cer- tainly by 9. 1. 10. 0.0 or 9.2.10.0.0, when the first stela was erected at Copan, this custom changed, and thereafter, instead of erecting monuments to com- memorate such events, which in the very nature of the case had to be after the events had taken place, the practice was introduced of erecting the stelffi on the very days the events took place, namely, at the expiration of successive divisions of the Long Count. PROBABLE FUNCTION OF THE MAYA MONUMENTS. 397 A very obvious and natural reason for this change is suggested by the Maya method of counting time, i. e., in terms of elapsed units, which kept the priests continually looking forward to a date which would close the curre7it period, not backward to a date that had already passed. Their con- ception of time was such that they were always planning ahead, waiting for a future date which, when it arrived, closed a current time-period for them. Instead of erecting monuments to inaugurate new time periods, therefore, they erected them to commemorate the passage of elapsed ones. It was not the first day of the New Year which was of ceremonial importance to them as it is to us, but the last day of the Old Year. When a time period was finished and its corresponding monument erected, the priests were done with it, and were already looking forward to and preparing for the next period- marker. Any other procedure, such as erecting monuments after the occur- rence of the events they were to commemorate, must have been more or less upsetting to Maya psychology and contrary to their general conception of time. And thus, possibly even to obviate some such a feeling as this, there grew up the practice of selecting, in advance of the dates of actual dedication, the dates upon which the monuments were to be erected. This procedure had the very practical advantage of giving the priests ample opportunity to prepare for the important occasions which these period-endings were — - a preparation, indeed, which must have required a great deal of time. First the block of stone had to be quarried and from it the stela roughly shaped. Next it had to be transported, sometimes for a distance of several kilometers, to the site where it was to be erected, and there set up in pre- viously prepared foundations.^ The transportingof blocks of stone weighing sometimes as high as 50 tons (Stela E, Quirigua), was of itself no small undertaking, and must have taxed the resourcefulness of the rulers not a little. The method probabl}' followed was to use logs as rollers, which could be had in all sizes, of excellent hardwoods, and in unlimited quantities in the forests nearby, and to drag the blocks over these. The forests also pro- vided an abundance of natural ropes, lianas, hanging vines, and the like, and with these simple expedients and plenty of labor the monoliths were eventually moved from the quarries to the cities and erected in their ap- pointed places. But this was only half the undertaking. Before a monument could be put into formal use, that is, dedicated, it had to be sculptured, and such are the intricacies of the inscriptions and other designs that it is absolutely necessary to assume that the sculptors who executed them had working- drawings of the designs constantly before their eyes. The inscriptions, as we have seen, contain many fairly complicated calculations which had to be worked out in advance of the dates they dealt with, and probably were written down on paper or skin or wood to serve as 'These foundations, as noted more than once in the preceding chapters, are sometimes of an elaborate nature, cruciform subterranean chambers built of stone, over which the stelae were set up, their bases held in sockets formed by surrounding slabs. Such constructions, made of cut stone, had to be prepared in advance, and demanded coordinated activities. 398 THE INSCRIPTIONS AT COPAN. working-drawings for the sculptors, whose business it was to transfer these data and designs to the monuments themselves. All these labors consumed much time, and it thus became necessary to select, in advance of the time of actual dedication, the exact date upon which the ceremony itself was to be celebrated, i.e., the date upon which the monument was to be formally put into use, dedicated; in short, its con- temporaneous date. And so the writer believes the ends, first of the katuns next of the lahuntuns, and finally of the hotuns, were chosen for this purpose. That the katun-endings were chosen first rests upon the purely hypo- thetical ground that the katun would appear to be the best fitted period of the Maya chronological system available for this purpose. The tun was too short; it would have made the dates for the erection of stelae come around too frequently. The katun, the next higher period in their system of numer- ation, however, came to an end only once every 20 years, and, in the very beginning at least, this would have been often enough for the struggling young cities to have undertaken such an ambitious project. On the other hand, the archaeological evidence at Copan seems to indicate that the half-katuns, the lahuntuns, were the periods at first selected for this purpose. The three earliest surely deciphered monuments at Copan, Stelae 24, 15, and 9, all record lahuntun-endings, 9.2.10.0.0, 9.4.10.0.0, and 9.6.10.0.0 respectively, and the earliest on stylistic grounds, Stela 20, probably does also, 9.1.10.0.0. In fact, the earliest surely deciphered katun-ending does not occur until 9.9.0.0.0, or 128 years after the first lahuntun-ending, unless the reading suggested for Stela 9 at Tikal, 9.2.0.0.0, be accepted as correct. Nevertheless, the writer is inclined to believe that the katun- endings were first used for this purpose, and that at first it was the custom to erect monuments only at the end of every 20-year period, and not until later, when the Maya had grown more powerful, i. e., after they had reached Copan for example, were the lo-year periods also similarly commemorated. Thus, on the evidence furnished by Stelae 20, 24, 15, and 9 above, it appears probable that by the time the Maya founded Copan they were able to erect stelae at lo-yeai intervals, in which case the stelae commemorating the earlier katun-endings, 9.2.0.0.0, 9.3.0.0.0, 9.4.0.0.0, and 9.5.0.0.0 eithei still lie buried somewhere in the valley or they have been destroyed, or even more probable, all the early katun-endings may not have been thus marked, and stelae recording some of these dates may never have been erected. The earliest certain first or third hotun-marker known is Stela 25 at Piedras Negras in 9.8.15.0.0, ten years earlier than the earliest now known at Copan, i.e., Stela E in 9.9.5.0.0; but from this time onward the practice of marking the expiration of the five-year periods became the general rule and persisted down almost to the end of the Old Empire, except at the smaller cities, where sometimes the lahuntun-endings were used instead. (See Ap- pendix VII for further discussion of this practice.) It was stated in Chapter I, page 46, that the record of the Initial, Supplementary, and Secondary Series and Period Ending and Calendar PROBABLE FUNCTION OF THE MAYA MONUMENTS. 399 Round dates fills about half of the average Maya text, but in some inscrip- tions, the proportion will run much higher even than this. Thus, for example, on Altar W (see page 365), there are only two glyphs, one record- ing the day and the other the month of the Calendar Round date 1 1 Ahau 18 Mac; in other words, this text is 100 per cent deciphered. Or again, on a round altar near Stela i at Yaxchilan there are only eight glyphs, the first seven of which record the Initial Series 9.18.3.1.5 11 Chicchan (8 Kankin), the month being suppressed. Here only the last glyph is of unknown mean- ing, or, in other words, this text is 87.5 per cent deciphered. Such cases, however, are rare, and considered as a whole the Maya inscriptions may be said to be not more than 50 per cent deciphered. The above texts have been cited here only to illustrate the great importance which the Maya attached to the element of time, and possibly as tending to indicate that time in its different phases may even have been the chief content of their inscriptions. The above counts except the Supplementary Series deal with the fixing of specific dates in Maya chronology. We will examine some of these further. It was stated in the beginning of this section that the earliest Initial Series known do not record the ends of even periods of the Long Count, but that by the time the Maya had reached Copan this practice had undergone a change; and from this time on, the monuments, in the very great majority of cases, were erected at the ends of such periods. A further study of the six earliest surely deciphered monuments at Copan, Stelae 24, 15, 9, 7, E, and P, shows that five of them have the period-endings they were erected to commemorate recorded as their corresponding Initial Series. But in the case of Stela E, for the first time at Copan at least, although there is an earlier example at Piedras Negras (Stela 25), we find a slightly difi^erent condition. On Stela E the Initial Series is neither a hotun-ending nor the contemporaneous date of the monument, but an irregular date prior thereto, and the contemporaneous date, which is a hotun-ending, is recorded by a Secondary Series later on in the text. This new departure from the older practice became very popular elsewhere in the Old Empire and is only slightly less frequent at Copan, where it appears on the following monuments: Stela E, 19, 10 (?), I, I, 5, A, the west altar of Stela 5, the altai of Stela i, Altar H', and Altar P. The following examples from Copan, Naranjo, Piedras Negras, and Quirigua will illustrate its use: Copan, Stela I. 9.12.3.14.0 Initial Series 1. 4.0 9.12.5. 0.0 Period Ending and contcmpo raneous date Naeanjo, Stela 24. 9.12.10. 5.12 Initial Series 5. 7.15 9.12.15.13. 7 11. 8. 1 9.13. 7. 3. S 2.14.12 9.13.10. 0. Period End- ing and con- temporane- OU.S date Piedras NEGRAfi, Stela 3. 9.12. 2. 0.16 Initial Series 12.10. 9.12.14.10.16 1. 1.11.10 9.13.16. 4. 6 3. 8.15 9.13.19.13. 1 4.19 9.14. 0. 0. second Initial Series and contein po- raneous date Quirigua, Stela F. 9.14.13. 4.17 Initial Series 13. 9. 9 9.15. 6.14. 6 9.14.13. 4.17 repeated 1.16.13. 3 9.16.10. 0. second Initial Series and con tern po- raneous date 400 THE INSCRIPTIONS AT COPAN. In the first example above, the inscription starts with the Initial Seiies, 9. 1 2. 3. 14.0, which a single Secondary Series number brings forward to the current hotun-ending, 9.12.5.0.0, the contemporaneous date of the monu- ment. This is the customary, although not the invariable, practice at Copan, where there is usually not more than one intermediate date between the Initial Series date and the contemporaneous date. The second example begins with the Initial Series, 9. 12. 10. 5. 12, which three Secondary Series numbers, 5.7.15, 11. 8.1, and 2. 14. 12, reaching two intermediate dates, bring forward to the contemporaneous date, the lahuntun-ending 9. 13. 10. 0.0, nearly a katun later. The third example begins with the Initial Series 9. 12.2.0. 16, which four Secondary Series, 1 2.10.0, i.i.ii.io, 3.8.15, and 4.19 bring forward to the contemporaneous date, the katun-ending 9.14.0.0.0. On the front of this monument this is recorded as a Period Ending date, but on the side it is repeated as an Initial Series. The fourth example above begins with an Initial Series, 9. 14. 13. 4. 17, which one Secondary Series, 13.9.9, brings forward to an important inter- mediate date, and which another, 1. 16. 13. 3, also proceeding from the Initial Series date direct, brings down to the contemporaneous date, the lahuntun- ending 9. 16. 10. 0.0, also repeated as an Initial Series on one of the sides. Analyzing these data, we find in every case that the opening date is not at the end of any particular period of the Long Count, but is a date which was apparently determined by some historical event or astronomical phe- nomenon. This odd date is brought forward in every case, however, by one or more Secondary Series numbers to the particular hotun-ending which each of the above monuments was erected to commemorate. Let us examine the case of Stela 3 at Piedras Negras somewhat further. The previous hotun-ending at this city, 9. 13. 15. 0.0, was commemorated by the erection of Stela i, and the following hotun-ending, 9.14.5.0.0, by the erection of Stela 5. That is, it seems fair to assume Stela 3 must have been quarried, transported, erected, and sculptured between 9. 13. 15. 0.0, the previous hotun-ending, and its own contemporaneous date, 9.14.0.0.0, five years later. But on the other hand, the first two dates on it, 9.12.2.0.6 and 9. 1 2. 14. 10. 16, are approximately 38 years and 26 years earlier than its con- temporaneous date, and thus lie without the hotun which it was erected to commemorate, i. e., the hotun ending in 9.14.0.0.0. The same condition also obtains in regard to the first two dates on Stela F at Quirigua and the first two on Stela 24 at Naranjo, both pairs lying without the spans of the hotuns these monuments were respectively erected to commemorate. The above stelae, moreoever, are no exception to the general rule, and it will be seen that the contemporaneous dates are quite as frequently recorded by Period Ending or Calendar Round dates, particularly in the Middle and Great Periods, as by Initial Series, and finally that the Initial Series fre- quently precede the corresponding contemporaneous dates by several years or decades, and in a few instances even by many centuries. PROBABLE FUNCTION OF THE MAYA MONUMENTS. 4OI There remains to be considered the Supplementary Series, a more detailed description of which will be found in Appendix VI. First of all, it may be accepted as proved beyond all reasonable doubt that the Supple- mentary Series is a lunar count, and that it never occurs without an accom- panying Initial Series. The principal data set forth in this count may be summarized as follows: 1. Glyph A declares the kind of month, i. e. whether composed of 29 or 30 days, in which the accompanying Initial Series date falls. 2. Glyph C probably declares the position of this month in a group composed of 5 or 6 such months, that is, whether it was the first, second, third, fourth, fifth, or sixth division of such a group; and the length of this larger lunar period, formed by grouping together 5 or 6 such months, seems to have depended in some as yet undetermined way upon the eclipse phenomenon, certainly of the sun, possibly of the moon, and possibly even of both. (See Appendix VI.) 3. Glyphs D and E in some way further ampHfy the data set forth in Glyph C. 4. Glyph X declares still further data, the nature of which still remains inde- terminate. The three remaining glyphs of the Supplementary Series, B, F, and G, are constant non-numerical signs, probably having some generalized mean- ing such as " here ends the diurnal count " and " here begins or ends the lunar count," and in no way change the meanings of the different Supplementary Series in which they occur. In this latter respect they probably play a much less important role than the Initial Series introducing glyphs do in the mean- ing of the Initial Series, for in the latter glyph the variable elements change according to some unknown law in the different Initial Series, while in the Supplementary Series, Glyphs B, F, and G, except for minor and probably unessential stylistic differences, always remain the same. It should be noted, moreover, that these lunar and eclipse data are given only for Initial Series dates, and that except for those very rare cases, eight in number,^ where two Initial Series are recorded on the same monument, only one Sup- plementary Series appears on each monument. In summing up all the foregoing evidence as to the nature and probable function of the Maya stela;, it appears possible to state the following general conclusions concerning them: 1. They were period-markers erected to commemorate the passage of suc- cessive units of the Maya Era. 2. The unit chosen for this purpose was at first the katun, later the lahuntun, and still later, at the height of the Old Empire, the hotun. 3. The Initial and Secondary Series, Period Ending, and Calendar Round dates record specific days in the Maya chronological era, and the Supplementary Series set forth certain lunar and eclipse data concerning the Initial Series dates which they respectively accompany. The record of the information given in No. 3 above exhausts approxi- mately so per cent o f the Maya glyphs, and what have we? The con- ■Tikal, Stela 17; Yaxchilan, Stela 11; Copan, Stela 3; Piedras Negras, Stelae I and 3; and Quirigua, Stelse D, E, and F; the two Initial Series on the Tikal stela, as well as the two on the Yaxchilan stela, in each case recording the same date. 402 THE INSCRIPTIONS AT COPAN. temporaneous dates of the monuments, and occasionally other associated dates which were astronomically or historically important at the time, and finally certain lunar and eclipse data pertaining thereto; in short, we have used up half of the signs in setting forth these few matters. The meaning of the remaining half of the inscriptions is largely a sealed book as yet, and here if anywhere we must look for the subject-matter of Maya history. Judging from the glyphs already deciphered, this would hardly seem to be a particularly promising field, but happily, with the above three numerical counts out of the way (the Initial, Supplementary, and Secondary Series), there appear to remain in the undeciphered glyphs very few of a numerical nature, and we may possibly look forward with some degree of confidence to finding, among the latter, place-names, personal- names, and signs of generalized meaning, by the aid of which we will event- ually be able to fill in the background of Maya history as successfully as we have already constructed its chronological framework. THE ORIGIN OF THE MAYA CIVILIZATION. As pointed out in Chapter I, the Maya have been variously derived by one authority or another from Egypt, Carthage, Java, and southern India, even the lost continent of Atlantis having competed at one time for the honor of their origin. Lord Kingsborough, in nine costly volumes, sought to trace their descent from the Ten Lost Tribes of Israel, and Le Plongeon, reversing this at least usual order of procedure, tried to people the Old World from the New, believing Egypt to have been colonized from Yucatan more than ii,ooo years ago.^ Unfortunately, these highly improbable hypotheses have not entirely disappeared before the advance of the science, since the Egyptian connection has been revived recently by Smith on the basis of erroneous identifications and purely superficial similarities. (See note 4, page 28.) Happily, in all this maze of extravagant speculation and improbable conjecture, we are not without some trustworthy lines of direct evidence, linguistic as well as archaeologic, which throw light upon this important ques- tion. These are: 1. The provenance of the Tuxtla Statuette, upon which is inscribed the earliest date in the Maya hieroglyphic writing. 2. The provenance of the Huasteca, the only Maya-speaking tribe or group which is not contiguous to the main body of the Maya linguistic family, being entirely surrounded by other linguistic stocks, chiefly Nahuatlan and Totonacan, and the only Maya group, moreover, which lives in a region showing no traces of typical Maya archaeological remains. 3. The provenance of the earliest dates in the region covered by the remains of the Maya civilization. ' Le Plongeon (1886, p. 113) says in this connection: "In my work The Monuments of Mayax I have shown how legends accompanying the images of several of the Egyptian deities, when interpreted by means of the Maya language, point directly to Mayax as the birthplace of the Egyptian civilization." THE ORIGIN OF THE MAYA CIVILIZATION. 403 Let US examine these several lines of evidence in the above order. The Tuxtla Statuette was found near San Andres Tuxthi, in the State of Vera Cruz, Mexico, about 1902, and was acquired for the United States National Museum by Holmes in 1903. (See fig. 63.) It is of nephrite, 16.5 cm. high, 9.5 cm. in diameter at /;.'.'' J^''\ the broadest place, and represents a bird-like figure //■■'i'aA^C::,^^ with a human head. On the front, as already noted M '\\5 i|^ // )''\. several times, is inscribed the date 8.6.2.4.17, which /\ \\H |: / \ is some 160 years earlier than the next earliest date, \ 'irt--' / \ i. e., on the Leyden Plate, and is the earliest date yet discovered in the Corpus Inscriptionum Mayarum. Bowditch has challenged the accuracy of this read- ing, and indeed that these glyphs record an Initial Klii; H \ J- ] fl/"/M Series at all;' but on insufticient grounds the writer u '4^ ' " ^ ■■■■' X"^ believes,^ and the general opinion now is, that this ■;; \ -^ -_; f^^ specimen is not only genuine, but that its date also §f .'i gi? ' / probably corresponds with the time of its execution. ■■-tr Jl ^.--^ This date, as the writer has shown elsewhere,^ ^'" olTuSX't" «".'""" is recorded in bar-and-dot numerals without the accompanying period-glyphs, the value of the latter being determined by their positions from bottom to top in the number itself, like the Initial Series in the Dresden Codex.^ The provenance of this specimen, because of its very early date, is thus of unusual importance as possibly indicating where the Maya first began to record their chronology, and particularly so in view of the fact that it was recovered from a region where distinctive Maya remains have not been found. Indeed, San Andres Tuxtla is in Nahua^ country about 225 kilo- meters north of west from Comalcalco, the westernmost Maya site now known. '^ In other words, it lies some distance outside of the territory now definitely associated with Maya remains, such as the typical Maya archi- tecture and sculpture, for example. Turning next to the second line of evidence mentioned above, the Huasteca, a Maya-speaking people living on the Gulf Coast-plain of Mexico between Tuxpam and the Panuco River, in the States of Vera Cruz, Hidalgo, San Luis Potosi, and Tamaulipas, a unique condition is found. (See fig. 64.) Here we have a Maya linguistic island surrounded by a sea of Totonaca, Nahua, Otonii, and Tamaulipeca, that is to say, a branch of the Maya linguistic stock entirely separated by intervening stocks from the main body of the family, the nearest bianch of which to the Huasteca is the Chontal, 750 kilometers to the southeast.' Nor is this linguistic isolation the only extraordinary feature about the Huasteca. Although speaking a Maya dialect, their prehistoric culture, judging from its remains, shows no resem- 1 Holmes, 1907, pp. 695,696. =Morley, 1915, pp. 194-196. <* Xhomas, 191 1, map. = Ibid., pp. 696-700. ■■ Ibid., pp. 266-273. " Charnay, 1887, pp. 194-210. 'It is possible that the Totonaca were also originally a Maya branch. See Thomas, 1911, p. 49. 404 THE INSCRIPTIONS AT COPAN. blance whatsoever to that of the Maya, either of the Old or New Empire, or even to those more culturally distant Maya, such as the Quiche, Cak- chiquel, Tzutuhil, Tzotzil, Tzendal, Chontal, and Mame, to mention only the more important tribes, inhabiting the highlands of southeastern Mexico and southern and western Guatemala, for the distribution and linguistic affiliations of which see Appendix XII and figure 91. Fig. 64. — Map showing provenance of earliest Maya inscriptions and probable line of migration of the Maya into their historical habitat. The Huasteca have no hieroglyphic writing, no highly developed cal- endar system based upon a 260-day Sacred Year and a 365-day solar year, and no chronology recorded in terms of a vigesimal system of numeration, the first two elements of which (the 260 and 365-day years), spread from the Maya to the several Zapotec, Miztec, Nahua and Totonac tribes of southern Mexico and Central America about the close of the Old Empire (10.2.0.0.0) } Their material culture, architecture, sculpture, and ceramic art, etc., show no resemblances to the corresponding phases of the Maya civiliza- tion, and, indeed, so dissimilar are the two cultures in all things, save only 1 Brlnton says in this connection (1893, p. 260): "There is no direct evidence that it [the hieroglyphic-writing and calendar] had extended to the Huastecas, of Maya lineage on the Rio Panuco, but it was in vogue among the Totonacos, their neighbors to the south on the Gulf of Me-xico." THE ORIGIN OF THE MAYA CIVILIZATION. 405 that of a common language, that it is necessary to postulate a separation of the Huasteca from the other members of the Maya stock before the latter had developed the highly distinctive civilization by which they were charac- terized in later times. ^ Linguistically considered, it is necessary to regard both as having had a common origin, however remote; but culturally con- sidered, the two must just as surely have separated before the Maya devel- oped their civilization. But how was this separation effected ? Were the Huasteca left behind in a general southward movement of the Maya race, or do they represent a branch which pushed northward and away from the parent stock? This question can not be definitely answered until the Gulf Coast-plain of Mexico from the Grijalva to the Panuco River has been carefully examined, and excavations made at the principal archaeological sites, but already it seems probable that the former explanation is the more likely to be correct. It will be seen in figure 64 that the provenance of the Tuxtla Statuette is about midway between the territory occupied by the Huasteca and the northern Peten region, where the earliest inscriptions within the Maya area have been found; that is, the earliest of all Maya texts comes from a region where typical Maya remains have not been found, which, however, stands about midway between the historic region occupied by the Maya farther south, and a region now occupied by Maya-speaking people, whose remains show no Maya cultural characteristics. Sapper attempts to solve this problem on the basis of certain linguistic evidence, laying particular stress on the two following factors: (i) that the name for pine-tree among the different Maya-speaking peoples shows a remarkable agreement, and (2) his discovery of a small Maya tribe in the highlands of southeastern Chiapas, the Chicomucelteca, which he claims is more closely related to the Huasteca linguistically than any other tribe of the Maya stock. From the first he argues that the original home of the Maya was in a pine-tree country, i. e., a mountainous land; and because of the second he suggests the highlands of Chiapas and Guatemala as the most likely place for this to have been : "Then it seems as most probable that the mountainous country of Chiapas- Guatemala is the original seat of the Maya family of peoples, from whence at an early date the Maya and Choi tribes must have migrated to the lowland, while still earlier the Huasteca emigrated from Chiapas along the Atlantic coast-plain to the north and settled at their present seats. "" This, of course, may be true, but it is easier to conceive the Chicomu- celteca, whom Sapper estimated doubtfully at not more than 4,000 in 1897, as moving southward to their present habitat in southeastern Chiapas at some early time, than it is to conceive the Huasteca, at present numbering at least 42,000, as moving northward from the highlands of Chiapas to their 1 See Seler, igoi-igoS, vol. II, pp. 168-183; Prieto, 1873; Joyce, 1914, and Staub, 1919, for descriptions of the Huasteca and their material culture. 2 See Sapper, 1897, p. 398. 4o6 THE INSCRIPTIONS AT COPAN. present habitat at an equally early date. And in conclusion, the writer believes the most probable place of origin for the Maya civilization is some- where on the Gulf Coast-plain of Mexico between i8° and 22° north latitude. Indeed, in a later publication (1905) Sapper himself would seem to have reached a similar conclusion: "On the other hand it is very difficult to distinguish the causes of the great distance separating the Huasteca from the Mayas with whom they belong ethno- logically. And this difficulty has increased since I succeeded in finding in the vil- lage of Chicomucelo in Chiapas, near the Guatemalan frontier, a language which is very like the Huastec and is in striking contrast to the other members of the Maya stock of languages. Before we knew of the Chicomucelteca, we could assume that in the original immigration of the Maya nations from the north, a part remained behind in the Huasteca, and there the language developed in a peculiar way, because the connection with its kindred was interrupted. But how came the Chicomu- celteca in their present home, forming a linguistic island in the midst of Maya peoples? Can it be that they are only a Huasteca colony, which has recently settled there. "^ Joyce believes the Maya civilization originated where it reached its zenith during the Old Empire, namely, in northern Guatemala: "The isolation of this definitely Maya branch [i.t'., the Huasteca] would seem to imply that the Maya in the earliest days of all must have spread from Guatemala, up the east coast of Mexico as far as Tampico, penetrating into Chiapas, and possibly into Oaxaca, and colonizing the Mexican valley, where they found a primitive people akin to the earliest population of Michoacan. In times subsequent to what I may call this proto-Maya movement, there took place in the southern fertile region a great cultural development, culminating in the organization of a calendar, the invention of a script, and construction of the ruined 'cities. '"- This is a far-reaching extension of the Maya indeed, and one which the archaeological evidence will hardly justify, certainly not as to their coloniza- tion of the Valley of Mexico, and probably not as to that of Oaxaca either. Joyce's hypothesis of the autochthonous origin of the Maya civilization, he admits, is open to the serious a priori objection that no beginnings ot the culture have been found in this region, "that civilization springs," as it were, "full-blown from the earth." He overcomes this difficulty by ascribing the failure to find these earlier traces to the lack of systematic excavation in this region (negative evidence at best) and to the fact (generally admitted) that the art of carving was first developed on some perishable material such as wood, and that therefore the earlier remains have not survived. Finally, he dismisses the evidence afforded by the Tuxtla Statuette with the brief comment that it is "artificial or mythical."'' The writer finds himself unable to agree with any of these several con- clusions. In the first place, for reasons already stated, he believes the date inscribed on the Tuxtla Statuette was contemporaneous with the date of its execution; in the second place, that the Huasteca are much more likely to represent a backward branch of the Maya, who have always remained in or 1 See Sapper, 1905, p. 5. ^ See Joyce, 1914, p. 36S. ' Ibid., p. 254 note I. THE ORIGIN OF THE MAYA CIVILIZATION. 407 near their original habitat, than that they were a more progressive element who pushed out first and more distantly than any other Maya branch; and finally, he believes sufficient evidence has been presented to show that it is extremely improbable the Maya developed their civilization in the region where it reached its zenith during the Old Empire, i.e., generally speaking, the Peten region of northern Guatemala. The hypothesis suggested below, on the other hand, should by no means be accepted as proved. As yet the evidence upon which it or any other is based, which seeks to explain the origin of the Maya civilization, is too insufficient to permit final conclusions, but the writer ventures to believe that it better meets the conditions imposed by the archaeologic and linguistic evidence than any other. At some remote epoch, sufficiently prior to 8.6.2.4.17 for them to have developed such a complex calendar, chronology, and hieroglyphic writing as they possessed even at that early date (about 100 b. c, see Appendix II), the Maya may have lived somewhere north of their habitat during the Old Empire (see fig. 64); and since a Maya-speaking people still inhabit such a region between Tuxpam and the Panuco River, this may possibly have been the place. Before developing their calendar, chronology, hieroglyphic writing, and distinctive civilization, by which they were characterized in later times, the great mass of the stock moved south, leaving behind, perhaps, the more backward elements, who later developed a far lower and different culture, but who continued to speak the mother Maya tongue, and who later became the Huasteca of historic times. Somewhere between the above region and San Andres Tuxtla, if our hypothesis be correct, the Maya civilization had its origin, and their calendar and chronology had been perfected to such a point that by 8.6.2.4.17 they were able to carve upon a very hard stone (the Tuxtla Statuette has a hard- ness of about 7) the earliest date yet found in their hieroglyphic writing. How long prior to this date it took them to make and record the necessary astronomical observations on the sun and moon, upon which their calendar is based, and having at last sufficient data at hand, how long it took them to perfect their remarkable chronological system, is of course impossible to say. The first process, however, would appear a priori to have been much the longer of the two. Since, once certain astronomical facts, such as the apparent revolutions of the sun and moon around the earth, had been deter- mined, the invention of the whole elaborate calendar and chronology, including the arithmetical and notational systems, might have been the work of a single individual. Such highly complex and arbitrary inventions, while based upon data slowly and laboriously acquired over long periods of time, are apt to flower quickly once a certain stage is reached — a sudden liquidation of long-accumulated intellectual investments; and so the actual construction of the Maya calendar and chronology may have come swiftly, 408 THE INSCRIPTIONS AT COPAN. once the astronomical data upon which they were based had been accumu- lated in sufficient quantities to establish therefrom certain dependable astronomical laws. And possibly this invention may have taken place not long prior to 8.6.2.4.17, the date on the Tuxtla Statuette, since no certain earlier contemporaneous date has yet been found in the Maya writing.^ The hypothesis outlined above rests on the three following postulates: 1. That the date on the Tuxtla Statuette is 8.6.2.4.17, 2. That this date is the contemporaneous date of the specimen, 3. That the specimen was made in the general region where it was discovered, i. e., that it was not brought from somewhere else by trade. Concerning the first, the writer has no doubt as to its accuracy. The Initial Series number 8.6.2.4.17 leads to the terminal date 8 Caban o Kankin, and the day-sign coefficient 8 appears regularly in its proper place below this number, attached to a glyph the main element of which has unfor- tunately been effaced. (See fig. 63.) This is too close an agreement to be the result of coincidence only, and practically establishes, first, that the number is an Initial Series, and second, that it reads 8.6.2.4.17, even though the corresponding month-part of the Initial Series terminal date does not appear in the rest of the text, at least in a recognizable form. Possi- bly the month-part may have been suppressed altogether, as was not infre- quently the case; for the discussion of which point see pages 138, 139, 157. Concerning the second postulate above, since 8.6.2.4.17 is the only date on the specimen, the logical assumption is that this date was present time when it was inscribed, namely, that it is the contemporaneous date of the inscription. The third postulate above is less certain than the other two, although the writer believes it is probably correct. The Tuxtla Statuette is small, however, 16.5 cm. high and 9.5 cm. maximum diameter, and could easily have been carried long distances in trade, straying far from its original place of execution. Still, it was found near San Andres Tuxtla, and perhaps the burden ot proof rests on those who would argue that its original provenance was elsewhere. Weighing all the evidence, therefore, and giving due weight to the fact that we have a Maya-speaking people 400 to 500 kilometers still farther to the northwest (see fig. 64), it seems not improbable that the Maya were in the San Andres Tuxtla region in 8.6.2.4.17, when the Tuxtla Statuette was made, and that they may have come hither from some region farther north and possibly from as far north as the present habitat of the Huasteca. ^ Seler (1902-1908, vol. II, p. 30) assumes that the Maya graphic system, chronology, and calendar must have been developed at least two centuries before its first record on stone: "This would place the nephrite plate from the Rio Gracioza (?) [the Leyden Plate] approximately in the year goo [a. d.] and according to my previous assumption fix upon the year 700 as the latest limit, which we should have to assume for the discovery of the elements of the writing, the invention of the calendar, and the age of the kingdom of Tollan." It should be noted that at the time Seler wrote the above passage ( 1902) the Leyden Plate was the earliest dated object known. Aside from the inaccuracy of his equivalent dates in Christian chronology (see Appendix II, pp. 528, 534), this minimum limit of 200 years appears too great to the writer, since the date on the Tuxtla Statuette could conceivably, although improbably, have been recorded during the lifetime of the inventor of the graphic system and calendar. THE ORIGIN OF THE MAYA CIVILIZATION. 4^9 One other point tends strongly to confirm the accuracy of this conclu- sion, namely, the existence of another tribe, the Totonaca, who occupy the coast-plain just north of Vera Cruz, i. e., between the Huasteca and San Andres Tuxtla, and who are linguistically and cuhurally said to be related to the Maya. Speaking of their linguistic affiliations, Swanton says: "The present tendency of linguistic opinion is to place the Totonac language in the Mayan family, thus bringing it into relation with the Huasteca. The long friendly relations between the two tribes corresponds with this opinion. Orozco y Berra" expressed his beHef in the relationship of the two dialects."- The material cuhure of the Totonaca, moreover, shows strong resem- blances to that of the Maya; although this is probably due to actual con- tact with the Maya during or after the Great Period rather than to an earlier common origin of the two cultures. The so-called laughing heads of the Totonacan region, modeled in clay, are clearly Mayan in feeling, and in ceramic motives direct connections may be traced.^ Says Spinden in speaking of Totonacan art : "This apparent connection in language is all the more interesting in view of the character of Totonacan art, which also shows a strong strain of Mayan feehng and technique in certain products but an unmistakable likeness to the archaic art of the Mexican highlands in certain other products. The pottery faces in the archaic style are advanced beyond the average of such work and probably represent a late phase. It is possible to bring forward examples of every degree of transition from the archaic style to the classical Mayan of Tabasco and Chiapas. Curiously enough it does not seem possible to extend these linking likenesses to the Huastecas. Finally, in a passage from Brinton, already quoted on page 404, note i, he states that the Totonaca used a hieroglyphic writing and a calendar system. Taken together, these several factors point to the former racial unity of the Maya and the Totonaca, and in the present connection tend to indicate that the region where the Maya civilization originated lay somewhere to the northwest of San Andres Tuxtla and possibly even as far south as the present habitat of the Totonaca. To trace the probable history of the Maya before the separation of the Huasteca from the main body of the stock becomes largely a matter of specu- lation. Spinden has shown that, coincident with the invention and primary dissemination of agriculture, possibly somewhere in and from the highlands of central Mexico, there seems to have spread over the greater part of Middle America and the northwestern corner of South America, largely restricted to arid tropical regions and avoiding the low Atlantic Coast-plain with its heavy rainfall and rich alluvial soil, an early homogeneous culture characterized by a simple and undeveloped religion, an unsymbolic art, pottery-making, and loom-weaving, which he calls the Archaic Horizon.^ 1 See Orozco y Berra, 1864, p. 214. ^ Thomas, 191 1, p. 49- ..,,/; 3 See Strebel, 18S4; ibid, 1885-1889; ibid, 1904; and Spinden, 1917, PP. 145-15°- Spinden, 1917. PP- Hs. H^- 'Spinden, igiS.PP- 45 1-459,467-469. ^bid., 1917. PP- 43^4! ^'d., 19m, PP- 181-188; ib,d., 1917*. PP- 269-276. 410 THE INSCRIPTIONS AT COPAN. He further believes this culture was the common product of tribes then living in the highlands of central Mexico, but that the Nahua led in its development and dissemination, and that it was carried by them southward down the Pacific Coast-plain of Central America to Guatemala, Salvador, Nicaragua,^ and as far south as the Isthmus of Panama.^ It appears as not unlikely that before this period the Maya may have found their way to the Gulf Coast-plain, possibly to the general region now occupied by the Huasteca or Totonaca. If so, they must have been largely a hunting and fishing people, depending only partly upon the many wild fruits and plants of the tropical forest to supplement their food-supply, moving to and fro in their quest for food, and not held to fixed abodes by the exigencies of an agricultural life, their time filled with and their energies absorbed by the struggle for bare existence. To such a people, living in such an environment richly endowed by nature with a fertile soil and a warm, moist climate, the factors most essen- tial for the growth of crops, and wanting only cultivation in order to yield a maximum return for a minimum effort, there may have come, from the highlands to their west, knowledge of the practice of agriculture, probably first as applied to the cultivation of corn. Soon, because of the several factors just mentioned, the returns in pro- portion to the effort expended became very much greater than on the arid highlands; nature herself lent a more helping hand; the harvests became more and more abundant, until from scarcely sufficing for the general needs of the tribe from one harvest to the next, reserve supplies of food began to be accumulated, thus releasing from purely economic production energies which could be directed toward other ends, religious and esthetic. The introduction of agriculture brought about a tremendous change in the lives of the groups which it touched. Instead of moving hither and thither, driven by the necessities of a game, fish, and only casually vege- tarian dietary, living in temporary houses under a very loose social and governmental organization, agriculture for the first time made possible, indeed compelled, the establishment of permanent homes and developed the need for property rights. Larger social units than the family became pos- sible, such as the village, clan, and tribe; and with less and less time being absorbed in the food-quest, more and more time was devoted to the development of the household arts; pottery-making and loom-weaving were invented; religion became more complex, and esthetic instincts wider and more elaborate in their expression. Under some such conditions as these the Maya emerged from a nomadic, hunting, and fishing life to a sedentary agricultural one, and because their habitat was so richly endowed by nature to begin with, and far more fertile than that of the arid-highland peoples, the resulting civilization which they were able to develop gradually surpassed all surrounding cultures, and event- ' Spinden, 1917, p. 43. ^ Ibid., 1917b, p. 269. THE ORIGIN OF THE MAYA CIVILIZATION. 411 ually under the Old Empire became the finest expression of the aboriginal American mind. In some such a way, then, the Maya civilization had its origin, presum- ably somewhere on the Gulf Coast-plain of Mexico, probably between the Grijalva and Panuco Rivers, 18° to 22° north latitude, and the writer believes, although this point is yet incapable of direct proof, some time toward the end of the second or the beginning of the first millenium before the birth of Christ. Finally, although the earliest dated text, the Tuxtla Statuette {circa 100 B. c), is doubtless at least a thousand years later than this begin- ning, it is probable that future excavations in this region, archaeologically so strategic in the solution of the Maya problem, will bring to light still earlier texts that will carry us back still nearer to the beginning of the Maya hiero- glyphic writing, which, so far as the calculations involved in the Initial Series proper are concerned, is as perfected and as finished on the Tuxtla Statuette as in the latest Initial Series known, some 1,100 years later. ^ Turning next to the consideration of the third line of evidence mentioned on page 402, namely, the provenance of the earliest dates in the Maya area proper, that is, exclud- ing the Tuxtla Statuette, which we have just seen was found without this region, we at last reach firm historical ground. The next earliest surely dated contemporaneous texts in the Corpus Inscriptionum Mayarum are the following: I. The Leyden Plate, 8.14. 3. 1.12 8. 14. 10.13. 15 9. 2.10. o. o 9. 2.13. o. o 9. 3. II. 2. o 9. 3.13. o. o 9. 4.10. o. o 9. 6. 3. 9.15 9. 6.10. o. o In addition to these, there are 16 other monuments at Copan and 10 others at Tikal which are earlier than the ear- liest monument now known at any third site, namely. Stela 25 at Piedras Negras, recording the hotun-ending 9.8.15.0.0. Analyzing, first, the provenance of these earliest dated texts, we find that the earliest one of all is on a small neph- {\r?S7]Vi?r^^ rite celt, the Leyden Plate (see fig. 65), found just west of U^y/f^Si the mouth of the Motagua River near the Rio Graciosa. It ^^jf^j^ is only 21.6 cm. long, 7 to 8.5 cm. wide, and 0.2 to 0.5 cm. thick, in fact, so small an object (much smaller and lighter than the Tuxtla Statuette) that it could easily have been carried long distances with little or no inconvenience. When this is taken into consideration with the fact that jade (neph- 'The latest Initial Series known is that on the back wall of the Temple of the Initial Series at Holactun (Xcalum- kin), Yucatan. As shown by the writer elsewhere (19180, p. 274), this probably records the date 1 1. 2. 8. 4. 9, although 10.9.8.4.9 is not an impossible reading. Uaxactun, Stela 9, 3. Copan, Stela 24, 4. Tikal, Stela 3, 5. Tikal, Stela 10, 6. Uaxactun, Stela 3, 7. Copan, Stela 15, 8. Tikal, Stela 17, g. Copan, Stela 9, 0/^5 Fig. 65. — Inscription on Leyden Plate. 412 THE INSCRIPTIONS AT COPAN. -Inscription on Stela 9 at Uaxactun. rite or jadeite) was the most precious of all materials to the ancient Maya, it will be seen that the precise locality where such a small object was found is of little value in determining its place of origin. It was so small of bulk, so light of weight, so precious of material, and so cherished because of its hoary antiquity even as early as the Middle Period of the Old Empire, that it would have been, and possibly was, carried far from its place of manu- facture. This same objection, however, can not be raised against the next inscrip- tion, that on Stela 9 at Uaxactun (see fig. 66), which is less than 8 years later than the Leyden Plate, and is the oldest monument, i. e., large, monolithic remain, yet found in the Maya area, or indeed recorded in the Maya hiero- glyphic writing. This site was discovered by the Carnegie Institution Central American Expedition of 1916, in the Department of Peten, Guatemala, north of Lake Peten Itza, some 25 to 30 kilometers northwest of Tikal and 600 kilometers south of east from San Andres Tuxtla (see plate i and fig. 64).^ This monument is a large shaft of limestone, 2.9 meters high (above the ground), 1.22 meters wide at the base, 76 cm. wide at the top, and 60 cm. thick, and weighs several tons, much too heavy an object, in fact, ever to have been moved far from its place of manu- facture. It is still standing, although leaning far out of the perpendicular, with its front forward, in which position the back or surface having the single glyph-panel has suffered more from weathering than the front. Even if there were not other very early monuments at Uaxactun in addition to Stela 9,^ we are justified, on the evidence aff^orded by this monument alone, in assuming that we have here a stela in situ, recording a contemporaneous Cycle 8 date; in other words, that on the basis of the dated remains in situ Uaxactun is the oldest Maya site yet discovered. Omitting Stela 20 at Copan, although the writer believes its date is 9. 1. 10. 0.0 as suggested, the next earliest monument is Stela 24 at Copan, which is 160 years later than Stela 9 at Uaxactun; and then during the next 80 years follow Stelae 3 and 10 at Tikal, Stela 3 at Uaxactun, Stela 15 at Copan, Stela 17 at Tikal, and Stela 9 at Copan. Analyzing next the dates of these nine monuments, it will be found that Stela 9 at Uaxactun is 166 years later than the Tuxtla Statuette; in other words, that during the interval between 8.6.2.4.17, the date of the latter, and 8. 14. 10. 13. 15, the date of the former, the Maya had continued their migration southeastward and had occupied the northern part of the 1 See Motley, 19163, pp. 339-341. This site was named Uaxactun because of the discovery of the Cycle 8 Initial Series on Stela 9 here, ua.\ac being Maya for 8, and lun Maya for stone, i. e., "8 stone." ^Other early monuments at Uaxactun are: Stela 5, S.i5.io.3.i2(?); Stela 3, 9.3.13.0.0; and Stela 6, 9.6, 7, or8.?.?.3. THE ORIGIN OF THE MAYA CIVILIZATION. 413 region (see plate i and fig. 64), where three centuries later they were to attain such cultural brilliance. Uaxactun is approximately 600 kilometers south of east from San Andres Tuxtla, which latter place, as we have already seen, is about half- way between the historic habitat of the Huasteca and the northern Peten region, the earliest known historic habitat of the Maya. Still another 160 years, seven generations later, and we find that some branch of the Maya had found its way 300 kilometers farther south; had reached the Copan Valley, and had established itself there with sufficient confidence and permanency to be able to execute and erect Stela 24 in 9.2.10.0.0, or possibly Stela 20, a katun earlier. On the basis of the provenance and dates of the earliest surely deci- phered contemporaneous inscriptions, therefore, it appears probable that Copan is not so old as Uaxactun by some 160 years; and even if we admit that the date of Stela 20, the earliest monument at Copan on stylistic grounds, is correctly deciphered as 9.1.10.0.0, this only cuts down the priority of Uaxactun by 20 years. But this important question as to which is the older city and which region the first occupied does not rest on the evidence supplied by these two monuments alone. Whereas Stela 20 is the only monument earlier than Stela 24 at Copan that we can possibly admit even on stylistic grounds, there are several other eai'ly monuments at Uaxactun and Tikal, besides those given on page 411. Thus, for example. Stela 5 at Uaxactun may have another Cycle 8 Initial Series, 8. 15. 10. 3. 12, within 20 years of that on Stela 9; and again, although the earliest surely deciphered date at Copan (Stela 24) is 3 years older (or if we accept the reading suggested by the writer for Stela 20, 23 years older) than the earliest surely deciphered date at Tikal (Stela 3), there is this important difference between these two great cities: At Copan we have nothing earlier than Stela 20, even on stylistic grounds, whereas at Tikal there are at least 7 other stelse the dates of which have not yet been deciphered, but which are fully as eai-ly as Stela 3 on stylistic grounds, and 4 of which are almost cer- tainly even earlier. Maler' describes 17 sculptured stelae at Tikal, and Tozzer'- enumerates 51 plain ones. Of the former only 4, Nos. 5, 11, 16, and 6 (?), belong to the Great Period, all the rest dating not only from the Early Period, but also from its earlier half.-^ Of the remaining 13 sculptured stelse, 2, Nos. 14 and 15, are so badly injured that Maler took no photographs of them, and of ' Maler, 191 1, pp. 62-91. ^Tozzer, 191 1, p. 102. ^Thus, for example, there are no dates at Tikal after 9.6.3.9.15 (Stela 17) until 9.14.0.0.0 (Stela i6), an Interval of 150 years. It is highly probahle, however, that the 51 plain stelae here were erected during this period, their inscriptions and designs being painted upon them instead of engraved. If we suppose them to have been the hotun-markers (and of the 17 sculptured stelae only 2, Nos. 11 and 16, record hotun-endings, 10.2.0.0.0 and 9.14.0.0.0 respectively), and if we suppose further that only the lahuntuns and katuns were marked up to 9.10.0.0.0, there would have been required between 9.3.0.0.0 (the next katun after Stela 3) and 10.2.0.0.0, the date of Stela 11, 63 stelae, and deducting 2, for Stelae II and 16, because they record two katun-endings within this period, we will have left 61 stelae, or only 10 mote than the number of plain stelae already discovered. It seems not unlikely, therefore, that the hotuns at Tikal may have been marked by painted, plain stelas instead of carved ones. 414 THE INSCRIPTIONS AT COPAN. the remaining ii, on stylistic grounds 4 are surely earlier than Stela 3,^ 3 as early,- and only 3 are later. ^ That is to say that whereas at Copan we have only i stela (Stela 20) which may be earlier than 9.2.10.0.0 (Stela 24), at Tikal we have at least 4 (Stelae 4, 7, 8, and 13), and possibly 3 more (Stelae i, 2, and g), which are earlier than 9.2.13.0.0 (Stela 3). This evidence at Tikal, coupled with that at Uaxactun, where we have one monument (Stela 9) if not two (Stela 5) recording dates actually 140 years earlier than the earliest possible contemporaneous date now known at Copan, proves on the chronologic side almost conclusively that Uaxactun- Tikal is considerably older than Copan. This is established not only by the actual priority of the Uaxactun dates, but also by the mass of the evidence, more earlier monuments being known at Tikal, although not exactly dated, than at Copan. Finally, the geographic location of these two great Maya centers is such as to make it extremely probable that Tikal is the older. Not only is Tikal nearer the center of the Old Empire, and Copan far out on the southeastern frontier (see plate i), but also the general trend of early Maya migration was from northwest to southeast, the earliest dated Maya object known being found 300 kilometers nearer Tikal than Copan. Summing up these several lines of evidence as to the origin of the Maya civilization, it appears as not improbable that the introduction of agricul- ture from the highlands of central Mexico to the Gulf Coast-plain may have been the primary factor in releasing the Maya from complete absorption in the continuous struggle for bare existence. Cultivation applied to this naturally rich region yielded a far more abundant return than in the arid highlands, and the exigencies of the agri- cultural year, the clearing, planting, and harvesting seasons, must soon have turned the minds of the Maya priesthood toward the accurate measure of time and the study of the seasonal year and of the sun and moon. After many generations of recorded observations on these bodies, cer- tain natural laws became deducible therefrom, and then some Mayan Hipparchus invented the calendar, possibly first a 260-day Sacred Year (the tonalamatl) built up on the permutation of 13 numbers and 20 names, per- haps next a 365-day solar year (the haab), composed of 18 periods of 20 days each and a closing period of 5 days, and still later a combination of the two, in which the 260 differently named days were fitted into the 365 posi- tions of the year, giving a new period (the Calendar Round) composed of 18,980 dates, 52 years of 365 days each or 73 years of 260 days each. Still later some one devised the remarkable Maya vigesimal numerical system, numeration by position, i. e., from bottom to top, and the ingenious Maya arithmetical notation of bars and dots, and probably later the head-variant numerals, and thus the calendar and chronology gradually took shape. Some time prior to these discoveries, however, the Maya would seem to have begun a general movement southeastward, in which the possibly ' Stelae 4, 7, 8, and 13. '' Stelae i, 2, and 9. ^gtels 10, 12, and 17 THE ORIGIN OF THE MAYA CIVILIZATION. 415 more backward elements, later to become the Huasteca, were lett behind in the region between the Panuco River and Tuxpam in the State of Vera Cruz. Moving southeastward, slowly we may feel sure, the Maya would appear to have been established in the region around San Andres Tuxtla in 8.6.2.4.17, by which time their chronology, hieroglyphic writing, notational system, and proficiency in stone-carving were sufticiently perfected to enable them to carve upon a hard, refractory material such as nephrite an inscription in their graphic system. Up to this point our history of the probable movements and activities of the Maya has been largely speculative, based principally upon the exist- ence of a Maya-speaking people of non-Mayan culture far to the northwest of the recognized Maya culture and linguistic areas, and upon the prove- nance and date of the Tuxtla Statuette, but from this time onward we enter upon firm historical ground. By 8. 14. 10.13. 15, some 165 years later, we find the Maya established at Uaxactun in northern Peten, 600 kilometers still farther south and east, and by 9.2.10.0.0 and probably by 9.1. 10. 0.0, 160 or 140 years later respectively, we find them in the Copan Valley and sufificiently at home there to be quarry- ing, carving, and erecting monuments and presumably in building a city. If the hypothesis advanced here is correct, namely, that Uaxactun and Tikal are much older than Copan, at least by a century, we need not look for the beginnings of the Maya hieroglyphic writing at Copan at all, or indeed for inscriptions much earlier than Stela 20, since under this hypothesis the Maya are assumed to have developed and perfected their chronological sys- tem and hieroglyphic writing, even to the point ot recording it upon stone monuments, long before they reached Copan, and the beginnings of it must be sought elsewhere, possibly in northern Peten, but more probably some- where on the Gulf Coast-plain of Mexico between the Panuco and Grijalva Rivers. HISTORY OF COPAN DURING THE OLD EMPIRE. THE EARLY PERIOD. As deciphered in Chapters II, III, and IV, probably 95 per cent, of the Copan dates are correct as given; certainly those which are recorded as Initial Series, Secondary Series, and Period Endings have no greater propor- tion of error than i out of 20; and of the remaining Calendar Round dates at least 75 per cent, are probably correctly deciphered. Moreover, if the last also happen to be hotun-endings as well, such as the contemporaneous dates of Altars Z, G3, Q, W, W, G2, and Gi, Temple 21a, and the Reviewing-stand in the Western Court for example, the percentage of accurately deciphered dates is even higher. Thus, in spite of effacement due to erosion and break- age, and even the loss of essential parts of the record on still missing fragments, it is probable that less than 5 per cent, of the readings suggested in the fore- going pages are incorrect. 4l6 THE INSCRIPTIONS AT COPAN. In Utilizing these chronologic data for the reconstruction of the his- torical background at Copan, it is necessary to bear in mind the fact that all dates recorded were not contem-poraneous, even at the times the monu- ments presenting them were severally erected, and that in order to analyze their significance properly, it is necessary to discriminate sharply between the surely contemporaneous dates and those which were either historical i-e., in the past, or prophetic, i. e., in the future, at the times they were recorded. If this distinction is not made, the considerable preponderance of late dates due to the extensive use of Secondary Series and Calendar Round dating in texts of the Great Period will unduly emphasize the importance of that period, and mislead especially as to the number of texts emanating there- from, as compared with those from the Middle and Early Periods. It has been shown in the preceding section that the branch of the Maya which founded and built Copan probably reached there with a full knowledge and ample experience in the use of their peculiar chronologic and graphic systems; that is to say, they probably did not develop either in the region of Copan. Indeed, the evidence presented in the preceding section tends to establish that they had been engraving their records on large stone monu- ments (Stela 9 at Uaxactun) for at least 156 years before the earliest certain date at Copan (Stela 24), and on smaller stone objects (the Tuxtla Statuette) for at least 321 years before; in short, it would appear that the Maya reached the Copan Valley at a high level of cultural attainment and set about the intensive occupation of the region with little loss of time. And the next question is, when and where did this first occupation take place? It has been shown in Chapter II that, on stylistic grounds. Stela 20 is pretty surely the oldest monument at Copan, and the best reading of its fragmentary date, 9.1. 10. 0.0, agrees with this; first, because of this fact, i. ^., there being no earlier monuments known at the site, and second, because by this date the Maya had already been carving their inscriptions on stone for such a long time (more than three centuries), that the carving of stelae could have offered no real difficulty to them even in a newly occupied region; the writer is disinclined to push back the date of their arrival in the Copan Valley much before 9.0.0.0.0. Indeed, if Stela 20 were the first hieroglyphic monument to be carved at Copan, and certainly no other of earlier style has yet been found there, it is even possible that they may have arrived shortly after 9.0.0.0.0. In round numbers, we may probably say that the Maya reached this region not much before the close of Cycle 8, and that by the beginning of Cycle 9 its occupation was under way, and finally that as early as 9.1. 10. 0.0, less than 30 years later, the capital, or principal settlement in the valley, had been located at Group 9, which answers the question propounded above as to when and where this occupation first took place. It may be objected that this reading for the date of Stela 20 is doubtful; but even if this were true, the date of Stela 24, which is surely deciphered, is only 20 years later. MORLEY PLATE 33 o C < C 'Si o o 3 -a c o o c '5 c4 HISTORY OF COPAN DURING THE OLD EMPIRE. 417 Or again, it may be held that earher monuments will yet be found at Copan which will carry back the occupation of the site much earlier. While this is of course possible, it is nevertheless significant that such earlier monu- ments have not come to light as yet, in spite of the fact that Group 9, appar- ently the oldest settlement in the valley on the basis of the dated remains, has been more thoroughly excavated than any other place in the valley, the Main Structure not excepted, owing to the fact that the modern village was built on its site. On the contrary, such monuments as have been found under conditions indicating secondary usage are all not only of later style than Stela 20, but also of later date. Or finally, it may be contended that the earlier monuments at Copan were made of wood and have consequently disappeared; but against this objection may be adduced the fact that there is at least one stone stela known, at Uaxactun, which is surely 137 years older than the date suggested for Stela 20, and there are a half a dozen others at Tikal probably from 20 to 50 years older than Stela 20. In a word, stone stelae had been known for five or six generations at least before the date suggested here for Stela 20. In view of all the foregoing evidence, (i) the lack of a single text at Copan, which, even on stylistic grounds, can possibly be older than Stela 20 in spite of all the archaeological work, especially excavation, that has been done there during the past 35 years, (2) the presence of inscribed stone stelae elsewhere of every much earlier date, and (3) the dates of Stelae 20 and 24 themselves, the writer believes it probable that the Maya reached the Copan Valley with a full knowledge and understanding of their chronology and hieroglyphic writing and with an ample previous experience in the art of stone-carving, about the beginning of Cycle 9, and probably settled first at Group 9, Stela 20 possibly being the first monument erected there. That Group 9 was the first settlement in the valley rests on strong archaeological evidence, as we have already seen in Chapter II: 1. The two earliest surely dated monuments. Stelae 24 and 15, are both found here, as well as the one which we have seen is probably the earliest of all, Stela 20. 2. More early monuments are found here than at all the other groups in the valley combined, including the Main Structure. Of the 22 monuments of the Early Period under observation, not including Fragments V and Fragment S', 12, or more than half, were found at this group. 3. Of the 10 monuments outside of this group, practically every one occurs in positions indicating secondary -usage, i. e., in positions for which they were not originally designed. 4. Of the 15 Early Period fragments now known at Copan, all but one, Frag- ment S', were found here. Nor could the ancient Maya have chosen a better site in the whole valley for their first settlement. At this point a bench of the foot-hills extends out into the valley from the north side. Nearby, along its eastern base, flows the Rio Sesesmil, offering at all times an abundant supply of potable water fresh from the mountains; and below, to the east, south, and 4l8 THE INSCRIPTIONS AT COPAN. west, for a kilometer or more, stretches the valley-bottom, a rich alluvial plain, subjected to annual overflow in its lowest parts bordering on the Co- pan River, and capable, under cultivation, of supporting a large population. In later times, so well chosen was this spot and so admirably adapted for a small settlement, that the modern village was also located in the same place, the plaza of the one being roughly coincident with the plaza of the other, and unfortunately, the building material of the one furnishing the building material for the other. It appears probable that the religious center of this first settlement was the mound of Stela 7, since of the twelve early monuments found at Group 9, eight were recovered here. Altars P' and Q' and Stelae 20, 24, 25, 15, 18, and 7 (fig. 22, d', b', V, q, e', s, x, and respectively) ; and of the remaining four, two, Stelae 21 and 22, show unmistakable signs of secondary usage in ancient times, the former having been built into the hearting of the high mound at the southeastern corner of the plaza (fig. 22, k) and the latter being found in one of the stone stairways around the small court at the southwestern corner of Group 9 (fig. 22, w'), and the remaining two. Altars L' and M', must have come from nearby. (See fig. 22, g and i, respectively.) Possibly the first monument to be erected in the original settlement on the banks of the Rio Sesesmil was Stela 20 in 9.1. 10. 0.0 (fig. 22, v), and from this time on for the next 160 years. Group 9 grew steadily, increasing in size until its religious and civic center occupied practically the whole of this little bench jutting out into the valley. It has been pointed out that practically all the monuments of the Early Period which have been found outside of Group 9 occur in positions strongly indicating secondary usage, as follows: Altars J' and K', in the foundations of Stela 10 (9. 10.19. 13.0) at Group 12. Altar X, in the foundations of Stela 5 (9. 13. 15.0.0 or 9.14.0.0.0) at Group 8. Altar Y, in the foundations of Stela 4 (9. 17. 12. 13.0) at the Main Structure. Altar A', in the foundations of the Hieroglyphic Stairway (9.16.5.0.0) at the Main Structure. Stela 16, found broken at the Main Structure, possibly not reused. Stela 17, reused as a building-block in the Great Plaza at the Main Structure after 9.12.5.0.0. Stela E, re-erected on the terrace west of the Great Plaza at the Main Structure after 9.12.5.0.0. Stela P, re-erected in theWestern Court at thcMain Structure about 9.17.0.0.0. Fragment S', reused as a building-block in the facing of Mound 9 (9. 11. 15. 0.0) at the Main Structure. That is, with the possible exception of Stela 16, they all occur in posi- tions for which they were not originally designed, and hence they may have been, and doubtless were, carried to these later positions from some earlier group or groups. Finally, since more than half of the early monuments have been found at Group 9 (nearly two-thirds, if we include Altar X and Stela 9 at Groups 8 and 10 respectively, each within a kilometer of Gioup 9), it HISTORY OF COPAN DURING THE OLD EMPIRE. 419 appears as not improbable that all the outlying monuments of the Early- Period originally had been erected at Group 9, which we may perhaps appropriately call Old Copan, being in all likelihood the first settlement in the valley. The dates and provenance of Stelae 7, E, and P, moreover, greatly strengthen this hypothesis. It has been shown: 1. That Stela 7 is the only monument of the Early Period which was practically in situ (i.e., fallen directly above its original foundations) when found. 2. That Stelae E and P at the Main Structure are surely not zn situ where now found, since they occur in places which were not built until long after their con- temporaneous dates. 3. That Stelae 7, E, and P commemorated three successive hotun-endings in the Long Count, 9.9.0.0.0, 9.9.5.0.0, and 9.9.10.0.0 respectively. It seems reasonable to infer from these facts that Stelae E and P, which marked the next two hotuns after 9.9.0.0.0, the date of Stela 7, had originally been erected at Group 9, probably near Stela 7, and that some time later, after the Great Plaza was built (i. e., after 9.12.5.0.0), Stela E and its altar were removed from Group 9 and carried to the Main Structure 2 kilometers east and re-erected there on the terrace on the west side of the Great Plaza, and further, that some time after the Western Court had been completed (about 9. 1 7.0.0.0), Stela Pwas similarly removed from Group 9 and re-erected in the Western Court. Stela P is the last monument of the Early Period, and between it and Stelae 12, 2, 10, 19, 23, 13, and 3, the first stelae of the Middle Period, there is a lacuna in the sequence of the monuments of 25 or 30 years. Moreover, as will appear presently, not one of these seven earliest stelae of the Middle Period is at Group 9, and only two of them are at the Main Structure, the rest being scattered from Santa Rita (Group i), 14 kilometers east of Group 9, to Hacienda Grande (Group 13), 3.5 kilometers west of Group 9. It appears probable, therefore, that at the end of the Early Period (after 9.9. 10. 0.0) a tremendous expansion took place, in the course of which the whole valley was intensively occupied for the first time, a number of smaller groups, Nos. i, 2, 3, 12, and 13 being established (see plate 3); and coincident with this centrifugal movement. Group 9 began to decline in importance, no monuments at all being erected there during the Middle Period, and only five during the Great Period. Assuming, then, that all the monuments of the Early Period, wherever found, were originally erected at Group 9, let us next examine the dates of the 13 stelae now known from the period: Stela 20 Stela 16 Stela 24 Stela 7 Stela 21 Stela 17 Stela 15 Stela E Stela 22 Stela 18 Stela 9 Stela P Stela 25 Of these, the 7 in the first two columns are only doubtfully deciphered, although Stelae 20 and 25 are probably correct as given, and Stelae 17 and 18 possibly so. The six in the last two columns are surely deciphered. 420 THE INSCRIPTIONS AT COPAN. Beginning with the surely deciphered group, it has been shown in Chapter II that every one records a hotun-ending,^ and that all but one, Stela E, a second or fourth hotun-ending, that is, a lahuntun or a katun- ending. Moreover, of the remaining 7, at least 4 (Stelae 20, 25, 17, and 18), also probably recorded katun or lahuntun-endings. This important practice of erecting stelae at the expiration of second and fourth hotuns, i. e., lahuntun and katun-endings respectively, is encoun tered fox the first time here at Copan in 9.1.10.0.0 or 9.2.10.0.0, and at the expiration of first and third hotuns in 9.9.5.0.0, although the latter custom appears 10 years earlier in 9.8.15.0.0 at Piedras Negras (Stela 25), and by the Middle Period is found everywhere in the Old Empire. Because of this fact, and also because all of the surely dated early monu- ments elsewhere (see p. 411), the Tuxtla Statuette, the Leyden Plate, Stelae 9 and 3 at Uaxactun, and Stelae 3, 10, and 17 at Tikal, were not erected on hotun, lahuntun, or katun-endings, it appears reasonable to infer that this important custom, which in one form or another survived until the close of the New Empire in 1541 in northern Yucatan, a period of more than 1,200 years, may have originated at Copan as early as 9.1. 10. 0.0 and spread elsewhere from here.- If the 13 stelae of the Early Period at Copan, all record lahuntun or katun-endings except Stela E, which records a third hotun-ending, let us ascertain how many lahuntuns and katuns there were between the dates of the earliest and latest of these monuments. Between 9.1. 10. 0.0 (Stela 20) and 9.9.10.0.0 (Stela P) inclusive, there are 17 lahuntuns and katuns, for which we have only 12 stelae, exclusive of Stela E; that is, apparently 5 or 6 are still missing. An attempt has been made in the following list to assign 10 of these 12 stelae to these 17 period- endings, the monuments marked (?) probably being correctly deciphered as given, those marked ( ? ?) being very doubtful. Stela 25 probably records the same lahuntun-ending as Stela 24; and it is impossible to say exactly where Stela 16 belongs, probably somewhere between 9.4.10.0.0 and 9.7.10.0.0. Stela 20, 9.1.10.0.0 (?) Stela 17, 9.6. 0.0.0 (?)■' 9.2. 0.0.0 Stela 9, 9.6.10.0.0 Stela 24, 9.2.10.0.0 Stela 18, 9.7. 0.0.0 (?) 9.3. 0.0.0 Stela 21, 9.7.10.0.0 (??) Stela 22, 9.3.10.0.0 (??) 9.8. 0.0.0 9.4. 0.0.0 9.8.10.0.0 Stela 15, 9.4.10.0.0 Stela 7, 9.9. 0.0.0 9.5. 0.0.0 Stela P, 9.9.10.0.0 9.5.10.0.0 ' The altar of Stela E, which actually records the current hotun-ending, has been regarded as textually one with Stela E in this study. (See pp. 112, 128). 2 If the readings suggested on page 392, note i, for Stels 8 and 9 at Tikal, 9.0.10.0.0 and 9.2.0.0.0 respectively, should be correct, these monuments would be the earliest hotun-markers now known, and they would nullify the above conclusion, making Tikal the first Maya city to have inaugurated this important custom instead of Copan. As already noted, however, these two readings are so doubtful that they have been disregarded in the conclusions set forth in this chapter. ' Under the postulate that the Early Period stela: prior to 9.9.5.0.0 record either lahuntun or katun-endings, Stela 17 can only be either 9.6.0.0.0 or 9.6.10.0.0, since its katun coefficient is surely 6 and the former is chosen here because the latter date is recorded on Stela g. HISTORY OF COPAN DURING THE OLD EMPIRE. 42I Even if the doubtfully deciphered stelae, 22, 17, 18, and 21, are correctly- assigned above, and allowing Stela 16 for one of the unfilled period-endings, there are still six lacunae in the sequence of the early monuments. Since the first 4 early monuments concerning which there is little or no doubt all record lahuntun-endings, i.e., Stelae 20, 24, 15, and 9, perhaps at first, and up to 9.6.0.0.0, stelae were erected only on lahuntun-endings, and the katun- endings were permitted to pass by without being thus marked. Such an explanation is very unsatisfactory, as it presupposes the half-katun periods were of more importance than the katuns themselves, which would have been an unusual belief, to say the least. On the other hand, the 3 earliest surely deciphered stelae all record lahuntun-endings, and if this were the case, and assuming Stela 22 may be referred to the lahuntun-ending 9.3.10.0.0, there would be only one break in the monumental sequence prioi to 9.6.0.0.0, namely, 9.5.10.0.0, to which we may possibly assign Stela 16. If the katun-endings were also marked from 9.6.0.0.0 on, we have two lacunae in the latter half of the Early Period, namely, at 9.8.0.0.0 and 9.8.10.0.0, and even if Stela 21 be assigned to either of these dates, it still leaves two lacunae, as in that case there is no monument for 9.7.10.0.0. These results are not altogether satisfactory. Some of the readings sug- gested are very uncertain, and the assumption that lahuntun-endings were of greater importance than katun-endings, in spite of the fact that the remains apparently indicate such a fact, is even more doubtful. However, the following facts may be accepted as established : 1. Surely as early as 9.2.10.0.0 (Stela 24), and probably as early as 9.1.10.0.0 (Stela 20), the practice of marking the expiration of the successive lahuntuns and katuns was inaugurated at Copan. 2. Of the 6 surely dated monuments of the Early Period at Copan, 5 record lahuntun and katun-endings and the sixth a third hotun-ending. 3. On the basis of this periodicity for the erection of the stelae, i. e., no first and third hotuns marked by stelae prior to 9.9.5.0.0, there are only 6 out of the 17 possible period-endings, between 9.1.10.0.0 and 9.9.10.0.0 inclusive at Copan, for which no corresponding stelae have been found. 4. Finally, this practice is found to have prevailed so much earlier at Copan than anywhere else, the first occurrence elsewhere being at least 125 years later, if we exclude the doubtfully deciphered Stelae 8 and 9 at Tikal, that there are e.xcellent grounds for believing it may have originated at Copan, and from there spread elsewhere in the Old Empire and later carried over into the New Empire, having been the determining factor in the erection of stelae for more than 1,200 years. One more point in connection with the Early Period at Copan demands our attention. In Chapter III, it will be remembered, Spinden expresses the belief that the two primitive human figures found under Altars X and Y in the foundations of Stelae 5 and 4 respectively are the earliest sculptures at Copan, but since neither has any glyphs carved upon it, exact dating is impossible. This lack of exact chronological data about either is particularly unfortunate in view of the fact that both indubitably resemble similar 422 THE INSCRIPTIONS AT COPAN. sculptures of the Archaic type found in the highlands of Salvador and Guatemala, south and west of Copan respectively, notably a stone figure at the Hacienda of Miraflor, just outside of Guatemala City. (See fig. 67, a, b, and c, and pp. 208, 209.) Thus, had it been possible to date either of these two sculptures from Copan in the Maya chronological system, there would a b c Fig. 67. — Anthropomorphic figures of the Archaic type: a, Copan, foundations of Stela 5; b, Copan, foundations of Stela 4; c. Hacienda Miraflor near (niatemala City. have been established at least one definite point of contact between the Old Empire and the Archaic culture, which seems to have had a far vaster exten- sion, and doubtless a much earlier origin than the Maya civilization as pointed out in the preceding section. Possibly these two figures may have been taken from some of the earlier inhabitants of the region, some Archaic people living south and west ot Copan, and were placed in the foundations of Stelae 4 and 5 as objects of unusual importance and sanctity. Or again, they may be very early Maya copies of still earlier Archaic sculptures found by the Maya when they first reached the region.^ Or again, they may date from a period at Copan before the hieroglyphic writing had been transfferred to stone, a view the writer does not share, however, since he believes the Maya had been carving their inscriptions on stone for several centuries before they reached Copan. In any case, they do not controvert any of the conclusions reached above, and they may probably be referred to the same general period as Stela 20. Sum.ming up the history of Copan during the Early Period, it appears probable that the branch of the Maya who colonized this region reached the Copan Valley shortly before, or not later than, the beginning of Cycle 9. In this connection the provenance and date of the Leyden Plate should be borne in mind. This small nephrite plate, as already noted on page 411, was found near the mouth of the Motagua River, some 130 kilometers northeast of Copan, and bears the very early date 8. 14.3. 1. 12, which is about 145 years earlier than the reading here suggested for Stela 20. Possibly this object may have been left behind at some intermediate stopping-place of the tribe on their migration southward from northern Peten (see plate i and fig. 64), and it doubtless antedates the first settlement in the Copan Valley. ' Lothrop found a similar sculpture, although with a second figure on the back of the first, the whole being very crudely executed in block-like outlines similar to the sculptures under discussion, at La Florida, 50 kilometers northeast of Copan, in 1916. HISTORY OF COPAN DURING THE OLD EMPIRE. 423 This first settlement appears to have been located at Group 9, and here for the next 200 years all the monuments were probably erected, and here may be said to have been the capital, the religious and administrative center of the region. Judging from the provenance of the early monuments, it appears as not improbable that Group 9 was the only settlement in the valley during the Early Period, at least the only one which attained sufficient wealth and importance to have been able to execute and erect monuments. During this period the valley was doubtless put under cultivation from end to end, and reserves of wealth and experience accumulated which were to be utilized in the great expansion that took place at the beginning of the next period. Each lahuntun and katun as it passed was probably marked by the erection of a corresponding stela at Group 9, and monument by monument we may see reflected the increasing prosperity of the tribe. Perhaps as early as 9.7.0.0.0 the priests attempted to portray the human figure on the fronts of their stelse (Stela 18), and before the end of the period, i. e., after 9.9.0.0.0, the all-glyphic stela began to pass out of fashion. We must believe that the rulers of the tribe during this period, whether hereditary or elective, civil, military, or ecclesiastic in character, were wise administrators, who occupied themselves in developing the resources of the surrounding region, in building and embellishing their capital, and in gen- erally extending their influence and the cult of their tribal deities. Finally, by the end of the period, the tribe had become so powerful and wealthy that it was able to expand the sphere of its activities beyond Group 9 and to establish other important settlements throughout the valley. THE MIDDLE PERIOD. The Middle Period at Copan opens with a hiatus in the sequence of the monuments, followed by a tremendous outburst of sculptural activity all over the valley in 9.1 1. 0.0.0, on which latter date no less than 7 dift'erent stelae were erected at 6 different groups. After Stela P in 9.9.10.0.0, a period of decentralization seems to have set in, during which the intensive occupation of the whole valley seems to have taken place, and 30 years later, almost as if by common consent, the current katun-ending, 9.1 1. 0.0.0, was commemorated by the erection of stelae at 6 difl'erent groups: Group i at the eastern end of the valley, 14 kilometers distant from Old Copan; Group 2 on the north bank of the Copan River, 8.5 kilometers distant; Group 3 on the summit of a hill on the eastern side of the valley, 4.5 kilometers distant; the group which was later to become the Main Structure, where two stelae bearing this date were erected, 2 kilo- meters distant; Group 12 on the summit of a hill at the western end of the vaUey, 2.5 kilometers west of Old Copan, and Group 13 in a little side-valley entering the main valley from the north, 3.5 kilometers west of Old Copan. Groups 7, 1 1, 14, and 15 may have been founded at this same time, although no dated monuments have been found at any of them. (See plate 3.) 424 THE INSCRIPTIONS AT COPAN. But this centrifugal force, which must have threatened to disrupt the tribe, and which certainly cost Group 9 its position of preeminence in the valley, appears to have been of short duration. Of the six different groups where the katun-ending 9. 11. 0.0.0 was commemorated by the erection of a stela, only one has any other inscribed monuments, and that is the group which later became the Main Structure. In attempting to explain this condition, one point at least appears reasonably certain, namely, that these groups, with the exception of the Main Structure, did not continue to hold important positions in the valley, which is indicated by the fact that, with the single exception noted, no sub- sequent monuments were erected at any one of them. Perhaps the best explanation of the facts observed is that after this decentralizing wave had reached its crest in 9. 11. 0.0.0 there was a reaction, during which the former tendency toward centralization developed during the 200 years Group 9 had been the only settlement of importance in the valley, reasserted itself, and as a result the Main Structure became the principal settlement in the valley and in the following century the next to largest city in the Old Empire, being second in size only to Tikal. The next question confronting us is, when was the group at the Main Structure founded ? And in seeking the answer to it, we must again turn to the evidence afforded by the dated monuments. The earliest monument at the Main Structure, which is not obviously in a position indicating secondary usage, is Stela E; but, as we have already seen, neithei Stela E noi Stela P can be regarded as in situ where now found, strictly speaking, both probably having been removed to the Main Struc- ture from Group 9 long subsequent to their original erection at the latter site. The next earliest monuments are Stelae 2 and 3 in the Middle Court, which may or may not be in their original positions, and the earliest monu- ment at the Main Structure surely in situ is Stela i, also in the Middle Court. Let us next examine the dates and provenance of these three stelae. Stelse 2 and 3 both date from 9. 11. 0.0.0, as already noted, and Stela i from 9. II. 1 5.0.0, the provenance of all three, as just stated, being the same, i. e., the Middle Court. The first is very doubtfully in its original position; the second probably so, because a cache of objects, though not in a stone- lined chamber, was found under its foundation-stone; and the third surely so, the chamber under its foundation-stone having been built at the same time as Mound 9. It was suggested in Chapter III that Stelae 10 and 12 at Groups 12 and 3 respectively, both of which were erected at approximately the same time as Stels 2 and 3,^ may have had something to do in determining the location of a settlement at the Main Structure, possibly by defining some particular line of sight, which now cuts across the southern slope of the Acropolis. We have seen how, under the influence of a strong decentralizing move- ment, possibly due to the normal expansion and growth of a people who had 'The only date on Stela lo is only lOO days earlier than 9.11.0.0.0, i.e., 9.10.19. 13.0. HISTORY OF COPAN DURING THE OLD EMPIRE. 425 been living and prospering for 200 years in one place, the whole valley was occupied and a number of smaller settlements established in 9.1 1. 0.0.0, among others possibly the Main Structure, since the earliest monuments possibly m situ there present this date. Certainly 15 years later a settle- ment had been made here, a temple built (Mound 9), and a stela erected (Stela i). The fact that the three earliest stelae at the Main Structure which can possibly be regarded as m situ all occur within 75 meters of each other and in the same plaza, suggests that this is the earliest part of the Main Struc- ture, and the fact that one of them was built in the foundations of Mound 9 suggests that Mound 9 probably is the oldest building now extant, in its original form at least, at the Main Structure. The temple on its summit was excavated by Gordon in 1895 and was found to be devoid of sculptural decoration, in which respect it is unlike most of the other temples of the Acropolis group at the Main Structure, Nos. II, 2ia, 22, and 18 for example, and therefore presumably earlier. The chief objection to the hypothesis that Mound 9 is the oldest build- ing now extant at the Main Structure, and that Stelae 2, 3, and i were the first stelae erected there, is the possibility that the Acropolis could not have been built in the 120 years between 9.1 1. 0.0.0 and 9.17.0.0.0, when it is known to have been completed. (See the dates of Temple 2i<3 and the Reviewing-stand in the Western Court, pp. 318, 321.) The cross-section of the Acropolis exposed by the river has a maximum height of 40 meters and shows five or six earlier plaza-levels, mostly in the lower half, and below the present floor-level of the Eastern Court. It has been argued^ that to have built this vast artificial construction a long period of time was necessary, several centuries at the very least, but after a study of all the evidence available the writer has not reached this conclu- sion; on the contrary, he believes that the Acropolis could have been built, and in fact probably was built, in the 120 years between 9.1 1. 0.0.0 and 9.17.0.0.0, when the Eastern and Western Courts were completed. In sup- port of this view it may be pointed out that the construction of Mound 26, which is an integral part of the Acropolis, was apparently not started until 9. 13. 18. 17.9, the date on the lowest step of the Hieroglyphic Stairway, that is, something less than 50 years before its completion, less than 60 years before the completion of Temple 11 nearby, and less than 75 years before the completion of the whole Acropolis. It has been shown that the Great Plaza was not laid out until after 9.12.5.0.0 (Stela I) and was probably completed by 9.13.10.0.0 (Stela J), 25 years later; and, judging from the dates on the Acropolis itself, the com- pletion of successive units of that construction is to be measured by decades rather than centuries. Assuming for the moment that no temples stood on the river-plain here until after 9.1 1. 0.0.0, we have Temple 9 being dedicated 15 years later ' Gordon, 1896, p. 10. 426 THE INSCRIPTIONS AT COPAN. (9.1 1. 1 5. 0.0) and the foundations of Mound 26 being laid 42 years later (9. 13. 18. 17.9). Doubtless we must also assume that during these 57 years other temples had been built east of the site of Mound 26, corresponding to the lower plaza-levels in the cross-section of the Acropolis exposed by the river. Mound 26, the Hieroglyphic Stairway on its western side, Temple 26 on its summit, and Stela M at its western base, an undertaking involving a great outlay of labor, had been completed and were dedicated in 9.16.5.0.0, 45 years later. Meanwhile, work on the substructure on the south side of the Court of the Hieroglyphic Stairway was going forward, and 5 years later, in 9. 1 6.10.0.0, Stela N was erected, and less than 3 years later, in 9. 16. 12. 5. 17, the very important Temple 1 1 was dedicated and another large section of the Acropolis completed. Although we have no dates to guide us here, we may doubtless assume that toward the close of the Middle Period, i.e., during the course of con- struction of Mound 26, the part of the Acropolis lying to its south was also changing shape, gradually being built higher, and reaching the next higher plaza-levels in the cross-section exposed by the river on the east side. During the 8 years between 9. 16. 12. 5. 17 and 9.17.0.0.0, the Acropolis received its final additions. The whole southwestern corner marking the Western Court was probably filled in and dedicated in 9.17.0.0.0, as indi- cated by the date on the Reviewing-stand at the northern end of this court. At the same time the whole eastern side of the Acropolis was also raised to its final levels. Temples 21 and 22 being dedicated some time between 9. 16. 12. 5. 17 and 9.17.0.0.0, and Temple 21a on the latter date. Judging from the rapidity with which these several parts of the Acrop- olis seem to have been successively completed, as indicated by the dates actually recorded upon integral members of this architectural complex, the writer believes it is probably safe to assume that the Acropolis was not commenced much, if any earlier than 9.1 1. 0.0.0 and possibly not before 9. 13. 18. 17. 9, and was completed in 9.17.0.0.0, that is to say, it was probably something between 60 and 120 years in building. Returning now to the general history of Copan during the Middle Period, we may probably assume that the Main Structure was founded in 9.1 1. 0.0.0, or at least became the most important settlement in the valley from that date on, soon eclipsing Group 9, which had hitherto been the capital of the region. Moreover, of the other five groups founded at the same time, not one has another monument of any date, and it seems probable that, at least in so far as these particular groups were concerned, all the architectural and sculptural efforts of the tribe were centered on the construction and embellishment of buildings at the Main Structure. This appears clearly in the second column of Appendix IX, where, after 9. 1 1 .0.0.0, it will be seen that the only settlement in the valley outside of the Main Structure at which monuments were erected during the Middle Period was Group 8, midway between the Main Struc- HISTORY OF COPAN DURING THE OLD EMPIRE. 427 ture and Group 9. In short, the extensive occupation of the valley had been effected, and now for the next century and a half the principal efforts of the tribe were devoted to beautifying and enlarging their new capital. Some time during the first 15 years, work on Temple (Mound) 9 was commenced, and in 9.1 1. 15. 0.0 both this temple and Stela i were dedicated, the former facing south, where, during the next century, the Acropolis was slowly to take shape. (See plate 6.) The next hotun-ending, 9.12.0.0.0, also a katun-ending as well, was commemorated by the erection of two round altars, both presenting Initial Series, being the first examples of altars which may have been used inde- pendently of stelae yet encountered, one associated with Stela i and Temple 9, and the other at a new settlement, Group 8, which seems to have been founded about this time. The next hotun-ending, 9.12.5.0.0, was marked by the erection of Stela I, an extremely important monument, since its location probably determines the date of construction of the Great Plaza. Stela I stands in a niche on the terrace on the east side of the Great Plaza in such a way as to indicate that it was erected before this terrace was built, that is to say, this terrace could not have been built until after 9.12.5.0.0 (see plate 6). Probably shortly after this date the Great Plaza was laid out in its present form, and possibly completed or at least well under way before the erection of Stela J, 25 years later. The next hotun-ending was commemorated by the erection of Stela 6 at Group 8, and then, with 9. 12. 15. 0.0 missing in the monumental sequence, the next, 9.13.0.0.0, also a katun-ending as well, was commemorated by the dedication of two rectangular table-like altars, H' and 1'. These are now found in the Western Court, but the writer supposes them to have been brought hither from some other part of the Main Structure, assuming the Western Court not to have been built until 80 years later. The next hotun is not represented by any known monument, but the next, 9. 1 3. 10. 0.0, is recorded on Stela J at the southeast corner of Mound 3, one of the complex of terraces surrounding the Great Plaza. (See plate 6.) The last monument of the Middle Period is Stela 5 at Group 8, dating either from 9. 13. 15. 0.0 or 9.14.0.0.0; and then the closing 20 years of the period, 9.14.0.0.0 to 9.15.0.0.0, are represented by no monuments, being a complete blank, so far as the inscriptions are concerned. We have seen, however, that this was an epoch of great building activ- ity at Copan. The construction of Mound 26, and probably of Mound 1 1 and the whole eastern side of the Acropolis, had just been started, and was absorbing the energies of all the artisan class. At the beginning of the Great Period, i. e., in 9.15.0.0.0 and 9.15.5.0.0, as will appear presently, time was taken to erect three stelae and an altar, but after 9.15.5.0.0 there is another 20-year hiatus in the monumental sequence, unless we except the 428 THE INSCRIPTIONS AT COPAN. doubtfully dated and unimportant inscribed steps on the southern side of Mound 2. Work was doubtless being pushed forward on the whole Acropolis complex from 9.14.0.0.0 to 9.16.5.0.0, at which latter date we have seen Mound 26, Temple 26, the Hieroglyphic Stairway, and Stela M were all completed and dedicated, and during the next 15 years the series of mag- nificent temples surrounding the Eastern and Western Courts, 11, 16, 18, 19, 21, 21(2, and 22, were erected, which amply accounts for the almost complete absence of inscriptions dating from these 45 years, elsewhere so productive. Another factor which may have tended to cut down the output of inscriptions at Copan during the closing decade of the Middle Period is the possible absorption of her resources in the colonization of the neighboring city of Quirigua, probably by emigrants from Copan, in 9. 14. 13. 4. 17, and additions thereto in 9. 15. 6. 14.6, the latter date being actually recorded at both sites. All lines of evidence point to this fact. In the first place, Quirigua is not more than 60 kilometers north of Copan in an air-line, but is several hundred kilometers south of the nearest large Maya city to the north. Again, the art and architecture of the two cities are practically identical; indeed, the art of Copan and Quirigua shows closer relationship in technique as well as in subject-matter than does that of any other two cities in the Old Empire. Finally, the dates at Quirigua indicate that it was founded in 9. 14. 13. 4. 17, a fact substantiated in a general way by the art there, which shows no archaistic features whatsoever, but even on the earliest monuments is already perfected and in full flower. The writer believes the earliest monument at Quirigua is Altar M, dating either from 9.15.0.0.0 or 9.15.3.2.0, some 7 or 10 years after the city was founded. Even assuming that the colonists came fully equipped from Copan, it would probably have taken them at least that time to have felled the forest, put the land under cultivation, laid out their city, located the quarries, taken out and transported the stone, and finally to have carved the first monument. Indeed, the first hotun-ending commemorated by the erection of a stela was 9. 15. 15.0.0, 15 years later, although after this latter date not a single hotun-ending is omitted in the monumental sequence for the next 65 years. It appears as not unlikely that the probable foundation of Quirigua by colonists from Copan in or about 9. 14. 13. 4. 17 withdrew from the mother-city a number of her skilled artisans, especially stone-workers, masons, and sculptors, and this, coupled with the fact that the work on the Acropolis was also drawing heavily on the resources of the tribe, doubtless explains the absence of inscriptions from the closing katun of the Middle Period, and with few exceptions from the opening katun of the Great Period. The century from 9.10.0.0.0 to 9.15.0.0.0 was an important one for this southern branch of the Maya. During the previous period the tribe had grown beyond the capacity of its original capital at Group 9, and during the HISTORY OF COPAN DURING THE OLD EMPIRE. 429 first katun of the Middle Period established settlements throughout the valley. After 9.11.0.0.0 the capital was shifted from Group 9 to the Main Structure, i. e., nearer the middle of the valley, and an extensive building program inaugurated there. Finally, toward the close of the period, the neighboring city of Quirigua was founded, probably by colonists from Copan, at which time the provincial phase of the tribe's history may be said to have come to an end, and from this time on for the next 70 years, i.e., the first half of the Great Period, Copan was the most brilliant city architecturally and sculpturally, if not indeed the most powerful, in the Old Empire. THE GREAT PERIOD. The Great Period at Copan opened with the erection of two handsome stelae at the Main Structure and an altar at Group 9, the last being the first monument to be set up at the former capital for more than a century, and the next hotun-ending, 9.15.5.0.0, was commemorated by the erection of another imposing stela in the Great Plaza, especially interesting as being the earliest example known of the exclusive use of full-figure glyphs. We have already seen how, before the erection of these several monu- ments, the scene of sculptural and architectural activity had shifted to the Acropolis, and after the erection of Stela D for the next 20 years work went steadily forward on different parts of that great artificial substructure. In 9.16.5.0.0, as we have already seen, a large unit of this construction was completed — Mound 26, Temple 26, Stela M, and the Hieroglyphic Stairway — the last being by far the largest text in the Corpus Inscriptionum Mayarum. A hotun later, Stela N was dedicated, the last stela to be used for such a purpose at Copan, and less than two and a half years later the most important date in the Great Period at Copan, 9. 16. 12. 4. 17, was recorded as the dedicatory date of Temple 1 1 and Altars V, R, and U. Unfortunately, the nature of the event which rendered this particular date of such importance in the annals of Copan is unknown. It is different from the great majority of the dates heretofore encountered, being at the end of no particular division in the Maya chronological system, such as a tun, hotun, lahuntun, or katun, and for that reason probably is to be inter- preted as referring to the occurrence of some actual historical event or astro- nomical phenomenon, although which we are unable to say. On the evidence afforded by the glyphs thus far deciphered in the inscriptions, which are practically limited to the characters used in recording Initial, Secondary, and Supplementary Series, Period Ending, and Calendar Round dates, we would be justified in assuming that it was some important astronomical phenomenon which had taken place in 9. 16. 12. 5. 17; but, judged by the evidence as a whole, and particularly the three factors enum- erated at the top of the next page, it seems necessary to admit that this event may possibly have been of an historical nature, such as the death or acces- sion of a ruler, a notable conquest, or even the dedication of an important building devoted to religious uses, like Temple 11 where it was recorded. 430 THE INSCRIPTIONS AT COPAN. 1. Manj^ of the Aztec codices, which were doubtless patterned after Maya models originally, are wholly historical in character. 2. The Books of Chilan Balam, actual Maya redactions in the Spanish script of native Maya originals now lost, contain chronicles which are obviously copies of historical records in the Maya hieroglyphic writing. 3. The direct statements of the Spanish chroniclers of the sixteenth and seventeenth centuries to the effect that the Maya had the practice of recording their history in their books. (See pp. 42, 43.) Whatever may have been the nature of this event, whether historical or astronomical, it was of sufficient importance that, 20 years later, its first katun anniversary, 9. 17. 12.5. 17, although not a tun, hotun, lahuntun, or katun-ending, was commemorated by the erection of two monuments (Altar T and Stela 8) and possibly by a temple or hieroglyphic stairway (Fragment E')- After Stela N the hotun-endings at Copan appear to have been marked by larger constructions, such as Temple 2\a or the Reviewing-stand in the Western Court, or by altars, the subsequent stelae apparently being devoted to the record of tonalamatls, Calendar Rounds or other unusual counts. But the Acropolis was now nearing completion; as we have already seen, both the Eastern and Western Courts being finished in 9.17.0.0.0, and the last monument in the latter being erected either on the following hotun-ending, 9.17.5.0.0, or 15 years later, 9.18.0.0.0 (?).^ Moreover, the occupation of the city itself was also now drawing to an end. In the tonala- matl from 9. 1 7. 1 2. 0.0 to 9. 1 7. 1 2. 1 3.0 the last five stelse at Copan were erected, four in the Great Plaza at the Main Structure and the fifth a kilometer west of Old Copan at Group 10. This particular tonalamatl was of unusual importance, as has already been pointed out, since within its space fell the first katun anniversary of the important date 9. 16. 12.5. 17. Two stelae, C and H, were dedicated at the beginning of this tonalamatl, a third, Stela 8, commemorates the first katun anniversary of 9. 16. 12. 5. 17, and two others, F and 4, being dedicated at its end in 9. 17. 12. 13.0, the latter being not only the latest but also the most beautiful of all the Copan stelae. The three latest dates at Copan are the hotun-endings, 9.18.0.0.0, 9.18.5.0.0, and 9. 18. 10. 0.0, recorded on Altars W, G2, and Gi respectively, after which there are no later monuments, the inscriptions cease, the record becomes a blank, and the curtain falls for the last time on the scene of Maya activity in the valley, and indeed in the whole surrounding country. Doubtless the city was occupied for a few years longer, but some time during the next 20 years and before the beginning of Cycle 10 both Copan and Quirigua, and indeed most of the other Old Empire cities, were aban- doned, never to be reoccupied, and the Maya again set forth on another long exodus which finally brought them to other lands, other destinies, and five centuries later to their renaissance. 'The provenance of the monument presenting this date, Altar W, is doubtful. (See page 364). HISTORY OF COPAN DURING THE OLD EMPIRE. 43 1 The history of Copan during the Great Period is the history of the Main Structure, and here, in altar, stela, stairway, temple, palace, court, and plaza, we may read the record of the tribe's growing power and wealth. Her increasing influence outside of the valley has already been noted, as, for example, the colonization of Quirigua by emigrants from Copan at the close of the Middle Period. But during the Great Period more distant colonies were established, at Paraiso, Rio Amarillo, and even on the other side <; of the divide in the Chamelecon Valley at Los Higos. (See figure 57.) Fortunately, at the last-mentioned site the lahuntun-ending 9. 17. 10. 0.0 is recorded, so that we know at least one of her colonies besides Quirigua was occupied at the height of the Great Period; in fact, this latter date is only 2 years earlier than the last group of stelae in the Great Plaza. Her sphere of esthetic influence we may imagine to have extended far beyond this cluster of southern Maya cities, of which she was easily the leader, and to have been felt far to the south, southeast, east, and northeast, where peoples of much lower cultures eagerly copied, as best such outer bar- barians might, her art, sculpture, and ceramics, decorative motives of the latter being traceable as far to the southeast as Costa Rica. In the northwest, i. e., the Peten region, she must have come in contact, if not indeed in open conflict, with some of the great northern cities, Tikal, Nakum, or Naranjo, for example, and in this direction her sphere of actual dominion probably did not extend beyond the Golfo Dulce or the Sarstoon River. (See plate i and fig. 57.) In size Copan was second only to Tikal, and in learning and art she had no peers. The wealth of her inscriptions, probably comprising as high as 40 per cent, of the Corpus Inscriptionum Mayarum, has already been noted. In sculpture she enjoyed a similar preeminence not only in mass of material but also in superiority of workmanship, technical processes, and the like. In the extent of her architecture alone she may have fallen somewhat behind her great northern rival. Judged as a whole, however, Copan may be aptly called "the Athens of the New World," a title the writer has been wont to bestow upon her in drawing analogies from the ancient cities of the Old World; and in closing this summary of her history it may be claimed with perfect assurance that no other city of aboriginal America ever attained so high a level of cultural achievement. The curve of civilization at Copan, based upon the prevalency of the monuments, is shown graphically in figure 68, the data upon which it is based appearing in Appendix IX. The abscissae of the curve are the suc- cessive hotuns of the Long Count shown by the vertical lines, every fourth one of which, corresponding to a katun-ending as well, being heavier. The dates of these several hotuns appear above, only those corresponding to the second and fourth ones, i.e., the lahuntuns and katuns respectively, being written out. The brackets above indicate the three periods of the Old Empire. The ordinates of the curve are the 7iumber of times any given 432 THE INSCRIPTIONS AT COPAN. hotun occurs as the contemporaneous date of different monuments, all con- temporaneous dates not hotun-endings being plotted as at the ends of their corresponding hotuns. Thus, for example, all the monuments which date from 9.16. 12. 5. 17 have been plotted as dating from 9. 16. 15. 0.0, the current hotun-ending. In the Early and Middle Periods, when practically all of the monuments were erected at hotun-endings, this has no appreciable effect on the curve, but after 9. 16. 10. 0.0, when this practice began to fall into disuse, it introduces certain minor variations between hotun-endings not shown in figure 68. The black dots on the curve indicate the number of OLD EMPIRE EARLY PERIOD MIDDLE PERIOD GREAT PERIOD ooooooooooooooooooooooooooooooooooooooooo dodddcidodbobdddoodcjdddoooddodddddododb dddd3d^cjQdCjdSdgdod2'^2'^^'^2^2°9'^9°3'^2^2'^S'^2 000 6 ci d cj d -i -i oi 000 7 S l/l 5 ^ 1 \ 2 1 \ 4 t \ •' ' ^ \ / \ / \ n \ \ M ^ J Jj Fig -Dir.gram showing the chronologic distribution and frequency of the dated monuments at Copan. monuments assigned to the corresponding hotun-ending; when the curve passes through a vertical line without such a dot, it indicates that no monu- ments have yet been found which date from that particular hotun. The outstanding features of the curve are: 1. The practically stationary position at the first ordinate above the base- line (o) throughout the Early Period, with stations only at second and fourth hotun- endings until 9.9.10.0.0 is reached. This may be interpreted as indicating that throughout the Early Period the lahuntun and katun-endings were for the most part commemorated by the erection of but a single monument.^ 2. The first long minimum, from 9.9.10.0.0 to 9. 11. 0.0.0. This return of the curve to the base-line for 25 years, i. e., until the hotun-ending in 9. 11. 0.0.0, may be interpreted as indicating that during this period the extensive occupation of the valley was under way and new groups were being established, after the long period of quiescence at Group 9, during which the tribe was slowly gaining in strength. 3. The sudden upward swing of the curve in 9. 11. 0.0.0, to the first maximum. This may be interpreted as indicating that the extensive occupation of the valley after the close of the Early Period and during the first katun of the Middle Period had been completed by this date. 4. The sudden drop of the curve to the second ordinate above the base-line after 9. 11. 0.0.0 and its continuation there with but minor variations until the last 1 Stela IS and Altar Q', Stela l6 and Altar Y, and Stela 17 and Altar X have been plotted in figure 68 as three monuments instead of six, each pair being regarded as a single monument like Stela E and its altar. In all probability Altar Q' was formerly associated with Stela 15, since both record the same date, and the other two associations have been suggested as not improbable. Altars J', K', L', M', ?', and A' and Stelje 22 and 21 have been omi«('J from figure 68 as being of too uncertain date to plot properly, and the inscribed peccary skull from Tomb I (see pages 379-381) on the grounds that it is not a monument and hence does not properly belong to the monumental sequence. OTHER CITIES OF THE OLD EMPIRE. 433 katun of the Middle Period. This may be interpreted as indicating that the tribe, while unable to maintain the high level of seven monuments for each hotun-ending, was able to erect two on most of these occasions, showing a tendency, however, to drop back to second and fourth hotun-endings, as in the Early Period. A fact not disclosed by figure 68 is that most of the monuments after 9.11.0.0.0 were erected at the Main Structure, which from this tim.e on became the capital of the region. 5. The second long minimum from 9.14.0.0.0 to 9.16.5.0.0 (not very apparent in fig. 68). This second long return of the curve to the base-line for 45 years, i. e., until the hotun-ending in 9.16.5.0.0, save for the erection of four monuments only, may be interpreted as indicating the period during which the Acropolis was under construction. 6. The several and frequent maxima after 9.16.5.0.0, until the second and third high points of the curve are reached in 9. 16. 15. 0.0 and 9. 17. 15.0.0 respectively. This may be interpreted as indicating that the height of sculptural and architec- tural activity was reached at Copan, as everywhere else, in the eighteenth katun, i. e., from 9.17.0.0.0 to 9.18.0.0.0. 7. The sudden drop to the third ordinate in 9.18.0.0.0 and to the first ordinate in 9.18.5.0.0 and in 9.18.10.0.0, and its final return to the base-line before 9. 18. 15. 0.0. This may be interpreted as indicating a swift loss of power before 9.18.0.0.0, though no loss of technical proficiency, and the final abandonment of the city in the follow- ing katun before 9. 18. 15.0.0. OTHER CITIES OF THE OLD EMPIRE. Before closing this study of the Copan inscriptions, it appears advisable to describe briefly the dates of the other principal cities of the Old Empire, so that the chronology of Copan may be compared with that of her con- temporaries. OLD EMPIRE EARLY PERIOD MIDDLE PERIOD GREAT PERIOD 00000000000000 qoooooooqoooo 0000 o'Qoooqoooo 0000000 0000 opqo' 00 ddOdbOddddodddq<5qcJbdoddoodoodooc5c3ciododcJoacjC)dc>ooo'ddcidqo'tio'(ici UAXACTUN TIKAL COPAN PIEDRAS NEGRAS NARANJO ALTAR OE SAC-EL PAB PALENQUE NAKUM YAXHA YAXCHILAN TZENDALE5 CHICHEN ITZA ITSIMTE QUIRIGUA LA HONRADEZ SEIBAL EL CAYO LOS HIGOS IXKUN LA MAR CANCUEN AGUAS CALIENTES FLORES UCANAL BENQUE VIEJO QUEN SANTO 06 2 9' tri (6 d s d CO CO s ad 9 06 d d d d d iri d d d N 2 w d d ri 9 01' d d 'Si d d d d t'i rri S d 3 d 05 d d a a si d d d d 2- a-; d a d There is a possibility that the earliest date at Piedras Negras may be 60 or 70 years earlier than this, Stela 29 possibly dating from as early as 9.5. 15.0.0 or 9.5.5.0.0. This monument is fragmentary and its date has not been exactly deciphered as yet. OTHER CITIES OF THE OLD EMPIRE. 437 During the Middle Period the occupation of the southern Maya field went steadily forward; Palenque (9.10.10.0.0), Yaxchilan (9.11.3.10.13), and Tzendales (9.13.0.0.0) were founded in the west; Yaxha (9. 11. 5. 0.0) in the north; Itsimte (9.14.0.0.0) (?)inthe center, and Quirigua (9. 14. 13.4. 17) in the south; and by the end of the period almost all the larger cities had probably been founded, except perhaps Seibal, which, on the basis of its monuments, appears to have been the last big city established in the south. Even Nakum, the earliest sure date at which is 9.17.0.0.0, was almost certainly founded as early as the Middle Period. This is a very large site belonging to Class 2 in the writer's classification of the Old Empire cities (see p. 441), and in addition to its 3 sculptured stelae has 12 plain ones.^ Perhaps, as suggested for Tikal, the hotun-endings at Nakum were marked by plain stelae which were painted, thus carrying the monumental sequence back 3 katuns before 9.17.0.0.0 and into the Middle Period. Tozzer's map^ shows that the city was extensive, and the number of temples there indicates a date of foundation considerably prior to 9.17.0.0.0. In figure 69, on the assumption that the 12 plain stelae were either lahuntun or katun-markers like the carved ones, and that they marked the 12 katun and lahuntun- endings previous to 9.17.0.0.0, a possible foundation date of 9.1 1. 0.0.0 is suggested for Nakum; or 9.14.0.0.0 if they marked the lahuntuns and katuns prior to 10. i. 0.0.0, the latest date known there. One other city of the Middle Period requires some further comment, namely, Palenque. This important site, perhaps better known than any other center of the Old Empire, with the possible exception of Copan, is the only one where the chronology and art criteria are apparently contradictory. Spinden maintains that on the basis of the architectural remains, as well as the stylistic criteria, particularly that presented by the stucco-work, Palen- que is very late, in all probability being occupied after 9.18.0.0.0.'' On the other hand, the latest apparently contemporaneous date yet found there is 9.13.0.0.0, on the tablets in the Temples of the Inscription, and the Foliated Cross, and on the stela in front of the Temple of the Cross; and, so far as the chronologic side is concerned, the evidence is indisputable that the tablets in these two temples were dedicated in 9.13.0.0.0, although the possibility remains that originally they may have been parts of earlier buildings than those in which they are now found. Spinden believes the stucco-work in particular is late, i.e., after 9.18.0.0.0. It is especially unfortunate, therefore, that the only two Initial Series in stucco known at Palenque, one on the left pier of the facade of the Temple of the Sun and the other on the back wall of the outer chamber of the Temple of the Beau-relief, should both be almost entirely destroyed and impossible of decipherment. The writer believes it is possible, even probable, that Palenque may have been occupied down to 9.18.0.0.0 or thereabouts, though hardly any 1 Tozzer, 1913, pp. 162, 163. 2 Ibid., p\ates 32 and 33. ' Spinden, igi/f, p. 177- 438 THE INSCRIPTIONS AT COPAN. later. The latest date in the whole western part of the southern Maya field is 9.18.5.0.0 on Stela 12 at Piedras Negras, and about this time, on the evi- dence of the dates at least, it is necessary to postulate that the whole western region, including Palenque, was abandoned. One other event of the Middle Period deserves especial mention because of the tremendous influence it was to exert a century and a half later. Some time during the closing katun of the Middle Period Chichen Itza was dis- covered, thus opening up to the Maya a vast new territory to the north, devoid of previous inhabitants^ and admirably adapted to their peculiar type of civilization. With the beginning of the Great Period in 9.15.0.0.0 the horizon of Maya history broadens widely, and in the next century twice as many cities, in all parts of the southern Maya region, were founded as in the pre- vious four centuries. By this time the Maya were a rich and powerful peo- ple and the establishment of cities and the erection of temples and monu- ments had become, from the technical side at least, an easy matter, and the curve of civilization and cultural attainment surged upward. (See also figs. 68 and 70.) The first city to be founded in the Great Period, based on the monu- mental record, was either Seibal, 9.16.0.0.0, in the rich valley of the Pasion River in southern Peten, or La Honradez, in the extreme northeastern corner of Peten, possibly a little earlier. The earliest surely deciphered date at the latter is 9.17.0.0.0, but there are several other monuments there which on stylistic grounds are still earlier. After 9.17.0.0.0 the new sites follow each other in quick succession. El Cayo in 9.17.5.0.0 (?), Los Higos and Ixkun in 9. 17. 10. 0.0, La Mar in 9. 17. 15. 0.0, and Cancuen and Aguas Calientes in 9.18.0.0.0, when the zenith appears to have been reached, more monuments having been found which record this last hotun-ending than any other during the Old Empire. (See Appendix VIIL) After 9.18.0.0.0 no cities appear to have been founded until the last group, Elores, Ucanal, and Benque Viejo in 10. i. 0.0.0; indeed, there is a break in the monumental sequence of the Old Empire after 9. 19. 10. 0.0, not a single monument having been discovered which dates from the 30 years between 9. 19. 10. 0.0 and 10. i. 0.0.0.'- ' Mercer (1896, pp. 162-167), '" ^n excellent study of the caves of Yucatan, during tlie course of wliich 29 were examined and 10 excavated, reaches the firm conclusion that Yucatan had never had an earlier occupation than that of the Maya: "But results more important than these had rewarded our close examination of the position and contents of the human rubbish heap everywhere present in the caves. Though this layer was the only culture- layer our digging had fairly proved at Oxkintok, Loltun, and Sabaka, and though we had often failed to reach rock bottom at other caverns, there was nowhere ground for supposing that deeper digging or blasting would have upset our inference. An earlier people visiting Yucatan under its present topographical conditions must needs have left their trace in the caves, and because the undisturbed earth beneath the culture-layer discovered always failed to show trace of any deeper, older, or more primitive human visitor, the conclusion was that no such earlier people had seen the region while its stony hills, its torrid plain, and its damp caves were as they now are." ^ It is a curious fact that, as important as the date 10. 0.0. 0.0 must have been to the Maya priests, the end of the cycle during which they attained such cultural brilliance, not one contemporaneous monument has been found dating therefrom. OTHER CITIES OF THE OLD EMPIRE. 439 These last three cities could not have been occupied very long, since the last date recorded anywhere in the southern area, 10.2.0.0.0, is only 20 years later than their earliest date, and doubtless even at the time of their foundation the movement which was emptying the Old Empire region of its inhabitants was already nearing its end. After 10. i. 0.0.0 no more new sites were founded, and soon after 10.2.0.0.0 the few that had survived the general exodus up to this time were abandoned and the Old Empire was at an end. Much of the data upon which the foregoing conclusions are based are, of course, not only incomplete, but also of doubtful quality. Thus, for example, there are two important sections of the Old Empire area which have yet to be explored — the extreme northern part of Peten along the Mexican boundary and the western bank of the Usumacinta River from Tenosique southeast to Salinas de los Nueve Cerros and thence back to the base of the Cordillera, the latter comprising the northeastern quarter of the State of Chiapas, Mexico. The writer confidently believes a thorough exploration of these two densely forested and, in most parts, uninhabited regions will result in the location of new Maya cities and the discovery of new hieroglyphic texts, although it appears highly improbable that another city as large either as Copan or Tikal can still be hidden in either, or even one as large as the cities of Class 2 on page 441, since notice of such a site at one time or another would almost certainly have been brought out by some mahogany-cutter, chicle-bleeder, or even by Lacandon Indians who roam these forests. Again, it is obviously hazardous to assume that the earliest deciphered monument at a site is at the same time also the earliest one actually erected there, and similarly, that the latest deciphered monument is also the latest one which was erected there; and doubtless future discoveries will change some of the minor aspects of both figures 69 and 70. But the writer believes that in both these cases, i. e., the discovery of new texts, as well as the future dating of known undeciphered ones, such new data will not affect the general features of these figures, and that the broad outlines of Old Empire history may safely be accepted as laid down here. Another objection to the data upon which the above conclusions are based is that the sculptured monuments are not always an exact criterion as to the lengths of the periods of occupation, the most striking example of this kind being at Nakum, where only 12 of its 15 stelae, and these all late, are sculptured. This objection is perfectly valid, but the answer to it is that in the very few cases where it applies there are also found plain stelae which may doubtless be assumed to have taken the place of the sculptured stelae as the period-markers at such sites. Weighing all the evidence, the well-nigh universal custom of the Maya during the Old Empire to erect their monuments, and in some cases their temples as well, at the ends of even hotuns in the Long Count, the writer believes most of the data in figures 69 and 70 will be validated rather than 44° THE INSCRIPTIONS AT COPAN. controverted by such future discoveries, always admitting that new texts will undoubtedly be found from time to time which will introduce minor variations. The classification of the Old Empire cities into four groups, as given on page 441, based upon their s'ize, monumental and architectural remains, and varying degrees of importance, is perhaps hazardous, since surveyed maps are lacking for all of them save only Tikal, Copan, Palenque, Nakum, and Quirigua. It is offered here, however, only by way of suggestion in answer to the oft-repeated question, what was the relative importance of the different cities of the Old Empire; and it should be noted in using it that save for Classes i and 4, where the classification is fairly obvious, the assignments suggested are only tentative and open to revision any time additional data are available. The real difficulty in any such a classification is the proper distribution of the cities in Classes 2 and 3, or indeed whether it would not be preferable to group these cities together in one class. The principal objection to the latter solution is that it involves the grouping together of such important sites as Yaxchilan and Nakum with such relatively unimportant sites as Yaxha and Ixkun. The problem is further complicated by the fact that, architecturally considered, Palenque, Yaxchilan, and Nakum are very important, while their monumental remains, especially those of the first and last, are rela- tively insignificant ; whereas at Piedras Negras, Naranjo, and Quirigua we have the opposite condition, namely, important monumental sequences but rela- tively insignificant architectural remains. The only escape from this latter dilemma, since we do not know which the ancient Maya esteemed the more — a handsome series of stelae or an imposing assemblage of temples — would appear to be to give each an equal value and group them all together. Although this makes Class 2 as large as Class 3, its six cities are more nearly of a size, while at the same time the six cities of Class 3 all have about the same number of monuments. OLD EMPIRE EARLY PERIOD MIDDLE PERIOD GREAT PERIOD Oociocioo 00000000 'o 000 00000 000000000] 00 0000000 o| 0000000000000 q[ 00 OcidOoocicJOc50c>dcidC)OCJOCiOc)OooboboOC)CioloocJac50oooo'cJOocjooopqqq^ D"C)Oc3dOdo"CJbc)CjbbbodoOgbdc5obobbdodocigbOObdooGoooo*oooooooooooo S3;2idS^!^^2222'^°"^'~'^^^'^'*''*''^'^^"''"^'*°^''^ CD V- ^ MrjMfoi!3Sil92S2t^ti2292i^'^'~''^*^'^'^ rooJcdcDoJcbaJcbtOcocOcoincnai cn aim ti crioSoioicrimtTiCT-jCr) CTicTJcricri cTiCTjtnoi ffy C^ Ti (T) (Jl(T)ai(JlCT101 , „ 1j , 1 \ _ t\ : \ I \. \ I. '' \ ' „ \. \ f 1 ' ^--^ 3 - ' - .. - A \ - / ^.,.. " i' / ir \ ' '- -T Si V ■ g> 0^ KiG. 71. — Diagram showing supposed variation in rainfall in the Maya area from 500 B. c. to 1000 A. v., after Huntington. The striking agreements mentioned above between the climatic curve obtained from the Sequoia washingtoniana and the writer's correlation of Maya and Christian chronology appear clearly in figure 71, which shows the variation in rainfall in California for 1,500 years, during the last 500 years before Christ and the first millennium after Christ, as established by the varying thicknesses of the rings of 450 Sequoia washingtoniana.- Hunting- ton gives the following general explanation of this curve: ' Penck, 1914, p. 290. Other recent investigations in this field will be found in the bibliography under the following titles: Arctowski, 1910-1913, vol. xlh, pp. 270-282, 481-495; vol. XLiv, pp. 598-606, 745-760; vol. XLV, pp. 1 17-13 1; Brooks, 1916, pp. 249-255; Helland-Hansen and Nansen, 1916, pp. 1-341; and Huntington, 1918, pp. 483-491. ^Huntington, 1917, p. 158. THE FALL OF THE OLD EMPIRE. 451 "According to our hypothesis the high parts of the curve mean abundant rainfall in southern California, but diminished rainfall and a pronounced dry season m the Maya area. Therefore they are the favorable periods. The low shaded areas [the black sections, below the line ab in figure 71, the level of present pre- cipitation], on the contrary, indicate times of drought in California, but of abund- ant rain at all seasons in the Maya area, with consequent dense forests, difficult agriculture, overwhelming disease, and enervating damp heat at all times. "^ During the first five hundred years before Christ and down to the be- ginning of the third century after Christ, when it is necessary to assume the Maya were developing their peculiar culture and remarkable calendar and chronologic system, extremely favorable climatic conditions, denoted by the general height of the curve above the line ab in figure 71, seem to have prevailed between 14° and 18° north latitude, the region of the Old Empire civilization, according to Huntington's hypothesis. It will be noted further, that beginning with the third century after Christ, a steady decline in the curve sets in, which continues with but few minor variations until the middle of the fifth century. That is to say, according to the writer's correlation of Maya and Christian chronology, the climatic conditions during the Early and Middle Periods grew steadily worse, although up to the early part of the Middle Period they were still fairly good. By the middle of the fifth century, however, the annual rainfall had become as heavy as it is to-day in this region, a fact indicated by the curve having dropped to the horizontal line ab in figure 71, and living conditions, accord- ing to Huntington's hypothesis, had become intolerable. At this point, however, the curve sweeps upward and so continues with few minor varia- tions for nearly a century, until 540 a. d., indicating a return of favorable climatic conditions to the southern Maya area, and in the writer's correla- tion of the two chronologies, coinciding with the last katun of the Middle Period and the first two and a half katuns of the Great Period, the katuns of maximum sculptural and architectural activity. From 540 to 560 (approximately 9.18.10.0.0 to 9. 19. 10. 0.0, according to the writer) the curve again makes a sudden drop to the unfavorable level of present conditions, and then rises rapidly between 560 and 610 (approxi- mately 9.19.10.0.0 to 10.2.0.0.0, according to the writer), only to drop as suddenly to new low levels after 610, indicating heavier annual rainfall and more unfavorable conditions than even to-day, which continued for more than two and a half centuries. And just here is perhaps the most satisfac- tory agreement between the dates on the monuments and Huntington's diagram. No two katuns of the Middle and Great Periods are represented by fewer monuments than the two between 9.19.0.0.0 and 10. i. 0.0.0 (551 to 590); indeed, there are only two monuments now known (Stela D at Nakum and Stela 32 at Naranjo) which date from this period. The beginning of this lacuna in the monumental sequence falls almost in the middle of Hunting- ton's second minimum after the birth of Christ, and under his hypothesis is 'Huntington, 1917, p. 158. 452 THE INSCRIPTIONS AT COPAN. to be interpreted as indicating a falling-off in the erection of stelae owing to the prevalence of extremely unfavorable climatic conditions. A number of monuments, however, were erected in lo.i.o.o.o and 10.2.0.0.0 (591 to 610 according to the writer), which Huntington would have us believe was due to a return of more favorable climatic conditions, evidenced by the contin- uous rise of the curve from 560 to 610. Finally comes the sudden cessation of all monuments in the Old Empire after 10.2.0.0.0 (610 a. d. according to the writer), evidenced by a drop of the curve to new low levels of unfavora- bility and its continuance there for the next two and a half centuries, during the first part of which period the Maya are known to have abandoned the southern cities, as Huntington believes, due to the prevalence of the worst climatic conditions they had ever experienced. In spite of these satisfactory, not to say almost startling, agreements, the writer feels unable to accept this hypothesis as the principal explanation why the Maya abandoned such a large region as that covered by the Old Empire, although admitting that climatic changes may have been partially responsible therefor. The principal objection to Huntington's hypothesis, as already pointed out, is not his data on the variation of rainfall in southern California, which appear to be fairly well established by the varying thickness of the rings of the Sequoia washingtoniana, nor is it so much a question of possible inac- curacy in the writer's correlation of Maya and Christian chronology, which now appears from all indications to be correct with a maximum margin of error of not more than one year (see Appendix H), but rather uncertainty as to the accuracy of his basic postulate that climatic changes in southern California were accompanied by and coincident with diametrically opposed changes 4,000 kilometers distance to the southeast. This is a far-reaching and fundamental assumption, and even in spite of the apparently corroboratory results recently obtained by Penck, Arctowski, Helland-Hansen and Nansen, Hildebrands- son, and Brooks in this field, it seems safer to withhold unqualified accep- tance thereof until the laws governing climatic changes have been more thoroughly worked out than at present, and their nature and operation more clearly understood. There remains to be considered but one more hypothesis, which seeks to explain the extinction of the Old Empire civilization, namely, that sug- gested by Cook,^ that the system of agriculture practiced by the Maya eventually reduced the soil to such a state of unproductivity that they were literally starved into moving elsewhere. He sums up his conclusions in the following words: "Apait from dangers of war or pestilence to which the ancient communities of Central America may have been exposed, their existence was definitely limited by methods of agriculture which denuded the country of its forests, and destroyed the fertility of the soil. Civilization is at an end when an agricultural country ceases ' Cook, 1909. THE FALL OF THE OLD EMPIRE. 453 to be adapted to agriculture. To recognize these natural limitations of the primi- tive civilizations of Central America should make us more careful to appreciate and to correct the harmful tendencies of some of our own systems of agriculture."* Cook's conclusions are based upon personal studies in the highlands of Guatemala, and in the State of Chiapas, Mexico, among the descendants of the Old Empire Maya, the Quiche, Cakchiquel, Tzutuhil, and other modern representatives of the Maya stock in this region, where the methods of agriculture in vogue have changed little if any since pre-Columbian times. The Maya method of agriculture, ancient as well as modern, may be briefly summarized as follows: cutting, burning, planting, and sometimes weeding. As soon as the rainy season is over a new piece of forest is cleared, usually in January or February, and the fallen trees and underbrush are allowed to dry under the fierce heat of the March and April sun. When sufficiently dry to burn readily, usually in March and not later than April, the clearing is burned. Throughout the Maya area, north as well as south, the skies in April are covered with a pall of smoke, the sun setting each night a ball of fiery red. It is the time of the milpa (cornfield) burning, just before the end of the dry season. After the first rains, usually during the first half of May, the corn is planted among the fallen, charred trees, some of which have not been entirely consumed. A sharpened fire-hardened stick is generally used, and the corn planted 5 to 7 cm. deep. In some places weeding is practiced, in others not, the burning being deemed sufficient to retard the growth of weeds until after the corn has a good start. The harvest is usually not garnered at all. In August, when the corn has ripened, the ears are bent down and left hang- ing on the stalks to be gathered only as they are needed, in some cases being left on the stalks until the end of the dry season, when the last are picked and brought in. This method of harvesting is not so casual as it first appears, since the ears are much less subject to attack by insects, decay, etc., when left hanging on the stalks in the open air than when they are picked and stored in floorless thatched huts, where deterioration from all causes is more rapid. Whatever may have been the practice in ancient times, to-day the same field is not usually put under cultivation two successive seasons; but a new piece of forest is cleared and the same process repeated. This is done because the second season's crop from the same field is from 40 to 50 per cent, less than the first season's yield, and the natives, as a rule, prefer the larger return even at the cost of the greater labor involved in clearing new pieces of the forest each year. After lying fallow from 2 to 5 years, and in some places even 7 years, sufficient trees and bushes have grown up in a clearing to permit its being put under cultivation again, and then the same process is repeated and the cycle of ^nilpa rotation is complete. But each time the same clearing is ' Cook, igog, p. 23. 454 THE INSCRIPTIONS AT COPAN. burned, more and more of its humus soil is destroyed by the fire, and it requires longer and longer intervals of time for the woody growth upon which such a system of agriculture depends to return. After each successive burning more and more coarse grass grows up, and fewer and fewer trees and bushes, until finally nothing but coarse grass will grow and agriculture as practiced by the Maya is no longer possible. Cook describes this process in the highlands of Guatemala as follows: "The usual system of corn culture involves the repeated burning off of the weedy growth and a resulting exposure of the soil. This causes a gradual deterior- ation of the crops of corn and a slower renewal of the woody vegetation. New clearings in the forest are soon covered again with bushes, and can be cut, burned, and planted again within a year or two. With each cutting the interval has to be lengthened, until finally the land becomes thoroughly occupied by coarse grasses which are not killed by fire. The Indians can then make no further use of the land for agricultural purposes."' This method of agriculture, he goes on to say, has been carried in the highlands of Guatemala, at San Pedro Carcha near Coban, for example, to the point where the barren grassy zone is of such an extent that the Indians plant their milpas in the Cajabon district, 80 kilometers distant, and carry the crop home on their backs. As applied to the Old Empire, such a system of agriculture would have required a vast extent of territory to have supported the large population which formerly occupied this region, and if this hypothesis is correct, per- haps we are to imagine the ancient inhabitants of Copan, Tikal, and the other southern Maya cities as being driven farther and farther from their homes in order to find suitable forested regions in which to make the clear- ings for their milpas. Such a system of agriculture, if pursued long enough, would eventually have exhausted all the available forest lands within practicable carrying distances of the centers of population, and with grassy savannas stretching far out on every side replacing formerly forested regions; as Cook says: "civilization is at an end when an agricultural country ceases to be adapted to agriculture." This hypothesis appeals to the writer personally more than any of the others described. To begin with, it best explains the progressive abandon- ment of the Old Empire cities, which we have seen took place not all at once, but scattered over a period of about a century. This replacing of the forests by grassy savannas, and the end of cultivability so far as the Maya agricultural methods were concerned, must have come about gradually, reaching really acute stages at the different cities at different times, depend- ing upon such variable factors as their relative sizes and ages, and the general fertility of their surrounding regions. Thus the point at which complete abandonment and migration else- where came to be generally recognized as the only remaining solution for 1 Cook, 1909, p. II. THE FALL OF THE OLD EMPIRE. 455 their desperate economic situation must in the very nature of the case have been reached at a different time at each city, giving rise to different closing dates in each, the extremes possibly covering a century, and thus conform- ing better than any of the other hypotheses with the chronological data actually found. This hypothesis also explains the greater salubrity and healthfulness of the climate in the southern Maya field during the Old Empire than at present, upon which Huntington lays such stress. For with the forest cleared from the immediate vicinities of the cities and the surrounding country under cultivation, less standing water would have collected, and the malarial mosquito would have occurred in far less abundance than in these same regions to-day, making the living conditions far more healthful than they are now. The writer has stated that every Old Empire city with which he is familiar is, or was when first discovered, covered with a luxuriant vegeta- tion, a dense tropical forest. Cook offers convincing evidence, both botanical and zoological, tending to show that these forests are not original primeval forests, but are examples of reforestation over previously cleared areas. "Reforestation can be traced through a succession of temporary types of vegetation, such as pines, oaks, Curatella, Acrocomia, Cecropia, Castilla, and Attalea. These are abundant in regions undergoing reforestation, but are ex- tremely rare in virgin forests or in those sufficiently old for tropical hardwood trees to have grown to maturity and occupied the land, along with their attendant hosts of epiphytes and shade-tolerant undergrowth. It thus becomes evident that many of the existing forests are not permanent or primeval, but show the intermediate stages of a process of reforestation which probably requires several centuries to reach a stable condition "Central America is the home of many species of the Chamaedorea, and other small palms which live among the undergrowth in the shady depths of the forests. Nevertheless many localities affording conditions apparently suitable for these palms are without any representatives of the group. The undergrowth palms re- main abundant only in regions which have not been completely deforested for agricultural purposes, and especially in districts too mountainous and broken for agricultural use."^ His zoological evidence on this point is equally satisfactory: "Localities which contain remnants of ancient forests can be recognized by the presence of complete faunas of humus-inhabiting forest animals, such as the millipeds and centipeds, and some of the lower orders of insects and arachnids. In districts which are frequently cleared by cutting and burning many of the humus- inhabiting groups are exterminated. Even if they escape the fire they are unable to resist the exposure to the heat, sunlight, and dryness of cultivated lands. As long as the surface soil retains its humus and remains loose and pervious to water some of the smaller subterranean forms will persist, but when denudation is com- plete, or when the soil becomes sticky and impervious the humus-inhabiting types entirely disappear, as in many of the tenacious 'gumbo' soils of the Texas prairies. ^ Cook, 1909, pp. 12, 13. 4^6 THE INSCRIPTIONS AT COPAN. "Many forested places in Central America, which now afford conditions favorable for these humus-loving animals, are occupied by small and incomplete faunas. This shows that the period of reforestation has not been long enough to permit these sedentary, slow-moving creatures to spread again over the reforested areas. Thus in the valley of Ocosingo [an old Empire city, see plate i] in southern Mexico are many such tracts of new forest in which the humus fauna is still very poorly represented."^ Cook's suggestion that many of the Central American forests are of recent growth was corroborated by Whitford's investigations in the Motagua Valley in June 1919. Under Cook's hypothesis we are to conceive the Maya as founding their first cities in northern Peten in the midst of a vast primeval tropical forest, and as gradually felling this forest in the vicinities of their settlements and putting the cleared lands under cultivation. Later, penetrating southward through the Peten forest, some of the Maya eventually reached Copan and repeated this same process of clearing and cultivating there. Still later, other cities were founded, and gradually the whole region covered by the Old Empire was brought under intensive occupation. Perhaps as early as the Middle Period, judging from certain archaeologi- cal evidence to be presented shortly, there began to be concern over the economic condition, the increasing difficulties in the way of securing ade- quate supplies of corn, the great Maya staple. As early as the Middle Period, if Cook's hypothesis be correct, the increasing distances to which the people had to go to find suitable land for their milpas, especially around the older cities, where the zones of grassy savannas had become the largest, must already have occasioned the rulers and priestly caste considerable anxiety. But before the end of another century, i.e., in the Great Period, the situation must have become so acute as to have caused general dissatis- faction with, if not indeed actual distrust of, deities, rulers, and priests who could permit such a condition to continue unchecked. The writer imagines many of the religious ceremonies of the time must have been specially devised for meeting this grave national crisis, and for seeking the aid of their deities, particularly those of fructification and fer- tility, to avert the threatened extinction of the food-supply, which was drawing ever nearer and nearer. That no solution for this urgent national economic problem was found would appear to be indicated by the fact that from 9.18.0.0.0 on, one by one the southern cities were abandoned; at least no more monuments were erected in any one of them after 10.2.0.0.0, and indeed, a few of them — Palenque, Altar de Sacrificios, and Itsimte for example — had possibly been abandoned much earlier. (See fig. 69.) The earliest abandonments would naturally have taken place at the oldest cities, or those where the total area available for cultivation had not ' Cook, 1909, p. 14. THE FALL OF THE OLD EMPIRE. 457 been large in the first place. Thus, under Cook's hypothesis, it is not sur- prising to find that Copan was one of the first of the larger cities to have been abandoned, 9.1 8.10.0.0 being the last contemporaneous date found there. Not only was it one of the very oldest of the cities, but also it was located in a small valley where the area available for cultivation was comparatively restricted. Piedras Negras, located in a small semicircular flat surrounded by hills on one side and the Usumacinta River on the other, appears to have been abandoned about a hotun earlier, i.e., 9.18.5.0.0, and Yaxchilan, on the opposite (west) bank of the same river, in a similar location higher up, about the same time, 9.18.3. 1.5. The closing dates at Quirigua and Uaxactun, located in the midst of fairly broad plains, are the same, namely, 9.19.0.0.0, and that at Naranjo, similarly located, is only 10 years later, i. e., 9. 19. 10. 0.0. Thirty years later, in 10. i. 0.0.0, we get the closing dates at Benque Viejo, Ucanal, and Nakum, all in the northeastern corner of Peten, and 20 years later, in 10.2.0.0.0, the last dates of the Old Empire at Flores, Tikal, and Seibal, the last two being located in the midst of very large plains, the for- mer in the northeastern section of Peten, and the latter in the rich valley of the Pasion River, 150 kilometers farther south. Probably not long after 10.2.0.0.0 the last of the Maya moved out of the Peten region, some going north into Yucatan and others south into the high- lands of Guatemala, and the cities of the Old Empire were left deserted, to be reclaimed eventually by the same tropical forest from which they had originally been carved. Happily we are not without direct archsological evidence as to the two-fold direction of this Mayan exodus which com- pletely depopulated the Old Empire, the nature of which we will now pro- ceed to examine. To begin with, as early as the Middle Period, in 9.13.0.0.0 or 9.14.0.0.0 (according to the writer's correlation of Maya and Christian chronology), we have documentary evidence of the discovery of the region lying to the north of the Old Empire, i.e., the Peninsula of Yucatan. Three of the five chronicles in the Books of Chilan Balam, those from the Mani and Tizimin manuscripts, and the first chronicle in the Chumayel manuscript, record the discovery of the Province of Ziyancaan or BakhalaF in 9.13.0.0.0 or 9.14.0.0.0. The Mani manuscript describes this event in the following words: "Then [9.14.0.0.0] took place the discovery of the Province of Ziyancaan or Bakhalal; 4 Ahau [/. e., 9.15.0.0.0 4 Ahau 13 Yax], 2 Ahau [i. c, 9.16.0.0.0 2 Ahau 13 Tzec], 13 Ahau \i. e., 9.17.0.0.0 13 Ahau 18 Cumhu], three score years they ruled Ziyancaan when they descended here; in these years that they ruled Bakhalal, it occurred then that Chichen Itza was discovered. "- ' Brinton (1882, p. 124) gives the following etymology for these two names: Bakhalal "cane-brakes," halal the cane and hak roll or inclosure, possibly referring to the cane-brakes around the shores of the lagoon of this name in the southeastern part of the peninsula; and Ziyancaan, "the birth of the sky" zi'yan, birth and caan, sky. Brinton believes the latter is a picturesque allusion to the view from the seashore nearby, where the sky appears to rise from out of the water. - Ibid., pp. 95, 96, 100, and loi. 458 THE INSCRIPTIONS AT COPAN. The Tizimin manuscript places this event 20 years earlier: "8 Ahau [i. e., 9.13.0.0.0 8 Ahau 8 Uo] it occurred that Chichen Itza was learned about; the discovery of Ziyancan took place."' The Chumayel Manuscript agrees with the Mani text as to the date: "In 6 Ahau [i. e., 9.14.0.0.0 6 Ahau 13 Muan] took place the discovery of Chichen Itza."^ Let us next ascertain the location of the Province of Ziyancaan or Bakhalal. The former name has not survived as that of any known locality in the Maya area at the present time, but Bacalar, the Hispanicized form of the Maya Bakhalal, is the name of the large lagoon in the southeastern part of the peninsula, some 80 kilometers northwest of Santa Rita Corozal, British Honduras. (See plate i.) This is the first place-name mentioned in the Books of Chilan Balam which still attaches to a definite geographical locality; and a glance at plate i will show that the region west of Lake Bacalar, i. e., the Province of Bakhalal, lies directly in the path of a migra- tion from the northeastern corner of Peten, where we have seen the Old Empire Maya survived the latest at Uaxactun, Nakum, Ucanal, and Benque Viejo (10. 1. 0.0.0) andTikaland Flores (10.2.0.0.0) to Chichen Itza, where we will find them dedicating a temple as early as 10.2. 10. 0.0, only 10 years later than the closing dates in the last cities of the Peten region. The Peabody Museum Central American Expedition of 19 12, under Merwin, explored this region, and discovered a number of important new sites, Rio Beque, Ramonal, and Porvenir, and although no inscriptions were found at any of them, they probably belong to the Transitional Period of the New Empire, 10.6.0.0.0 to ii.i.o.o.o.^ The purely archaeological evidence is fully as satisfactory. In 1900, Thompson found at Chichen Itza, in that part of the city called Old Chichen Itza, a lintel with the Initial Series 10. 2. 9. 1.9 inscribed on its under side, which was assumed to have been its contemporaneous date. In 1918, however, the writer deciphered on the front of this lintel the lahuntun-ending 10.2.10.0.0 2 Ahau 13 Chen, which conforms with the usual practice of dedi- cating monuments, temples, etc., at even hotun-endings and may conse- quently be regarded as its contemporaneous date, being in fact less than a year later (331 days) than the Initial Series. Now, this date is not only the earliest contemporaneous date known at Chichen Itza, but also throughout the New Empire, and it is, moreover, only i lahuntun later than the closing dates of Tikal (Stela 11), Flores (Stela i), and Seibal (Stela i). Thus, on the chronological side, it becomes evident that the final abandonment of the Old Empire was roughly coincident with the growth of the New Empire, and it is even possible that the rise of the latter may have been partially responsi- ble for the fall of the former. ' Brinton, 1882, pp. 138, 144. '^ Ibid., pp. 153, 158. ' Morley, 1917^) pp. 140, 146. The writer understands this important material is now in course of publication. THE FALL OF THE OLD EMPIRE. 459 This colonization of Yucatan from the southeast agrees, moreover, with a tradition gathered by Father Lizana, who wrote as early as 1601 : "They [the first Spanish priests] knew that the natives came, a part from the east [the original colonization of the country here in question], a part from the west. So in their old language they call the east by another name than which they use to-day. Now they call the east, LiKin, which amounts to saying the place from which the sun rises upon us. And the west they call ChiKin, which means the fall or the end of the sun, or rather, where it hides itself in relation to us. But in old times they called the east, 'Cenial, The Little Descent,' and the west, 'Nohenial, The Great Descent,' (the former) referring to the few people who came on the one side, and (the latter) to the great multitude who came on the other, whoever they may have been."^ The former, the Little Descent, doubtless refers to the comparatively few people who first settled at Chichen Itza arriving from the southeast, and the latter probably refers to a more general movement into Yucatan, the Great Descent, from the southwest. Tabasco (.?), which seems to have taken place during the tenth century. - It was stated on page 457 that the migration which depopulated the region of the Old Empire was two-fold in direction; that in addition to the Ivlayia who moved nbrthward and colonized the peninsula of Yucatan, others moved southward and into the highlands of Guatemala, and later became the Quiche, Cakchiquel, Tzutuhil, and other related Maya tribes of historic times. The only dates in the Maya hieroglyphic writing known from this latter region are the two Initial Series on the Quen Santo (Sacchana) stelae found by Seler in 1895 at the Hacienda of Sacchana, in the State of Chiapas, Mexico, but which he states had been brought thither from the neighboring ruins of Quen Santo in the Department of Huehuetenango, Guatemala.^ (See plate i.) The dates of these two monuments are 10.2.5.0.0 (Stela i) and 10.2. 10. 0.0 (Stela 2), the latter being exactly the same date as that on the Chichen Itza lintel. Here, indeed, is an interesting archaeological condition, two monuments 600 kilometers apart, both without the territory occupied by the Maya dur- ing the Old Empire and both recording precisely the savie date, which date is only 10 tuns later than the latest closing dates in Seibal and Tikal, the nearest large Old Empire cities in each case; and therefore if Yucatan (Chichen Itza) was colonized from cities in the northeastern part of Peten, such as Tikal, Uaxactun, Nakum, Naranjo, and La Honradez, for example, the highlands of Guatemala (Quen Santo) would appear to have been colonized from cities in the southern part of Peten on the upper reaches of the Pasion, Chixoy, and Lacantun Rivers, such as Seibal, Cancuen, Aguas Calientes, Altar de Sacrificios, Ocosingo, and Tzendales. (See plate i.) 1 Lizana, 1893, pp. 3, 4. 'Landa (1S81, pp. 74, 75) mentions great numbers of Maya-speaking people immigrating into Yucatan from the south about this time, which he conjectures must have come from Chiapa, "because many words and com- positions of the verbs are the same in Chiapa and in Yucatan." 2 Seler, 1902-1908, vol. 11, pp. 251, 252. 460 THE INSCRIPTIONS AT COPAN. We are now in possession of the most important archaeological and docu- mentary evidence bearing upon the problem of the extinction of the Old Empire civilization and have reviewed the principal hypotheses seeking to explain the reasons therefor. The writer wishes in a few closing words to outline what appears to him to be the best explanation of the facts observed, admitting, at the same time, that the evidence is as yet too insufficient to warrant final conclusions. Probably as early as the Middle Period, the increasing difficulties of the agricultural situation, i. e., the farther and farther from their homes they were obliged to go in order to find suitable new land for cultivation, coupled with the deterioration of the cleared areas nearer the cities, had already begun to prove burdensome. This must have been especially true of the older and larger cities like Copan and Tikal, which had been occupied not only for longer periods but also by larger numbers of people; also, it is possible there may have been increasingly unfavorable climatic conditions. About the same time, according to the documentary evidence (the Books of Chilan Balam), and possibly because of this very fact, the region to the north of the Old Empire, the Province of Ziyancaan Bakhalal was dis- covered. This region is only 100 to 150 kilometers north of La Honradez, the northeasternmost Old Empire city now known (see plate i), and colonists pushing out either from there or from one of the other cities of this region in search of new agricultural lands, owing to the increasing deteriora- tion of the lands nearer home, discovered Lake Bacalar, i. e., the Province of Ziyancaan Bakhalal, in 9.13.0.0.0 or 9.14.0.0.0. The water of the lake was sweet, the country to the west fertile, and gradually the new region became colonized. While they were at Bakhalal, 60 years according to the Mani manu- script (from 9.14.0.0.0 to 9.17.0.0.0 according to the writer), in the words of the native chronicles, "it occurred then that Chichen Itza was discovered." Possibly hunting or exploring parties pushing still farther north from Bakhalal in search of new lands discovered the two great natural wells or cenotes at Chichen Itza, something less than 200 kilometers farther north, and because of this abundance of sweet water, all the more remarkable for being in such a parched and generally waterless country as the whole northern half of the Yucatan Peninsula is, they settled there. It is not to be supposed that these discoveries of new fertile lands far to the north at first attracted general attention in the Old Empire cities, but later, toward the middle of the Great Period after 9.18.0.0.0, when the writer believes the economic situation may have begun to become menacing, peo- ple's minds, particularly in the northern Peten cities, turned more and more toward these new lands, where living conditions were reported to be so easy. We may probably imagine the priesthood as opposed to the exodus already under way, because of its heavy investment in permanent build- ings, temples, and dwellings, and the rich monumental series, particu- THE FALL OF THE OLD EMPIRE. 461 larly in the larger cities;^ but during the last four katuns ot the Great Period from 9.18.0.0.0 to 10.2.0.0.0, when we get the last dates, there was probably a continuous movement of the people both to the north and to the south, a great outpouring from the Old Empire cities in both directions. Although we lack documentary evidence for this migration to the south, unless indeed the Popol Vuh or Sacred Book of the Quiche'- and the Annals of the Cakchiquels'' refer to an earlier period than now appears probable, there remains the indisputable linguistic fact that the Indians of the highlands of Guatemala to-day still speak dialects of the mother Maya tongue, morphologically probably no more distant one from another than the different branches of the Romance linguistic group — modern French, Spanish, Portuguese, and Italian. And to this must be added the direct archaeological evidence of the Quen Santo stelae, not more than 200 kilo- meters distant from Seibal in an air-line and only 10 years later than Stela I at that site chronologically, and finally a large body of ceramic material showing many Old Empire designs from all over the adjacent parts of the highlands of Guatemala. After 10.2.0.0.0 the region covered by the Old Empire was entirely abandoned, not to be reoccupied for another eight centuries, until the collapse of the New Empire about 1447, again set the Maya wandering over a large area in search of new homes for the third time within a period of 1,500 years. Probably such a large movement of people as that of the Maya from the Old Empire region is not to be explained by any single cause, and possi- bly more than one of the hypotheses described may have contributed to the final result. The writer is well aware that the time has not yet come when general problems in the Maya field can be definitely settled. Not only is the evi- dence still incomplete, but also the interpretation of the data at hand is in some cases open to doubt and revision. Thus, for example, such a problem as the probable cause or causes which brought about the fall and extinction of the Old Empire civilization is incapable of exact solution in the present state of knowledge, too many factors being uncertain or even unknown. The hypotheses described, on the other hand, cover in their range the differ- ent causes which have been variously suggested to account for the archaeo- logical fact, now generally recognized and already several times stated here, namely, the sudden cessation of the monumental series in the individual cities, but the gradual cessation over the area of the Old Empire as a whole. ' A similar situation arose in Guatemala after the great earthquake of 1773, which almost completely destroyed the capital of that country. The civil and military authorities were in favor of removing the seat of government to another valley, 25 kilometers farther east, to the present site of the capital. This step, however, was strongly opposed by the Church, because of its heavy property-holdings and investment, 43 large churches, monasteries, and convents alone, to say nothing of the many smaller buildings and elaborate equipment with which all were provided. Indeed it was not until 3 years later, in 1776, that the new capital was formerly established In its present location, and the old capital, Antigua Guatemala, has continued to be occupied down to the present day. ^Brasseur de Bourbourg, 1861. 'Brinton, 1885. 462 THE INSCRIPTIONS AT COPAN. Always admitting the possibility of an unknown factor in the situation, it appears to the writer that some, at least, of the several hypotheses sug- gested in the foregoing pages must have played an important part in pre- cipitating the exodus which the archaeological evidence demands; and of these, Cook's suggestion of an agricultural collapse appears to be the most probable. Possibly operating singly, but more probably working with other stimuli, such as climatic changes, fear, and superstition (the two last always potent forces among primitive peoples), bringing in their train attendant dis- ease, social unrest, and loss of confidence in themselves, their rulers, and their deities, these several factors may finally have brought about such an intol- erable condition toward the close of the Great Period that abandonment of the whole region ultimately came to be generally accepted as the only solu- tion for their extremity. And here we may leave the Maya. Their history in their new homes to the north and south, their brilliant cultural recovery and renaissance, particularly in Yucatan in the eleventh, twelfth, and thirteenth centuries, the Spanish Conquest of the southern Maya under Pedro de Alvarado in 1524 and of the northern Maya under Francisco Montejo the younger in 1541, and the final extinction of the last remnant of the Maya civilization around Lake Peten Itza by Martin de Ursua in 1697, are all chapters of another story which lie without the province of the present investigation, and which must await another occasion for adequate presentation. APPENDIX I. A PETROGRAPHIC DESCRIPTION OF THE MATERIAL OF THE COPAN MONUMENTS. Fred. E. Wright. Small specimens from three monuments, Stelae D and 3 and Altar U, were examined and found to be essentially the same in general appearance and type. In each one of these rock-masses, small inclusions of a denser, harder rock occur which differ from the host only in the fact that they are more indurated and contain more quartz. The rocks as a group are so highly altered that their original char- acters are difficult to decipher with certainty. They are evidently of tufFaceous origin and range in composition from dacites to andesites high in silica. The lack of ferromagnesian minerals, such as amphiboles, pyroxenes, and micas, is note- worthy. In the ground-mass, dusty areas, more or less rhombic in outline, may be seen here and there and may represent former amphiboles; but no definite proof was obtained of the presence of any one of the ferromagnesian minerals in the original rock-mass. In general appearance the specimens are dull, porous, fine-grained rocks, pale green and yellow-green in color, mottled here and there with colored angular areas, such as abound in altered tuffaceous rocks. The clastic texture is more clearly marked in the dark-colored inclusions than in the inclosing rock-mass. At first glance these inclusions appear to be fresh, vitreous rocks containing phenocrysts of clear, glassy plagioclase and quartz; but on closer inspection with a magnifying glass, and especially in the thin section under the microscope, they are seen to be clastic in nature and not essentially different from the host, except for the greater abundance of secondary quartz. In these indurated fragments the sharply curved intersections and interstitial spaces between the glass fragments of the original tuff are well preserved and give to the rock an unusual and characteristic appearance. Many of the angular and rounded cavities are lined with a white to pale green crust of soft material which under the microscope is cryptocrystalline to microcrystal- line, and weakly birefracting, with an average refringence of about 1.535. Many of the cavities contain, in addition, secondary quartz. The spherical shape of some of the cavities, I to 2 mm. in diameter, suggests bubbles in an original glassy lava. In the inclusions these gashes and other cavities are common and demonstrate the tuffaceous origin of the rock. In the less indurated rocks the cavities are still present, but they are not so apparent in the hand specimen because of the generally altered, dull condition of the samples, which has obliterated all contrasts. Scattered through each specimen are clear, glassy crystals of plagioclase feld- spar m.easuring up to 3 mm. in diameter. These crystals are developed as stocky prisms elongated along the axis 001:010; tabular development after the side pina- coid (010) is less common. The forms identified on the crystals are: (ooi), (oio), (no), (iTo), loi). Cleavage after 001 is well developed. The feldspar sections show slight zonal structure and average in composition an albite-oligoclase of the composition AbeAui. Fine albite twinning lamellae are common; Karlsbad twin- ning lamellae were observed in a few of the feldspar sections. Many of the feldspar crystals are irregular in shape and are evidently fragments of fractured crystals. Quartz grains are less abundant than the feldspars. They occur in water-clear 463 464 THE INSCRIPTIONS AT COPAN. grains, generally rounded in shape. Many of the primary quartz grains are cor- roded and embayed. Inclusions of long, needle-shaped crystals of high refringence were observed in several of the quartz sections. A brown weathering crust was observed on one of the specimens. The junc- tion between this crust and the adjacent interior is sharply defined; the weathered shell is more or less banded; but even in this shell the plagioclase feldspar still preserves its clear, glassy character. Under the microscope some of the feldspar sections are more or less regular in shape, but many are irregular and corroded. They are embedded in a much altered matrix which is not easy to decipher in all its details. In each thin section angular, irregular, short lines of dark, dusty material appear in the field; these are commonly curved and resemble the fracture hues of an origmal tuff rich in glass and pumice fragments. These are now profoundly altered and the interstitial spaces are more or less filled with secondary material, much of which is submicroscopic in size. Some secondary quartz is present in fine veinlets and aggregates; also calcite in thin, thread-like lines. Chlorite is comimon and fills cracks and mterstitial spaces; in some cases it appears to be an alteration product of original hornblende, but the evidence is too indefinite to be decisive. Small grains of iron oxide surrounded by a brown alteration zone are common; they are probably titaniferous magnetite. In the ground-mass, much of the material is too fine for satisfactory identifica- tion. There is present some argillaceous material, some secondary calcite, some quartz; aggregates of radial spheruhtes of a colorless, weakly birefracting mineral of positive elongation and refractive index about 1.480 occur, especially near cor- roded plagioclase feldspars. In the ground-mass there is present an isotropic or weakly birefracting substance of refractive index about 1.460. In view of the profound alteration of the matrix, the relatively unaltered con- dition of the plagioclase crystals is surprising. It appears that the glassy tuff frag- ments suffered rapid devitrification and subsequent alteration, whereas the primary plagioclase and quartz crystals and fragments remained unchanged, except for mar- ginal attack and alteration. The evidence, so far as it can be gathered from the few small specimens, points definitely toward their tuffaceous origin. The conditions which led to induration of the fragments of cuff which now appear as inclusions in the larger masses of rock can be ascertained only by field work. Both the inclusions and their host may properly be designated andesite tuffs; the presence of quartz in some of the speci- mens and the lack of ferromagnesian minerals indicates that the rocks are high in silica and alkalies and approach dacite tuffs in composition. The causes which led to the profound alteration of these rocks and even the details of the changes as presented in the few specimens at hand can not be profit- ably discussed without further field evidence. Geophysical Laboratory, Carnegie Institution of Washington. APPENDIX II. THE CORRELATION OF MAYA AND CHRISTIAN CHRONOLOGY. NATURE OF THE PROBLEM. No problem in Maya archaeology has excited wider interest or provoked more general consideration than the correlation of the Maya chronological system with our own Gregorian calendar; and indeed, few problems of similar importance in any archaeological field offer equal promise of ultimate exact solution. The nature of the Maya inscriptions upon which this correlation must neces- sarily depend is such as to indicate that an exact alinement of the two chronologies to the very day is not only a possible but also even a probable outcomiC of pending investigations. It has been frequently stated in these pages that a large part of the Maya inscriptions, indeed practically all of the deciphered glyphs, treat of the subject of time in its various manifestations, such as the lengths of the apparent revolutions of the sun, moon, and other planets around the eaith, and probably also of the eclipses which the first two occasionally undergo. In fine, it appears highly prob- able that actual astronomical phenomena of determinable nature are recorded in the Maya inscriptions; and it only awaits the exact identification of any one of these, such as any particular solar or lunar eclipse which was visible in northern Central America during the first six centuries of the Christian Era for example, to make immediately possible an exact correlation of Maya chronology with our own Gregorian calendar. Pending the solution of this problem by the astronomical method, which, however imminent it may be, has yet to be achieved, the writer wishes to suggest a correlation of the two chronologies based upon other data, which he believes is probably correct to within 4 months and possibly to within 49 days. It was stated in Chapter I that Maya chronology is a highly artificial but exceedingly accurate system for m.easuring elapsed time, with the day as the basic unit of the count. The current day is given in terms of the total number of elapsed days which separate it from the starting-point of the system, a point more exactly fixed in time than the zero-point of our own calendar, i. e., the buth of Christ. It finds its closest modern analogy not with our Gregorian calendar, but with our Julian Period, used by astronomers and chronologists in measuring elapsed tim.e, the hypothetical starting-point of which is the year 4713 B.C. Indeed, the two systems are so similar that Professor R. W. Willson, of the Harvard Astronomical Department, has suggested to the writer the propriety of describing any Maya date by its corresponding Maya day number after the practice of astronomers and chronologists in designating any given date in the Gregorian calendar by its cor- responding Julian day number. As a matter of fact, the whole problem of the proper correlation of Maya and Christian chronology may be reduced to precisely this: the correct engage- ment of the Mayan and Julian Periods at any single point; for if it were possible to establish a single point of contact between the two, every date in Maya chron- ology could be transcribed into its corresponding Julian or Gregorian equivalent, and the dates on the Maya monuments would suddenly become more accurately fixed in our own chronology than any event of Old World history prior to the birth 46s 466 THE INSCRIPTIONS AT COPAN. of Christ, and this solely because of the extraordinarily exact character of the native Maya chronological system, which within itself is absolutely accurate. Having stated the problem, let us next examine the evidence upon which the correlation here suggested is based. During the Old Empire, Maya dates were recorded in terms of their corresponding Initial Series numbers; but, as we have already seen, even before the close of the Old Empire this method had begun to give way to Period Ending dating, which, however, was only a more abbreviated form of the same system. In Initial Series dating the total number of elapsed days from the starting-point to the day recorded is given, while in Period Ending dating only the positions of the periods, whose ending-dates are recorded in the periods next higher, are given. These two methods, however, are but different expressions of the same system, as noted above; and as the following example will show, they are interchangeable. Thus, the Initial Series 10.2.0.0.0 3 Ahau 3 Chen may be expressed by the following Period-Ending date: 3 Ahau 3 Chen, End of Katun 2. Or, reversing this process, 3 Ahau 3 Chen, End of Katun 2 may be expressed by its corresponding Initial Series 10.2.0.0.0 3 Ahau 3 Chen. It will be noted in this reverse process that it is necessary to assume that the number of the current cycle was 10, but in the Old Empire the cycle-coefficient was always either 8, 9, or 10, and, as between these three, there is never any doubt on stylistic grounds as to which one was intended in a Period Ending date. Moreover, the record of the date upon which the specified katun ended renders it impossible for such a Period Ending to recur, fulfilling all the given conditions, until after a lapse of 18,980 katuns or about 374,153 years; so that for all practical purposes, as used in the Old Empire, Period Ending dating is as accurate as, and indeed is interchange- able with, Initial Series dating. Coming down to the New Empire, however, a very much less exact type of Period Ending is found, although even in these cases the Initial Series intended can usually be worked out. As used in the New Empire, in the few inscriptions that have come down to us (with but one exception),^ only tun-ending dates appear to have been recorded. These, moreover, frequently lack the month-parts of their corresponding terminal dates, and consist only of the record of a specified tun, together with the day on which it ended. When the corresponding month-parts are not omitted, such tun- ending dates are accurate within a period of 18,980 tuns or about 18,707 years, but when they are omitted, as is usually the case, the resulting dates are only accurate within a period of 260 tuns or about 256 years.^ A natural development out of the latter for use in the manuscripts, where long historical summaries had to be kept, but still only a further abbreviation of the original system, was the u kahlay katunob or sequence of the katuns, in which a katun was named after the day on which it ended, as 7 Ahau, 5 Ahau, 3 Ahau, for example, and no record was made of its corresponding month-part, or more im- portant still, of its position in the period next higher, that is the cycle. An example of this kind of count has already been given in Chapter I (page 43), where the 71 kahlay katunob will be seen to have consisted of nothing more than a series of the ending-days of the succeeding katuns accompanied by the record of the more important events, if any, which occurred in each. Finally, and this is most important of all in the present connection, the u kahlay katunob are more or less accurately correlated with Christian chronology by means of several events in them, the dates of which are given in terms of both chronologies. 'The inscription on the capstone in the outer chamber of the East Range of the Monjas Quadrangle at Uxmal. (See figure 74.) "See Morley, 19180, pp. 270-275. CORRELATION OF MAYA AND CHRISTIAN CHRONOLOGY. 467 The correlation here suggested therefore may be divided into two steps: (i) The correct alinement of the u kahlay katunob with Christian chronology. (2) The correct alinement of the Initial Series with the u kahlay katunob. When these two points of contact have been established, the dates of the Old Empire, the Initial Series, can be reduced to terms of our own chronology, through the medium of the common equivalents afforded by the u kahlay katunob} This, in brief, describes the method of procedure which the writer and every- one else has followed, and the different results reached by the several investigators of the problem (reviewed later), are due to the several interpretations placed upon the same evidence, chiefly the u kahlay katunob in the Books of Chilan Balam pub- lished by Brinton in 1882, under title of The Maya Chronicle sr Unfortunately, as will appear later, these native records are more or less con- tradictory, especially in regard to the exact Maya equivalents, for specific days in the Christian Era, for which reason it appears hazardous to push this evidence to the point of deriving from it a correlation for which accuracy is claimed to the very day. This is attempting to read the vernier-scale of our instrument more accurately than the instrument was built to register. On the other hand, if exact correlation is not demanded, and if certain obviously contradictory dates in the Christian calen- dar are eliminated, and finally if we will accept approximate correlation, i. e., a larger point of contact, say anywhere within a given year of the Christian Era, the writer believes satisfactory, and within these larger limits accurate, results can be obtained. SOURCES FOR THE CORRELATION OF THE U KAHLAY KATUNOB AND CHRISTIAN CHRONOLOGY. Taking up the first step mentioned above, i. e., the alinement of the u kahlay katunob with Christian chronology, let us first examine the original sources upon which this part of our correlation must be based. The writer finds twelve authori- ties of the sixteenth, seventeenth, and eighteenth centuries, nine native and three Spanish, which may be utilized in this connection as follows, arranged in chrono- logical order: I. The Chronicle of Chacxitlubchen, written about 1562 by Nakuk Pech, the native Maya chief of that town. II. Relation of the Things of Yucatan, written not later than 1566 and not earlier than the close of 1561 by Diego de Landa, fourth Bishop of Yucatan. III. The u kahlay katunob from the Book of Chilan Balani of Mani, copied not later than 1595 by some native Maya. IV. The u kahlay katunob from the Book of Chilan Balam of Tizimin, copied about the close of the sixteenth century, also by a native Maya. V. Page 66 of the Chronicle of Oxkutzcab, containing entries for 1532-1545, copied from "an ancient book," on May 29, 1685, by Don Juan Xiu.' (See figure 72.) VI. History of Yucatan, written in 1656 by Diego de Cogolludo, twelfth Bishop of Yucatan, and published in 1688. VII. An ancient Indian painting, bearing the date 1536 figured in the preceding. (See figure 73.) VIII. History of the Conquest of the Province of the Itza, written after 1697 and published in 1 701 by Juan de Villagutierre Sotomayor. 'The correlation of Maya and Christian chronology suggested in this Appendix was first proposed by the writer in igog at the Baltimore meeting of the Archaeological Institute of America (Morley, igioa, p. 193), when it was announced that Stela 9 at Copan (g. 6. 10. 0.0) dated from 284 to 304 A. D., depending upon which tun of Katun 13 Ahau 8 Kanlcin coincided with the year 1536. Subsequent investigations, described in this Appendix, have convinced the writer that it was the closing tun of this katun which coincided with the year 1536, and that Stela 9 at Copan therefore dates from 304 A. D. ^See Brinton, 1882. 'This is page 66 in the Gates pagination, or page 80 in the Breton pagination of this manuscript. With a few minor changes the Gates pagination appears to be the better of the two, 468 THE INSCRIPTIONS AT COPAN. IX. The first u kahlay katunob from the Book of Chilan Balam of Chumayel, copied in 1782 by Juan Josef Hoil. X. The second u kahlay katunob from the Book of Chilan Balam of Chumayel, copied in 1782 by Juan Josef Hoil. XI. The third u kahlay katunob from the Book of Chilan Balam of Chumayel, copied in 1782 by Juan Josef Hoil. XII. Page 85 of the Book of Chilan Balam of Chumayel, copied in 1782 by Juan Josef Hoil.i As used hereinafter, these several authorities will be cited under the corre- sponding Roman numerals above. Concerning the relative merits of these sources, in matters pertaining to the native chronology, in general the writer places greater confidence in the statements of the native writers than in those of the Spanish historians (II, VI, and VIII), and naturally the earlier the authority the more likely he is to be correct. Most trust- worthy of all appears to be I. Nakuk Pech was a memberof the noble house of Pech of Cumkal, and was himself the hereditary batab or chief of Chacxulubchen. He speaks of having been an adult in 1519, and he must have been of mature years during the period of the Spanish Conquest, in which he took no small part, aiding the Spaniards whenever possible. He thus grew up under the native regime, re- ceiving his education, which must have been of the best, since he was to become chief one day, at the hands of the Maya priesthood before the Spanish Conquest, and therefore while the native institutions were still flourishing. In short, he possessed first-hand knowledge of what he wrote, and his statements, especially those regarding the native calendar and chronology, are to be accepted with greater confidence probably than those of any of the other authorities cited. Even in matters relating to Spanish history, such as the first appearance of whites in the peninsula (the wrecking of Geronimo de Aguilar and his companions on the east coast in 1511), the arrival of Cortes at Cozumel on February 28, 1519, the fall of Tenochtitlan on August 13, 1521, the first arrival of the Spaniards at Merida in 1541, and the foundation of the city on January 6, 1542, Pech gives the correct year, and in the case of the fall of Tenochtitlan, even the correct day. Such accuracy on the part of a native in regard to Spanish events shows that he was an exceedingly careful writer, and gives to any statements he may make about his own calendar the highest degree of reliability. - Of the second source, Bishop Landa's Relation, little further need be said here, his work having already been described in Chapter I (pages 28-30), as being the sine qua non of our knowledge of Maya chronology. It was composed prior to 1566,^ but not before the close of 1561,* although some of the material may have been 'This is page 85 of the Gordon reproduction of this manuscript (see Gordon, 1913), or the face of folio 46 of the original. 'Unfortunately the original Pech manuscript has disappeared. Gates is of the opinion that the text in Don Rafael de Regil's collection in Merida was only a copy of the original made by Pio Perez. In igi8 the eminent Yucatecan scholar, Don Juan Martinez y Hernandez, found a duplicate chronicle by Ah Naum Pech, mentioned by Nakuk Pech as being the head of the family in his time, which he assures the writer is practically a word-for-word transcription of the Nakuk Pech chronicle, with only the name of the author changed. See Brinton, 1882, pp. 189-259, for the text and translation of the Nakuk Pech version. 'The title of the copy of Landa's manuscript in the archives of the Royal Academy of History at Madrid bears this date: "Relation of the things of Yucatan taken from what Father Diego de Landa of the Order of San Francisco wrote, mdlxvi." •■Landa mentions in his Relation (1881, pp. 79, 80, 103) that he held an auto dafe at Mani, where many idols, hieroglyphic manuscripts, etc., were burned, and Cogolludo fixes the date of this event as having taken place after September 13, 1561, and before the early part of August 1562. See Cogolludo, 1688, pp. 308-310 and 322. Brinton says this event took place in 1562 (Brinton, 1882, p. 90). Shortly after his quarrel with Bishop Total in 1562, Landa returned to Spain, where he was tried in 1565, and it therefore seems most probable that he wrote his Relation at some Franciscan establishment in Spain in 1563-1565, while he was waiting for his trial. CORRELATION OF MAYA AND CHRISTIAN CHRONOLOGY. 469 gathered as early as 1553.' One of his chief informants, he tells us," was that Nachi Cocom, halach vinic of Sotuta, who figures so sanguinarily in connection with the murder of Napot Xiu at Otzmal in 1536, to be described later (Event C). Un- fortunately, Landa gives little information as to the correlation of the two chronolo- gies, and even his single statement bearing upon this point is probably incorrect. He does, however, give the correct version of Event C, as will appear later, which indeed must have been a matter of common knowledge at that time, having hap- pened only 13 years before he first came to Yucatan in 1549.^ The third source, the u kahlay katunob from the Book of Chilan Balam of Mani, was compiled not later than 1 595, according to internal evidence in the text.^ This chronicle and the other four u kahlay katunob (IV, IX, X, and XI), the writer be- lieves were copied by native Maya, perhaps directly from Maya historical codices, which have since been either lost or destroyed. In spite of the fact that both III and IV have several lacunae in their series of the katuns, in essential points they agree remarkably with IX and X, and occasion- ally even with XI, the least complete of them all; and they constitute, in the writer's opinion, perfectly reliable sources for the reconstruction of the main aspects of New Empire history. The fourth source, the u kahlay katunob from the Book of Chilan Balam of Tizimin, so closely resembles III, not only in phraseology but also in the positions and lengths of its several lacunae, as to indicate that both must have been copied from the same original, and, as already noted, probably about the same time.^ This close similarity is all the more remarkable in view of the fact that III is said to have come from Mani,^ in northwestern Yucatan, to which the Xiu removed after the abandonment of Uxmal in the middle of the fifteenth century, and IV from Tizimin in northeastern Yucatan, in the territory of the Itza. There has been some attempt in IV to fill the later lacunae in its series, since this chronicle agrees closely with III down to the fall of Chichen Itza and the end of the League of Mayapan, but after these events several katuns have been interpolated, which make a duplication in its series, some sections being recorded twice. A case in point is the interpolation of 13 katuns (from Katun 11 Ahau to Katun II Ahau), after the Katun 2 Ahau in which the Spaniards are said to have 'The specimen Maya year which Landa gives (i88i, pp. 90-102) begins with the day 12 Kan [ibid., p. 97) which fell in 1553, according to almost all the early sources. See pages 495-497. ^Landa, 1881, p. 76. Cogolludo, 1688, p. 268. ■■See Brinton, 1882, p. 70. 'See ibid., p. 136. "The Mani manuscript is first mentioned by Stephens, who says Pio Perez furnished him with a copy of "a fragment of a Maya manuscript, written from memory by an Indian, at some time not designated, and entitled: Principal epochs of the ancient history of Yucatan." (1841, vol. 11, pp. 278-280.) Brinton, who had access to Berendt's notes, says the latter states that the u kahlay katunob published by Stephens was from a manuscript in the possession of a native schoolmaster of Mani named Balam. He quotes Berendt as follows on this point: "The historical data which Stephens published in the Appendix of his work were extracted from such a book of Chilan Balam in the possession of an Indian of Mani, master of the school, who, because he had the same name Balam, pretended to be a descendant of the priest of the Maya, who gave his name to this class of writings." {Chilan Balam, Articulos y Fragmentos en Lengua Maya. MSS. Advertencia, p. vii. See Brinton, 1882, p. 91, note I.) Finally, Brinton gives an extract from the Codice Perez, a mass of material copied by Pio Perez from various sources, among others this u kahlay katunob, which states that this chronicle was in the possession of the master of the chapel at Mani' "Here ends the book entitled Chilambalam (from which IH was extracted] which is preserved in the town of Mani in the possession of the master of the chapel." {Ibid, same page and note.) Unfortunately, this most important manuscript has not been seen or heard of since the War of the Castes, which devastated Yucatan in 1847 and 1848, and the writer greatly fears it was destroyed at that time. It is not known surely whether a copy of it may be in existence or not. .jQ THE INSCRIPTIONS AT COPAN. first reached Yucatan (Event A), and for which not a single event is recorded, the whole being merely a repetition of the previous 13 katuns. With the other ukahlay katunob to check by, however, especially IX, such repetitions may be eliminated, and both III and IV can be brought into satisfactory agreement with IX and X.^ The fifth source is page 66 from the Chronicle of Oxkutzcab, which is a col- lection of titles, baptismal certificates, and probanzas de hidalguia (proofs of nobility) family papers, etc., of the Xiu family, covering the period from 1608 to 1817, which remained in the possession of the family down to within the past two decades, and which is now in the Peabody Museum at Cambridge, Massachusetts.' 'The Tizimin manuscript along with the Chumayel, Calkini, Kaua and others was copied by Berendt in 1868. After Berendt's death his collection was acquired by Brinton, and most of the latter's investigations in this field were based upon it. Bishop Carrillo y Ancona of Merida, says that the Tizimin manuscript was in his library in 1 870. See Carrillo y Ancona, 1870, p. 128; also Brinton, 1882, p. 136. And while still in his possession in 1887, Teobert Maler'made photographic copies of the Tizimin, Chumayel, Calkini, Kaua, and other manuscripts. A complete set of these prints, mounted and annotated by Maler, is now in the Gates collection. After Carrillo's death the Tizimin manuscript next came to light in the possession of the lawyer who administered his estate, Don Ricardo Figueroa of Merida. When the writer was in Yucatan in 191 3 he examined the Figueroa collection several times, but did not see it, although he was told that it was still in Figueroa's possession at that time; however, it has never been seen since, and its pre.^ent whereabouts are unknown. The writer was extremely fortunate in obtaining a copy of this manuscript in Merida in 1913, which Gates believes to be the work of an educated Maya, written about 30 to 40 years ago. On the front page is copied the following notation: "The cura who subscribes himself gives this book to Sefior D. Crescendo Carrillo Pbro. for the use which he wishes to give it. Tizimin, March 23, 1870. Manuel Luciano Perez. Rubric. Chilan Balam 'Codice Tizimin' "From the collection of the Pbio. Cres. Carrillo Ancona. Rubric. Note "This Codex is also called 'anonymous' inasmuch as the name of the Indian who wrote it does not appear. This copy is now in the Gates collection. It is written in a cleai, fine hand, exceedingly legible throughout, and is perhaps the best copy of the Tizimin manuscript now extant. The above note indicates that it was in the possession of the Cura of Tizimin prior to Match 23, 1870, and it doubtless emanates fiom that vicinity. ^This manuscript was obtained from Bernabe Xiu (1839-1911), or someone close to him, by Thompson in 1900 or 1901, and is now in the Peabody Museum. This Bernabe Xiu in all probability was a lineal descendant in the fifteenth generation of the Napot Xiu, who met such a violent death at Otzmal in 1536 (see Event C), and thus was descended from the former royal house of Uxmal. When the writer was at Oxkutzcab, Yucatan, in 1918, he obtained from Dofia Felipa Xiu, the youngest daughter of Bernabe Xiu, genealogical data of the family from 1918 back to the close of the eighteenth century, and°through recent collaboration with Mr. William Gates, it has been possible to connect the modern representa- tives of the family with the genealogical tree on one of the pages of the Chronicle of Oxkutzcab. This page is unfortunately frayed at the edges, but in the lower left-hand margin can clearly be distinguished a part of the day- sign Ahau, and below this to the right there is a black dot and just below this, the word "katun." The coefiicient here can only have been 2, 3, or 4, and of these, 3 is the only one historically probable, if not indeed possible. In the cofielation of the u kahlay katunob and Christian chronology suggested here, this can hardly have been other than the Katun 3 Ahau, which ended in 1635, which agrees well with the probable date of authorship of this part of the tree, as will appear below. The four generations at the left of the tree and also the Katun 3 Ahau just mentioned were probably added in 1635, nearly a century after the main part of the tree was drawn (1550), by Don Juan Xiu, born about 1622, who succeeded to the headship of the family in 1640, and who was living as late as 1689. His is the latest name on the tree, and, ingeniously enough, as if to indicate his authorship of these later additions, a hand points to his name. On the basis of four generations to a century, and the known dates of birth of the heads of the Xiu family during the seventeenth century, it is evident that the tree begins with the Tutul Xiu born about 1397, the last halach vinic to rule at Uxmal, who led the Maya chieftains against the Cocom of Mayapan in Katun 8 Ahau (about 1447) and destroyed their power. As a result of this victory, the largest cities were all abandoned and the Xiu themselves, presumably under this same individual, moved their capital to Mani. This break in the Xiu family history supplies the reason why at a later date, about 1550 (the generation contemporary with the Spanish Conquest is the last complete one, showing that the main part of the tree was drawn about that time), when the tree was made to establish their nobility and right to exemption from taxation, granted by the Spanish Crown to members of the native ruling families, their descent was traced only from the first member of the family who ruled at Mani, rather ihan from some earlier ancestor at Uxmal. From this Tutul Xiu, born about 1397, down to the little Dionisio Xiu, whom the writer saw playing on the mud floor of a thatched hut near Ticul, Yucatan, as a child of 3 in 1918, is a total of 22 generations. CORRELATION OF MAYA AND CHRISTIAN CHRONOLOGY. 471 The only document of this collection which concerns us here, page 66, is exactly dated by the following colophon at the bottom (see figure 72) : "Now I, on May 29, 1685, have copied this from an ancient book, that is characters by name Anares. I, Don Jhoan Xiu." 1 f 5 ? f ^j^^ A ^-" ^"^ ^t ^ h.^^^ ■}^ ^Oj> cayn a. -^Lcc Ai. ff^ ^ it, - -^ ^n/^rtO^^J^PCjf^ ^'^ ■no ^e. ^^ *n**i- a< >v,AT*»'o-!^/ve/- 7;^ ^^ <"^2^X cP Fig. 72— Page 66 of the Chronicle of Oxkutzcab. •i 472 THE INSCRIPTIONS AT COPAN. Again we have a native source of the greatest value, copied by a member of the most important native family in Yucatan at the time of the conquest, namely, the Xiu, from "an ancient book" in "characters by name Anares." In Anares we can hardly fail to see the word "analtehes," which Villagutierre uses less than 15 years later (1701) to describe the hieroglyphic manuscripts of Canek, the last independent Mayan ruler: "Because their king (Canek) had read it in his analtehes, they had knowledge of the provinces of Yucatan, and of the fact that their ancestors had formerly come from them {analtehes or histories bemg one and the same thing). "^ And in "characters" (Spanish caracteres) we have precisely the same word, which Landa, Lizana, and Ponce use to describe the hieroglyphics themselves." This evidence is so convincing that the writer believes the "ancient book" from which Don Juan Xiu copied the entries on page 66 of the Chronicle of Oxkutzcab on May 29, 1685, was nothing less than a hieroglyphic manuscript, in short a codex, and that his copy thereof may be regarded as an original source of the highest order of credibility. The sixth source, Cogolludo's History of Yucatan, contams no material for cor- relating the u kahlay katunob with Christian chronology. It gives, however, a full account of the history of the Roman Catholic Church in Yucatan during the first century of the Spanish regime, and is particularly complete in regard to the activi- ties of the friars. Its especial importance in the present connection lies in the fact that it fixes the dates in Christian chronology of Event C (the death of Napot Xiu), Event D (the foundation of Menda), Event E (the arrival of Bishop Toral in Yuca- tan), and Event F (the death of Bishop Landa), which the native authorities give in terms of their own chronology m the u kahlay katunob. Molina Solis says CogoUudo wrote his history in 1656, although it was not published until 1688 in Madrid,'' and we may doubtless assume that his information as to these early events of Spanish history in Yucatan is substantially correct. The seventh source is insepar- able from the preceding, since it was published therein, and is in fact the only illustration the book contains.^ Fortunately we have independent corroboration of its existence by Stephens in 1841. It is an old Indian painting (see figure 73) which, ac- cording to Cogolludo's interpreta- tion, sets forth the death of Napot Xiu at Otzmal. In the page facing this plate the Bishop states that the original painting had the year 1536 on it, although he tries to prove this is an error for 1541.^ This painting shows a circle of 1 3 human heads with FiC. 73 — Representation of a Katun-wheel (?) on page 133 of Bishop Cogolludo's Uisloria de Yucatlian. 'Villagutierre Sotomayor, 1701, p. 353. ^Landa, 1881, p. 103; Lizana, 1893, p. 3; and Ponce, 1872, torn. 11, p. 392. 'Molina Solis, 1904, p. 66. On page 127 of his history, Cogolludo speaks of "how I went this year of 1655 personally" to Campeche for the purpose of veiifying the date of its foundation. ^Cogolludo, 1688, p. 133. ^Ibid., p. 132. CORRELATION OF MAYA AND CHRISTIAN CHRONOLOGY. 473 the eyes closed, a name being written below each, that of Napot Xiu being at the top on the left. The thirteenth head, counting sinistrally from Napot Xiu, that of Ah Kin Chi, is slightly different from the others. Although the eyes are closed, an arrow also pierces it (see figure 73). This, Cogolludo explains as indicating that the life of one of the Xiu envoys, this Ah Kin Chi, was spared so that he might carry the tidings of the massacre back to Mani, although he was first blinded by having his eyes pierced with an arrow. Finally, within the circle, according to Stephens, is the Tree of Otzmal, where the massacre is said to have taken place. As will appear later under Event C, a somewhat different interpretation of this painting is more probable here. Fortunately, when Stephens was at Mani in 1841, the original was still in existence : "Albino had inquired of the cacique for the ancient relics of which we had heard ac- counts, and the Indians brought a copy of Cogolludo, wrapped up and treasured with great care in the casa real. This did not astonish us much, and they opened the book, and pointed out a picture, the only one in it, being a representation of the murder of the ambassadors of Tutul Xiu; and while we were looking at it they brought out and unrolled on the floor an old painting on cotton cloth, being the original from which Cogolludo had the engraving made. The design was a coat of arms, bordered with the heads of the murdered ambassa- dors, one of which has an arrow fixed in the temple, intended to represent the ambassador who had his eyes put out with this weapon. In the center is a tree growing out of a box, rep- resenting the sapote tree at Zotuta, under which the murder was committed, and which the Indians say is still standing. This tree I shall have occasion to mention again hereafter. The painting had evidently been executed by an Indian, and probably very near the time of the occurrence which it was intended to commemorate. Cogolludo refers to it as an interesting and ancient relic in his time, and, of course, it is much more so now. It is an object of great reverence among the Indians of Mani."' In this ancient painting we again have a native and probably contemporaneous source of highest credibility, and although in Cogolludo's engraving of it the year 1536 is omitted, his calling attention thereto in the accompanying text as an error on the part of the native painter serves to associate that date with the death of Napot Xiu in an unusually convincing manner. The eighth source, Villagutierre Sotomayor's History of the Conquest of the Province of the Itza, written between 1697 and 1701, contains but a single item of importance in the present connection; but this is no less than the direct statement that 1618, the year of Father Fuensalida's visit to Tayasal, the island capital of the Itza in Lake Peten Itza, occurred in a Katun 3 Ahau.^ This is particularly significant because, as will appear later, although the Itza had migrated from Yucatan two and a half centuries before 1697, their chronology was still in agreement with that prevalent in Yucatan down to and even after the Spanish Conquest a century and a half earlier. This source also agrees with III, IV, IX, and XI in placing the fall of Mayapan in Katun 8 Abau.'^ Our remaining sources, IX, X, XI, and XII, the three u kahlay katunob and page 85 from the Book of Chilan Balam of Chumayel, may best be compared to- gether. All four were copied by Juan Josef Hoil of the village of Chumayel, 24 kilometers east of Mani, about 1782. Page 81 of this manuscript shows Hoil's signature and the date January 20, 1782. They are doubtless redactions of much earlier originals now either lost or destroyed. The first u kahlay katunob from the Chumayel manuscript is, in the writer's opinion, much the best one of the five that have come down to us, since it is the only one in which the sequence of the katuns is absolutely without a single lacuna from beginning to end — an uninterrupted series of 61 katuns from the discovery of 'Stephens, 1843, vol. 11, pp. 260, 261. ^yiHagutierre Sotomayor, 1701, pp. 83, 84, 105, 106. 'Ibid.,pp. 105, 106. 474 THE INSCRIPTIONS AT COPAN. Chichen Itza (said to have taken place in a Katun 6 Ahau), down to a Katun 3 Ahau, nearly a century after the Spanish Conquest, and thus it presents an un- broken section of New Empire history for more than 12 centuries. ^ Although it does not reach as far back as the Mani and Tizimin u kahlay katunob by something Hke 275 years, it nevertheless begins with the earliest event of New Empire history, the discovery of Chichen Itza (some time between 432 to 452 A.D. in the correlation of the « kahlay katunob and Christian chronology suggested here), and unlike the other four, once started it continues without a break right down to and past the time of the Spanish Conquest. This chronicle is our most trustworthy authority for New Empire history, and it has proved of invaluable aid in checking up the lacunae and repetitions in III and IV, and in bringing them into agreement with it. It is not quite so detailed as III and IV, but its chief value lies in the continuity of its sequence of katuns, and it is no exaggeration to say that without it there would be no single u kahlay katunob in which entire confidence could be placed. It is preceded by the following heading: "This is the record of the count of the katuns from when took place the discovery of Chichen Itza; this is written for the town in order that it may be known by whoever wishes to know as to the counting of the katuns."^ In spite of the fact that it emanates from Xiu territory (Chumayel), it is clearly an Itza chronicle, beginning with the discovery of the site of the Itza capital, and mentioning "the men of the Itza" or "those of Itza" five different times, and the Xiu only once, and that at the very end, i. Q) appears in at least />^ one instance (Stela II at Piedras Negras) emerg-'^L) \3ing from Glyph A, V@}li;^i.f., the moon-sign proper, Si^f] and on the basis of this association, as well as the fact that it always has anj^^ ending-sign, and finally because of its position, always immediately preced-^^^ing Glyph A, the writer has suggested for it the general meaning: "here ends the count of the moon" or "next follows the current lunar month." This concludes the non-numerical signs, which, as already mentioned, doubtless had Httle real effect upon the meanings of the Sup- plementary Series. Turning next to the numerical glyphs, the most important of these is prob- ably Glyph A, the last sign of the Supplementary Series (see figure 79, a-h), and the key by means of which the general meaning of the whole count was first worked out. To begin with, the glyph itself, which is the moon-sign, is very constant in its normal form (figure 79, a-e) as well as in its head variant (figure 79, /) ; in fact, there is only one other variant known (figure 79, g and h), and that occurs but thrice.' The most constant characteristic about this glyph, however, and the one which gave the first clue as to its meaning, is its coefficient, which is always 9 (figure 79, a, b, e, and h) or 10 (figure 79, c, d, f, and g), and which is always attached to the right of the glyph (figure 79, b, d, e, and/) or at the bottom (figure 79, a, c, g, and h); that is, never at the left or above, as in the case of all other numerical coefficients.^ Forstemann had shown long before, in his study of the Dresden Codex, that the moon-sign there has a numerical value of 20, and in 191 5 Professor R. W. Will- son, of Harvard University, suggested to the writer that Glyph A of the Supple- mentary Series, which was nothing more than the moon-glyph with a coefficient of 9 or 10, was a sign for the 29 and 30 day month respectively, the nearest approxima- tions possible in terms of whole days of the exact length of a lunation. He further suggested that the close resemblance of the moon element in Glyph A to those forms so often found m the Dresden Codex, where the moon-glyph is used as a numerical sign for 20, when taken into consideration with these coefficients of 9 or 10, is of itself convincing proof that the Maya once used a lunar calendar consisting of alternate months of 29 and 30 days — such an arrangement as is in use in the Mohammedan calendar. And in a recent letter he refers the writer to a similar usage in the Babylonian lunar calendar in which the months were labeled i or 30 accordingly as they contamed 29 or 30 days.^ It then became apparent why these coefficients of 9 and 10 were attached to the right of or below the moon-sign, instead of in the usual positions at the left or above. This was done in order to indicate thereby that they were added to the moon-glyph, giving totals of 29 {i.e., 20-I-9) and 30 {;i.e., 20-|-io) instead of being multiplied by it, giving totals of 180 {i. e., 20X9) and 200 (20X10) as Maya coefficients do when they stand in the regular posi- tions to the left or above. The Maya had no fractions, and the only way they could keep the lunations correct in terms of whole days was to have some months composed of 29 days and ' (i) Stela A, Copan; (2) Stela N, Copan; and (3) the Hieroglyphic Stairway at Naranjo. ^There are only four exceptions to this known, the last three texts at Quirigua, Stela I, Stela K, and Structure i, 9. 18. 10. 0.0, 9.18.15.0.0, and 9. 19.0.0.0 respectively, and on the west jamb of the north doorway of Temple II here at Copan: 9. 16. 12.5 . 17. When the first three were inscribed, however, the purpose for which this differentiation of position had been devised was so well known that no mistake in meaning could arise about them, and their coefficients were allowed to go back to the regular positions for other coefficients. And for the explanation of the irregularity of the last, see pages 311-313. 'See Ginzel, 1906-1914, vol. I, p. 124. THE SUPPLEMENTARY SERIES. 555 Others of 30 days; and on the basis of this meaning the writer suggested that Gylph A declared the kind of month {i. e., whether composed of 29 or 30 days, a condition shown by Glyph A itself in each case), in which the accompanying Initial Series date fell, a hypothesis which has since become generally accepted. For example, Glyph A on Stela F at Quirigua has a coefficient of 10, and the whole glyph, there- fore, is to be interpreted as indicating that the corresponding Initial Series terminal date, 9. 16. 10. 0.0 I Ahau 3 Zip, fell in a 30-day month. The decipherment of this glyph at once established the general meaning of the Supplementary Series as a lunar count of some sort, which is further proved by the fact that no less than 6 of the 8 glyphs of which it is composed, in fact all except Glyphs G and F, at one time or another, and some of them all the time, have the moon-glyph as an essential part. Scarcely less important than the preceding sign is Glyph C (see figure 79, i-p), the fifth sign from the left, which occurs as frequently as Glyph A (in 97 per cent, of the texts under observation) and which, together with Glyphs A and X, are the three most important signs in the Supplementary Series. Glyph C, like Glyph A, is again constant, being composed of four elements which are always present in one form or another and a fifth, an ending prefix or superfix, which, since the Maya themselves omitted it in about two-thirds of the texts under observation, we may conclude was not essential to the meaning of the sign. These elements are: 1. A hand, always present and never changing in form. 2. A variant of the moon-sign, always present and never changing in form. 3. A bar-and-dot coefficient, always either 2, 3, 4, 5, 6, or in some cases no coeffi- cient at all, which, since I itself is never found, is probably to be interpreted as the equivalent of i in this glyph, as m algebra la and a are the same. 4. A human head, always present, though varying considerably, passing through a number of different types. 5. The ending prefix or superfix alluded to above, wanting in about two-thirds of the examples. The first two are so constant that they may be passed with brief comment. The hand in the Maya inscriptions has always been found to mean "end of," "close of," and hence even "zero." As attached to a variant of the moon-sign, therefore, it might perhaps indicate that a lunation or at least some longer lunar period had come to an end, or that possibly a whole month was in question. The third element, the bar-and-dot coefficients, of 2, 3, 4, 5, 6, and no coefficient at all, which hereafter will be called i, is, on the contrary, probably the most im- portant part of this glyph. R. K. Morley was the first to explain this coefficient as indicating in each case the position of the month declared by Glyph A, in a higher lunar period composed sometimes of 5 and sometimes of 6 of these 29 and 30 day lunar months. In other words, that these coefficients were ordinary positional numerals, like any regular Maya coefficients, those of the katun for example, and that they fixed the positions of the current month in a higher lunar period, no coefficient corresponding to the first or opening position. His basis for this explanation is pages 51 to 58 of the Dresden Codex, where there are recorded 405 successive lunations, arranged in a series of groups, some of which contain 5 lunations each, but more 6 lunations each. The individual luna- tions vary from 29 to 30 days in length, but these are so cleverly combined in each group, some having 177 days (/. e., 3X29 + 3X30), others having 178 days (i. e., 2X29 + 4X30), and still others 148 days {i. e., 2X29 + 3X30), that at no single group-ending in the entire period covered, nearly 33 years, is the cumulative error as much as a single day out with the total number of days in the corresponding total of lunar revolutions. 556 THE INSCRIPTIONS AT COPAN. This peculiar grouping of the 29 and 30 day lunar months on pages 51 to 58 of the Dresden Codex into higher lunar periods of never more than 6 months and never less than 5 immediately suggested that Glyph C of the Supplementary Series, which was known to treat of the moon, and the coefficient of which was never found to be higher than 6, was in fact a positional indicator in the same sort of a lunar group, the bar-and-dot coefficients from i to 6 indicating the position of Glyph A in a 5 or 6 month lunar period, an explanation which has since met with general acceptance. But this did not explain why the Maya, either in the lunar calendar on pages 51 to 58 of the Dresden Codex, or in Glyph C of the Supplementary Series, should have grouped the lunations into larger periods of 5 and 6 lunations each. The first to suggest the true explanation of this characteristic, or at least so far as pages 51 to 58 are concerned, was Meinshausen, who in 191 3 showed that the totals of days on pages 51 to 58 of the Dresden Codex very closely agree with the intervals be- tween eclipses of the sun and moon: "It will occur to everyone who observed the eclipse of the sun last summer (191 2) what an impression such a phenomenon of nature makes; therefore the fact will be easily under- stood that among the ancient civilized peoples who were much occupied with astronomy there were none which has failed to leave conspicuous records of such events. When, there- fore, starting from this fact, I searched the Dresden Codex for eclipses of the sun and moon, I was nevertheless not a little surprised instead of the expected slight and brief remarks to find, besides others, a long chapter that seemed to me to deal with such phenomena. A careful study of the periodicity of the eclipses of the sun and moon soon brought me to the conclusion that my assumption, in spite of the contrary explanation of Professor Forste- mann of this part of the manuscript, was wholly correct. The following comparisons of the periods between eclipses of the sun on the one hand and the moon on the other with the numbers of the Codex leaves no doubt that the latter actually arose from the observation of such appearances. "Eclipses of the moon ordinarily recur for the whole earth, as appears from the follow- ing table, after 177 days, but occasionally after 502 days. If 502 be divided by 177 the remainder is 148. The Codex shows exactly these numbers. The agreement is, moreover, of such sort that a coincidence appears to be excluded."' Meinshausen then gives a series of lunar eclipses which took place in the years 177810 1811 (first table, page 557), and aseriesof solar eclipses which took place in the years 1775 to 1808 (second table, page 557), as recorded in the Berlin Astronomical Almanacs. The first three columns in each table describe the date of the eclipse, viz, (day) 20, (month) 7, (year) 1778. The fourth column shows the dilFerences in days between two successive eclipses; the fifth column, the total number of days from the beginning of the series to each succeeding eclipse; and the last column, the numbers in the Dresden Codex. When the latter disagree with the corresponding totals in the fifth column by more than i day, they are inclosed in parentheses. In the first table, which shows the lunar eclipses, out of a total of 51 num- bers quoted from the Dresden Codex, 23 or 45 per cent, agree exactly; 12 or 24 per cent, are i day off; 2 or 4 per cent, are 2 days off"; and only 14 or 27 per cent, are 28, 29, and 30 days off". Meinshausen suggests that the reason there are any dis- agreements at all between the actual eclipse periods and the manuscript is due to the fact that not all the eclipses are visible at any one point on the earth's surface, and consequently that the remainder of 148 in the Codex was always placed arbitrarily immediately before an observed eclipse. The agreements in the second table, that showing solar eclipses, are even more significant. Here, of the 69 numbers quoted from the Dresden Codex, 28 or 40 per cent, agree exactly; 26 or 38 per cent, are i day ofF; 2 or 3 percent, are 2 days off; and only 13 or 19 percent, are 28, 29, or 30 days off. These agreements are again so 'Meinshausen, 1913, p. 221. THE SUPPLEMENTARY SERIES. 557 Table 0/ Lunar and Solar Eclipses. Eclipses of the Moon. Christian date. Day. 20 4 30 23 18 12 29 21 18 11 7 30 14 11 4 30 24 9 3 29 23 18 12 25 21 14 U 4 31 14 9 3 29 23 9 2 30 22 19 11 26 22 15 11 4 21 14 10 3 29 23 10 Mo. 7 12 5 11 5 11 3 9 3 9 3 8 1 7 1 6 12 5 11 4 10 4 10 2 2 7 12 6 12 5 11 4 10 3 9 3 9 1 7 1 7 1 5 11 5 11 4 10 3 Yr. 1778 1778 1779 1779 1780 1780 1782 1782 1783 1783 1784 1784 1786 1786 1787 1787 1787 1789 1789 1790 1790 1791 1791 1793 1793 1794 1794 1795 1795 1796 1797 1797 1798 1798 1800 1800 1801 1801 1802 1802 1804 1804 1805 1805 1806 1807 1807 1808 1808 1809 1809 1811 Differ- ences. Totals. 602 502 502 177 679 679 177 856 856 177 1033 1033 178 1211 1211 502 1713 (1742) 176 1889 (1919) 178 2057 (2096) 177 2244 2244 178 2422 2422 176 2598 2598 502 3100 (3130) 178 3278 3278 177 3455 3455 177 3632 3632 177 3809 3809 602 4311 (4340) 178 4489 4488 177 4666 4666 177 4843 4842 177 6020 6020 177 5197 5197 502 5699 (5728) 177 5876 (6905) 177 6053 (6082) 178 6231 6230 177 6408 6408 177 6585 6585 502 7087 (7116) 177 7264 7264 177 7441 7441 177 7618 7618 178 7796 7795 502 8298 (8326) 176 8474 8474 179 8653 (8651) 176 8829 8828 178 9007 9006 176 9183 9183 602 9685 (9714) 178 9863 (9891) 177 10040 10039 177 10217 10216 177 10394 10394 602 10896 (10925) 177 11073 (11102) 178 11251 11250 177 11428 11427 177 11605 11604 177 11782 11781 503 (11960) (11958) Codex Noa. Eclipses of the Sun. Christian date. Day. 26 21 19 15 14 9 5 30 24 19 16 14 8 4 27 23 17 12 6 3 1 27 26 20 16 9 5 30 25 20 19 15 9 4 27 24 17 14 14 3 27 22 16 12 5 31 1 27 25 20 16 10 5 29 24 18 15 8 4 28 23 18 14 13 8 7 4 28 21 17 11 6 1 30 27 26 21 16 10 6 29 25 25 Uo. 8 1 2 7 8 1 7 12 6 12 5 6 12 5 10 4 10 4 10 3 4 8 9 2 1 7 12 1 6 12 6 11 5 11 4 5 10 11 4 9 3 9 3 9 1 1 7 1 7 12 6 12 5 11 5 10 4 10 3 4 9 10 3 8 2 8 2 8 1 1 6 7 12 6 12 8 11 4 5 Yr. 1775 1776 1776 1776 1776 1777 1777 1777 1778 1778 1779 1779 1779 1780 1780 1781 1781 1782 1782 1783 1783 1783 1783 1784 1784 1785 1786 1786 1786 1786 1787 1787 1787 1788 1788 1789 1789 1790 1790 1790 1790 1791 1791 1792 1792 1793 1793 1794 1794 1794 1794 1796 1795 1796 1796 1796 1797 1797 1798 1798 1799 1799 1800 1800 1801 1801 1801 1801 1802 1802 1803 1803 1804 1804 1805 1805 1805 1805 1805 1806 1806 1807 1807 1808 1808 Differ- ences. 177 177 148\ 29/ 1471 30/ 148 177 178 176 178 1481 29/"' 177 148 176 178 177 177 177 1481 29/'" 1481 30/"'* 147 178 177 177 178 176 "8^78 30/"** 147 177 178 176 178 177 1481 30/"** Totals. Codex Nos. 1471 176 29/ 148 177 177 178 177 177 1481 29/"' 1481 29]"' 148 177 178 177 177 177 177 148 177 177 177 177 178 1471 30/"' 1481 29/"' 148 177 177 177 178 176 1491,, 29/"*' 1481 29/"' 148 177 177 178 176 1481 30/ 178 177 354 502 679 857 1033 1211 1388 1565 1713 1889 2067 2244 2421 2598 2775 2953 3100 3278 3455 3632 3810 3986 4164 4311 4488 4666 4842 6020 5197 5375 6551 5699 5876 6053 6231 6408 6585 6762 6939 9361 9538 9686 9863 10040 10217 10395 10571 10749 10926 11074 11251 11428 11606 11782 11960 177 353 602 679 856 1033 1211 1388 1565 (1742) (1919) (2096) 2244 2422 2598 2776 2953 (3130) 3278 3455 3632 3809 3986 4164 (4340) 4488 4666 4842 5020 5197 5374 5551 (5728) (5906) (6082) 6230 6408 6585 6762 6939 7087 (7116) 7264 7264 7442 7441 7619 7618 7795 7795 7973 7972 8160 8149 8298 (8326) 8475 8474 8662 8651 8829 8828 9006 9006 9184 9183 9360 9637 (9714) (9891) 10039 10216 10394 10571 10748 10925 (11102) 11250 11427 (11604) 11781 11958 558 THE INSCRIPTIONS AT COPAN. Striking as to make it practically certain that pages 51 to 58 of the Dresden Codex are an eclipse calendar, possibly of the sun, possibly of the moon, or possibly even of both. Through independent investigation Willson came to a similar conclusion about these same pages of the Dresden manuscript and in 1916 told the writer he believed them to be tables of possible solar eclipses;^ and during that and the following year Dr. Carl Guthe made the lunar count, especially as presented in these pages of the Dresden Codex and in the Supplementary Series at Quirigua, the subject of an ex- haustive investigation in connection with work for the doctorate at Harvard Uni- versity." He has also assured the writer that these pages of the Dresden manu- script can have no other interpretation, although he disagrees with the details of Meinshausen's conclusions. Returning to the Supplementary Series again, the restriction of the coefficients of Glyph C to the numerals I, 2, 3, 4, 5, and 6 only, can hardly be interpreted under the circumstances as indicating other than a similar condition, and it therefore appears probable that Glyph C is in some way connected with the eclipse phenom- enon, Glyph A showing the number of days in the month, whether 29 or 30, in which the accompanying Initial Series falls, and the coefficient of Glyph C, showing the position of that month in a group of 5 or 6 months, the length of which was determined in some way by the eclipse phenomenon. There remains to be explained one more important element of Glyph C, namely. No. 4, on page 555, the human heads of varying types. When the writer first approached this problem in 1907 he noticed that this head element is occa- sionally replaced by another element which looks like an eye,(5\@y?-^and which is the name-glyph of God M.^ Subsequently, however, ^|^~-—^^^~^ through the resemblance of these heads to head-variant numerals, he ^X^^ came to regard them as another set of numerical coefficients, of which the sign above was the zero, and he thus identified this element in 1916.* These heads indubitably resemble the head-variant numerials, indeed are identical with them, as reference to figure 79 will show, j and p being clearly the heads for 10, k the head for 7, m the head for 4, / the head for i or 8, and n the head for 6, the last being unmistakable. Note the cross element in the eye. These resemblances are so close as to indicate the identity of these heads with those of the numbers given, but the correct explanation thereof now appears to the writer to be more in the line of his first identification of them as signs for specific gods than that they are numerical coefficients. It has long been suspected that certain numbers were associated with the different deities of the Maya Pantheon, 10 with the God of Death (A of the Schell- has classification); either i or 8 with the Maize god (God E); 4 with the old god, possibly Itzamna (God D), 5 with the god of the 5 closing days of the year, Uayeb (God N); and 7 with the god with the cruller-like ornament over his nose (perhaps God K). It now appears to the writer, on the basis of these associations of specific head-variant numerals with specific deities, that this element of Glyph C is to be explained not as a series of numbers, but as a series of names of deities who presided over the corresponding periods, a case in point being the eye element above (see figure 79, i and 0), which is surely the sign for God M. 'See Motley, 1916, p. 394. ^The results of this investigation are in course of publication by the Peabody Museum as volume 6, No. 2, in its series of archaeological and ethnological papers. See Guthe, 1920. 'Compare Schellhas, 1904, figs. 45 and 46. ■■See Morley, 1916, pp. 380, 381. THE SUPPLEMENTARY SERIES. 559 As yet what period these deities presided over is doubtful. Indeed, before suggesting what appears to be the most hkely period in this connection from the Hmited evidence available, it is necessary to describe first the closest parallel afforded by Aztec mythology to the Maya practice, which we are supposing may be indicated by these head-elements in Glyph C, namely, the so-called Nine Lords of the Night of the Aztec pantheon. Bowditch has shown that, as portrayed in the Codex Borbonicus, the Nine Lords of the Night are there associated with a series of tonalamatls or 260-day periods, each Lord presiding in turn over a day thereof, the tenth day having the same Lord as the first day, the eleventh day the same Lord as the second day, and so on, each sequence being continuous, with but one exception. Since 9 and 260 contain no common factors, and since the latter is divisible by the former with a remainder of 8, it follows that if the Nine Lords of the Night were applied to the days of the succeeding tonalamatls without a break in either series, each tonal- amatl would begin with a different lord, and not until the tenth tonalamatl came around would the First Lord, Xiuhtecuhtli, coincide with the beginning day of a tonalamatl again. That is, all of the other eight lords would preside over the begin- ning days of tonalamatls before the First Lord would occupy that position again. As a matter of fact this is not the case, since the First Lord, Xiuhtecuhtli always pre- sided over the first day of all tonalamatls, and this could only be achieved by drop- ping the Ninth Lord, Quiahuitl, at the end of each tonalamatl, and having the Eighth Lord, Tepeyollotli, who always presided over the two hundred and sixtieth day, be followed by the First Lord in this one position alone. Bowditch clearly sums up the situation in the Codex Borbonicus as follows: "Apparently, therefore, the Tonalamatls succeeded each other, continuously lapping over from one year [365-day year] to the other, while the Lords of the Night accompanied the Tonalamatls and lost one of their number [always the Ninth Lord, Quiahuitl] with the ending of each Tonalamatl."' Let us ne.xt apply this general idea, namely that of a fixed series of deities, (not necessarily limited to 9 in number, however) presiding over a series of consec- utive time-periods (not necessarily the day), to this element of Glyph C, and see how it agrees with the data recorded. In the writer's study of the Supplementary Series in 1916 he found 11 pairs (one a triplet) of Initial Series, the members of each pair of which record the same date, as follows: Pair I. Stela 2, Copan Stela 12, Copan Pair 2. Stela 13, Copan Stela 3, Copan Pair 3. Stela I, Piedras Negras Stela 3, Piedras Negras Pair 4. Stela 11, Yaxchilan Stela II, Yaxchilan Trip. 5. Stela I, Yaxchilan Stela F, Quirigua Stela N, Copan Now, if these deities, presided over the successive days of the Maya chronologi- cal era, under the assumption that the head-element of Glyph C declares which deity it was that presided over the day recorded by the accompanying Initial Series, 9 10 15.0. Pair 6. Stela 24, Naranjo 9 12 10. s. 12 9 10 iS.o. Stela 29, Naranjo 9 12. 10.5. 12 9 II 0.0. Pair 7- Stela 7, La Honradcz 9 17 0.0, 9 11 . 0.0. Stela E, Qmrigua 9 17 0. 9 1 2 . 2.0. 16 Pair 8. Stela 13, Naranjo 9 17 10. 0. 9 12. 20. 16 Zooph. B Quirigua 9 17- 10. 0. 9 16 1 .0, Pair 9- Stela 1, Ixkun 9 18 0.0. 9 16. 1 .0. Zooph. 0, Quirigua 9 18 0.0. 9 16 10. 0. Pair 10. Stela 8, Naranjo 9 18 10. 0. 9 16 10.0. Stela I, Quirigua 9 18 10. 0. 9 16 10. 0. Pair II. Stela T, Quirigua 9 16 S-o. Stela M, Copan 9 16 5-0 'Bowditch, igoo, p. 152. 56o THE INSCRIPTIONS AT COPAN. we should find the sa7ne head-elements in Glyph C of the Supplementary Series in each of the above pairs, because each pair records the same day. In these 1 1 pairs, however, the head-elements of Glyph C are effaced or wanting in one member of 4 different pairs (Nos. i, 2, 7, and 9 above), which leaves but 7 upon which we can base our comparisons. In figure 80 the head-elements in Glyph C in these seven pairs are shown, the two in Pair 3 in a and h, the two in Pair 4 in c and d, the three in Triplet 5 in e,f, and g, the two in Pair 6, in h and i, the two in Pair 8 in/ and k, the two in Pair 10 in / and m, and the two in Pair 1 1 in n and 0. Of these, the heads of 5 pairs agree perfectly, Pair 10 from Naranjo and Quirigua (figure 80, / and m) both having the sign for God M; Pair 3, both from Piedras Negras (figure 80, a and b), both having the head of God A, Pair 4 both from the same monument. Stela 11 at Yaxchilan (figure 80, c and d) both having the head of a youthful deity, probably God E, Pair 11 from Quirigua and Copan (figure 80, w and 0) both having the head of God A again. Pair 8, from Naranjo and Quirigua (figure 80, / and k), both having the head of God A again. Fig. 80. — Head-elements, or name-glyphs of deities in Glyphs C of Supplementary Series, which accompany the same Initial Series: a, b, Piedras Negras, Stelae i and 3 (9. 12.2.0. 16) ; f, (/, Yaxchilan, Stela 11 (9.16.1.0.0); ^ /, g, Yaxchilan, Stela i, Quirigua, Stela F Copan, Stela N (9.16.10.0.0); h, i, Naranjo, Stela 24 and 29 (9. 12. 10. 5. 12); y, k, Naranjo, Stela 13, Quirigua, Zoomorph B (9.17.10.0.0); /, m. Naranjo. Stela 8, Quirigua, Stela I (9. 1 8.10.0.0); 71, 0, Quirigua, Stela J, Copan, Stela M (9.16.5.0.0). The other two. Pair 6 and Triplet 5, show a disagreement in this element. The first member of Pair 6, Stela 24, at Naranjo has the sign of God M (figure 80, h) and the second member, Stela 29 at Naranjo (figure 80, i), the head of God E, a clear contradiction. Triplet 6 has two of its three members in agreement. Stela i at Yaxchilan and Stela N at Copan both having the sign of God M again (figure 80, e and g); the third member, however. Stela F at Quirigua (figure 80, /) is aberrant, having the head of God E. The disagreements in the last two cases are so positive that in spite of the satisfactory agreements in the other five, it is probable that the day could not have been the period presided over by these deities, and that in this respect at least, they differ from the Aztec Nine Lords of the Night. Another possibility is that this deity in Glyph C may have presided over the corresponding lunar month, although, if so, the series cut across the 5- or 6-month lunar periods, because we find different deity heads associated with the same bar- and-dot coefficients, as for example in figure 79,7 and k, where the bar-and-dot coefficients are both 2, but the head in; is the head of God A, and that in k is the head of the God with the cruller ornament, possibly God K; or again in figure 79, n and 0, where the bar-and-dot coefficients are both 5, but the head in n is that of an unknown god (head-variant for the number 6), atid the corresponding element in 0, the sign for God M. THE SUPPLEMENTARY SERIES: 561 Finally, it is even possible that these deities presided over still longer periods like the lunar groups of 177, 178, or 148 days. In this connection, R. K. Morley has recently called the writer's attention to a significant characteristic of the Glyphs C in Pair 6 and Triplet 5, the two aberrant groups above. The coefficients of Glyph C in Pair 6 are i (Naranjo, Stela 24) and 6 (Naranjo, Stela 29), and in Triplet 5 they are again i (Yaxchilan, Stela i and Copan, Stela N) and 6 (Quirigua, Stela F). In other words, where the heads in Glyph C differ for Supplementary Series accompanying the same Initial Series, the corresponding coefficients — in the only five examples available for comparison — are either i or 6. The most obvious explanation of this phenomenon is that these differing deities presided over the larger lunar periods of which the coefficients of Glyph C denote the subdivisions, that is, over the periods of 177, 178, or 148 days, if we may follow the parallel afforded by pages 51 to 58 of the Dresden Codex. This explains why these deities could differ, and yet belong to Supplementary Series accompanying the same Initial Series, since under this explanation the head of God E in Pair 6 (Naranjo, Stela 29) with its coefficient of 6 indicates that God E was the deity who presided over the lunar period of 6 months ending on the day 9. 12. 10. 5. 12 (Initial Series of Stela 29), and the sign of God M (Naranjo, Stela 24) with its coefficient of i indicates that God M was the deity who would preside over the following lunar period of 5 or 6 months, which began on the same day. Again, in the case of Triplet 6, the head of God E (Quirigua, Stela F) with its coefficient of 6 would indicate that God E presided over the lunar period of 6 months ending on the day 9. 16. 10. 0.0 (Initial Series of Stela F), and the signs for God M (Yaxchilan, Stela i, and Copan, Stela N) with their coefficients of i would indi- cate that God M presided over the following lunar period of 5 or 6 months, which began on the same day. These two groups, comprising five examples in all, are, of course, tooinsufficient to establish this point, but Motley's hypothesis here, that the deities shown in Glyph C presided over the longer lunar periods of 177, 178, or 148 days as the case might be, satisfactorily explains both the agreements and the disagreements ob- served in this element of Glyph C, and fits the archaeological evidence better than the day or lunar month hypotheses. On the basis of this assumption the deities indicated by this element in Glyph C might possibly be called Eclipse Gods, since the eclipse phenomenon would appear to be associated with these 5-month and 6- month lunar groups. It would be premature to accept this hypothesis as proved, but the important fact in connection with this element now is that in it we probably have the sign of the deity who presided over the period in which the accompanying Initial Series date fell, hardly the day, or the lunar month, but more probably one of these larger lunar groups, or even some as yet unknown period. Returning to the Supplementary Series, the next sign. Glyph D (see figure 79, q-v), is the fourth from the left and immediately precedes Glyph C. It only occurs in about half of the texts under observation and therefore could hardly have been essential to the meaning of the count. When present, however, it is extremely constant, being composed of 4 elements, as follows: 1. A hand always present and never changing in form. 2. A variant of the moon-sign, always present and never changing in form. (Note that these two elements of Glyph D are identical with the corresponding ele- ments in Glyph C.) 3. A bar-and-dot coefficient varying from 2 to 19 inclusive and no coefficient at all, which is probably to be interpreted as i. 4. A subfix, probably unessential. 562 THE INSCRIPTIONS AT COPAN. Since the first two elements are identical with two of the elements of Glyph C, and because, when present, Glyph D always immediately precedes Glyph C, it seems reasonable to infer that the two characters are closely connected in meaning. R. K. Motley has pointed out by far the most important characteristic of this glyph, namely, that when Glyphs E and D both occur in the same text, as is fre- quently the case, Glyph D never has a coefficient and Glyph E always has one. The latter may be only i, it is true, as on Stela 8 at Naranjo, but wherever Glyphs E and D are both present the latter never has a coefficient. This characteristic very strongly suggests that Glyph D contains within itself the data for Glyph E also, that is, that Glyph D could be expanded into Glyphs D and E at will by attaching its coefficient to Glyph E, and either omitting Glyph D altogether or recording it without a coefficient. This may possibly explain why Glyph E is present only in 30 per cent, of the texts under observation. When it was desired to record them both, however, perhaps to fill in a space on the monument, then Glyph D was recorded without a coefficient, its coefficient going over to Glyph E. In fact, whatever these two glyphs may mean, it is evident that they are very closely connected, if not indeed actually synonyms, since we get all three com- binations of them possible: (l) Glyph D by itself; (2) Glyph E by itself; and (3) both together; but what we never get is D zvith a coefficient when E is present. Glyph E (figure 79, iv-b'), as the writer has already shown, ^ is probably reduci- ble Co a number of days. It is composed of but two constant elements, the moon- sign (the same variant as in Glyph A) and a series of numerical coefficients ranging from I to 19. Since the variant of the moon-sign is the same as in Glyph A, it must have a numerical value of 20, and since the coefficients attached to it are always either to the left or above (see figure 79, zu-V) they are doubtless to be regarded as multipliers, so that the glyph can stand for as low as 20 days (i. e., i X 20) or as high as 380 days {i. e., 19 X 20). Furtherm.ore, at Yaxchilan it is sometimes modi- fied by the same superfix as the cycle-sign is in the great-cycle glyph, that is to say, by an element that multiplies it by 20, making it 400.- And in one case at least (Stela 24 at Naranjo) this is modified by a bar-and-dot coefficient on its left as high as 18, making it equal to 7,200 days (i. e., 18X400) under this assumption, and with 7,600 days (i. e., 19 X 400) as its possible maximum. However probable it now appears that Glyphs D and E are synonymous and that both may be reduced to a number of days, we are yet entirely in the dark as to what determined the number of days they appear to record. Any suggestion that might be brought forward at this time would be purely speculative, but in closing the presentation of these two glyphs the writer wishes to note that he regards it as highly probable that this number has something to do with the accompanying Initial Series in each case, possibly indicating the distance therefrom, either before or after some specific observed eclipse phenomenon. We come next to the last glyph of the Supplementary Series, Glyph X (see figure 79, c'-j') which is the sixth character from the left or the third from the right. This sign is the least understood of all the characters of the Supplementary Series, is the most variable of all (hence its name Glyph X), and in importance ranks with Glyphs A and C. In the other glyphs of this count we have been able to establish essential elements or characteristics which remain constant, no matter through what changes the accompanying coefficients may pass. But in Glyph X for the first time we reach a glyph which possesses no element common to all of its examples, but on the contrary which passes through a number of changes. Happih^ it appears possible to classify these, at least roughly. The commonest element in 'Morley, 1916, pp. 384-387. -Ibid, 1915, pp. 117-119. THE SUPPLEMENTARY SERIES. 563 Glyph ,'' — ^ J^-"^""^ X is a pair of crossed human legs without the upper part of the body 1^^^ / I (see figure 79, c'-e'). In most cases these are combined with t . ^& the m.oon-sign (figure 79, c' and d'), but in a few cases with heads (figure 79, (•') or other forms. This pair of legs occurs in 11 out of the 61 cases where Glyph X is recognizable, or 18 per cent. The next commonest form of Glyph X is the head of God C, sometimes found with a coefficient of o (figure 79,/'), sometimes with a coefficient of i (figure 79, g'), sometimes with a coefficient of 3 (figure 79, h'), and sometimes with no coefficient at all. More rarely still we find other grotesque heads with coefficients of o (figure 79, i') and sometimes scattering geometric forms such as figure 79, ;'. The presence of the moon-sign not infrequently might indicate that Glyph X declared the accompanying phase of the moon in each Supplementary Series, a not unlikely part of the record in any lunar count. Or again, it may denote special planets or configurations which were prominent at the time of the accompanying Initial Series date; but here again we are as yet in the dark. Enough has been determined about the Supplementary Series, however, to establish beyond all doubt that it is primarily a lunar count, and probably that it represents an attempt to arrange groups of complete lunar revolutions so as to coin- cide with possible recurrences of some eclipse phenomenon, either solar or lunar, and possibly even of both. In collaboration with R. K. Morley and Guthe, the writer has formulated the following points which it appears safe ro accept as more or less surely established in regard to this count : 1. The Supplementary Series is a lunar count. 2. It appears to be an attempt to arrange a series of 29 and 30 day months, so as to make the diurnal calendar (z. e., the Long Count) measure the actual lunations without the resulting remainders reaching a whole day m any case. 3. Glyph C appears to show some further arrangement, probably of the months expressed by Glyph A, denoted by coefficients of from i to 6, inclusive. 4. The Dresden Codex, (pp. 51-58) shows a lunar count of 29 and 30 day months arranged in groups of 6 months of 177 days each, or more rarely of 178 days each, and occasionally groups of 5 months of 148 days each. 5. It seems reasonable to infer from points I to 4 that the Supplementary Series in the inscriptions and pages 51 to 58 of the Dresden Codex treat not only of the same subject, but also are constructed according to the same general plan. 6. This peculiar 6-5 grouping of the 29 and 30 day months is probably caused by the attempt to record or predict solar or lunar eclipses, or possibly even both. (See Meinshau- sen, 1913.) 7. The arrangement used m the Dresden Codex appears to have been followed, possi- bly with minor variations, in the inscriptions as well. Although all attempts to fit the former to the latter unchanged have failed, it seems necessary to believe that the same general scheme prevailed in both, because of the internal evidence supplied by the Supple- mentary Series themselves. 8. Finally, in attempting to ascertain the system used in the inscriptions, the most promising line of investigation appears to be to try first to discover the system followed in any one city (Quirigua and Piedras Negras, because of the regularity of the hotun-markers, are the best adapted for this purpose) from the internal evidence of its own inscriptions alone, since it is already apparent that the Supplementary Series are by no means con- structed according to the same universal system as are the Initial Series. That such should be true should not be wondered at, when it is considered that it is a valiant attempt to correlate two incommensurable time periods, and further, that the method of correlation may have been influenced by the eclipses visible at any one point. APPENDIX VII. THE HOTUN. Probably no single phase of the Maya inscriptions is more noteworthy than that of the periodicity of the monuments upon which they are engraved, and cer- tainly no other characteristic of the monuments themselves is of greater importance in determining their function or the nature of the phenomena which regulated the dates of their erection. While this periodicity is not characteristic of the earliest Maya monuments, as we shall presently see, it is encountered, nevertheless, far back in the Old Empire, and even by the early part of Cycle 9, especially at Copan, it had become the controlling factor in the erection of the monuments, and so continued down to the very end of the New Empire, more than 13 centuries later. ^ The writer's attention was first attracted to this phenomenon in 1907 in connection with his first work on the Supplementary Series, during the course of which he discovered that the monuments at Quirigua were erected at intervals of every 1,800 days. Seler had noted this condition at the same city (where it is most apparent) as early as 1899,- although when the writer made his discovery he was unaware of Seler's work; and in 1910, Bowditch made this phenomenon the subject of a special appendix in his The Numeration, Calendar Systems, and Astronomical Knowledge of the May as, ^ in which he reaches the following conclusion: " It might be deduced from this table [i. e., a table showing katun, lahuntun, and hotun- endings on the monuments] that in early times the Mayas marked the lapse of each katun by some kind of stone record, as Bishop Landa and others report, and that, as time ran on, they made this record more often, perhaps at the end of each half-katun, and that in still later times the record was made at the end of each quarter-katun. But as the early monu- ments are probably more defaced than the later ones, and as a large number of monuments are found the glyphs of which are too much worn to be identified, and as undoubtedly there are many monuments yet to be discovered, it is not at all improbable that the quarter- katuns were recorded from the beginning."'' r^ ^(F~^: m n p q f s Fig. 81. — Glyph for the hotiin: a, Piedras Negras, Stela 12; ^>, Quirigua, Stela C; c, Copan, Stela 1; d, Piedras Negras, Lintel 2; c, Piedras Negras, Stela 22; /, Piedras Negras, Stela 36; g, Yaxchilan, Altar 9; /;, Quirigua, Stela K; i, Quirigua, Stela J; ;', Copan, Stela I; k, Piedras Negras, Stela 6; I, Piedras Negras, Stela 25; m, Quirigua, Stela H; n, Piedras Negras, Stela i5; 0, Yaxchilan, Lintel 3; p, Pie- dras Negras, Stela 9; q, Copan, East Altar of Stela 5; r, Quirigua, Zoomorph G; s, Quirigua, Stela D. 'This is under the writer's correlation of Maya and Christian chronology; under Goodman's it would only be two and a half centuries less, however. (See Appendix IL) ^Seler, 1899, pp. (67o)-(738); republished in Seler, 1902-1908, vol. i, pp. 712-836. ^Bowditch, 1910, Appendix VIII, pp. 310-318. Hbid., pp. 3 10-3 11. S^i S66 THE INSCRIPTIONS AT COPAN. In 1910 the writer first identified the glyph for this 1,800-day period^ (see figure 81), and during the field seasons of 1912, 1913, 1914, and 1915 increasing evidence as to the widespread prevalence of this custom was accumulated, particularly in 1914 and 1915, during the course of trips to the Peten region of Guatemala, where this phenomenon was noted in 1914 at Naranjo, Ucanal, Ixkun, Seibal, Aguas Calientes, Altar de Sacrificios, El Pabellon, and Piedras Negras, and traces of it at Tikal and Yaxchilan, and in 1915 at Cancuen, Itsimte, Flores, Nakum, Benque Viejo, and La Honradez" (see plate i). In 191 5, at the meeting of the Nineteenth International Congress of Ameri- canists in Washington, these results were summarized in a paper entitled: "The hotun as the principal chronological unit of the Old Maya Empire; "^ where the following thesis was presented: "The Maya monuments, and especially those of the stela type, seem to have been used, perhaps primarily, to mark the passage of time, stelae being erected at intervals of every hotun (i,Soo days), or multiples thereof, such as every lahuntun (3,600 days), or katun (7,200 days), throughout the Old Empire, approximately 200 to 600 A. D."* In this paper, also, the name hoturv' was suggested for the 1,800-day period, and lahuntun for the 3,600-day period on the following etymological grounds: The Maya called their 360-day period tun, and their 7,200-day period {i.e. 20 X 360) katun, the latter probably being a contraction for kaltiin, kal being the Maya for 20, and kaltun, therefore, 20 tuns. Since ho means 5 in Maya, and lahun 10, on the basis of the above parallel the writer suggested hotun and lahuntun as names for the periods of 1,800 days and 3,600 days respectively." The glyph for the former, as noted above, is shown in figure 8i,the winged-Cauac element representing the tun, the bar above 5, and the superfix or prefix having the value of "end of" or "close of." The whole glj'ph, therefore, may be read as "end of 5 tuns"or"endof ahotun." 1< I m n o p q Fig. 82. — Glyph for the lahuntun: a, Naranjo, Stela 31; b, Naranjo, Stela 22; c, Naranjo, Stela 24; d, Tikal, Temple III; e, Tikal, Temple I; /, Seibal, Stela 7; g, Naranjo, Stela 13; h, Copan, Stela F; i, Naranjo, Stela 12; y, Quirigua, Stela F; t;, Tikal, Structure 10; /, Piedras Negras, Stela 4; m, Copan, Altar H'; n, Copan, Stela 6; 0, Palenque, Temple of the Inscriptions; p, Copan, Stela 15; q, Copan, Stela P; r, Quirigua, Stela F; s, Copan, Stela J; t, Copan, Stela A. The glyph for the lahuntun (see figure 82) was first identified by Goodman: "This glyph occurs immediately after a date that is the beginning of a loth ahau [i. e., the end of a loth tun]. The first character is the sign commonly employed, by itself, to denote that circumstance. It is a modification of the quadrated sign for 20 which forms Its characteristic feature. There are many variants oi it — in fact, I doubt if it ever occurs twice in the same shape; but, whatever the variation, it never fails to indicate a loth ahau, or an even 10 ahau [/. e., lo-tun] reckoning."^ Bowditch calls this interpretation doubtful, on the ground that the element immediately above the quadrated sign for 20, or better, sign for zero, is sometimes 5, 'Although the writer first identified this glyph in 1910, and described it in 191 1, owing to his frequent absences from the countty, publication thetcof was delayed until 4 years later. (See Morley, 1915, p. 166 and fig 67.) ^Morley, 1915a, pp. 343-346. '-'Ibid., igiyb. 'Ihid., p. 201. ■■This name had been previously suggested in ibid., 1915, p. 166. "Morley, 1917*, pp. 196 and 197. 'Cioodman, 1S97. p. 99. THE HOTUN. 567 sometimes 10, and sometimes 15, i. e., one, two, or three bars.' It is true that this element erodes down, so that frequently it resembles one or other of these three numbers (see figure 82, a, h, 0, and t), jjut when perfect it practically always has interior decorations, which show that it is not 10 or 15 and probably not 5 (see figure 82, b-g, i-n, p-r). The best explanation for this element would seem to be that it cuts the zero-element in half, as it always does, and cutting any ei'/io/i? period in half, i. e., one whose coefficient was zero, in the Maya vigesimal system gave 10 units of the next lower order in all places save periods of the third order, where it gave 9. As if in support of this morphology for the lahuntun-sign, we find it applied only once to a katun-sign (see figure 82, 0), where it accompanies the date 9.10.0.0.0 in the Temple of the Inscriptions at Palenque. But Katun 10 is precisely the only katun to which it could be applied and still retain the essence of the meaning sug- gested for it here, i. e., as indicating half of the period next higher, since Katun 10 is exactly half of a cycle. However, no matter how this sign acquired its meaning, Goodman correctly identified it, as the writer was able to prove in 191 5, and Bow- ditch's objection may doubtless be disregarded. Field work subsequent to the publication of the above paper on the hotun, moreover, has further corroborated the writer's earlier conclusions. Former pre- valence of this custom was established at Uaxactun in 1916," at Los Higos in 1917,'^ and even in the New Empire at Tuluum and Chichen Itza in 1918.'* In short, since the hotun-glyph was first identified in 1910, it has become increasingly evident that this period, or its second or fourth multiple (i. e., the lahuntun and katun), was the controlling factor in determmmg upon what dates the Maya erected their monu- ments throughout both the Old and the New Empire. In outlining the history of this practice let us seek its origin first, then trace its development during the Old Empire, and finally follow it north into Yucatan, and down to the period of the Spanish Conquest. For the Old Empire, our sources are exclusively archaeologic, but when we come to the New Empire we will find documentary evidence bearing directly upon this custom in both the Spanish and native sources. It has been stated in Chapter V that the three earliest dated objects known, the Tuxtla Statuette, the Leyden Plate, and Stela 9 at Uaxactun, and possibly even Stela 5 at the same site, 8.6.2.4.17, 8. 14. 3.1. 12, 8. 14. 10. 13. 15, and 8. 15. 10.3. 12 (?) respectively, show no traces of this custom. From which it appears probable that the first monuments were not erected at the hotun-endings, but that some other factor, possibly an actual historic event or astronomic phenomenon, gave rise to their manufacture and determined the dates recorded upon them. In this con- nection also it should be noted that none of these four texts have Supplementary Series accompanying their respective Initial Series, at least in a recognizable form. That is, it appears probable that the 29 or 30 day lunar month and the eclipse features were not added to this type of Maya record until some time after their magnificent chronological system had been devised. The earliest possible occurrence of a period-ending date is on Stela 8 at Tikal, where the writer believes he may have found the lahuntun-ending 9.0.10.0.0, and the earliest certain example is on Stela 24 at Copan, 9.2.10.0.0, 40 years later. The earliest possible occurrence of a katun-ending is on Stela 9 at Tikal, where the writer believes he may have found the katun-ending 9.2.0.0.0, and the earliest certain occurrence is on Stela 7 at Copan, 9.9.0.0.0, 140 years later. Finally, the earliest possible occurrence of a first or third hotun-ending is on Stela 16 at Copan, 9.4.15.0.0 (very doubtful), and the earliest certain occurrence is on Stela 25 at Piedras Negras, 9.8.15.0.0, 80 years later. Even eliminating these doubtful earlier examples, it is 'Bowditcli, 1910, p. 247. 'Morley, 1916^, pp. 339-341. ^Ibid-, igiyr, pp 288. 2S9. *Ihid., I9i8«, p. 274. 568 THE INSCRIPTIONS AT COPAN. clear that the custom began not later than the third lahuntun of Cycle 9 (Copan, Stela 24), and admitting them, as early as the first lahuntun of Cycle 9 (Tikal, Stela 8); in other words, we get our first traces of it some 140 or 160 years after the earliest stela known was erected (Uaxactun, Stela 9). On the basis of the evidence now available. Stela 8 at Tikal (9.0.10.0.0), Stelae 20, 24, 15, and 9 at Copan (9.1.10.0.0, 9.2.10.0.0, 9.4.10.0.0, and 9.6.10.0.0 respec- tively), it appears as though the lahuntun-endings were actually commemorated in this way before the katun-endings, although the latter is much the more important period of the two. This is so contrary to the logical order for this custom to have followed in its development, i. e., first the katuns, then the half-katuns (the lahuntuns), and finally the quarter-katuns (the hotuns), that even in spite of the above evidence to the contrary, the writer feels that at first monuments were only erected at the katun- endings; that is to say, as first worked out, possibly during the close of Cycle 8, at Tikal, the katun-endings {i. e., once every 20 years) were the only times at which stelae were erected. Later, as the Maya became more prosperous, in a second stage (the first at Copan), they were able to erect the stelae on lahuntun-endings as well (i. e., once every ten years), and finally, toward the close of the Early Period (9.8.15.0.0), they became sufficiently prosperous to be able to erect them on the first and third hotun-endmgs as well (i. e., once every 5 years). This surely is the logical order of development, and the fact that the only evidence available tends to indicate the priority of the lahuntun over the katun as the period first chosen for this purpose, the writer believes is due to the chance survival of these few earliest lahuntun-markers and the chance destruction or non-recovery of the earliest katun-markers, rather than to any actual priority of the former over the latter. Bowditch's conclusion (see page 565) that the quarter-katuns were probably marked from the first appears to the writer incorrect. He believes the evidence on this point at least is sufficient to show that the hotun-endings were not marked until some time after the katun and lahuntun-endings — in fact, until toward the close of the Early Period. The custom, moreover, appears to have varied in different cities at the same time, as the foUowmg brief synopsis of the dates at each will indicate: Tikal. — Even admitting the accuracy of the readings suggested here for Stelae 8 and 9, after 9.2.0.0.0 (Stela 9) we do not get another hotun-marker until 9.14.0.0.0, 240 years later (Stela 16), and after Stela 16 not another until 10.2.0.0.0, 160 years later (Stela 11); that is, out of the 17 sculptured stelae at this site, only 2, or possibly 4, are hotun-markers. But as already noted in Chapter V, there are 51 plain stelae at Tikal, which probably were painted, and possibly were period-ending markers, and if these were erected only at the lahuntun and katun-endings — and no first and third hotun-endings have yet been found here — they would largely fill this gap, otherwise inexplicable in such an important site as Tikal, no less than the largest city of the Old Empire, and indeed of the whole Maya civilization. Copan. — The situation here has been fully explained in Chapters II, III, and IV. The earliest stelae are lahuntun-markers and toward the close of the Early Period katun, first and third hotun-markers appear. There are several lacunae in the sequence of the monuments at Copan, which it has been shown were probably coincidental with corresponding periods of building activity and temple construction. Piedras Negras. — This city contains one of the two best series of hotun-markers known. Here for a period of nearly 2 centuries probably not a hotun-ending passed without the erection of a corresponding monument to commemorate the event. This is shown graphically in figure 83, where the map of the city appears with a chronological diagram below indicating the dates of the different monuments. THE HOTUN. 569 DATED MONUMENTS ■Bi STRUCTURES ^ "%. , ""^ \ '-^ 2 ARABIC MUMERALS, MONUMENTS ^ '^'f'^ .\ ''"^ '" H R0MAMNUMtRALS,5TRUCTURES ^ -^"^ ''n~ /// -'^ jn ,y tW«t\ , - ' « I. .,>»>■ '** .'""^/I -. Il* ' *"% Fig. 83. — Map of Piedras Negras, Guatemala, showing location of the dated and the undated monuments. each line corresponding to a hotun-ending. When there is a black square and a number in the first column, the number indicates the name of the stela, or in one or two cases altar, which records the corresponding hotun-ending. If the first column is empty this indicates that no monument hat yet been found recording this particular hotun-ending. In such cases, however, it should be noted that there are other monu- ments at the city, the dates of which have not yet been deci- phered, shown as white squares on the map in figure 83, which in all probability fill these lacunae. There are 37 stelae and 2 large altars at Piedras Negras, the dates of 23 of which, or about 60 per cens., have been deciphered; and since all of the deciphered monuments have been found to be hotun-markers, it appears highly probable that the remaining 16 when deci- phered will be found to fill some of these lacunae. The ear- liest date at Piedras Negras is 9.8.15.0.0 (Stela 25) and the latest 9.18.5.0.0 (Stela 12). Stylistically considered also. Stela 25 is one of the earliest, if not the very earliest monu- ment, and Stela 12 is surely the latest, so that it is reasonably safe to regard most, if not all, of the other monuments as fall- ing between these two limits. There are, moreover, just 39 ho- tuns between and including these two dates, and we have seen that there is a total of just 39 monuments known here. These results are so close and satisfactory that the writer has little hesi- tancy in concluding that the as yet undated monuments at this site will eventually be found to fill most if not all of these lacunae. Naranjo. — At this city there is evidence that at first only the katun and lahuntun-endings were marked, 9. lo.o.o.o (reused Monu- ments. 251 311 3S« 36b 37* 4" IB 3" 5" 7" 91 101 2'l 221 161 13" 141 121 Hotuns. Early Period. 9- 9- 9- 9- 9- 8. ij.o.o 9. 0.0.0 9. 5.0.0 9. 10. 0.0 9.15.0.0 9.10. 0.0.0 Middle Period. 9.10. 5.0.0 9. 10. 10. 0.0 9.10. 15.0.0 9.11. 0.0.0 5 10 II. 15.0.0 12. 0.0.0 12. 5.0.0 12. 10. 0.0 12. 15.0.0 13. 0.0.0 13. 5.0.0 9. II. 9. II. 9. II. 9.12. 9.12. 9 9. 9. 9- .0.0 .0.0 9.13.10.0.0 13.15.0,0 14. 0.0.0 14. 5.0.0 14. 10. 0.0 14. 15.0.0 15. 0.0.0 Great Period. 9.15. 5.0.0 9. 15. 10. 0.0 9. 15. 15.0.0 9.16. 0.0.0 9.16. 5.0.0 9. 16. 10.0.0 9.16.15.0.0 9.17. 0.0.0 9.17. 5.0.0 9. 17. 10. 0.0 9.17. 15.0.0 9.18. 0.0.0 9.18. 5.0.0 S70 THE INSCRIPTIONS AT COPAN. lintel in the Hieroglyphic Stairway) and 9.10.10.0.0 (the Hieroglyphic Stairway). In 9. 13. 15. 0.0 (Stela 21) and again in 9. 14. 15.0.0 (Stela 18), however, third hotun- endings were recorded, but this practice does not appear to have continued, as the closing group of stelae were all erected on lahuntun or katun-endings again, 9.1 7. 10. 0.0 (Stelae 13 and 19), 9.18.0.0.0 (Stela 14), 9.18.10.0.0 (Stelae 8 and 12), 9.19.0.0.0 (Stelae 7 and 10), and 9.19.10.0.0 (Stela 32). However, there are 32 sculptured stelae at this site, the dates of only about half of which have been deciphered, so that some of the undeciphered monuments probably fill the existing lacunae in the series of hotun-markers there. Palenque. — The evidence here is very limited. There is only one stela known and a few wall tablets. As far as they go, however, these conform with this prac- tice: 9. 10. 10. 0.0 (Temple of the Sun), 9. 11. 0.0.0 (Palace Group and a small slab of unknown provenance), and 9.13.0.0.0 (Temples of the Inscriptions and Foliated Cross, and the stela in front of the Temple of the Cross). ^ \ D DATED M0NUMef1T5 UNDATED MONUMENTS ma STRUCTURES CAPITAL LETTERS MONUMENTS ARABIC NUME,RAL5 STRUCTURES ScflL£ If Pr 6 iO ITO IM J06 +>3 Fig. 84. — Map of Qiiirigiia, Guatemala, showing location of the dated and the undated monuments. Yaxchilan. — The evidence at this site is the most unsatis- factory of all, though even here traces of the custom appear. Of the 20 stelae, only one records a hotun-ending, 9.16.10.0.0 (Stela i). An altar was dedicated in 9. 15. 15. 0.0, and two lin- tels in 9.16.5.0.0 (Lintel 3) and 9.17.0.0.0 (Lintel 31). Many of the dates here have not been deciphered, and many more, particularly on the lintels, which were the commonest media for hieroglyphic treatment at Yaxchilan, present odd dates at the ends of no periods in the Long Count, and thus possibly refer to historic or astronomic events. Nakum. — There are only 3 sculptured stelae at this site, but each one of these conforms to this custom: 9.17.0.0.0 (Stela U), 9. 19. 10. 0.0 (Stela C), and 10. i. 0.0.0 (Stela D). There are 12 other plain stelae here, and assuming that they were either lahuntun or katun-markers like the 3 sculptured ones, and that 10. 1. 0.0.0 is the latest date, these would carry the series of period-markers at this site back to 9.14.0.0.0 for its beginning. Monu- ments. Hotuns. M SB H» J" Fb Db Eb A&C B Bb Gb Ob Pb JB Kb I B 9 G 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 1 15. 0.0.0 ''eat Period. 15. 5.0.0 15 . 10. 0.0 15. 15.0.0 16. 0.0.0 16. 5.0.0 16. 10. 0.0 16. 15 .0.0 17. 0.0.0 17. 0.0.0 17. 10. 0.0 17.15.0.0 18. 0.0.0 18. 5.0.0 18. 10. 0.0 1 8 . 1 5 . . 19. 0.0.0 THE HOTUN. 57I Quirigua. — The series of hotun-markers at this site is the most satisfactory in the Old Empire, for there is not a single gap after the erection of the first stela in 9.1 5.1 5.0.0 (Stela S) down to and including the last hotun-ending recorded here, 9.19.0.0.0 (Structure i). This is shown graphically in figure 84, where a map of the city appears with a chronological diagram below giving the dates of the difi^'erent monuments after the same scheme as in figure 83. It will be noticed in figure 84 that there are no undated stelae at Quirigua, and only 4 undated altars, N, L, Q, and R, the latter being small and relatively unimportant. The earliest hotun- ending is 9.15.0.0.0 (Altar M) and the next 9. 15. 15. 0.0 (Stela S), after which the series continues without a single break for 65 years down to and including 9. 19.0.0.0 the last date at Quirigua (Structure l), one hotun, namely that ending in 9.17.5.0.0, being comm.emorated by two monuments (Stelae A and C). This is the most satisfactory sequence of all, so much so, in fact, that before the discovery of the last monument at Quirigua, Stela S, in 191 3, the writer had predicted that if another monument were found there it would record either ofthe hotun-endings 9. 15. 15. 0.0, or 9.19.0.0.0, i.e., the two open ends ofthe sequence at that time. Subsequently Mr. M. D. Landry, ofthe United Fruit Company, found Stela S a kilometer south- west of the main group, and this proved to record the date 9. 15. 15.0.0, one of the two which had been predicted for it. Uaxactun. — Of the 1 1 sculptured stelae known at this site, at least 3 record katun-endings, 9.14.0.0.0 (?) (Stela i), 9.16.0.0.0 (Stela 2), and 9.19.0.0.0 (Stela 7). There are other early monuments here, 8. 14. 10.13. 15 (Stela 9), 8. 15. 10. 3. 12 (?) (Stela 5), 9.3.13.0.0 (Stela 3), and 9.8.6. ?.3 (?) (Stela 6), however, which do not conform to this practice, and 4 others which are undecipherable (Stelae 4, 8, 10, and 11). Yaxha. — Only one monument has been dated at this site (Stela 6), and that very doubtfully as 9.1 1.5.0.0. La Honradez. — The 10 sculptured stelae here are very much effaced, owing to the inferior quality of the limestone upon which they were carved. Only one has been surely dated, but it records the katun-ending, 9.17.0.0.0 (Stela 7). Other doubtfully deciphered monuments here are 9. 17. 10. 0.0 (?) (Stela 6), 9.18.0.0.0 (?) (Stela 5), and 9.18.10.0.0 (?) (Stela 4), although there is not much doubt as to the dates of the last two. The latest monument at this site on stylistic grounds is Stela 4, and assuming that the 6 as yet undeciphered stelae recorded lahuntun or katun-endings prior thereto — and no first or third hotun-endings have been found here — these would carry the series of monuments back to 9.14.0.0.0. Seihal. — The custom at this sice appears to have been to commemorate only the lahuntun and katun-endings, no first or third hotun-endings having been found here. The following dates have been deciphered: 9.16.0.0.0 (Hieroglyphic Step), 9.17.0.0.0 (Stela 6), 9.18.0.0.0 (Stela 12), 9. 18. 10. 0.0 (Stela 7), lo.i.o.o.o (Stelae 8, 9, 10, 11), and 10.2.0.0.0 (Stela i). The dates of Stelae 2, 3, 4, and 5 have not been deciphered yet; they may have marked some ofthe missing lahuntun-endings. Ixkun. — Only one monument has been surely dated at this site, 9.18.0.0.0 (Stela i), although the lahuntun-ending 9. 18. 10. 0.0 is probably recorded on Stela 5. There is a third monument here (Stela 2), none of the dates of which, although per- fectly legible, appears to conform to this custom. Itsivite. — The dates of 4 monuments have been deciphered at this site, 2 surely and 2 doubtfully; all 4, however, conform to this custom, 9.14.0.0.0 (?) (Altar i), 9.14.10.0.0 (?) (Altar 2), 9.15.0.0.0 (Stela 5), and 9. 15. 10. 0.0 (Stela 2). Altar de Sacrificios — El Pabcllon. — Of the 4 decipherable monuments at these two sites, which are so near each other as to be really parts of the same center, 3 record katun or lahuntun-endings, 9.10.0.0.0 (El Pabellon, Stela i), 9. 10. 10. 0.0 (Altar de Sacrificios, Stela 4), and 9.14.0.0.0.0 (Altar de Sacrificios, Stela 7). 572 THE INSCRIPTIONS AT COPAN. The remaining sites each have one or two datable monuments only, but in every case these are found to conform with this practice. Tzendales. El Cayo. Los Higos. La Mar. Aguas Callentes. Cancuen. UcanaL Benque Viejo. Flores. Quen Santo. Chichen Itza. g.13. 0.0.0 (Temple of the tablet) 9.17. 5.0.0 (?) (Stela 1). 9.17.10.0.0 (Stela i). 9.17.15.0.0 (Stela l), 9.18. 0.0.0 (Stela i). 9.18. 0.0.0 (Stela 2), 10. I. 0.0.0 (Stela 3). 10. I. 0.0.0 (Stela i). 10. I. 0.0.0 (Stela l), 10. 2. 5.0.0 (Stela l), 9. 18. 15.0.0 (Stela 2). 9.18. 5.0.0 (Altar 2). 10. 2. 0.0.0 (Stela 2). 10. 2.10.0.0 (Stela 2). 10. 2.10.0.0 (Temple of the Initial Series). 86 000 Fig. 85. — Design and part of inscription on front of Stela i at Tuluum, Yucatan. Fig. 86. — Design and inscription on clitf at the Hunacab month of the Cave of Loltun, Yucatan. Fig. 87. — Inscription on front of lintel in the Temple of the Initial Series at Chichen Itza, Yucatan. It is apparent from the foregoing summary that this custom was practiced throughout the Old Empire, even those small sites which have only one or two THE HOTUN. 573 monuments conforming to it. In some places, notably at Piedras Negras and Quirigua, the custom of marking every hotun-ending by the erection of a stela was rigidly adhered to, but more often only the lahuntun and katun-endings were thus commemorated, and in some places the custom even seems to have changed from time to time, as at Naranjo, for example. Turning next to the New Empire, we find not only archaeologic evidence of the prevalence of this custom, both on the monuments and in the codices, but also direct documentary proof thereof in the post-conquest sources, Spanish as well as native. Mention has been made elsewhere that the contemporaneous date of the Chichen Itza lintel was originally assumed to have been declared by the Initial Series 10. 2. 9. 1.9 on its under side, but that in 1918 the writer was able to show that it was declared by the Period Ending date on the front instead, which records the lahuntun-ending 10.2.10.0.0 2 Ahau 13 Chen.^ This is shown in figure 87, where the inscription on the front of this lintel is represented. The first glyph, a1, is the lahuntun-sign, here shown in a very unusual manner. Instead of the regular lahuntun-sign (see figure 82), Al is composed of the "winged-Cauac" variant of the tun-sign ivith a coefficient of 10, making the whole glyph 10 tuns or a lahuntun. The last glyph on the front, 12, declares the day on which this lahuntun ended, namely, 2 Ahau, and the glyph just above, il, the day of the Initial Series terminal date, 9 Muluc, 331 days earlier. Al, i2, therefore, may be interpreted as "Tun 10, ending on the day 2 Ahau," which, in conjunction with the Initial Series on the under side, indicates that the lahuntun 2 Ahau here intended was 10.2. 10. 0.0 2 Ahau 13 Chen, thus conforming to the regular practice. The Chichen Itza lintel is the earliest hotun-marker known in the New Empire. The next occurrence on a New Empire monument is Stela i at Tuluum (see figure 85). The contemporaneous date of this monument also was first thought to be declared by the Initial Series, which Howe correctly deciphered as 9.6.10.0.0 8 Ahau 13 Pax.- In 1918, however, the writer found a missing fragment of this stela, giving a later Period Ending date, which, with other glyphs on the back (see figure 88, a and b), indicates that its contemporaneous date was the lahuntun- ending 10.6. 10. 0.0 7 Ahau 18 Yaxkin instead, exactly i cycle later than its Initial Series, and about 80 years later than the Chichen Itza lintel.'' (Note the lahuntun- sign in figure 85 at A15, and in 88, b, at a2, and the day 7 Ahau in figure 85 at A16, in 88, a, at a2, and in 88, b, at Al.) Thus the Tuluum stela is also seen to conform with this practice. We have already seen that Initial Series dating fell into disuse in the New Empire, being replaced by Period Ending dating. So it is not surprising to find our next example from the Cave of Loltun near the modern village of Oxkutzcab, northern Yucatan, recorded as a Period Ending date, a Katun 3 Ahau. On a vertical wall at the Hunacab entrance of this cave there is sculptured a large deity, ruler, priest, or warrior, 3 meters high, with a ceremonial staff in his right hand and a club (perhaps an atlatl or spear-thrower) in his left hand. (See figure 86.) Above and to the left is a panel of 3 glyph-blocks, the first of which is the day 3 Ahau. The whole composition is doubtless to be interpreted as the ending of some Katun 3 Ahau, the anthropomorphic figure being the deity or ruler who presided over the particular katun recorded. Turning to the u kahlay katunob on page 503 we will see that in all probability this Katun 3 Ahau was the one which ended in 1379 a. d., that is, the only one after the end of the League of Mayapan, corresponding to the Initial Series 1 2.1. 0.0.0 'Morley, 1918a, p. 274. 'Howe, 191 1, p. 546. ^Modey, op. cit., pp. 274, 275. See also Morley, 1916a, pp. 338, 339, and 1917, pp. 190, 192, 193, 201, 202. rjA THE INSCRIPTIONS AT COI'AN. 3 Ahau i8 Kayab. This is so because the only other three Katuns 3 Ahau in the ukahlay katunob subsequent to the discovery of Chichen Itza (10.2.0.0.0 3 Ahau 3 Ceh, 610 A. D., 10. 15. 0.0.0 3 Ahau 18 Pop, 866 a. d., and 11. 8.0. 0.0 3 Ahau 18 Chen, 1122 A. D.) are too early to be probable on historic grounds, as indicated by the following evidence. The atlatl, which the deity or ruler in figure 86 holds in his left hand, is a Nahuatl weapon, and the Nahuatl influence did not become strong in Yucatan until after 11. 12.0.0.0 8 Ahau 3 Mol (1201 a. d.), when, owing to the victory of Hunnac Ceel, the halach vinic of Mayapan, over the Itza in that katun, Chichen Itaa seems to have been turned over to the Nahua troops, who had aided Hunnac Ceel in its conquest. The last Katun 3 Ahau before the Spanish Conquest i.e., that ending in 1379, therefore is probably the one intended here. P 0^ Fig. 88. — Parts of inscription on back of Stela I at Tuluura, Yucatan. Fig. 89. — Design and part of inscription on front of Stela I at Ichmul, Yucatan (top only recovered). A third example of doubtful date is Stela i at Ichmul (see fig. 89). This is only the upper part of the monument, the lower part having disappeared. It is now built into the wall of a house on the Hacienda of Ichmul, some 40 kilometers east of north from Chichen Itza. Two human figures hold in their right and left arms respectively a large central glyph-block, which records the day 6 or 7 Ahau, probably the latter. Across the top of the monument runs a row of glyph-blocks, of which originally there would seem to have been 24. (1 he left side of the stela is missing, see fig. 89.) Unfortunately, owing to the uncertainty as to whether the day 6 Ahau, or as the writer is inclined to believe, 7 Ahau, is recorded here, it is impossible to date this monument exactly. The following readings are the best possibihties, however: II. 13.0.0.0 6 Ahau 3 Zip (1221), II. 19.0.0.0 7 Ahau 13 Chen (1339), and 12.6.0.0.0 6 Ahau 3 Zac (1477), with the second, 11. 19. 0.0.0, as the best of all. Another New Empire katun-marker is Stela 9 at Mayapan, the former capital of the Cocom, and during the fourteenth and early fifteenth centuries the leading city in Yucatan. This monument (see fig. 90, a) was undoubtedly one of those described by Landa as having been discovered by him "in the plaza of that city."' It was rediscovered by the Abbe Brasseur de Bourbourg in 1866, three centuries later,- and was subsequently removed to the casa principal of the Hacienda of Xcanchakanon the lands of which the ruins of Mayapan lie, and there built into a wall in the corridor, where it is excellently preserved. Again we see two figures facing each other, the right-hand one apparently Itzamna, God D of the Schellhas classification, the day-sign 10 Ahau in front of a baton he holds in his right hand.^ This is clearly to be interpreted as Katun 10 'For this whole passage, see page 577. ^grasseur de Bourbourg, 1867, pp. 246-249, and figure 4. ^The details of this day-sign as well as those of the other glyphs on this stela were painted, not carved, and have consequently disappeared; the coefficient 10, however, is unusually clear. THE HOTUN. 575 Ahau, and this monument is one of those very "stones" wliich Landa says the natives told him they were accustomed to erect "every 20 years, which is the num- ber they use for counting their ages." Referring to the u kahlay katunob on page 503, we will find that there are only two Katuns 10 Ahau to which this monument could possibly have belonged, namely, ii.ii. 0.0.0 10 Ahau 3 Mac (1182) and 12.4.0.0.0 10 Ahau 18 Uo (1438). The writer inclines to the latter of these, as the carving on Stela 9 is excellently preserved, and the more recent date is therefore the better of the two. Fig. 90. — a. Principal part of design on front of Stela 9 at Mayapan, Yucatan, b. Part of middle section of page ii of the Codex Peresianiis. If 12.4.0.0.0 10 Ahau 18 Uo was the Initial Series corresponding to this Katun 10 Ahau, as the writer believes, then Stela 9 was the last monument to be erected at Mayapan, because before the next one fell due, i. e., at the end of the next katun, Katun 8 Ahau (1458), the city was destroyed by a league of Maya chieftains under the leadership of Tutul Xiu, halach vinic of Uxmal, and the Cocom forced to seek new homes elsewhere. Coming next to the codices, the writer has stated in Chapter I (see page 43) that the middle sections of pages 2 to 11 of the Codex Peresianus appear to record an u kahlay katunob. Here we can see a series of 10 pictures, each having two anthropomorphic figures facing each other, just as in figure 89, and more particu- larly in figure 90, a, and similarly between the two figures in each picture there is a day Ahau, decreasing from left to right by 2 in each picture, beginning with 2 Ahau on page 2 and continuing as follows: 13 Ahau (p. 3), 11 Ahau (p. 4), 9 Ahau (p. 5), 7 Ahau (p. 6), 5 Ahau (p. 7), 3 Ahau (p. 8), i Ahau (p. 9), 12 Ahau (p. 10), and 10 Ahau (p. 11), possibly parts of an 11 kahlay katunob. The picture on page 1 1 of this codex is shown in figure go, b, where we see two anthropomorphic figures facing each other. Here the left-hand one is Itzamna, God D; the right-hand and smaller one, sitting on a platform, is badly effaced. Itzamna offers in his hand the head of Kukulcan, the Long-nose God (B of the Schellhas classification), and just in front of this head is again the day 10 Ahau. The writer believes this picture shows the same Katun 10 Ahau as the one on Stela 9 at Mayapan, and it should be noted that the same deity, God D, is the principal figure in each tableau. If this identification is correct, pages 2 to 1 1 of the ^76 THE INSCRIPTIONS AT COPAN. Codex Peresianus cover the period from 1241 (beginning of Katun 2 Ahau) to 1438 (end of Katun 10 Ahau). Unfortunately, it appears more probable from the nature of the pictures in this section of the codex that they refer to rites and ceremonies which took place at the ends or installations of these katuns rather than to his- torical events which occurred in the corresponding periods. Always we have the same principal elements — a large figure on the left, apparently a deity, offering the head of Kukulcan to a smaller figure seated on a platform or dais. Above there hovers a bird, which differs in the different pictures, and below we see a vessel holding corn, i. e., the sign for the day Kan. The similarity of these 10 pictures strongly suggests that they refer to certain rites and ceremonies which took place when one katun ended and another began, ^ and the large figure present in each is probably to be interpreted as the deity presiding over the corresponding katun. In a word, it appears probable that these pages of the Codex Peresianus treat of ritualistic rather than historical matters. Up to this point all the evidence examined has been drawn exclusively from archaeological sources; let us next turn to the documentary evidence touching upon the prevalence of this custom. So explicit is Nakuk Pech upon this point that a quotation from his chronicle, already previously cited, will bear repetition here: "In this year [1517] the katun ended, and then ended the putting in place of the town- stone, for at each twentieth stone they came to place the town-stones, formerly, when the Spaniards had not yet come to Cuzamil, to this land; since the Spaniards came, it has ceased to be done."^ The entries in the second chronicle from the Book of Chilan Balam of Chu- mayel bearing upon this point tend to indicate that just prior to the time of the Spanish Conquest at least, several towns combined to celebrate this festival, the town-stone (m tunil balcah) first being put up in one town, and that for the next katun, in another, and so on: [Katun] 12 Ahau: the stone of Otzmal was taken. [Katun] ID Ahau: the stone of Zizal' was taken. [Katun] 8 Ahau: the stone of Kancaba was taken. [Katun] 6 Ahau: the stone of Hunnacthi was taken. [Katun] 4 Ahau: the stone of Ahtiku was taken; in this katun took place the pestilence, in the fifth tun of Katun 4 Ahau. [Katun] 2 Ahau: the stone of Chacalna was taken. [Katun] 13 Ahau: the stone of Euan was taken. 'Says Lancia in describing this ceremony: "If the Spaniards had not been here they would have adored the idol of II Ahau until the year of [ijlji, which are lo years and at the loth year they W'ould have set up another idol, 9 Ahau, and they would have honored it, guiding themselves by the prognostications of 1 1 Ahau [from the year of 1541] until the year of [15)61, and then taking it away from the temple and putting there the idol of 7 Ahau [surely a mistake for 9 Ahau, see following] and guiding themselves by the prognostications of 9 Ahau for another 10 years [apparently until 1571], and thus they made a complete round; in this manner they venerated their katuns for 20 years [each], and [for ?] 10 they arrange their superstitions and tricks, which are so many and of such a sufficiency as to deceive simple people, who look not at things as do those who are versed in the affairs of nature, and the dealings which the demon has with them." (Landa, 18S1, p. 103.) In this highly confused and confusing passage it is hard to make out just what method of procedure Landa is trying to describe, introduc- ing as he does the idols for three katuns, 11 Ahau, 9 Ahau, and 7 Ahau, in a period of 30 years (1541 to 1571). The only natural assumption is that each deity presided over a full katun, reigning from the end of the preceding katun to the end of its own, i. e., 20 years. This introduction of a half-katun might, however, be construed as indi- cating that the idol of a specific katun was not placed in the temple until the katun over which it presided was half completed, and that it was not taken therefrom until the first half of the next katun had passed; but this seems to be going far out of the way indeed, to help a passage which is beyond help ^Brinton, 1882, p. 227. ^Possibly the modern Sisal, on the northern coast of Yucatan, the principal port during the Spanish Period. THE HOTUN. 577 [Katun] II Ahau: in the time of its beginning the stone of Coloxpeten was taken; in this katun died the watei-hringer Napotxiu, in the first tun of [Katun] ii Ahau; it was also in this katun that the Spaniards first arrived here in this land, in the seventh year of Katun ii Ahau; also Christianity began in the year of fifteen hundred and nineteen, the year of our Lord 15 19. [Katun] 9 Ahau: no stone was taken at this time; in this katun first came the bishop Brother Francisco Toral; he arrived in the sixth tun of Katun 9 Ahau.' The idea conveyed by this passage is that a group of towns possibly joined in putting up the same katun-marker, first at one of their number and then at another; thus both this and the Nakuk Pech quotation clearly refer to the erection of monu- ments at the katun-endings. The Spanish authorities are no less explicit, Bishops Landa and Cogolludo both making unmistakable reference to the same custom. Says Landa in this connection: "There was discovered in the plaza of that city [Mayapan] seven or eight stones, each 10 feet in length, round at the end, and well worked. These had some writings in the char- acters which they use, but were so worn by water that they could not be traced. More- over, they think them to be in memory of the foundation and destruction of that city. There are other similar ones, although higher, at Zilan,- one of the coast-towns. The natives when asked what these things were, replied that they were accustomed to erect one of these stones every twenty j'ears, which is the number they use in counting their ages."' And Cogolludo has the following: "Their lustras having reached five in number, which made 20 years, which they call a katun, they place a graven stone on another of the same kind laid in lime and sand in the walls of their temples and the houses of the priests, as one still sees to-day in the edifices in question, and in some ancient walls of our own convent at Merida, about which there are some cells. In a city named Tixhualatun, which signifies 'place where one graven stone is placed upon another,' they say are their archives, where everybody had recourse for events of all kinds, as we do to Simancas."^ The foregoing quotations, the writer believes, leave little room for doubt but that the practice of erecting monuments at the ends of successive katuns of the IVIaya chronological era persisted down to the time of the Spanish Conquest, and, what is even more important, they corroborate and explain the archseologic evi- dence and constitute nothing less than direct documentary proof of the former existence of this custom. Reviewing IVIaya history from its beginnings, we may conclude that although this custom was an early development of their civilization, there was a time when monuments and smaller objects were not dedicated at the ends of even periods of their chronological era. 'Brinton, 1882, pp. 171, 172. ^In February 1918, the writer found here parts of two stelre, which may have been the very ones referred to by Landa in the above passage, another instance of the rehabihty and trustworthiness of his statements. Only the lower part of Stela l has been recovered. It presents a band of 5 flyphs at the bottom, the first 2 of which record the Calendar Round date 7 Muluc 2 Kayab. Above is a crouching human figure upon whose back stands the prin- cipal figure; the break occurs at the knees of the latter, and the rest of the stela is still missing. The fragment recovered is built into a back wall of the cabildo, on the southern side of the plaza, and some local artist has mod- eled in stucco the missing parts of the legs, torso, head, and arms, reconstructing the figure as that of a Mexi- can or German (?) soldier, helmet on head, and gun, with fixed bayonet, in hand. Stela 2 is built into a wall on the north side of the church-yard. It is much more effaced than Stela i and all that can be distinguished is a standing human figure with elaborately plumed head-dress, and 7 glyph-blocks, 4 before and 3 behind; al! are too effaced to decipher. Both these stela: seem to have been found in excavating a platform which runs in front of a high, long mound, just west of and facing the church. Stela i is said to have been built into the back wall of the cabildo in 1900. ^Landa, 1881, p. 75. ^Cogolludo, 1688, p. 1S6. ^yg THE INSCRIPTIONS AT COPAN. From all the evidence now available it appears probable that the Maya chronological system was devised some time before 8.6.2.4.17, although not neces- sarily long prior thereto, to keep account of periods longer than the Calendar Round, i. e., 52 j^ears of 365 days each. For this purpose a remarkable vigesimal mathematical system, including numeration by position, a fixed hypothetical starting-point, and an ingenious arithmetical notation of bars and dots and several specialized characters for zero, was invented, but underlying this, and of a still earlier date, there were (i) a sacred or divinitory year composed of 20 names com- bined with 13 numerals, making 260 days, and (2) a solar year fixed at 365 days in length, i.e., composed of 18 months of 20 days each and a closing period of 5 days, as the two basic elements of Maya chronology. At first this extraordinarily accurate chronology was utilized for the record of any date, regardless of when it occurred, as evidenced by the casuality of the earhest dates, 8.6.2.4.17, 8. 14.3. 1. 12, 8. 14. 10.13. 15, and possibly 8. 15. 10. 3. 12, but very early, perhaps within a century of the last one of these, i. e., about the begin- ning of Cycle 9, an exceedingly important change was introduced, no less than the practice of restricting the erection, of the larger monuments at least, to the ends of even periods in their chronological era. Several fairly obvious factors must have contributed largely to the origin and development of this practice. In the first place, to erect a monument or dedicate a smaller object after the event which it was to commemorate, was to violate the whole Mayan conception of time. The Maya conceived, measured, and recorded time in terms of elapsed units, which kept them continually looking forward into the future for their dates of ceremonial importance, i. e., those which would close their time-periods. This of itself would tend to restrict the erection and dedica- tion of monuments to dates determined in advance by the passage of successive units of their chronological system, rather than to encourage the memorialization of early events at fortuitously chosen later times. Another very practical advantage to which this custom gave rise was the ample opportunity it afforded the priests to make the necessary arrangements for these important occasions, which the writer has pointed out in Chapter V required considerable forethought, elaborate planning, and coordination of activities in the not inconsiderable mechanical labor involved in quarrying, transporting, and erecting the monuments, to say nothing of the artistic efforts involved in their sculpture. A third factor was that of economic necessity. The work of many kinds, required to bring a monument to the point of completion must have withdrawn from purely economic production many laborers at not infrequent intervals, and as the period-endings in a very brief time shift all around the cycle of the seasonal year, it is obvious that careful planning was necessary to prevent these religio- esthetic activities from interfering with the m.ore urgent need of producing corn and cotton, i. e., food and clothing, in sufficient quantities to meet the requirements of the community. The latter demands were imperative. The beginning ot the dry season fixed the time when the bush had to be felled, the beginning of the rainy season when the fields had to be planted. These labors brooked no delay, and other activities, religious, architectural, etc., had to be made to conform to the exigencies of agriculture. Thus the selection, a long time in advance, 5, 10, and 20 years, of the dates upon which their monuments were to be erected, gave the Maya ample time to prepare for these events, without interfering with the production of those food and clothes-stuffs upon which the very life of the community depended. And so, about the beginning of Cycle 9, the custom of erecting their monuments only on THE HOTUN. 579 period-endings was inaugurated, and the original custom of erecting them after the events they memoriaHzed had taken place, as evidenced by the casuality of the dates of the earliest monuments as opposed to the periodicity of the dates of the great majority of the later ones, was discontinued. It must not be concluded that important dates, not coinciding with period- endings, were never recorded. On the contrary, many such were, as for example the important date 9. 16.12. 5. 17 6 Caban 10 Mol at Copan, 9. 14. 13.4. 17 12 Caban 5 Kayab and 9. 15.6. 14. 6 6 Cimi 4 Tzec at Quirigua, 9.1 1. 12. 7. 2 2 Ik 10 Pax and 9.13.9. 14.15 7 Men 18 Kankin at Piedras Negras; 9.15. 10. 17. 14 6 Ix 12 Yaxkin at Yaxchilan, and 9.12. 10.5. 12 4 Eb 10 Yax, 9.12. 15. 13.7 9 Manik o Kayab, and 9.14.I.3.19 3 Cauac 2 Pop at Naranjo. But the important point in all these cases is that these casual dates (with the exception of 9. 16. 12. 5. 17) are never the con- temporaneous dates of the monuments upon which they are recorded, but are always prior thereto, the contemporaneous date in each case being a subsequent period-ending. Thus, for example, at Quirigua four different monuments begin with the date 9.14.13.4.17 12 Caban 5 Kayab, but Secondary Series in each case bring forward the beginning dates to subsequent period-endings which are in each case the corresponding contemporaneous date of the monument, i.e., Stela J, 9.16.5.0.0; Stela F, 9. 16. 10. 0.0, Stela E, 9.17.0.0.0, and Zoomorph G, 9. 17. 15.0.0. By beginning their inscriptions with such dates as these, or reaching them by Secondary Series elsewhere in the texts, the Maya were able to record events of astronomical or historical importance to them, but upon monuments which were erected at fixed predetermined periods. The period-markers were in efi:"ect, 5, 10, or 20 year almanacs issued at the ends of these respective periods, which covered important matters that had come to pass therein, or even earlier. The unit of their chronological system first selected for this important purpose was probably the katun, or 20-year period, in spite of the fact that all our earliest examples are of lahuntuns, but as the Maya waxed in strength, wealth, and pros- perity they w-ere able to erect monuments more frequently, and presently this interval was cut in half, and the lahuntun-endings were similarly commemorated. Still later, as we have seen, this interval was again cut in half at some cities, and every hotun or 5-year period marked by the erection of a corresponding monu- ment. And thus the matter continued throughout the Old Empire. As the suc- ceeding hotun-endings came around, all over that considerable region from Palen- que and Ocosingo in the west, across the valleys of the Usumacinta, Lacantun, Pasion, and Motagua Rivers, and the rolling plains and interior drainage area of northern Peten toQuirigua and Copan in the southeast (see plate i), on the same day monuments were being dedicated, with elaborate and impressive ceremonies we may feel sure, at all the occupied cities. Especially was this true of every fourth hotun, the katun-endings which from first to last throughout both the Old and the New Empires continued to be of greater importance than first, second, or third hotun-endings. For proof of this see the several maxima in figure 70 at 9. 1 1 .0.0.0, 9.13.0.0.0, 9.14.0.0.0, 9.15.0.0.0, 9.16.0.0.0, 9.17.0.0.0 and 9.18.0.0.0. At first the hotun-endings were marked by stelae exclusively, but later, as at Quirigua for example, low boulderlike stones were used, the so-called zoomorphs, and elsewhere even small altars, and finally, toward the end of the Old Empire, when the Maya were truly at their cultural zenith, their esthetic and intellectual apogee, whole temples were dedicated, especially at the katun-endings.^ 'Examples of this are the dedication of Temple 10 at Piedras Negras in 9.12.0.0.0, of theTemples of tht Cross, Foliated Cross, and Inscriptions at Palenque in 9.13.0.0.0, of Temple 21a and the Reviewing-Stand in the Western Court at Copan, and of Temple 10 at Yaxchilan in 9.17.0.0.0, of Temple 17 at Naranjo in 9. iS. 0.0.0, and of Temple I at Quirigua in 9.19.0.0.0. ego THE INSCRIPTIONS AT COPAN. At the very end of the Old Empire there is an obvious return to the first prac- tice of commemorating only the katun-endings. No first, second, or third hotun- endings have been found at any of the Old Empire cities after 9.19.10.0.0, and the few that seem to have been occupied during the first two katuns of Cycle 10, the little group in the northeastern corner of Peten, Flores, Tikal, Nakum, Ucanul, and Benque Viejo, and Seibal in central Peten, only commemorated the katun- endings lo.i.o.o.o and 10.2.0.0.0. After the extinction of the Old Empire civilization, Maya culture poured out of the Old Empire region in two directions, south into the highlands of Guatemala and north into the plains of Yucatan. The custom we are examining just managed to reach the northern edge of the first region (Stelae i and 2 at Quen Santo, see plate i), but for some unknown reason could not, or at least did not, survive there. In the north, however, this practice fared better, and we have two lahuntun-end- in^s declared on Initial Series monuments within the first century after the close of the Old Empire {i. e., the Chichen Itza lintel and Stela i at Tuluum). But coin- cident with the collapse of Initial Series dating in the New Empire, it would seem to have become restricted in its application principally, although probably not exclusively, at least at first, ' to the katun-endings only. As has been shown in the foregoing pages, we have archaeologic proof of this in such monuments as Stela 9 at Mayapan, Stela i at Ichmul and the Cave of Loltun, as well as preconquest manuscript evidence thereof in the Codex Peresianus, and direct contemporary documentary corroboration therefor in the accounts of Nakuk Pech, Juan Josef Hoil (the redactor of the Chumayel chronicles), Landa, Cogolludo, and even Villagutierre Sotomayor, and the indirect evidence furnished by the pas- sages quoted from the u kahlay katunob in the Books of Chilan Balam. In closing this description of the hotun, it is perhaps well to point out that monuments were not erected exclusively at these times, as, for example. Stela 11 at Yaxchilan, which bears the date 9. 16. i. 0.0, and Stela 6 at Naranjo, erected in 9.17.1.0.0; but that the very great majority were can not be gainsaid. It was one of the most important facts of ancient Maya life. It must have influenced and absorbed the activities of a great part of the population of every Maya city; and finally it must have determined not only the times for some of their most important feasts and ceremonies, but also profoundly afl^ected their very religion as well. It was, in fine, the most fundamental phase of their religion and life of which we have archaeologic evidence, and the persistence with which it prevailed, practically unchanged for more than 15 centuries (i. c, down to 1697), is ample proof that it was also one of the most deep-rooted customs of their own times. 'In addition to the lahiintun-endings on the Chichen Itza Hntel and Stela i at Tuluum, the writer found a Tun II ending on a column in the Temple of the High Priests' grave at Chichen Itza, 11.19.11.0.0 2 Ahau 18 Xul (see figure 76), and a Tun 13 ending in the Temple of the Initial Series at Holactun, either 10.9.13.0.0 2 Ahau 8 Yax or 11. 2. 13. 0.0 2 Ahau 8 Cumhu, and possibly others. APPENDIX VIII. LIST OF MONUMENTS MARKING THE HOTUN-ENDINGS DURING THE OLD EMPIRE. This Appendix contains a list of all known hotun-markers, including those at Copan; it is, in short, a concordance of all the examples of this type of Maya monu- ment now known, and has been frequently cited in the synoptic headings in Chapters II, III, and IV for purposes of comparison. The writer believes it to be as complete as it is now possible to make it, although future investigation, particularly explora- tion in the two regions mentioned on page 439, i. e., the eastern part of the State of Chiapas, Mexico, and the northern part of the Department of Peten, Guatemala, will doubtless greatly amplify it. Such exploration will certainly lead to the dis- covery of additional inscriptions from the Middle and Great Periods, while exca- vation, particularly at Uaxactun, Tikal, and other sites in the immediate vicinity, may be as confidently depended upon to yield additional texts, although fewer in number, dating from the Early Period. The writer finds 165 different inscriptions on stelae, altars, lintels, hieroglyphic steps and stairways, piers, tablets, and cornices which may now be included in this list of hotun-markers. Of these, 134 are surely deciphered as given, 21 prob- ably so, and 10 very doubtfully so. Possibly as high as 1 50, and certainly not lower than 140, may be accepted as correctly deciphered. Of these 165 texts, the con- temporaneous dates are declared by Initial Series in 72 cases (44 per cent.) and by Period Endings in 93 cases (56 per cent.). The data pertaining to these texts are presented as follows: The successive hotun-endings from 9.0.0.0.0 to 10.6.10.0.0 appear in the first column. In the second column are given the names of the sites at which the corresponding monu- ments in the third column are found. The nature of the contemporaneous dates, i. e., whether they are expressed by Initial Series or Period Endings, is given in the fourth column. When there are no monuments dating from any given hotun- ending, the last three columns for that particular hotun have no entries, as, for example, the first two hotun-endings in the list. In some cases hotun-endings are recorded as secondary dates on monuments, but not as their contemporaneous dates, as for example, 9.0.0.0.0, the first hotun- ending in the list, which appears on several monuments, but nowhere as a contem- poraneous date. It is obvious that no monuments can be referred to any given hotun-ending on such a basis, but in these cases a note indicates on what monu- ments such secondary non-contemporaneous dates have been found. Dates probably, but not surely, correctly deciphered are indicated by a single interrogation point, thus (?), after the name of the corresponding monument in the third column, and dates very doubtfully deciphered, by the double interrogation point, thus (??). The natural growth and expansion of the Old Empire civilization is clearly indicated in the following table. At first the hotun-markers are few and scattering, and, in fact, so continue until the close of the Early Period. Beginning with the Middle Period, however, they become more frequent and regular, increasing in these respects until the height of the Great Period, when the katun ending on 9.18.0.0.0 was commemorated at no less than nine different cities. After this latter date they decrease in frequency, ceasing temporarily between 9. 19. 15.0.0 and lo.o. 15.0.0 inclusive, and finally (so far as the Old Empire was concerned) after 10.2.0.0.0. 581 S82 THE INSCRIPTIONS AT COPAN. List of Hotun-markers during the Old Empire. Early Period. Date. Site. Monument. Kind of date. 9.0.10.0.0^ Tikal Stela S (?) Period Ending. 9.1.10.0.0^ Stela 20 (?) Initial Series. 9.2. 0.0.0 Tilcal Stela 9 (?) Period Ending. 9.2. 5.U.U 9.2.10.0.0^ Stela 24 Initial Series 9.3. 5.U.U 9.4.10.0.0 9-4-iS-O-o Initial Series. Period Ending. Period Ending. Altar Q' Copan Stela 16 (??) 9'S- 5-0.0 PiecJras Negras Stela 2q (?') Initial Series. 9.5.15.0.0 9.6. 0.0.0 Stela 17 (?) Initial Series. g. 5.10.0.0' Initial Series. 9.7. 0.0.0' Stela 18 (?) Initial Series. 9.7.15.0.0 1 1 iThis important date, the beginning of the cycle during which the Maya reached their first great cultural efflorescence, is recorded, although not as a contemporaneous date, at several places: here at Copan on Stelre 15 and 3 for example (see pp. 88, 89, and 157); again on the tablet in the Temple of the Cross at Palenque (see Motley, 1915, p. 228); and finally as the opening entry in the u kahlay Izatunob from the Books of Chdan Balam of Mani and Tizimin, according to the writer's correlation of Old and New Empire chronology (see Appendix II, P- 503)- , ,. , -This is possibly the oldest contemporaneous hotun-endmg known. 3This date is recorded by the Initial Series on the west side of Stela C at Quirigua, the contemporaneous date of which is shown by a Period Ending date to have been about 325 years later, i. e., in 9.17.5.0.0. (See Morley, 1915, pp. 173-175, 179, note I, 226.) "■This is probably the earliest contemporaneous date at Copan. 'This is the earliest surely deciphered hotun, lahuntun, or katun-ending known. <50n the tablet in the Temple of the Inscriptions at Palenque, the 10 katun-endings from 9.4.0.0.0 to 9.13.0.0.0 inclusive are recorded, the first as an Initial Series, the others as Period Ending dates, the last being the contem- poraneous date of the temple. (Sec Morley, 1915, p. 84, note I.) 'This date is also recorded by the Initial Series on Stela I at Tuluum. The contemporaneous date of this monument, however, 10.6.10.0.0, just one cycle later, is expressed by a Period Ending date. (See Morley, 1916(2, PP- 338' 339; ibid., 1918(7, pp. 274, 275.) «This date also appears as a tun-ending on a square altar at Chinikiha (plate i), although not as its contem- poraneous date, the month-sign being the unusual dog-head variant of the month Kankin fitst deciphered by the writer in 1914, and the tun-sign "the winged-Cauac" variant, deciphered by the writer in 1918. (See Maler, 1901, plate 2, A-c.) "This date is recorded as a Period Ending, although not the contemporaneous date, of the tablet in the Temple of the Inscriptions at Palenque. (See Maudslay, 1889-1902, vol. IV, plate 57.) LIST OF MONUMENTS MARKING THE HOTUN-ENDINGS. List of Hotiin-markers during the Old Empire — continued. 583 Early Period — continued. Date. Site. Monument. Kind of date. 9.8.15.0.0' 9.9. 0.0.0 9-9- S-o.o 9.9.10.0.0 9-9-I5-0-0 9.10.0.0.0- Piedras Negras. Copan Copan Copan Stela 25. Stela 7. . Stela E. , Stela P.. Period Ending. Initial Series. Period Ending. Initial Series. Naranjo. . . . El Pabellon. Reused lintel in Hier. Stairway. Stela I Period Ending. Initial Series. Middle Period. 9.10. 5.0.0.. 9.10.10.0.0^ 9.10.I5.0.0''. 9.1 1. 0.0.0.. Altar de Sacrificios . . Piedras Negras Altar de Sacrificios. . . Naranjo Palenque Stela 5 (??) Stela 31 (??) Stela 4 Hieroglyphic Stairway (?). Temple of the Sun Period Ending. Initial Series. Period Ending. Initial Series. Period Ending. 9.11. 5.0.0. 9.11.10.0.0. 9.11.15.0.0. 9.12. 0.0.0. 9.12. 5.0.0. . 9.12.10.0.0^. 9. 12. 15.0.0. . 9.13. 0.0.0. . 9.13. 5.0.0. Copan Copan Copan Copan Copan Copan Copan Palenque Palenque Yaxha Piedras Negras. Copan Copan Piedras Negras. Piedras Negras. Copan Copan Piedras Negras. Copan Copan Piedras Negras. Copan Copan Palenque Palenque Palenque Palenque Piedras Negras. Tzendales Piedras Negras. Stela 12 Stela 2 Stela 10 (?) Stela 19 Stela 23 Stela 13 Stela 3 Palace, House C Slab Stela6(?) Stela 35 (?) Stela I East Altar of Stela 5 Stela 36 Lintel 2 West Altar of Stela 5 Altar of Stela i Stela 37 Stela I Stela 6 Stela 6 Altar H' Altar I' Temple of the Inscriptions. . . . Temple of the Foliated Cross. . Temple of the Cross (?) Stela in front of the Temple of the Cross Stela 8 (?) Temple of the Tablet Stela 2 Period Period Period Period Period Initial Initial Period Period Initial Period Period Initial Period Period Period Period Initial Period Initial Initial Period Initial Period Period Period Ending. Ending. Ending. Ending. Ending. Series. Series. Ending. Ending. Series. Ending. Ending. Series. Ending. Ending. Ending. Ending. Series. Ending. Series. Series. Ending. Series. Ending. Ending. Ending. Period Ending. Period Ending. Period Ending. Initial Series. 'This date also appears as the Initial Series of Stela 4 at Copan, although not the contemporaneous date of that monument. (See p. 357.) =This date is also recorded on Slab 6 of the Hieroglyphic Stairway at Naranjo, but not as the contempo- raneous date of that construction. (See Maler. 19080, plate 27, 03.) ^It is possible that this date on the Hieroglyphic Stairway at Naranjo may not be contemporaneous, although the writer is inclined to accept It as such. ■■This date Is recorded by the Initial Series on Stelae 12 and 2 at Copan, but in neither case is it probably the contemporaneous date of the monument. (See pp. 132, 135, 136, 160.) ^Thls date is also recorded by a Period Ending on Altar H', but the style of this monument, as well as the calculations in its Inscription, arc so closely connected with those of Altar I', the contemporaneous date of which is 9.13.0.0.0, that it appears necessary to regard the contemporaneous date of both as having been the same, namely, 9.13.0.0.0. (See pp. 186, 189, 193, 194.) S84 THE INSCRIPTIONS AT COPAN. List of H otun-markers during the Old Empire — continued. Middle Period — continued. Date. Site. Monument. Kind of date. 9.13.10.0.0' 9. 13. 15. 0.0 9.14. 0.0.0 9-14- 5-o.° 9.14.10.0.0- 9.14.15.0.0' Copan Stela J .... Initial Series. Period Ending. Period Ending. Initial Series. Period Ending. Initial Series. Period Ending. Period Ending. Period Ending (?) Period Ending. Initial Series. Period Ending. Initial Scries. Initial Series. Initial Series. Period Ending. Initial Series. Initial Series. Period Ending. Stela 22 Naranjo Stela 24 Piedras Negras Naranjo Piedras Negras Altar de Sacrificios. . . . Copan Stela 4 Stela 21 Stela I Stela 7 Stela 5 (?) . Itsimte Naranjo . ... Altar I (??) Stela 23 Stela 3 Stela 16 Piedras Negras Tikal Uaxactun Piedras Negras Itsimte Naranjo Piedras Negras Naranjo Piedras Negras Stela I (?) Stela 5 Altar 2 Stela 31 .... Stela 7 Stela 18 Altar i'(?) Gre.^t Period. 9.15. 0.0.0'' 9I5- S-o.o 9.15.10.0.0'^ 9. 15. 15. 0.0 9.16. 0.0.0 9.16. 5.0.0 9.16.10.0.0 Copan Stela A Period Ending. Initial Series. Initial Series. Initial Series. Initial Series. Initial Series. Initial Series. Initial Series. Initial Series. Initial Series. Period Ending. Period Ending. Initial Series. Period Ending. Initial Series. Initial Series. Initial Series. Initial Series. Period Ending. Initial Series. Initial Series. Initial Series. Stela B Altar S Itsimte . . Stela 5 ... Piedras Negras Stela II Stela D Piedras Negras Itsimte Stela 2 Piedras Negras Stela 10 Altar near Structure 39 Altar 2' ... Piedras Negras Quirigua Stela H Seibal Hieroglyphic Stairway Stela 2 Stela M Copan Piedras Negras Stela '»2 Stela J Lintel 3 Stela N Copan Stela 6 (??) Stela F ^Tliis date is also recorded as a Period Ending, in Date 21 of tin Hieroglyphic Stairway here at Copan. (See pp. 256, 257.) -This date also appears as a Period Ending on Stela F at Copan, although not the contemporaneous date of that monument. (See p. 354.) 'This date is also recorded as a Period Ending on Stela 4 at Copan, although not the contemporaneous date of that monument. (See p. 357.) ■•This date appears on a number of monuments, although in no others is it the contemporaneous date: on Altar U at Copan (see pp. 304, 305), on Stela E at Quirigua (see Morley, 1915, pp. 235-239), on Stela F at Quirigua (see ibid,, 1915, pp. 239, 240), on Altar M at Quirigua (see ibid., 1915, pp. 240-242), and on Altar 2' at Piedras Negras. 'This date appears on three wooden lintels from Temples I and III, and the Palaceof the Five Stories at Tikal. (See Maudslay, 18S9-1902, vol. iii, plates 72, 74, 77, 78.) LIST OF MONUMENTS MARKING THE HOTUN-ENDINGS. List of Hotun-viarkers during the Old Empire — continued. S8S Great Period — continued. Date Site. Monument. Kind of date. 9.16. 9.16. lo.o.o.-continued 15-0.02 9.17. 0.0.0. . 9.17. 5.0.0. 9. 17. 10. 0.0. 9.17. 9.18. 5.0.0. g.iS. 5.0.0. 9. 18. 10. 0.0. Yaxchilan Piedras Negras. . Quirigua Copan Copan Copan Copan La Honradez. . . . Nakum Piedras Negras. . Quirigua Seibal Yaxchilan Copan Copan Copan Copan Quirigua Quirigua Ixkun Los Higos La Honradez Naranjo Naranjo Quirigua La Mar Quirigua Aguas Calientes. Cancuen Copan Copan Ixkun La Honradez. . . . Naranjo Piedras Negras. . Quirigua Seibal Cancuen Copan Piedras Negras. . Quirigua Copan Ixkun La Honradez Naranjo Naranjo Naranjo Quirigua Seibal Stela i' Stela 16 Stela D Altar G3 Altar Z Rev. stand in Western Court. Temple 2ia Stela 7 Stela U Stela 13 Stela E Stela 6 Lintel 31 Altar Q Altar W Stela II (??) Temple 18 (??) Stela A Stela C Stela 2 (?) Stela I Stela I (?? Stela 13 Stela 19 Zoomorph B Stela I Zoomorph G Stela I Stela 2 Altar W Fragment X' (??) Stela 1' Stela5(?) Stela 14 Stela I4(?) Zoomorph O Stela 12 (?) Altar 2 Altar G2 Stela 12 Zoomorph P Altar Gi Stela 5 (?) Stela4(?) Stela 8 Stela 12 Stela 28 Stela I Stela 7 Initial Initial Initial Period Period Period Period Initial Period Initial Initial Period Period Period Period Period Period Initial Period Period Period Initial Initial Period Initial Period Initial Period Period Period Period Initial Initial Period Period Initial Period Period Period Initial Initial Period Period Initial Initial Period Period Initial Period Series. Series. Series. Ending. Ending. Ending. Ending. Series. Ending. Series. Series. Ending (?). Ending. Ending. Ending. Ending. Ending. Series. Ending. Ending (?). Ending. Series. Series. Ending. Series. Ending. Series. Ending. Ending. Ending. Ending. Series. Series. Ending. Ending. Series. Ending. Ending. Ending. Series. Series. Ending. Ending. Series. Series. Ending. Ending. Series. Ending. 'The Initial Series actually recorded on this monument is 9.11.10.0.0 l Ahau 3 Zip, the month-sign not appear- ing in its regular position after Glyph A of the Supplementary Series, but farther on in the text. This Initial Series number, however, will not lead to this terminal date, but a correction of 5 in the katun coefficient, that is, adding one bar to the two bars and a dot actually recorded, making the katun coefficient 16 instead of 11, will reach this date, and 9.16.10.0.0 may therefore be accepted as the correct reading for this Initial Series. ^This date is recorded as a Period Ending on, although not the contemporaneous date of, Altar G3 at Copan. (See p. 325.) 'The katun coefficient of this Initial Series number is incorrectly recorded as o instead of 18. The terminal date, however, is fairly clear on the front, and even clearer on the back as 11 Ahau 18 Mac, and 9.18.0.0.0 may therefore be safely accepted as the contemporaneous date of this monument. 586 THE INSCRIPTIONS AT COPAN. List of Hotun-markers during the Old Empire — continued. Great Period — continued. Date. Site. Monument. Kind of date. 9.18.15.0.0 9.19. 0.0.0 La Mar Stela 2 Period Ending. Initial Series. Period Ending. Period Ending. Initial Series. Initial Series. Quirigua Stela K Stela 10 Quingua . . . Temple 1 . 9.I9.IO.O.O Nakum Stela C . Period Ending. Period Ending. Naranjo . . . Stela 32 TO Coo lo.i. 0.0.0 Period Ending. Period Ending (?). Period Ending. Period Ending. Period Ending. Period Ending. Period Ending. Period Ending. Nakum Stela D Seibal Stela 8 Stela 11 Ucanal . Stela 3 TOT T C 10.2. 0.0.0 Flores Stela I Period Ending. Period Ending. Initial Series. Seibal Stela I Tikal . ... List of Hotun-markers during the Netv Empire. CoLONiz.\TioN Period (begins in 9.14.0.0.0). 10.2. 5.0.0.. . 10.2.10.0.0. . . 10. 2. 15.0.0. . . Quen Santo. . Quen Santo. . Chichen Itza. Stela I Stela 2 Temple of the Initial Series, Initial Series. Initial Series. Period Ending. Transitional Period. 10.3. 0.0.0.. 10.3. 5.0.0. . 10.3.10.0.0. . 10.3.15.0.0. 10.4. 0.0.0. 10.4. 5.0.0. 10.4.10.0.0. 10.4.15.0.0. 10.5. 0.0.0. 10.5. 5.0.0. 10.5.10.0.0. 10. 5. 15.0.0, 10.6. 0.0.0. 10.6. 5.0.0. 10.6. 10. 0.0. Tuluum. Stela I (?). Period Ending. 'Curiously enough, although it is recorded on a number of different monuments as a prophetic date, as, for example, on Stelre J and 8 and Altar S at Copan (see pp. 197, 198, 342, 343, and 227-229, respectively), on Altar I at Piedras Negras, on Stela 11 at Seibal (see Morley, 1915, pp. 230, 231), and on Zobmorph G at Quirigua (see ibid., 1915, pp. 229, 230), this important period-ending has not yet been found anywhere as a contempo- raneous date. APPENDIX IX. THE PROVENANCE AND DATES OF THE COPAN MONUMENTS. Note. — The monuments marked with asterisks, thus, Stela E*, do not have their contemporaneous dates ex- pressed by their corresponding Initial Series dates but by subsequent Period Ending dates. Since such inscrip- tions in every case begin with Initial Series, however, they have been classified as such in column 4. I. The Monuments of the Early Period. Monument. rovenance. Date. Kind of date. Class. Altar T' Altar k' Altai L' Altar M' Altar Q' Altar P' Altar X Altar Y Altar A' Stela 22 Stela 25 Stela 20 Stela 24 Stela 16 Stela 15 Stela 17 Stela 9 Stela 21 Stela 18 (Inscribed peccary skull) Stela 7 Stela E* Stela P Fragments V. . . . Fragment S' Group 12 Group 12 Group 9 Group 9 Group 9 Group 9 Group 8 Main Structure Main Structure Group 9 Group 9 Group 9 Group 9 Main Structure Group 9 Main Structure Group 10 Group 9 Group 9 Main Structure Gioup 9 Main Structure Main Structure Group 9 Main Structure 9.0. 9.0. 9.4. 9-4- 9.4. 9.0. 9.5. 9-7- 9-3- 9-3- 9.2. 9.1. 9.2. 9.4. 9.4. 9.6. 9.6. 9.6. 97- 97- 9-9- 9-9- 9.9. o. o. o. o. o. o. o. o. 10. o. 0. o. 19.12. 1. 7. o. o. o. o. 10. o. ID. O. 10. o. 9.17. 10. o. o. o. 10. o. o. o. o. o. o to 9.5.0. 0.0. o to 9.5.0. 0.0. O to 9.6.0. 0.0. o to 9.6.0. 0.0. o(?) o to 9.5.0. 0.0. i8(?) 6(?) o to 9.7.0. 0.0. o to 9.5.0. 0.0. o(?) o(?) o o or 9.7.2.12.0. o o(?) o o to 9.7.0. 0.0. o(?) Period Ending. Calendar Round. Calendar Round. Initial Series (?). Initial Series (?) Initial Series. . . . Initial Series. . . . Initial Series. . . . Initial Series. . . . Initial Series. . . . Initial Series. . . . Initial Series. . . . Initial Series. . . . 8. o. o. o. o. o. 5. o. o. 10. o. o. Period Ending. Initial Series. . . Initial Scries. . . Initial Scries. . . Early Period. Early Period. II. The Monuments of the Middle Period. Stela 12* Stela 2* Stela 10 Stela 19 Stela 23* Stela 13 Stela 3 Stela I* East Altar of Stela 5.. West Altar of Stela 5' Altar of Stela I* Stela I* Stela 6 Altar K Altar H'* Altar r Stela J Stela 5* Fragment Y' Group 3 Main Structure Group 12 Group 13 Group I Group 2 Main Structure Main Stiucture Group 8 Group 8 Main Structure Main Structure Group 8 Main Structure Main Structure Main Structure Main Structure Group 8 Group 8 9.11. 9.11. 9.10. 9.10. 9.11. 9.11. 9.11. 9.11. 9.11. 9.12. 9.13. 9.12. 9.12. 9.12. 9-I3- 9-I3- 9.13. 9-13 o. 0.0. o. 0.0. 19.13.0. 19.15.0. o. 0.0. 0.0. . . . 0.0. . . . 0.0. . . . CO.. . . 0.0. . . . 0.0. . . . 0.0. . . . 10. 0.0. . . . 16. 7.8.... o. 0.0. . . . O. 0.0. . . . 10. 0.0. . . . 15. 0.0 or 9.14 Middle Period. Initial Initial Initial Initial Initial Initial Initial Initial Initial Initial Initial Initial Initial Initial Initial Initial Initial Initial Series- Series. Series. Series. Series. Series. Series. Series. Series. Series. Series. Series. Series. Series. Series. Series. Series. Series. Initial Series (?). 587 S88 THE INSCRIPTIONS AT COPAN. III. The Monuments of the Great Period. Monument. Stela A* Stela B Altar S Stela D Hiei. Step of Mound 2 Hier. Stairway of Mound 26 Temple 26 Stela M Stela N Altar L Altar B' Altar C Altar D' Altar V Altar R Altar U Temple 11 Temple 22 Temple 21a Rev. stand in West- ern Court Altar Z Altar G3 Altar Q Altar W Altar T Fragment E' Stela 8 Stela C Stela H Stela F Stela 4* Altar W Altar G, Altar Gi Fragment X' Stela II Altar 0' Temple 18 Altar F' Altar G' Altar N' Altar T' Altar U' Shrine R'. . . . Fragment Z'. Provenance. Main Structure. Main Structure. Group 9 Main Structure. Main Structure. Main Structure, Main Structure. Main Structure. Main Structure. Main Structure. Main Structure. Main Structure. Main Structure. Main Structure. Main Structure. Group 9 Main Structure. Main Structure. Main Structure. Main Structure Main Structure Main Structure Main Structure Group 5 Group 9 Group 9 Group 10 Main Structure Main Structure Main Structure Main Structure Main Structure. ... Main Structure Main Structure Main Structure (?) Main Structure Main Structure. . . , Main Structure Main Structure Main Structure Main Structure. Group 10 Group 9 Group 7 Group 4 Date. o. o. o. . . . 15. o. o. o. . . . 15. o. o. o. . . . 15. s. o. o.... 15.17. o. o (?). 16. 5. 16.10. ,9.17 ,9.17 16. 5. o. o. . . 16. 5. o. o (?) o. o. . . o. o. . . 16.11. o. 5. . . 16.10. o. o to 16.10. o. o to 9. o(?) 16.12. 5.17 16.12. S.I7 16.12. 5.17 16.12. 5. 17 16.12. 5.17 to 9.17.0.0.0. 17. o. o. o 7. o. o. o. 17. o. o. o. 17. o. o. o. 17. 5. o. o 17. 5. o. o. 17.12. S.I7. 17.12. 5.17 .o(?). ,0(?). (?). 17.12. b. 17.12. o. 17.12. o. 17. 12. 13. 17. 12. 13. 18. o. o. 18. 5. o. 18.10. o. 18. o. o. 17. 5. o. rear Period 16.14. 16. 6 7. 4. I. II IS. 4.17. I (?). (?). or 9. (?).. or 9. '(?) Great Period. . . . Great Period . . . , Great Period. . . . Great Period (?) . Great Period. . . . Kind of date. Initial Series. . . Initial Series. . . Initial Series. . . Initial Series, . . Period Ending. 16 Initial Series. . . . Initial Series (??)... Initial Series Initial Series Calendar Round, . . . Calendar Round Calendar Round (?) Calendar Round Calendar Round. . . . Calendar Round. . . . Calendar Round 2 Initial Series Calendar Round Peiiod Ending Period Period Period Period Period Calend Initial Calend Period Period Calend Initial Period Period Period Ending. . . Ending. . . Ending. . . Ending. . . Ending. . . ar Round. Series. . . . ar Round. Ending. . . Ending. . . ar Round. Series'. . . Ending . . Ending . . Ending . . Period Ending (?) . Calendar Round. Calendar Round. Class. I. Monuments of the Early Period 25 II. Monuments of the Middle Period 19 III. Monuments of the Great Period 45 Total. 89 'This makes a total of 59 Initial Series at Copan exclusive of what may have been destroyed by the collapse of the Hieroglyphic Stairway, but inclu.sive of the very doubtful Initial Scries on Temple 26. APPENDIX X. LIST OF THE DAY-SIGNS AND MONTH-SIGNS FOUND IN THE COPAN INSCRIPTIONS. The following list of the day and month-signs found in the inscriptions of Copan is nearly, although not quite, exhaustive. It is believed to be complete for all the texts described in Chapters II, III, and IV, with the single exception of the Hieroglyphic Stairway of Mound 26, the utterly ruinous condition of which makes it impossible to identify and classify all of its day and month-signs. Of this important text, the longest in the Corpus Inscriptionum Mayarum, only such day and month-signs as occur in the 28 dates described in pages 237-274 are included below. Further study of the disconnected fragments would doubtless yield other identifiable day and month-signs, but it would in no way alter the great preponder- ance of the day-sign Ahau noticeable below, and no attempt has been made to include the day and month-signs on these fragments. Nor has any attempt been made to include the signs for Ahau and Imix when these occur in passages other than where they denote specific days, which is not infrequently the case. These two characters doubtless had other meanings. Thus, for example, in at least two inscriptions. Stela C (north side) at Quirigua,^ and Stela I" at Piedras Negras, the Ahau-sign inverted is used as a sign for the kin or day, the lowest unit of the Maya chronological system. Again, in two other inscriptions, Stela I at Aguas Calientes and Stela 2 at Cancuen, neither of which has been pub- lished, Imix is the main part of the sign for the month Mac, as in the Dresden Codex.^ On Altar S here at Copan it has already been suggested (page 229) that a combination of the two signs may indicate the close of one time-period and the beginning of the next. Such uses as the foregoing clearly lie without the range of day-signs proper, and, as noted above, have not been included in the following list. Of the 158 day-signs included in this list, 104, or very nearly 60 per cent., are Ahau; this is to be explained by the fact that all units of Maya chronology above the kin ended on some day Ahau, and the preponderance of period-endings over all other kinds of Maya dates accounts for the great majority of the day-signs recorded being Ahau. Three day-signs, Imix, Akbal, and Ix, have not been found in the Copan inscriptions at all, although they are by no means unknown elsewhere, particularly Imix, which is fairly common. All of the ig divisions of the haab are represented in the Copan inscriptions except Uayeb, the closing period, and so far as the writer is aware, the sign for this month occurs only thrice in the Corpus Inscriptionum Mayarum — twice at Palen- que, in the Temple of the Foliated Cross (see Maudslay, 1889-1902, vol. iv, plate 80, D8) and in the Temple of the Inscriptions {ibid., plate 59, Q3), and once at Naranjo (see Maler, 1908, plate 30, i, second glyph on the staff, and Morley, 1909, p. 549). This period was composed of 5 days, and, as compared with all the other divisions of the haab, the 18 months of 20 days each, it was only quarter as long as any one of them. Therefore, in the very nature of the case, Uayeb should be found on an average, only one-fourth as many times as any other division of the haab. As a 'See Maudslay, 1889-1902, vol. 11, plate 19, Glyph e. ^See Maler, 1901, plate I2, F3. ^See Bowditch, 1910, plate 8. S89 590 THE INSCRIPTIONS AT COPAN. matter of fact, however, its actual occurrence is very much less even than this, and it almost seems as though there had been a deliberate attempt to refrain from recording this particular month in the inscriptions. Says Pio Perez in speaking of these 5 days: "Some call them u yail kin, or u yail haab, which may be translated, the sorrowful and laborious days or part of the year; for they [the Maya] believed that in them occurred sudden deaths and pestilences, and that they were diseased by poisonous animals, or devoured by wild beasts, fearing that if they went into the field to their labors, some tree would pierce them, or some other kind of misfortune happen to them."^ This belief alone may have been sufficient to account for the observed rarity with which Uayeb appears in the inscriptions, very much less in fact than any other division of the haab, even discounting its shorter length. In the following table the first and third columns give the names of the monu- ments on which the day and month-signs will be found, and the second and fourth columns give their exact positions on the monuments after the method of glyph designation described in Chapter I, page 50. List of the Day-signs in the Copan Inscriptions. Day- sign Jhau . Monument. Altar K' Altar X Stela 35 Stela 24 Stela 16 Stela 15 Inscribed peccary skull. Stela 9 Stela 7 Stela E Stela E Stela E Stela E Altar of Stela E Stela P Stela 12 Stela 12 Stela 2 Stela 2 Stela 10 Stela 19 Altar of Stela 19 Stela 23 Stela 23 Stela 13 Stela 13 Stela 3 Collar of Stela 3 Stela I Stela I East Altai of Stela 5. . West Altar of Stela 5. . West Altar of Stela C . Altar of Stela i Stela I Stela I Altar of Stela I Altai of Stela I Stela 6 Glyph. One end (?) H2 Lower left corner. B5- C3i. b6. AI. B5. B5a. C2, 1. h. ci3,l. h. Dio, u. h. D12, I. h. Tib. hC,a, BS. CIO. B5A. csh. A')a. C2. B. EU4. G7. b8. Next to last glyph. F, __ j__ CI. Y>6a. Ka N2rt, 1. h. oii. la. B5i. Clfl. ct. E(J. B4/;. Day- sign. Ahau . Monument. Stela 6 Altar H' Altar r Altar I' Stela Jot Stela ]w Stela ]w Stela ]vi Stela Jn Stela 5 Stela An Stela Kw Stela \w. Stela Kw Stela B Altar S Altar S Hier. Step, Mound 2.. Altar W Hieroglyphic Stairway . Hieroglyphic Staiiway . Hieroglyphic Staiiway . Hieroglyphic Stairway. Hieroglyphic Staiiway. Hieroglyphic Stairway . Stela M Stela M Stela N Stela N Altar D' Altar R Altar U Altar U Temple 11 Temple 21a Temple 2ia Temple 2lfl Rev. stand West. Court Rev. stand West. Court 'See Stephens, 1843, vol. I, p. 437, and Landa, 1864, p. 384. Glyph. h. h. h. hja. Mb, 1 nb, u. la, u. I. 14, u. h. 2<)b. 31- Bia, 1. h. b6, 1. h. A^b. C2a. cub. C12b. A7- Efl. Ha. ub. A. Step D, vb. Step O, Db. Step R, aa. Step S, M. Date 21, B. Date 22, cb. B2b. Bgb, A7- B16. A. J2. N4fl. 01. H4. E. lb. ob. A4. Qb. h. LIST OF DAY-SIGNS AND MONTH-SIGNS IN THE COPAN INSCRIPTIONS. 591 List of the Day-signs in the Copan inscriptions — continued. Day- sign. Ahau . Monument. hnix Ik.. Akbal. Kan . . Chicchan Cin Manik . Lamat. Rev. stand West. Court Rev. stand West. Court Altar Z Altar G3 Altar G3 Altar Q Altar Q Altar W' Stela 8 Stela C Stela C Stela C Stela C Stela C Stela C Stela C Stela H Stela F Stela 4 Stela 4 Stela 4 Altar G2 Altar Go Altar Gi Stela II Temple 18 Not represented. Altar U Stela 8 Not represented. Altar of Stela 13 Pedestal of Stela N Altar Q Altar W' Aitai H' Hieroglyphic Stairway. . Hieroglyphic Stairway. . Hieroglyphic Stairway. . Altar Y Hieroglyphic Stairway. . Altar V Hieroglyphic Stairway. . Stela 23 Stela I Glyph. u. e'4- B2. AI. A2. B3. d6. Bii. GI. AZb. Ayi. A9a. Bzb. B7fl. Biia. Bi4a. Ala. BI. A3a. x6a. hja. AI. A2a. AI. AI. AI. I-5- DI. A. DI. e6, u. h. Bzb. La, u. h. Step E, si (?). Step F, Db. Date 26, fig. 41. AI. Date 23, Di. A. Date 15, vb. H4. cja. Day- sign. Laviat. Muluc. Oc... Chuen Eb... Ben. . Ix... Men.. Cib... Caban Einab . Cauac. Monument. Altar K Stela ]e Hieroglyphic Step, Mound 2 Hieroglyphic Stairway. Hieroglyphic Stairway. Hieroglyphic Stairway. Hieroglyphic Stairway. Temple 22 Altar H' Hieroglyphic Stairway. Hieroglyphic Stairway. Hieroglyphic Stairway. Hieroglyphic Stairway Hieroglyphic Stairway. Altar of Stela I Altar F' Altar U Temple 22 Altar Q Not represented. Pedestal of Stela N.... Pedestal of Stela N. . . . Temple 11 Hieroglyphic Stairway. Altar V Altar R Altar U Altar U Temple 11 Temple 11 Altar Q Altar Q Altar T Altar T Stela 8 Altar X Hieroglyphic Stairway Hieroglyphic Stairway Hieroglyphic Stairway Glyph. C2. 27. Step D, sa. Step E. Db. Date 24, ob. Gordon 1902, pi. 13, D, ca. A, u. h. Da, I. h. Step I, r). Step K, .\b. Step L, A. Step P, Ma. Step Q, Mb. ja. Bib, u. h. 03- P2. CI. AI. BI. H7- Date 27, B. C2. AI. AI. KI. AI. b6. Front, top glyph. AI. Front, head of left figure. Front, head of right figure. AI. AI. Step H, A (?). StepH. o(?). Step K, v.b. List of the Month-signs in the Copan inscriptions. Month- sign. Monument. Glyph. Month- sign. Monument. Glyph. Pop.... Uo Stela P B6b. }b. Step Q, A. B2. BI. P3- f8 (?)• BI. Bii, 1. h. BI. 1 Uo .... Zip .... Stela E C3. B2b. e4. Ei, 1. h. e6, 1. h. B7i. Biii. Bli, 1. h. Hi. AIS. Altar of Stela I Hieroglyphic Stairway Pedestal of Stela N.... Altar U Altar of Stela E Altar of Stela I Altar r Altar Q Altar U Stela C Temole II Stela C Altar Gs Stela ]n Altar F' Altar S Altar Y Stela N 59^ THE INSCRIPTIONS AT COPAN. List of the Month-signs in the Copan inscriptions — continued. Month- sign. Monument. Zip. Zotz Tzec . . . Xul. Yaxki; Mol. Chen. . . Yax. . . . Altar R Rev. stand West. Court Rev. stand West. Court Altar T Stela 8 Stela 8 Stela 8 Stela 7 Stela I Stela 6 Altar H' Altar I' Hier. Step, Mound 2. . Hieroglyphic Stairway Stela M Stela 20 Stela E Hieroglyphic Stairway Hieroglyphic Stairway Hieroglyphic Stairway Altar of Stela I Hieroglyphic Step, Mound 2 Stela i6 Hieroglyphic Stairway Stela lo Stela 23 West Altar of Stela 5. . Altar of Stela I Pedestal of Stela N. . . Altar V Altar Q Altar Q Stela I East Altar of Stela 5. . West Altar of Stela 5. . Altar H' Hieroglyphic Stairway Altar V Altar R Altar U Altar U Temple II Temple II Altar Q Altar T Stela 8 Altar of Stela 19 Stela D Altar K Stela A Stela B Altar S Altar U Glyph. KI. BI. VI. Front, in hand of right figure. A2. HI. E3- B7fl. C2. B6i. Na, u. h. lb, u. h. H2. Step D, lb. A3a. Bg. Dio, 1. h. Date 23, E. Step R, v,b. Step S, N. Dfl. Ni. C3a. Step H, ?b. \C)b. 14. N2i. lb. A2. B. BI. A4. d6A. lb. D2a. cb, u. h. Gordon, 1902, pi 12, L, sixth block D2b. BI. LI. MI. Bib. B7. Front.bot'm glyph. Front, in hand of left figure. BI. c. A^b. G2. BI2a. a8. Eb. PI. Month- sign. Yax. Zac. Cell. Mac. Kankin Muan . Pax. . . Kayab Cutnhu Monument. Uayeb . Stela 4 Altar B' Altar D' Temple II Altar W' Stela 4 Altar Gi Stela 15 Stela 12 Stela 2 Stela 23 Stela 13 Stela 3 Stela I Altar U Altar G2 Temple 11 Stela 12 Hieroglyphic Stairway Hieroglyphic Stairway Stela F Altar W Altar X Stela A Hieroglyphic Stairway Hieroglyphic Stairway Hieroglyphic Stairway Altar Q Stela C Stela C Stela H Stela 9 Stela 5 Stela 15 Altar H' Stela 5 Hieroglyphic Stairway Hieroglyphic Stairway Altar Q Altar W' Stela C Stela C Stela 25 Ctela 23 Altar r Stela Jfo Stela ]e Stela A Hieroglyphic Stairway Hieroglyphic Stairway Temple 21a Rev. stand West. Court Rev. stand West. Court Altar Z Altar G3 Stela C Not represented. Glyph. Ajb. BI (?). d4 (?). H7- CI (?). A6i. BI. d6. DIO. c6a. FI. A9- B8fl, u. h. cib. ^\b. BI. h8(?). A9. Step D, Sfl. Step E, Ha. A2. B. BI. cib. Date 15, ub. Step K, si. Step L, cb. DI. AgJ. Bub. Alb. b8. B7a, u. h. C3- lJ, u. h. a8, u. h. Step E, Ta. Step F, la. EI. A2. A3a. ABa. Lower light corner. f8. Ha, 1. h. 33- 12, u. h. Aqb. Date 21, ca. Step P, pa. ra. RI. e'5. A3- B2. B3a. APPENDIX XI. A DESCRIPTION OF THE RUINS OF COPAN, BY JUAN GALINDO, IN 1834.i Report of the Scientific Commission Appointed to Make a Survey of the Antiquities of Copan in Compliance with a Decree Dated January 15, 1834, Issued by Dr. Mariano Galvez, Commander-in-Chief of the State of Guatemala. Copan, June ig, iSj^. Citizen Secretary of the General Office of the Supreme Government of the State of Guatemala: I have the honor to submit to you the following statement of the investiga- tions which I have been able to carry out among the ruins of this old city and the neighboring country, in compliance with the mission by which I was honored by that Supreme Government on the i6th of January ultimo. It is impossible for us to fix in an accurate manner the beginning of the exis- tence of the planet which we inhabit; nevertheless, we can clearly see that since that epoch it has undergone great changes. Water used to cover what is now dry land, whole races of gigantic creatures have disappeared, and tropical animals inhabited the boreal regions. Of all the living species of the globe, that to which we belong is the one which particularly attracts our attention. Man appears in six different races, namely, the American Indian, the Esquimaux, the Tartar, the Malay, the African, and the Caucasian, and among all these the oldest is undoubtedly the Indian. The mis- taken and foolish pride of the descendants of the Caucasian makes them claim 'The manner in which the original manuscript of the Galindo report on the ruins of Copan fell into the writer's hands after the first chapter of this book was already in galley proof is so unusual as to warrant a brief account of this timely discovery by way of a preface to the report itself, a translation of which into English is given here. Colonel Juan Galindo was an officer in the service of the government of Central America, and previous to his visit to Copan in 1834 had been commandant at Flores, in the Department of Peten, where he had also under- taken other archajological investigations. (See page 18.) In April 1834, he was sent to Copan by the government of Central America to make an official report on the ruins, and while there he wrote several letters to scientific societies and periodicals both in Europe and America, notably to The London Literary Gazelle and Journal of Belles Lettres, Arts and Sciences (see Galindo, 1835); The American Antiquarian Society (see Galindo, 1835a); and The Societe de Geographic de Paris (see Galindo, 1836 and l836fl). All three of these letters were written under the same date, June 19, 1834 (the same date as that of his report, see above), and in the one to the American Antiquarian Society he states that he was engaged in the preparation of a report which "the Government of Central America intends publishing" (Galindo, 18353, p. 545). This report, however, never seems to have been published, and its existence was only known through the above letters. Indeed, after a protracted search for it in the government archives in Guatemala City, during which the writer enlisted the aid of his friend, the Licenciado Don Adrian Rccinos, Undersecretary for Foreign Affairs in the government of Guatemala, he reluctantly reached the conclusion that it had been destroyed in one of the many revolutions which had swept over Guatemala since Galindo's time, and in Chapter I of the original manuscript of this volume had so stated. Last summer (August 1919), during the course of a visit to Baltimore, where Mr. William Gates, of Point Loma, California, then had his large collection of Maya manuscripts housed, the latter placed in the writer's hands a folio manuscript of 46 pages in Spanish, which he said he had received from abroad several years ago. This manuscript proved to be none other than the long-lost original of the Galindo report on his mission to Copan, written in Galindo's own handwriting at Copan on June 19, 1834. It was directed to Dr. Mariano Galvez, then Commander-in-Chief of the State of Guatemala, and the 25 figures, which had formerly accompanied it — maps, drawings of the monuments, etc., direct reference to which is frequently made in the text of the report — had been removed before it came into the possession of Mr. Gates. Where this report had lain hidden all those 80 years, and how it came into the hands of the collector from whom it was purchased by Mr. Gates, it has been impossible to ascertain, but the writer regards it as a peculiarly happy coincidence that it was "rediscovered" in the Gates collection during the past summer, and that the owner has graciously consented to its publication here for the first time. 593 594 THE INSCRIPTIONS AT COPAN. the greatest antiquity, when, in reahty, they are the youngest. In a comparatively recent epoch they migrated from the Caucasian Mountains, took possession of Europe, and have been able to spread themselves throughout the greater part of America, and, aided by the strength of their youth and talent, they are now invad- ing Asia and Africa. The Indian race, on the contrary, is in a decrepit old age, having passed m_any centuries ago through youth, civilization, and even decadence. The North Americans solved the problem by beheading the unfortunate natives of the continent, or driving them away to the west, where they gradually perished.^ The new governments of that part of America that formerly was Spanish admitted them into their societies and endeavored to have them share the benefits of civilization; but this policy, though it honored its authors, will be useless. The Indian race is in the last centuries of its age, and soon will disappear from the earth. As a general rule, power and civilization travel westward. Looking for their origin toward the East, we go from France to Greece, from the latter to Egypt, to China, and, finally, we come to America in the fartherest eastern end. China, the oldest nation of the Trans-Pacific Hemisphere, is about to disappear, and perhaps will become a colony of a far-distant island. The Indian race, which was the ances- tor of China or Tartary in the evolution of civilization, has in a greater degree than China reached an old age incapable of regeneration. We see this clearly in Central America. The extensive Mosquito Coast, inhabited by natives who are entirely free from foreign domination, surrounded by civilized colonies and states, and in spite of the fact that the English authorities have endeavored to educate the chil- dren of their leading men, that people are still in a state of most degrading bar- barism. It is necessary to consider the Indian race in olden times in order to render to it the respect it deserves; it is necessary to state other facts in order to destroy the fullest vanity of the white or Caucasian race of being the mother of the Indian race or the origin of its civilization. We have seen that this is, by analogy, the oldest human species of the globe; if we admit the blunder that the Indians descend from the Africans or the Europeans, and that the climate has changed the color of their skin, how is it, then, that the American Indian preserves in all climates the same bronze color.? If the climate of the West Indies darkened the descendant of the European, the snowy mountains of Canada, Quito, and Patagonia should have turned their skin into the same whitish color. If the old nations of the Trans- Atlantic hemisphere should have had authentic information concerning these regions, they would have communicated it to posterity. Some learned men of that hemisphere, even before Columbus, suspected that there were lands here, just as we reasonably suppose that there are such in the center of the unknown portion of the Southern Ocean. But why should we endeavor to make the American a de- scendant of the other continent.? Is it possible that there have come thence the tapir, the llama, and so many other animals of the hot lands of the American tropics ? There is no trace of their race in the other hemisphere, and it is impossible that whole species or kinds of animals should come without leaving behind a single specimen of their race. They could not have come swimming by sea; from the north they could not have arrived, at least in the present state of the globe, because the cold would have killed them. No man could have brought ferocious wild animals in ships. Based on these reasons and other data, I establish the two following epochs of American civilization : Of the primitive, as is to be supposed because of its antiquity, 'This is an interesting commentary upon our treatment of the Indians, written 85 years ago by a foreigner, who felt himself under no obligations to refrain from telling the truth as he saw it. DESCRIPTION OF COl'AN BY GALINDO. 595 there remain no tangible traces whatever. Its destruction must be attributed to an horrible convulsion of the world, to plagues, hunger, to a barbarous invasion of the extremities of the continent, or perhaps to an uprising of the slaves. The colonies, or remains of this primitive civilization, having passed to the eastern coast of Asia, prepared the enlightenment of Japan and of China, and this primitive civilization, although dispersed, likewise left traces for the second. The Chinese, Hindus, Persians, Chaldeans, and Egyptians all were very much alike in their character and other characteristics, which indicates a common origin which may be looked for in America, and perhaps the Sanskrit language originates in this continent. America lapsed into barbarism, and one century after the destruc- tion of Rome by the barbarians of the North, there appeared from our northern region the Toltecs, bringing with them some enlightenment and a partial civiliza- tion, and who settled around Anahuac and founded an empire. Later on the Incas of Peru endeavored to revive the old American civilization. The Toltecs came imbued with the remembrances of the first epoch of enlight- enment which the ancients perhaps left on their journey toward the west. Their conquests and the colonies of the Toltecs were extended to Central America; they mixed their language and customs with those already existing in these regions and formed several states. From the analogy of their language, writing, and places where sacrifices were made, it is deduced that Copan originated from a Toltec colony, and that its king dominated the country extending to the east of that of the Mayas, or Yucatan, reaching from the Gulf of Honduras almost to the Pacific Ocean, and comprising an area of over 10,000 square miles, at present included in the modern States of Guatemala, Honduras, and Salvador. Throughout this extent of land the Chorti language was spoken and is still spoken, and from these and other data it is inferred that the peoples of Cuagini- quilapa, Los Esclavos, Quesaltepeque, San Jacinto, Santa Elena, San Esteban, San Juan Ermita or del Rio, Jocotan, Camotan, San Jose, Chimalapa, Sacapa, and San Pablo, in the State of Guatemala, formed a part of this empire. Chiquimula and Esquipulas were governed by subordinate princes of the King of Copan. In Honduras, Omoa, the mineral district of San Andres, Sensenti, Ocotepeque, Tipalpa, La Brea, and other places were comprised in their dominions. In Salvador the same empire comprised Texis, Dulce Nombre, Metapas, Tejutla, and Sitala. The large city of Copan, Copante, or Copantli was the capital of the nation and residence of the monarch, being situated at 14° 45' north latitude and 90° 52' west longitude from Greenwich.^ This city is built on the right bank of the river of the same name, extending along said river a distance of over a mile and a half (see map No. i).- Hills and mounds of unwrought quarried stones indicate the site of the city and of the principal buildings, all of which have fallen. In all that place there are found obelisks, some standing and many lying on the ground, wrought tables, busts, and several fragments of statuary and earthenware. The principal and highest building was the temple (see plan No. 2), built at the eastern end of the city and perpendicular to the bank of the river. They used iQallndo's latitude is approximately correct, but his longitude is more than 160 kilometers too far west. A longi- tude of 90° 52' west would locate Copan in the western part of Guatemala, not far from Santa Cruz Quiche. See page I, note I. ^As already noted at the beginning of this Appendix, all the illustrations, maps, and drawings in Galindo's report had been removed therefrom before it came into Gates's hands in igiS. The writer suspects that the "ten drawings well enough executed," mentioned in Galindo, 1S36, p. 268 (see page 19, note l), and seen by Hamy as as late as 18S6 in the archives of the Societe de Geographic at Paris, are some of the original illustrations of this report, or at least duplicates thereof, which Galindo himself sent to France. The numbers, which follow, are Galindo's references to his own illustrations, now unfortunately separated from his report. 596 THE INSCRIPTIONS AT COPAN. to ascend from the neighboring plain to all the eminences and places where sacri- fices were made, by steps which in many places are now deteriorated and lost. The little plaza [the Eastern Court] surrounding the church is 21 yards above the level of the river, and ascent from it is made also by steps which surround it on four sides to the place of sacrifice, to the north, to the west, and to the precipice on the bank of the river. (This is shown from the opposite side of the river under No. 3.) It is a wall of partially quarried stone, the height of which, as now shown, is 47 yards. It contains three openings, called windows, the principal and highest . of which, wherein the body of a man can be placed, is i yard and 13 inches high and 30 inches at its greatest width. (See fig. 4.) (I estimate the vara or yard at 32 inches, and the latter is a half inch longer than the Castillian vara or yard.) It is constructed, without mortar, of large quarried stones, and extends directly into the hill at the same level and size until it terminates at the same height as the floor of the little plaza, and communicates only with the latter; it is diflBcult to guess its use, because it evidently was not used for drainage. The window farthest to the north can not be entered on account of its small size, inasmuch as the bottom thereof being filled with earth, constituting a solid terrace, there is only left an aperture 16 inches wide by 8 inches high. The window nearest the water is larger, but is filled in a yard from its mouth or opening. The whole of said wall is very much deteriorated and broken up, although it still has two whole sections built of small, square stones. Among many excavations, I made one at a place ending in the small plaza [the Eastern Court], where the highest window is located. First I encountered the opening of this window, and excavating to a greater depth, we entered a sepulchral chamber, the floor of which is 4 yards and 9 inches lower than that of the small plaza. It is 2 yards and 6 inches high, 2 yards wide, and 3 yards and 19 inches long, extending directly from north to south, in accordance with the compass, which in these countries has a variation of 9° to the east. The outline of its form is shown in No. 5. It has two niches on each side which are 18 inches high from the floor and which are 16 inches deep, 19 inches high and 28 inches wide, and both the niches and the floor of the sepulcher were filled with pieces of red earthenware coated with tar, such as dishes, wash-bowls, frying-pans, and pitchers. I took out more than 50 of superior workmanship, and some of these were full of human bones, mixed with lime, sharp razors, and with a material that the Mexicans used to call "itzli," with a small head (see No. 6), which seemed to represent the head of a corpse, the eyes almost closed, the lower jaw fallen, and protruding lips, there being many symmetrical holes in the back, as for hanging or shaking the same, the whole head being made of a fine stone covered with green enamel, ^ the same as two strings of beads which I also found in the vault, together with niany shells of snails and oysters, which undoubtedly were brought from the sea in compliance with some superstition. Besides, there were stalactites brought from some cave to be deposited here. The whole floor of the subterranean vault was filled with fragments of bones, and under them there was a layer of lime on the solid pavement of stone. The stones from which this vault is constructed are 10 inches thick and 10 inches wide and long, and were not set in mortar. To the west of the square and a little above the steps, there is the gigantic bust (No. 7), called by the common people "The Bull"; it is 2 yards high, although the crest is made of stone separated from the head.^ On the same steps, but a little below and in the direction of the place where sacrifices were performed, there is a 'Galindo here doubtless refers to jade. =See Stephens, 1S41, vol. I, plate facing page 143. DESCRIPTION OF COPAN BY GALINDO. 597 gigaatic monster the shape of which resembles that of a large toad standing, with human arms and the claws of a tiger.^ The steps of the temple generally are 12 inches high and more than half a yard deep, with an inward inclination; the first two, which lead from the plaza or square upwar to the place where the sacrifices were performed, are of the aforesaid size, but the others are from 1.5 to 2 yards high. From the side of the narrow pass to the third step the wall of the place where sacrifices were performed is perpendicular. On the other side of the latter there is an obelisk, 3 yards and 22 inches high [Stela P]. On its western front there is worked in semi-relief figure No. 11 ; on the opposite side there are two perpendicular rows of double squares with characters and on each side appears the same thing, but with single squares. (No. 18 represents one of the latter.) The whole column is whitewashed, and over it can still be seen parts of the red color which in ancient times must have covered the whole stone, but which water and time have caused to disappear almost entirely. A short distance to the left of this obelisk there is a kind of solid table, very remarkable, somewhat raised from the ground, together with other smaller stones [Altar Q]. It is i yard and 23 inches long, is of the same width, and is 27 inches thick or high; the upper part is divided into 49 squares with characters, having a cornice 4 inches thick, and surrounding the 4 faces of the table there are 16 human figures, seated on cushions or benches, with legs crossed, and with fans, or some- thing else which I can not accurately describe, in their hands, as drawn and shown in Nos. 20, 21, 22, and 23. On the steps which lead up from the place where the table is situated to the place where sacrifices are performed [Mound 16, plate 6], there are many gigantic skulls carved on the extremities of large loose stones. To the left of the place where sacrifices are performed there is a soHd stone, the exterior shape of which resembles that of a canoe; No. 9 shows its outline, while No. 10 represents its upper surface. It is i yard and 20 inches high, 26.5 inches wide, and 29 inches in thickness or height. Near to the corner of this pyramid of sacrifice [Mound 16] there is found a rectangular table or stone [Altar H'], elevated above the ground like that already described [Altar Q] by smaller stone supports; it is 2 varas 21 inches long, and i vara 20 inches wide, and 13 inches thick or high; three sides of its edge contain characters, which are in squares (casillas) four to a block; figure 17 represents one of these rectangular groups; the edges of the stone thus contain 24 squares on the long side, and 16 on the smaller sides; the other long side, which faces south, and the top and bottom are plain. At the other side of the canoe there is a stone, or table, similar to the last described, but very broken [Altar I']. The three tables referred to [Altars Q, H', and I'] — the one called canoe and the obelisk [Stela P] — are found in the plan on the level of the little plaza [the Western Court], and from thence descent is made toward the south and west to the ground. From the northern side ascent is made by steps to a part of the temple whose height is equal to that of the place of sacrifice, and exceeds that of the river wall [Temple 11]. At the foot of this elevation, on the level of the ground and near the corner which the temple forms to the northwest, is another obelisk with human figures on its front and back, and a row of smaller figures on the two sides, with evidence of all having been of a red color [Stela N]. In front it has a small altar or circular table at a distance of 2 yards. In the narrow passageway of the temple is a circular stone i yard 4 inches in diameter and 16 inches thick. Although it somewhat resembles a millstone, but as 'This gigantic monster resembling a toad is one of the pair of rampant jaguar figures which flank the Jaguar Stairway on the west side of the Eastern Court, at the Main Structure. ^g8 THE INSCRIPTIONS AT COPAN. it has not a hole in the center, I do not divine its use [possibly the altar of Stela i]. There are other circular stones among the ruins of more or less the same size and some of them with holes in their middles. Some distance from the temple toward the northwest there are five obelisks still standing (see No. 24), the northernmost, at the foot of a great fallen edifice [Mound 2], has on the face opposite the building the representation of a man [Stela D]. It should be borne in mind that all of the standing figures have the hands resting in the same manner, across the breast, and the shoulders are not well delineated. The sides of the obelisk are curiously wrought and the back is divided into 16 squares, wider than they are high, each one containmg one or two little figures squatting in various postures. Hard or fine stones are encountered set in the obelisk.^ The latter and other works of the ruins are of a soft kind of stone, but, nevertheless, it is admirable how the ancient architects wrought them, they being ignorant, as is known, of the use of iron. Did they work with chisels of gold or silver, or did they use chisels of chaya? In front of this column there is an altar or wrought table with corners toward the principal figure and 5 yards distant from it. To the south of this obelisk there is another, whose eastern face contains figure No. 14, and in the opposite face the drawing No. 15 [Stela B]. The height of the stone is 4 yards, the width is i yard 13 inches, and the thickness i yard 4 inches. The sides contain squares with characters (hieroglyphics). It is seen that the figure of this obeUsk, like the others, has bracelets on the wrists and ankles, the feet have sandals tied on them, the laces passing between the first, second, third, and fourth toes. The dress of this figure extends only to the upper part of the thigh. A little to the south, and close to another fallen building [Mound 4], there is an obelisk [Stela A], the back of which contains the characters shown in No. 16, which are placed in double squares, that is to say, 2 in each quadrangle. Those which appear in the lower row are no longer legible. The Indian writing found among all these relics is quite interesting, especially because heretofore we did not know that the art of writing was known on this continent before the voyage of Co- lumbus. This writing is hieroglyphic-phonetic, representing sounds, and is greatly superior to the paintings of the Mexicans and the symbolic hieroglyphics of the Egyptians, which only represented things." A human figure appears on the front of this obelisk, and on each of the two sides there are 2 single rows of squares with characters; its height is 3 yards 21 inches, i yard 2 inches wide, and i yard thick. It is evident that it had had red paint, and it seems to m.e that all the obelisks were painted, although in some of them the influence of time has destroyed the color. To the east, and in front of figure No. 14 [Stela B], there is another standing stone, the back of which contains 20 squares arranged in groups of 4, as can be seen in No. 19 [Stela F]. The upper ones seem to be hanging, and the same is true of the two squares of No. 15, all of which lead us to believe that this nation used fo write their letters on wooden tablets which they afterward hung in their homes with strings or ribbons. The front of the obelisk is occupied by a human figure in Iialf relief, in front of which there is a large square table or altar, having a hole in the middle. The obelisk nearest the temple is located at a distance of about 200 yards to the northeast, being to the south of the last one described, separated by a dis- tance of 25 yards [Stela H]. The figure on its face is shown in drawing No. 12. The trousers he wears are curious; perhaps it was a woman's dress. No. 13 represents 'Galindo here refers to the roughly spherical Inclusion composed of a harder, more indurated material found in the rock-mass of Stela D, for a description of which see Appendix I. 'It is hardly necessary to point out that Galindo is in error here as to the real character of the Maya hiero- glyphic-writing. While it is induhitably true that phonetic elements occur here and there in the texts, the great majority of the signs are idrngraphic, that is, they stand for ideas rather than sounds. (See Morley, 1915, pp. 33-30.) DESCRIPTION OF COPAN BY GALINDO. 599 the foot of the same figure drawn to the right. In front of the figure, at a distance of 3 yards, there is a table or altar with the corner toward the figure, the upper part being straight. The other 3 sides of the obelisk are curiously wrought. Near by there is a stone (figure No. 8), almost spherical in shape, surrounded by a belt, its horizontal diameter being i yard 28 inches, while its smallest or per- pendicular diameter is i yard 6 inches. In the upper part there is a small circular cavity crossed by a serpentine line. I wonder what whim of art this could be? There is also here, fallen and in ruins, the head of a gigantic alligator or lizard, between the j aws of which there is placed the half body of a monstrous figure, whose face is human, but with the claws or feet of an animal. On the other side of the Copan River, on a hill in front of the windows, and at a distance of half a league or more from the latter, there is a monumental stone, now fallen and broken into 2 parts, a small piece of its head being missing [Stela 12 at Group 3]. Although it stands on a commanding site, at its back there is, at a certain distance, a still higher mountain. This obelisk, like all the others, is parallelepipedal in shape. It was standing with one of its wider sides overlooking the windows, that is to say, west 29° 15' north, surrounded by a pavement a little higher than the ground, 8 yards long, extending from the northeast to the southwest, being from 4 to 5 yards wide. This obelisk, monumental stone, or column was not a perfect parallelepiped, as its width and thickness gradually diminish from top to bottom, and although its actual total length was 3 yards 28 inches, its greatest width at the top is 24 inches, and at the bottom it is only 22 inches wide. In the upper part the greatest thickness of the obelisk is 19 inches, and at the bottom it is only 16 inches. The stone was buried to the depth of 24 inches, and the 12 mches from that point to the squares of its inscription are smooth. On the two widest faces of the column there are 48 squares in 4 perpendicular rows of 12 each. On the other 2 sides there are 52 squares, also in 4 rows, 2 on each side. The 4 lower squares of the widest faces are 9 inches square, their width upward bemg the same, but their height gradually diminishes until the height of the 4 lower squares of the same faces is only 6.5 mches. The same is true of the 2 thick sides of the obelisk, where the 4 lower squares have a width of 8 inches and a height of 7 inches, the width continuing uniform upward, but the height diminishes gradually, inasmuch as the upper squares become equal to those of the other two faces at a height of 6.5 inches. The margin makes the difference in the size of the column. The red paint which covers this stone and causes the common people to call it painted is so well mixed that even now it is noticeable throughout the entire surface, and seems to constitute an essential part of the obelisk. Upon a hill, at a point even more prominent than the one referred to, and about 2 leagues distant west of the windows, there is another monumental stone [Stela 10 at Group 12]. This obelisk is smaller than the one just described, and is com- pletely fallen. Its height from the bottom of the squares is 24 inches, its total length being 3 yards 20 inches. It is rectangular in all its parts and contains 72 squares, arranged in 8 perpendicular rows of 9 each. On the two widest sides ot the obelisk, that is to say, those which are 24 inches wide, the squares and their characters are very distinct and marked. On the other 2 narrower faces, the 2 rows of squares can scarcely be distinguished, and the characters are confused for lack of space. These letters, signs, or characters, like most of the others which have been noticed, are painted red. Many piles of fallen edifices are found in all the neighboring countiy. Half a league to the north of the western boundary of the city there are immense quarries, forming great precipices and ravines, from which the ancient architects obtained 600 THE INSCRIPTIONS AT COPAN. material for their buildings, obelisks, etc. The stone, of course, is the same as that which is encountered in the ruins. These quarries are in the ocote-pine groves, in the direction of the hill of Cutilca. The Cave of Tibulca, of which Father Juarros speaks so fabulously, must be the Cutilca Cave, as there is none other in all that vicinity, and it should be added that this corruption of names has been common. On the side of the hill of Cutilca, some 800 yards above the Sesesmil Canyon, which divides said mountain from the "Cerron" or large mountain, called by this name because of its greater height, facing the west , is the mouth of the cave commonly called "Tigra" [Tigress], because some 20 years ago a ferocious animal of that species took refuge therein. As the hunters desired to starve this animal, they closed the entrance with a fence, but being tired, after 7 days waiting, they entered the cave and killed the wild animal with their lances. The entrance to the cave is 5 yards high and 2 wide, and the fence, which completely closed the same, still exists. This cave is quite inferior in extent, beauty, and interest to the Jobitsina Cave near the Lake of Peten, the latter cave being one of the most beautiful works of nature, but the nearness of the Cutilca Cave to the great city of Copan must have made it famous. According to figure No. 25, the gallery at the entrance is level until it turns to the north, then rapidly falls to another level which, on account of its greater width, forms a hall. From this hall ascent is made to another space a little higher up. There are 80 paces from the farthest extremity of the cave to its mouth. Only a few bats dwell there. It does not have many nor beautiful stalactites, and these are not very hard, and can be loosened almost by the force of the fingers. Since the death of the tigress no one has dared enter the cave. The hill of Cutilca is, furthermore, very remarkable because of its picturesque slope. Near the top and looking toward the south there is an immense precipice of red rock. The common people speak of a cave supposed to exist at the foot of this ravine. After many difficulties on account of the steepness of the mountain, I went up to this place, which has no cave other than some small open cavities. From the foot of this natural wall a large extent of the States of Guatemala, Salva- dor, and Honduras, the Gilotepeque volcano, etc., n seen. At a distance of 3 leagues from Copan, across the little hills of ocote-pme forests called Llano Primero, half a league to the west of the rancho of Llano Grande, through which passes the main highway from Chiquimiula to the plains of Santa Rosa, are found quantities of trunks of petrified ocote-pine, particularly three kinds, namely, one of great specific gravity or weight, another of flint stone, and another much used for sharpening iron instruments. Great trunks of petrified pine are seen almost buried, and those which are on the surface are of different sizes, but are generally split horizontally by the force of the sun, in widths of from 4 to 5 fingers. In a neighboring oak grove are found pieces of petrified oak, equally good for whet and spark-stones. Petrified wood is found in great masses wedged in the fallen trunks, and large pieces of the same are half buried. Crossing a creek through the oak forest, which stream dries up in the summer, it is found that all of the said ocote-pine and oak groves are in a dry place, which causes us to wonder as to the cause of the petrification. This is one of the curiosities which makes Central America the country where nature has been most lavish in exhibiting its wonders. Pines and oaks grow in the midst of petrified timber. Only in the place indi- cated is this petrification found, since there is nothing like it in the neighboring country, not even in the swamps. This petrification process has doubtless sug- gested to the present inhabitants of these regions the foolish idea which they enter- tain, concerning the liuman figures which are found in the ruins of Copan, that DESCRIPTION OF COPAN BY GALINDO. 6oi formerly they were living beings who, because of their heathenism and sins, became stones by divine wrath. The Copan River rises in the mountain to the east, runs through La Brea and beneath the windows. Farther on the Sesesmil Canyon joins the river on the north, this canyon rising also in the said mountain, passes near the famous Cave of Cutilca. It used to divide the western suburbs of Copan from the rest of the city, and flows into the river within the limits of the city, as may be seen on map No. i. The Copan River runs to the west, receives the Jupilingo River, runs through the village of Xupa and the villages of Camotan and Jocotan; and 7 leagues below the Chiquimula River empties into it. It passes between Sacapa and Estansuela, and empties into the Motagua, its course extending 20 leagues, counting from this place [Copan] to its mouth. Thence to the sea, through the Motagua River, is a distance of 65 leagues, following the course of the river. The Copan River con- tains a great quantity of fish and in no place is navigable, although in the winter season logs are floated down it. Comparing these ruins with those of Palenque, it can be seen immediately that their similarity suggests a common origin, in spite of the fact that they differ in essential points. Palenque was ruined and forgotten before the conquest, while the Spaniards found Copan in all its splendor, and yet the buildings and other works in Palenque are in a better condition than in Copan, owing to their superior architecture. Here in Copan there are no houses standing, as there are many in Palenque. Its build- ing-stones are of diverse character, while those of Palenque are not more than 2 inches thick. The roofs in Copan were m.ade of inclined stones, while those of Palenque are always horizontally placed. In Palenque they are cemented with mortar, while m Copan they are not. In ancient times, with the exception of Palenque, Copan was undoubtedly the most remarkable city of Central America, since, if the capitals of the Quiches and Cachiqueles equaled it, there would have been left some signs of their superiority. There is more fineness and perfection in the human figures there [Palenque], and they are nearly always placed in profile, while these on the contrary are most commonly found with front views. I did not see obelisks nor carved tables at Palenque. The circular stones of both places are very similar, and also I find their writings always placed in almost square blocks containing faces and hands and other identi- cal characters. This similarity may come from the similarity of the Maya and Chorti languages, or perhaps because the inscriptions are in a dead language, the common mother of the two languages mentioned. In the same manner modern nations frequently use Latin under the same circumstances. The following is a short vocabulary of the Chorti language: Quin Sun. Unen Son, Daughter. Uj^ Moon (a month). Sacun Eider brother. Ek Star. Uitsin Younger brother. Kak Fire. Jor Head. J? Water. Sutsernijor Hair. lokar Cloud. Unacaut Eye. Uinik Man. Chiquin Ears. Ixik Woman. Ti Mouth. Tegerom Bov. Caab Hand. Ikchok Girl. Tigere Tapir. Tata Father. Masa Deer. Tu Mother. Mut Bird. ^Tliis is the Spniiisli 7, atul oqu.ils the English /;; a; eqti.ils English sli. 6o2 THE INSCRIPTIONS AT COPAN. Chai Fish. Otot House. Guina Hunger. Uitsir Hill. Jacatiniti Thirst. Noja River. Aingiie To eat. Inte I. Unchi To drink. Chate 2. Inguaian To sleep. Uxte 3- Inchamai To die. Chante 4- Te Tree. Jote 5- Ucabte Branch of a tree. Uakte 6. Tajte Ocote-pine. Uaxikte' 7- Ixim Maize. Ukte 8. Ajan Ear of green corn. Boronte 9- Uchigtun-cha Grindstone. Launte- TO. Tun Stone. In the orthography of the indigenous languages of Central Americaj x has the same value as the French ch. The climate of Copan is excellent, and the site of the city has all the advan- tages which the ancient Americans generally chose for their principal towns. Copan is located 770 yards above the level of the sea, and consequently has a temperature similar to that of Amatitlan. Now that the governing class of this hemisphere has a direct interest in its fame and a filial love for its history, the ancient history of America will begin to be properly considered and written. The study of the history of their own country will give to the people of Central America a more refined patriotism and a character peculiarly its own. Probably the founding of the Empire of Copan was contemporaneous with that of Peru, that is to say, about the eleventh century of the Christian era. The Spanish conquerors invariably assert with pride that the indigenous nations of the continent spontaneously offered themselves as vassals of their king on the first intimation or hint given them. This is undoubtedly affirmed, to justify their invasions, but that such an assertion is false is inferred from the long wars that they always waged with the natives. The well-known expedition which marched against Copan was under the com- mand of Fernando Chaves, who came from Guatemala some time in April 1530, with an army of Europeans and native allies. He advanced not without opposition, through Jalpatagua, Mita, and Esquipulas, and then laid siege to this city. No person could point out the place of the victory of Fernando Chaves over the King of Copan. These inhabitants do not know that such a battle took place. The ignorance and apathy concerning the origin and history of the ancient monuments are really surprising. The owner of the adjoining ranch, Ornillos, who, being a prominent and wealthy gentleman in this part of the country, was supposed to be somewhat intelligent, does not seem to know anything at all. The Spaniards taught all of them to look with hatred and contempt on the Vv'orks of the natives, because the latter were not instructed in the mysteries of our holy religion, and it is still difficult to eradicate and destroy this feeling. Four or five leagues from Esquipulas is the place popularly called "Conquest" (see figure No. 26). It is a peninsula formed by the Lempa River, and in part by a brook which unites with that river to the south, the waters running through a very 'Galindo lias evidently interchanged the words for 7 and 8 here. The Maya word for 7 in the Books of Chilan Balarn (northern Yucatan) is uuc, and for 8, uaxac, which are very similar to the Chortl words Galindo gives for these two numbers, uaxikte and ukte respectively, only reversed. The te is an unessential difference, being only a particle used in counting, as pel and even te itself in northern Yucatan, for example, hunp'cl haab, i year; oxp'el hanb, 3 years, and te hunte Pop, the first of Pop. -The corresponding forms in the Books of Chilan Balam are: I, hunp'el; 2, cap'cl; 3, oxp'el; 4, canp'el; S, hop' el; 6, uacp'el; 7, uucp'el; S. uaxarp'el; g, holonp'el; and 10, Inhunp'el. DESCRIPTION OF COPAN BY GALINDO. 603 deep canyon in the vicinity, leaving to the south a single narrow entrance. After traveling some distance through it, a stone trench is encountered, with many quarried stones, which crosses the isthmus where it is not wider than 20 yards. After passing the trench, a plain of about 200 yards in diameter is found, surrounded by the river and elevated above it by perpendicular precipices more than 100 yards. The side opposite the peninsula is level with it, but at a distance of a rifle shot. As a point of defense it was excellent, the besieged having supplies of provisions and the means of drawing up water from the surrounding abyss. Bearing in mind the advancement the aborigines of Copan had made in art, science, and civilization, the coldest and most indifferent soul revolts against the barbarous conquerors who allowed even the memory of such an interesting nation to perish. It is due to the memory of the priests, who were themselves the victims of the Spanish rule, that, during the first years of the conquest, they were the strongest, or, it may well be said, the only defenders of the unhappy Indians, and they brought upon themselves, therefore, the universal hatred of the conquerors. Guatemala will always be proud of having had as one of its citizens that noble and untiring defender of the Indians, Bartolome de las Casas. A number of Indians driven from Copan after the conquest, founded, it is believed, the neighboring towns of Jocotan and San Juan Ermita. Camotan, a village nearer to Copan, was founded many years afterwards by natives of the Chorti nation, who emigrated from Tachaluya, in the State of Salvador. Copan continued to be inhabited, even after the conquest, but in a state of perpetual decadence. Some 75 years ago the cultivation of tobacco was brought from there to the plains of Santa Rosa, and the population gradually decreased to a village of three houses situated to the west of the Sesesmil Canyon, which formerly comprised the western suburb of the city. The site of the ancient capital is now entirely included in the lands of a chaplaincy founded by the curacy of Guatemala. I have the honor to express to you, citizen minister general, the assurance of my most humble respect and devotion. God, Union, Liberty. Juan Galindo. NOTE. Through a fortunate accident, when this volume was already on the press, the writer learned that Colonel Juan Galindo was born in Ireland, and was a Central American by adoption only. Thanks are due for this timely information to Dr. Don Policarpo Bonilla, former President of Honduras, and now Envoy Extraordi- nary and Minister Plenipotentiary on Special Mission from that republic to the government of the United States. Galindo, it seems, after his return from Copan in 1834, was sent bv the govern- ment of Central America to Great Britain to effect a settlement of the boundary dispute then pending between the two countries, but when he arrived in London the British government refused to receive him as a diplomatic agent in the negotiations on the ground that he was a British subject, having been born in Ireland, and was thus disqualified from representing the government of Central America. The whole correspondence, so far as it affected the United States government, to which Galindo first appealed before going to England, was published in United States Senate Documents, second session. Thirty-second Congress, 1852-18^^, Senate Document No. 27, pp. i-ij. Miss M. W. Williams, in her Anglo-American Isthmian Diplomacy, iSi^-igi^, also gives a brief summary of the affair. See Williams, 1916, PP- 33' 34- 6o4 THE INSCRIPTIONS AT COPAN. Squier describes Colonel Galindo as an intelligent Irishman (1855, p. 52), and gives the following biographical sketch and a bibliography of five titles (ibid., p. 390). Unfortunately it has been impossible to ascertain his real name from any of the several sources mentioned. "Galindo, John, an Irishman, who entered the service of the old Republic of Central America about the year 1827, received the rank of colonel in the army, was governor of the Department of Peten in Guatemala, subsequently named representative of the republic to the court of St. James, but was refused recognition on the ground of bemg a British sub- ject, and was finally killed in an Indian town in Honduras. He was far from being a close observer, nor was he a man of large information. He nevertheless was industrious, and gave the world many interestmg facts, coupled with crude speculations, on the states of Central America and the country in general. After Juarros, he was, I believe, the first to direct public attention to the ruins of Copan." APPENDIX XII. THE DISTRIBUTION OF THE SEVERAL BRANCHES OF THE MAYANCE' LINGUISTIC STOCK. William Gates. In the following pages I have endeavored to set forth three points, as follows: (i) To indicate the probable nature of the tongue spoken at Copan during the Old Empire, based upon all the evidence now available, including much new unpub- lished material in my own collection of Mayance manuscripts and rare, if not unique, copies of early published grammars, texts, doctrinas, reports, and the like. (2) To establish by means of geographical, phonological, grammatical and vocabulary comparisons, the proper relation of that language to the other members of the Mayance linguistic stock. (3) To collate briefly this evidence with the historical and chronological data worked out by Morley, the stylistic data accumulated by Spinden, and certain native traditional and historical data as yet only partly published. In figure 91 the distribution of the several Mayance linguistic branches is shown. This map is based upon Stoll's ethnographic map of Guatemala- and Thomas's linguistic map of Mexico and Central Am.erica,'' together with certain emendations of my own, indicated by unpublished material in my collection. The most important new contribution to the subject brought out by this study is the proof of the essential unity of the Cholti and the Chorti dialects as members of the same linguistic branch, a condition absolutely demanded by the archaeological evidence, but one which heretofore it has been impossible to admit on the linguistic side, because of the incorrect filiation of the Chorti with the Pokom group through Stephens's curious error in 1839, when he collected 21 Pokoman words from a Pokoman Indian in Chorti territory, i. e., at Zacapa. On the basis of this word-list, StoU later filiated Chorti with the Pokom branch, thereby creating a gap between the archaeologic and linguistic evidence which it has been impossible to account for. The true position of the Chorti, i. e., filiated with the Cholti, however, clears up this discrepancy and for the first time brings these two lines of evidence into agreement with each other. Our earliest information as to the language spoken in the vicinity of Copan comes from Palacio (1576), who after recounting the tradition that people from Yucatan had anciently conquered the provinces of Ayajal (probably Tayasal), Lacandon, Verapaz, Chiquimula, and Copan, adds that "it is certain that the Apay language, which is spoken here, is current and understood in Yucatan and the aforesaid provinces." In the name Apay we probably have the same as in Payaqui, given in Isagoge^ as applying to the Corregimiento of Chiquimula, "in which are the edifices of Copan." The authority here is a manuscript probably quoted in the unpublished part of Fuentes, but which I have not by me while writing, to verify. The great Balam Quiche, ninth king of Utatlan, is there given as the first king of those of 'Gates uses the term "Mayance" to designate the entire family of Maya dialects, drawing his analogy from the use of the word Romance in European linguistics to indicate the modern representatives of the old Roman or Latin linguistic stock. However, he still follows the general usage in speaking of " the Maya civilization, Maya art," etc. ^See Stoll, 1S84, map. 'See Thomas, 191 1, map. ■'See hagogc liijtorico, 1892, p. 348. 60s 6o6 THE INSCRIPTIONS AT COPAN. Payaqui, having united the two kingdoms, eleven reigns before the coming of the Spaniards. Upon the name Apay I find no further hght, the vocabulary of the Paya, in northeastern Honduras, being so completely diflFerent as to be out of con- sideration. The next reference, and the first Chorti vocabulary yet found, is in the Galindo report of 1834, the chance discovery of the original manuscript of which in my col- lection last summer (1919) led to the writmg of this Appendix. This list contains 51 words, including the first ten numbers, and is definitely good Chorti, having come from Copan itself. Fig. 91. — Map showing distribution of the several brandies of the Mayancc hnguistic stock. I. Maya. la, Maya lb, Itza (Icaichc, Santa Cruz) ic, Lacandon II. Tzental 2a, Tzental zb, Tzotzil 2f, Chontal 2d, Chanabal III. Cholti. 3a, Cholti 3^, Chorti IV. Mame. Ha, Mame 4b, Ixil 4.C, Aguacateca 4J, Solomeca 4t', Jacalteca 4/, Chuje 4g, Chicomucelteca 4k, Motozintleca V. Quiche. Sa, Quiche Sb, Cakchlquel 5 and arriving by largely different roads, Morley and I had both come to see this great culture as stretching across this "land of Tezulutlan," north of the central Guatemalan Altos — the Peten region — that is, through a tierra caliente, a broken country — the land of the great rivers. Of these great cities of the Old Maya Empire, Copan appears to me to have been the chief exoteric religious center, Palenque the esoteric and most sacred religious center, and Tikal as a non-sacred city larger than either, possibly the mart. The chronological and cultural unity of these three sites is unmistakable, and the whole district exactly maps this Cholti field we have been considering, and which is properly outlined on StoU's map, plus what we now call Itza territory, which is now linguistically Yucate- can Maya. At this point a further element enters. About the year 600 (we can only say) "something happened"; the culture breaks, and even the large cities were entirely abandoned, much the same sort of thing as seems to have taken place in Yucatan eight centuries later after the fall of Mayapan, and there due to a mere defeat in war, and also just what happened at a stroke at Tayasal when Ursua conquered that city in 1697. We know that after the fall of Mayapan the Itza abandoned their homes and went south to Peten, long before the Spanish period; and Morley's idea here, that this was a return to an ancient homeland, seems to rc.t most apt. What we do know and may count upon as certain is that with the break-up of the Old Empire the technique of the inscriptions failed som.ewhat, though the knowledge of the writing was not lost altogether. The thread is carried on for at least another 800 years of m.onum.ents, and then through chronicles and manuscripts — the codices — to past the time of the Spanish entry. Broadly speak- ing, Maya science did two things: it went north, and it deteriorated greatly, without, however, dying out entirely, even to this day. The contents of the different Maya manuscripts, the medical lore, and traces of astronomical learning show this. This does not mean that much was not preserved in the south also; indeed much was, and T am satisfied learning of even a higher philosophical character than that of the north. The south has not yet yielded what it holds — has only glimpsed it to us. But the technique of the architecture and the inscriptions did go north, and the later southern cities were much inferior in type, not only to those of the Old Empire, but also to those of the New Empire in the north. Returning once more to our linguistic theme, perhaps the most marked point is the separateness structurally of the Quiche-Pokom, the highland Guatemalan branches from the others; and the cultural gap in that region, from say 600 a. d. to perhaps iioo a. d., is also far greater than in the north. We go back through the Quiche manuscripts to origins, to mythology and cosmogony, no doubt, but with a cultural and historical gap. Balam Quiche, who united (new) Chiquimula to Utatlan, eleven generations before the Spanish Conquest, that is to say, about 1200, was the "ninth" in the Quiche line and probably the first historical personage, since the one before him was Hunahpu, the divine youth, who was the hero of the Popol Yuh; and two reigns before Hunahpri was Acxopil, who reigned 200 years! All this agrees not only with the failure of Ordonez to include a Quiche realm in his four divisions of the Votanide empire, but also with the linguistic evidence. If by "Yucatan" in the citation from Palacio at the head of this Appendix, we DISTRIBUTION OF THE MAYANCE LINGUISTIC STOCK. 615 understand "Laguna de Terminos," it fits in completely with what is said by Ordonez. Linguistically the language of Tezulutlan, or what we have heretofore called Cholti, is rooted in northeastern Chiapas, with the Tzental tongues; structurally closest stands probably the little-known adjoining Mame, next closest the north- ern Maya, and then last the highland pre-Conquest kingdoms of the Quiche and Pokom. The letter r is unknown to all the other Mayance, save the last two; and when they developed it, the same influences seem to have expelled the / from the by that time divorced far-eastern Cholti, and left them Chorti. Languages grow and diversify under the spur of up-springing civilizations. The separate growth of Quiche and Maya, therefore, would correspond to the growth of the two new kingdoms; and that the Maya should have stayed closer to the Old Empire tongue, the Tezulutleca one may call it, than the Quiche, is to be expected from the now estab- lished closer relationship between the Old and New Empires than between the Old Empire and the later Guatemala highland kingdoms, both in cultural tradition and sciences. Meanwhile the Tzental, Mame, and Cholti would live on, in the past, and changing or separating less — just as Iceland has done, compared say to Sweden. In closing I may say that if we disregard the traditions of the Votanide Empire, as pertaining rather more to the realm of mythology than to that of history, and if we do not carry history back of the actual Old Empire period, just as later we have to reject the events back of Balam Quiche, as probably mythological in nature, and if we posit the four kingdoms, as Ordonez gives them: (i) Tul-ha or Ocosingo, (2) Na-chan or Palenque, (3) Chiquimula or Copan, and (4) Yucatan or Tikal, as we have indicated; if we bring this period to an end with the events of about 600 A. D., followed by a dispersion and rebirth in the north, toward 900 or 1000, with Uxmal as the chief center; and the Nahuatl influx coming in about 1200, resulting in the growth of new Chichen Itza, followed by the events of the chronicles in the Books of Chilan Balam down to the fall of Mayapan, and the return south of the Itza to Tayasal, spreading the newly-grown Maya-Itza tongue over what may have been Tezulutleca in Old Empire days; and if meanwhile a historical connec- tion was maintained in the south of which we have no present knowledge, resulting in a Quiche rebirth somewhat later than that at Uxmal, becoming definitely his- torical with Balam Quiche, and his incorporation of the eastern Chiquimula sur- vival of the Copan branch — we will have stated little that is new in itself, but we will have brought together the ethnographic and linguistic evidence, with the his- torical and chronological evidence which has been growng under Morley's work, and with the cultural and very positive stylistic evideice which Spinden has so skillfully brought together. There remains to be mentioned one final point of no small importance. Prior to Mayance times, the Central American territory washed by languages generally classed in with northern South American, and now reduced to scantiest remains north of Nicaragua, namely, to the little Chiapanec "linguistic island," in the mountain knot of Chiapas, occupying a Imguistic position similar to that of the Pyrenees or Wales. And finally, that the Tezulutlan empire, starting from around Ocosingo, and perhaps by entry through the Usumacinta delta, as Ordonez relates, should then have spread first across the fertile northern regions through Peten to Golfo Dulce, leaving the mountain range and the Pacific slope to the earlier inhabi- ta/nts, is just as natural as it is that its dispersion should have later been down over these same northern plains into Yucatan, leaving the highland occupation, with the resulting historical Quiche kingdoms, to a later date. BIBLIOGRAPHY. Aguilar, p. Sanchez de. 1639. (See Sanchez de Aguilar, P.) Al.FORD, W. V. 1899. Fragments from prehistoric America. (Home magazine, Jan. 1899, p. 8-17.) Allen, E. A. 1885. The prehistoric world; or, vanished races. Cincinnati. Allen, G. M. 1910. (See Tozzer, A. M., and G. M. Allen.) Alston, E. R. 1879-1882. Mammalia. (Biologia Centrali-Amer- icana, zoology.) Ancona, E. 1878-1905. Historia de Yucat.in desde la epoca mas remota hasta niiestros dias. 5 vols. Merida. Arctowski, H. 1910-1913. Studies on climate and crops. (Amer- ican geograph. soc, bull., vol. xlii, p. 270-282, 481-495; vol. xliv, p. 598- 606, 745-760; vol. xlv, p. 1 17-13 1.) Arnold, C, and F. J. T. Frost. igog. The American Egypt. A record of travel in Yucatan. New York. AuBiN, J. M. A. 1893. (See Codex Aubin, 1576.) Bancroft, H. H. 1882. The native races. 5 vols. San Francisco. Beltran de Santa Rosa, Maria P. 1746. Arte de el idioma Maya. Mexico. 1859. Same. Ed. 2. Merida. Beuchat, H. 1912. Manuel d'archeologie Arnericaine. p. l-xli, 1-773. Paris. BODDAM WhETHAM, J. W. 1877. Across Central America. (185 pages. Copan mentioned on p. 168.) London. BownncH, C, P. 1900. The lords of the night and the tona- lamatl of the Codex Borbonicus. (American anthropologist, new set., vol. 2, p. 145-154.) 1901. A method which may have been used by the Mayas in calculating time. Cambridge. 1901, a. Memoranda on the Maya calendars used in the Books of Chilan Balam. (American anthropologist, new ser., vol. 3, p. 129-138.) 1901,^. On the age of Maya ruins. (American anthropologist, new ser., vol. 3, p. 697-700.) 1901, c. Notes on the report of Teobert Maler in Memoirs of Peabody museum, vol. ii, no. i. Cambridge. 1903. Notes on the report of Teobert Maler in Memoirs of Peabody museum, vol. ii, no. ii. Cambridge. BOWDITCH, C. P. 1903, «. A suggestive Maya inscription. Cam- bridge. 1904. Mexican and Central American an- tiquities, calendar systems, and his- tory. (Twenty-four papers by Ed- uard Seler, E. Forstemann, Paul Schellhas, Carl Sapper, E. P. Diesel- dorff. Translated from the German under supervision of C. P. Bowditch. U. S. bureau of American ethnology, bull. 28.) Washington. 1906. The temples of the cross, of the foliated cross, and of the sun at Palenque. Cambridge. 1909. The dates and numbers of pages 24 and 46 to 50 of the Dresden Codex. (Putnam anniversary volume, p. 268- 298.) New York. 1910. The numeration, calendar systems and astronomical knowledge of the Mayas. Cambridge. Bowditch, C. P., and F. C. Lowell. 1891. Explorations in Honduras. Boston. Bradford, A. W. 1841. American antiquities and researches into the origin and history of the red race. New York. Bransford, J. F. 1882. Report on explorations in Central America in 1881. (Smithsonian insti- tution, report for 1882, p. 803-825.) Brasseur de Bourbourg, C. E. 1857-1859. Histoire des nations civilisees du Mexique et de I'Amerique-centrale, durant les siecles anterieurs a Chris- tophe Colomb. 4 vols. Paris. 1861. (See Popol vuh. 1861.) 1866. Palenque et autres ruines de I'anci- cnne civilisation du Mexique. Paris. 1867. Rapport sur les ruines de Mayapan et d'Uxmal au Yucatan. (Archives de la commission scientifique du Mexique, tom. ii, p. 234-288.) 1867, a. Extracts from two lettets written on July 15, 1865, and Dec. 2, 1865. (Archives de la commission scien- tifique du Mexique, tom. ii, p. 298-311.) (Brasseur de Bourbourg's visit to Copan in 1863 is described in these letters.) 1 869-1 870. (See Codex Tro-Cortesianus.) Brinton, D. G. 1882. The Maya chronicles. (Brinton's Library of aboriginal American lit- erature, no. i.) Philadelphia. 1S82, a. American hero-myths. A study in the native religions of the westetn continent. Philadelphia. 617 6i8 BIBLIOGRAPHY. Brinton, D. G. 1882, b. Books of Chilan Balam. (Penn monthly, vol. xiii, p. 261-275; (Re- printed by the Numismatic and anti- quarian society of Philadelphia, p. 1-19.) 1885. The annals of the Cakchiquels. The original text, with a translation, notes, and introduction. (Brinton's Library of aboriginal American literature, no. vi.) Philadelphia. 1886. On the ikonomatic method of pho- netic writing with special reference to American archaeology. (Proc. American philos. soc, vol. xxiii, P- 503-514-) 1886, a. The phonetic elements in the graphic system of the Mayas and Mexicans. (American antiquarian and oriental journal, vol. viii, p. 347-357.) 1887. Critical remarks on the editions of Diego de Landa's writings. (Proc. American philos. soc, vol. xxiv, p. 1-8.) 1890. Essays of an Americanist. Phda- delphia. 1893. The native calendar of Central America and Mexico. A study in linguistics and symbolism. (Proc. American philos. soc, vol. xxxi, p. 258-314.) 1894. What the Maya inscriptions tell about. (Archzeologist, vol. ii, p. 325-328-) 1895. A primer of Mayan hieroglyphics. (Univ. Penn. pub., series in phi- lology, literature, and archaeology, vol. iii, no. 2.) 1896. The myths of the new world. Brooks, C. F. 1916. World-wide changes of temperature. (Geographical review, vol. ii, p. 249-255.) Carmablich, J. G. F. R. 1824. (See Hassel, G., and J. G. F. R. Carmablich, 1824.) Carpenter, F. G. 1892. Relics of past ages. Interesting dis- coveries in Honduras. (Boston Herald, July II, 1892.) Carrillo y Ancona, C. 1865. Estudio historico sobre la raza indigena de Yucatan. Vera Cruz. 1870. Disertacion sobre la historia de la lengua Maya 6 Yucateca. (Revista de Merida, p. 128.) 1871. Compendio de la historia de Yucatan. Merida. 1883. Historia antigiia de Yucatan seguida de las disertaciones del mismo autor relativas al propio asunto. Ed. 2. Merida. Catherwood, F. 1844. Views of ancient monuments in Cen- tral America, Chiapas, and Yucatan. London. Central America. 1853. Correspondence as to claims of Great Britain relative to the Mosquito Coast, Honduras, and Yucatan. U. S. Senate doc. 27, 32d Cong., 2d Sess., p. 1-13. Chamberlain, A. F. 1899. In memoriam: Daniel Garrison Brin- ton. (Jour. American folklore, vol. xii, p. 215-225.) Charnay, D. 1887. The ancient cities of the new world, being travels and explorations in Mexico and Central America from 1857-1882. Translated from the French by J Gonino and Helen S. Conant. London. 1903. Les explorations de Teobert Maler. (Societe des Americanistes de Paris, Journal, new ser., tom. i, p. 289-308.) Chilam Balam of Chumayel. 1913. The book of Chilam Balam of Chu- mayel, with introduction by G. B. Gordon. (Univ. Penn., museum, anthropological pubs., vol. v.) Phila- delphia. Codex Aubin 1576. 1893. Histoire de la nation mexicaine depuis le depart d'Aztlan jusqu'a I'arrivee des conquerants espagnols (et au dela 1607). Manuscrit figur- atif accompagne de texte en langue nahuatl ou mexicaine suivi d'une traduction en fran^ais par feu J. M. A. Aubin. Reproduction du codex de 1576. Paris. Codex Boturini. 183 1. Fac-simile of an original Mexican hieroglyphic painting from the col- lection of Boturini. (Kingsborough, E. K., Antiquities of Mexico, vol. i, no. 3.) London. Codex (Mapas) de Acolhuacan. (1889?) Mapas de Acolhuacan. Plate iii. Mappe de Tepechpan. Histoire synchronique et seigneuriale de Tepechpan et de Mexico. Codex Cortesianus. 1S83. Photographic et public pour la primiere fois par Leon de Rosny. Paris. (This is the first part of the Codex Tro-Cortesianus). Codex Dresdensis. 1880. Die Maya-Handschrift der Konig- lichen Bibliothck zu Dresden; her- ausgegeben von Prof. Dr. E. Forst- emann. Leipzig. 1S92. Same. Ed. 2. Dresden. Codex Mendoza. 183 1. Copy of the collection of Mendoza preserved in the Selden collection of manuscripts, in the Bodleian library at Oxford. (Kingsborough, E. K., Antiquities of Mexico, vol. i, no. i.) London. BIBLIOGRAPHY. 619 Codex Peresianus. 1864. Manuscrit (lit Mexicainc No. 2 de la Bibliotlieque Imperiale. Photo- graphic par ordre de S E. M. Diiruy, Ministre de I'lnstruction Publiqiie, President de la Commission Scienti- fiqiie de Mexiqiie. Paris. 1S87. Manuscrit hieratique des anciens Indiens de I'Amerique Centrale conserve i la Bibliotheque nationale de Paris, avec une introduction par Leon de Rosny. (Public en coleurs. Ed. 2.) Paris. 188S. Same as 1887, but not in colors. 1909. Redrawn and slightly restored, and with the coloring as it originally stood, so far as possible, given on the basis of a new and minute examina- tion of the codex itself. Drawn and edited by William Gates. Point Loma. Codex Rios. 1900. (See Codex Vaticanus 3738.) Codex Telleriano-Remensis. 1S99. Manuscrit mexicain du cabinet de Ch. M. Le Tcllier, archeveque de Reims a la Bibliotheque nationale (Ms. mexicain, no. 385) reproduit en photochromogiaphie aux frais du Due de Loubat. 1899. Codex Tro-Cortesianus. 1892. Codice Maya denominado Corte- siano que se conscrva en el Museo arqueologico nacional (Madrid). Reproduccion fotocromolitografica ordcnada en la misma forma que el original hccha y publicada bajo la direccion de D. Juan de Dios de la Rada y Delgado y D. Jeronimo Lopez de Ayala y del Hlerro. Madrid. (To this should be added the plates re- producing the Codex Troanus taken from Abbe C. E. Brasseur de Bour- bourg's Manuscrit Troano. Paris. 1869-70. These two codices are parts of one original.) Codex Vaticanus 3738. 1900. II manoscritto messicano vaticano 3738 detto il Codice Rios. Ripro- dotto in fotocromografia a spese di sua eccellenza il duca di Loubat per cura della biblioteca vaticana. Roma. Codice Pi^rez. Cronologia antigua de Yucatan y examen del metodo con que los Indios contaban el tiempo; sacada de varios documcntos antlguos por Don Juan Pio Perez. (Manuscript in library of museum of University of Pennsylvania, ji. 1-258.) CoGOLLUDO, D. Lopez de 168S. Historia de Yucathan. Madrid. 1842. Same. Ed. 2, 2 vols. Campeche. 1867-186R. Same. Ed. 3, 2 vols. Merida. Cook, 0. F. 1909. Vegetation affected by agriculture in Central America. (U. S. bureau of plant industry, bull. 145.) CoRONEI,, J. 1620. Arte en lengua de Maya. Mexico. Cresson, H. T. 1892. Remarks upon the graphic system of the ancient Mayas. (Science, vol. XX, p. 25, 26.) 1892, a. The antenns and sting of Yikilcab as components in the Maya day signs. (Science, vol. xx, p. 77-79.) 1892, /;. Phonetic value of the Ch'i glyph in the Maya graphic system. (Science, vol. XX, p. 101-102.) Cui.lN, S. 1900. Bibliography of Dr. Brinton. (Brin- ton memorial meeting. Report of the memorial meeting held Jan. 16, 1900, under the auspices of the American philosophical society by 26 learned societies in honor of the late Daniel Garrison Brinton, M. D., p. 42-67.) Philadelphia. Dellenbaugh, F. S. 1901. The North Americans of yesterday. (Brief mention of Copan on p. 242, 35I.) Dieseldorff, E. p. 1904. Two vases from Chama, by E. P. Dieseldorff, Eduard Seler, and E. Forstemann. (U. S. bureau of Amer- ican ethnology, bull. 28, p. 635-666.) Duron, R. E. 1909. Las ruinas de Copan. (San Salvador. Museo nacional, Anales, tom. 4, p. 100-109.) Eisen, G. 1893-1S94. RuinsofCopan. (Great divide, vol. x.) Fernandez, L. 1881. San Salvador y Honduras el aiio 1576. Con prefacio y notas del Dr. A. Frantzius . . . tr. del ale- man por Don Manuel Carazo. (Col- lcci6n de documentos para la historia de Costa Rica, tom. i, p. 1-52.) 1892. Lenguas indigenas de Centro-Amer- ica en el siglo XVIII, segun copia del Archivo de Indias, hecha por el licenciado don Le5n Fernandez y publicada por Ricardo Fernandez Guardia y Juan Fernandez Ferraz para el 9° Congreso de Amcricanistas. San Jose de Costa Rica. Fewkes, J. W. 189S. Archaeological expedition to Arizona in 1895. (U. S. bureau of Amer- ican ethnology, 17th annual report, 1S95-96, p. 519-744.) 1904. Two summers' work in pueblo ruins. (U. S. bureau of American ethnol- ogy, 22d annual report, 1900-01, p. 1-195.) Fi.oRES, Fray I. 1753. Arte de la lengua metropolitana del reyno Cakchiquel 6 Guatemalteco. Guatemala. Forstemann, E. W. 1880. (See Codex Dresdensis.) 1887. Zur Entzifferung der Mayahand- schriften. Dresden. (For transla- tion, see U. S. bureau of American ethnology, bull. 28, p. 393-472.) 1892. (See Codex Dresdensis.) 620 BIBLIOGRAPHY. FoRSTEMANN, E. W. 1894. Die Mayahieroglyphen (Globus, Bd. XLVI, p. 78-80.) 1901. Commentar zur Mayahandschrift der Koniglichen offentlichen Bibliothek zu Dresden. Dresden. (For transla- tion, see E. W. Forstemann, 1906.) 1902. Der zehnte Cyklus der Mayas. (Globus, Bd. l.\.\.Kii, p. 140-143.) 1902, a. Commentar zur Madrider Maya- handschrift. Dantzig. 1902. b. Eine historische Maya-Inschrift. (Globus, Bd. Ixxxi, p. 150-153.) 1903. Commentar zur Pariser Mayahand- schrift. Dantzig. 1904. Aids to the deciphering of the Maya manuscripts. (U. S. bureau of Amer- can ethnology, bull. 28, p. 393-472.) 1904, a. Die Stela J von Copan. (Globus, Bd. Ixxxv, p. 361-363.) 1906. Commentary on the Maya manu- script in the Royal public library of Dresden. (Harvard univ., Peabody museum papers, vol. iv, no. 2, p. 49-266; translation of E. W. Forste- mann, 1901.) Franck, H. 1916. Tramping through Mexico, Guate- mala, and Honduras. (Brief mention of Copan on p. 291, 292.) New York. Frantzius, A. VON. 1873. San Salvador und Honduras im Jahre 1576. (Amtlicher Bericht des Licen- ciaten Dr. Diego Garcia de Pal- acio.) Berlin. Frost, F. J. T. 1909. (See Arnold, C, and F. J. T. Frost.) FuENTES Y Guzman, F. A. de 1689. Historia de Guatemala, 6 Recorda- cion Florida. Pt. 2, 4 vols. (Photo- graphic reproduction by W. Gates of manuscript copy by Mariano Padilla. Original Fuentes y Guz- man manuscript now in the archives of the cabildo of Guatemala city, Guatemala.) 1882-1883. Historia de Guatemala 6 Recorda- cion Florida escrita el siglo xvii . . . Publicada por primera vez con notas e illustraciones. D. Justo Zaragoza. Pt. I, 2 vols. Madrid. Gaceta Oficial de Honduras. 1868. Relacion importante. (Gaceta oficial de Honduras, torn. 6, no. 91, p. 1-4, Nov. II, 1868.) Galindo, J. 1834. 1835- Informe de la Comision Cientifica formada para el reconocimiento de las Antiguedades de Copan, por decreto de 15 de Enero de 1834 del c? gefe supremo del Estado de Guate- mala, Dr. Mariano Galvez. (Manu- script Gates collection.) (For trans- lation, see Appendix XI.) Central America. (London literary gazette and journal of belles lettres, arts, sciences, no. 965, p. 456, 457.) Galindo, J. 1835, a. 1836. 1836, a. Gallatin, Albe 1845. The ruins of Copan, in Central America. (Archaeologia Americana. Transactions and collections of the American antiquarian society, vol. ". P- 543-550-) Unelettreen36pages datee de Copan avec dix dessins assez bien executes. (Bulletin de la societe de geo- graphic, deuxieme serie, torn, v, p. 253-291. Paris. See also Walckenaer, Baron, and others.) Notions transmises par M. Juan Galindo, officier superieur de I'Amer- ique Centrale, sur Palenque et autres lieux circonvoisins. (Anti- quites mexicaines, tom. i, p. 67-76.) Paris. Gann, T. 1900. 1919. Garcia Pelaez, 1S51-1852. Gates, W. E. 1900. 1910. Gatschet, a. S. 1897. 1902. GiNZEL, F. K. I906-I9I4. Godman, F. D. 1897-1904. Goldenweiser, I9I6. rt. Notes on the semi-civilized nations of Mexico, Yucatan, and Central Amer- ica. (Trans. American ethnological society, vol. i, p. 1-352.) New York. Mounds of Northern Honduras. (U. S. bureau of American ethnology, 19th annual report, 1897-98, pt. 2, p. 661-692.) The Maya Indians of southern Yu- catan and northern British Hon- duras. (U. S. bureau of American ethnology, bull. no. 64.) F. de p. Memorias para la historia del anti- guo reino de Guatemala. 3 vols. Guatemala. The Maya and Tzental calendars. Comprising the complete series of days, with their positions in the month, for each one of the fifty-two years of the cycle, according to each system. Cleveland. Commentary upon the Maya-Tzental Codex Perez. (Harvard univ.. Pea- body museum papers, vol. vi, no. I, p. 1-64.) Die Erforschung der Ruinen von Copan in Honduras. (Globus, Bd. Ixxi, p. 99.) The hieroglyphic stairway, ruins of Copan. By George Byron Gordon. (A review.) (American anthropolo- gist, new set., vol. 4, p. 302, 303.) Handbuch der mathematischen und technischen Chronologic. 3 vols. Leipzig. (See Salvln, O., and F. D. Godman.) A. A. Diffu.sion vs. independent origin: a rejoinder to Professor G. Elliot Smith. (Science, new ser., vol. xliv, P- 53I-533-) BIBLIOGRAPHY. 621 Goodman, J. T. 1897. 1905. Gordon, G. B. 1896. iKgS, a. 1898, h. 1901 1902 1902, a. 1902, h. 1909. 1913- 1914. 1916. igit The arclialc Maya inscriptions. (Bio- logia Centrali-Americana, aiclKcol- ogy. Appendix.) London. Maya dates. (American anthropolo- gist, new ser., vol. 7, p 642-647.) Prehistoric ruins of Copan, Hon- duras. A prehminary report of the ex.plorations by the museum, 1891- 1S95. (Harvard univ., Peabody museum memoirs, vol. i, no. i, p. 1-48. For Spanish translation see G. B. Gordon, 1914.) The mysterious city of Honduras. An account of recent discoveries in Copan. (Century, vol. Iv, p. 407- 419-) Researches in the Uloa Valley, Hon- duras. Report on explorations by the museum, 1896-97. (Harvard univ., Peabody museum memoirs, vol. i, no. 4, p. 93-136. For Spanish translation, see G. B. Gordon, 1914.) Caverns of Copan, Honduras. Re- port on explorations by the museum, 1896-97. (Harvard univ., Peabody museum memoirs, vol. i, no. 5, p. 137-148. For Spanish translation see G. B. Gordon, 1914) The ruined city of Copan. (American geograph. soc, bull., vol. xxxi, p. 39- SO.) The vponderful city of Copan. (Daily Picayune, New Orleans, La. Oct. 20, 190 1.) The hieroglyphic stairway, ruins of Copan. Report on explorations by the museum. (Harvard univ., Pea- body museum memoirs, vol. i, no. 6, p. 149-186. For Spanish translation see G. B. Gordon, 1914.) On the interpretation of a certain group of sculptures at Copan. (Amer- ican anthropologist, new ser., vol. 4, p. 130-143.) On the use of zero and twenty in the Maya time system. (American an- thropologist, new ser., vol. 4, p. 237- 27S0 Conventionalism and realism in Maya art at Copan, with special reference to the treatment of the macaw. (Putnam anniversary vol- ume, p. 191-195.) New York. (See Chilam Balam of Chumayel.) Ruinas prehistoricas de Copan. Tra- ducido del Ingles por Juan A. Soto Mayor. (El Comercio, San Pedro Sula, Honduras, Febr. and March, 1914.) A contribution to the archeology of Middle America. (Holmes anni- versary volume, p. 137-141.) Wash- ington. The trail of the golden dragon. (Univ. Penn. museum journal, vol. ix, p. 29-38.) Gratacai', L. p. 1885. Ruins at Pakmiue and Copan. (American antiquarian and oriental journal, vol. vii, p. 193-207.) GUARDIA, R. F., V J. F. Ferraz. Guernsey, S. J. 1919. GuTHE, C. E. 1920. Hamy, J. T. E. i886(?) (See Fernandez, L., 1892.) (See Kidder, A. V., and Guernsey.) S. J. A possible solution of the number series on pages 51 to 58 of the Dres- den Codex. (Harvard univ., Pea- body museum papers, vol. vi. No. 2. Essai d'interpretation d'un des monu- ments de Copan (Honduras). (De- cades americane, Memoires d'ar- cheologie et d'ethnographie ameri- caines, i'' and 2" Decades, p. 83-89.) Paris. Hassel, G., and J. G. F. R. Carmablich. 1824. Vollstandige und neueste Erdbc- schreibung von Reiche Mexico, Gua- temala und West Indien. Weimar. Helland-Hansen, B., and F. Nansen. 1916. Hewett, E. L 1911. Temperaturschwanlcungen des Nord- atlantischen Ozeans und in der Atmosphare. Einleitende Studien iiber die Ursachen der kllmatologi- schen Schwankungen. (Videnskabs- selskabet-Skrifter, Math. Naturv. Klasse, 1916, Art. 9, p. 1-341.) Two seasons' work in Guatemala. (Archaeological institute of America, bull., vol. ii, p. 1 17-134.) 1912. The excavations at Quirigua in 1912. (Archasological institute of America, bull., vol. iii, p. 163-171.) 1916. Latest work of school of American archaeology at Quirigua. (Holmes anniversary volume, p. 157-162.) Washington. Hilderbrandsson. 1916. Holden, E. 18S1. Holmes, W. H 1907. (Summary in Nature, vol. 97, p. 228, of Hilderbrandsson's article appearing in Kungl. Svenska Vetenskapsakede- mines Handlingar, Bd. 51, no. 8.) Studies in Central American picture- writing. (U. S. bureau of American ethnology, 1st annual report, 1879- 80, p. 205-245.) On a nephrite statuette from San Andres Tuxtla, Vera Cruz, Mexico. (American anthropologist, new ser., vol. 9, p. 691-701.) Honduras E.xpedition. 1891. Hough, W. 1914. The Honduras expedition. (Harpers weekly, vol. xxxv, no. 1823, Nov. 28, 1891, p. 943.) Culture of the ancient pueblos of the upper Gila River region. New Mexico and Arizona. Second Museum- Gates expedition. (U. S. national museum, bull. 87.) Washington. 622 BIBLIOGRAPHY. Howe, G. P. 1911. The Ruins of Tuloom. (Ameiican anthropologist, new. set., vol, 13, p. 539-550.) Hrdlicka, a. 1903. The region of the ancient "Chichi- mecs," with notes on the Tepecanos and the ruin ot La Quemada, Mexico. (American anthropologist, new ser., vol. .5, p. 385-440.) Humboldt, A. de. 1827. litat present de la Republiqiie de Centro-America ou Guatemala, d'apres des documents manuscrits. (Nouvelles annates des voyages, vol. 35, p. 2S1-330.) Paris. Huntington, E. 1905. Explorations in Turkestan, with an account of the basin of Eastern Persia and Sistan. (Carnegie Inst. Wash. pub. no. 26, p. 157-315-) 1907. The pulse of Asia. A journey in Central Asia, illustrating the geo- graphic basis of history. Boston. 191 1. Palestine and its transfoimation. Boston. 1913. Guatemala and the highest native American civilization. (Proc. Amer- ican philos. soc, vol. lii, p. 467-487.) 1914. The climatic factor as illustrated in arid America. . . with contributions by Charles Schuchert, Andrew E. Douglass, and Charles J. KuUmer. (Carnegie Inst. Wash. pub. no. 192.) 1915. Civilization and climate. New Haven. 1917. Maya civilization and climatic changes. (Proc. International con- gress of Americanists, 19th sess., p. 150-164, 1915.) Washington. 1918. The sun and the weather; new light on their relation. (Geographical review, vol. 5, p. 483-491.) IsAGOGE HISTORICO. 1S92. Isagoge histofico apologetico general de todas las Indias. Madrid, Joyce, T. A. 1913. (See Maudslay, A. P., and T. A. Joyce.) 1914. Mexican archaeology. New York. JUARROS, D. 1808. Compendio de la historia d( la ciudad de Guatemala. Guatemala. 1823. Same. English ed. London. 1857. Same. Ed. del museo guatemalteco, 2 vols. Guatemala. Kidder, A. V., and S. J. Guernsey. 1919- Archjeological explorations in north- eastern Arizona. (U. S. bureau of American ethnology, bull. 65.) Wash- ington. KiNGSBOROUGII, E. K., 1831-1848. Antiquities of Mexico. 9 vols. Lon- don. Landa, D. de. 1864. Relation des choses de Yucatan. . . texte espagnol et traduction fran- (faise. . . par I'Abbe Brasseur de Bourbourg. Paris. Landa, D. de. 18S1. Relaci6n de las cosas de Yucatan. 1566. Reprint 1881. (Rosny, L. de. Ensayo sobre la interpretaci5n de la escritura hieratica de la America Central. Traduccion anotada y preccdida de un prologo por D. Juan de Dios de la Rada y Delgado. ,'\pcndice i, p. 71-114.) Madrid. 1900. Relacion de las cosas de Yucatan. (Colecci6n de documentos ineditos relatives al descubrimiento, con- quista y organlzaci6n de las antlguas posesiones espaiiolas de Ultramar, ser. 2, torn. 13, p. 265-411.) La Rochefoucauld, F. A. de. 1888. Palenque et la civilisation maya. Pans. Leemans, C. 1878. Description de quelques antiquites americaines conservees dans le Musee royal neerlandais d'antiquites, a Leide. (International congress of Americanists, 2d session, Luxem- bourg, 1877; Comptes-rendus, torn. 2, p. 283-302.) Lehmann, W. 1909. Methods and results in Mexican research. (Translated from the Ger- man by Seymour de Ricci. Paris. Originally published in the Archiv fiir Anthropologic, vol. vi, 1907, p. 113-168.) 1910. Ergebnisse einer Forschungsreise in Mittelamerika und Mexico 1907- 1909. (Zeitschrift fiir Ethnologic, Bd. 42, p. 687-749.) Leon, N. 1907. Bibhografra Mexicana del siglo xviii, seccion I. pt. 4. Mexico. (Con- tains part of Ordonez y Aguiar, ca. 1790.) Leon Pinelo, A. de. 1639. Relacion en el Consejo Real de las Indias, sobre la pacificaci6n y poblaci6n de las provincias del Manche; Lacandon, que pretende hazer Don Diego de Vera Ordonez de Villaquiran. Le Plongeon, A. 1885. The Maya alphabet. (Scientific .American, supplement, vol. xix, no. 474. P- 7572-7S73-) 1886. Sacred mysteries among the Mayas and the Quiches, 11,500 years ago. New York. Lizana, B. de. 1893. Historia de Yucatan. Devocionario de Ntra. Sra. de Izamal y conquista espiritual. (Reprint, including fac- simile of title-page of original edition, Valladolid, 1633.) Mexico. Lopez de Ayala y del Hierro, J. 1892. (See Codex Tro-Cortesianus.) Lopez de Cogolludo, D. (See Cogolludo, D. Lopez de, 16S8.) LOUBAT, J. F. Due DE. 1899. (See Codex Telleriano-Remensis.) 1900. (See Codex Vaticanus 373S.) BIBLIOGRAPHY. Lowell, F. C. 1891. LUMHOLTZ, C. 1902. Maler, T. I90I. (See Bowditch, C. P., Lowell ) and F. C. Unknown Mexico. 2 vols. New York. Researches in the central portion of the Usumatsintia valley. Report of explorations for the museum, 189S- 1900. (Harvard univ., Peabody mus- eum memoirs, vol. ii, no. I, p. I-75.) 1903. Researches in the central portion of the Usumatsintia valley. Reports of explorations for the museum. (Harvard univ., Peabody museum memoirs, vol. ii, no. 2, p. 77-208.) 1908. Explorations of the upper Usuma- tsintia and adjacent region. Altar de Sacrificios; Seibal; Itsimte-Sac- luk; Cankuen. Reports of explora- tions for the museum. (Harvard univ., Peabody museum memoirs, vol. iv, no. I, p. 1-49.) 1908, a. Explorations in the department of Peten, Guatemala, and adjacent region. Topoxte; Yaxha; Benque Viejo; Naranjo. Reports of explora- tions for the museum. (Harvard univ., Peabody museum memoirs, vol. iv, no. 2, p. 53-127.) 191 1. Explorations in the department of Peten, Guatemala. Tikal. Report of explorations for the museum. (Harvard univ., Peabody museum memoirs, vol. v, no. I, p. 1-91.) Mappe de Tepechpan (See Codex (Mapas) de Acolhuacan.) Maudslay, A. C. and A. P. 1899. A glimpse at Guatemala and some notes on the ancient monuments of Central America. London. Maudslay, A. P. 1886. Exploration of the ruins and site of Copan, Central America. (Proc. Royal geographical society, new ser., vol. viii, p. 568-595, plate and map opposite p. 608.) London. 1889-1902. Archaeology. (Biologia Centrali- Americana, 4 vols.) London. 1892. The ancient civilization of Central America. (Nature, vol. 45, p. 617- 622.) Maudslay, A. P., and T. A Joyce. 1913. A study of Maya art. By Herbert J. Spinden. (Review, in Current an- thropological literature, vol. ii, p. 238-247.) Means, P. A. 1916. Some objections to Mr. Elliot Smith's theory. (Science, new ser., vol. xhv, p. 533-534.) 1917. History of the Spanish conquest of Yucatan and of the Itzas. (Harvard univ., Peabody museum papers, vol. vii.) Cambridge. 1917, a. A survey of ancient Peruvian art. (Trans. Connecticut acad. arts and sciences, vol. 21, p. 315-442.) Means, P. A. 1918. Medina, J T. ■9I.V Meinshausen, I9I3. Membreno, A. 1895- 1897. 1912. Mercer, H. C. 1896. Molina Solis, 1896. 1897. 1904. MORLEY, R. K. 1918. MoRLEY, S. G. 1909. I9I0. 1910, a. I9II. 1911, a. 1911, b. 1911, c. 623 Las relaciones entre Ccntro America y Sud-America en la epoca prehis- t5rica. (Sociedad geografica de Lima, Boletin, torn, xxxiii, p. 151- 170.) Fray Diego de Landa inquisidor de los Indios en Yucatan. (Proc. International congress of Ameri- canists, i8th sess., 1912, pt. I, p. 484-496.) London, M. Uher Sonnen- und Mondfinsternisse in der Dresdener Mayahandschrift. (Zeit.schrift fiir Ethnologic, Bd. xlv, p. 221-227.) Hondurenismos. Tegucigalpa. Same. Ed. 2. Tegucigalpa. Same. Ed. 3. Notablemente corr. y aun. Mejico. The hill-caves of Yucatan. Phila- delphia. .T. F. Historia del descubrimiento y con- quista de Yucatan con una resefia de la historia antigua de esta penin- sula. Merida. El primer obispado de la naci6n mejicana. Merida. Historia de Yucatan durante la do- minacion Espariola. 3 vols. Merida. On computations for the Maya calendar. (American anthropologist, new ser., vol. 20, p. 49-61.) The inscriptions of Naranjo. north- ern Guatemala. (American anthro- pologist, new ser, vol. 11, p. 543- 562.) A group of related structures at Uxmal, Mexico. (American jour- nal of archaeology, 2d ser., vol. xiv, p. 1-18.) The correlation of Maya and Chris- tian chronology. (American jour- nal of archaeology, 2d ser., vol. xiv, p. 193-204.) The historical value of the books of Chilan Balam. (.'\merican journal of archaeology, 2d ser., vol. xv. p. 195-214.) Ancient temples and cities of the new world. Chichen Itza. (Pan- American union, bull., vol. xxxii, p. 453-468.) Ancient temples and cities of the new world Uxmal, the city of the Xlus. (Pan-American union, bull., vol. xxxii, p. 627-642.) Ancient temples and cities of the new world. Copan, the mother city of the Mayas. (Pan-American union, bull., vol. xxxii, p. 863-878.) 624 BIBLIOGRAPHY. MORLEY, S. G. I9I2. I913. I913, a. I914. I915. 1915, a. 1916. 1916, a. 1917. 1917, a. 1917, b. 1917, f. 1918. 1918, a. 1919. 1919, a. Morris, E. H 1919. QuirigLia, an American town 1,400 years old. (Scientific American, vol. cvii, p. 96, 97, 105.) Excavations at Quirigua, Guatemala. (National geographic magazine, vol. xxiv, pt. I, p. 339-361.) Archaeological research at the ruins of Chichen Itza, Yucatan. (Car- negie Inst. Wash. pub. no. 200, p. 61-91. Reports upon the present condition and future needs of the science of anthropology presented by W. H. R. Rivers, A. E. Jenlcs, and S. G. Morley.) Washington. Archaeology. (Carnegie Inst. Wash. Year Book, no 13, p. 233.) An introduction to the study of the Maya hieroglyphs. Washington. (U. S. bureau of American ethnol- ogy, bull. 57.) Archaeology. (Carnegie Inst. Wash. Year Book, no. 14, p. 343-346.) The supplementary series in the Maya inscriptions. (Holmes anni- versary volume, p 366-396.) Wash- ington. Archaeology. (Carnegie Inst. Wash. Year Book, no. 15, p. 337-341.) The ruins of Tuloom, Yucatan. The record of a visit of the Carnegie Institution Central American expedi- tion, 1916, to an important but little known ancient Maya city. (Ameri- can museum journal, vol. xvii, p. 190-204.) The rise and fall of the Maya civil- ization in the light of the monuments and the native chronicles. (Proc. International congress of Ameri- canists, 19th sess., Washington, 1915, p. 140-149. republished in Proc, Pan-American scientific congress, 2d sess., 1915-1916.) Washington. The hotun as the principal chrono- logical unit of the old Maya empire. (Proc. International congress of Americanists, 19th sess., p. 195-201, 1915.) Washington. Archaeology. (Carnegie Inst. Wash. Year Book, no. 16, p. 285-289.) The Guatemala earthquake. (Amer- ican museum journal, vol. xviii, p. 200-210.) Archaeology. (Carnegie Inst. Wash. Year Book, no. 17, p. 269-276.) Archaeology. (Carnegie Inst. Wash. Year Book, no. 18, p. 317-321.) Review of J. T. Goodman's life and work. (American anthropologist, new ser., vol. 21, p. 441-445.) Notice sur le Yucathan. The Aztec ruin. (American museum of natural history, anthropological papers, vol. xxvi, p. 1-108.) Nadaillac, J. F. A. DU Pouget, marquis de. 1883. L'Amerique prehistorique. Paris. Nansen, F. 1916. (See Helland-Hansen, Nansen.) 1843. Ordonez y ca. 1790. B., and F. Notice sur le Yucathan tiree des ecrivains espagnols. (Nouvelles an- nales des voyages, tom. 97. p. 30-52.) Paris. Aguiar. Historia de la creacion, conforme al sistema de la gentilidad ameri- cana. Manuscript. (See N. Leon, 1907, p. 1-272.) Orozco y Berra, M. 1864. Geografia de las lenguas y carta etnografica de Mexico. Precedidas de un ensayo de clasificacion de las mismas lenguas y de apuntes para las inmigraciones de las tribus. Mexico. Palacio, D. G. de. 1576. English Editions. See Squier, 1858, p. 241, 242. See Squier, i860, p. 88-97. See Maudslay, 1889-1902, vol. i of text, p. S-7- See Gordon, 1896, p. 45-48 Spanish Editions. See Pacheco, 1866, tom. vi, p. 37-39. See Gaceta Oficial (of Honduras) tom no. 91, p. 1-4, Nov. II, 1868. See Fernandez, L., 1881, tom. i, p. 1-52. See Gordon, 1914. French Editions. See Ternaux Compans, 1840, p. 42-44. See Notice sur le Yucathan, 1843, tom. xcvii, p. 38-40. German Edition. See Frantzius, 1873, p. 1-62. Pacheco, J. F. Relacion hecha por el Licenciado D. Palacio al Rey D. Felipe II, en la que describe la provincia de Guatemala, las costumbres de los Indios y otras cosas notables. (Pacheco, J. F., and others, ed. Coleccion de documentos ineditos relativos al descubrimiento, conquista y colonlzacion de las posesiones espafiolas en America y Oceania, tom vi, p. 5-40.) Madrid. 6, 1866. Peccorini, a. 1909. 1909, a. 1909, b. 1909, c. 1910. 1918. Penafiel, a. 1885. Excursionistas Salvadoreiios en Hon- duras. (Diario del Salvador, Agosto 9, 1909.) San Salvador. Ruinas de Copan. (Revista Centro America intelectual, p. 47-56.) San Salvador. Ligeros apuntes sobre el dialecto Camotan (Revista Centro America intelectual, p. 79-83.) San Salva- dor. Las ruinas de Copan (Diario del Salvador, Septiembre 10, 1909.) San Salvador. Copan. (Enciclopedia universal ilus- trada Europea-Americana, tom. xv, p. 368-373.) Conferencia sobre la civilizacion de los Mayas y las ruinas de Copan. San Salvador. Nombres geograficos de Mexico. Mexico. BIBLIOGRAPHY. 625 Penck, a. 1914. PlNART, A. 1881. Pineda, V. Ponce, A. 1872. POPENOE, W. I919. 1919, a. POPOL VUH. I86I. The shifting of the climatic belts. (Scottish geographical magazine, vol. XXX, p. 281-293.) Catalogue de livres rates et precieiix, maniiscrits et imprimes, principale- ment sur TAmeriqiie et sur les langues du moiide entiers, composant la bibliotheqiie de M. Alph. L. Pinart et comprenant en totalite la biblio- theque Mexico-Guatemalienne de M. I'Abbe Brasseur de Bourbourg. Paris. Historia de las siiblevaciones in- digenas habidas en el estado de Chiapas. Gramatica de la lengua tzel-tal. Chiapas. Relacion breve y verdadera de algunas cosas de las muchas que sucedieron al Padre Fray Alonso Ponce, en las provincias de la Nueva Espafia, siendo comisario general de aquellas partes. . . escrita por dos religiosos. (Coleccion de documentos ineditos para la historia de Espafia. vols, xvii and xviii.) Madrid. The avocado in Guatemala. (U. S. department of agriculture, bull. 743.) Washington. The useful plants of Copan. (Ameri- can anthropologist, new ser., vol. 21, p. 125-138.) Popol vuh. Le livre sacre et les mythes de I'antiquite americaine, avec les livres heroiques des Quiches . . . Texte quiche et traduction fran9aise, accompagnee de notes philologiques et d'un commentaire sur la mythologie et les migrations des peuples anciens de I'Amerique par I'Abbe Brasseur de Bourbourg. Paris. Prieto, a. 1873. Historia, geografia y estadistica del estado de Tamaulipas. Mexico. 191 1. Tribus indigenas de Tamaulipas. Victoria. Rada y Delgado, Juan de Dios de la. 1881. (See D. de Landa. 1881.) 1892. (See Codex Tro-Cortesianus.) Relaciones de Yucatan. 1898-1900. Relaciones historio-geograficas de las provincias de Yucatan. 2 vols. Madrid. (Coleccion de documentos ineditos relativos al descubrimiento conquista y organizacion de las anti- guas posesiones espafiolas de Ultra- mar., ser. 2, torn. II, 13.) Reynoso, Fray D. 1897. Arte y vocabulario en lengua Mame. Mexico, 1644; reprint, Paris, 1897. Rosny, L. de. 1876. Essai sur le dechifFrement de I'ecri- ture hieratique de I'Amerique Cen- trale. Paris. Rosny, L. de. 1881. Salvin, O. 1863. Codex Cortesianus. Manuscrit hieratique des anciens indiens de I'Amerique centrale. Paris. (See Codex Peresianus.) Description of photographs of ruins of Copan. The ruins of Copan. 1864. (Illustrated London news. Jan. 16, 1864. Perhaps contains first repro- ductions made from photographs of the monuments, taken by Salvin before 1863.) Salvin, O., and F. D. Godman. 1897-1904. Aves. (Biologia Centrali-Americana, zoology, 4 vols. London.) San Buenaventura, Fray G. 1684. Arte de la lengua Maya. Mexico. Sanchez de Aguilar, P. 1639. Informe contra idolorum cultores del Obispado de Yucatan. Madrid. Sapper, K. 1897. Das nordliche Mittel-Amerika nebst einem Ausflug nach dem Hochland von Anahuac. Reisen und Studien aus den Jahren 1888-1895. Braun- schweig. 1899. Uber Gebirgsbau und Boden des nordlichen Mittelamerika. Gotha. (Petermann's Mitteilungen Ergiin- zungsband xxvii, Erganzungsheft no. 127.) 1905. Der gegenwartige Stand der ethno- graphischen Kenntnis von Mittel- Amerika. (Archiv fijr Anthropologie, neue folge, Bd. iii, p. 1-38.) Saville, M. H. 1892. Explorations on the main structure of Copan, Honduras. (Abstract.) (Proc, American association for ad- vancement of science, 41st sess., p. 271-275.) Rochester. 1892, a. Explorations at Copan, Honduras. (Scientific American, supplement, vol. xxxiv, no. 870, p. 13904.) 1894. A comparative study of the graven glyphs of Copan and Quirigua. A preliminary paper. (Journal of Amer- ican folk-lore, vol. vii, p. 237-243.) 1916. The glazed ware of Central America, with special reference to a whistling jar from Honduras. (Holmes anni- versary volume, p. 421-426.) Wash- ington. SCHELLHAS, P. 1904. Representation of deities of the Maya manuscripts. Ed. 2. Translated by Selma Wesselhoeft and A. M. Parker. (Harvard univ., Peabody museum papers, vol. iv, no. I, p. 1-47.) Scherzer, K. ritter von. 1857. Travels in the free states of Central America: Nicaragua, Honduras, and San Salvador. (2 vols. London. Translation from the author's Wan- derungcn durch die Mittel-Ameri- kanischen Freistaaten.) 626 BIBLIOGRAPHY. Schmidt, J. 1S83. SCHRAM, R. 1908. SliLER, E. 1900. 1902-15 1904. 1904, a. Short, J. T. 1880. Simeon, R. 1885. Smith, G. E. 1915-1916, Die Steinbildcweike von Copan iind Quirigua. (English translation by A. D. Savage in 1883, published in Nevf York.) Kalendariographische und chronolo- gische Tafeln. Leipzig. Die Moniimente von Copan rind Quirigua und die Altar-Platten von Palenque. (Berliner Gesellschaft fur Anthropologie, Ethnologic und Urgeschichte. Verhandlungen, Jahr- gang 1899, p. (67o)-(738.) Published in Zeitschrift fiir Ethnologie, Bd. xxxi.) Einiges mehr iiber die Monumente von Copan und Quirigua. (Berliner Gesellschaft fijr Anthropologie, Eth- nologie nnd Urgeschichte. Verhand- lungen, Jahrgang 1900, p. (188)- (227.) Published in Zeitschrift fiir Ethnologie, Bd. xxxii.) Gesammelte Abhandlungen zur Am- erikanischen Sprach- und Alter- thumskunde. 3 vols. Berlin. The Mexican picture writings of Alexander von Humboldt in the Royal library at Berlin (U. S. bureau of American ethnology, bull. 28, p. 123-229.) Venus period in the picture writings of the Borgian Codex group. (U. S. bureau of American ethnology, bull. ^8, p. 353-391-) The North Americans of antiquity. New York. Dictionnaire de la langue nahuatl ou mexicainc. Paris. Pre-Columbian representations of the elephant in America. (Nature, vol. 96, p. 340,341, 425, 593-595) 1916-1917. The origin of the pre-Columbian civilization of America. (Science, new ser., vol. xliv, p. 190-195, vol. xlv, p. 241-246.) A. (See G. B. Gordon, 1914.) Soto-Mayor, J, 1914. Spinden, H. J. 1910. 1912. 1913- Table showing the chronological sequence of the principal monu- ments of Copan, Honduras. (Amer- ican museum of natural history.) The chronological sequence of the principal monuments of Copan (Hon- duras). (Proc, International congress of Americanists, 17th sess., Mexico, 1910. Resena de la segunda ses- sion, p. 357-363.) A study of Maya art. (Harvard Univ., Peabody museum memoirs, vol. vi.) Cambridge. Spinden, H. J. 1915- 1916. 1916, a. 1917. 19 17, a- 1917, h. 1917. c. Squier, E. G. isss- iSsS. i860. 1865. 1869. IS83. Notes on the archeology of Salvador (American anthropologist, new ser., vol. 17, p. 446-487.) Pre-Columbian representations of the elephant in America. (Nature, vol. 96, p. 592-593.) (Criticism of G. E. Smith, 1915-1916.) Portraiture in Central American art. (Holmes anniversary volume, p. 434- 450. Washington.) Ancient civilizations of Mexico and Central America. New York. (American museum of natural his- tory, handbook series, no. 3.) The invention and spread of agricul- ture in America. (American museum of natural history journal, vol xvii, p. 181-188.) The origin and distribution of agri- culture in America. (Proc, Interna- tional congress of Americanists, 19th sess., p. 269-276, 1915.) Washington. Recent progress in the study of Maya art. (Proc , International congress of Americanists, 19th sess., p. 165- 177, 1915.) Washington. Notes on Central America; particu- larly the states of Honduras and San Salvador. New York. The states of Central America; their geography, topography, climate, pop ulation, resources, productions, etc., comprising chapters on Honduras, San Salvador, Nicaragua, Costa Rica, Guatemala, Belize, the Bay Islands, the Mosquito shore, and Honduras interoceanic railway. New York. Collection of rare and original docu- ments and relations, concerning the discovery and conquest of America. Chiefly from the Spanish archives. No. I. Carta dirijida al Rey de Espana, por el Licenciado Dr. Don Diego Garcia de Palacio. . . Ano 1576. Being a description of the ancient provinces of Guazacapan, Izaico, Cuscatlan, and Chiquimula, in the audiencia of Guatemala: with an account of the languages, customs, and religion of their aboriginal inhabitants, and a description of the ruins of Copan. New York. Die Staaten von Central America, insbesondere Honduras, San Salva- dor, und die Moskitokiiste. In deutscher Bearbeitung hrsg. von Karl Andree. Neue Ausg. Leipzig. (German ed. of E. G. Squier, 1858.) Tongues from tombs; or the stories that graves tell. No. 5. Central America. (Frank Leslie's illustrated newspaper, June 26, 1869, p. 236.) Ancient laborers and princes of Grand Chimu and New Granada. (Frank Leslie's Magazine, vol. xv, no. 4, p. 468-480. April 1883.) BIBLIOGRAPHY. 627 Staue, W. 1919. Stephens, J. L. 1841. 1843. Stoll, O. 1S84. Some data about the pre-Hispanic and the now living Huastec Indians. (Mexico antlguo, torn, i, no. 3, p. 49-65 •) Incidents of travel in Central /\meri- ca, Chiapas, and Yucatan. 2 vols. New York. Incidents of travel in Yucatan. 2 vols. New York. 1888-1896. Strebei., H. 1884. 1885-1889. 1904. Zur Ethnographic dcr Repuhlik Guatemala. Zurich. Maya-Sprachen der FokomGruppe. 2 pts. Vienna, 1886, and Leipzig, 1896. Die Ruinen von Cempoallan im Staate Veracruz (Mexico). Mittcilungen iiber die Totonaken der Jetztzeir. (Naturwissenschattlicher Verein von Hamburg-Altona, Abhandlungen, Bd. viii, Abt. I.) Alt-Mexico. (Archaologische Bei- trage zur Kulturgeschichte seiner Bewohner, 2 vols.) Hamburg. Uber Ornamente auf Thongefassen aus Alt-Mexico. Hamburg. SwANTON, J. R. 191 1. (See Thomas, C Tapia Zentano, C. 1767. Noticia de la Mexico. Ternaux-Compans. 1838. 1911.) lengua Huasteca. Itineraire du voyage de la flotte du roi catholique a I'lle de Yucatan dans ITnde. Fait en I'an 1518, sous les ordres du capitaine general Juan de Grijalva, redige et dedie a S. A. (i), par le chapelain en chef (2) de ladite flotte. (Recued de pieces re- latives a la conquete de Mexique. Paris, p. 1-47.) 1840. Description de la province de Guate- mala, envoyee au roi d'Espagne en 1576. Traduit sur le manuscrit inedit que se trouve dans la biblio- theque de M, Ternaux-Compans. (Recueil de documents et memoires originaux sur I'histoire espagnoles dans I'Amerlque, p. 5-45) Paris. A study of the manuscript Truano. (U. S. departmeni of interior, con- tributions to North American ethnol- ogy, vol. V.) Aids to the study of the Maya codices. (U. S. bureau of American ethnology, 6th annual report, 1884- '85, P- 253-371O 1893. Are the Maya hieroglyphs phonetic? (American anthropologist, vol. vi, p. 241-270.) 1894. The Maya year. (U. S. bureau of ■American ethnology, bull. iS.) Washington. Thomas, C. 1882. 1888. Thomas, C. 1894, a. 1897. 1900. 1904. 1911. TozzER, A. M 1907. 1907, a. igii. 1913- 1916. TozzER, A. M 1910. Valentini, P 1879. 1880. Report on the mound explorations of the Bureau of ethnology. (U. S. bureau of American ethnologj', 12th annual report, i89o-'9i, p. 3-742.) Day symbols of the Maya year. (U. S. bureau of American ethnol- ogy, i6th annual report, i8g4-'95, p. 199-265.) Researches in the Uloa Valley, Hon- duras, and caverns of Copan, Hon- duras. . . By George Byron Gor- don. . . (A review.) (American archaeologist, vol. 2, p. 309-310.) Maudslay's archjeological vpork in Central America. (American anthro- pologist, new ser., vol. i, p. 552-561.) Mayan calendar systems. (U. S. bureau of American ethnology, 19th annual report, i897-'98, pt. 2, p. 693-819.) Mayan calendar systems. II. (U. S. bureau of American ethnology, 22d annual report, igoo-'oi, pt. I, p. 197-305.) Indian languages of Mexico and Central America and their geographi- cal distribution, by Cyrus Thomas, assisted by John R. Swanton. (U. S. bureau of American ethnology, bull. 44.) Washington. Ernst Forstemann. (American an- thtopologlst, new ser., vol. 9, p. 153-159-) A comparative study of the Mayas and the Lacandones. New York. Prehistoric ruins of Tikal, Guate- mala. (Harvard univ., Peabody museum memoirs, vol. v, no. 2, p. 93-135 ) A preliminary study of the pre- historic ruins of Nakum, Guate- mala. A report of the Peabody museum expedition. 1909-1910. (Harvard univ., Peabody museum memoirs, vol. v, no. 3, p. 137-197.) Pre-Columbian representations of the elephant in America. (Nature, vol. 96, p. 592.) (Criticism of G. E. Smith. 1915-1916.) , and G. M. Ali.en. The animal figures in the Maya codices. (Harvard univ., Peabody museum papers, vol. iv, no. 3, p. 273-372.) J- J; The katunes of Maya history. Translated from the German, by Stephen Salisbury; Jr. (Proc. Amer- ican antiquarian society. Oct. 21, 1879, no. 74, p. 71-117) The Landa alphabet; a Spanish fabrication. (Proc. American anti- quarian society, April 28,:i88o, no. 75. P- 59-91-) 628 BIBLIOGRAPHY. ViLLAGUTIERRE SoTO-MaYOR, J. DE. 1701. Historia de la conquista de la pro- vincia de el Itza, reduccion, y pro- gresses de la de el Lancandon, y otras naciones de Indios barbaros, de la me- diacion de el reyno de Guatimala, a las provincias de Yucatan, en la America septentrional. Priniera parte Escriuela Don Juan de Villagutierre Soto-Mayor. Madrid. Walckenaer, baron, and others. 1836. Rapport sur le concours relatif a la geographic et aux antiquites de I'Amerique centrale, par une com- mission composee de MM, Le Baron Walckenaer, de Larenau- diere, et Jomard, rapporteur. (Bul- letin de la societe de geographic, deuxieme serie, tom. v, p. 253-291. Same as Galindo, 1836.) Walsh, H. C. 1897. Copan: a city of the dead. (Harper's weekly, vol. xli, no. 2124, Sept. 4, 1897, p. 879-883.) Williams, M. W. 1916. Anglo-American isthmian diplomacy, 1815-1915. (Prize essays of Amer- ican historical association, 1914.) Work on Copan's Ruins. (New York Sun, July 1892.) XlMENEZ, F. 1857. Las historias del origen de los Indios de esta provincia de Guatemala, traducidas de la lengua Quiche al Castellano. . . por el R. P. Fran- cisco Ximenez. . . publicado por la primera vez, y aumentado con una introduccion y anotaciones por el Dr. C. Scherzer. Viena. INDEX Abandonment, climatic change suggested as cause of, 448 Cook's hypothesis, 452 Earthquake hypothesis rejected, 443 Huntington's hypothesis, 448 Latest cities to be affected, 444 Of cities, progressive, 434 Of Copan, period of, 23 Of most Old Empire cities (prior to Cycle 10), 430 Of region, gradual only, 442 Achi, defined, 611 Acropolis, growth by accretion, 7, 11 Date of, 9 Height of, 425 Rapidity of completion, 426 Vertical cross-section, 211 View from, 1 1 Adams, Percy, 27 Addition and multiplication, different positions of numeral modifier in each, 312 Agouti, 4 Agricidtural collapse most probable cause of end of Old Empire civilization, 462 Agriculture, Cook on, 454 Debilitation of soil from successive burnnigs, 454 Development in archaic period, 410 Factor in development of civilizatixjn, 414 Maya method of, 453 Aguacate, 3 Aguas Calientes, 333, 433, 438, 441, 459 Stela I, 135, 572. 585, 589 Aguilar, Garcia de, 544 Aguilar, Geronimo de, 476 Aguilar, Sanchez de, quoted, 42 Ahau-sign, as infix to Imix, 230 Inverted, used as glyph for kin or day, 589 Occurs 104 times out of 15S day-signs at Copan, 589 Unusual element in, 380. Ahpulha, meaning of word, 509. Napot Xiu (see Xiu, Napot) Ah-tz'un, meaning of, 479, 509 Ahuitzotl, death of, 37 Ah-ziyah, meaning of, 479 Akbal, not found as day-sign at Copan, 589 One of year-bearers, in New Empire and Codices 516, 518 Alford, 8 Allen, 8 Alphabet, general for all Mayance tongues, 611 Altar de Sacrificios, 391, 433, 436, 441, 459 Abandonment of, 456 Forms one site with El Pabellon, 571 Stela 4, 571, 583 Stela S, 583 Stela 7, 571, 584 Altar types, chronological sequence of, 395 Altars, associated, at times with hotun-markers, 214 Drum-shaped, no Of Early Period, description of, 395 Subordinate to stelae, 127 Truncated cone, 377 Amalin, Pedro, 17 Analtehes, Maya codices, 42, 43 Ancona, History of Yucatan, 526 Annals of Quauhtitlan, 42 Anniversary of 6 Caban 10 Mol, 336 Anona, 3 Ant-eater, 4 Apay language, 605, 606 Archaic culture, 422 Earlier origin of, 422 Vaster extension of, 422 Archaic horizon, in Middle America, 409 Archaic stel;e, characteristics of altars of, 62 Anthropomorphic figures, 422 Classification of, 125 Sculptures under Altars X and Y, 208 Archaisms, at Tikal, 54, 76 At Uaxactun, on Leyden Plate, on Tuxtla Statuette, 76 Architectural achievement, zenith of, 315 Arnold, quoted, 224 Artistic considerations, coercive influence of, 170 In glyph cutting, 265 Astronomical elements (in Stelaa 10 and 12), 133, 134, 143 Data, probably recorded here, 134 Astronomical phenomena, 33, 465 Athens of the New World, 431 Atlatl, Nahuatl weapon, 574 Aubin Codex (1576), 38, 41 Autochthonous origin of civilization doubtful, 406 Aztec calendar, 36 Codices, 36, et seq Glyphs, decipherment of, 41 Place names, 36 Time-count grafted on Maya count, 479, 514 Babylonian lunar calendar, parallel with Maya Supple- mentary series, 554 Bacalar or Bakhalal, 457, 502 Brinton on etymology, 457 Balam Quiche, present fixed Quiche history begins with, about 1 2th century, 614 United Chiquimula to Utatlan, 605 Ball Court a purely Nahuan institution, 514 Only two in Yucatan, 514 Bancroft, H. H., references to Copan, 21 Baraona, Sancho de, 16 Baskets, 4 Bat glyph, 45 Beans, 3 Benque Viejo, 433, 438, 441, 444, 458 Abandonment of, 457 Stela I, 278, 528, 572, 586 Berendt, Choi word- list, 610 Copies Stephens's erroneous Chorti word-list, 607 Beverages, 3 Bezerra, Bartolome, 16 Biologia Centrali-Americana, publication of, 22 Birds, s Bissextile shift in Maya New Year, 486 Bogran, 24 Bonilla, Policarpo, on Galindo, 603 Bcrbonicus Codex, 559 Boturini Codex, 41 629 630 INDEX. Bowditch, Charles P., bibliography, 32 Correlation of chronology, 529 Error noted, 342 Lords of the Night, 559 Method of notation, superseded those of Goodman, Forstermann, and Seler, 49 Numeration, Calendar Systems, and Astronomical Knowledge of the Mayas, 32 Periodicity of monuments, 565 Quoted, 139, 231 Work of, 24 Brasseur de Bourbourg, on Chcrti language, and Chiqui- mula kingdom, 607 Chronology, 525 Visits to Copan, 21 Brinton, D. G., bibliography, 32 Contributions to Middle American archaeology, 32 Doubling of katuns, 492 Error of, corrected, 477 "Father of Maya history," 32 Katun-wheel in Cogolludo, 481 Prevision, 35 Quoted, 33, 35, 404 Caban and Ik, at Quirigua second m importance only to Ahau, ;i6 Only day-signs in variable element in Initial Series, 516 Probably Old Empire year-bearers, 516, 518 Caban series of year-bearers, in Old Empire, 516, 518 Cabildo, museum in, at Copan, 106, 117 Cacao, 3 Cachapa, 147 Cache under Stela I, 424 Under Stela 3, 155 Under Stela 7, 105 Under Stela C, 346 Under Stela I, 177 Under Stela M, 278 Caches in foundation chambers at other sites, 278 Cakchiquel Annals, 461 Calabash, 3 Calculations of Maya dates, means for, 75 Calendar and chronology, probable swift development of, 407 Calendar-Round date, 294 Dating, uncertainty of, 236 Camargo, Diego, 545 Camotan, founded by Chorti emigrants from Copan, 603 Cancuen, 433, 438, 441, 459, 5^0 Altar 2, 572, 58s Stela 2, 156, 572, 585, 589 ^ Capital at Copan shifted from Group 9 to Main Structure in Middle Period, 429 Cardenas, Diego de, on number of natives in Cholti region, 608 Cardona, Antonio de, 546 Carpenter, A. W., 27 Carrillo y Ancona, Bishop, 526 Carving, excellence of, on Stela P, 116 Catherwood, drawings, 20 Accuracy of, 226, 352 Cauac-sign, 108 Cauac, winged, as tun-sign, 153, 163, 380, 566 At Chinikiha, 582 Used as tun-sign in New Empire, 393 Cayo, El, 220 Ceh, forms for, 66, 89 Center of population shifting, 129 Center, Mr., visit to Copan, 21 Ceramic finds in foundation chambers, 278 Ceremonial importance of last day of time periods, 397 Chachalaca, 5 Chac Mool, 520 Chamelicon Valley, 6, 431 Chaves, Hernando de 17, 544, ct seq. Chay, obsidian lances, 545 Chel New Year, according to Goodman, 491, 49S Chels move to Tikoch, 487 Chen, forms for, 66 Chiapanec, linguistic island, of Nicaraguan origin, 615 Chichen Itza, 34, 391, 433> 44i. 4SS, 5^7 Discovery of, 49S Earliest hotun-marker in New Empire, 573 Early occupation, 435 High Priest's Grave, 511, 512, 580 Occupations of, 502 Structure near casa principal, 512 Temple of Initial Series, 393, 512, 572, 586 Temple of two lintels, 358, 512, 573, 580 Temple at Ula, 512 Chichen Itza Initial Series lintel, 520, 527, 531 Lintel, Old Empire, 458 Re-used during Nahua period, 511 Same date on Old Empire lintel as on Stela 2 at Quen Santo, 459 Chichimec, History, 42 Chicomucelteca, 405 Chicozapote, 220, 441 Chilan Balam books, 578 Chile, 3 Chinikiha, 44 Altar date, 582 Chiquimula, 542 Chiquimula de la Sierra, conquest of, 16 Cholti, linguistic affiliations, 610, 611 And Chorti one tongue, 605, 610 Meaning of word, 610 Territory, extent, 609, 610 Vocabulary, compared with Maya and Quiche-Pokom, 6n Chorti language, 595 Not one of Pokom language group, 605 Vocabulary, earliest known, 606 Word-list in Galindo manuscript, 601 Christianity, official introduction of, in Yucatan, 489 Chronological sequence of altar types, 395 Of monumental types, 394, 395 Chronological systems, two in use at time of conquest, 512 Chronology, Maya, correlation with Christian dates, 465 Chronology of Copan, Spinden on, 53 Chumayel manuscript, 473, 578 Extract, translated, 479 Itza records in, 474 Quoted. 43, 458, 474, 485, 576 Chunvis, 441 Civilization, Maya, autochthonous origin of, doubtful, 406 Extension of, 403 Influence on Zapotec, Nahua, etc., 404 Origin of, 402 Transfer of after Old Empire to Yucatan, 614 Climate, 2 Climatic change suggested as cause of abandonment, 448 Climatic conditions, effect on erection of stelae, according to Huntington, 452 Hilderbrandsson on, 449 Penck on, 450 Climatic chart, Huntington quoted in explanation, 450 INDEX. 631 Cloisonne, Mexican, wide distribution of, 34 Cocom, 445 Driven to Tibulon and Sotuta, 487 New Year, according to Goodman, 491, 49S Cocom, Juan, one of Landa's informants, 29 Codices, month coefficients in, 513 (see under Dresden, Peresianus, and Tro-Cortesianiis) Coefficients, unusual forms of: 4, coefficient, unusual form of, 162 10, fleshless lower jaw used as zero in three places at Copan, 137 11, glyph for, 166 15, bar with death's head for, 166 15, composite numeral, possible reason for, 169 CogoUudo, History of Yucatan, 472 Katun wheel in, 472 On erection of stelae, 577 Quoted, 43, 480, 481 Cojolito, 5 Colonies of Copan, 43 1 Quirigua, 428 Colonization of Yucatan, 459 Period of New Empire, 458 Colors on monuments, 132 (see also Red paint) Comet, over Mexico in 1489, in Codices, 40 Conquest of Copan, told by Fuentes y Guzman, 17 Told by Juarros, 17 Constellation bands, 152 Conventionalization in Maya glyphs, 232, 259 Cook's studies in agricultural methods, quoted, 452, 455 Copal, incense, 4, 365 Copan: Colonies of, 431 (also see Los Higos, Paraiso, Rio Amarillo, Santa Rita, Quirigua) Important dates at, 516, 579 (see also 6 Caban 10 Mol) Location, i Name, different etymologies suggested, 15 Outlying groups, 13 Preeminence of, 431 Site clearing, 58 Site ol, 6 Supplementary Series at, 551 Valley, intensive occupation of, 14, 25 Copan, Cabildo, museum in, 106, 117 Copan Calel, 17, 544 Copan, divisions of site: Court of Hieroglyphic Stairway, 9 Eastern Court, 10 Mentioned by Galindo, 596 Great Plaza, 129 Date, 12, 177, 220, 425 Date determined by Stela I, 427 Hiatus in monuments during its construction, 130 Removals from Group 9 to, 419 Group 9: Favorable character of location, 417 First center of intensive occupation, 122 Most thoroughly excavated group at Copan, 417 Mound of Stela 7, earliest center of intensive occupation at, 106 Oldest settlement in valley, 417 Removals from, to Great Plaza, 419 Sketch map of, 124 Group 13, plan of, 144 Main Structure, chief center from Middle Period, 129 Earliest monument at, 424 General description, 7, 8 Most important center after 9. 1 1 .0.0.0, 163 Mound 9 probably oldest building left at Main Structure, 425 Copan, division of site — Continued. Middle Court, 12 Western Court, date of, 115 Copan Early Period (see also Earliest, etc.): Altars, description of, 127 Inscriptions, 53 Monuments, provenance and periodicity, 121 None in situ, 122 Re-used, list of, 418 Stela; and altars, tentative correlation of, 12S Steljc, list of dates, 419, 420 Copan, Middle Period: Expansion in, 437 History summarized, 210 Inscriptions of, 129 Last monument of, 427 Steloe, classified, 212 All date from hotun-endings, 214 All show Initial Series, 212 Copan, Hieroglyphic Stairway, 51, 243, 244, 257, 263 Classification of dates, 266 Discussion, 237 Hiatus in monuments during Its construction, 130 Landslide at, 268 Summary of chronology, 272 Copan monuments: Altar of Stela I, 598 Discuss'on, 174 Altar A', discussion, 68 Its stela probably lost, 128 Altars B' and C, discussion, 291 Altar D', discussion, 293 Altar F', discussion, 372 Now in Peabody Museum, 373 Altar Gi, discussion, 366 Altar Gs, discussion, 365 Altar G.3, discussion, 325 Altar G', discussion, 374 Altar H', 566, 597 Altars H' and I' removed at early date, 194 Discussion, 186 Inscription continued on Altai I', 189 Altar r, 597 Discussion, 189 (see Altar H') Altar J', discussion, 56 Altar K, discussion, 185 Altar K', discussion, 56 Altar L, discussion, 289 Altar L', discussion, 57 Altar M', discussion, 58 Altar N', discussion, 375 Altar O', discussion, 370 Altar P', discussion, 63 Altar Q, 597, Discussion, 326 Altar Q', discussion, 61 Altar R, discussion, 29S Alrar S, closely connected with Stehc A and B, 23c Discussion, 230 Suggested translation for, 230 Altar T, 344 Discussion, 334 Altar T', discussion, 375 Altar U, 516 Discussion, 299 Petrographic description, 463 Altar U', discussion, 376 Altar V, discussion, 296 Altar W, discussion, 364 63: INDEX. Copan monuments — Continued. Altar W, discussion, 330 Altar X, archaic sculpture beneath, 20S Discussion, 63 Altars X and Y, and Stehe 16 and 17. closely related, 65,91 Altars X, Y, and A', similarity of, 12S Altar Y, archaic sculpture beneath, 20S, 421 (see Altar X) Discussion, 66 Altar Z, discussion, 324 East Altar of Stela 5, 164, 16S, 56; Discussion, 163 Fragments V, discussion, 117 Fragment E', discussion, 33S Fragment S', discussion, 120 Fragment X', discussion, 36S Fragment Y', discussion, 209 Fragment Z', discussion, 378 Hieroglyphic Stairway of Mound 26: Date I, 241; Date 2, 243; Date 3, 244; Date 4. 245; Date 5, 24s; Date 6, 247; Date 7, 247; Date 8, 247; Date 9, 247; Date 10, 249; Date II, 251; Date 12, 252; Date 13, 252; Date 14, 253; Date 15, 254; Date 16, 255; Date 17, 255; Date 18, 256; Date 19, 256; Date 20, 256; Date 21, 236; Date 22, 257; Date 23, 257; Date 24, 258; Date 25, 261; Date 26, 262; Date 27, 264; Date 28, 264; Date 23, probably historical event, 258 Reappears at Quirigua 258 Date 24, probably earliest of late group, 269 Probable position of, at top of stairway, 269 Hier. Steps, south side Mound 2, discussion, 233 Stela I, i,6s Altar of, 174 Discussion, 161 Stela 2, 559 Discussion, 136 Resemblances to Stela 12, 140 Stela 3, 155, 559 Discussion, 155 Petrographic description, 463 Stela 4, discussion, 355 Stela 5, discussion, 204 East Altar of, 163 Relation of West Altar of, to Stela i, 181 West Altar of, 171 Stela 6, 566 Discussion, 183 Stela 7, 79, 567 Cross-section of mound of, 104 Excavation of mound of, 104 General discussion, 102 Mound of, first religious center at Copan, 418 Plan of mound of, 103 Stela 8, discussion, 340 Stela 9, 368 Comparison with Stela 15, 94 Discussion, 93 Stela 10, discussion, 141 Stela II, discussion, 369 Stela 12, 559 Discussion, 132 Resemblances to Stela 2, 140 Stela 13, 559 Discussion, 152 Copan Monuments — Continued. Stela 15 566, 568 Comparison with Stela 9, 94 Discussion 86 Stela 16, 567 Discussion, 84 Stylistic study, 92 Stela 17, discussion, 89 Stela; 16 and 17 and Altars X and Y closely related, 65, gi Stela 18, discuss on, 96 Shows first human figure in front view, 60 Stela 19, altars with, 145 Discussion, 143 Stela 20, 72, 74, 568 Discussion, 72 Oldest monument at Copan, 60, 77 Stela 21, discussion, 95 Stela 22, discussion, 68 Stela 23, discussion, 146 Stela 24, 80, 567, 568 Discussion, 78 Stela 25, discussion, 69 Stela A, 47, 566, 598 Closely connected with Stela B and Altar S, 230 Discussion, 221 Stela B, 598 Closely connected with Stela A and Altar S, 230 Discussion, 223 Stela C, discussion, 345 Stela D, 27, 59S Discussion, 230 Petrographic description, 463 Stela E, discussion, loi Probably brought from Group 9, 107 Stela; E and I and their altars, analogies, 113 Stela F, 566, 598 Discussion, 353 Stela H, 598 Discussion, 351 Stela I, 565 Discussion, 177 Relationship to West Altar of Stela 5, 181 Stela J, 244 Discussion, 194 Stela M, 559,560 Cruciform chamber under, 278 Discussion, 277 Stela N, 559, 560, 561, 597 Discussion, 279, 289 Stela P, 566, 597 Discussion, 114 Excellence of carving, 116 Temple II, 597 Discussion, 307 Temple iS, discussion, 371 Temple 2irt, discussion, 318 Possibly Temple of Venus, 319 Temple 22, 10 Beauty of, 317 Discussion, 316 Head from, at Peabody Museum, 317 Temple 26, discussion, 274 Copan, similar monuments to, at adjacent sites: Los Higos, Stela i, 384 {see Los Higos) Rio Amarillo, Altars I and 2, 382 (see Rio Amarillo) Copan, peccary skull from Tomb i, 379 Discussion, 379 Now in Peabody Museum, 379 INDEX. 633 Copan, Revlewlng-stand in Western Court, 579 Discussion, 321 Copan, Shrine R', discussion, 377 Copan, Summary of monuments relating to 6 Caban 10 Mol, 34.3 (see also 579) Summary of Stela' 4, C, F, and H, 358 Copper bells, 34 C5rdoba, Hernando de, 490 Cornette, visit to Copan, 21 Cornice, hieroglyphic, cases of, 276 At Quirigua, 387 Correlation of chronologies, probably assured within 49- day range, 505 Basic evidence for, 466 Historical bases for, 475 Proposition by Morley, 467 Sources for, 467 Correlation schools, American, 529, et seq. French, 525 German, 526 Guatemalan, 525 Cortes, Hernando, 549 In Yucatan, 490 Cotton, 3 Criteria of archaism, 77 Crops, 3 Cruciform underground chambers, 177 Under Mound 9, 121 Under Stela i, 161, 163 Under Stela 3, 155 Under Stela 7, 105 Under Stela I, 177 Under Stela M, 278 Under Stela C, 346 Other similar locations, 105 Cultural connection with Old World unsupported, 224 Cultural gap greater in Guatemalan history than in Yuca- tecan, 614 Culture intrusions in other localities, 34 Culture levels, 8 Curassow, 5 Current time-system grafted on elapsed time-system, 479, Curve of civilization at Copan, 432 Cutilca Cave, 600 Cycle renewal in codices, 41 Cycle 10, record of, 198 No contemporaneous monument found recording, 43S Cycle II dates, 250 Cycle-sign, 83 Daly, Cesar, visit to Copan, 21 Dardon, Juan Perez, 16 Date, anachronism of 13 Ahau 8 Xul developed by the Oxkutzcab chronicle, 510 Most important in Great Period, 429 Date glyphs, percentage of, in inscriptions, 399 Date, latest in southwestern Maya field, 438 Date (see also Important dates) Date of dedication of monuments, necessarily selected in advance, 398 Date, probable position of 13 Ahau 8 Kankin in Long Count, 510, 51S Date, 6 Caban 10 Mol, a historical date, most important at Copan, 297 Anniversary of, 336 Date, starting-point of Maya chronological system (4 Ahau 8 Cumhu) occurrence of, on monuments, 150 Dates, decimal notation for Christian, 477 Important group of, 344 Dates of Copan m.onnments, table, 587, et seq. Calculation of, 75 Contemporaneous on monuments, 26, 27 Decipherment of, 46 In lower half of Hieroglyphic Stairway, probably relate to historical events of Earli' Period, 268 Latest on one monumunt, is usually the contemporane- ous, 227 Probable percentage of error in decipherment of, 415 Dates, "prophetic," 27, 197, 198, 333, 342 Dating system, accuracy of, 27 Day, basic unit of time count, 465 Count (tonalamatl), inviolability of, 522 Positions shifted after 1200 A. D., 478 Day-sign, 83 Day-signs at Copan, list of all, 589, et seq. Death god, 558 Death's-head characteristics in numerals, 172 Glyph, 81 Decipherment, of Aztec glyphs, 41 Maya date, 46 Maya hieroglyphic writing, 28 Means for, 75 Probable percentage of error in, 415 Decipherment, Maudslay's contribution, 31 Phonetic school, 30 Decorated element, to avoid distortion, 203 Deer, 4 Deer hunts in Codex Tro-Cortesianus, 4 Deities, presiding over periods, 559 Delgado, Joseph, among Choles, 609 Destruction of monuments and buildings due to tropical vegetation, 442 Diaz, Padre Juan, with Grijalva, 519 Different plaza levels in Acropolis at Copan, 425 Diocese of Yucatan, 29 Dolores, location of, 609 Doves, 5 Drawings, scale of, 524 Difficulties in, 52 Dresden Codex, 589 Bowditch on contemporaneous date of, 519 Forstemann on contemporaneous date of, 519 List of lunar and solar eclipses, 556, 557 Lunar and eclipse calculations, 555, 557, 558, 561, 563 Meinshausen on, 556 Moon-glyph in, 554 Serpent numbers, 281 Venus-solar periods in, 35, 150, 182, 527 Year-bearers, 516, 517 Dress, 5 Drum-shaped altars, no Earliest center of intensive occupation, 106 At group 9, 122 Earliest center of population, 123, 125 Earliest certain contemporaneous texts in Old Empire, 41 1 Earliest deciphered date at Copan, 78 Earliest entry in Maya chronicles (9.0.0.0.0), 501 Earliest Great Period cities, 43S Earliest human form in front view, 97 Earliest monuments at Copan, 55 Stela 20, 77 At Main Structure, 424 Earliest stela at Copan, 60 Early and late groups of dates on Hieroglyphic Stairway, stylistic difference between, 270 Early Period at Copan (see Copan, Early Period) 634 INDEX. Early Period, monuments, list of re-used, 41S None in situ, 122 Provenance and periodicity, 121 Early Period stelae, altars, description of, 127 Inscriptions, 53 List of dates, 419, 420 Stelae and altars, tentative correlation of, 128 Early visits to Copan, Stephens, 2 Earthquake hypothesis as cause of abandonment, rejected, 443 Earthquakes, in Aztec Codices, 39 Eclipse phenomena in Supplementary Series, 401 Eclipses, 35, 465 In Aztec Codices, 39 Egyptian cultural connection unsupported, 224 Eisen, G., on Chorti language and on Copan natives, 607 El Cayo, 220, 391, 433, 438, 441 Stela I, 572 El Chicozapote, 220, 441 El Encanto, 441 El Mecco, 520 El Pabellon, 391, 433, 436, 441 Four monuments from, deciphered, 571 Site same as Altar de Sacrificios, 571 Stela I, 571, 583 Elapsed time concept, 282 Elapsed time in Maya counting, 48, 64 "Elephant-head" decoration, 28 "Elephant-trunk" element, 224 Elevation of Copan, i Encinas, Fernando, 546 Ending-sign (see Hotun and Lahuntun) Engraving of records practiced long before foundation of Copan, 416 Environment at Copan, i Errors in inscriptions, 280 In month-signs, very unusual, 332 Espiritu Santo Bay, 520 Esquipulas, surrender, 17 Excavations, objects found, 25 Exodus from Old Empire region, iwo-fold direction of, 4^7 Expansion at end of Early Period, 419 In Middle Period, 437 Of civilization under Old Empire, 434, 435 Extinction of Old Empire civilization, causes suggested, 442 Eye pierced by arrow in katun-wheel figure, 483 Eznab, resemblance of, to death's head, 81 Fall of Old Empire 442 Famine in 1533-34, Landa and Juan Xiu on, 509 Fauna, 4 Feather head-dresses, 5 Tassel decoration, 360 Feathers of turkey, macaw, quetzal, toucan, and parrot, 5 Ferraz, Lenguas Indigenas de Centro America, 610 Fibers, 4 Figures on two faces, monuments with, uncommon, 279 Fire-sticks, distribution in ancient America, 41 Flora at Copan, i Floras, 433, 438, 441, 444, 45S Abandonment of, 457 Stela I, 458, 528, 572, 586 Stela 2, 458, 52S, 572, 586 Florida, La, 422 Food staples, 3 Food-supply, national crisis in Great Period, 456 Probable effect on leligious ceremonies of time, 456 Forearms of human figures, position of, 141 Forestation, i Forests, present not primeval, 455 Botanical and zoological evidence for this, 455 Cook, corroborated by Whitford, 456 Forstemann, Ernst, 31, 78 Life and bibliography, 30 On correlation of chronology, 199 On moon-glyph in Dresden Codex. 554 Quoted, 527 Work on decipherment, 30 Fragments, inscribed, importance of preserving, 120 Frog, relation to Uo, uinal, and moon-glyphs, 232 Fruits, 3 Fuensalida, Padre, visit to Tayasal, 472 Fuente Albores, Juan Jossef de la, Choi vocabulary from Tila, 610 Fuentes y Guzman, chapters on Copan, 543 Digested copy by Mariano PadiUa, 543 Manuscript, 543 Quoted, 15 Full-figure glyphs, 231 Function of monuments, probable, 396 Galindo, Juan, a British subject born in Ireland, 603 Discovery of his report on Copan, 593 Drawings, 19 Excavations and finds, 11 Letters, 19 Manuscript, 18 Our treatment of the Indian, 594 Quarries, 6 Quoted, 132, 187 Gallardo, Pablo, 548 Gailatin, prints Stephens's erroneous Chorti word-list, 607 Gann, copies of wall paintings, at 'Fuluiim, 519 Garcia de Medina, Probanza, 488 Garcia Pelaez, cites Isagoge on the Payaqui, 607 Garnica, Diego Martinez de, 549 Garrido, Bartolome, 545 Gates, William, collection, 18, 593, 594 Maya post-conquest manuscripts in, 475 On Maya passage, 530 Quoted, 477, 482, 483 Translation and notes, Oxkutzcab Chronicle (p, 66), 508 Translation by, 479, 485 Translation of Fuentes y Guzman, 543 Geology of Copan, 1, 463 Glyph, defined, 50 Conventionalization, 259 For lahuntun and katun, 201 Value, 337 Glyph-blocks, defined, 50 Delineation, characteristics of earliest style, 393 Earliest style most extensive at Tikal, 393 Measurements of, as criterion, 117 Glyphs, full-figure, 231 Form variance in, 64 Inversion of, 322 Right to left reading of, 313 L^nusual order of, 156 God A, and number 10, 558 God D, and number 4, 55S God E and numbers i or 8, 558 God K, and number 7, 558 God N and number 5, 558 Godman and Salvin, work on fauna and flora, 4 Goldman, i.. A., revision of fauna nomenclature, 4 Goodman, J. T., work on decipherment, 30 Correlation of chronology, quoted, 530 Contradictions involved in his correlation, 531 Date for Palenque, 530 I INDEX. 63 s Goodman, J. T. — Continued. Error corrected, 242 Mistranslates Maya passage, 530 On Itza, Cliel, and Cocom New Year, 491, 49S, 523 On lahuntun-glyph, quoted, 566 On Old Empire year-bearers, 516 Quoted, 283 Review of his work, 31 Gordon, George B., 2 Date, reading of, for Altar X, 65 On inscriptions of Hieroglyphic Stairway, 241, et seq On monument nomenclature, 537 Quoted, 271 Gorgas, W. C, on malaria, 446 Gourd, 4 Great Cycle, etc., 349, 351 Length of, 281 Great Cycle glyphs, 376 Occurrences of, 281 Great Cycle, previous, 250 Great Cycle 19, 283 Great-Great Cycle glyph, 346, 347 Great-Great Cycle II, 283 Great-Great-Great Cycle I, 2S3 Great-Great-Great-Great Cycle i, 283 Great Period, earliest cities of, 438 Inscriptions of, 219 (see also Copan, Great Period) Grijalva, Juan de, 490 Expedition, 519 Guacas, hidden treasures, 549 Guatemala Highlands, settlement of by migration from the north, 459 Guava, 3 Guelva, Juan Sanchez de, 544 Guthe, Carl, 551, 558 On Supplementary Series, 563 Hacienda Grande, 6, 14 Plan of, 144 Hacienda Mirallor, 422 Archaic figures found at, 422 Halach vinic, Maya term for ruler, 487 Hardcastle, Mr., visit to Copan, 21 Head facing to right on Stela 2, reason for, 140 Head glyph, unusual position of, 167 Held, John, jr., 28 Hewett, E. L., 27 Hiatus in monumental sequence, possible explanation there- for, 130 Hieroglyphic Stairway, date, 24 Its beauty of composition, 261 (sec also Copan, Hieroglyphic Stairway) Hieroglyphic steps at other localities, 220 Hieroglyphic writing, study of, 232 Hilderbrandsson on climatic conditions, 449 Historical bases for a chronological correlati(m, 475 Data in inscriptions, 36 In Aztec Codices, 36-40 History in Maya manuscripts, 42, et cq. History of Copan, Early Period, Old Empire, 415 Great Period, 429 Middle Period, 423 Holactun, Temple of Initial Series, 250, 358, 391, 393, 411, 500, 512, 580 Holmes, W. H., 27 Study of tools, 5 Hondureiiismos, cited, 607 Honradez, La (see La Honradez) Hotim, defined, 74 Chronological unit in Old Einpire for 250 years, 127 Hotun-ending, earliest known, 398 Earlier of two, invariably contemporaneous date, 227 Later of two, "Future time," 227 Later of two, invariably of unusual importance, 22S Single, Invariably contemporaneous date, 227 Hotun-Glyph, 565 First identified by writer, 566 Widespread use of, 566 Hotun, lahuntun, or katun endings, priority among, 421 Hotun-markers, at Quirigua, 214 At Tikal and Nakum possibly painted, 437 At times associated Altars used as, 214 Custom at different cities, detailed statement, 568 Disuse of in later period, 364, 388 Erection of, 90, in In Early Period, 126 Missing, probable reason for, 216 Possibly originated at Copan, 420 Tabulated list of, 581 Use continued through Middle Period, 213 Hotun numeral coefficients, 75 Hotun-sign, unique variant of, 162 Howe, 573 Huasteca, 402, et seq. A Maya linguistic island, 403 No evidence of hieroglyphic writing among, 404 Separation of, before development of civ lization, 405 Huitzilipochtli, dedication of temple, in Aztec Codices, 39 Human figure, range of period, 395 Representation, 60 Hunnac Ceel, victory over Itza, 574 Hunter, Annie, drawings, 22 Drawings by, corrected, log, 115, 139, 142, 162, 302 Huntington, Ellsworth, theory of climatic changes and rainfalls, 448 Hypothesis, chief weakness of, 452 Ichmul, stela i, 580 Ik and Caban, only day-signs, in variable element in Initial Series introducing-glyphs, 516 At Quirigua second in importance only to Ahan, 516 Old Empire year-bearers, 516, 518 Imix not found as day-sign at Copan, 589 As part of month-sign Mac, 5S9 Important dates, other than Period Endings, 579, 582 9.0.0.0.0. recorded at several places, including Copan, Palenque, Books of Chilan Balam, 582 (see notes to App. vrii, 582, et seq ) 9. 14. 13 .4. 17 most important date at Quirigua, 25S, 273. 274 Possibly date of foundation ol Quirigua, 272 9. 15 .0.0.0 at various places, 584 10. 0.0. 0.0, not found as contemporaneous date, 5S6 9. 16. 12. 5. 17 most important date of Great Period at Copan, possible significance of, 429 Its anniversary commemorated, 430 6 Caban 10 Mol, at Copan, 516, 579 12 Caban 5 Kayab at Quirigua, 516, 579 6 Cimi 4 Tzec at Quirigua, 579 4 Eb 10 Yax, 9 Manik O Kayab, 3 Cauac 2 Pop at Naranjo, 579 2 Ik 10 Pax at Piedras Negras, 579 6 Ix 12 Yaxkin at Yaxchilan, 579 7 Men 18 Kankin at Piedras Negras, 579 Incense, 4 Inclusions in roc': masses, 463 In Altar Gi, 367 In Altar U, 300 Petrographic description of, 463 636 INDEX. Inclusions in rock masses 463 — Continued. In Stela 2, 140 In Stela 3, 157 Petrographic description of, 463 In Stela D, 233 Petrographic description of, 463 Indians, Galindo on our treatment of, 594 Initial Series, 34, 466 In stela;, 29 known at Copan, iii Name given by Maudslay, 31 System perfected at time of Tuxtla Statuette, 411 59 out of 178 known are at Copan, 391 Initial Series found on all Early Period Stelae, 126 At Copan nearly always record katun, lahuntun, or hotun-endings, 70, 74 Caban and Ik in introducing glyphs of, 516 Contemporary date on early stelse, 399 First discovery of in Yucatan, 500 Found on all iVIiddle Period Stel:e, 212 Latest at chief sites, 392 Latest known, 411 List of known, at all sites, 391 Method not used in latter part of Great Period at Copan, 307, 388 None at Seibal, or Nakum, 393 Only 3 known in New Empire, 393 Stelae with two, list of, 401 Unique introducing glyph, 144 Initial Series, terminal date, suppression of, 138, et seq. Inscribed lintels at other localities, 220 Inscriptions, Goodman's work on, 31 At Copan, constitute 40 per cent of all known, 391 Of Early Period, 53 Of Great Period, 219 Of Middle Period, 129 Inscriptions continuous on Altars H' and I', 189 On architectural features, 220 On stelae concluded on their associated altars, 182 Interlacing glyph bands, 19; Introducing glyph not at beginning, 356 In second position, 190, 191 Introducing glyphs without Initial Series, use confined to Copan, 347 Inverted L-shaped glyph panels, 151 Irregularity of glyph outline, best criterion of archaism, 77 Isagoge historico, cited, 605, 607, 5o8 Isla de Mugeres, 520 ItJ.a, 445 Migrate south to Peten, 4t'7, 614 Retained Old Empire time-count, 523 Itza chronology, its persistence, 473 Count, different from Xiu, 512 Records preserved, 473 New Year, according to Goodman, 491, 498 Itzamna, 558 On Codex Peresianus, 575 On Mayapan stela, 574 Itsimte, 391, 433, 437, 441 Abandonment of, 456 Four monuments deciphered, two doubtfully, all are hotun-markers, 571 Itsimte Monuments, Altar i, 434, 571, 584 Altar 2, 434, 571, 584 Stela I, 571, 584 Stela 2, 571, 584 Ix, not found as day-sign at Copan, 589 Ixkun, 391. 433. 43S, 44i Ixkun Monuments: Only one stela deciphered, not a hotun-marker, 571 Stela I, 559, 571, 5S5 Stela 2, 571, 585 Stela 5, 571, 585 Jade, mentioned by Galindo, 596 Objects of, below Stela 7, 10; Jaguar, 5 Jaguar-skin cloaks, at Palenque, on Chama vase, on Stela 20 at Yaxchilan, 5 Jalisco, objects from, at Chichen Itza, 34 Jaw, fleshless lov/er, sign for zero, 159 Jobitsina Cave, 600 Jocotan, founded by people driven from Copan, 603 Jocote, 3 Joyce, quoted, 406, 443 Correlation of chronology, quoted, 531 Error in Chichen Itza date, 531 Error in Tuluum date, 531 Other errors, 532 Juarros, History of Guatemala, 18 On distribution of Chorti, 607 Julian and Maya periods, similarity of, 48 Julian period, analogy with, 465 Kan series of year-bearers in Codex Tro-Cortesianus and Books of Chilan Balam, 517, 518 Introduced after 1200 A. D., 520, 521 Only series in use at time of Conquest, 518 Katun, 566 Katun anniversaries, 181, 336, 337 At Quirigua, iSl Katun-circuit, 483 Katun dates, correlation, 495 Katun-ending, earliest certain, 39S Katun-ending records, 184 Katun-endings only, found after 9. 19. 10. 0.0., 580 Katun, lahuntun, and hotun endings, importance of, 65 Katun, lahuntun, and hotun monuments, review, 578 Katun-sign, S3, 201 Katun-stones, erection noted by Landa and CoguUudo, 577 Katun-wheel in Cogolludo, 472 Its explanation, 4S3 Lists in Mani, Kaua, and Tizimln manuscripts, 482 Seen by Stephens in 1 841, 472 Katun II, commemorated by stelae at six different Copan groups, 423 Katun 13 Ahau 8 Kankin, 510, 51S Katun 13 Ahau 8 Xul, 510 Katuns and towns associated, 4S4 Kaua manuscript, 482 Kin-coefficient, unusual, 178 Kin-sign variant, 168 Kukulcan, in Codex Peresianus, 575 Labna, Palace, 358 Lacandon, a dialect of Cholti in Early Period, 611 Lacandon territory, 542 La Florida, 422 Laguna de Terminos, 609 La Honradez monuments, 391, 433, 438 Stela I, 585 Stela 2, 434 Stela 3, 434 Stela 4, 571, 585 Stela 5, 571, 58s Stela 6, 434, 571, 584 Stela 7, 59, 5S9, 571. 585 Stela 8, 434 Stela 9, 434 INDEX. 637 Lahuntiin, or 3,600 day period, 566 Glyph for, 61, 88, 116, 1S3, 201 Laliuntun ceremoni s, 576 Laliuntun-ending, earliest, 398 Laliuntun glyphs at Copan, list of, 354 At different sites, 566 F'irst identified by Goodman, 566 Possible explanation of form, 567 La Mar, 433, 438,441 Stela I, 572, 585 Stela 2, 572, 5S6 Lamat, normal form, 64 Unusual form of, 150 Lamat, with Venus-sign on Altar K, 1S6 Lamat and Venus-sign, 203, 237, 243, 244, 265 Possible connections, 150 Lamat and Venus year, 182 Lambert, E., drawings, 22 Lances of chay, obsidian, 54; Landa, Diego de. History of Yucatan, 29, 42, 44, 222, 304, 322, 576, 577 Date of death, 491 Hieroglyphic manuscripts and objects burned bv, 44, 46S Trial, 29 Landa "alphabet," 29 Landry, M. D., finds Stela S at Quirigua, 571 Landslide at Hieroglyphic Stairway, 239 Latest dates at Copan, 430 Latest Initial Series known, 411 Latest monument of Middle Period, 437 La Tibuica, Cave of, 548 Lehmann, Walter, correlation of chronology, 528, 529 Quoted, 45 Le5n Pinelo, Relacion on Lacandon territory, 608 Letras heridas, use of in different Mayance languages, 612 Leyden plate, 84, 391, 393, 403, 411, 420, 516, 52S, 567 Description, 411 Provenance of, 422 Linguistic separation, probable course, 615 Lintels, inscribed, 220 Lizana, 42 On settlement of Yucatan, 459 Llano Grande, 14 Local traditions, none at Copan, 542 Loltun, Cave at, 438, 573, 580 (see also Mercer) Hunacab mouth, 572 Long Count, defined, 64 Correlation with Christian chronology, 499 Lopez, Gonzalo, 545 Lords of the Night, 559 Bowditch on, 559 Los Higos, 6, 3S1, 391, 431, 433, 438, 441, 567 Discussion, 384 Stela I, 333, 435, 572, 585 Loss of 205 day positions in Yucatan year, 523 Possible reason therefor, 522 Probably occurs in Xiu manuscript only, reason, 522 Lothrop, S. K., 27 Find of archaic figure, 422 Lunar period in Supplementary Series, 401 Mac, forms tor, 135 Unusual form for (?), 154 Magafia, Lucas, 544 Maize, 3 Maize god, 558 Malaria, influence on development of civilization, 446 Endemic in region, 446 Maler, photographs of Chilan Balani manuscripts, 475 Mani, Lord of, visits Montejo, 480 Mani manuscript, 436, 457, 469, 482 Mani and Tizimin passages, errors in, 506 Difficulties, 533 Manuscripts, hieroglyphic, burned by Landa, 468 Marin Cristobal, 545 Martinez y Hernandez, Juan, 468, 479 Matting, 3 Maudslay, Alfred P., 21, 23, 147, 187 Error, noted, 324, 341, 349 Nomenclature of monuments, 537 Visits to and work at Copan, 21, 24 537 Maya and Julian periods, correspondence of, 48 Maya Calendar, basic unit the day, 48 Maya chronicles, editions, 32 Maya chronology, starting point of, 34 Maya civilization, a native American product, 28 Hypotheses on origin of, 28, 402, et seq. Maya codices, contents of, 43 Maya dates, equivalents in Christian chronology, 49 Method of transcription, 47 Maya inscriptions, history in, 44 Maya, linguistic relation to Cholti, 61 1 To Quiche-Pokom, 611 Maya New Year, 516 On July 16, o. s., 484 Mayapan, Stela 9, 574 Compared with page 11, Codex Peresianiis, 575 Stela; at, mentioned by Landa, 577 Mayance, linguistic stocks, 605 Manuscripts, importance of, 613 Tongues, consonant mutations among, 610 Use of term, 605 Means, Philip, studies on Peruvian chronology, 34 Mecco, El, 520 Melendez, Luis, 545 Membreiio, Alberto, Chorti vocabulary, 607 Menche Tinamit, 23 Mendoza Codex, 36, 38, 41 Mercer, excavation of caves of Yucatan, 438 On occupation of Yucatan, 532 Merida, act of incorporation, 488 Foundation, 488 Mexican objects in other localities, 34 Meye, visit to Copan, 21 Mezquita, Martin de la, 544 Migration of the Maya, eastern, via Bakhalal to Chichen Itza, 458 Probable line of, 404 Trend of early, 414 Migration lines followed by Nahua, 444 Moctezuma U, death of, 37 Mohammedan lunar calendar, parallel with Maya Supple- mentary series, 554 Molina Solis, J. F., 44, 491, 526 Mongolian cultural connection unsupported, 224 Monjarras, Fernando de, 54S Monkey, s Montejo and Lord of Mani, 4S0 Montejo's entries into Yucatan, dates of, 488 Month-coefficients, shift of, 512, 513, 515 Cause, 521 In Books of Chilan Balam, 513 In Dresden, Percsianus, Tro-Cortesianus Codices, 513 In New Empire, 513 In Old Empire, 513 Probable date, 514, 520 Month-sign in terminal date replaced by unusual glyph, 13S 63 8 INDEX. Month-sign indicator at Yaxchilan, Stela i, Stela ii, Lintel 29. 138 Month-signs at Copan, list of all, table, 589, et seq. Monuments, classification of, 50 Majority of, belong to Great Period, 220 Method of designating, 50 Method of treatment, 46, 49 Periodicity of, 565 Quartz inclusions in stone of, i, 2, 563, 564 Recent destruction of, 51 Sudden cessation of, at cultural zenith, 442 Monumental types, chronological sequence of, 394, 395 Moon-glyph, 152 And God D, 153 In Dresden Codex, 554 In Supplementary Series, 554 Mopanes, spoke Itza and not Cholti, 609 Moran, Francisco, enters Lacandon and Choi territory, 60S His manuscript, 608 Manuscript given by Mariano Galvez to Amer. Philos. Soc. in Philadelphia, 609 Motley, R. K., 551, 555 Glyph C, 551, 561 Glyphs D and E, 562 On Supplementary Series, 563 Morris, Earl, 27 Motul Maya Dictionary, manuscript, 16 Multiplication and addition, different positions of numeral modifier in, 312 Municipality of Copan (organized 1893), 86 Muiioz calls attention to Copan, 15 Murga, Alonso de, 545 Nachijn, or Palenque, 615 Nahua influence, 15, 16, 521 Date of first, 502 Lines of migrations followed by, 15, 444 Period at Chichen Itza, 500 Year-bearers, 521 Nahuan customs, effect on calendar, 480 Nakulc Pech, see Pech, Nakuk Nakum, 433, 437, 441, 444, 458, 459 Abandonment of, 457 Sculptured and unsculptured monuments, 439 Nakum Monuments: Stela C, 333, 570, 586 Stela D, 451, 528, 570, 586 Stela U, 570, 585 Nanche, 3 Naranjo, 220, 391, 433, 441, 444, 459. 589 Abandonment of, 457 Important date at, 579 Supplementary Series Glyph E at, 562 Naranjo Monuments: Hieroglyphic Stairway, 281, 389, 554, 570, 583 Stela 6, 5S0 Stela 7, 333, 570, 586 Stela 8, 392, 559, 560, 570, 585 Stela 10, 333, 570, 586 Stela 12, 392, 566, 570, 585 Stela 13, 278, 559, 560, 566, 570, 585 Stela 14, 135, 570, 5S5 Stela 15, 278 Stela 18, 570, 584 Stela 19, 570, 5S5 Stela 21, 570, 584 Stela 22, 566, 584 Stela 23, 151, 584 Stela 24, 399, 400, 559, 560, 561, 566, 584 Stela 28, 392, 585 Naranjo Monuments — Continued. Stela 29, 151, 559, 560, 561 Stela 31, 566, 584 Stela 32, 451, 570, 586 Structure 16, 220 Temple 17, 579 Natural phenomena in Aztec Codices, 40 Nephrite heads, found under Stela 3, 155 New and Old Empire Periods, classification of, 505 New Empire, Colonization period 458 New Empire month coefficients, 513 New Year ceremonies in Yucatan, 222 New Year Days, 304, 306 Less important than last day of year, 397 Perhaps referred to on Altar U, 304 Nieto, Gonzalo, in Yucatan, 490 Nomenclature of monuments, 537 Non-Maya archaeological remains, complete absence of, in Old Empire region, 445 Nonohual, 498, 501, 502 Nueve Cerros, 6og Numbers, and deities, relationships, 558 Numetal ornaments, omission of, 64 Modifiers, different positions for multiplication and addition, 312 Numerals, composite, 136, 166 Examples of, 166-169 Form for 17, 169 Ornamental fillers with, 59, 64 Lack of, a sign of archaism, 59 13 to 19, characteristics of, 172 Nusbaum, J. L., 27 Obsidian lances, 545 Occupation of Old Empire Cities, periods of, 433 Length of, at Copan, 436 Longest at Uaxactun, 436 Ocelot, 5 Ocosingo, 391, 441, 459 An Old Empire capital, 615 Old and New Empire Periods, classification, 505 Old Chichen Itza, description, 123 Lintel discovered by Thompson at, 500 Old Copan, 123 Old Empire area, important sections yet to be explored, 439 Cities, classification of, 440, et seq. Cities, other than Copan, 433 Civilization, migration of, 578 Month coefficients, 513 Region reoccupied about 1447 A. D. after fall of New Empire, 461 Old Maya Empire, divisions of, 46 Fal of, 442 Oldest monument in Maya area, 412 Openings bored in Stela 3 in introducing glyph, 156 Orantes tablet, 281 Ordoriez y Aguiar, Canon, his work on Votanide traditions and early Empire, 613, 614 Ordonez de Villaquiran, Diego, 608, 609 Ordufia, Visitador, 17 Origin of Maya civilization, 402 Probably about 1000 B. c, 411 Probably between the Rio Panuco and the Rio Grijalva, 407 Osuna, Juan Vasquez de, 545 Otzmal, place of Napot Xiu's death, 481 Oxkintok, caves at, 438 Oxkutzcab Chronicle, 470 Page 66 probably copied from hieroglyphic codex, 472 Text, translation, and notes, 507, et seq. (also see Xiu) INDEX. 639 Owens, John G., death and burial, 25 Work at Copan, 24 Pabell5n, El (see El Pabellon) Pacific Guatemalan slope, non-occupanon of by early Maya, 615 Padilla, Mariano, digested copy of Fuentcs y Guzman, photographed by Gates, 543 Palacio, report on Copan, 541 Editions, 15, 541 On local dialects, 605 Quoted, 15 Palenque, 391, 433, 437, 441 Abandonment of 456 Chronological and art criteria apparently contra- dictory at, 437 Initial Series at, 100 Katun-ending records at, 184 Goodman on date of, 530 Spinden on date of, 437 Subject of figures at, 443 Palenque Monuments: House C, 220, 389, 583 House C, Stairway, 150 Palace Group, 99, 515, 570 Slab, 583 Temple of the Cross, 150, 166, 167, 168, 386, 570, 579, 583. Calculation, 98 Error in month-sign on, 332 Stela of, 583 Temple of the Cross, Foliated Cross, and Sun, 2S2 Temple of the Foliated Cross, 167, 570, 579, 583, 589 Temple of Inscriptions, 237, 281, 346, 347, 358, 376, 566, 570, 579, 582, 583, 589 Temple of the Sun, 153, 570, 583 Palm, 3 Palm nuts, 3 Pan, Nahuatl suffix, 15 Papaya, 3 ParaUo, 6, 38 1, 431 Pataxte, 3 Paved plaza floors in vertical section of Acropolis, 211 Payaqui, 605, 607 Peabody Museum of American Archaeology, work of, 24 Nomenclature of monuments, 537 Photographs, 52 Publications, 25 Work at Copan, 537 Peccary, 4 Peccary skull, engraved, 379 Now in Peabody Museum, 379 Peccorini, Atilio, Chorti word-list from Copan, 607 Pech, Ah Naum, chronicle found by Martinez Hernandez, 468. Pech Nakuk, 578 His chronicle, 468 His position, 468 Quoted, 475, 476, 498, 576 Penck on climatic conditions, 450 Peresianus Codex, 578 Compared with Mayapan stela, 575 Probably ceremonial rather than historical, 576 Year-bearers on pages 19 and 20 of, 517 Perez, Juan PJo, on chronology, 525 On katun dates, 495 Period-ending date, at Loltun Cave, 573 On Chichen Itza lintel, 500 Period-ending dating, 466 Superseded Initial Series, 392 Period-endings, predominant in records, 191 Stela; to commemorate, first a Copan practice, 420 Period-glyphs, specialization In, 83 Period-markers, practice of erecting, when introduced, 396 Stelx used as, for 13 centuries, 391 Period of greatest development, 396 Periods, different, limits of, 53, 54 Division in Early, Middle, and Great necessary accord- ing to Morley, 54 Periods of Old and New Empire, classification, 505 Periodicity of monuments, 565 Bowditch on, 565 Seler on, 565 Periodicity of recorded dates in Early Period, 125 Perry, E. W., concession at Copan, 24 Peten region abandoned after close of Old Empire, 444 Entirely deserted for over 800 years, 444 Peten Itza, Itza migrate to, 487 Petrographic description of Copan monuments, 463 Phonology of different Mayance branches, 613 Photographic record of Copan, 25 Piedras Negras, 220, 391, 433, 471 Abandonment of, 457 Important dates at, 579 Latest southwestern date found at, 438 Subject of figures at, 444 Supplementary Series at, 551 Piedras Negras, Monuments: Altar 1', 150, 227, 228, 229, 584 Altar 2', 282, 584 Lintel 2, 45, 166, 444, 565, 583 Stela I, 156, 400, 401, 559, 560, 584, 586, 589 Stela 2, 583 Stela 3, 156, 228, 399, 400, 401, 559, 560, 584 Stela 4, 227, 22S, 566, 584 Stela 5, 400, 584 Stela 6, 565, 583 Stela 7, 227, 228, 584 Stela 8, 583 Stela 9, 565, 584 Stela 10, 152, 579, 584 Stela 1 1, 228, 584 Stela 12, 45, 392, 438, 444, 565, 569, 585 Stela 13, 585 Stela 14, 5S5 Stela 16, 135, 565, 585 Stela 22, 565, 584 Stela 25, III, 398, 399, 565, 567, 569, 583 Stela 29, 436, 582 Stela 31, 583 Stela 35, 583 Stela 36, 227, 565, 583 Stela 37, 583 Piedra Pintada, 13 Pigeons, 5 Pineapple, 3 Pine-tree country. Sapper on early Maya habitat, 405 Planetary motions, 35 Playa Carmen, 520 Polanco, Caspar de, 544 Ponce, quoted, 42 "Pop counted in order," 519 Popenoe, LTseful Plants of Copan, cited, 4 Popol Vuh, 461, 607, 614 Porvenir, 458 Pre-Maya sculptures under Altars X and Y, 208 Probanza de Votan, 613 Profile portrayal of principal figure, at Copan and else- where, 151 "Prophetic" dates, 27, 197, 198, 333, 342 Provenance of Copan monuments, table, 587, et seq. 640 INDEX. Puchutla, 608 Pueblo Bonito, 34 Putnam, Frederic W., work of, 24 Putun, defined, 610 Quarries, 6 Quartz inclusions in monuments, 463 Quirigui, 6, 391,433,437,441 Abandonment of, 457 Colonization of, 428 Earliest date at, 272, 435 Excavation, 27 Hotun-markers of, 214 Important dates at, 329, 344, 579 Katun-ending records at, 1S4 Probable foundation of, 272 Subject of figures at, 443 Supplementary Series at, 551 Tonalamatl, 334 Quirigua, Monuments: Altar L, 334, S7I Altar M, 571, 584 Altar N, 571 Altar Q, 334, 571 Altar R, 334, 571 ^ Hieroglyphic cornice, 387 Stela A, 571, 585 Stela C, 150, 273, 282, 52S, 565, 571, 585, 589 Stela D, 156, 181, 231, 259, 320, 401, 565, 585 Stela E, 156, 181, 227, 258, 273, 287, 301, 386, 397, 399, 400, 401, 559, 579, 584, 585 Stela F, 258, 273, S5S. SS9. 560. 561, 566, 579, 584 Stela H, 194, 56s, 584 Stela I, 312, SS4, 559, 560, 565, 585 Stela J, 258, 273, 559, 560, 579, 584 Stela K, 312, 554, 565, 586 Stela S, 214, 571, 584 Structure i, 214, 220, 259, 276, 312, 389, 392, 554, 571, 579- 586 Zoomorph B, 214, 231, 386, 559, 560, 579, 5S5 Zoomorph G, 214, 227, 228, 258, 273, 320, 565, 585, 586 Zoomorph O, 214, 559, 585 Zoomorph P, 214, 585 Quebrada Seca, 13 Quelen, defined, 610 Quen Santo, stela, same date as Chichen Itza lintel, 459 Stela I, 580, 586 Stela 2, 249, 527, 528, 572, 580, 586 Quetzalcoatl, 502 Quiahuitl, Ninth Lord of the Night, 559 Quinteros, Miguel, 544, 545 Rainfall, its effect on growth and decay of civilization, 448 Rainfall figures, 2 Rainy season, 2 Ramonal, 458 Range of inscriptions at Copan, 335 years, 391 Recent destruction of monuments at Copan, 93 Recovery of fragments, 164 Red paint on monuments, 83, 96, 10;, 132, 141, 346, 376 Religious center of first settlement, 418 Religious ceremonies during Great Period, probable effect of food crisis on, 456 Removals of monuments from other localities, 194 From Group 9 to Great Plaza, 419 In early times, 21 1 Representations, general character of not warlike, 443 Reverse direction of reading glyphs, right to left, 312, 313, 3 22 Rio Amarillo, 381, 431, 441 Altars I and 2, discussion, 382 Rio Beque, 458 Rio Chixoy, 459 Rio Graciosa, referred to by Seler, 528 Rio Grande, 441 Rio Lacantun, 459 Rio Pasion, 438, 459 Rio Sesesmil, 123 River course, 7 River washing, 7 Rocks, I Rock material of Copan monuments described, 463 Rope pattern, 57, 59 Round altar, first found, 396 Rosny, de, katun correlation, 526 Ruano Suarez, Chorti vocabulary, 607 Rubber, 4 Rulers of Copan, their character, 423 Sabaka, caves at, 438 Sacchana, stelae, 459 Sacred fire, renewal of, in Aztec Codices, 41 Salto de Agua Choi word-list, 610 San Diego Museum, 52 San Jose de Motul, 441 San Juan Ermita, founded by refugees from Copan, 603 Santa Cruz Quiche, archaic sculpture at, 209 Santa Rita, 6, 13 Santiago, Jose de, 54S Saville, Marshall H., work at Copan, 24 Sapper, on correlation of chronology, 526 On early habitat of Maya, 405, 406 Scherzer, Carl, 21 Scope of present investigation, 33 Scidptured monuments not always a criterion of length of occupation, 439 Secondary Series, contemporary date on later monuments, 399 Developed for brevity, 392 Numbers counted backward, 162 Order of, 264 Term defined, 46 Secondary usage of monuments, 55, 82, 92, 96, 122, 123, 124, 12;, 341 Examples of, 82 Of Early Period monuments, list of instances, 418 Possible reason therefor, 82 Scibal, 220, 433, 438, 44T, 444, 458, 459 Abandonment of, 457 Seibal Monuments: Hieroglyphic step, 389, 571, 584 Stela I, 458, 528, 571, 586 Stela 5, 220 Stela 6, 220, 571, 585 Stela 7, 220, 566, 571, 585 Stela 8, 528, 571, 586 Stela 9, 528, 571, 586 Stela 10, 528, 571, 586 Stela II, 228, 528, 571, 586 Stela 12, 571, 585 Stete 2, 3, 4, 5, not deciphered, 571 Seler, Eduard, on correlation of chronology, 528 Periodicity of monuments, 565 Quoted, 114, 116, 285, 408 Strong Mexican bias in his Maya studies, 528 Studies on inscriptions, 31 Works, 3 1 Separation of monuments of the three periods, 54 Sequence of archaic stelse, 125 Series of 18 Tuns, on Stela J at Copan, 201 Serpent-head decoration on earliest five altars, 62 INDEX. 641 Sesesmil River, 14 Settlement at Copaii probably not long before 9.0.0.0.0., 416 Site, history of, 14 Of ruins not that of city conquered by Chaves, 18, 23 Skeletal characteristics, 81 Smith, Elliott, quoted, 224 Sotuta, last Cocom capital, 487 Southwestern Maya field, latest date in, 438 Southwestern United States, Mexican objects in, 34 Speech scroll, emblem of, royalty in Aztec codices, 38 Sphere of Copan influence, 6, 381-386, 431 Spinden, H. J., on correlation of chronology, 531, et seq. On date of Palenque, 437 On extravagance in art and cultural decay, 447 On Stela 15, 87 On Stela 18, 97 On Totonacan art, 409 Provisional chronologies, 34 Publications, 26 Quoted, 45, 53, 130, 14'. 208, 226, 229, 326, 358 Reliability of his stylistic criteria, 358 Work on Maya art, 26 Squash, 3 Squier, E. G., on distribution of Chorti language, 607 Stairways, hieroglyphic, at different places, 389 Stelae, as time-markers, 51, 127 Intervals of erection, .5 1 Method of transport, 397 Removal of during occupation, 115 Sequence of archaic, 125 So used for 13 centuries, 591 36 now known at Copan, III Steps, Hieroglyphic, at other localities, 220 Stephens, John L., error in Chorti word list from Zacapa, 605, 606 Mission to Central America, and work';, 20 On location of monuments, 539 On quarries, 6 Quoted, 224, 239, 473 Stiles, Charles Wardell, on malaria, 446 Stoll, Cholti and Chorti vocabularies, 610 Ethnographic, cited, 608 Prints Stephens's erroneous "Chorti" word-list, 607 Stone of monuments, i Character of, 6, 463 Stone carving, development of, 393 Style, archaic, similar to that of Tikal, 54 Style interchange with northern sites (on Stela 23), 151 Stylistic criteria, 26 Stylistic sequence, 53 Apparent conflict with chronological, 354, et seq. Stylistic similarity of the seven g. II .0.0.0 stete, 160 Supplementary Series, 46 Always preceded by an Initial Series, 74, 401 Analyzed, 401 Bowditch on, 551 Description, 551 Detailed analysis of glyphs, 552 Discussion, 551 Eclipse phenomena, 401 Goodman on, 551 Glyph A, 554 Lunar-month sign, 554 Glyph C, description, 555 Probable explanation, 561 Glyph D, description, 561 Relation to glyph E, 562 R. K. Motley on, 562 Supplementary Series — Continued. Glyph E, analysis, 562 At Naranjo, 562 At Yaxchilan, 562 Glyph X, description and analysis, 562, 563 Possibly refer.s to phases of moon, 563 Similarity to God C, 563 Summary of conclusions, 563 Parallels with Mohammedan and Babylonian lunar calendars, 554 Pairs, head elements in Glyph C, 559, 560 Term defined, 46 Swanton, 409 Sweet potato, 3 Symbolism of 20 figures, 335 Symm.etry considerations, 322 Syntactic parallels between Chorti, Maya, and Quiche- Pokom, 612 Tapr, 4 Tassel ornaments, 6g, 70, 83 Tayasal, visited by Fuen^alida, 473 Tecuantepec, conquest of, in Aztec codices, 38 Telleriano-Remensis Codex, 38, 39, 41 Tenochtitlan, fall of, 468, 534 Tepechpan, Mapa de, 36, 41 Tepeyollotli, Eighth Lord of the Night, 559 Terminal date, month-sign suppression in, 157 Ternaux-Compans, quoted, 519 Testera, Padre, entry in Yucatan, 490 Tezulutlan, seat of Old Empire, 614 lezulutleca, possible name for Old Empire language, 614, 615 Thatch palm, 3 Thomas, Cyrus, commentary on Goodman's work, 31 Thomas and Swanton, error in, 60S Thompson, E. H., 500 Tibulca Cave, 548, 600 Tibulon, Cocom driven to, 4S7 Tikal, 220, 391, 433, 437, 441, 444, 458, 459 Abandonment of, 457 And Uaxactun, practically one locality, 414 (see also Uaxactun) Archaic style at, 54 Archaisms at, 76 New Year Day perhaps referred to on lintels, J04 Similar archaisms on Leyden plate, 76 Stone stelae at, long before settlement of Copan, 417 Subject of figures at, 443 17 sculptured stelae at, 413 51 plain stelae at, probably painted, hotun-markers, 568 Tikal Monuments: Altars, 13s Palace of five stories, 5S4 Stela I, 414 Stela 2, 414 Stela 3, 76, 77, 83, 35S, 380, 41 1, 412, 413, 414, 420, 441 Stela 4, 393, 414, 434 Stela 5, 76 Stela 6, 76 Stela 7, 393, 414, 434 Stela 8, 76, 392, 393, 396, 39S, 414, 420, 421, 434, 567, 568, 582 Stela 9, 392, 414, 420, 421, 434, 568, 5S2 Stela 10, 34, 76, 190, 191, 249, 2S1, 282, 283, 288, 346, 347, 349, 351, 356, 376, 4", 4^°, S29 Stela II, 76, 392, 413, 458, 528, 568, 586 Stela 13, 393, 414 Stela 16, 76, 413, 568, 584 ^ \ 642 INDEX. Tikal Monuments — Continued. Stela 17, 156, 401, 411, 412, 413, 420 Structure 10, 566 Temple I, 566, 584 Temple III, 566, 584 51 plain steliE at, 568 Tikoch, new Chel capital, 487 Time concept of Maya, 282 Time-count, Maya method, 4S Accuracy of, 34 Predominance of in inscriptions, 33 Time counts, conflict of, 522 Time element in adjoining cultures, 34 Time periods, higher than cycle, 28 1 Tixhualatun, place of archives, 577 Tizimin manuscript, 436, 457, 470, 519 Tlahtouani, 37 Toad, Shrine of, 378 Tobacco, 3 Tonalamatl, 222 Important instance, 334 Occurrence in inscriptions, 182 On Stela 3, 158 Prominence of on stelie (C, F, H, 4), 351, 364 Tool-work on Stela 3, delicacy of, 159 Tools: metal, basalt, diorite, flint, 5 Total, Bishop, arrival of, 29, 489 Torches, 4 Totonaca, possible Maya afiinities of, 403, et seq. Totonacan art, showing strong strain of Maya feeling, 409 Towns and katuns, association of, 484 Tozzer, A. M., 25 Work at Harvard, 33 Traditions at Copan, 542 Trethevvay, A. J., drawings, 32 Trees, clearance of, 2 Tro-Cortesianus Codex, deer-hunts in, 4 Only Calendar-Round date in, 519 Probably from Tuluum region, 520 Similarity to figures at Tuluum, 519 Year-bearers, 517, 518 Tropical vegetation alone lesponsible for general destruc- tion of monuments, 520 Truncated-cone altar, 377 Tulapan, 498, 501 Tul-ha, or Ocosingo, 615 Tuluum, 391, 567 Figures at, similar to those in Tro-Cortesianus, 519 Stela I, 393, 512, 531, 572, 573, 580, 582, 586 Style of architecture, 519 Wall paintings at. copied by Gann, 519 Tun-sign, 83 Synonymity with Winged Cauac, 153, 380, 573 Unusual form, 153 Used as zero coefficient, 99, 101 Tun coeflncient, unusual, 165 Tun 13, importance of, 358 Tuns, series of 18 on Stela J, 201 Tutul Xiu (see Xiu) Tuxtla Statuette, 77, 209, 391, 393, 412, 416, 420, 500, 501, 515.567 Archaisms on, 76 Provenance and description, 403 Twenty figures, possible symbolism of, 335 Tzalac, San Lucas, del Choi, 608, 609 Thrice established, 609 Tzec, forms for, 75 Tzendales, 433, 437, 441, 459 Temple of the tablet, 572, 583 Tzental-Tzotzil family, Choiti a branch of, 611 Tzuc, defined, 477 Uaxactun, 391, 433, 441, 444, 458, 459 Abandonment of, 457 And Leyden Plate show same archaisms. 76 And Tikal, practically one locality, 414 Early monuments at, 412 Occupied for 541 years, 436 Oldest city known 436 Reason for name, 412 Stone stehe at, long before settlement of Copan, 417 Supplementary Series at, 551 Uaxactun Monuments: Stela I, 571, 574 Stela 2, 571, 584 Stela 3, 76, 358, 411, 412, 420, 571 Stela 4, 571 Stela 5, 76, 412, 413, 414, 420, 567, 571 Stela 6, 312, 412, 571 Stela 7, 571, 586 Stela 8, 412, 571 Stela 9, 76, 77, 393, 411, 412, 416, 420, 441, 500, 501^ 568,571 Stela 10, 571 Stela II, 571 Uayeb-sign not found at Copan, only three tnnes else- where, 589 Landa on Uayeb days, 590 Uayeb and God N, 558 Ucanal, 433, 438, 441, 444, 458 Abandonment of, 457 Stela 3, 435, 52S, 572, 586 Uinal forms, 232 Uinal-sign, 83 U kahlay katunob, or record of the katuns, 43, et seq. Contradictions in probable explanation of, 512 Similarity to Aztec annals, 44 Ulloa, Andres de, 544 Underhill, G., 27 Unexplored sections of Old Empire area, 439 Uo, forms for, 66 Relation to uinal, 232 UuD katun, or katun circuit, 483 Uxmal, Ball Court ring, 511, 512 Description, 515 Monjas Quadrangle, 466, 511, 512, 514, 519, 522 Uyajal, 542 Valdivia, 476 Valentin!, on katuns, 526 Closeness to Motley's results, 526 Vandalism, 93, 292 Vanilla, 3 Vaticanus Codex, 37, 38, 39, .:l Vegetation, i Venus-sign, 150 For Lamat, 203, 237, 243 For Lamat, possible connections, 150, 265 In introducing glyph, 185 On Temple 2ta, 319 Venus years, 182 And Lamat, 182 Vera Paz, 542 Vierra, Carlos, paintmg by, 52 Villaguiterre Sotomayor, Conquest of Itzas, 43, 472, 473, 492 Volcanic activity, I Eruptions shown in Aztec Codices, 39 Rocks in earth below Stela 7, 105 Votanide traditions, 613 "Votanide" Empire, four divisions of, 615 INDEX. 643 Warlike subjects, entirely absent on stelae and altars at Copan, 219 Rare occurrence of, elsewhere in Old Empire, 443, 445 Weight of stelae, 397 Willoughby, C. C, 25 Willson, R. W., 465, 551, 554, 558 Winged-Cauac tun sign, 573 (see also Tun) Writing, decipherment of, 28 Wright, Fred. E.,on material of Copan monuments, 463 Xcanchakan, Mayapan stela at, 574 Xihuitlmolpia, or 52 year cycle among Aztec, 41. Xiu Chronicle (see Oxkutzcab.) Xiu coimt, different from Itza, 512 Changed after fall of Mayapan, 523 In western Yucatan lost Old Empire count under Nahua influence, 523 Xiu genealogical tree in Oxkutzcab Chronicle, 485 Genealogy, 22 generations from 1397 to 1914 ,a. d., 470 Xiu, Juan, quoted, 471 Errors by, 522 Probable origin of, 522 Xiu, Napot, Bowditch on, 529 Classification of data on, 493, 494 Death of, 478 Resume, 487 Summary of evidence, 485 Xiu, Tutul, 445 At Uxmal, 487 Xiuhtecutli, First Lord of the Night, 559 Xiuhtzontli, turquoise-studded Aztec crown, 37 Xupa, 441 Yax, forms for, 66 Yaxchilan, 23, 220, 433, 437, 447 Abandonment of, 457 Important date at, 579 Subject of figures at, 443 Supplementary Series at, 551 Glyph E in, 562 Yaxchilan Monuments: Altar (near Stela i), 392, 399 Altar 9, 565 Altar (near Structure 39), 584 Lintel in Berlin Museum, 168 Lintel 3, 565, 570, 584 Lintel 9, 100, 516 Lintel 21, 168 Lintel 29, 168 Lintel 31, 570, 585 Lintel 32, 151 Lintel 33, 135 Yaxchilan Monuments — Continued. Lintel 42, 151 Lintel 43, 135 Lintel 46, 151 Stairway (on Structure 6), 389 Stela I, 559, 560, 561, 570, 585 Stela 6, 166 Stela II, 151, 156, 242, 401, 559, 560, 580 Stela 12, 203 Stela 19, 151 Stela 20, 5, 151 Structure 5, 220 Yaxha, 391,433, 437,441 Only one stela deciphered, 571 Stela 6, 435, 571, 583 Yaxkin, first day of, as New Year's Day, loi Year, beginning of, 134 Year-bearers, 478, 515, et seq. In Codices Dresdensis and Peresianus, 516, 518 In Codex Tro-Cortesianus, 517, 518 In New Empire, 516 In Old Empire, 516 Nahua, 521 Yucatan, colonization of, 459 Consequences of civil war in, 445 Discovery of, 457 Early entries of fathers, 490 East coast ruins, type of, 520 Immigration from Tabasco in loth century, according to Landa, 459 Zac, forms for, 66 Zapata, Munoz, of Oxkutzcab, 48S Zapote, 3 Zaragoza, Springs of, 546 Zero element, coefficient signs, loi Fleshless lower jaw, sign for, 159 Month coefficient, an unusual form for, 99, 100 Position in Maya months, 64 Unusual form, 137 Zilan, stelae at, mentioned by Landa, 577 Stelae found by Motley, 577 Zip, forms for, 66, 67 Special form at Copan in decade following 9. 16. 10. 0.0, 67 Unusual variant, 290 Variant for, 2S0 Ziyancaan, 457 Location of, 458 Zuiva, 501, 562 M ' 1«^ ^s - ■O^ <^V ^ d\\ »K //>! o 'J^, /^^ i.- ^/% l^^^^P;^ ^'%^V ^ » J- .- ■■:,; . -^ > = , 00 h^ ■:*-, ■\ «^*' % LIBRARY OF CONGRESS 015 838 243 A =f*