PEABODY MUSEUM OF AMERICAN ARCHAEOLOGY AND ETHNOLOGY HARVARD UNIVERSITY MEMOIRS VOLUME VI CAMBRIDGE, MASS. PUBLISHED BY THE MUSEUM e jO V, ip Copyright, 1913, By the Peabody Museum of American Archaeology and Ethnology, Harvard University University Press: John Wilson and Son, Cambridge, U.S.A. V IHE GJIY CENTER UBRARY MEMOIRS OF THE PEABODY MUSEUM OP AMERICAN ARCHAEOLOGY AND ETHNOLOGY, HARVARD UNIVERSITY Vol. VI A STUDY OF MAYA ART ITS SUBJECT MATTER AND HISTORICAL DEVELOPMENT BY HERBERT J. SPINDEN WITH TWO HUNDRED AND EIGHTY-SIX ILLUSTRATIONS IN THE TEXT TWENTY-NINE PLATES AND MAP CAMBRIDGE PUBLISHED BY THE MUSEUM 1913 AUTHOR’S PREFACE The study of Maya art, here presented, is based upon a thesis for the degree of Doctor of Philosophy submitted May 1, 1909, in Harvard University. While the matter has expanded greatly under further study, still the thesis presented contained an exposition of the chronological sequence of the monu¬ ments, which the writer considers the most noteworthy contribution, as well as chapters on the analysis of the designs and the principles of the architecture. It was thought wise to present the portion relating to the historical develop¬ ment of art at Copan before the Congress of Americanists at Mexico City in September, 1910, otherwise the subject matter has not been given to the public. The attempt has been made to be precise and exoteric in the discussion of this most involved subject. It is with gratitude that the writer acknowledges his indebtedness to the many persons who have aided and encouraged him in this work. The inception of this research took place in Anthropology 9, a course on Mexican and Central American archaeology offered in Harvard University by Dr. A. M. Tozzer. Its continuance has been largely due to the support and cooperation of the small band of students of Maya culture headed by Mr. C. P, Bowditch. Thanks are also due to Mrs. Zelia Nuttall of Mexico City, to Mr. E. H. Thompson of Chichen Itza, Yucatan, and to Mrs. W. M. James of Merida, Yucatan, as well as to many other persons in Mexico, Guatemala and Honduras. In the revision of the manuscript the writer received valued assistance from Dr. Tozzer. For the revision of the proof he is indebted to Professor Putnam and to Miss Mead and to the latter also for the preparation of the index. Mr. C. C. Willoughby has given his kind attention to the preparation of the plates and to the making of the blocks for the illustrations in the text. The map of the region covered by the Maya civilization was drawn by Mr. L. M. Hen¬ drick, Jr., according to data compiled from several sources. Since his connection with the American Museum of Natural History the writer has been greatly indebted to his superiors in the Museum, who have done everything in their power to further his labors in this field. H. J. S. American Museum of Natural History, New York, November 28, 1912. CONTENTS Page Editorial Note . v Preface . vii List of Plates . xiii List of Figures . xv Introduction. 1 Area. 1 Relation to Surrounding Cultures. 1 Natural Divisions . 2 Early Notices. 3 Native Accounts. 10 Political and Religious Ideas. 11 Present Population. 12 Materials Available for Study. 12 Previous Studies. 13 I. General Consideration of Maya Art. 15 Homogeneity of Maya Art. 16 The Human Form. 21 Subjects Represented. 21 Poses and Groupings. 24 Foreshortening and Perspective . 27 Expression. 31 Composition. 31 The Serpent. 32 The Origin of the Serpent in Art. 32 Zoological Observations. 33 Idealization. 34 The Serpent in Design. 36 Simplification. 39 Elaboration. 41 Elimination. 43 Substitution. 46 The Serpent and Geometric Art. 46 The Serpent in some of its Religious Aspects. 49 The Ceremonial Bar. 49 The Manikin Scepter. 50 The Two-headed Dragon. 53 Composite Types and Miscellaneous Variations. 56 The Serpent Bird. 60 The Long-nosed God. 61 The Roman-nosed God. 69 Other Subjects. 76 The Jaguar. 76 Birds and Feathers. 77 Miscellaneous Animals. 82 Bones and Death. 85 Grotesque Figures. 87 The Maize God . 88 Other Divinities. 90 Astronomical Signs. 91 CONTENTS. Page II. Consideration of the Material Arts. 96 Architecture. 96 Assemblage. 96 Function of Buildings. 98 Ground-plans. 99 Elevation Plans.102 Correlation of Buildings.103 Substructures.105 Walls.107 Vaults .108 Roof Structures.110 Columns .113 Cornices.114 Doorways and Wall Openings.115 Application of Decoration.115 Realistic Decoration.116 Mask Panel.118 Profile Mask Panel.124 Geometric Decoration.127 Stelae.129 Altars.130 Color.131 Prototype of the Maya Temple.132 Minor Arts.133 Ceramics.133 Precious Stones .142 Metal Working.145 Basketry.146 Textiles.147 Tattooing.150 Minor Carvings.150 Illuminated Manuscripts .152 III. Chronological Sequence.155 First Epoch.155 Statement of the Problem.155 Copan .155 Table I. Chronological Sequence of Copan Monuments, facing .164 Tikal. 165 Earliest dated Objects.170 Quirigua.173 Naranjo.177 Seibal.183 Yaxchilan.184 Piedras Negras.189 Palenque.192 Other Sites .195 Summary of the First Epoch.198 Second Epoch.198 The Period of the Transition. 199 The Period of the League of Mayapan.202 The Period of Influence from the Valley of Mexico.205 Correlation of Maya and Christian Chronology.215 Table II. Historical Correlation of Inscriptions and Chronicles, facing .... 216 Connection with other Cultures.219 Maya Influence in Nahua and Zapotecan Art.219 Chronological Sequence of Art in Mexico.225 Problems of Cultural Connection outside of Mexico.231 Pyramids.232 Religious Ideas Connected with the Serpent.236 Symbolism and Art.237 Conclusion.247 CONTENTS. xi Page Table of Nomenclature.249 Ruins and Monuments.249 Codices.260 Bibliography.263 Index.277 Plates .l - 29 LIST OF PLATES Plate 1. Quirigua sculptures: Fig. 1, Altar G, riewed from the northwest; fig. 2, Altar G, west side; fig. 3, Altar B, south side; fig. 4, Altar B, east side. Photographs by Peabody Museum Expeditions (Gordon). Plate 2. Quirigua sculptures: Fig. 1, Altar P, west side; fig. 2, Altar P, detail of east side. Photographs by Peabody Museum Expeditions (Gordon). Plate 3. Copan structures: Fig. 1, Mound 21, view looking north; fig. 2, portions of Structures 21a and 22, view looking west. Photographs by Peabody Museum Expeditions (Saville). Plate 4. Vault construction: Fig. 1, Copan, small vaulted chamber; fig. 2, Labna, view of a portion of the Temple. Photographs by Peabody Museum Expeditions (Fig. 1, Saville, fig. 2, Sweet). Plate 5. Maya construction: Fig. 1, Chichen Itza, partial view of a structure belonging to the Group of the Columns; fig. 2, Uxmal sealed portal vault of the House of the Governor. Photographs by Peabody Museum Expeditions (Fig. 1, Thompson; fig. 2, Sweet). Plate 6. Roof structures: Fig. 1, Uxmal, general view of the roof comb, House of the Doves; fig. 2, Sabacche, small temple with flying fagade. Photographs by Peabody Museum Expeditions (Sweet). Plate 7. Columns: Fig. 1, Chacmultun, portion of Edifice 1; fig. 2, Chichen Itza, structure belonging to the Group of the Columns. Photographs by Peabody Museum Expeditions (Thompson). Plate 8. Mask panels: Fig. 1, Kabah, portion of the fagade of Structure 1; fig. 2, Uxmal, portion of the fagade of the House of the Governor. Photographs by Peabody Museum Expeditions (Sweet). Plate 9. Mask panels: Fig. 1, Uxmal, portion of the fagade of the Western Range of the Nunnery Quad¬ rangle; fig. 2, Labna, East Wing of Upper Range of the Palace. Photographs by Peabody Museum Expeditions (Fig. 1, Sweet, fig. 2, Thompson). Plate 10. Mask panels: Fig. 1, Labna, West Wing of Lower Range of the Palace; fig. 2, Labna, building north of the Portal. Photographs by Peabody Museum Expeditions (Sweet). Plate 11. Mask panels and profile panels: Fig. 1, Hochob, Left Wing of Principal Structure; fig. 2, Hochob, Middle portion of Principal Structure. Photographs by T. Maler. Plate 12. Mask panels and profile panels: Fig. 1, Tabasqueno, North Fagade of Temple; fig. 2, Dsib- ilnocac (Iturbide) Temple with Sealed Doorway. Photographs by T. Maler. Plate 13. Mask panels and profile panels: Fig. 1, Uxmal, House of the Magician; fig. 2, Chichen Itza, Eastern fagade of the East Wing of the Monjas. Photographs by Peabody Museum Expeditions (Fig. 1, Sweet, fig. 2, Thompson). Plate 14. Geometric decoration and mask panels: Fig. 1, Uxmal, House of the Governor; fig. 2, Uxmal, Eastern Range of Nunnery Quadrangle. Photographs by Peabody Museum Expeditions (Sweet). Plate 15. Geometric decoration and mask panels: Fig. 1, Chichen Itza, Iglesia; fig. 2, Labna Portal arch from the southeast; fig. 3, Sabacche, Principal Structure, East Fagade. Photographs by Pea¬ body Museum Expeditions (Sweet). Plate 16. Contrasted styles of decoration: Fig. 1, Kabah, Structure 3; fig. 2, Xlabpak of Maler, Fagade of Principal Temple; fig. 3, Uxmal, North Range of the Nunnery Quadrangle. Photographs by Peabody Museum Expeditions (Sweet). Plate 17. Terra cotta figurines: Figs. 1-3, whistles from Uloa Valley, Honduras; figs. 4 and 7, whistle and figurine from Jonuta, Tabasco, Mexico; figs. 5 and 6, figurines from mounds at Ivamela, Rio Salinas (Chixoy), Guatemala; fig. 8, modern cast from terra cotta mould, Rio Salinas, Guatemala; All the above figurines are in the collections of the Peabody Museum. Photographs by the Mu¬ seum. Figs. 9-12, figurines from the Island of Jaina, Campeche, Mexico, in the collection of Mrs. W. M. James, Merida, Yucatan. Photographs by T. Maler. Plate 18. Copan: Series of archaic stelae: Fig. 1, Stela 7; fig. 2, Stela E; fig. 3, Stela P; fig. 4, Stela 6. Photographs by Peabody Museum Expeditions (Saville and Gordon). Plate 19. Copan: Series showing later development of sculpture: Figs. 1-2, fragments of Stela 5 show¬ ing sculpture on opposite sides; fig. 3, Stela 3; fig. 4, Stela N; fig. 5, Stela H. Photographs by Peabody Museum Expeditions (Saville and Gordon). XIV LIST OF PLATES. Plate 20. Series showing development of hieroglyphs. Copan: Fig. 1, fragment of an old altar found in the Hieroglyphic Stairway; fig. 2, one side of Altar K; figs. 3-4, blocks from the Hieroglyphic Stairway; fig. 5, part of inscription on back of Stela 9; fig. 6, inscription on back of Stela 6; fig. 7, hieroglyphs on back of Stela A; fig. 8, hieroglyphs on back of Stela D; Quirigua: fig. 9, hieroglyphs on east side of Stela D; fig. 10, hieroglyphs on the east side of Stela F. Photographs by Peabody Museum Expeditions (Gordon). Plate 21. Tikal: Series showing development of sculpture: Fig. 1, Stela 7; fig. 2, Stela 9; fig. 3, Stela 1; fig. 4, Stela 16. Photographs by Peabody Museum Expeditions (Maler). Plate 22. Tikal sculptures: Fig. 1, Stela 5; fig. 2, three hieroglyphs on west side of Stela 13; fig. 3, three hieroglyphs on east side of Stela 9; fig. 4, eight hieroglyphs, some incomplete on north side of Stela 1; fig. 5, part of hieroglyphic inscription of west side of Stela 5. Photographs by Peabody Museum Expeditions (Maler). Plate 23. Early and late stelae. Quirigua: Figs. 1 and 3, front and back of Stela I; Copan: fig. 2, one side of Stela 15 bearing initial series inscription. Photographs by Peabody Museum Expedi¬ tions (Gordon). Plate 24. Naranjo: Series showing development of sculpture: Fig. 1, Stela 25; fig. 2, Stela 30; fig. 3, Stela 12. Photographs by Peabody Museum Expeditions (Maler). Plate 25. Miscellaneous sculptures. Seibal: Fig. 1, Stela 7; fig. 2, Stela 10; Piedras Negras: fig. 3, Stela 13; Ocosingo: fig. 4, lower portion of Stela 1; fig. 5, portion of Stela 2; Cankuen: fig. 6, lower portion of Stela 1. Photographs by Peabody Museum Expeditions (Figs. 1, 2, 3 and 6, Maler, figs. 4 and 5, Tozzer). Plate 26. Sculptured heads. Copan: Fig. 1, stone head from Mound 32; fig. 2, stone head from Mound 41; fig. 3, stone head from d6bris of Temple 22. Photographs by Peabody Museum. Palenque: Figs. 4 to 6, three views of a stone head, after Maudslay. Plate 27. Chichen Itza: Fig. 1, Casa de Monjas, Second or Main Range of rooms; fig. 2, Casa de Monjas, East Wing, from the north. Photographs by Peabody Museum Expeditions (Sweet). Plate 28. Chichen Itza: Details of Casa de Monjas: Fig. 1, reused stones in wall of Upper Chamber; fig. 2, geometric panel, Second Range; figs. 3 and 4, mask panel in frieze of the Foundation; fig. 5, mask panel, East Fagade of East Wing; fig. 6, corner masks on northeast corner of East Fagade of East Wing. Photographs by H. J. Spinden. Plate 29. Fig. 1, Sculptured Column, North Temple of Ball Court; fig. 2, Fresco Painting, Temple of the Jaguars; fig. 3, Pilaster, Temple of the Tables; fig. 4, Atlantean Figure; fig. 5, Chacmool Figure at San Salvador; fig. 6, Sculptured Columns, Temple of the Tables; fig. 7, Jaguar Relief, Mau¬ soleum, Mound 13; all except fig. 5 are Chichen Itza. Photographs by F. M. Chapman (fig. 2); E. Mosonyi (fig. 5); T. Maler (fig. 7); H. J. Spinden (figs. 1, 3, 4 and 6). Plate 30. Map of the Principal Archaeological Sites of the Maya Area. Compiled from various sources by H. J. Spinden. Drawn by L. M. Hendrick. LIST OF FIGURES IN THE TEXT Page Figure 1. Heads with curled nose ornament: a, Sun shield on the tablet of the Temple of the Sun, Palenque; b, Stone head with a tenion at back, from Copan, original in the Peabody Museum; c, Sculptured stone block from Pyramid 37, Quen Santo, Seler, 1 1901, c, p. 112; d, Sculptured stone head from Labna, in the Peabody Museum; e, Pottery vessel from near Coban, in the collection of Mr. E. P. Dieseldorff, Seler, 1901, c, p. 178; /, Pottery mask from Panzamald, Seler, 1901, c, p. 179. 17 Figure 2. Plant motive: a, Chichen Itza, Lower Chamber, Temple of the Jaguars, Maudslay, III, pi. 47; b, Palenque, Palace, House D, Maudslay, IV, pi. 37; c, Madrid Stela, L6on de Rosny, 1882, pi. 2 . 18 Figure 3. Plant and fish motive: a, Copan, Altar T, Maudslay, I, pi. 96; b, Palenque, Palace, House A, Maudslay, IV, pi. 10; c, Chajcar, Maudslay, IV, pi. 93, 1; d, Chichen Itza, Lower Chamber, Temple of the Jaguars; e and/, Palenque, Temple of the Cross, Maudslay, IV, pi. 71; g, Ixkun, Stela 1, Maudslay, II, pi. 69; h, Palenque, Frieze of Temple of the Cross, Maudslay, IV, pi. 68, a; t, Copan, Stela N, Maudslay, I, pi. 82; j, Chichen Itza, Lower Chamber, Temple of the Jaguars, Maudslay, III, pi. 51; k, Nebaj, on a pottery vessel reproduced by Seler, 1902- 1908, III, p. 718. 19 Figure 4. God B wading in water, Dresden Codex, p. 67. 19 Figure 5. Astronomical signs in bands: a, Uxmal, Fagade of the Annex to the House of the Magi¬ cian; b and c, Pottery fragments from Alta Vera Paz, Seler, 1902-1908, III, p. 615; d, Chichen Itza, East fagade of East Wing, Monjas, Maudslay, III, p. 13, b. 19 Figure 6. Astronomical signs combined with bird and animal heads: a, Quirigua, back of Stela I, Plate 23, fig. 3; b, Copan, Stela H, Maudslay, I, pi. 61; c, d and /, Peresianus Codex, L 6 on de Rosny, 1888, pis. 3, 5, and 22; e, Dresden Codex, p. 46; /, Naranjo, Stela 32. 20 Figure 7. Detail showing a kneeling worshiper, Yaxchilan, Stela 7, after a photograph by Maler. 21 Figure 8 . Presiding priest, Palenque, Tablet of the Temple of the Sun, Maudslay, IV, pi. 88 . . 22 Figure 9. Man wearing mask, Yaxchilan, Stela 11. 22 Figure 10. Memorial of conquest, Yaxchilan, Lintel 12, drawn from photograph by Maler and a cast in the Peabody Museum. 23 Figure 11. Atlantean figures on pilasters, Temple of the Tables, Chichen Itza. 26 Figure 12. Warrior from the frescos of the Inner Chamber of the Temple of the Jaguars, Chichen Itza. 26 Figure 13. Group of figures, Dresden Codex, p. 60. 27 Figure 14. Seated worshiper in profile, Palenque, Palace, House A, see Maudslay, IV, pis. 8 and 11. 27 Figure 15. Elaborated aprons: a , Copan, Stela N, Maudslay, I, pi. 82; b , Quirigua, Stela F, Maud¬ slay, II, pi. 36, a; c, Ixkun, Stela 1, Maudslay, II, pi. 69; d, Tikal, Stela 11; e, Palenque, Tem¬ ple of the Inscriptions, Maudslay, IV, pi. 55. 28 Figure 16. Ornament suspended from headdress, Tikal, Stela 5, Plate 22, fig. 1. 28 Figure 17. Drawing of sculpture, Yaxchilan, Lintel 8 . 29 Figure 18. Kneeling figures, La Mar, Stela 1. 30 Figure 19. Representation of drapery, Palenque, Temple of the Cross, Maudslay, IV, pi. 76 . 30 Figure 20. Realistic serpents in Dresden Codex: o, p. 23; 6 , p. 36. 34 Figure 21. Idealized serpent, Chichen Itza, frescos of the Inner Chamber of the Temple of the Jag¬ uars, after Miss A. Breton, 1906, pi. 8 , fig. 1. 34 1 References to author and year are according to the Bibliography given at the end of the volume. All references to Maudslay are to the archeological portion of the Biologia Centrali-Americana, 1889-1902. Where the name of the Bite or monument is stated but no reference to a publication is given the reader may ordinarily gain additional in¬ formation by consulting the Table of Nomenclature beginning on page 249. Here the sites are arranged alphabeti¬ cally and the principal contributions to our knowledge of each site is noted and correlated. xvi LIST OF FIGURES IN THE TEXT. Page Figure 22. Feathered serpent, Chichen Itza, Lower Temple of the Jaguars, Maudslay, III, pi. 49. 34 Figure 23. Serpents with grotesque heads in their mouths, Copan, Stela D, Maudslay, I, pi. 45 . 35 Figure 24. Divine serpent with human head in the mouth, Yaxchilan, Lintel 15, Maudslay, II, pi. 83. 35 Figure 25. Details showing influence of serpentine forms: a-d, Tro-Cortesianus Codex, pp. 14, 30 and 35; e, Palenque, Palace, House A, Maudslay, IV, pi. 11; /, Palenque, Temple of the Foli¬ ated Cross, Maudslay, IV, pi. 81; g, Chichen Itza, Lower Chamber, Temple of the Jaguars, Maudslay, III, pi. 46; h, Palenque, Temple of the Cross, Maudslay, IV, pi. 72; i, Quirigua, Altar P, Maudslay, II, pi. 60, b; j, Detail of a glyph, Copan; k, Quirigua, Altar P, Maudslay II, pi. 61; l, Tikal, Lintel of Temple IV, Maudslay, III, pi. 7S; in, Quirigua, Part of initial glyph Stela E, Maudslay, II, pi. 32; n, Quirigua, Part of initial glyph Stela J, Maudslay, II, pi. 46; o, Tikal, Lintel of Temple II, Maudslay, III, pi. 73; p, Labna, Ear plug of mask panel ... 37 Figure 26. Plumed serpent with elaborated head, Altar 0, Copan. 38 Figure 27. Reptilian radicles, H. Allen, 1881, p. 315. 39 Figure 28. Drawing of rattlesnake’s head. 39 Figure 29. Serpent heads with tenons, Mound 14, Chichen Itza. 39 Figure 30. Typical conventionalized serpent head. 40 Figure 31. Part of underlying design, Altar P, Quirigua. 41 Figure 32. Part of overlying design, Altar P, Quirigua. 42 Figure 33. Elaborated serpent head, Altar P, Quirigua. 43 Figure 34. Modified serpent heads: Copan, a and d, Stela H, Maudslay, I, pis. 61 and 59; Palenque: b, Temple of the Cross, Maudslay, IV, pi. 68; c, Palace, House E, Maudslay, IV, pi. 43 . . . 44 Figure 35. Serpent heads at sides of aprons: a, Ixkun, Stela 1, Maudslay, II, pi. 69; b, Quirigua, Stela F, Maudslay, II, pi. 36; c, Copan, Stela B, Maudslay, I, pi. 34; d, Copan, Stela N, Maud¬ slay, I, pi. 82; e, Naranjo, Stela 6; /, Seibal, Stela 10; g, Copan, Stela A, Maudslay, I, pi. 26 44 Figure 36. Serpent heads conventionalized in the flamboyant manner: Chichen Itza, Temple of the Jaguars, Maudslay, III, pi. 35. 45 Figure 37. Nose plug representing a conventionalized serpent head, Piedras Negras, Stela 14 . . 45 Figure 38. Plant form modified by serpent features, Palenque, Palace, House D, Maudslay, IV, pi. 35. 45 Figure 39. Plant form modified by serpent features, Chichen Itza, Lower Chamber, Temple of the Jaguars, Maudslay, III, pi. 45. 45 Figure 40. Plant form with stem modified into a serpent’s head, Quirigua, Altar P, Maudslay, II, pi. 62. 45 Figure 41. Detail of an ear plug with attached serpent heads, Copan, Stela 2. 46 Figure 42. Serpent heads showing multiplication of nose plugs, Chichen Itza, Lower Chamber, Tem¬ ple of the Jaguars, Maudslay, III, pis. 49, 51 and 47. 46 Figure 43. Breast ornaments, Copan: a, b, d, e and g, from interior step of Temple 11, Maudslay, I, pi. 8; c and /, Altar Q, Maudslay, I, pi. 92. 47 Figure 44. Serpent head modified by a fret turning down, Copan, Southern stairway of Temple 11. 47 Figure 45. The Ceremonial Bar, Leiden Plate. 50 Figure 46. The Ceremonial Bar, Copan: a, Stela P, Maudslay, I, pi. 87; b, Stela N, Maudslay, I, pi. 82. 50 Figure 47. The Manikin Scepter: a, Yaxchilan, Lintel 1; b, Quirigua, Altar P, Maudslay, II, pi. 62 51 Figure 48. The Manikin Scepter on potsherd from Santa Cruz Quich6, Guatemala. Original in Peabody Museum. 51 Figure 49. The Manikin Scepter as a new-born child (?), Palenque: a, Temple of the Inscriptions, Maudslay, IV, pi. 55; b, Temple of the Foliated Cross, Maudslay, IV, pi. 81. 52 Figure 50. The Manikin Scepter with body reduced to a staff, Ruins of Tzendales. After a field sketch by Dr. Tozzer. 52 Figure 51. The Ma niki n head on a staff, Palenque, Palace, House A, Maudslay, IV, pi. 10 . . . 52 Figure 52. The Two-headed Dragon, simplest form on an oblong block of stone, Copan, Maudslay, I, pi. 114. 53 Figure 53. View of central portion of Altar M, Copan. 53 Figure 54. Trefoil scroll on joints of Two-headed Dragon. 53 LIST OF FIGURES IN THE TEXT. xvii Page Figure 55. Rear head of Altar M, Copan. 54 Figure 56. Rear head of Altar D, Copan. 54 Figure 57. The Two-headed Dragon with elongated body: a to c, Palenque, Palace, House E, Maudslay, IV, pi. 43; d, Piedras Negras, Stela 25. 55 Figure 58. Manikin Scepter combined with Ceremonial Bar, Tikal, Stela 1. 56 Figure 59. Rear end of modified Ceremonial Bar, Seibal, Stela 10. 56 Figure 60. Fragment of carved bone from the river front, Copan. Original in the Peabody Museum 56 Figure 61. Ceremonial Bar held in tilted position, Yaxchilan, Stela 1. 57 Figure 62. Man holding two-headed serpent of flexible type, Yaxchilan, Lintel 39. 57 Figure 63. Ceremonial Bar on pottery box from Coban, Dieseldorff Collection. 58 Figure 64. Degenerate form of Ceremonial Bar, Copan, Stela 11, Maudslay, I, pi. 112. 58 Figure 65. Substitute for Ceremonial Bar, Tikal, Stela 16. 58 Figure 66 . Substitute for Ceremonial Bar, serpent heads attached to chain-like objects: a, Quirigua, Stela F, Maudslay, II, pi. 36, a; b, Palenque, Temple of the Cross, Maudslay, IV, pi. 67 . . 59 Figure 67. Two-headed Dragon, degenerate form, Yaxchilan, Lintel 25, Maudslay, II, pis. 87 and 88 . 60 Figure 68 . Serpent Bird, Palenque, Temple of the Foliated Cross, Maudslay, IV, pi. 81. 60 Figure 69. Staff representing a bird on a tree, Yaxchilan, Lintel 2. 60 Figure 70. Serpent Bird, Copan, Stela H, Maudslay, I, pi. 61. 61 Figure 71. Small clay head with serpent wing panel attached to ear plug. Dieseldorff Collection . 61 Figure 72. The Wing Panel: a, Copan, Stela H; b, Yaxchilan, Lintel 1; c, Quirigua, Stela Q; d, Quirigua, Stela F. 62 Figure 73. God B and the serpent: a, God B issues from serpent, Dresden Codex, p. 33; 6 , God B with serpent body, Dresden Codex, p. 35. 62 Figure 74. God B holding serpent in his hand, Dresden Codex, p. 40. 63 Figure 75. Head of Long-nosed God as secondary ornamentation, Quirigua, Altar P, Maudslay, II, pb 64. 63 Figure 76. Head with elaborated nose, Quirigua, Stela F, Maudslay, II, pi. 36. 64 Figure 77. God K in his relation to the serpent and to God B: a, God B attached to serpent with the head of God K, Tro-Cortesianus Codex, p. 31; b, God K, Dresden Codex, p. 12; c, God B wearing head of God K, Dresden Codex, p. 34. 64 Figure 78. God B, the Long-nosed God, in relation to rain and corn growing: a, scattering seeds in the rain (?), Tro-Cortesianus Codex, p. 10; b, holding sign for corn in hand, Tro-Cortesianus Codex, p. 8 ; c, falling with sprouting corn in hand and leaves attached to body, Dresden Codex, P-15. 65 Figure 79. The Long-nosed God as incised decoration on a bowl. Original in American Museum of Natural History. 65 Figure 80. The Long-nosed God with the sun sign on forehead, Copan, Stela D, Maudslay, I, pi. 46 . 66 Figure 81. Long-nosed death heads attached to serpent tails: a, Copan, Stela D, Maudslay, I, pi. 46; b, Yaxchilan, Lintel 14 (design reversed), Maudslay, I, pi. 23. 66 Figure 82. Hieroglyphs of the rear head of the Two-headed Dragon, Palenque, Temple of the In¬ scriptions, Maudslay, IV, pi. 61. 66 Figure 83. The rear head of the Two-headed Dragon attached to band of astronomical symbols, Palenque, Temple of the Cross, Maudslay, IV, pi. 76. 67 Figure 84. Objects that may represent falling water: a, Quirigua, Stela H; b to c, Palenque, Temple of the Cross, Maudslay, IV, pis. 75 and 71. 67 Figure 85. Object falling from hands, Yaxchilan, Stela 7. 67 Figure 86 . The Long-nosed Death God as a period glyph, Copan, Stela D, Maudslay, I, pi. 46 68 Figure 87. The inverted head of Long-nosed Death God, Copan, Stela B, Maudslay, I, pi. 37 . . 68 Figure 88 . Head in serpent mouth, Tikal, Stela 1. 68 Figure 89. The Roman-nosed God as Sky God: a, Yaxchilan, Stela 1; b, Yaxchilan, Stela 4; c, Dresden Codex, p. 56; d, Dresden Codex, p. 55. 69 Figure 90. Two-headed monster with face of God D, Dresden Codex, pp. 4 and 5. 71 xviii LIST OF FIGURES IN THE TEXT. Figure 91. The Kin period glyph and the Roman-nosed God as Sun God: a, Copan, Stela 9, Maud- slay, I, pi. 110; b, Copan, Stela I, Maudslay, I. pi. 31; c, Yaxchilan, Lintel 23, Maudslay, II, pi. 98; d, Chichen Itza, Casa Colorada, Maudslay, III, p. 24; e, Copan, Stela A, Maudslay, I, pi. 24; /, Palenque, Temple of the Sun, Maudslay, IV, pi. 88 ; g, Copan, Stela M, Maud¬ slay, I, pi. 74. 72 Figure 92. Conventionalized head of Roman-nosed God, Piedras Negras, Stela 14. 73 Figure 93. Sculptured block showing Two-headed Dragon and the face of the Roman-nosed God, Northeast group, Copan. 73 Figure 94. Pottery flask with face of Roman-nosed God, Uloa Valley, Gordon, 1898, a, p. 19. Ori¬ ginal in Peabody Museum. 74 Figure 95. Jadeite slab representing a Sun God, Ocosingo. Squier collection, American Museum of Natural History. 74 Figure 96. Intermediate series: Long-nosed God and’Roman-nosed God: a, Copan, Stela D, Maud¬ slay, I, pi. 46; b, Copan, Altar 0, Maudslay, I, pi. 85; c, Copan, Stela A, Maudslay, I, pi. 26; d, Copan, Stela 2, Maudslay, I, pi. 101; e, Palenque, Temple of the Cross, Maudslay, IV, pi. 71; /, Copan, Stela H; g, Tikal, Lintel of Temple III, Maudslay, III, pi. 78; h, Copan, Stela I, Maudslay, I, pi. 63. 75 Figure 97. Close association of God B and God D : a and b, Dresden Codex, pp. 41 and 12 . . 75 Figure 98. God G, the Sun God, Dresden Codex, p. 15. 75 Figure 99. Altar of Stela F, Copan, Maudslay, I, pi. 114. 76 Figure 100. Realistic drawing of a jaguar, Dresden Codex, p. 8 . 77 Figure 101. Jaguar on a potsherd from Mound 36, Copan. Original in Peabody Museum ... 77 Figure 102. Various representations of birds: a, Copan, back of Stela B; b, Copan, sculptured stone from Hieroglyphic Stairway; c, Copan, detail from front of Stela B, Maudslay, I, pi. 37; d, Codex Borgia, p. 51; e, Palenque, Temple of the Cross, Maudslay, IV, pi. 71; /, Codex Per- esianus, pi. 6 ; g, Dresden Codex, p. 36; h, Quirigua, Altar B, Maudslay, II, pi. 14; i, Tro-Corte- sianus Codex, p. 85; Codex Peresianus, pi. 4; k, Palenque, Palace, House C, Maudslay, IV, Pi- 28. 78 Figure 103. Quetzal bird on conventionalized tree, Yaxchilan, Lintel 5. 79 Figure 104. The Moan bird, Dresden Codex, p. 11. 79 Figure 105. Highly modified birds as period glyphs: a, d and g, Copan, Stela D, Maudslay, I, pi. 48; b, e and h, Quirigua, Stela D, Maudslay, II, pi. 25; c, j and t, Quirigua, Altar B, Maudslay, II, pi. 14. 80 Figure 106. Examples of sculptured feathers: a, Copan, Stela A, Maudslay, I, pi. 26; b, Copan, Stela D, Maudslay, I, pi. 45; c, Piedras Negras, Stela 7; d, Copan, Stela D, Maudslay, I, pi. 46; e and/, Copan, Stela H, Maudslay, I, pis. 61 and 56; g, Copan, moulding representing feathers, original stone in Peabody Museum; h, Seibal, Stela 10; i, Piedras Negras, Stela 7; Uxmal, House of the Turtles, Stephens, 1843, I, pp. 311. 81 Figure 107. Head showing manipulation of feathers, Yaxchilan, Lintel 16, Maudslay, II, pi. 84 . 82 Figure 108. Shells and figures associated with them: a, Codex Peresianus, pi. 6 ; b, engraved pot from Chama, Dieseldorff, 11893, p. 378; c, Dresden Codex, p. 41; d, Dresden Codex, p. 37; e, Tikal, Lintel of Temple III, Maudslay, III, pi. 78; /, Codex Nuttall, p. 75; g, Codex Borgia, p. 8 ; h, Palenque, Temple of the Foliated Cross, Maudslay, IV, pi. 81. 83 Figure 109. God N, Dresden Codex, p. 23. 84 Figure 110. Man with shell attached, Chichen Itza, Seler, 1908, p. 197. 84 Figure 111. Girdle ornaments of shell: a, Palenque, Temple of the Inscriptions, Maudslay, IV, pi. 5; b, Ixkun, Stela 1, Maudslay, II, pi. 69; c, Copan, Stela B, Maudslay, I, pi. 34; d, Copan, Stela P, Maudslay, I, pi. 87. 84 Figure 112. Anthropomorphic bat, Copan: a, Stela D, Maudslay, I, pi. 4S; b, Stela D, Maudslay, I, pi. 46; c, Altar T, Maudslay, I, pi. 96. 85 Figure 113. Realistic drawing of a deer, Dresden Codex, p. 60. 85 Figure 114. Representation of human sacrifice, Piedras Negras, Stela 11. 85 Figure 115. Various representations of bones and death: a, Chichen Itza, Temple of the Jaguars, Lower Chamber, Maudslay, III, pi. 45; b, Copan, Altar Gl; c, Copan, Altar R, Maudslay, I, pi. 95; d, Tikal, Altar 16; e, Tikal, Lintel of Temple III, Maudslay, III, pi. 78; /, Copan, Stela H, Maudslay, I, pi. 56; g, Palenque, Temple of the Cross, Maudslay, IV, pi. 72; h, Tro-Cortes- LIST OF FIGURES IN THE TEXT. xix Page ianus Codex, p. 33; i, Copan, Back of Stela B; j, Copan, Stela I, Maudslay, I, pi. 63; k, Palen- que, Palace, House D, Maudslay, IV, pi. 37; l, Uxmal, Torso from the House of the Governor, collected by Le Plongeon, original in the American Museum of Natural History; m, Dresden Codex, p. 53; n, Tikal, Detail of frieze, Temple 55; o, Sculptured stone, environs of Uxmal, Stephens, 1843, I, p. 367; p, Palenque, Palace, House D, Maudslay, IV, pi. 37; q, Mexican Codex. 86 Figure 116. Death God, Dresden Codex, p. 12. 87 Figure 117. Glyphs indicating death: a, Dresden Codex, p. 63; b, Dresden Codex, 10; c, Quirigua, Stela F, east side, Maudslay, II, pi. 40; d, Copan, Stela I, Maudslay, I, pi. 65. 87 Figure 118. Grotesque heads: a, Palenque, Temple of the Foliated Cross, Maudslay, IV, pi. 81; b, Quirigua, Altar P, Maudslay, II, pi. 60 a. 88 Figure 119. Fish with a face on its back, attached to shell. Collection of E. P. Dieseldorff ... 88 Figure 120. Girdle strap wrappings modified into an inverted face, Copan, Stela B, Maudslay, I, pi. 34. 88 Figure 121. Glyphs of the day Kan: a, Tro-Cortesianus, p. 83; 6, Tro-Cortesianus, p. 24; c, Yax- chilan, Lintel 21; d, Palenque, Temple of the Cross, Maudslay, IV, pi. 77. 88 Figure 122. Various representations of the Maize God: a, Dresden Codex, p. 14; b, Dresden Codex, p. 12; c to e, Peresianus Codex, pi. 19. 89 Figure 123. The Maize God in the sculptures: a, Palenque, Palace, Enclosed Corridor, sculpture over vault in passage; b, Tikal, Lintel of Temple IV, Maudslay, III, pi. 78; c, Copan, Interior step of Temple 11, Maudslay, I, pi. 8; d, Palenque, Temple of the Foliated Cross, Maudslay, IV, pi. 81; e, Copan, Stela IL, Maudslay, I, pi. 59; /, Quirigua, Stela H. 89 Figure 124. God C and his hieroglyph, Dresden Codex, p. 5. 90 Figure 125. God C with a mottled green body, Codex Peresianus, pi. 18. 90 Figure 126. God C, Tro-Cortesianus Codex, p. 10. 90 Figure 127. Hieroglyphs containing face of God C: a, Palenque, Palace, House C, Maudslay, IV, pi. 24; b, Copan, Stela A, Maudslay, I, pi. 30; c to e, Tro-Cortesianus Codex, pp. 21, 74 and 78 90 Figure 128. Sun symbols: a, Palenque, Palace, House A, IV, pi. 10; b, Yaxchilan, Stela 4; candd, Dresden Codex, pp. 66 and 55. 91 Figure 129. Sun disk represented in fresco, Chichen Itza, Inner Chamber of the Temple of the Jaguars. 92 Figure 130. Symbols that may represent the moon: a, Dresden Codex, p. 55; b, Palenque, Palace, House E, Maudslay, IV, pi. 43; c, Palenque, Palace, House A, Maudslay, IV, pi. 11 .... 92 Figure 131. Venus symbols: a, Palenque, Palace, House E, Maudslay, IV, pi. 43; b, Copan, Altar R, Maudslay, I, pi. 94; Dresden Codex, p. 56. 93 Figure 132. Astronomical symbols: a, Copan, Stela N, Maudslay, I, pi. 82; b, Dresden Codex, p. 68 ; c, Palenque, Temple of the Inscriptions, Maudslay, IV, pi. 56; d, Dresden Codex, p. 53; e, Dresden Codex, p. 54; /, j and k, Palenque, Temple of the Inscriptions, Maudslay, IV, pis. 54, 55 and 56; g, Palenque, Palace, House A, Maudslay, IV, pi. 11; h, Dresden Codex, p. 52; i, Dresden Codex, p. 53. 93 Figure 133. Composite monster and its glyph, Dresden Codex, p. 45. 94 Figure 134. Series showing modifications of glyphs: a, Peresianus Codex, pi. 24; 6, Dresden Codex, p. 6; c, Tro-Cortesianus Codex, p. 17; d, Landa, 1864, p. 244; e, Copan, Interior step of Temple 11, Maudslay, I, pi. 8;/, Quirigua, Stela E, Maudslay, II, pi. 31; g, Quirigua, Stela F, Maud¬ slay, II, pi. 40; h, Quirigua, Stela E, Maudslay, II, pi. 31; i, Quirigua, Stela F, Maudslay, II, pi. 40; j, Quirigua, Stela D, Maudslay, II, pi. 26. 94 Figure 135. Hieroglyph with entire figures, Copan, Stela D, Maudslay, I, pi. 48. 95 Figure 136. Representations of Indian huts in fresco, Chichen Itza. 99 Figure 137. Series showing development of the sanctuary: a, Copan, Temple 22a; b, Tikal, Temple I; c, Copan, Temple 22; d, Chichen Itza, Castillo; e, Palenque, Temple of the Sun .... 99 Figure 138. Structures showing the extreme development of the portico, Chichen Itza, Group of the Columns.100 Figure 139. Partial ground-plan of the Group of the Monjas, Chichen Itza, Maudslay, III, pi. 3 101 Figure 140. Ground-plan, House of the Turtles, Uxmal.101 Figure 141. Ground-plan of Akat’cib, Chichen Itza.102 Figure 142. Ground-plan, Santa Rosa Xlabpak, Maler, 1902, p. 226 . 102 XX LIST OF FIGURES IN THE TEXT. Page Figure 143. Elevations, Santa Rosa Xlabpak: a and b, cross-section, Maler, 1902, p. 226 . . . 103 Figure 144. Assemblage of Edifice 5, Chacmultun, Thompson, 1904, p. 19.104 Figure 145. Restoration of balustrade, Hieroglyphic Stairway, Gordon, 1902, p. 13.109 Figure 146. Typical cross-section of an interior wall.107 Figure 147. Cross-section of Temple 5, Tozzer, 1911, p. 122.110 Figure 148. Plans of Yaxchilan temples: a, Structure 25, Maudslay, II, pi. 77; b, Cross-section of Structure 33, Maudslay, II, pi. 77; c, Ground-plan of Structure 33, Maudslay, II, pi. 77 . . 112 Figure 149. Plans of the Casa Colorada, Chichen Itza.113 Figure 150. Cornice forms: a, Southern area; b, Usumacinta area; c, Northern area; d, Lower cornice of the Iglesia, Chichen Itza; e, Palenque, Temple of the Cross; /, Palenque, Palace; h, Second Range, Monjas, Chichen Itza; i, Caracol, Chichen Itza; j, Monjas, Chichen Itza; k, Labna; l, Chacmultun.114 Figure 151. Modifications of door jamb: a, common form; b, Uxmal; c, Chichen Itza.115 Figure 152. Gargoyle in form of serpent head, Copan, Gordon, 1896, p. 7.117 Figure 153. Ornamental niche on fagade, Uxmal, The Nunnery, North Range, Catherwood, 1844, pi. 15.118 Figure 154. Faces limited to rectangular spaces, Chichen Itza: a, Temple of the Jaguars, Lower Chamber; b, Temple with two pilasters made of re-used blocks, Group of the Columns ... 118 Figure 155. Face in profile, Stela B, Copan, Maudslay, I, pi. 37.118 Figure 156. Mask panel at base of Stela 4, Yaxchilan.119 Figure 157. Mask panel at base of stela, La Hondradez.119 Figure 158. Mask panel on tower, La Hondradez. After sketch by Dr. Tozzer.119 Figure 159. Mask panel, partly restored, Nakum. After sketch by Dr. Tozzer.119 Figure 160. Mask panel, partly restored, Nakum. After sketch by Dr. Tozzer.120 Figure 161. Corner mask built up mosaic fashion, Palace, Labna .120 Figure 162. Mask panel, Palace, Labna.120 Figure 163. Mask panel over doorway, Xkichmook.121 Figure 164. Simplified mask panel.121 Figure 165. Superimposed mask panels showing elaboration, East Range of the Nunnery, Uxmal 122 Figure 166. Mask panel over doorway, showing extreme elimination, Palace, Labna.123 Figure 167. Mask panel, East Range of the Nunnery, Uxmal, Catherwood, 1844, pi. 15 .... 123 Figure 168. Mask panel showing substitution, House of the Dwarf, Uxmal, Catherwood, 1844, pi. 13 124 Figure 169. Geometric panel, Dsibiltun.124 Figure 170. Geometric panel, Dsibiltun.124 Figure 171. Profile panel, Yaxchilan, Stela 7.125 Figure 172. Panel at base of Stela 6, Yaxha .125 Figure 173. Profile at side of doorway, Copan, Temple 11.125 Figure 174. Remains of wall decoration, Nakum. After a sketch by Dr. Tozzer.126 Figure 175. Profile mask panel, Hochob, Principal Structure.126 Figure 176. Assemblage of profile and front view mask panels, ruined temple, Hochob.126 Figure 177. Mosaic elements, Labna. Original stones in Peabody Museum.128 Figure 178. Mosaic elements with examples of their development. Chichen Itza.128 Figure 179. The shield as a mosaic element, Labna .128 Figure 180. Mosaic elements used in fagade decorations, Labna. Original stones in Peabody Museum.129 Figure 181. Typical banded section of the banded column, Labna.129 Figure 182. Richly ornamented wall with projecting sculpture, Uxmal, Temple of the Magician. The sculptured head now removed to Museo Nacional, Mexico City.129 Figure 183. Detail of fagade decoration of banded columns, Tantah.130 Figure 184. Potsherds showing different processes of ornamentation: a, b, c, and e, Copan; d, Peten region. Original specimens in Peabody Museum.134 Figure 185. Decoration on bowl from near Pet'o. Original in Peabody Museum.135 Figure 186. Bowl from Calcetok. Collection of Senor Enrique Camara. 130 LIST OF FIGURES IN THE TEXT. xxi Page Figure 187. Bowl from the Island of Jaina. Mus 6 e de Trocaddro, Hamy, 1897, pi. 27 ... . 136 Figure 188. Engraved potsherd from Santa Cruz Quiche. Original in Peabody Museum .... 136 Figure 189. Potsherds showing applied relief decoration. Quen Santo, Seler, 1901, c. pp. 140 and 171 . 137 Figure 190. Jaguar head vase, Copan, Tomb 1, Gordon, 1896, p. 48. Original in Peabody Museum. 137 Figure 191. Crude painted figures on food bowls, Copan.138 Figure 192. Design on finely painted bowl from Copan. Original in Peabody Museum.139 Figure 193. Design on bowl, El Jicaro, Guatemala. Original in Peabody Museum.140 Figure 194. Geometric motives used in pottery decoration. 141 Figure 195. Carved jadeite amulet with inscription. Bishop Collection, Metropolitan Museum of Fine Arts. 143 Figure 196. Jadeite amulet with inscription on back, Island of Jaina. In American Museum of Natural History.. Figure 197. Perforations of a jadeite amulet. In Field Museum of Natural History.144 Figure 198. Jadeite amulet, Chichen Itza. In Field Museum of Natural History. 144 Figure 199. Jadeite ear plug, Ocosingo. Squier Collection, American Museum of Natural History 145 Figure 200. Clay figurine showing use of jadeite ear plugs, Coban, Seler, 1895, d, p. 34.145 Figure 201. Portion of design on marble vase from Honduras. Squier Collection, American Museum of Natural History. 145 Figure 202. Copper bell with coarse wire decoration, Honduras. In American Museum of Natural History. . Figure 203. Baskets represented on sculptures: a, Yaxchilan, Lintel 24; b, Chichen Itza, Lower Chamber of the Jaguars, Maudslay, III, pi. 49. 146 Figure 204. Imitation of basket weaves on painted potsherds, Uloa Valley. In Peabody Museum 147 Figure 205. Details of textile ornamentation from the sculptures: a, Copan, Hieroglyphic Stairway; b, f, g, Piedras Negras Stelae 35, 1 and 14; c and d, Details of lace work; e and h, Yaxchilan, Lintel 25. 24 g Figure 206. Garments represented in sculptures and codices: a and 6 , Chichen Itza, Lower Chamber Temple of the Jaguars; c-e, Dresden Codex, pp. 5, 23, 27 . 149 Figure 207. Woman richly attired, Yaxchilan, Lintel 15. 149 Figure 208. Facial tattooing: a-f, Details mostly from figurines, Schellhas, 1890, p. 213; /, g, Yax¬ chilan, Lintel 24. 15 Q Figure 209. Pottery whistle in shape of head showing tattooing, Tecolpa, Tabasco. In American Museum of Natural History.. Figure 210. Peccary skull with incised drawings, Copan. In Peabody Museum. 151 Figure 211. Ceremonial staff ending in a hand: a, Dresden Codex, p. 27; b, Codex Borgia ... 153 Figure 212. Cross-sections of Copan stelae. I 57 Figure 213. Proportions of the body on Copan stelae. 158 Figure 214. Altar 14, Copan. 16 i Figure 215. Altar of Stela I, Copan. 16 i Figure 216 . Rectangular altar, Copan, Sculpture Y, Gordon, 1902,6, pi. 14. 161 Figure 217. Glyphs from jadeite statuette, San Andres Tuxtla, Holmes, 1907, pi. 36.171 Figure 218. Introducing glyph on Leiden Plate. 171 Figure 219. Head of jadeite statuette, San Andres Tuxtla, Holmes, 1907, pi. 34.172 Figure 220. Foot of figure on Leiden Plate showing serpent-head ankle ornament.173 Figure 221. Feet of Copan stelae showing serpent-head ankle ornaments: a and 6 , Stela D, Maud¬ slay, I, pis. 45 and 47; c, Stela N, Maudslay, I, pi. 82. 173 Figure 222. Foot showing conical ankle ornament, Quirigua, Stela F, Maudslay, II, pi. 36 . . 176 Figure 223. Altar N, Quirigua. 177 Figure 224. Plan of Main Temple, Naranjo, showing locations of stelae.178 Figure 225. One end of Ceremonial Bar of Stela 32. 179 Figure 226. Ceremonial Bar, Stela 6 , Naranjo. 179 Figure 227. Ceremonial Bar, Stela 7, Naranjo.182 xxii LIST OF FIGURES IN THE TEXT. Page Figure 228. Bound captive, Stela 1, Ixkun, Maudslay, II, pi. 69.182 Figure 229. Ceremonial spears: a and 6, Chichen Itza, Temple of the Jaguars, Maudslay, III, pis. 47 and 4S; c, Palenque, Temple of the Sun, Maudslay, IV, pi. 88.183 Figure 230. Head on fragment of Lintel 41, Yaxchilan.188 Figure 231. Short wands with head of the Long-nosed God: a, Palenque, Temple of the Foliated Cross, Maudslay, III, pi. 81; b, Yaxchilan, Lintel 14 .193 Figure 232. Stela at ruins of Tzendales. After a field sketch by Dr. Tozzer.197 Figure 233. Diagram showing growth of the Substructure of the Casa de Monjas, Chichen Itza 203 Figure 234. Details of re-used mask panels, Chichen Itza: a, Lateral mouth ornament of mask panel of frieze of the Substructure of the Casa de Monjas; b and c, Details from center mask of the flying fagade of the Iglesia.204 Figure 235. Serpent head with human head in its mouth, Uxmal, Nunnery Quadrangle, West Range.205 Figure 236. Atlantean figure, Temple of the Initial Series, Chichen Itza.207 Figure 238. Atlantean figure, Tlascala, Seler, 1902-1903, p. 92 . 207 Figure 238. Breast ornaments representing birds, Chichen Itza: a to d, Temple of the Jaguars, Maudslay, pis. 49 and 50 . 208 Figure 239. Sun and star symbols of the Nahua type: a, Mitla, Seler, 1895, c, pi. 1; b, Totonacan area, Strebel; c, Santa Rita, Gann, 1897-1898, pi. 29; d, Santa Rita, Gann, 1897-1898, pi. 31; e, Codex Porfirio Diaz; /, Codex Fej6rv£ry-Mayer, p. 25; g, Vienna Codex; h, Chichen Itza, Temple of the Jaguars, Maudslay, III, pi. 38; i, Mitla, Seler, 1895, c, pi. 1.209 Figure 240. Speech scrolls: a, Hieroglyph for Cuernavaca (Cuauhnahuac), Penafiel, 1885, pi. 91; b, Hieroglyph for Cuicatlan, Penafiel, 1897, pi. 30; c and d, Xochicalco, Penafiel, 1890, pi. 190; e, Santa Lucia Cosumahwalpa, ILabel, 1878, pi. 5; /, h, i, k, l, m, n, Chichen Itza, Temple of the Jaguars, Maudslay, III, pis. 46, 49 and 50; g, Dresden Codex, p. 34; j, San Juan Teotihuacan, Penafiel, 1899, pi. S2; o, Chavero, 1900-1901, Pt. I.210 Figure 241. Speech scroll representing Long-nosed God, Chichen Itza, Temple of the Jaguars, Maud¬ slay, III, pi. 51.211 Figure 242. Terra cotta tile from Tezcoco. In American Museum of Natural History.211 Figure 243. Nahua subjects similar to those of Chichen Itza: a, Codex Vaticanus, 3773, p. 14; b, Codex Borbonicus, p. 17.212 Figure 244. Design showing Maya style, Chichen Itza, Temple of the Jaguars, Maudslay, III, pi. 46 212 Figure 245. A stucco panel at Acanceh.213 Figure 246. Figure that may represent Manikin God, Santa Rita, Gann, 1897-1898, pi. 30 . . . 213 Figure 247. Head of Roman-nosed Sun God, Santa Rita, Gann, 1897-1898, pi. 31.214 Figure 248. Head of figurine with animal headdress, Santa Rita, Gann, 1897-1898, pi. 32 ... 214 Figure 249. Figurine from Cerro de los Idolos, Tabasco. In Mus6e de Trocad6ro. Hamy, 1897, pi. 24.215 Figure 250. Typical Lacandone incense burner.215 Figure 251. Human heads in animal jaws: a, Piedras Negras, Stela 26; b and c, Figurine from Oaxaca; d, Stone sculpture from Oaxaca, Seler, 1901-1902, p. 41; e, Piedras Negras, Stela 7; /, Onyx slab, Chaleo, in Field Museum of Natural History, Holmes, 1S95-1897, m, p. 305; g, Chichen Itza, Temple of the Jaguars, Maudslay, III, pi. 51; h, Sculptured bowlder, in Field Museum of Natural History, Holmes, 1895-1897, pi. 56 . 221 Figure 252. Human head in animal jaws, Oaxaca, Teotitlan del Camino, Seler, 1895, c, pi. 13 . . 222 Figure 253. Human head in jaws of serpent, Calendar Stone, Museo Nacional, Mexico.222 Figure 254. Serpent head on pottery, Cholula, Seler, 1902-1908, II, p. 297 . 222 Figure 255. Serpent head on engraved bowl from Cholula. In American Museum of Natural History.222 Figure 256. Serpent head on bowl from Cholula. Museo de la Academia de Bellas Artes, Puebla 223 Figure 257. Serpent heads on Cholula pottery: a, Seler, 1902-1908, vol. 2, p. 296; b, Museo Na¬ cional, Mexico; c, In American Museum of Natural History; d, Museo Nacional, Mexico . . 223 Figure 258. Conventionalized serpent heads arranged in a swastika, Museo Nacional, Mexico . . 223 Figure 259. Lineal forms derived from serpent head, Cholula pottery; a, American Museum of Nat¬ ural History; b and c, Museo Nacional.223 LIST OF FIGURES IN THE TEXT. xxiii Page Figure 260. Serpent, curvilinear style, Codex Vaticanus, 3773, p. 49 . 224 Figure 261. Serpent head, angular style, Codex Vaticanus, 3773, p. 71.224 Figure 262. Double-headed serpent forming a bowl, Codex Vaticanus, 3773, p. 55 . 224 Figure 263. Serpent heads painted on pottery, State of Vera Cruz, Strebel, 1899, pi. 9.224 Figure 264. Detail of stone yoke, State of Vera Cruz, Holmes, 1895-1897, p. 211.225 Figure 265. Reptile head on stone yoke, State of Vera Cruz, Holmes, 1895-1897, p. 316 .... 225 Figure 266. Human figure, Xochicalco, Penafiel, 1890, pi. 190 . 225 Figure 267. Small pottery objects of earliest type, Mexico. In American Museum of Natural History: a-c, Atzcapotzalco; d, San Juan Teotihuacan; e, Tuxpan, Jalisco; /, Zapotlan, Jalisco; g, Cuernavaca.228 Figure 268. Design on interior of bowl, Calchaqui area, Argentine. In Field Museum of Natural History.239 Figure 259. Snakes on vessels from Chimbote, Peru. In American Museum of Natural History . 239 Figure 270. Fragment of a garment, Pachacamac, Peru, Ulile, 1903, pi. 4.239 Figure 271. Detail on pot from Pachacamac, Peru. In American Museum of Natural History . 240 Figure 272. Serpent heads on pottery from Nicaragua. In American Museum of Natural History 240 Figure 273. Nahua sculptures, Nicaragua: a, Subiaba, Squier, 1852,1, p. 319; b, Zapatero, Squier, 1852, II, p. 52; c, Pensacola, Squier, 1852, II, p. 36 . 240 Figure 274. Head of basalt from near Tegucigalpa, Hamy, 1896, pi. 1.241 Figure 275. Plumed serpents on Casas Grandes pottery. In American Museum of Natural His¬ tory .241 Figure 276. Reclining figure with animal headdress, Casas Grandes pottery. In American Museum of Natural History.241 Figure 277. Plumed serpent of Zuni: a, Cushing, 1882-1883, p. 515; b, National Museum, Washington.242 Figure 278. Horned serpent on prehistoric bowl from Puye, New Mexico. In Museum of the South¬ west, Sante Fe.242 Figure 279. Horned serpent and cloud symbols on prayer meal bowl from Perige (Old San Ildefonso). In Museum of the Southwest, Santa Fe.242 Figure 280. Shell gorget with rattlesnake design, Tennessee, Holmes, 1880-1881, pi. 65 . 242 Figure 281. Winged and horned snakes: a, Alabama, Moore, 1905, p. 229; b, Arkansas, Holmes, 1898-1899, p. 91.243 Figure 282. Serpent engraved on a vase from Georgia, Holmes.243 Figure 282. Copper platesfrom the Mound Area: a, Illinois, Thomas, 1890-1891, p. 309; b, Georgia, Thomas, 1890-1S91, pi. 17; c, Thomas, 1890-1891, p. 304; d, Thomas, 1890-1891, p. 305; e, Thomas, 1890-1891, pi. 17. 244 Figure 284. Designs on shell gorgets: a, Kentucky, Moorehead, 1910, II, p. 130; b, Alabama, Moore, 1907, p. 398 . 245 Figure 285. Shell masks: a, Tennessee, Holmes, 1880-1881, pi. 69; b and c, Virginia, Holmes, 1880- 1881, pi. 69 . 246 Figure 286. Horned snake on buffalo skin lodge, Dakota Indians, Weygold, 1903 . 246 MAYA ART INTRODUCTION Area. The region in which remains of the pre-Columbian Maya civilization are found corresponds closely with that still inhabited by Indians speaking dia¬ lects of the Maya linguistic stock. Roughly it lies between 87° and 94° west longitude and 14° and 22° north latitude. More exactly it comprises, in Mex¬ ico, the states of Tabasco and Chiapas and the peninsula of Yucatan (with the states of Campeche and Yucatan and the territory of Quintana Roo), in addi¬ tion to the whole of British Honduras, the two-thirds of Guatemala lying north of the Motagua River, and a considerable portion of Honduras including the head-waters of the Copan River, the lower course of the Uloa, and, in all prob¬ ability, the rich central valley of Comayagua. Relation to Surrounding Cultures. Upon the west the Maya area adjoins those of the Zapotecan and Nahua cultures. Although there is hardly a doubt concerning the common origin of these three most important civilizations of Mexico and Central America, yet environmental, chronological and linguistic differences have made them at least superficially distinct. In all three there were apparently two or more periods of widespread high culture, each followed by a period of disintegration and lower culture. It has been argued that all were branches of an early civilization located on the plateau of Mexico and referable to the legendary Toltecs. A detailed dis¬ cussion of this question will be taken up further on, after evidence has been pre¬ sented. In this place it is only necessary to point out that, owing to the imper¬ fections of Nahua reckoning, all dates before 1325, the year generally accepted for the founding of Tenochtitlan, the Aztec capital, must be regarded as largely fictitious. It will be shown that the Toltec or pre-Aztec remains were for the most part contemporary with the brilliant period of the cities of northern Yuca¬ tan, but much later than the first florescence of southern Maya art. The Zapo¬ tecan and Nahua cities, found in a flourishing condition by the Spaniards, apparently rose after the Maya culture had declined. It seems unnecessary to consider at length the various wild speculations concerning Old World origins of New World civilizations. Lord Kingsborough’s attempt to identify the nations of America with the lost tribes of Israel was in keeping with the speculative age in which he wrote. The far-fetched theories of Dr. Le Plongeon must be laid to an over vivid imagination, although there is no gainsaying the painstaking enthusiasm of this unfortunate student. But no reasonable excuse can now be found for writers who, on the strength of this or that similarity, cheerfully leap the bounds of space, time and reason to derive the religious and artistic conceptions of the Maya from Egypt, India or China. The evidence these writers present is always insufficient and usually wrong. 2 MAYA ART. Where real similarities exist they probably can be explained by pure chance or by psychic unity. In determining origins, however, account may well be taken of the single outlying group of the Maya-speaking peoples, the Huasteca, who inhabit the low coast region north of Vera Cruz, and in whose territory many remains of cities as yet undescribed are known to occur. It is possible but not likely that a careful study of this disconnected group will indicate a northern origin for Maya arts. An origin to the south of the stated limits is hardly conceivable, owing to the great and sudden falling off in handicraft and ideas once the south¬ ern frontier has been crossed. Such similarities as do exist may easily be ac¬ credited to the Nahua colonies which, in the last centuries before the coming of Europeans, were planted even farther south than Lake Nicaragua. No matter, however, to what other region fuller investigation may refer the humble begin¬ nings of Maya art, the indisputable fact remains that in all essential and char¬ acteristic features it was developed upon its own ground. Natural Divisions. The Maya area, as above defined, contains three prin¬ cipal natural divisions. In each of these the differences in climate, in natural resources, and in topography are marked enough to have had a decided effect upon the material culture of the inhabitants. The first of these divisions com¬ prises the peninsula of Yucatan; the second, the great central valley; the third, the cordilleran plateau on the south and west. Since the entire region lies south of 22°, it is distinctly tropical except where the altitude counteracts, and is sub¬ ject to the doldrum rains under the high sun. The duration of this summer rainy season is less in Yucatan than in the two other regions of greater land relief. The peninsula of Yucatan is a limestone plain of recent geological formation, with its highest ridges but a few hundred feet above the sea. 1 It has no river valleys because, owing to the porous and soluble nature of the limestone, the drainage 3 is subterranean. There are many caverns and sink-holes. The cav¬ erns seldom show signs of former habitation and then only as retreats. 3 The sink-holes are often very large and form natural wells or cenotes. These cenotes determined the location of most ancient and modern towns. Often, however, artificial reservoirs and cisterns, called chultunes, 4 were constructed. In the southeast several large lakes occur, Lake Peten being the most important. The soil of Yucatan is shallow, and although trees grow rapidly and in dense masses they seldom attain great height. The universal building stone is limestone, which also is burned for lime. The wide valley plain of the meandering Usumacinta and its maze of tribu¬ tary streams is a region little known and poorly mapped. It supports at present a small, roving population of wood-cutters and a few hundred squalid Lacan- done Indians, though it must formerly have been the seat of wealth and power, to judge from its ruined cities, such as Yaxchilan, Piedras Negras, and Seibal. Like Yucatan, the rocks are young and calcareous. Maler is probably in error when he refers to sandstone at Piedras Negras. In the great alluvial valley stone may be had at but a few points where the hills come close to the river. Conse¬ quently many sites show now only the earthen foundation mounds from which the wooden superstructures have long since vanished. Timber is plentiful, the 1 For a discussion of the geology see Sapper, 1896. 3 Thompson, 1897, a; Mercer, 1896. 2 Casares, 1905; Mercer, 1896, p. 21, footnote. 4 Thompson, 1897, b ; Stephens, 1843, II, p. 227. INTRODUCTION. 3 whole region being covered with a dense tropical forest of mahogany and other large trees. The rivers form the highways of travel. The surface of the land is marked by extensive swamps and a number of lakes; hills of moderate elevation vary the topography, and on the southern and western margin the land rises suddenly to the continental plateau. This plateau attains an average height above the sea of about 8000 feet, but is deeply dissected by the Chiapas, Usumacinta and Motagua river sys¬ tems. The crest of the continental range lies so close to the Pacific that no large streams flow into that ocean. The plateau swings to the east round the head of the Usumacinta basin and reaches the shores of Lake Izabel in long narrow spurs, while outlying ridges extend well into British Honduras. The flora of the plateau region is characterized by the oak and the pine, but much of the country is fairly open and well adapted to agriculture. These uplands formed the highway for migrations north and south, and supported a large heterogeneous population, but were apparently never the seat of such high culture as obtained in the lowlands. Copan and Quirigua are both situated on valley floors. Ledges of old blue limestone and of a soft volcanic tuff furnished an abundant supply of excellent building material at the former, while at the latter city a much harder stone of similar volcanic origin was encountered. Early Notices. The number of early historical references to the Maya Indians is small, partly due to the fact that the principal theatre of action for the Spaniards lay in the valley of Mexico. Few of the soldiers of those strenuous days found time to lay aside the sword. As for the Spanish priests, most were as deeply imbued with fanaticism as were the natives whose culture they sought to destroy. They were incapable of comprehending the real character of the native religion, which they summed up as devil worship. They were true icon¬ oclasts, and went about throwing down idols, burning ancient chronicles and de¬ stroying everything that would keep alive the remembrance of old times. Most of the first-hand information on the culture of the Maya must be gleaned from the writings of Cogolludo, Landa, Lizana and the “Relations of Yucatan” which consist of reports sent in by the heads of various towns and provinces. Excel¬ lent second-hand information is found in the works of the great historians, Herrera, Oviedo and Villagutierre. The first expeditions from Cuba to the mainland made a number of landings along the coast of Yucatan, 1 where the Spaniards met bands of natives and visited their towns. Bernal Dias del Castillo, who sailed and fought with Francisco Hernandez de Cordoba in 1517, with Juan de Grijalva in 1518 and with the redoubtable captain Hernando Cortes in 1519 and for many years afterwards, describes Maya temples and sculptures. Dias 2 writes as follows: “They led us to some large houses very well built of masonry, which were the Temples of their Idols, and on the walls were figured the bodies of many great serpents and snakes and other pictures of evil looking Idols. These walls surrounded a sort of altar covered with clotted blood. On the other side of the Idols were symbols like crosses, and all were coloured. At all this we stood wondering, as they were things never seen or heard of before.” At first reading this might be considered an adequate description, from the blunt pen of a sixteenth-century soldier, of the run of Maya architecture and 1 For an early map see Valentini, 1902. 1 Dias del Castillo, 1908, I p. 1 4 MAYA ART. art. But on examining the ruins 1 of the buildings referred to in this and similar notices (on Cozumel Island, near Cape Catoche, etc.), it is evident that they must have been decidedly inferior to the great temple structures of interior Yucatan. Though the same principles of construction were applied in both localities, yet the workmanship of the temples on the seacoast was much cruder, and the ornamentation of the fagades much less permanent. Beyond doubt, the same people erected all these buildings but during different stages of culture. It is possible that Tuloom, on the east shore of Yucatan, was visited or observed from the sea by the expedition of Juan de Grijalva. His chaplain, Juan Dias, speaks 2 enthusiastically of a city facing the sea and having great walls and towers. He compares this city to Seville, and mentions that it was inhabited by a large population. Unfortunately Tuloom has since been visited only by Stephens, who found it in ruins, and by Dr. Howe in 1911, who was forced to limit his stay to two days. Stephens 3 comments on the fresh appear¬ ance of the walls, and expresses his belief that Tuloom was inhabited until after the Conquest. This may be the case, because in several architectural features Tuloom varies from cities known to be early. In regard to the inland cities of Yucatan there are early notices of Chichen Itza and Uxmal that are worthy of consideration. These passages indicate that both cities were fallen from their ancient glory, although still the centers for certain religious rites. There is no evidence that the stone buildings of either city were actually inhabited at the time of the Spanish conquest. Bishop Diego de Landa, 4 writing about 1566, says of Chichen Itza: “The elders among the Indians say that they remember to have heard from their ancestors that in that place there once reigned three Lords who were broth¬ ers, and who came to that land from the west. And they brought together in these cities a great number of towns and people, and ruled them for some years with justice and in peace . . . that soon they split into factions, so wanton and licentious in their ways, that the people came so greatly to loathe them that they killed them, laid the town waste and themselves dispersed, abandoning the buildings and this beautiful site ...” Landa 5 also gives a fairly accurate description and a crude plan of the famous Castillo, and mentions survivals of the ancient religious practices in connection with the sacred cenote. Stephens 6 quotes at length from the title paper to the land upon which the ruin of Uxmal is situated. It appears that the Regidor received by royal grant certain meadows and places, uncultivated and useless except for pasturage, whereby a great service would be done to God because “it would prevent the Indians in those places from worshiping the devil in the ancient buildings which are there, having in them idols, to which they burn copal, and performing other detestable sacrifices, as they are doing every day notoriously and publicly.” This grant having been contested by an Indian claimant, the matter was settled by payment. A later document runs as follows: “In the place called the edifices 1 Le Plongeon, quoted in Salisbury, 1S78, pp. * Landa, 1864, p. 340. Passage translated by 76-84; Stephens, 1843, II, pp. 365-378, 415-417; Maudslay, 1889-1902, III, pp. 6-7. Holmes, 1895-1897, pp. 57-78. • 1864, pp. 342-344. 1 Dias, Juan, 1838, pp. 11-12. 8 1843, I, pp. 322-325. 3 1843, II, p. 406. INTRODUCTION. 5 of Uxmal and its lands, the third day of the month of January, 1688, ... he walked with me all over Uxmal and its buildings, opened and shut some doors, etc.” The last statement, in regard to the opening and shutting of doors, seems a mere legal formula to indicate acts of possession. Maya temples, in all proba¬ bility, never had doors that opened and shut. While these passages prove con¬ clusively that the buildings were still held sacred by the natives, they cast little light upon the question when Uxmal ceased to be a real city. Of more value in deciding this vexatious question is an account of Uxmal dated a full century earlier (1586). This account, 1 written by one of the com¬ panions of Alonzo Ponce, a Franciscan delegate, is so accurate and detailed that it deserves to be given in full. Not all the buildings of the city were examined and described by this early traveler, yet one can recognize with ease each struc¬ ture taken up, for the descriptions of the outward appearances apply to-day with hardly the change of a word. The curious reader may compare this passage with the excellent modern description of the same buildings by Mr. Holmes. 2 The earlier description is as follows: “On the north of the ranchos where the father delegate was lodged, as has been seen, which is about twenty leagues from Merida, to the south of that city, stands a ku or mul [artificial pyramid], very tall and made by hand. It is very difficult to ascend this by its 150 stone steps, which are very steep and which, from their being very old, are very dilapidated. On the top of this mul a large building [House of the Magician] has been built, consisting of two 3 vaulted rooms, made of stone and lime, the stones being carved with great care on the outside. In old times they took the Indians who were to be sacrificed to these rooms, and there they killed them and offered them to the idols. The father delegate went up this mul as soon as he arrived there, and this surprised the others greatly, since many others did not dare to go up and could not have done so if they had tried. Close to this mul and behind it on the west, there are lower down many other buildings built in the same way with stone and lime and with arches. The stones are carved with wonderful delicacy, some of them having fallen and others badly injured and ruined, while others can still be seen, and there is much in them worth examining. Among these there are four very large and handsome buildings [Nunnery Quadrangle] set in a square form, and in the middle is a square plaza, in which grew a thicket of large and small trees, and even on top of the building there were very large and dense trees growing. The building [South Range of the Nunnery Quadrangle] which faces the south, has on the outside four 1 rooms, and on the inside eight others, all arched with cut stone, and as carefully joined and put together as if very skillful workers of the present had built them. These arches, and all the other old arches which have been found in this province, are not rounded over in the form of a cupola nor like those which are made in Spain, but are tapered as the funnels of chim¬ neys are made when built in the middle of a room, before the flue begins, since 1 Relacion Breve, 1872, LVIII, pp. 455-461. 1 1895-1897, pp. 86-96, and panoramic drawing. Comparison may also be made with Stephens, 1843, I, pp. 166-180; 299-308; 312-318. 3 It should be noted that the padre ascended this pyramid by the main stairway on the eastern side. He thus could gain access only to the two end rooms of the main temple. The center room of this build¬ ing has its only doorway on the west, looking out upon the roof of the two-chambered annex. For the names given to these buildings the reader is referred to the table of nomenclature at the end of this vol¬ ume, p. 252. * Here again the numbering of the rooms is slightly at fault. This building has eight rooms on the outside as well as the inside. There are also two rooms at either end. Such minor inaccuracies need not be wondered at when one considers the luxuri¬ ant tropical vegetation which covered everything. The padre properly noted that this building faced outward as well as inward. It served, in fact, as the fagade of the entire group of four correlated structures. b MAYA ART. both sides draw together little by little and the space between becomes more na rr°w, till on the top one wall is separated from the other by about two feet and there they place a layer, which extends inwards four or five inches on each side and over this they place flags or thin flat stones in a level position, and with these the arch is closed, so that there is no key to the arch, but with the great weight of stone and mortar, which is placed on top and which strengthens the sides, the arch is closed and remains fixed and strong. The ends of this arched building are continuous and straight from top to bottom. At the door of each of the rooms of this building on the inside, there are four rings of stone two on one side and two on the other, — two of them being high up and two lower down and all coming out of the same wall. The Indians say that from these rings those who lived in these buildings hung curtains and portieres, and it was to be noticed that no one of these rooms, nor of all the others, which we found there, had any window, small or large. The rooms were therefore rather dark, especially when they were made double, one behind the other, so that even in this, this idolatrous race gave evidence of the darkness and obscurity of the error in which it was enshrouded. The high lintels of all these doors were made of the wood of the chico zapote, which is very strong and slow to decay as could well be seen, since most of them were whole and sound, although they had been in position from time immemorial, according to the statements of the old Indians. The door jambs were of stone carved with great delicacy. 1 On the fagades of the building, both on those which face the plaza or courtyard, aS j Ve l!- a i S j 0n ^ ose f ace outward, there are many figures of serpents, idols and shields, many screens or latticework, and many other carvings which are very beautiful and fine, especially if one look at them from a distance like a painting of Flanders, and they are all carved from the same kind of stone. In t this building a great arch is made, so that it takes in all the depth of the building, and therefore it is the entrance to the courtyard or the above- mentioned plaza. It would appear that this entrance had been plastered and that on the plaster paintings had been made in blue, red and yellow color, since even now some of them remain and can be seen. Nearly all the rest of the stones had been plastered but not painted. The building [Eastern Range of the Nunnery Quadrangle] which stands at the west, behind the previously mentioned mound of sacrifices, was in the best condition and uninjured. It had four doors which opened on to the courtyard or plaza with as many rooms, arched in the same way as the others, and beyond each room was another, so that there were eight in all. Between these four doors, two on one side and two on the other, there was still another door which opened on the patio, and within this was a very large hall, long and broad, with two small rooms on the sides; and beyond this hall there was another — a little smaller, with two other small rooms — one on each side, so that inside of this one door there were six rooms, four small and two large, making, with the other eight, fourteen rooms which this building contained. On the inside fa¬ cades and ends of this building, there were carved many serpents in stone, and heads of savages and other figures in the manner of shields, and at the four corners (since each building stood by itself and not joined or connected with the other) there were many other carvings cut in the round like a half curve, with tips, which looked like serpent heads, and which stood at half a vara from the rest of the carvings. The building on the north [North Range of the Nunnery Quadrangle] is the tallest, and has more carvings and figures of idols, serpents and shields and other very beautiful things about it, but it is very much injured and the most 1 Stephens states, 1843, I, p. 308, that the door- stances of interior decoration at Uxmal. No repro- ways of the central group of chambers in the Eastern ductions of these sculptures are known. Range are ornamented with sculptures, the only in- INTRODUCTION. 7 of it has fallen. It has ten 1 doors which open on the plaza and another which opens on the eastern end, and inside each one there are two rooms, and so among them all there are twenty-two rooms in that building made of stone and lime, and arched like the others, but the most of them, especially those inside, have fallen. Before the ten doors above mentioned there has been made a terrace, paseo, or walking-place, somewhat broad and open on all sides, to which one ascends from the plaza by steps which are now half in ruins. All this terrace has below it other arched rooms with doors opening on the same plaza, and these are covered and stopped up with stones and earth and with large trees which have grown there. “The building on the west [Western Range of the Nunnery Quadrangle] is very elegant and beautiful on the outside fagade, which looks on the plaza, since serpents made of stone extend over the whole of it so as to enclose it from end to end, making many turns and knots, and they finally end with the head of one of them, on one end of the building, joined with the tail of the other, and the same thing happens on the other end of the building. There are also many figures of men and idols, other figures of monkeys, and of skulls and different kinds of shields — all carved in stone. There are also over the doors of the rooms some statues of stone with maces or sticks in their hands, as if they were mace- bearers, and there are bodies of naked Indians with their masteles (which are the old-fashioned loin-clothes of all New Spain, like breeches), by which it is shown that these buildings were built by Indians. In this building are seven doors, 2 of which six open on the patio and the seventh on the end which faces the north, and inside of each door are two rooms, so that there are fourteen rooms in all, arched like the others. “Besides these four buildings there is on the south of them distant from them about an arquebus shot, another very large building [House of the Gov¬ ernor] built on a mul or hill made by hand, with abundance of buttresses on the corners, made of massive carved stones. The ascent of this mul is made with difficulty, since the staircase by which the ascent is made is now almost destroyed. The building, which is raised on this mul, is of extraordinary sumptuousness and grandeur, and, like the others, very fine and beautiful. It has on its front, which faces the east, many figures and bodies of men and of shields and of forms like the eagles which are found on the arms of the Mexicans, as well as of cer¬ tain characters and letters which the Maya Indians used in old times — all carved with so great dexterity as surely to excite admiration. The other fagade, which faces the west, showed the same carving, although more than half the carved part had fallen. The ends stood firm and whole with their four corners much carved in the round, like those of the other building below. There are in this building fifteen doors, of which eleven face the east, two the west 3 and one each face the north and south, and within these doors there are twenty- four rooms arched like the others. Two of these rooms are in the northern end, and two others in the southern end, while two are in the west front, and all the rest in the eastern front — all made with special accuracy and skill. “The Indians do not know surely who built these buildings nor when they were built, though some of them did their best in trying to explain the matter, but in doing so showed foolish fancies and dreams, and nothing fitted into the facts or was satisfactory. The truth is that to-day the place is called Uxmal, and an intelligent old Indian declared to the father delegate that, according to what the ancients had said, it was known that it was more than nine hundred 1 According to the plans of Mr. Holmes this building has twelve doors which open on the plaza and one door at each end, making fourteen in all. All the rooms are double. 2 According to Holmes all seven doors open on the court. 3 The small chambers under the great arches of this building must have been counted on both the eastern and western face. The central chamber of the eastern front has three doors; apparently only one of these was counted. 8 MAYA ART. years since the buildings were built. Very beautiful and strong they must have been in their time, and it is well known from this that many people worked to build them, as it is clear that the buildings were occupied, and that all about them was a great population, since this is now evident from the ruins and remains of many other buildings, which are seen from afar; but the father delegate did not go to these ruins, since the thicket was very close and dense, and there was no opportunity to open and clear out a path so as to reach them. And now they all serve only as dwellings and nests for bats and swallows and other birds, whose droppings fill the rooms with an odor more disgusting than delightful. There is no well there, and the farmers of the vicinity carry their drinking water from some little pools of rain-water which there are in that region. It may be easily suspected that these buildings were depopulated for want of water, al¬ though others say that this is not so, but that the inhabitants departed for an¬ other country, leaving the wells which were there choked up.” Similar notices of the evident antiquity of Tiho 1 (Merida), Izamal, 2 and other cities might be quoted. The complete silence in regard to other important centers of northern Yucatan, such as Labna and Kabah, tells the same story of desertion and desolation. 3 Cortes, during his wonderful march from Vera Cruz to Honduras, seems to have found none of the stone-built cities of the Usumacinta region inhabited. The identifying of the village of Teutiercar with Palenque is surely incomplete. “This village,” says Cortes, 4 “is very pretty, and is called Teutiercar by the natives. There are in it very handsome mosques or idol-houses, where we took up our abode, casting out their gods, at which the natives showed no great dis¬ content . . .” There is no reason to suppose that the idol-houses were built of stone. Indeed, in speaking of a near-by village of equal importance, he writes: 5 “Cagoatespan was entirely burnt down, even to the mosques and idol-houses.” Many passages indicate that the idols of this region were carved of wood and not of stone. When more permanent structures are suggested, there are no modern remains to test conclusions. Bernal Dias 6 thus describes a town on Lake Peten: “We proceeded towards a place named Tayasal, situated on an island, the white temples, turrets and houses of which glistened from a distance.” Although this town was a capital of a province, no noteworthy remains are found on its site. 7 At the end of the journey, had Quirigua still been the center of such wealth and power as its monuments bear witness to, Cortes would prob¬ ably have found food there and would not have been forced to ascend to the highlands. As Maudslay 8 points out, the praise that Cortes bestows on the town of Chacujal, where the present remains are of the most meager sort, is a pretty sure indication that he visited none of the really great cities. Yet, had these great cities still been maintained, he could hardly have missed them all. Copan was visited in 1576 by Diego Garcia de Palacio, 9 who saw there “ruins and vestiges of a great population and of superb edifices, of such skill and splen¬ dor that it appears that they could never have been built by the natives of that province.” The natives informed him “that in ancient times there came from 1 Bienvenida, 1877, p. 71. * Dias del Castillo, 1803, pp. 117 et seq. * Lizana, pp. 3 et seq., and Landa, 1864, p. 32. 7 Maler, 1910, p. 169. For an account of the de- * Charnay, 1885, p. 329, is evidently in error in struction of Tayasal in 1697 see Villagutierre, 1701, thinking Landa refers to these cities as recently pp. 481-483. abandoned. 8 1889-1902, II, p. 29. 4 1868, p. 36. See also, Seler, 1895, c, p. 22. 9 Gordon, 1896, pp. 47-48. 8 Cortes, 1868, p. 25. INTRODUCTION. 9 Yucatan a great Lord, who built these edifices, but at the end of some years returned to his native country, leaving them entirely deserted.” Maler 1 is doubtless right in identifying Yaxchilan with the ruined city dis¬ covered by Alzayaga during the Lacandone wars. It is described 2 thus: “They arrived at a place, where it was plain that there must have been once a very ancient city, owing to the great number of stone foundation-walls, and enor¬ mous ancient ruins of edifices which they found; which city must have meas¬ ured more than a league in circumference.” The fact that the wild and untaught Lacandone Indians to this day bring offerings of copal to the old ruined temples of Yaxchilan is worthy of note in weighing the evidence above quoted in regard to Uxmal. Regarding Comalcalco, perhaps the westernmost Maya city of importance, the ruins of which lie on the right bank of the Rio Seco about forty miles west of Frontera, there is also credible evidence of desolation. Charnay 3 attempts to identify this city with the historical Cintla, where Cortes fought his first great battle, but all his arguments are signally refuted by the independent re¬ searches of Rovirosa 1 and Brinton. 6 At the coming of the Spaniards the in¬ habitants of this portion of Tabasco spoke a Maya dialect and probably belonged to the Tzendal tribe. But they had evidently fallen away from the high culture of theii^ncestors. Upon the highlands of Guatemala and southern Mexico certain large towns are known to have been occupied at the time of the conquest. Alvarado,” in a dispatch to Cortes, describes Utatlan. In this description the Spanish captain pays special attention to fortifications and leaves the bare impression that the town consisted of inflammable buildings crowded together. The ruins found on the site of this old town show small mounds, one of which in Stephens’ 7 time still retained part of its stone facing and traces of frescos. There is nothing here, however, fit to be compared with the monumental remains in the low¬ lands. Indeed, it appears that at no period, historic or prehistoric, did archi¬ tecture on the plateau reach a high stage of development. But the ceramic and other remains 8 of minor arts prove undoubtable connection at some time with the lowland civilization. Since, however, this open plateau lies upon the frontier of the Maya area and upon the main road for migration north and south, it is but natural that it should be the first to feel the effects of an ascendant neighboring culture. It will be shown that influence from the Nahua cities to the north was marked. The above and similar notices from the accounts of the first European ob¬ servers, referring to the various parts of the Maya area, make it pretty evident that when white men set foot on the shores of Mexico the golden age of Maya civilization had long since passed. Not a single great city was maintained in its ancient splendor. It is equally evident that certain phases of the ancient cul¬ ture, such as referred to religious ideas, were still kept up and that the art of writing and recording time were still understood at least by a portion of the people. The decadent culture was surely a survival of the higher and earlier 1 1903, pp. 106-108. - Villagutierre, 1701, p. 362. 3 1885, pp. 163-177. 4 1897, pp. 16 et seq. 6 1896, pp. 262-264. 6 1838, p. 112. 7 1841, II, p. 184; Maudslay, 1889-1902, II, pp. 30-38. 8 Bulletin 28, pp. 77-121 and 639-670. 10 MAYA ART. one in the same area, and the Maya of historical times were the descendants of the builders of the monuments. Native Accounts. The conclusions stated in the preceding paragraph are borne out by certain native literary material. This native material is of two kinds, pre-Columbian and post-Columbian; the one written in hieroglyphs and the other in European script but with Maya words. As is well known, the Indians of Mexico and Central America possessed a compound system of ideographic and phonetic writing and were on the very threshold of the alphabet. The Nahua hieroglyphs of personal and place names are readily solved; first, because the glyphs are so strongly pictographic that the component parts may be recognized; secondly, because the method of writing was maintained after the conquest and in part mastered and described by the Spanish priests. The Mexican place name glyphs or cartouches show an elab¬ oration of the rebus method in which advantage was taken of the position, color and all the possibilities of punning pictures. The system of writing in vogue among the Maya was probably the same, but with a greater degree of conventionalization. Only three pre-Columbian Maya books or illuminated manuscripts are known to exist. They are known by the names of, 1st, the Dresden Codex; 1 2d, the Tro-Cortesianus; 2 3d, the Peresianus. 3 These treat subjects much more complex than many of the Nahua codices and afford no easy beginnings for their elucidation. Thus far it has only been possible to work out their meanings in a general way, except where numbers are concerned. They treat of the calendar and of associated religious ceremonies. 4 * Among the books destroyed by the zealous Spanish priests there are said to have been some on civil and religious history and some on rites, magic and medicine. They seem to have been held in great veneration, and in all proba¬ bility had either been handed down from former times or else carefully copied from earlier originals. The manuscripts were capable of withstanding wear and tear, being written on both sides of strips of prepared deerskin or stout paper of maguey fiber sized with fine lime. 6 These strips were folded screenwise between boards. Although most of these invaluable records had been lost, yet educated na¬ tives attempted to save something from the wreck of ancient culture by writing down in European script certain digests of chronicles. These make up the so- called Books of Chilan Balam. 6 The different redactions from different towns vary in details, but all agree in carrying back Maya history many hundred years. Two Maya tribes from the highlands of Guatemala have preserved some¬ what similar ancient accounts. Both in the Annals of the Cakchiquels 7 and in 1 Forstemann, 1880 and 1892. 1 Codex Troano published by Brasseur de Bourbourg, 1869-70, and the Codex Cortesianus by Rady y Delgado in 1892. For convenience the two parts are usually put together and the numbering of the pages made consecutive as is shown in the Table of Nomenclature at the end of this volume. 3 Codex Peresianus reproduced by L6on de Rosny, 1887 and 1888. The pagination was probably as shown in the Nomenclature but references are made to the plates as numbered by de Rosny. 4 By far the most important single contribution to the study of the codices is Forstemann’s Com¬ mentary on the Dresden Codex. 6 Important early references to codices are Peter Martyr in Brasseur de Bourbourg, 1869-70, I, pp. 2-3; Aguilar, 1639, p. 88; Alonzo Ponce, Relacion breve, LVIII, p. 392; Landa, 1864, p. 44; Villaguticrre, 1701, pp. 393-394. 8 Brinton, 1882, b and d. 7 Brinton, 1885, a. INTRODUCTION. 11 the Popol Vuh 1 of the Quiche there is a mythological preamble identical in regard to certain place names (Zuiva, Nonoual, etc.) with that which intro¬ duces the definite historical sequence in the Books of Chilan Balam. Without doubt a careful study of these three accounts — considerable portions of which are still unpublished — will make possible a valuable outline of the ancient his¬ tory of the Maya. Political and Religious Ideas. The Spaniards found the Maya-speaking people divided into many small tribes, each independent of the others and under the direction of its hereditary chief. About twenty such tribes are recorded for the peninsula of Yucatan alone. 2 There was an organized priesthood and a well-marked nobility with strict regard for descent. Probably the priesthood and the nobility were more or less closely joined. Nepotism was apparently the prevailing system under which the chiefs assigned secondary political offices such as that of headman of a village. Practically nothing is known regarding the qualifications of the priests or their divisions into classes. It is clear, how¬ ever, that the priesthood and the nobility held a monopoly of learning. It seems necessary to postulate for the period of national greatness a much more centralized form of government than existed at the time of the conquest in order to account for the magnitude and splendor of the temples and public buildings. These could have been built only at great expense of wealth and labor and under a highly organized system of superintendence. Tradition, how¬ ever, refers to confederacies and not to a united empire. It seems possible that the Maya, like the Greeks, were religiously and artistically a nation while politically a number of sovereign states. Under the powerful stimulus of a re¬ ligious and artistic awakening of national scope, city after city may have arisen, in influence and wealth. Conquest, colonization, abandonment of old sites and migration to new ones may be inferred from striking similarities in the remains of certain cities. But, whatever the political conditions under which the Maya flourished, there were doubtless intervals of decadent culture due to civil strife. Finally, perhaps a scant century before the coming of Europeans, the entire political fabric fell apart. Little is known concerning the details of Maya religious ideas. A list of divinities is given by Cogolludo 3 and other information added by Lizana 4 and Landa. 5 There seems to have been belief in a supreme deity without form or substance. Outwardly religion was greatly concerned with the plumed serpent, especially in the personification known as Kukulcan. There were, however, many lesser divinities. 6 Some of these were closely connected with the plumed serpent, and seem to have been merely individual or functional expressions of this more generalised godhead. An idea of the symbolical complications which probably prevailed throughout Maya religion may be gained from the Popol Vuh, the cosmogonic myth of the QuichA The ceremonials seem to have been characterized by pageants and proces¬ sions, by incense burning, and to some extent by human sacrifice. It is clear that human sacrifice never reached among the Maya the horrible extreme that it held among the Nahua in Mexico City. For incense, both rubber and copal 1 Brasseur de Bourbourg, 1861. 4 1893, pp. 4-5. 2 Brinton, 1882, b, pp. 25-26. 5 1864, pp. 144-168; 206-232, etc. 3 1680, pp. 196-198. * Schellhas, 1904, and Brinton, 1894, b, pp. 37-68. 12 MAYA ART. gum, the latter burning with a cloud of white smoke and a pleasant perfume, were used. This feature of incense burning, coupled with the prescribed making of the pottery burners, was purely ritualistic. Consequently it was practiced by the mass of the people and has survived to this day, while the complex the¬ ology died with the priests and nobles. Thus Dr. Tozzer 1 has been able to con¬ nect the chief ceremonies of the pagan Lacandone Indians of the present day with the yearly renewal of the incense burners, as described by Landa in the sixteenth century. The insufficient direct knowledge of gods and ceremonies has been pieced out by the study of the ancient codices, and of the sculptured representations on stelae and on temple walls. Apparently astronomy, the understanding of which made possible the calendar, was of first importance. Planets and stars, as well as the sun and moon, were represented by divinities. The forces of na¬ ture, such as the rain, the wind, and fertility in its various forms, were conceived as individual or as variant gods. That warfare had its strong religious aspects is seen in the prevalence of bound captives in the sculptured groups and in the use of spears and shields in the ceremonial regalia. Many particulars regarding sacrifice are also to be gathered from these sources. These particulars support the conclusion that human sacrifice played but a minor role in the religious practice of the Maya. Present Population. The present population of the Maya area is largely made up of Indians of the original stock, showing no great amount of race mix¬ ture. In many regions tribal distinctions are still clear. The range of culture is remarkable. In northern Yucatan the Maya have long been civilized and under the sway of the Catholic church. Even here, however, they still use their native language in almost entire purity, while a careful observer can detect in the modern religious rites many remnants of ancient custom and superstition. There is no chance that an understanding of the ancient hieroglyphs now exists among any of the Indians here or elsewhere in the area. Along the southern coast of Yucatan, in the territory of Quintana Roo, some of the tribes are at present independent of the Mexican rule. The ruins of Tuloom and others in this region are practically closed to investigation. These wild Indians and the tribes of British Honduras and the Peten department of Guatemala show only a moderate degree of culture. 2 The status of the Lacandone Indians 3 of the Usumacinta Valley is lower yet. Scattered thinly in family groups, these people have indeed reverted to the wild. Although their religion is now of the primi¬ tive spiritual guardian type, the ritual still preserves features that point upward to the past, as also does the making of pottery and cloth. On the highlands the Quiche and other tribes live in agricultural communities and possess an interesting decorative art, making excellent textiles. This art seems to be quite distinct from the ancient Maya art. The estimates of numbers by the Spanish historians were doubtless exces¬ sive, but the country is capable of supporting a large population. The Maya¬ speaking tribes number to-day several hundred thousand. Large tracts of ter¬ ritory that show abundant remains of habitation are now entirely deserted. Materials Available for Study. Remains of Maya art are for some branches of the subject quite extensive, while for others they are wofully lacking. Tak- 1 1907, pp. 106 et seq. 3 Sapper, 1895, b. 3 Tozzer, 1907. INTRODUCTION. 13 ing all in all, however, there is no reason to complain, because the mass of ma¬ terial preserved for study is probably greater than that which has survived from the great art of Greece. The remains may be considered under two heads: first, architecture; second, minor arts. The entire Maya area is dotted with groups of structures, great and small, some admirably preserved, others ruined beyond repair. Some of these struc¬ tures were temples, while others may have been for secular use. Probably more domestic architecture has all passed away. Some of the complicated structures may have been chiefs’ palaces, but it is more probable that they re¬ sembled monasteries. The church and state were one. These structures still show much of the original embellishment in stone carving, wood carving, fres¬ cos and stucco work. The minor arts include ceramics, textiles — most of the data on which must be taken second hand from the sculptures — ornaments carved in semi-precious stones, a little metal work, and, most noteworthy of all, the ancient illuminated manuscripts. Previous Studies. Recent study in the field of Maya culture has been di¬ rected mostly towards the elucidation of the codices and the decipherment of the hieroglyphic inscriptions. As a result of the labors of Bowditch, Goodman and Thomas, the calendar system has been worked out in many of its finer details, while Forstemann, Seler and Schellhas have collected much data upon the nature of the gods and the ceremonies. The facts brought out by these investigators are of great value to the student of art, because they furnish a basis for the chronological sequence of forms and for the interpretation of de¬ signs and sculptures. Descriptions of the buildings and other monuments may be gleaned from early and modern writings. The scanty notices of the Spanish conquerors have already been considered. The first travelers to draw the attention of the world to the wonderful structures of Central America were Stephens and Catherwood. The detailed accounts of the former and the accurate drawings of the latter are still of the greatest service to the student. The drawings of Waldeck are beautiful but inaccurate. The voluminous writings of Brasseur de Bourbourg contain many valuable references, but most of the theories and conclusions are untenable. The same may be said of the works of Le Plongeon. The era of enthusiastic travelers was followed by that of trained observers. Holmes, 1 in particular, has explained the process of Maya construction, and prepared ad¬ mirable panoramic views of Chichen Itza, Uxmal and Palenque. For overshad¬ owing importance, however, first place must be given to Maudslay’s 2 elaborate publication. In the four volumes of plates are figured, both by photographic reproduction and by clear drawings, the most important sculptures and buildings of Copan, Quirigua, Tikal, Yaxchilan, Palenque and Chichen Itza. The text, however, gives little more than brief descriptions of the monuments, with hardly any stylistic comparison, the author apparently being content to let his splendid illustrations speak for themselves. With these deserve to be mentioned the accounts of explorations at Copan and along the Uloa River by Gordon, and explorations in Yucatan and in the Valley of the Usumacinta by Maler, as well as the more popular narratives of Charnay. All of the latter works deal only ob- 1 1895-1897. ’ 1889-1902. 14 MAYA ART. jectively with Maya art, and are in reality hardly more than storehouses of selected material. Unselected material of equal value may be found in the col¬ lections of photographs and maps which constitute the field reports of various expeditions of the Peabody Museum. Such reports include much unpublished material, as, for instance, Thompson’s explorations of Labna, Kabah and other sites in northern Yucatan. Miss Breton’s reproduction 1 of the frescos of Chichen Itza in color is of the greatest value, because it preserves a splendid example of a kind of perishable art that has survived in few places, and that is peculiar in giving intimate glimpses of the ordinary life of the people. It is to be hoped that this excellent work may soon be published to the world. Among institutions who have supported field work in the Maya area, first place must be accorded to the Peabody Museum which has sent out many expeditions both to explore and to excavate. The results are seen in the splendid collections in this museum and in the many publications by Thomp¬ son, Gordon, Maler and Tozzer, mostly appearing in the Memoirs. Dr. Seler 2 conducted the only systematic field work that has taken place on the highlands of Guatemala. Excavations were also made by Maudslay, Dieseldorff and others. Special notice must be given Dr. Gordon’s paper on the Serpent Motive in the Ancient Art of Central America and Mexico, 3 because this is the only attempt at a general consideration that has been made. The modern ethnology of Maya-speaking tribes has been covered by Stoll, 4 Starr, 3 Tozzer, 6 and Sapper. 7 In general, however, the survivals of the ancient art are apparently slight, and little has been done in collecting myths. Maya art is on a much higher scale than any art in America except possibly the textile art of Peru. It deserves earnest study for the contributions which it is able to make to comparative religion and to comparative art. 1 A complete reproduction of these paintings is on exhibition in the Peabody Museum. 1 1901, c. s 1905. 4 1889. 6 1900-1904. 6 1907. 7 1905. I. GENERAL CONSIDERATION OF MAYA ART The influence of a national religion upon a national art was never more unmistakable than in the case of the Maya. But, indeed, it is universally im¬ portant. Religion is able to furnish the deepest and truest inspiration which the human mind is capable of receiving. Being ideal in itself, it develops the imagination so that this in turn finds secret meanings in common things. More¬ over religion, as a communal element in the life of the nation, turns the atten¬ tion of all artists to a common purpose. Through this focusing of the attention religion leads inevitably to an intensive rather than a diffuse development of art. But once this intensive development has exhausted the possibilities of the established ideas, then religion throws its powerful influence against further disorganizing change. Thus religion enriches art and makes it permanent. In the case of the Maya the art might almost be termed the concrete expres¬ sion of the religion, since all the great monuments were apparently connected with religious practices and no minor object was too humble to receive decora¬ tions with religious significance. Clearly this wonderful art rose under the com¬ munal inspiration of a great religious awakening and was conserved by the per¬ sistence of ritual. Doubtless the art reacted strongly upon the religion which gave it birth, filling that religion with symbolism and imagery. The two worked hand in hand. The spreading of the religion meant a spreading of the art, and the graphic representations of the art rendered the religion intelligible. It was probably through the objective ritual on the one hand and the objective art on the other, that the religion of the Maya was enabled to leap the bounds of lan¬ guage and impress itself so strongly upon the Nahua and Zapotecan peoples. The student finds in the ancient masterpieces of Yucatan and Central Amer¬ ica a fine technique and an admirable artistic sense largely given over to the ex¬ pression of barbarous religious concepts. Upon the scale of development the art is many points higher than the religion, in spite of the close connection be¬ tween them. At first glance too exotic and unique to be compared with the art of the Old World, nevertheless Maya art furnishes upon examination many anal¬ ogies to the early products of the classic Mediterranean lands. Indeed upon technological grounds — such as the knowledge displayed of foreshortening, composition and design — Maya art may be placed in advance of the art of Assyria and Egypt and only below that of Greece in the list of great national achievements. The representation of the body of man himself was not all-important to the Maya as to the Greeks, for a good and sufficient reason, although it received a very considerable share of attention. The Greeks conceived and represented their divinites and mythical heroes in human form. Hence they idealized this form till it embodied the finest possible conception of strength and grace. Now the gods and culture heroes of the Maya had fundamentally the physical char- 16 MAYA ART. acteristica of reptiles, birds and lower mammals, or were, at best, grotesque figures of composite origin. However, these brute gods, as we shall see, were often more or less humanized, resembling in a general way the half-animal, half-human gods of Egypt and Assyria. Human beings appear only in the mundane guise of priests, worshipers, rulers, warriors, and captives. The strange subject matter of Maya art should not militate against its real artistic merits, for the finest products of an inspired imagination are always worthy of respectful study. The principal methods employed by the Maya in the graphic and plastic arts differed little from those of classic lands. Delineation and painting upon a variety of substances including paper and plaster, carving in wood and stone and modeling in clay and stucco were widely practiced. Terra cotta figurines made from moulds are very common. Metal working was highly developed as far as the technical processes are concerned, but the scarcity of materials was such that only ornaments were commonly made. The stones used in the tem¬ ples and monuments were cut and carved with stone implements. The Maya might have accomplished greater wonders if they had had fine-grained marble instead of coarse and uneven limestone, and iron or bronze chisels instead of stone knives. In Maya plastic art the three usual divisions may be made; namely, low relief, high relief and full round. Much of the high relief, however, shows no more modeling than does the low relief, the figures being simply blocked out in high relief, but still retaining a comparatively flat outer surface. Sometimes high relief shows flat sculpture upon two or more planes. Fine examples of blocked- out high relief of these two sorts are found at Copan and Yaxchilan. But high relief with excellent modeling also occurs, particularly in the stucco work of Palenque. Sculpture in the full round reaches its highest development at Copan, probably because the stone found there was very easy to work. There may have been another reason. The habit of representing faces and bodies in front view seems to lead directly to the full round treatment, especially of the face. Profile figures, on the other hand, appear best in low relief. At Copan the majority of figures are presented in front view, and there is a steady progression from low relief, through high relief, to the exact reproduction of the human body. At Quirigua and Piedras Negras, where other front-view figures occur, the greater part of the body is shown in low relief, but the face is generally carved in high relief or in the natural roundness. The full round method of representation is also accorded to figures seated in a niche that occur at the two cities just referred to. But low relief is by far the most common mode of sculpture in wood, stone and stucco, and may be studied to advantage in all the principal Maya cities. Homogeneity of Maya Art. The homogeneity of Maya art, in spite of the many necessary differences due to time and place, will prove itself as the de¬ scription proceeds. In the following pages the aim will be to give a general ex¬ planation of the most widespread phenomena. Illustrations on particular points will be taken from all parts of the area, and from both major and minor arts. For instance, the pottery decoration of the uplands will show features similar to the architectural decoration of the lowlands, or the drawings of gods in the codices will agree with stone and stucco figures on monoliths and temple walls. Simply by way of illustration two or three series of particular similarities GENERAL CONSIDERATION OF MAYA ART. 17 that cover nearly the entire Maya area may be given in detail. Fig. 1 presents a number of faces of diverse forms. Most are strikingly grotesque, and all pos¬ sess the curious feature of a cruller-like ornament over the nose. This ornament is adventitious and unnatural, and, although apparently insignificant, yet it furnishes the strongest kind of proof of cultural unity, because it is in the nature Fiq. 1. — Heads with curled nose ornament: a, Palenque; b, Copan; c, Quen Santo; d, Labna; e, Coban; f, Panzamala. of an unconscious admission. The first specimen, a, is a representation of the so-called sun shield on the tablet of the Temple of the Sun at Palenque; 6, is a somewhat similar design from Copan. It is carved upon a block of stone with a tenon at the back so that it could be set into a temple wall. Shields having decorative faces of the same general type as these are represented as worn on the left arms of many of the warlike figures on the monuments. Noteworthy examples occur at Tikal, lintel of Temple II, 1 Naranjo, Stela 21/ and Yaxchilan, Stela 11. 1 The same device appears on faces in other situations, that may or 1 Maudslay, 1889-1902, III, pi. 73. * Maler, 1908, b, pi. 35. * Maler, 1903, pi. 74. of Mr. E. P. Diesel- dorff of Coban, 1 2 with applied ornament in the form of an excel¬ lently modeled face. This vase was exca¬ vated from ruins near Coban. The last spe¬ cimen, /, comes from Panzamala, 3 some distance east of Coban. Other ex¬ amples of this pecul¬ iar feature might be given. 3 * Most, if not all the faces which have been described probably represent some form of the Sun God. This point will be discussed later. Other examples of widespread similarities may be given to include certain important cities where the face with the twisted nose ornament is not known to occur. 1 he representation of plants is rarely seen in Maya art, except in a very peculiar motive that Maudslay 6 has worked out rather fully and which he calls the Water-plant motive. This water plant as used in decorative bands is shown in Fig. 2, a and b, while c offers a more realistic presentation from the so-called Madrid Stela. 6 The design frequently occurs on pottery. The flower has somewhat the appearance of the water lily, and in many instances a fish is shown seemingly in the act of feeding upon the petals. Examples of the fish and water-plant design present much stronger proof of culture affinity among the cities where they occur than do the simple water-plant forms, for designs analogous to the latter are universal, whereas the association of fish and flower is very unusual. In Fig. 3 are given examples, taken from different parts of the Maya area, that illustrate this peculiar motive. It occurs in full vigor at Copan, Palenque and Chichen Itza, as well as at many other sites, both on the highlands and lowlands. Its exact meaning is somewhat difficult to determine, but it apparently carries the idea of water. It is attached as an ornamental detail to the bodies of animals and to the heads of divinities that are probably asso- 1 Seler, 1901, c, p. 112. 2 Seler, p. 1901, c, p. 178. 3 Seler, 1901, c, p. 179. * For instance, the feature seems to occur on a head from Bellote on the coast of Tabasco, Char- nay, 1885, p. 162, and on a vase from San Salvador, Seler, 1901, c, pp. 180-181. 6 1889-1902, IV, pp. 37-38 and pi. 93. 6 L6on de Rosny, 1882, pi. 2. GENERAL CONSIDERATION OF MAYA ART. 19 Fia. 3. — Plant and fish motive: a and i, Copan; b, e, / and h, Palenque; c, Chajcar; d and j, Chichen Itza; g, Ixkun; k, Nebaj. ciated with water. Fig. 4, from the Dresden Codex, represents one of the Maya gods — known as God B or the Long-nosed God — wading into the water and Dresden Codex. d, Chichen Itza. 20 MAYA ART. pulling up a water plant. Fish and shells are shown in the water, the lower depths of which are colored green. The planets and other astronomical bodies are generally represented by sim¬ ple oblong hieroglyphs arranged in strips or bands. These signs occur as details of ornamentation on the dress of human figures, particularly on the belts, as Fia. 6. — Astronomical signs combined with bird and animal heads: a, Quirigua; b, Copan; c, d, and /, Codex Peresianus; e, Dresden Codex; g, Naranjo. markings upon various ceremonial objects and upon the elongated bodies of monstrous creatures, as general motives of architectural enrichment, as well as in connection with many gods and with intricate astronomical calculations. The forms written in the codices are similar to those carved upon the monu¬ ments, but show a more cursive delineation. The general similarity between the bands of astronomical signs in various parts of the Maya area is brought out by the examples given in Figs. 5 and 6. Of these Fig. 5, a, is from the front of the annex to the House of the Magician GENERAL CONSIDERATION OF MAYA ART. 21 at Uxmal, b and c are pottery fragments from the neighborhood of Coban on the highlands of Guatemala, and c is from the eastern fajade of the Monjas at Chichen Itza. Fig. 6 presents a second series showing the astronomical signs combined with birds’ heads and upon seats or thrones across which are bound grotesque ani¬ mals. A very close parallel is evidenced in a and c, the first from the back of a monolith at Quirigua, the second from the Peresianus Codex. In neither of these speci¬ mens is the head very distinct. Often bands of astro¬ nomical signs are terminated by birds’ heads, as may be seen from b on the back of Stela H at Copan, e in the Dresden Codex and g, the base of the splendidly carved Stela 32 at Naranjo. The latter sculpture really repre¬ sents an elaborate throne, but the human figure seated upon it is so badly mutilated that it was not reproduced in the drawing. Across the top of this throne is a gro¬ tesque animal trussed and bound. The same sort of thing is represented by drawings in the Peresianus Codex (d and /), likewise connected with astronomical signs. This grotesque animal will be discussed more fully in a later section, where many other objects combined with star symbols will come up for consideration. The inter¬ pretation of the particular signs will also be postponed. As stated, the only reason for giving the preceding examples at this time is simply to remove all possible doubt concerning the homogeneity of the ancient art within the limits ascribed to the Maya culture. This is seen in important things as well as in the relatively unimportant ones just given. The latter, however, have the unusual quality of covering not only all the geographical range but most of the chronological range as well. The Human Fohm An easy understanding of Maya art may be had by starting with subjects least opposed to the familiar ones of the Old World and proceeding thence into the labyrinth of fantastic conceptions peculiar to the New World. Subjects Represented. The representation of the human figure seldom served as an end in itself. The men and women shown in the sculptures are seemingly engaged in religious ceremonies and acts of adoration. The divinities which they worshiped are more or less clearly indicated. As a rule, the human figures are those of priests or warriors. But even in the case of the latter the religious motive is rarely absent and human beings free from the artistic domination of gods of lower nature. There are, however, a number of sculp¬ tures which apparently memorialize success in war, and in these none of the usual religious paraphernalia appears. An example of adoration is seen in Fig. 7 that reproduces part of the beau¬ tifully carved but badly broken Stela 7 at Yaxchilan. This kneeling worshiper 22 MAYA ART. wears a loose cloak over the shoulders, a light garment that covers the thighs and a rather elaborate headdress in the form of a somewhat grotesque animal head. His face is turned upward and his hands are lifted to receive whatever -— - . the being above may care to bestow. From a pair of hands above him, that belong to a stand- *^^ ing figure so badly shattered that it cannot be /X / l reproduced, descend certain objects that may in- ^ dicate an answering of his prayer. Immediately ^ above the face of the kneeling figure is a sign resembling that for corn or maize, lower down is $ the symbol of the sun, while the scroll-work of Z' jn small circles may represent rain or, perhaps, fruit- D fulness. This tableau, then, might be considered III to represent a prayer for the corn crop and the sun and rain most necessary for its welfare. Such J\ an interpretation would be in keeping with the wfj principles of sympathetic magic according to which the thing desired is usually represented either / J pictographically or dramatically, i / Very often a human figure holds in his arms \( T or hands one or more ceremonial objects. In the \\ next drawing (Fig. 8) a man in simple attire stands [Mi upon the back of a grotesque being and supports ft \ \\ upon his uplifted palms a manikin. He is one of two 1H ll\ I) priest-like figures sculptured on the justly famous \ // tablet of the Temple of the Sun at Palenque. 1 \ // Many other examples of human beings engaged in religious services will be given in other places. <^ 2 ® /%> Many sculptures show human beings seated upon —• . f I thrones before which are ^ ^ /1 standing or kneeling wor- / Fia. 8. — A presiding priest: Palenque. shipers. These Seated per- | sons may represent rulers J |8|pP®|u ... or high priests who were worshiped as the embodi- ments of gods. The divinities themselves were of a low animal order. But the government as well as the re- ligion of the Maya was probably of totemic origin, and jJ \ so might be expected to emphasize close relationships f* between the temporal and spiritual rulers and the animal " gods. These seated figures frequently hold the same Fl maskTYaxchTian nng ceremonial objects as the standing ones, while in other cases the ceremonial objects are held aloft by attendants. The human overlord as the agent of the god is clearly shown on Stela 11 at Yaxchilan. 2 On this monument is a standing human figure wearing the mask of a grotesque god (see Fig. 9 for the mask) who threatens with his baton several bound captives kneeling before him. 1 Maudslay, 1889-1902, IV, pi. 88. 2 Maler, 1903, pi. 74. GENERAL CONSIDERATION OF MAYA ART. 23 A sculpture that seems to refer only to war and conquest is reproduced in Fig. 10. This is Lintel 12 at Yaxchilan. In the center is a chief with spear and shield and in full regalia. The head of a slain enemy hangs hair down from his breast, and cross bones decorate his dress. At the left is one of his assistants, likewise armed. Kneeling on the ground are four captives bound with rope. Upon the bodies of these captives are glyphs which may record their names and the date of their capture. At the upper part of the stone are two bands of glyphs, left blank in the drawing, which possibly contain the narrative of the 24 MAYA ART. and of a somewhat Hebraic cast. The lower lip is protruding and pendulous and the mouth kept slightly open. This type is not characteristic of the modern Maya Indians, who, however, do not retain the ancient practice of deformation. Hamy 1 compares the faces carved at Palenque with those of two Mexican mi- crocephalic freaks on exhibition in Europe about 1850. The striking resemblance he notes seems to be purely the result of chance. Besides cranial deformation other modifications of the natural form were in vogue. Filed teeth and teeth inlaid with jade and other minerals have been found in a number of burial sites, as well as cumbersome nose and ear plugs. In regard to evidence of style which may be associated with particular sculp¬ tors, it is by no means lacking. There is a widely advocated theory that prim¬ itive art is purely communal. To be sure, the first artists did not ordinarily sign their works, but strict regard for ownership of designs or of songs is no rare thing among primitive people. What reason is there to hold that artistic genius among the Maya was not essentially the same as in our own land, simply because the social organization of the nation and the subject matter furnished by the religion are different? Real contributions to human culture are always referable to individuals, and the fact that the records are lost matters not. But the individual lives and works within the mode of his nation and his epoch. He adds something of his own to art, theology or what not, and that something is more noticeable in present view than in retrospect. Upon analysis that some¬ thing frequently resolves itself into a new imaginative reconciliation of pre¬ viously known elements. Rarely, indeed, the individual may strike back to real origins and make a radical departure from traditional habits of thought and expression. The works of the most flagrant individualist of to-day will to-mor¬ row fall into an inevitable scheme of evolution. It is reasonable to suppose that each of the various groups into which the stelae of Copan may be divided was the work of a single sculptor, or of a school under the direction of a single artist. Each group shows, as we shall see, a conscious and typical arrangement of common elements. But through all the groups runs a thread of change and development of which the artists themselves may well have been entirely uncon¬ scious, except in its more obvious features. Poses and Groupings. The poses as exemplified on the monolithic monu¬ ments, commonly called stelae, most of which present a single human figure, or a single figure on the front and another on the back, will be taken up first. Then the more unusual poses and the complex groupings on the monuments and elsewhere will be briefly considered. In the case of the Copan stelae the pose is practically uniform throughout the long series. The priest, chief,, or whoever it is that is represented, stands in an erect attitude, with his heels together, and holds an object called the Cere¬ monial Bar against his breast. The body shows perfect bilateral symmetry. Certain changes in the pose, which, in a later section, will be co-ordinated with changes in manner of carving to establish in part the chronological sequence of the monuments, may here be mentioned. In the earliest stelae the upper arms lie close to the side and the forearms rise almost vertically. In the later stelae the forearms are almost horizontal. Again, on the greater number of the monu¬ ments, the feet are represented as turned directly outward, forming a straight 1 1875 . GENERAL CONSIDERATION OF MAYA ART. 25 angle. But at the end of what we will call the Archaic Period, the sculptors began to take advantage of the increased relief, furnished by the heavy apron, to turn the feet inwards, till in the latest examples the pose became almost natu¬ ral. Some of these stelae are reproduced in Plates 18 and 19. On a few of the stelae at Quirigua the feet are likewise set at less than a straight angle. It may be noted that in general the stelae of Quirigua are later than those of Copan, but that they show a reversion to less laborious construc¬ tion. The poses are much the same, though on some of the monuments a man¬ ikin figure on a staff, commonly called the Manikin Scepter, replaces the Cere¬ monial Bar. This substitution breaks up the bilateral symmetry, since the staff is held diagonally across the body and not horizontally. As before stated, low relief practically necessitates the profile view of the face. Many stelae at Tikal, Naranjo, etc. (Plate 21, figs. 1-4; Plate 24, figs. 1-3 and Plate 25, figs. 1-2) show the same pose as those of Copan except that the face is turned in profile. With the feet turned straight out this pose is an awkward one. Often, however, the body appears in profile as well as the face (Plate 22, fig. 1). In such cases the Ceremonial Bar is replaced by other ceremonial objects. Frequently a staff or scepter is held before the face with one hand, while the other holds a decorated pouch at the side. Sometimes spear and shield replace the ceremonial objects. The feet are either one behind the other, as though the person were taking a short step, or else the outer foot covers and conceals the inner one. The bodies are represented as erect and motionless. In a few instances two standing figures are brought face to face with each other, as on Stela 11 at Yaxchilan. 1 More often a warlike figure stands above or beside a bound captive. In other cases where two persons are shown one is seated while the other stands. Seated figures are rather common on stelae in the Peten and Usumacinta cities. The most important type of monument, especially at Piedras Negras, shows a figure in high relief seated cross-legged in a niche with the hands upon the knees. There seems little doubt but that the niche really represents a can¬ opied throne. Such a throne in profile with a seated personage is seen on Stela 5 2 at this city. As a rule, however, the royal thrones, while richly upholstered, do not have canopies. It is worthy of note that on most of the stelae presenting a tableau of several persons, the interest centers in a seated figure. A good ex¬ ample of this is the remarkable Stela 12 3 at Piedras Negras on which a number of individuals, including priests or warriors, and bound captives are arrayed before a seated being that may represent either a chief or a divinity, but more probably the former. As to the disposition of the legs in seated poses, there is considerable variety. Usually they are crossed Turkish fashion and represented either in front or side views. Sometimes only one leg is drawn up, while the other extends downward, the result being a free, graceful pose. Seated figures are also shown with both feet on the ground in the usual attitude. But when this method of sitting is represented in front view the knees are bent outward and the heels raised, as on Stela 2 at Cankuen. 4 Maler, 1903, pi. 74, fig. 2. Maler, 1901, pi. 15, fig. 2. 3 Maler, 1901, pi. 21. 4 Maler, 1908, a, pi. 12, fig. 2. 26 MAYA ART. Poses which show motion are rather rare on the stelae. Of course the mere act of offering is common enough. One monument at San Juan de Motul 1 apparently shows two heavily dressed beings in the act of dancing. The knees are turned outward and the heels are raised so that the figures seem to stand on tiptoe in a somewhat squatting attitude. But it is probable that this peculiar pose is intended to represent sitting rather than dancing, as indicated by the Cankuen stela referred to above and by the principal figure on the lintel of Temple IV at Tikal. 2 The sprink¬ ling of maize or other small grains upon the ground is represented on Stela 13 at Piedras Negras (Plate 25, fig. 3). Various gesticulations are clearly shown on Stela 2 3 of La Mar. Here the poses exhibit unusual freedom. The poses and groupings on sculptured lintels, panels, steps, etc., show so much variety that it seems almost impossible to treat them in a general way. At Copan there are a number of sculptures which show small seated figures arranged in rows. The best examples are Altar Q 4 and the interior step of Temple 11. 6 The pose is slightly different in each case, the variety being more marked upon the last-mentioned monument. Most of the lintels of Yaxchilan show two figures facing each other, one with the body in front view and the face in profile and the other with both body and face in profile. As a rule, one figure is somewhat subordinated to the other. A number of lintels show more than two figures each, as, for instance, Lintel 12, which has already been presented (Fig. 9) as an example of a memorial of conquest. The panels and tablets of Palenque exhibit a great variety of grouping, but no remarkable departures from the poses of other cities. The processional arrangement of human figures so highly developed at Chichen Itza 6 is un-Maya and belongs, as we shall see, to the very latest period of Maya art, when influence from the highlands of Mexico had set in strongly. In these representations there is crude delineation, but often an admirable sense of action. Muscular effort also appears in the strained poses of the small atlantean figures carved on columns and door jambs at this city (Fig. 11). In the codices and frescos action is usually indicated, and the body is rep¬ resented in many positions. Warriors depicted on the walls of the inner cham¬ ber of the Temple of the Jaguars at Chichen Itza are especially active, as may be seen from the one reproduced in Fig. 12. A portion of this fresco represents a spirited battle with much lifelike detail. Part of page 60 of the Dresden Codex Fig. 11. —Atlantean figures on pilasters: Chichen Itza. Fig. 12. — Warrior in fresco paint¬ ings: Chichen Itza. 1 Maler, 1910, pi. 45. 1 Maudslay, 18S9-1902, III, pi. 78. tion see Table of Nomenclature, p. 256. 1 Maler, 1903, pi. 36. 4 Maudslay, 1889-1902, I, pi. 92. 5 Maudslay, 1889-1902, I, pi. 8. 6 Maudslay, 1889-1902, III, pis. 44 et s GENERAL CONSIDERATION OF MAYA ART. 27 Fig. 13. — Example of complicated grouping: Dresden Codex. is reproduced in Fig. 13, from which an idea of the more complicated poses and groupings in this wonderful manuscript may be obtained. Other illustrations of the points so far covered will appear as the discussion proceeds. Foreshortening and Perspective. As may be gathered from the foregoing description of poses and group¬ ings, the Maya had a considerable but by no means complete mas¬ tery of the technical difficulties of representing objects with three dimensions upon a surface with only two. High and low relief form something of a transition for this process. In foreshort¬ ening they greatly excelled the Egyptians and Assyrians, since they became sufficiently skilled to draw the entire body in pure profile, besides representing the legs and feet with ease and pre¬ cision in a variety of sitting and reclining positions. The real difficulty in the development of perspective is that the artist’s previous knowledge of the object interferes with his visual impres¬ sions. He cannot let the hand draw the picture as the eye sees it. He knows that a man possesses two arms and so feels constrained always to draw two arms in plain view. The Maya artists established a sort of compromise between appearance and reality. When they could not find a way to correct the drawing, they at least succeeded by graceful and pleasing treatment in distracting attention from the errors in delineation. The historical develop¬ ment of skill in foreshortening will be demonstrated in another section. Only the more perfect phases will be treated here. The mastery of the pure profile may be studied to advant¬ age at Palenque. A good example from this city is given in Fig. 14, which represents a seated individual with very little clothing to conceal the body. It will be noted that the upper part of the breast is drawn in profile, as well as the head and legs, and that the more distant arm does not appear in the picture. When, however, the body was covered with heavy drapery or elaborate ornaments, the difficulties of foreshortening all the details were sometimes be¬ yond the skill of the artists, especially when the profile pose was adopted. This ineffectiveness is best seen in the braided breast ornaments which seem to pro¬ ject outwards when they should lie flat on the breast. Examples may be seen in the right-hand priestly figure of the tablet of the Foliated Cross at Palenque 1 and in the drawing which Maler 2 gives of part of the incised tablet at Xupa. In all fairness to the Maya sculptors it must be stated that the difficulty with this detail seems to have been overcome on other monuments at Palenque-. 1 Maudslay, 1S89-1902, IV, pi. 81. 1 1903, p. 21. Fig. 14. — Seated fig¬ ure in pure profile: Palenque. i 28 MAYA ART. A change in pose of an exact 90 degrees, that is, from front view to profile or vice versa, was for the most part readily accomplished. Among the compli¬ cating details of the more elaborate dresses worn by human beings were small heads, probably of stone, with appendages of one sort or another. These heads were placed on the breast and on the middle and sides of the girdle. When the body was in front view, the head on the breast and the one on the front of the Fio. 15. —■ Elaborated aprons on the monuments: a, Copan; b, Quirigua; c, Ixkun; d, Tikal; e, Palenque. girdle were likewise drawn in the full face, while those at the sides of the girdle were presented in profile view. When the pose was shifted to the profile, the small heads on the breast and on the front of the girdle were likewise shifted into the profile. One of the heads at the side of the girdle was thrown into the front view, and the other disappeared behind the body. Other objects, such as disk-shaped and bar-shaped breast ornaments, aprons, shields, etc., that appeared entire in front view were ordinarily divided vertically in halves, and only the nearer half drawn when the pose was turned in another direction. The circle seems never to have been foreshortened into an ellipse, but instead was divided vertically as above. The modifications in circular breast ornaments when viewed from the side are easily seen on some of the lintels of Yaxchilan which show both profile and front- view figures with the same style of dress. 1 Shields in half view are worn by the sculptured warriors of the Temple of the Jaguars at Chichen Itza. 2 Fig. 15 presents a series of elaborate aprons of a wide¬ spread type. There is a grotesque face in the middle and a fret at each side. The latter, as we shall see, is really a highly modified serpent head. At the top are usually shell ing of feathers: Tikal. pendants, in groups of three, which project out over the rest of the apron. At the bottom are plumes, braided strips and tassels. Aprons of this sort are occasionally represented on human beings in side view, and in these cases the face in the center becomes converted into a profile face. The adjacent fret is retained without change, and the more dis¬ tant one disappears from view. Fig. 16 represents an object in side view, the lower part being very similar to the aprons we have just examined. This 1 See, for instance, Lintels 2 and 3 (Maler, 1903, 2 Maudslay, 1889-1902, III, pis. 38, 49 and 50. pis. 47 and 48). GENERAL CONSIDERATION OF MAYA ART. 29 object is attached to the headdress and hangs down at the back of a human figure (Plate 22, fig. 1). The fret is readily seen, and the face in profile in front of it can be made out after a little study. Higher up appears the group of three shell pendants. But the feathers that issue from the side of this object are splendidly foreshortened by the use of sweeping curves. The front view of feather projections on similar objects attached to headdresses may be examined for comparison on Lintels 1, 2, 3, 5, 33, etc., 1 at Yaxchilan and Lintel 2 2 at Piedras Negras. The human body is seldom represented in what might be called quarter-view, but there are a few interesting departures from the prevailing front and profile studies that deserve note. One of these is seen in the left-hand person on Lintel 8 at Yaxchilan (Fig. 17). This sculpture pictures two warriors bending over and grasping two partially fallen captives. The attitudes are all excep¬ tionally free, but the figure to which special attention has been called has few parallels in this respect. The face is in profile, but the rest of the body is twisted almost but not quite into front view. One foot is partially raised from the ground. Altogether the drawing shows with considerable success a 1 Maler, 1903, pis. 46-48 and 63. * Maler, 1901, pi. 31. 30 MAYA ART. pose having many difficulties. The carving on this lintel is, however, in such low relief that it appears to be hardly more than an incised sketch. On Stela 13 at Piedras Negras (Plate 25, fig. 3) is carved a superb figure ap¬ parently in the act of scattering grains of some sort. The shoulders in this instance are likewise twisted around, but not to so great an extent as in the drawing we have just examined. Other representa¬ tions of the human body in more or less twisted attitudes are seen on the stucco panels of House D of the Palace at Palenque. 1 The manipulation of drapery does not appear to any great extent in Maya art. As a rule, the dresses of both male and female subjects are covered with stiff and inflexible ornament. Some of the crouch¬ ing women at Yaxchilan 2 have the lower borders of their dresses extended over the ground in a rather picturesque manner. Two simply draped persons are presented in Fig. 18. The folds of the garments are represented by incised lines. Better examples yet may be seen on Stela 1 at Cankuen (Plate 25, fig. 6) and on Stela 7 at Yaxchilan (Fig. 7). But the highest development of drapery in the Maya area occurs at Palenque. One of the most interesting examples is given in Fig. 19. The apron and cloak fall in free and graceful folds, and a sort of twisted scarf hangs down behind the back. Perspective in its application to many objects, such as a crowd or a landscape, the Maya seem scarcely to have considered at all. Figures in dif¬ ferent planes of perspective may have been in¬ tended when they are shown in tiers one above the other. When two or more human beings are placed together, there is usually no unnatural dif¬ ference in size, to express kingly qualities as is so common in Egyptian scenes. On some of the stelae the captives and kneeling figures are clearly supplementary and are crowded into the available space. The small figures of men and animals in¬ terwoven so gracefully into the scroll ornaments of many of the Copan stelae seem to be primarily decorative. The frescos of Chichen Itza are about the only known attempt to handle a crowd or to portray everyday life. On the best one of these, against a green field, are shown many warriors of equal size engaged in combat. At the top of the picture, and apparently in the middle distance, are houses 1 Maudslay, 1889-1902, IV, pis. 32-37. ! Maudslay, 1889-1902, II, pi. 87. Fig. 19. — Representation of drapery: Palenque. Fig. 18. — Kneeling figures: Stela 1, La Mar. GENERAL CONSIDERATION OF MAYA ART. 31 near which are women preparing for flight. The scene is very natural, as if viewed from a considerable height, and the absence of perspective does not make itself felt. A group of trees in a corner of the scene is drawn with much decorative effect. 1 When, however, the artist puts his hand to drawing a mountain or hill with men clambering over it, he goes quite beyond his power. One or two such attempts are very crude. Expression. Except in the more or less grotesque figures, there is very little in the way of expression. The drooping eye in profile faces gives a certain air of sadness, and the finest faces in the full round have perfect serenity (Plate 26, fig. 3). Sometimes, however, there is a sullen expression upon the faces of the captives that is probably intentional, as may be seen from the four kneeling per¬ sons in Fig. 10. In grotesque conceptions, however, grimaces and scowls are admirably portrayed. Perhaps the best example of characterization is the old man smoking a tubu¬ lar pipe, who adorns one side of the doorway to the shrine of the Temple of the Cross at Palenque. 2 The thin lips, stooped back and weak knees of old age are extremely well presented. Composition. Composition in simple and in subtle kind finds high expres¬ sion in Maya art. Not only the direct opposition of practically equal figures, as on Stela 1 of Ixkun, 3 but also balance secured by difficult modes and meas¬ ures, distinguishes the work of the Central American people. Long feathers drooping in graceful curves as well as strips and blocks of glyphs are commonly employed to fill out corners and carry the lines of interest. The tablets of Palenque might serve as models of composition in which the most intricate methods are used. In the three tablets of the Sun, the Cross, and the Foliated Cross 4 the balance is across the vertical, medial axis. The attendant priests in each instance are of different heights, and blocks of glyphs and other devices are employed to give equal weight to the two sides of the picture. The pyramidal type of composition is illustrated in the stucco panels of House A of the Palace. 5 Here there is in each case a central standing figure who holds before himself a ceremonial staff. The staff balances the headdress of this principal subject. In each of the lower corners is a seated figure who looks upward. This arrangement gives weight and stability to the design as a whole. Many other examples of pyramidal grouping might be mentioned. The two lintels of Yaxchilan reproduced in Figs. 10 and 17 show agreeable composition. In both designs the bottom is made much heavier than the top, although the pyramid is not very obvious. Composition on the diagonal is not so common among the Maya as some other forms. This may be explained by the abhor¬ rence of blank spaces. Still in some sculptures a diagonal line of interest is maintained. Attention might be called in this regard to one of the most re¬ markable works of art that the Maya produced in the matter of composition and execution. The splendid lintel taken by Maudslay 6 from Yaxchilan and deposited in the British Museum represents a divine serpent which towers above a crouching female worshiper who holds up a basket of offerings. In the mouth 1 Maudslay has reproduced this but without en- * Maudslay, 1889-1902, IV, pis. 76, 81 and 88. tire success, 1889-1902, III, pi. 40. 5 Maudslay, 1889-1902, IV, pis. 8-11. 1 Maudslay, 1889-1902, IV, pi. 72. 0 Maudslay, 1889-1902, II, pi. 87. s Maudslay, 1889-1902, II, pi. 69. 32 MAYA ART. of the serpent appears the upper part of a being in human form — probably an anthropomorphic god — who threatens the woman below with a spear. The top of this picture possesses a heavier interest than the bottom, and the diagonal line of division is well marked. It has been frequently pointed out that the Maya did not subordinate suffi¬ ciently for our tastes and that they did not understand the contrast value of blank space. Of course it is probable that most of the more complicated designs were painted in different colors. Under this treatment much of the complexity would disappear. 1 After all, the principal reason the drawings seem involved to European eyes is because they are utterly unintelligible. But it might here be noted that there was a tendency to simplify and to limit the field of vision toward the end of the first Great Period. This tendency is mostly in evidence at Piedras Negras and Palenque. The principal figures or groups are often care¬ fully framed in by strips of astronomical signs, etc., and considerable blank space preserved as a background. The sense of careful and accurate composition was probably developed along with the carving of hieroglyphs. Each glyph is indeed a careful bit of compo¬ sition and design limited to a definite and uniform space. The matter repre¬ sented in a single glyph ranges from whole figures of men and animals to cryptic abbreviations. Examination shows that the spaces were nicely divided and mapped out before the finer details were added. A number of monuments have unfinished figures and inscriptions that show the preliminary blocking out in the rough. Altar L at Copan and Stela 1 at Tikal (Plate 22, fig. 4) may be given as examples. Rough free-hand drawings found on walls at Tikal 2 and elsewhere give some idea of the artists’ prelim¬ inary studies. The Serpent The Origin of the Serpent in Art. The unique character of Maya art comes from the treatment of the serpent. Indeed, the trail of the serpent is over all the civilizations of Central America and southern Mexico. Any attempt to explain the origin of the serpent in Maya art must take note of the following facts concerning the religion and social organization of the Maya: 1st. The belief in many animal gods, some being more powerful than others. 2nd. The association of these powerful gods with natural phenomena. 3rd. The marked progression of these animal gods towards anthropomor¬ phism. 4th. A strong political structure almost amounting to theocracy. 5th. A ruling class with careful regard for inheritance. 6th. The number and magnitude of public works of a religious nature. All of these conditions may be explained as direct indigenous outgrowths of generalized totemism. This is widespread among the American Indians as well as among primitive peoples in almost all parts of the world. Totemism, which as a religious and social institution varies widely in many details, may be said to have as its basis a primitive philosophical conception of 1 A fine instance of this is seen in Miss Breton’s Itza now installed in the Museum of the University restoration of the painted sculptures of the Lower of Pennsylvania. Chamber of the Temple of the Jaguar at Chichen * Maler, 1911, pp. 56-63. GENERAL CONSIDERATION OF MAYA ART. 33 the world. According to this conception all, or at least a part, of the objects of man’s physical environment, such as animals, plants, heavenly bodies, and various sorts of natural phenomena, are his equals or even his superiors in the possession of skill and intelligence and a will to help or hinder. The individual chooses from these his best friend in the society of nature. The methods by which the choice is made are legion, but fasting, revery and self-hypnotism are common features. The religious side of totemism concerns the worship of the acquired guardian spirit of the individual or the inherited guardian spirits of the clan. Of course this worship does not constitute the whole of the religion of any people unless the powers of the totems are extended to cover the more general activities and phenomena of nature. The governmental side of totemism is of even greater importance than the religious, from which, however, it is de¬ rived. Principally through its relation to ideas of inheritance totemism tends to emphasize the importance of the family or clan and gives rise to a strong and stable society with well-defined leadership. In a progressive community the different clan protectors do not long main¬ tain an equal status. An unusual or striking ritual, a popular myth or a change in the condition of life may elevate one clan totem over the totem of other clans of the same tribe. Or the political fortunes of a family may redound to the credit of the being worshiped by that family. When a god, as when a man, rises from the ruck of the commonplace, he attracts to himself the strong qualities of his inferiors, for even among gods nothing succeeds like success. Such a process of survival and absorption may partially account for the importance of the serpent and other animal forms in Maya religion and art. But there is good reason to suspect that the serpent was more potent in art than in religion and that its importance in the latter was partly reflected from the former. The peculiar form of the serpent’s body was able to furnish a richer theme and one with more obvious possibilities of artistic development than could that of any other animal in the early list of totemic divinities. Most of the more or less anthropomorphic gods of the historic Maya pantheon are distinct enough in powers and attributes and seem to have successfully cast off some earlier animal nature only to be endowed afresh with ophidian characters. On the other hand, there is no single god that can safely be called the serpent god to the exclusion of all others. As a result of its artistic extension the serpent seems to have lost its earlier religious intention and to have become merely a sign or an attribute of divinity in general. Zoological Observations. In any analytical study of this most complicated subject it is necessary to distinguish three aspects. The first aspect concerns the physical or zoological basis or explanation for any representation. The sec¬ ond aspect concerns methods of idealization or evidences of the reaction of re¬ ligious ideas and inspiration upon the given natural form. The third aspect concerns conventionalization, so called, or the modifications brought about by a sense of pure design. More briefly these three aspects are, 1st, physical, 2nd, religious, 3rd, decorative. The serpent is seldom represented realistically, but we may safely infer that the rattlesnake was the prevailing model. The common rattlesnake of Central America and southward is the Crotalus durissus, which has been thus described 1 1 Ditmars, 1910, pp. 353-354. 34 MAYA ART. as to its coloration: . the ground color is rich yellow or pale olive; a chain of large brown rhombs, bordered with light yellow, extends along the back.” Rarely, indeed, did the Maya draw snakes with the markings given above. Rattles are sometimes shown, but in many cases these too are omitted. Usually the picture conveys merely a general sug¬ gestion of the snake with every feature more or less modified. Still the lack of realism was not a b Fig. 20. — Realistic serpents: Dresden Codex. owing to the inability of the artist, as may be seen from a number of excellent drawings in the codices (Fig. 20). Parts of other creatures are frequently added to the body of the snake, but it is usually difficult to make a zoological identification of these additions. The most important are the plumes of the quetzal bird and orna¬ ments and features taken from the human form. The jaguar is also a close associate of the snake in Maya art. Clawed forefeet are often seen on some of the more complicated representations. Other animals make occa¬ sional contributions to the more or less grotesque con¬ ceptions, while some features appear that apparently have no zoological explanation. It will be shown that in the composite figures now one component and now Fig. 21. — Idealized serpent: Fresco painting: Chichen Itza. Fig. 22. — Feathered serpent: Chichen Itza. another comes to the front, but that all are artistically controlled by the sugges¬ tion of the snake. The unnatural com¬ binations are doubtless attempts to figure characters that appear in the mythology and religion. Idealization. But while the religion provided the gross composites just noted, as subjects for artistic expression, it also inspired a fine, spiritual idealization of them. This idealization was achieved by two methods which, although fundament¬ ally distinct, nevertheless worked hand in hand. According to the first method of ideal¬ ization the body and head of the serpent were elaborated by the additions of scrolls, spirals, undulating lines and other ele¬ ments essentially serpentine. Sinuosity received its ultimate expression. Thus the serpent of religion was distinguished from the serpent of nature by being made more ideally serpentine. Fig. 21 presents an example of such elaboration from the frescos of Chichen Itza. I he second method followed in the idealization of the serpent was progres¬ sively anthropomorphic. The serpent was endowed with certain human aspects GENERAL CONSIDERATION OF MAYA ART. 35 by the addition, in the first stage, of ornaments worn by human beings. Nose and ear plugs appear on comparatively unelaborated representations. Fig. 22 shows a plumed serpent whose head is decorated by nose plugs and by a feather headdress. The plumes on the body curve outward in much the same manner as the scrolls just considered. As a final stage in the evolution of the divine ser¬ pent, a human or grotesquely human head was placed in the wide-open jaws. This device is well illustrated in Fig. 23. This latter addition is perhaps the most striking and original feature Fiq. 23. — Serpents with grotesque of Maya art and has, as we shall see, a most heads in their mouths: Copan. significant development. Fig. 24 shows a beautiful representation of the divine serpent from Yax- chilan in which both the above-described methods of idealization receive lucid expression. In this example the body markings of the serpent are also brought into play. The explanation of this progressive anthro¬ pomorphism is simple and dependent upon uni¬ versal principles. It is but the graphic record of a process that takes place in all religions, from totemism upwards, in which the super¬ natural relations between men and animals are intensified. In gods, as in men, there is always the union of a body and a mind. The body of the god may be any natural form, vegetable, animal or human, but the mind is always like that of man himself. It is impossible for man, savage or civilized, to conceive of a divinity with an intellect essentially different from his own although the powers may be magnified or ex¬ tended. When the body of the god is imagined as that of an animal, there comes, with in¬ crease of culture, a growing sense of incongruity or inadequacy in the association of a superior mind with an animal body. As a first stage, the animal-like body of the god becomes larger or more mysterious than that of the every-day wild animal of the same species. Such a stage may be seen, for instance, in the totem gods and clan ancestors of the tribes of the northwest coast of America. Gradually the divinities assume human form and manners. The half-animal, half-human gods of Egypt, Assyria, India and Peru, as well as the deities of the Maya, show the middle stage. Complete anthropomorphism is seen in the gods of Greece. Even here, however, there are many scholars who maintain that an earlier animal nature of these gods is disclosed by the peculiar epithets, by sacred animals, plants, etc., and by sacrifices. Be this as it may, it is clear that after they had achieved human form the gods had to struggle to keep up with the ethical prog¬ ress of their worshipers. Fig. 24. — Divine serpent with human head and hand in mouth: Yaxchilan. 36 MAYA ART. It was, as before remarked, the special quality of human form in the Greek gods which directed Greek artists toward the human form as a principal subject for artistic treatment. Aside from human associations, the human body possesses little, if any, more absolute beauty than does the most humble object shaped by the refining hand of nature, be it bird, flower or stone. In any really great national art the choice of subjects from such a wide field is usually directed by the specialized enthusiasm of religious fervor. The Serpent in Design. It is necessary to consider the serpent in regard to modifications which result from its constant repetition as a decorative and symbolical motive. In this aspect of art the serpent combines with diverse objects, natural and artificial, and presents many phases of so-called conven¬ tionalization. The character of the delineation of any figure in decorative art is determined by a sort of survival of the fittest. These surviving forms show certain quali¬ ties of order — especially harmony of measures and dominant directions, or parallelism of lines — that constitute the basis for any successful appeal to the esthetic sense. The suggestion has already been made that the artistic success of the reptilian motive in design probably had much to do with making the same motive strong in religion. Many divinities of diverse animal natures seem to have been overcast by the serpent, and the actual intrusion of ophidian features into distinct representations can often be demonstrated. Any national or regional design is, of course, finite. Its scope is limited by one or more modes wherein it is intensively developed. Now, in the case of the Maya the physical nature of the serpent reacted strongly upon the national sense of beauty. Not that they saw beauty where there was none, but that they accepted the special beauty of the serpent and neglected the other kinds. The serpent appeared to them the ultimate expression of grace. It must be admitted that the snake’s body has a very simple but exceed¬ ingly graceful outline. Good artistic values can be obtained with little diffi¬ culty with this as a motive. The body swells and tapers. Within a little dis¬ tance the body can bend upon itself, and the curves produced by such bending are almost capable of being plotted by formulas, so simple are the factors which govern them. But the snake’s body does not ordinarily fall into the simple and uniform wave forms with which it is commonly associated. Instead, it makes a succession of quick curves which merge into tangents or into long, slow curves. The characteristic lines of the snake’s body in repose or motion seem, upon careful study, to be as nearly angular as the physical limitations allow. A snake will stretch out along a wall and fold itself as closely as it can into a corner. In Maya design the serpentine alternation of quick and slow curves strikes the dominant note. Fig. 25 presents a collection of parts of designs chosen from many different situations. In all of these examples the angular shaping of the curves is manifest. The angular drawing of the serpent itself appears in a to d, while the remaining figures show the use of comparable lines in a variety of other instances. The striking development of vertical or horizontal lines of interest should also be noted. This is not due to the suggestion of the serpent, but rather to the universal principle of harmony of directions. The skill with 37 GENERAL CONSIDERATION OF MAYA ART. which tapering masses are handled must elicit admiration. Here is a'relation of lines directly opposed to parallelism, yet the sense of the parallel|is preserved, while the grace and variety of converging lines only lend a subtle interest. The prevalence of tapering, flamelike masses is characteristic of Maya art (Fig. 24, e, /, h, l, etc.). These forms were doubtless suggested byrihe repre¬ sentations of the snake’s body. Two other prominent characters^may have Fig. 25. Details showing influence of serpentine forms: a-d, Tro-CorteBianuB Codex; e, /, and h, Palenque; g, Chichen Itza; i, k, m, and n, Quirigua; j, Copan; l and o, Tikal; p, Labna! had a similar origin. One of these is the use of the double outline and other applications of parallel lines. Parallelism of lines introduces into any design an emphasis of contours and a harmony of measures. The origin of the double outline may be traced to the common method of drawing the belly of the snake so as to distinguish the area of large ventral scale plates from the rest of the body. This demarcation may be seen in the first four examples given in Fig. 25. In m is given one of the important factors of the great glyph that introduces the so-called Initial Series dates. The three loops at the bottom and the single loop at the top may represent in a vague and symbolical manner a portion of a serpent’s body. But if the device of the double outline comes from this natural source its application was greatly extended, as may be seen from its occurrence 38 MAYA ART. in the remaining examples in the collection above noted. The suggestion may also have come from drawing the midribs of feathers, but this seems less likely. The common use of rows or series of small circles is another pleasing feature of many drawings. These circles are probably derived by suggestion from the small scales on the body of the snake. Altar 0 at Copan (Fig. 26) shows these body scales, but it must be admitted that they seldom appear on important works of art. A fine application of this decorative element is seen in Fig. 25, l, a detail from one of the wooden lintels of Tikal. Groups of circles frequently decorate the eyes of serpents and of gods. They are also used as symbols of water and fruitfulness. In the imaginative modification of any given natural figure, for purposes of decorative art, there are a number of rather definite processes. Each of these is amenable to the fundamental principles of design, such as balance, rhythm and harmony, as these terms have been elucidated by Dr. Ross. 1 Each process may show, moreover, the phases of conscious and unconscious manipulation of the subject matter. Lastly, these processes of intensive development of a design motive, like the already described methods of the idealization of the serpent, work both singly and in combination. It is possible to detect much of the counterplay. The processes are: 1. Simplification. 2. Elaboration. 3. Elimination. 4. Substitution. Careful analysis of one group of designs after another, during which special attention is paid to the changes in homologous parts, makes pretty clear the manner in which the imagination works. In the first place, imagination does not create, it merely reshapes and recombines, taking suggestions and material from any thing lying within the field of experience. It may be likened to a kaleidoscope. Instead of bits of vari-colored glass are shaken up elements dis¬ associated from originally composite ideas. Through some agent of order these are rearranged symmetrically, so that the result satisfies the logical sense. In any developed decorative art the student may find a graphic record of the prog¬ ress of imagination. The term “ conventionalized art” comprises a number of diverse mani- 1 1907; see also Batchelder, 1910. GENERAL CONSIDERATION OF MAYA ART. 39 °L£> 9 Fig. 27. — Reptilian radicles after Dr. Harrison Allen. festations. To be exact, all art is conventionalized of necessity. In any repre¬ sentation there is always a compromise with truth and a mental allowance for inadequacy. But when any idea other than that of giving the most realistic representation possible is uppermost in the mind of the artist, the result may with¬ out any quibble be termed conventionalized. As a rule the decorative idea is more important than the realistic, and is achieved by limiting the field of the design and by modifying the lines of the model in a purely formal manner. Simplification. In an early paper that has not received the attention it merits, Dr. Harrison Allen 1 discusses the relations between natural forms and art forms. He finds that the tendencies of conventional art are: 1st, to repeat the normal lines of the model; 2nd, to diminish the normal lines of the model; 3rd, to modify according to a symbol; 4th, to modify according to mythic or religious ideas. The primitive designs given in paintings, etc., he calls “primals”; the final forms which result from a series of variants he calls “ultimates”; and the more or less ideographic figures that preserve the essential lines of a natural series of variants he calls "radicals.” In the art of Mexico and Central America he finds a very common radical which he terms the “crotalian curve,” be¬ cause it preserves the supposedly essential’ lines of the profile of the rattlesnake. Examples of tins crotalian curve are given in Fig. 27. It may be pointed out that this radical is more characteristic of Nahua than of Maya art. Maya art was vital, original and construc¬ tive, while Nahua art was largely devoted to imitations and to derived forms. The phrase, “normal lines of the model,” must be allowed a very liberal interpretation. In almost all kinds of realistic art among people of low culture the normal lines of any nat¬ ural form are at best roughly approached. In Fig. 28 is given a sketch of a rattle¬ snake head. When the mouth is wide open, the forked tongue does not naturally protrude as it does in this drawing. Note particularly the dentition. At the top are two backward curving fangs, while at the bottom are a number of small raking teeth. Note also that immediately above the eye is a scale plate somewhat more prom ment than the other scales that cover the face. The nearest approach to this natural head in sculptural, decorative art oc¬ curs at Chichen Itza in carvings in the full round (Fig. 29, a and b). These ex- 1 1881, pp. 289 et seq. Fig. 28. — Drawing of rattlesnake’s head, showing the parts. - Serpent heads with tenons: Itza. 40 MAYA ART. amples show dentition quite different from the natural form. At the front of the mouth are several teeth resembling incisors, and behind these are a number of molar teeth. The tongue hangs out at the back of the mouth. In b the promi¬ nent scales around the mouth are clearly indicated, but the scales on the rear part of the head resemble feathers and in fact upon the neck are unmistakably so represented. The supraorbital scale plate is greatly enlarged, and a nose plug is added. In Fig. 30 is shown a typical serpent head in profile as developed by the Maya for decorative purposes, with the parts lettered and named. The nose and ear ornaments have been taken over from the human figure and perhaps the beard as well. The two kinds of teeth shown are more fanciful in shape than those just examined. One kind is pretty clearly molar, and the other kind may be called incisor for convenience. The spiral-shaped object at the back of the mouth ( n ) may have originally represented the articulation of the jaw, although it is commonly referred to as a curled fang and is identified by Gor¬ don 1 with the sheath of the tongue. The tongue itself is shown in front of this object. Of the added features one of the most important is the small object (e), extending along the top of the nose, that is labeled the “nose scroll” for want of abetter name. Through it is thrust the nose plug, which usually represents a bone. To the circu¬ lar ear plug (Z) is attached a flowing or¬ nament divided into three parts. This head exhibits all the parts that charac¬ teristically belong to the developed ser¬ pent head in Maya decorative art. As can be seen, it is very different from the head of a natural snake. All the scales on the head are omitted in the conventional form except the large scale above the eyes (k), and this is greatly en¬ larged. The nose is elongated and the upper jaw made considerably longer than the lower one. To sum up, the head lacks prominent natural features, the remaining natural features are greatly modified and a number of unnatural features are added. The delineation of this head shows, however, artistic skill of no mean merit. The lines of interest are either vertical or horizontal, although the masses them¬ selves are of varied contours. The subtle and skillful use of sinuous shapes is deserving of note. This head is an excellent example of simplification. All the details are rep¬ resented economically in few lines, and there is a splendid harmony of parts that defies analysis. Of course the simplification could be carried further by 1 1905, p. 138. Fig. 30. — Typical conventionalized serpent head: a, body; b, belly markings; c, back markings; d, nose; e, nose scroll;/, nose plug; g, incisor tooth; h, molar tooth; i, jaw; j, eye; k, supraorbital plate; l, ear plug; m, ear ornaments; n, curled fang; o, tongue; p, lower jaw; q, beard; r, incisor tooth. GENERAL CONSIDERATION OF MAYA ART. 41 omitting the extraneous features. Indeed, a sort of factoring out could be car¬ ried on till the irreducible characteristic was reached. According to Dr. Allen’s nomenclature, such an irreducible characteristic would be a “radical.” Elaboration. Of less real worth in the development of art but of more com¬ mon occurrence is the process of elaboration. This process amplifies rather than reduces and by means of adventitious ornament renders the original form more complex. The unnatural features that appear on the typical serpent head just described are evidences of elaboration of a sort. But the most interesting elaboration does not add new features so much as it makes the old ones more complex. In Fig. 26 is reproduced a plumed serpent from Copan. The head Fig. 31. — Part of underlying design, Altar P, Quirigua: a, incisor teeth; b, molar tooth; c, eye; d, nose scroll; e, supraorbital plate; /, fang of supplementary head. shows most of the features already noted. The three divisions of the ear orna¬ ment hang down over the neck, but the ear plug itself is not visible. The nose turns up and then back. The nose plug is seen directly over the supraorbital plate. The short lower jaw with the beard and the long upper jaw with the large teeth are easily made out. The flame-like object that issues from the mouth may represent breath. So far little evidence of elaboration has been mentioned. This, however, appears in the treatment of the upturned nose and the tongue. The end of the nose, that turns back horizontally, is modified into a grotesque face, best seen by turning the picture on end with the head down. The tongue, which hangs out at the back of the mouth just behind the two molar teeth, is itself the upper jaw of a serpent possessing nose plug, supraorbital plate, and teeth of two kinds. Extreme types of the elaboration of the serpent head are found on Altar P of Quirigua (Plate 2). This altar, as will be shown later, represents a curi¬ ous and grotesque conception known as the Two-headed Dragon. The orna¬ mentation that overlies the body of this monster is several layers deep. In particular, on each side of the body, are two pendent serpent heads lacking the 42 MAYA ART. lower jaws. Each serpent head is elaborated to the last degree, but the manner of enrichment is different in each case. Fig. 31 offers in a simplified and partial drawing the first of these heads. In a and b we see the incisor and molar teeth, respectively, in c the eye dec¬ orated with feather-like markings, in d the nose scroll, and in e the remains of the supraorbital plate which projects farther forward than usual. The up¬ turned nose of this serpent head, part of which appears at the left of the draw¬ ing, really lies along the ground on the great sculptured boulder. The details are so modified by the irregularities of the stone and so concealed by other overlying figures that they can be made out only with the greatest difficulty. I his nose itself consists of two small superimposed faces of which the eyes are Fig. 32. Overlying ornament, Altar P, Quirigua: a, ornament above the eye; b and c, water symbols. the only obvious features. The lower one is partly reconstructed in the draw¬ ing. The detail marked / is the curled fang at the back of the mouth of this subordinate face. But the elaboration of the great serpent head does not stop with this. The larger spaces are overlaid with grotesques, as may be seen from Fig. 32, a, which gives the two designs decorating the back part of the eye. Another grotesque occupies the nose scroll and the space above the incisor teeth. The smaller spaces are filled up with motives that seem to be modifications, for the most part, of a very common water symbol (Fig. 32, b and c). This incrustation of ornament would seem to have no relationship with the serpent heads beneath, or with the Two-headed Dragon which the serpent heads themselves overlie, except to embellish. A second serpent head from Altar P occupies a similar and adjacent posi¬ tion to the one just examined, which, in fact, it partly conceals. It is given in Fig. 33, a. The original parts of the serpent are here much more difficult to distinguish, for the ornamental details are more closely incorporated. A hu¬ man face, bearing a peculiar forehead ornament and a prominent nose plug, is readily seen in front of the serpent’s upper jaw. This human face and the ser¬ pent jaws that partially enclose it are upside down on the monument, but are set right side up in the drawing. Curling locks of hair are seen at the side of the face. Perhaps the forehead ornament is intended for a tuft of knotted hair. GENERAL CONSIDERATION OF MAYA ART. 43 The eye of the serpent is indicated by the sunken space in the center of the design just back of the human forehead. The teeth project from the angles of the jaw in the form of double scrolls. The upturned nose ends in a grotesque and highly modified face, redrawn in b. As a whole, the face is comparable to the one already noted on Altar 0 at Copan (Fig. 26). It possesses a forehead ornament analogous to that of the human face beneath (Fig. 33, a). The back or top of what was originally the supraorbital plate consists of a face greatly modified, the Pig. 33. — Elaborated serpent head, Altar P, Quirigua: a, entire head; b, grotesque face terminating the nose; c, grotesque bead at the back; d, details representing bones. parts of which are indicated in c. The mouth of this face incloses a second one. Note also on various parts of the representation the use of the motive given in d, frequently in partial form and in connection with circles. This motive appar¬ ently represents the end of a bone. As we shall see in another section, the rather grewsome use of bones is highly developed in the art of the Maya. In connection with these two elaborated heads on the side of Altar P it must be emphasized that they themselves merely serve to elaborate the body of the so-called Two-headed Dragon. Elimination. Elimination of one feature after another of a natural motive till only one or two survive is a common phenomena the world over in decorat¬ ive art. In Maya art the process is frequently observed in the case of the ser¬ pent. Very often the entire lower jaw is omitted, as in the examples of elaborated heads we have just examined. In fact, the upper part of the serpent head 44 MAYA ART. adapts itself to many situations, usually with little change in the relative posi¬ tions of the different features but with much change in their configuration. More complete elimination, leading to the survival of but one or two details, is rather rare in the best period. It is more frequent in ceramic decoration and in the sculptures of Chichen Itza and other late cities in northern Yucatan. Ex¬ amples of incomplete and highly modified heads running the gamut of change will now be given. In Fig. 34 are shown serpent heads that lack the lower jaw a b c j Fig. 34. — Modified serpent heads: o and d, Copan; b and c, Palenque. and occasionally other features such as the ear plug with its attachments. In some instances there is a compensation for the loss by the application of foreign bodies. These heads are parts of elaborate figures, but are here given as indi¬ vidual examples. Very often the elongated nose of the incomplete serpent head is bent back¬ ward to form a fret. Examples are furnished in Fig. 35. The significance of Fig. 35. — Serpent heads at sides of aprons: a, Ixkun; b, Quirigua; c, d, and g, Copan; e, Naranjo;/, Seibal. this and other geometric modifications will be discussed at length in another place. Heads of this type occur particularly as enrichment of a widespread form of apron that is seen on many of the heroic sculptures. This apron (Fig. 15) has characteristically a front-view face in the middle and a profile ser¬ pent head at each side. In Fig. 35, a and b, the serpent heads retain the eye with its supraorbital plate, the nose scroll pierced by a single or double nose plug, and both the molar and incisor types of teeth. In c of the same series the nose plug is eliminated and in d to g both the teeth and the nose plug are wanting. Formal heads of this character also occur in other situations. Sometimes ear and nose plugs of human beings are modified into serpent heads with the nose GENERAL CONSIDERATION OF MAYA ART. 45 turned back in a fret. 1 The detail also is found in some of the elaborate head¬ dresses represented on stelae and lintels. 2 In marked contrast to the angular development just described there are many incomplete serpent heads cast into flamboyant lines. Examples of these Fig. 36. — Serpent heads conven- Flo. 37. — Nose plug representing a con- tionalized in flamboyant manner: ventionalized serpent head: ^Piedras Chichen Itza. Negras. are given in Fig. 36. In a the eye is seen at the top, and in front of this is the double nose plug. The nose turns downward in a flamelike scroll. The inner division of the nose represents the incisor tooth, while the molar tooth is shown below and to the right of the eye. The ear plug survives in the object at the extreme right. A somewhat similar head looking in the op¬ posite direction is given in 6. The reptilian motive is very intrusive and is much used for the enrichment of all manner of objects. In fact, most of the examples of incomplete serpent heads given above are themselves details that serve to complicate other con¬ ceptions. Fig. 37 illustrates the fanciful development of the nose plug of an elaborate bird, while Fig. 34, c, is one side of a comparable nose plug upon a similar bird head. Fig. 38 probably represents some sort of plant growth. Each branch, however, is modified by the addition of an Fio. 38. — Vegetal form incomplete serpent head and by other foreign details. The featora- Mm”* so-called crosses on the tablets of Palenque probably repre¬ sent trees. The branches are greatly modified by reptilian details. Fig. 39 reproduces a section of a vinelike decoration at Chichen Itza that is limited to a narrow band. Flowers, fruits and fish are clearly represented. Serpent fea- Fig. 40. — Plant form with stem modi¬ fied into a serpent head: Quirigua. tures are attached to the stems in some places. In this example the eye, the nose plug and the teeth are readily seen on the upper bend of the vine. Another example of the same sort of modification in a floral motive by adding certain serpentine features is seen in Fig. 40, taken from the Altar P at Quirigua. 1 For example, Maler, 1903, pi. 72, fig. 3. J Maler, 1903, pis. 46 and 47. 46 MAYA ART. The stelae of Copan and Quirigua are notable examples of reptilian enrich¬ ment. Fig. 41 reproduces an ear plug with which are connected no less than four more or less complete serpent heads. The one which shows the greatest elimination is that which issues from the side of the ear plug. Only the tip of the nose survives in this instance. Examples might be multi¬ plied, but enough have been given to make clear the various processes and stages of change. Substitution. The process of substitution likewise plays a great part in all highly developed art., whether barbaric or civilized. The substitution of new and striking details for old and commonplace ones — even at the cost of the first meaning of the design — is one of the simplest and most natural ways by which the imagination can reconstruct and revivify worn-out subjects. The creative effort is much less in making a parody than an original production. For the parody preserves, in greater or lesser degree, the fundamental composition upon which much of the esthetic interest of the original de¬ pends. Especially in decorative art, details of a composi¬ tion realistic or geometric may be progressively replaced by other quite different details until in the end only a trace of the original setting remains. The true history of the design is made clear only through a study of the homol¬ ogous parts of a series of stages. Occasionally, as in the series of three heads given in Fig. 42, there is a sort of degenerate modification due to the redoubling or displacing of some feature. In a we see a rather simple serpent head with the nose plug projecting forward from the front of the eye. In b this object is doubled and projects from the top of the eye, while in c it is repeated many times upon the eye and in front of it. Elaboration and substitution are closely akin, but, in¬ deed, all the processes that have been described work hand in hand. Each has its special field where it may be studied to best advantage. Substitution may be studied best in the development of the Mask Panel, which will be taken up under Architecture. Since, in principle, this process is Fiq _ ge simpler than any of the others, the illustration of it will be modified IL-geiy by postponed till the consideration of the latter subject multiplication of none T , , , . , . , , J plugs: Chichen Itza. Test a false idea concerning the relation of the realistic motive to the geometric should follow from the examples of development and modifications that have just been given, it seems best to take up at this time a brief discussion of geometric art in its relation to the serpent and other life forms. The Serpent and Geometric Art. It has for some years been the vogue among students of primitive art to derive all geometrical elements in decorative Fig. 41. — Detail of an ear plug with attached serpent heads: Copan. GENERAL CONSIDERATION OF MAYA ART. 47 Fig. 43. — Breast ornaments: Copan. art from realistic forms through increasing conventionalization. Even in so elaborate an art as that of the Maya there have been attempts to derive the fret, the spiral, the guilloche, etc., from the serpent. 1 The usual method adopted to show such derivation is to arrange the designs in a “series” with a recogniz¬ able life form at one end and a pure geometric form at the other, and between these extremes to place a number of highly modified figures which show increas¬ ing similarity with one or other of the extremes. Such a method of study is highly useful and suggestive. But there are three possible lines of explanation. The change might be considered to move from the realistic to the geometric, from the geometric to the realistic, or from both the extremes inward. In order to prove either of the first two processes it is necessary to es¬ tablish two things: first, chrono¬ logical sequence, for the derived form must come after its origi¬ nal in point of time; second, a reasonable explanation why the change occurred. As regards the first point, it is often upon the same object that all the stages of change are represented, as, for instance, in the case in the carved paddles of the Hervey Islands, the study of which by Stolpe really led to the derivative method of explanation. Such a series in Maya art might be taken from the breastplates of the small human figures on the step of Temple 11 at Copan (Fig. 43), to prove beyond doubt that the tau sign is derived from the mouth of the jaguar. The successive elimination of one feature after another till only the opening of the mouth remains is so obvious that it needs no comment. All the stages of change here given are shown on this one monument, with the exception of c and / that illustrate an¬ other departure and are taken from Altar Q, showing the same style of sculpture. Few students would insist that any historical significance could be attached to the phenomenon under such circumstances. But chronological sequence from lost sculptures might be assumed and the possibility of sur¬ vivals invoked. At least it is reasonable to ask why the change took place. What natural quality is present in a jaguar’s face that would lead to the survival of the mouth in the form of a tau sign? And why should the cross hatching within this mouth (a) survive on a bar-shaped breast ornament (/) and not on the tau sign ( Maler, 1902, p. 223. Fig. 46. — The Ceremonial Bar: Copan, a, Stela P; b, Stela N. GENERAL CONSIDERATION OF MAYA ART. 51 The face of the manikin varies considerably, but is characterized by a long turned-up nose and a wide-open mouth which has in its upper jaw a prominent flame-shaped tooth. The lower jaw is usually much shorter than the upper one. These grotesque features are decidedly reptilian according to the Maya stand¬ ard for such things. Indeed, the upper part of the face will bear comparison part for part with the typical serpent head. Often a long celt- shaped object projects from the forehead. The body of the small figure has, as a rule, no covering except a belt with apron attached and such minor ornaments as arm bands and necklaces. There are usually oval markings on the legs, back and arms that may be intended to represent the scales of snakes or other reptiles. The proven¬ ance of the appendage is an interesting problem and several explanations may be given. Its purpose was doubtless to indi¬ cate still more clearly the ser¬ pentine nature of the figure, but there is great uncertainty whether it is a modified leg or phallus or even the umbilical cord. Only one leg is shown on the examples that are clear enough for detailed study, and it seems probable that the pendent serpent takes the place of the more distant leg. The general lack of sex significance in Maya art is an argument against phallic origin. It may, of course, represent the umbilical cord, but it does not begin at the right point. In Fig. 48 is reproduced a potsherd from the highlands of Guatemala, that has an incised representation of the Manikin Scepter so modified as to fill an oblong space. The serpent appendage is seen at the right. Unfortunately the design is not complete, but the decorative band on the base of the appendage resem¬ bles a leg or arm band and may indicate that the serpent appendage is really a modified leg. The same manikin type of figure appears in the guise of a newly born child at Palenque. In the Temple of the Inscriptions are four panels with stucco relief which show a human being holding a child in one arm, while the other arm is stretched out to support the ophidian appendage (see Fig. 49, a). The Fig. 47. — The Manikin Scepter: a, Yaxchilan; b, Quirigua. 52 MAYA ART. faces on all these representations are unfortunately destroyed. Maudslay’s drawings give both feet of the child, but a study of the photographs shows that the restoration of the more distant foot is doubtful, although the knee is clear enough. In two or three sculptures at Yaxchilan 1 the manikin figure with the flexible appendage is seated upon an inverted basket-like object on the top of a pole. Manikin figures without the flexible appendage are sculp¬ tured upon the tablets of the Temple of the Cross, the Foliated Cross and the Sun at Palenque. 2 They are represented either in a sitting or reclining attitude upon a folded cloth supported by the outstretched hands of a priest. The best preserved example is given in Fig. 49, b. The final stage of the Manikin Scepter is marked by the survival of the charac¬ teristic head upon some sort of staff. Proof of the actual connection of this type with the more complete ones just described ap¬ pears upon a stela found by Dr. Tozzer at a ruin on the upper Tzendales River. The object carved upon this stela is reproduced in Fig. 50. The head is clearly of the same character as heretofore. It is set upon a short staff which is held out opposite the face of the principal personage in the man¬ ner already noted. The body, however, is entirely eliminated. The staff itself is rigid for the greater part of its length, but the lower end bends outward and terminates in a serpent head, is clearly a survival of the ventral appendage. Examples of the head upon 1 Maler, 1903, pis. 50 and 67. * Maudslay, |1889-1902, IV, pis. 76, 81, and 88. Fig. 50. —The Man¬ ikin Scepter with body reduced to a staff: Tzendales. Fig. 51. — The Manikin head on a staff: Palenque. This staff, then, GENERAL CONSIDERATION OF MAYA ART. 53 a simple staff occur at Tikal, 1 Piedras Negras 2 and Palenque. Fig. 51 shows one from the last-named city. Two-headed Dragon. The monstrous creature to which Maudslay has given the name Two-headed Dragon” will next be considered. This grotesque ani- Fio. 52. — Two-headed Dragon: Copan. mal is seen in its simplest form on the side of a small rectangular altar at Copan (Fig. 52). The principal characters are as follows. There are two heads, one of which distinctly belongs to the front and the other to the rear, as may be seen from the direction in which the feet are pointed. The markings on the legs and belly are reptilian, and there is a prominent water symbol on the side. The feet in this example are clawlike, but in many other cases they resemble the cloven hoofs of deer or peccary. The front head is hard to characterize, but as a rule the face or snout is long and shaped somewhat like that of a crocodile. Often the eye is feathered and decorated with a diagonal cross. In the specimen before us a human head is seen in Fio. 53. — Middle part of Altar m : Copan. the open jaws, but this feature is out of type. The rear head is much more definitely marked. It is a rather grotesque face with an upturned nose. The head¬ dress is a triple symbol with a leaflike object in the center, a shell in top view or profile on one side and on the other an oval object marked with a diagonal cross. Upon the Fm. m. — Trifod scroll forehead is usually the kin or sun sign that resembles the on joints of Two-headed . , , , ,,,, . . Dragon. common ring and cross symbol, the lower jaw is repre¬ sented as a bleached bone, and sometimes the nose has a cavity that likewise indicates death. A more elaborate treatment of this motive is seen in a number of altars at Copan and Quirigua. The central portion of the Altar of Stela M is presented in Fig. 53. In this instance the body of the monster is carved on one huge block of stone, but the front and rear heads are each carved on a smaller block. The feet are all pointed the same way and are of the cloven type. The joints of the legs, on this example and many others, are marked with a peculiar scroll which generally assumes the trifoil form given in Fig. 54. A grotesque face bearing water symbols occupies each side between the front and hind legs, and another similar 1 Maudslay, 1889-1902, III, pi. 73. ’ Maler, 1901, pi. 15, fig. 2. 54 MAYA ART. Fig. 55. — Rear head of AI tor M: Copan. face adorns the top. Thus we have on this specimen five faces, three of which may be accounted for by the process of elaboration that has already been ex¬ plained. The separate blocks upon which the front and back faces of the mon¬ ster are carved have unfortunately suffered mutilation. A sketch of the rear face is given in Fig. 55. The three signs of the headdress are rather hard to make out excepting the middle member, but the kin sign on the forehead is very clear as well as the characteristic grotesque face with the bone grooves and crescents on the lower jaw. The front block shows a widely extended reptilian mouth enclosing a human head. In the effort of the Peabody Museum Expedition to set the two supplementary blocks of this altar in position they were unfortunately turned around so that the rear head appears at the front of the animal in the photographic reproduction and vice versa.' The Altar of Stela N 2 is somewhat similar, but lacks the upper face and is carved from a single block. Upon the top of this altar is the trifoil scroll that characteristically occurs on the joints. The Altar of Stela D is interesting on account of the syncopation that it shows. This monument is a more or less cubical block bearing two faces on diagonally opposite corners. One of these faces represents the front head of the Two-headed Dragon and the other the rear one. On each intermediate corner is a vertical bone with two clawed feet attached. These two leg bones with their double feet are all that remain of the body of the monster. The face that corresponds to the front is ornamented with water symbols. The rear face is given in Fig. 56. This clearly represents a death’s head, as may be seen by the nose and by the circles, crescents and wavy lines on the jaws. The peculiar triple symbol is absent, but the eyes are modified into the shape of the kin sign. Altar B of Quirigua (Plate 1, figs. 3 and 4) presents far greater elaboration than any speci¬ men so far considered. A more or less irregular boulder is completely covered with carvings in low relief and of very great detail. The front head has feathered eyes marked with the diagonal cross. In the open jaws is a human head with an animal headdress. The legs are doubled upon themselves frog fashion and have the trifoil scroll at the joints. The feet are reptilian in appearance. The rear head is crowded into an irregular space and is represented in profile in a horizontal position looking downward. The sun symbol is carved upon the forehead, and the middle element of the triple headdress is given in an elaborated style. The entire top of the altar is covered with a complicated scroll-work face. The legs conceal so much of the sides that there is no room for additional faces in these positions. These legs, however, are themselves overlaid with large hieroglyphs of the most elaborate type. Altars O and P at Quirigua belong to the same series as the preceding sculp¬ tures. The first of these is not in a very good state of preservation and the de- 1 Gordon, 1902, a, pi. 17. * Maudslay, 1889-1902, I, pi. 83. Fig. 56. — Rear head of Altar D: Copan. GENERAL CONSIDERATION OF MAYA ART. 55 tails are difficult to make out. Altar P (Plate 2), sometimes called the “Great Turtle Altar,” is perhaps the most complicated as well as the best preserved piece of sculpture in the entire Maya area. It is a natural boulder of great size and of hard stone, with carvings in fairly high but delicately modeled relief. Maudslay 1 regards this sculpture as a representation of a turtle, but a compar¬ ison of details shows that it belongs to the Two-headed Dragon group, although much modified by the several layers of ornament that conceal the animal form beneath. The front face shows a richly attired human figure in front view seated cross-legged upon the lower jaw of a great, open mouth. This human figure resembles very closely those carved on the stelae. In the right hand he holds the Manikin Scepter and in the left a shield. Concerning the great head that Fig. 57. — Two-headed Dragon with elongated body: a-c, Palenque; d, Piedrae Negras. contains this figure little can be said. The pointed teeth in both the lower and upper jaw are easily discernible. The eye is decorated with feathers and with the diagonal cross. The rear head bears evidences of death and is of a grotesque type, but the sun sign and the triple headdress are absent. The head on the top of the monument resembles that on Stela M at Copan, and is marked with many water symbols. The sides of this great altar are ornamented by conven¬ tionalized serpent heads that hang down from the sides of the face above and partly overlie each other. These heads have already been explained in some detail as examples of artistic elaboration (Figs. 31 to 33). A return to less labored presentation of the Two-headed Dragon is now in order. At Copan this motive is used in the adornment of the inner doorway 1 1889-1902, II, p. 17. 56 MAYA ART. of Temple 22. The partly destroyed design has been restored by Maudslay. 1 The pendent heads of the monster rest upon the hands of two seated human figures and the body stretches across the doorway. In the significant details of heads and legs this representation agrees with the type specimen that was described first. The feet are of the cloven type. The parts of the body adjacent to the heads show ventral scale plates, but the rest of it consists of a number of S-shaped devices in which are en¬ tangled small human figures with gro¬ tesque faces. Fig. 57, a-c, gives an example of the Two-headed Dragon with the body still further modified. This design stretches over the doorway and along each side F, 0 . 58. - Manikin &*pt«r«nd portion of Ceremonial of a room at p a l e nque, and is executed in stucco. The heads are fairly true to type and the legs as well, but the body is conventionalized into a long band of astronomical symbols. Upon the center of this band and directly over the doorway is perched a bird with wings extended, the head very much out of proportion to the rest of the body. The rear head of the monster is turned upside down, perhaps to emphasize its inferior position. The phase of the Two-headed Dragon shown in this figure is well established. Several fine examples occur on the stelae of Piedras Negras, one of which is given in Fig. 57, d. Here the astronomical band forms a framework for a human being seated in a niche, and the two heads are brought close together at the bottom just in front of a sort of throne. Note the legs with cloven feet and with trifoil scrolls at the knees. Other forms related to the Two- headed Dragon will be given pres- Fig. 59.— Rear end of Cere- ently. monial Bar: Siebal. Composite Types and Miscellaneous Variations. The three objects or conceptions whose principal develop¬ ments have just been described certainly appear distinct enough at first glance. But as a matter of fact each is more or less connected with the other, and all break down into variant types and gradually lose their indi¬ vidual characters. The intermediate stages will be pre- carved bone: Copan. sented first, in order, and then the decadent stages will be treated. The connecting link between the Ceremonial Bar and the Manikin Scepter appears at Tikal upon Stelae 1 and 2. The human beings represented on these 1 1889-1902, I, pi. 12. GENERAL CONSIDERATION OF MAYA ART. 57 monuments hold against their breasts unmistakable Ceremonial Bars of the second or straight-bodied phase. But sitting on the lower jaws of the serpent heads that terminate the bars is the complete Manikin Scepter, ventral append¬ age and all (Fig. 58). This little figure has quite evidently replaced the head or bust that usually appears in the serpent’s mouth on other repre¬ sentations of the Ceremonial Bar. Many of these heads or busts are found upon examination to re¬ semble the physiognomy of the manikin, but others are of a very different type. The Manikin Scepter is a common substitute for the Ceremonial Bar. It probably is not derived from this object but from the more generalized body of the serpent. At Seibal the Ceremonial Bar seems to have fallen under the influence of the elongated phase of the Two-headed Dragon as may be seen on Stela 9, 1 and Stela 10 (Plate 25, fig. 2). This object is held in the arms of the human being in a tilted position. The upper or forward end is developed into a curious head with flamboyant details. The head that usually appears in the jaws of the serpent has in this case been moved upward and attached to the serpent’s nose. This same feature appears on the Cere¬ monial Bar of Stela 7 at Naranjo. 2 The lower or backward end of the Seibal specimen, reproduced in Fig. 59, is modified into a likeness of the inverted rear head of the Two-headed Dragon. The details are represented in a flam¬ boyant style, but the triple headdress with the shell, the leaflike object and the saltire are discernible as well as the typical grotesque face with the kin sign on the forehead and the bleached bone for a lower jaw. The connection between the Manikin Scepter and the Two-headed Dragon is more difficult to demonstrate. Altar P at Quirigua represents, as we have seen, a very much elaborated Two-headed Dragon that has in the mouth of Fia. 61. — Ceremonial Bar held in tilted position: Yaxchilan. Fig. 62. — Man holding two-headed serpent of flexible type: Yaxchilan. the front head a human figure carrying a Manikin Scepter. But this circum¬ stance may have no special significance. The real connection between these two concepts is shown through the general similarity of the rear head of the monster to the head of the manikin. Aside from the symbols indicating special powers that are marked upon the rear head the physiognomy of the two figures is almost identical. It will be shown presently that these heads may be classed as different manifestations of a generalized god. In the meantime a more defi- 1 Maler, 1908, a, pi. 10, fig. 2. ! Maler, 1908, i>, pi. 22, fig. 1. 58 MAYA ART. nite idea of the affinity may be gathered from Fig. 60. This reproduces a carved fragment of bone, formerly painted red, that was picked up on the river front at Copan and is now in the Peabody Museum. The design shows a part of a reptilian monster with scaly legs and a head that is very similar to the manikin head. Although this head is most easily explained as the rear head of the Two- headed Dragon, still all the noted symbols are absent. The Ceremonial Bar is seen in a num¬ ber of variant and decadent forms. When held in a tilted rather than horizontal posi¬ tion, its symmetry begins to break down. Such a tilted bar is given in Fig. 61 from one of the Yaxchilan stelae. Note the dis¬ location of the lower head. A reversion of Fio. 63. — Ceremonial Bar c from Coban. i pottery box type to the original flexible serpent may be seen in Fig. 62, which likewise is found at Yaxchilan. The body of the bar is a flexible snake body that folds over the arm of the seated human figure. In the open jaws are likenesses to the head of the Manikin Scepter. Fig. 63 reproduces a piece of finely modeled pottery from the uplands of Guatemala upon which the Ceremonial Bar is rep¬ resented in a simple manner with a knife blade instead of a head in the serpent jaws. The same feature appears on Stela 25 1 at Naranjo. On the small stelae at Ocosingo (Plate 25, fig. 5) the Ceremonial Bar appears to end in ser¬ pent heads without any object in the Fl0 ' 64 ' — Degenerate (orm of the Ceremonial Bar: mouth. On Stelae A and C at Quirigua 2 C ° pan ' the Ceremonial Bar ends in small incomplete heads. The lower jaws are lack¬ ing and a long pendent object seems to be attached in their stead. A very decadent form of the bar is seen on Stela 11 at Copan (Fig. 64). Some objects appear as possible sub¬ stitutes for the Ceremonial Bar, being held in much the same way. On Stelae Fig. 65. Substitute for Ceremonial Bar: Tikal. 11 and 16 at Tikal appears the object shown in Fig. 65. In the center of an openwork staff is set a grotesque head of familiar type. On the south face of Stela F at Quirigua the human figure is represented with his hands held against his breast in the attitude taken when holding the Ceremonial Bar. However, no such object is in evidence. Below each hand is a serpent head that is sus¬ pended from a chain attached well up on the headdress. These two serpent heads may be survivals of the old order. It is interesting to note that an almost identical arrangement is seen at Palenque, as is made clear by the two 1 Maler, 1908, 6, pi. 40, fig. 1. ’ Maudslay, 1889-1902, II, pis. 4 and 16. GENERAL CONSIDERATION OF MAYA ART. 59 drawings given in Fig. 66, a and b. The decadent forms and the possible sur¬ vivals above noted are of great value in determining chronological sequence. There are a number of curious sculptures that are perhaps related to the Two-headed Dragon group. The most important are Altars G 1, G 2, G 3 1 and O 2 at Copan. These are vertical slabs of stone that are carved into reptilian forms. They make a series midway between the Two-headed Dragon and the ordinary representations of the Feathered Serpent. The Feathered Serpent on one side of Altar 0 has already been figured and described (Fig. 26). The design on the opposite side shows two serpents with greatly enlarged heads and small intertwined bodies, ending in “Ahau” sym¬ bols. The heads are similar in detail to the head on the opposite side of the altar, Fig. 66. — Serpent heads attached to chain-like objects: a, Quirigua; b, Palenque. and the upturned noses end in the same grotesque face. But under each head is a leg with a clawed foot. In the space enclosed by the intertwined bodies is a bunch of feathers. Altars G 2 and G 3 represent a double-headed serpent (the two heads being uniform in all particulars) with an arched body decorated by triangular and circular markings and by a mane of feathers. There are no legs on either of these reptilian forms. Altar G 1 is more elaborate. It also has two heads, one being smaller than the other, and a short body concealed beneath feather fringes and a double column of glyphs. The smaller head is similar to the heads of Altars G 2 and G 3 except that it has a Venus symbol marked on the eye and a grotesque bust in the mouth. Under this head is a leg, the character of which does not appear very clearly. The larger head also has a grotesque figure in the mouth. The lower jaw of this mouth consists of a bleached bone. The leg under the head also has bones marked by circles and wavy lines. From this description it is apparent that the last head comes pretty close to the type of the Two-headed Dragon, and that the series as a whole simply 1 Maudslay, 1889-1902, I, pis. 116-117. ! Maudslay, 1889-1902, I, pis. 84-85. 60 MAYA ART. emphasizes the lack of definite demarcation between the various conceptions in Maya art. Minor details on headdresses, etc., show two-headed reptile forms of a nondescript type, an example appearing in Fig. 67. Other phases of two- headed animals, now approaching pe of the Ceremonial Bar w the Two-headed Dragon, :eive still further considera- connection with material codices and the representa- f certain gods. The Serpent Bird. This name is applied to a bird motive with cer- Fig. 67. — Two-headed figure: Yaxchilan. tain reptilian features that occurs in many of the southern Maya cities. According to Maudslay, 1 the most essential character of the design seems to be the presence of a conven¬ tional snake’s head (without a lower jaw) in place of or overlying the bony structure of the bird’s wing.” He adds that the Serpent Bird may simply be another way of ex¬ pressing the idea intended to be conveyed by the Feathered Serpent. Maudslay ■ gives an entire plate to the explication of this complex figure, picking it out the various essential parts in different colors. It is a question whether the single feature of the wings is sufficient to show that the Serpent Bird repre¬ sents a fixed idea, d he head of the Serpent Bird assumes a number of distinct forms, and the head is usually the part that expresses the real individuality. This bird is seen in profile at Palenque and Piedras Negras. In the former city it is represented in two cases on the tops of the ceremonial trees, which so closely re¬ semble crosses that they have caused much foolish specu¬ lation. The heads of these two birds are similar to the long-nosed grotesque heads of the manikin figures which are represented elsewhere on the same tablets. One of these birds is reproduced in Fig. 68. The conventionalized serpent head may be easily seen on the under side of the wing in an inverted position. The general idea of a bird upon a cross-shaped tree occurs rather widely, but in other instances the serpent Fig. 69.-staff representing head on the wing does not make its appearance. For abird on a tree: Yaxchilan. : nQfoTW>£1 . instance, the idea is embodied in a sort of ceremonial wand that is seen a number of times at Yaxchilan (Fig. 69). Likewise in the illuminated manuscripts from the neighboring Zapotecan area, the bird 1889-1902, I, p. 63. 2 1899-1902,1, pi. 99. GENERAL CONSIDERATION OF MAYA ART. 61 on the cross-shaped tree plays an important part, but the bird is without reptilian features. 1 The Serpent Bird of Piedras Negras is perched upon the top of a grotesque head that forms a sort of canopy over a seated personage on Stela 5. 2 The face of this bird is not of the long-nosed variety, but is much more nearly human. The plumage, however, is much the same as that of the Palenque examples and doubtless imitates that of the quetzal. The front view representation of the Ser¬ pent Bird is much more common than the profile view. Fine examples may be seen on the back of Stela H at Copan (Fig. 70) 3 and at the top of the lintel of Temple IV at Tiled. 1 The serpent heads are arranged verti¬ cally on the inner sides of the wings in the Copan specimen and horizontally on the under sides of the wings at Tikal. In both cases long bones in pairs project outward from these serpent heads and extend across the fringe of feathers. In the last example the bird is really perched upon the arched body of the Two-headed Dragon. This same association is seen in Fig. 57, a and d, already described. In these cases the wings are fully spread, with the serpent heads in an inverted horizontal position. A stucco ornament over a door at Ocosingo represented a Serpent Bird of this type. An incomplete drawing of this by Catherwood 6 has been repeatedly miscalled a Winged Globe in various labored attempts to connect the civilization of the Maya with Egypt. Objects similar to the wings of the Serpent Bird are widely found on stelae and other sculptures as lateral ear ornaments of the richly dressed human figures. This device might be called the Wing Panel. A rather realistic instance of it is given in Fig. 71, and more conventionalized forms in Fig. 72. Of the examples given here a and b show a single serpent head at the side of the wing, while c and d show one head at the top and another at the bottom. It is quite possible that the Wing Panel Fie. ix. — Small clay head with wing was j nven ted and developed as an independ- ent ornament and was later used m the artis- tic elaboration of any sort of bird figure. We have seen throughout this study that the serpent was a very active element in art and was able to force itself into all sorts of designs. Birds in their more natural aspects will be taken up separately, when further evidence concerning the serpent head on the wing will be offered. The Long-nosed God. Having examined in some detail certain of the more important religious objects shown in the sculptures, we are now in a position to Fig. 70. — Serpent Bird. Stela H: Copan. 1 Seler, 1902-1903, pp. 77-81; Nuttall, 1901, pp. 187-190. 1 Maler, 1901, pi. 15, fig. 2. 3 Maudslay, 1889-1902, I, pi. 61. 4 Maudslay, 1889-1902, III, pi. 78. 6 Stephens, 1841, II, p. 259. 62 MAYA ART. attempt a correlation of the material on the monuments with that in the codices. The objective method gives safer results than the subjective and will be employed in most cases. We will first consider the multifold character and phases of a figure that must represent one of the principal Maya gods. From a persistently characteristic feature this deity is termed the Long-nosed God. Because of the natural exuberance of Maya art identification even of gods is far from easy. Fewkes 1 declares that in any attempt to classify the Maya Fig. 72. — The Wing Panel: a, Copan; b, Yaxchilan; c and d, Quirigui deities the character of the head must be taken as the basis. This statement is true within certain limits, simply because characterization is more easily expressed in the head than elsewhere, especially when the figures are largely anthropomorphic. But in many cases the character and decoration of the body are also significant and should be examined. It was remarked in the prelim¬ inary explanation that the more or less human head or bust in the mouth of the divine serpent was intended to express the fundamentally human intelli¬ gence of an animal divinity. More detailed study has shown that the serpent itself is merely a badge and cloak of godship. The personality and special powers of the individual gods who have more or less of the serpent character are ex¬ pressed largely by symbols and by grotesque modifications of the face and body. Schellhas, in his well-known paper on the “Representation of Deities of the Maya Manu¬ scripts,” calls the most common figure in the codices God B. He declares 2 that this god is “a universal deity to whom the most varied elements, natural phenomena and activities are subject.” Many authorities ‘"“Dris Code” 1 a ” d wi ‘ h “ rpent body: phenomena and activities consider God B to represent Kukulcan, the Feathered Sement. whose Typical examples of this god are shown in Figs. 73 and 74. The nose is, after 1894, pp. 260-262. 2 1904, p. 16. GENERAL CONSIDERATION OF MAYA ART. 63 Fig. 74. — God B holding serpent in his hand: Dres¬ den Codex. all, the most characteristic feature. This is long and usually rather pendulous, with a curled object attached to the top. The mouth shows a flame-shaped tooth at the front and frequently a somewhat similar object at the back. The representations of this god in the Tro-Cortesianus Codex are similar to those in the Dresden Codex as far as features are concerned, but the style of delineation is much coarser in the former manuscript. The general similarity of the face of this god to the face of the serpent is apparent: the former is simply the latter shortened and humanized to a slight extent. The curled object above the nose is clearly the homologue of the nose scroll of the serpent (Fig. 30), through which the nose plugs are thrust. On some of the faces of God B in the Dresden Codex the nose plugs are still attached to this object. A personage so important in the manuscripts as is God B could hardly escape representation in the sculptures. Indeed, with slight chance of error, he may be identified with a num¬ ber of figures, characterized by a long nose, that occur in many situations. In fact, some of the phases of a Long-nosed God of the sculptures have already been discussed under the title of the Manikin Scepter. We have seen the head of this grotesque deity thrust forth from the gaping jaws of the serpent, and we have seen the entire body —with its peculiar serpentine appendage that declares over and again the ophidian nature of the god — seated upon the under jaw of the serpent head on the Ceremonial Bar, or held up as a sacred thing before worshipers. In representa¬ tions in the Dresden Codex God B likewise issues from the jaws of the serpent (Fig. 73, a). In other representations he sits cross-legged upon the open mouth. He even appears in some drawings with the body of a serpent (Fig. 73, b). But in the majority of cases he has the body and the dress of a man. Reverting again to the long-nosed manikin god of the sculptures, it has been noted that the ventral appendage disappears in the more advanced stages and that the face expresses the exact nature of the divinity by its striking reptilian features. God B likewise has reptilian features, as has already been shown. A comparison of the two discloses a remarkable similarity of parts. The following features of the face are practically identical: 1st, long, sinuous nose with the nose scroll at the top; 2d, long, single or double tooth at front of mouth; 3d, curled fang at back of mouth; 4th, lower jaw much shorter than upper jaw. Besides these similarities in the face there are often comparable oval markings on the limbs and torso. In Fig. 74 we see £od B holding up a snake in the exact manner that the Mani¬ kin Scepter is held in the sculptures. On the monuments many long-nosed grotesque faces occur as details of artis¬ tic enrichment (Fig. 75) on human figures and other objects as well as in the mouths of serpents on the Ceremonial Bar and in other connections. The range of form is remarkable and the transitions smooth and without a Fig. 75. — Head of Long- nosed God as secondary ornamentation: Quiri- gua. 64 MAYA ART. Fig. 76.— Head with elaborated nose: Quirigua. break. In some cases the long-nosed faces receive a decidedly elaborate treatment (Fig. 76). In the codices there is a second kind of long-nosed figure with an extremely elaborate face who is called God K (Fig. 77, b). According to Schellhas, 1 God K is closely related to God B and yet distinct from him. He suggests that this god has some astronomical significance. Brinton 2 and Fewkes 3 consider him simply a special manifestation of God B, and Forstemann 4 holds that he is a storm god. The close relation between God B and God Iv is indicated in many ways. For instance, the former sometimes wears the latter’s head on the top of his own (Fig. 77, c). The face of God K seems to be derived from the face of the elaborated serpent that is often associated with God B. In the Tro-Cortesianus codex we find the body of God B attached to the middle portion of a snake bearing the head of God K (Fig. 77, a). In the Dresden Codex the serpent, from the mouth of which God B issues and upon the jaws of which he sits, has a remarkable likeness to the same god (Fig. 73, a). We have already seen that the face of God B itself resembles the serpent face, but the resemblance is not so striking as in the case of God K. Anthropomorphism is more com¬ plete in the case of the more important deity. It might be well before proceeding on another line of inquiry to consider briefly the functions of the Long-nosed God in the phases so far pre¬ sented (including Gods B and K of the codi¬ ces and the Manikin Scepter God and cer¬ tain similar forms on the monuments). This generalized deity is prominently associated with water and vegeta¬ tion. Leaflike objects, water plants, fish and shells are frequently re¬ presented in connection with him. In the codi¬ ces in the guise of God B, he is seen in the pouring rain (Fig. 78, a) or near bodies of water (Fig. 4). Sometimes he appears in the form of the water serpent (Fig. 73, b). Fig. 78, b, shows him associated with the kan or maize sign, and c represents him with leaves at¬ tached to his body and the growing maize plant in his hand. In the form of God K he appears in connection with a sacrifice the apparent object of which 77. — God K in his relation to the serpent and to God B: a, Tro-Cortesianus Codex; b and c, Dresden Codex. 1 1904, p. 32. 2 1894, b, p. 54. 3 1895, b, pp. 216-217. 4 1906, p. 60. GENERAL CONSIDERATION OF MAYA ART. 65 is to obtain good crops. 1 Drawings from other sources than the codices connect¬ ing the Long-nosed God with leaves, flowers and water are very common. A cylindrical terra cotta vase in the American Museum of Natural History bears an interesting design which is represented rolled out in Fig. 79. The principal subject is a head of the Long-nosed God, lacking the lower jaw. Bulblike Fig. 78. The Long-nosed God in relation to rain and corn growing: a and b, Tro-Cortesianue Codex; c, Dresden Codex. objects and flowers that resemble water lilies are attached to the forehead and to the ear plug. Nearby is a curious bird which probably is intended for a pelican. The association of water seems to be pretty clear in this instance. The Manikin Scepter and the Ceremonial Bar are evidences of worship, but offer little information concerning the powers of the object worshiped. On Stela 11 2 at Yaxchilan a human figure wearing a mask with an elaborated nose Fig. 79. — The Long-nosed God combined with flowers and other objects. (Fig. 9) holds in one hand the Manikin Scepter and in the other a club with which he appears to threaten three bound captives who kneel before him. It seems possible that this universal deity may have also been concerned with war. The astronomical significance of the Long-nosed God is not clear. In the cod¬ ices neither God B nor God K seems to be connected with the sun, but such a 1 Forstemann, 1906, pp. 59-60. 1 Maler, 1903, pi. 74, fig. 1 . 66 MAYA ART. Fig. SO. — The Long-nosed God with the sun sign on his fore¬ head: Copan. connection is indicated in Fig. 80 by the kin sign on the forehead of the Long- nosed God in the serpent mouth. The grotesque being here represented holds in his hand a leaflike object. In the phases so far considered it is most signifi¬ cant that the Long-nosed God seems to be entirely beneficent, since death signs do not occur in con¬ nection with him. There is another large group of representations that shows a Long-nosed God with features indis- tinguishable from those of the god just considered, but who seems to be connected unchangeably with death. It seems possible that these figures may symbolize the destructive extremes to which the generally beneficent sky god may sometimes go in causing flood or drought. Or they may indicate a dualism, pure and simple, in which each power for good is directly opposed by a second one for evil. Sometimes when the head of the Long-nosed God of the first type appears in the mouth of the serpent, the death’s head of the second type is attached to the serpent tail (Fig. 81, a and 6). Examples of such an opposition of good and bad are fairly common. A somewhat similar appearance of this Long-nosed God with the attributes of death has already been considered in connection with the Two-headed Dragon. The rear head of this monster, it will be remembered, is characterized by a long nose and by sym¬ bols that have been interpreted as referring to the sun, to water and to death. This head occurs in many situations detached from the body of the monster. Frequently it serves as a headdress for human figures on stelae, lintels and other monuments. 1 Its hieroglyph is perhaps to be seen in Fig. 82, which gives the details of its own peculiar headdress. In Fig. 83 the head of this god with all its attributes is placed in the center of a band of astronomical sym¬ bols, possibly to indicate that the powers of this god are of a heavenly nature. Fig. 84, a to c, furnishes examples of a peculiar object that in the first instance comes out of the end of a Ceremonial Bar where the bust of the manikin god with shield and spear is also featured, and in the next two instances is attached to the head of the other similar Long-nosed God, just considered, with the three signs as a head¬ dress. This second god has in c the kin sign upon his forehead, but in b he has Fig. 81. — Long-nosed death heads attached to serpent tails: a, Copan; b, Yaxchilan. Fig. 82. — Hieroglyphs of the rear head of the Two-headed Dragon: Palenque. 1 For instance, Copan, Stelae H and I, Maudslay, 1889-1902, I, pis. 61 and 63; Palenque, Palace House A, Maudslay, 18S9-1902, IV, pi. 10 and Yax¬ chilan, Lintel 14, Maler, 1903, pi. 55, and Piedras Negras, Stela 3, Maler, 1901, pi. 13. GENERAL CONSIDERATION OF MAYA ART. the sign which Seler 1 considers the general sign for all heavenly bodies. The pendent object with the symbols attached may indicate water descending in a flood. In Fig. 85 we have a somewhat similar object descending from the hands of a priest or deity. The kan or perhaps the imix symbol is seen as well as the sun sign. These representations deserve comparison with the last page of the Dresden Codex, where is depicted, according to Forstemann, the destruction of the world. In this picture a great flood of water gushes forth from the mouth of the composite monster that terminates the band of astronomical signs. Water also descends from the signs of the sun and moon that are attached to the lower side of the astronomical band. An old woman with a serpent upon her head, with crossbones on her skirt, and with jaguar feet inverts a bowl of water. 1 1901-1902, pp. 169-170. 68 MAYA ART. Upon the water that pours out of the bowl is the sign of the unlucky day, Eb, and the sign for zero, or completion. 1 At the bottom of the picture is a black god with the ominous moan bird perched upon his head. The old woman with the serpent upon her head will be reconsidered presently. There are many representations of the Long-nosed God that possess the bleached bone for a lower jaw and sometimes the kin sign on the forehead, but lack the characteristic shell, leaf and saltire symbols. Often heads of this type appear as hieroglyphs with numerals in connection (Fig. 86). Still other ex¬ amples fulfill some unknown function of symbolism or suggestion in the en¬ richment of stelae. Fig. 87 shows a small inverted head, attached to a vinelike object. The lower jaw, as before, is fleshless. When the head of the Long- nosed God appears as a headdress the lower jaw is often lacking, but when such is the case it is often possible to find other symbols of death upon the face. Fiq. 87.— The inverted Long-nosed Death God attached to a vinelike object: Copan. Fig. 86. — The head of the Long-nosed Death God as a period glyph: Co¬ pan. Fig. 88. — Head in serpent mouth: Tikal. Let us examine one more example. Fig. 88 shows a more or less human head enclosed in the jaws of a snake. The nose is not elongated, but the flamelike teeth seem to place this representation among those of the deity we have been considering. Upon the head itself are no death symbols, but immediately below the head we see two bonelike objects as well as another object that almost surely represents a leaf. In this case, then, attributes of life and death seem to be both indicated in connection with one individual. A mass of evidence has been presented in regard to the general conception of a Long-nosed God with constant affinity to the serpent. It has been shown that there are two groups of special manifestations, one of which is good and the other bad. Each one of these groups presents several distinct phases which, however, are found upon fuller investigation to merge into one another. A number of explanations for this state of affairs might be advanced: 1st, that each phase represents a distinct divinity; 2d, that each group represents a di¬ vinity of diverse interests who is directly opposed to the divinity of the other group of interests; 3d, that all the phases and the groups are merely attempts to differentiate the powers of one general and universal god. It is too early to make a choice of these explanations or of others that might be advanced in their stead. The study of primitive religions shows that in gen¬ eral the line of change is from many gods towards fewer gods and finally to one god. Assimilation is a much more common phenomenon than differentiation. Very often one god or a group of gods rises above the others and gradually absorbs 1 This picture is not on the last page in revised numbering. See the Nomenclature, p. 261. GENERAL CONSIDERATION OF MAYA ART. 69 or assimilates the less fortunate rivals. Such may have happened in the case of the Maya. The phases and the groups of phases that have just been described may represent many gods merged into one. The Roman-nosed God. One of the most important gods in the codices is God D, whose face seems likewise to be recognizable in the sculptures. In the codices he takes the form of an old man with a Roman nose and an eye orna¬ mented with a scroll, beneath which are small circles. The corners of his mouth are drawn back and surrounded by deep wrinkles. Sometimes a single tooth projects forward from the front part of the upper jaw, and when this is absent a stub tooth may appear in the lower jaw. But as often as not both jaws are toothless. Frequently he wears a flowing beard. According to Schellhas, 1 God D is a Moon and Night God; Fewkes, 2 as well as Thomas, Seler and Forstemann, consider this figure to represent Itzamna, Fig. 89. The Roman-nosed God as Sky God: a and b, Yaxchilan; c and d, Dresden Codex. while Brinton 3 thinks he is Kukulcan. It can be pretty definitely demonstrated that God D is a universal sky divinity with powers extending over the day as well as the night. lie is not so obviously connected with the serpent as is God B. The weight of evidence seems to incline towards the interpretation of this divinity as Itzamna rather than Kukulcan. Both God B and God D have the strongly deformed teeth which Cogolludo gives as the characteristic of Itzamna. Until the indentification is complete, however, it seems best to employ titles that are without prejudice. Some connection between God D and the sun is indicated by the kin sign that occasionally appears on the forehead of this god. The general connection with all the heavenly bodies is made clear by several passages in the Dresden Codex. In Fig. 89, c, we see the head of God D inclosed in a figure half white and half black that may symbolize alternating night and day. Over the head of the god is the usual sun sign, while above this is a strip of astronomical signs. 1 1904, pp. 22-28. ! 1895, b, pp. 208-216. > 1894, b, p. 56. ' u MAYA ART. A variant of the above appears in d, where the sun sign is placed on the forehead of the god. Let us now turn for a moment to the monuments. Fig. 89, a and b, presents two examples of a very complicated design, occurring several times at Yax- chilan, that resembles the familiar Two-headed Dragon. The legs of the more complete specimen (a) have trifoil scrolls at the joints and cloven feet exactly like those of the Two-headed Dragon figures of the elongated phase shown on page 55. Also the body consists of a band of astronomical symbols The two heads however, are similar to each other rather than strongly differentiated and other heads of more or less human forms are seen in the mouths and else¬ where in connection with the body. On the original monument from which a is taken (Stela 1) ‘ there is a bust of a human being or of a god directly over the center of the planet strip that forms the body of the two-headed monster, and its resemblance to God D of the codices is evident at the first glance. The Roman nose, the open mouth with the lips drawn back, the wrinkles on the cheek, the peculiar tooth project¬ ing outward, the ornamented eye and the flowing hair and beard are all features that occur m the codices in connection with God D. The air of old age is ad- mirably characterized. At either side of this central bust are representations of small human beings. Each of these figures is seated in a device which in one case is circular and pos¬ sibly represents the sun and in the other is crescent-shaped and may represent the moon. Each figure holds in his arms a Ceremonial Bar. At both ends of the Ceremonial Bars appear small faces of the principal deity in the mouths of the serpent heads (Fig. 61). Reverting to Fig. 89, a and b, it hardly needs pointing out that the face which so closely resembles God D likewise appears in the jaws of these two-headed monsters. As if this were not enough repetition, it occurs twice more on the under side of each one of the bodies. The latter examples are worthy of exam¬ ination. The faces look directly downward, but one is the obverse of the other, so that the lower parts of the heads are in conjunction. In a the two heads amalgamate into one, but in b they are separate and easily seen. The two heads of God D of the codices (c and d) that are attached to the planet signs and sun symbols acquire a new significance in the light of these sculptured pictures. It is impossible to state what connection exists between the Two-headed Dragon and the grotesque creature just discussed that is so completely loaded down with the faces that resemble God D. There is some reason, however, for believing that the Roman-nosed God is associated with the front head of this curious monster in somewhat the same way that the Long-nosed God is associ- ated with the rear head. On pages 4 and 5 of the Dresden Codex is represented a scaly green mon- ster with a head at each end (Fig. 90). In the open mouth of the front head is the face of God D. 2 Above this monster are the glyphs of a number of the principal gods, but no glyph that belongs exclusively to the monster itself. This monster may represent the same conception as the Two-headed Dragon that takes so many forms in the sculptures. 1 Maler, 1903, pi. 69. Forstemann, 1906, p. 68. GENERAL CONSIDERATION OF MAYA ART. 71 God D, according to Schellhas, 1 appears as a benevolent deity in the codices, but it seems certain that, like God B, he either has dual aspects or else is directly opposed by another divinity of similar form. In his benevolent appearance he assumes a close connection with maize, but his other powers and relationships are not clear. God E, who is the Maize God, seems to be rather lacking in real power and dependent upon the aid of other gods. He receives this aid from God B as benevolent rain god and God D as benevolent sun god. Fig. 90. —Two-headed monster with face of God D: Dresden Codex. The malevolent aspect of God D — or the evil-minded deity that assumes his form — is seen in the female figure that Schellhas 2 has named by the letter I and which he further describes as a destructive Water Goddess. It may be remarked at this time that sex seems to be a shifting and uncertain attribute among the Maya gods. Although generally masculine, nevertheless the prin¬ cipal deities sometimes assume female form. The Goddess I has already been brought before the reader in connection with the destructive flood symbols discussed on page 67. It seems clear that this deity has some affinity to the composite monster with the astronomical symbols on its narrow body that appears at the top of page 74 of the Dresden Codex. The floods that issue from the body of this monster are augmented by the water from an inverted bowl that the goddess holds in her hands. Goddess I appears several times more in conjunction with gushing streams of water both in the Dresden Codex and in the Tro-Cortesianus, and shows an indefinite relationship to God D. Her hieroglyph is uncertain, but the hieroglyph of God D is used in one instance where her picture occurs. The physiognomy of this goddess resembles strikingly that of God D. A constant and peculiar feature is a headdress consisting of a knotted serpent. Perhaps the natural opposition intended to be conveyed by God D and God¬ dess I is that which exists in nature between the clear sky in which appear the sun and stars, and the black storm clouds which blot out these orbs and deluge the earth with destructive floods. Fewkes, who offers much evidence of the close connection existing between Gods B, D and G, seems to be inplined to accept 3 the relationship implied in the Dresden Codex between God D and Goddess I. To return to the sculptures, the face of the Roman-nosed God —which is ‘ 1904 ’ PP- 22 - 23 - ’ 1904. PP- 31-32. . 1895, b, p. 210. I 72 MAYA ART. perhaps a safer title than God D for the general appearance of this divinity — appears frequently in the serpent mouths that terminate the Ceremonial Bar. Reference has already been made to one example of this at Yaxchilan. At Copan good examples are seen on Stelae P, 2 and I, and at Naranjo on Stelae 6, 7, 20 and 32, while numerous other citations from different cities might easily be given. In fact, it may be stated with some assurance that the heads on the Ceremonial Bar are nearly equally divided between representations of the Long- nosed God and of the Roman-nosed God. While the extreme types are clear and well fixed, the two types of heads blend into each other by almost insensible gradations. Still another manifestation of the Roman-nosed God is probably seen in the face form of the kin glyph, which, as every one knows, is the period glyph of the lowest order in the calendarical f 9 Fig. 91. — The kin glyph and the Roman-nosed God aa Sun God: a, b, e and g, Copan; c, Yaxchilan; d, Chichen Itza; /, Palenque. inscriptions, representing one day. If this god is, as we surmise, a god of both night and day but with the idea of the sun god uppermost, his face would serve nicely as a sign for the period, one day. The kin glyph is fairly uniform, examples from diverse monuments and cities being given in Fig. 91, a tod. Some¬ times the kin sign appears on the face, usually the nose is of the Roman type, a peculiar terraced tooth that is commonly described as filed projects from the front of the upper jaw, and a flowing beard is often present. The eye likewise shows similarities to the eye of the god we have been studying. The so-called normal form of the glyph simply abbreviates the above-described face to the kin sign and the flowing beard (Fig. 91, g ). This glyph occurs in other situations than the initial series inscriptions. It is one of the common glyphs in the so-called supplementary series that follows the initial series. The face numeral for four, which occurs only in the inscriptions, is likewise probably a variant form of the Roman-nosed God. The kin sign is usually placed on the cheek in front of the ear ornament. The face numeral for fourteen is similar except for some indication of death such as a bone for the lower jaw. The glyph of the month Yaxkin, in at least one instance 1 in the inscriptions, also offers evidence of the use of the face in question. One of the common astronomical signs shows the face of the Roman-nosed God curiously conventionalized. This has been interpreted by Seler 2 as a symbol of the sun. Fig. 91, /, gives a portion of the strip of planet signs at the base of the Tablet of the Sun at Palenque. This strip ends in the god's face in the phase most common as the day period glyph. Back of this face is a rectan- 1 Bowditch, 1910, pi. IX, 9. 2 1901-1902, p. 165. GENERAL CONSIDERATION OF MAYA ART. 73 gular panel containing a sign that resembles Caban and has been interpreted as referring to the moon. Next to this sign, and alternating with it during the entire course of this strip of symbols, is the conventionalized face of the Roman¬ nosed God looking upward. This type of face occurs many times at Yaxchilan, Piedras Negras and Palenque. A beautiful example greatly enlarged to frame in the entire side of a niche for a seated figure at Piedras Negras is given in Fig. 92. While discussing the homogeneity of Maya art (pages 20-21) there was presented a series of strips of astronomical symbols, some of which are combined with bird heads. These bird heads are in profile at the ends of the strips (Fig. 6, g). Later, while elucidating the development of the Two-headed Dragon, the occasional presence of the Serpent Bird upon the central part of the elongated body was noted (page 61). It seems possible that some connection may be established between these bird heads and the multiform Roman-nosed God, but the results are ambiguous. The face of God D certainly appears on birds in the Peresianus Codex. 1 The so-called Serpent Bird on Stela 5 2 at Piedras Negras presents a face of the same type. The front view example of the Serpent Bird from Palenque (Fig. 57, a) likewise shows similar features. An interesting sculpture from the northeastern group of mounds at Copan is sketched in Fig. 93. This sculpture is an abbreviated form of the Two-headed Dragon. The front head is true to type, while the rear head has the features of the Long- nosed God, but lacks the usual symbol of the shell, etc., on the forehead. In place of the monster’s body is a somewhat damaged face in front view that probably represents the Roman-nosed God. An analogous design is found as the headdress of the principal figure on the north side of Stela N 3 of the same city. But these examples, after all, may not indicate the survival of the Serpent Bird that in other cases is perched upon the body of the Two-headed Dragon. Nearly all the animal altars of Copan of the Two-headed Dragon type are elaborated by the intrusion of grotesque heads between the legs and upon the back. The real explanation of these anomalous conditions may be artistic exuberance rather than complication of religious ideas. The representations of the Roman-nosed God that are executed in the full 1 L4on de Rosny, 188S, pis. 4 and 8. ! Maler, 1901, pi. 15, fig. 2. 3 Maudslay, 1889-1902,1, pis. 77 and 79. Fig. 92. — Con- ventionalized head of Roman- nosed God: Piedras Negras. 74 MAYA ART. round show the same features as the more common profile studies and at the same time make possible the identification of the front view representations in low relief. Small heads on the belts of Stelae I and H at Copan likewise show the transition between profile and front-view faces. A large block of stone on the northeast corner of Mound 16 at Copan 1 is carved into an excellent por¬ trait of the divinity. Upon the forehead is the usual kin sign. A tassel-like nose ornament hangs down over the mouth, and at either side of the latter is a deep crease or wrinkle. Another similar head adorns the Jaguar Stairway. 2 In the Peabody Museum are a number of excellent original carvings from Copan that represent the Roman-nosed God with the terrace-shaped tooth (Plate 26, fig. 1). In the same collection there are several heads with the twisted or cruller¬ shaped ornament over the nose. Examples of faces of this type have already been presented in Fig. 1, and the wide distribution has received due notice. A re-examination of these faces will bring out many points of resemblance to one or another of the phases of this important god. It seems possible that this defi¬ nite manifestation may refer directly to the sun disk, the Tonatiuh of the Nahua. Seler, 3 however, considers the face with the twisted nose ornament to represent the god of the Evening Star. The face glyphs for seven and seventeen show this god in profile with the twisted ornament in view. On one of the wooden lintels at Tikal 4 is carved a human figure with this nose ornament and with the number seven on his cheek. Examples of the face glyphs for seven and seventeen agree with the face on this tablet in the matter of the twisted ornament over the nose. The sign for seventeen has in ad¬ dition the fleshless jaw bone, as is usual for numbers above ten. The twisted element appears also on heads in the Ceremonial Bar, an instance being Stela 2 at Copan. Many other representations, in the light of the variations and developments that have been noted, seem to represent the general- Fig. 94.—Pottery flask with face ized Roman-nosed God. Among of^Roman-nosed God: Uloa othe rs may be mentioned the atlantean figures that support the altar as well as those on whose backs stand the officiat¬ ing priests on the tablet of the Sun Temple at Palenque. 5 The old man smoking a tubular pipe from the same city is another case in point. 6 The same face stamped upon a pottery flask in the Peabody Museum is shown in Fig. 94. Let us now consider briefly some of the indeterminate Fl ° 95 ~ Jadeite sl “ b representations that lie between the Long-nosed God and God: Ocosingo. the Roman-nosed God. Fig. 95 is a drawing of a thin jadeite plate from Ocosingo which is alike on the two sides and has some features indicated by stencil-like perforations and others by low relief carving. The diagonal cross on the fore- 1 Maudslay, 1889-1902, I, pi. 10, b. * Maudslay, 1889-1902, III, pi. 73. 2 Maudslay, 1889-1902, I, pi. 18, a, and Stephens, s Maudslay, 1889-1902, IV, pi. 88. 1841, I, p. 143. ■ Maudslay, 1888-1902, IV, pi. 72. s 1902-1903, I, p. 317. I GENERAL CONSIDERATION OF MAYA ART. 75 head probably represents the sun symbol. The nose is slightly broken, but it is evident that it never projected much farther than now. It is impossible to say with assurance whether this face represents the Long-nosed God or the Roman¬ nosed God. In Fig. 9G we see eight heads. Some of these, such as a and e, are good examples of the Long- nosed God; others, such as d and h, represent the opposite conception. The remaining heads are much more am¬ biguous, although it seems likely that b and f fall in the Long-nosed group and c and g in the Roman-nosed group. The heads which terminate the upturned noses of elabo¬ rated serpents (see Fig. 26 and Fig. 33, b and c) almost all belong to an ambiguous middle series. In fact, Fig. 96, b and g, already dis¬ cussed, are found in such positions. Fig. 96. —Intermediate series. Long-nosed God and Roman-nosed God: a~d, / and h, Copan; e, Palenque; g, Tikal. Aside from these intermediate types there is abundant evidence of a close connection between the two generalized forms. Fig. 97 shows two drawings from the Dresden Codex in which the two deities are combined. In a we find God B, the principal phase of the Long-nosed God, seated upon the head of God D, the princi¬ pal phase of the Roman-nosed God. This latter head is marked with bunches of circles which have been interpreted as water symbols. In b, on the other hand, we see God D, with a sun symbol in his hand, wearing the head of God B as a headdress. We have observed that both the Long-nosed God and the Roman-nosed God occasionally bear the sun symbol. Now the Sun God, par excellence, of the Dresden Codex is God G according to the classification of This god takes the form of an old man with a Roman nose and a body marked with sun symbols (Fig. 98). The feature that distinguishes him from God D is an ornamental hook that is attached to his nose. It is possible to see in this characteriza¬ tion a sort of compromise between the natural nose of God D and the fantastic serpentine nose of God B. Schellhas 1 finds that God G is closely connected with God B. Fewkes 2 goes farther and groups Gods B, D and G together. God N may also belong here. 1 1904, p. 28. 5 1895, b, pp. 216-218. Fig. 97. — Close association of God B and God D: Dresden Codex. Schellhas. Fig. 98.—God G, the Sun God: Dresden Codex. 76 MAYA ART. In closing this discussion it seems best to reiterate the most significant point of all. Both the Long-nosed God and the Roman-nosed God are distinct enough in their general appearances and yet each blends into the other. Moreover each divinity is presented in a number of phases, which at first glance seem to be distinct and characteristic, but which upon further examination are all found to break down into a most chaotic state. The best explanation that can be offered is that the two most important gods in the pantheon became more and more important, and absorbed and assimilated their less powerful rivals. Then, too, the artistic importance of the serpent undoubtedly led to convergent evolu¬ tion of many forms. Other gods less closely allied to the serpent will soon be presented. Other Subjects The Jaguar. This animal received a great deal of attention from Maya artists and possessed a religious importance secondary only to the serpent. Many of the headdresses and breastplates represent the face of the jaguar. A fine b Fig. 99. — Altar of Stela F: Copan. series of such breastplates are shown on Altars Q, L and T, and on the sculp¬ tured interior step of Temple 11 at Copan. The Temple of the Jaguars at Chi- chen Itza gets its name from a frieze representing jaguars in a procession. The Jaguar Stairway at Copan is another notable occurrence of this animal in arch¬ itectural design. The jaguar seems to have served in a general way as a model for some of the animal altars of Copan and Quirigua. Altar G at Quirigua (Plate 1, figs. 1 and 2) presents a particularly fine jaguar head. Altar F at Copan consists of several carved stones fitted together that represent two jaguars bound to the side of two grotesque heads. Two views of this altar are reproduced in Fig. 99, a and b. A possible combination of the jaguar and the serpent in monstrous and grotesque creations may be seen in some of the altars of the Two- headed Dragon type. In particular the clawed feet seemed to be derived in some instances from this animal. Double-headed jaguar seats are represented at a number of sites, particularly Palenque 1 and Uxmal 2 and Chichen Itza. 3 Perhaps the finest example was the throne modeled in stucco upon the Tablet of the Beau Relief at the former city. This famous panel is now destroyed, but its original features are preserved more 1 Maudslay, 1889-1902, IV, pi. 44. 2 Stephens, 1843, I, p. 183. 3 Maudslay, 1889-1902, III, pi. 35, b, and pi. 50. GENERAL CONSIDERATION OF MAYA ART. 77 or less exactly in the drawing of Waldeck. 1 Jaguar seats occur at Chichen Itza on the carved lintels. The Chacs or Rain Gods of the Four Quarters were conceived in the form of jaguars, and the Balam or Jaguar Priests were an important religious institu¬ tion among the Maya. Jaguar priests which may or may not correspond to the Balam priests are represented upon Stela A 2 at Quirigua, Stela 10 3 at Piedras Negras and Stela 8 4 at Siebal. They are in human form and dress with the ex¬ ception of the hands and feet, which have the claws and markings of the jaguar. Many references to the jaguar occur in the codices, but their meanings are uncertain. Fig. 100 reproduces a drawing in the Dresden Codex where the ani¬ mal is realistically represented. The only unnatural feature is a flower resembling a waterlily that is attached to the fore¬ head. A jaguar design engraved upon a vase from Peto in northern Yucatan is figured in a later section of this paper (Fig. 185). This remarkable specimen shows a jaguar sitting in a floral circlet and wearing a cloak and breech cloth, not to mention arm and leg bands, nose plugs, etc. His headdress consists of the head of the Long-Nosed God and a small flower similar to that shown in the Dresden Codex specimen. A painted potsherd from Copan (Fig. 101) presents an analogous drawing of an elaborately dressed jaguar, with a so-called speech scroll issuing from his mouth, who wears over his forehead a leaflike ornament. The same leaflike design occurs again on one of the lintels of Tikal.* It seems possible that some connection may be established between the water plant and fish motive (page 18) that has already been de¬ scribed and this powerful beast of the jungle. The jaguar skin is frequently represented as a garment. Skirts showing the typical jaguar markings prevail on the figures at Copan and are common elsewhere. Sometimes the entire skin 6 with head and tail attached is repre¬ sented as thrown over the shoulder or about the waist. Birds and Feathers. The ceremonial and artistic importance of birds and feathers in Maya art can hardly be overestimated. Repre¬ sentations of the former occur in the glyphs and codices and upon the sculptured monu¬ ments in connection with some of the more recondite and peculiar features of religion and design. Feathers form a common motive for decoration on stelae and the fa¬ cades of buildings and are, as well, an integral part of the gala and everyday dress of the people as represented by sculptures and frescos. The Feathered Serpent and its reciprocal concept the Serpent Bird have Fig. 100. — Realistic drawing of a jaguar: Dresden Codex. 1 Waldeck, 1866, pi. 42. 2 Maudslay, 1889-1902, II, pi. 8. 3 Maler, 1901, pi. 19. 4 Maler, 1908, a, pi. 7. 5 Maudslay, 1889-1902, III, pi. 71. 6 For a good example see Maudslay, 1889-1902, IV, pi. 72. MAYA ART. already been discussed at some length. It was indicated that both simply form a general basis for a large part of the peculiarly involved art of this people and that further definite characterization was accomplished by adding specific details of one kind or another. It was pointed out that several species of birds (judg¬ ing by the head) might have the unnatural feature of a conven¬ tionalized serpent jaw lying along the wing, and that this unnatural feature was the only fixed characteristic of the so-called Serpent Bird. It was shown that this peculiar wing might even occur sep¬ arately in the device known as the Wing Panel, and that, as such, it frequently served as lateral ear or¬ naments for the more complicated figures on the monuments. The natural charac¬ ters of birds are some¬ times clearly given,but more often the repre¬ sentation is vague and grotesque. Many bird faces approach now the serpent and now the human type. Most of the more elaborate spe¬ cimens have ear plugs, nose plugs and teeth. The teeth are of the same two kinds as seen on the serpent jaws, namely, a rather realistic molar and a curious flame-shaped incisor usually divided into two parts. Very often the curled object at the back of the mouth likewise appears. Among the birds represented in the Maya codices Drs. Tozzer and Allen 1 have identified the following: herons, probably of several species, frigate bird, ocellated turkey, king vulture, black vulture, harpy eagle, Yucatan horned owl and screech owl, coppery-tailed trogan or quetzal, blue macaw and perhaps a few others. The bird reproduced in Fig. 79 doubtless represents the pelican, as 1 1910, pp. 324-346. See also Seler, 1909-1910, pp. 427-457, 784-846. Fig. 102.— Various representations of birds: a-c, Copan; d, Codex Borgia; e and k, Palenque; / and j, Codex Peresianus; g, Dresden Codex; h, Quirigua ; i, Codex Tro-Cortesianus. GENERAL CONSIDERATION OF MAYA ART. 79 may be seen from the character of the greatly enlarged bill which shows the knot that appears during the mating season. The length to which the Maya artist would go in representing a single spe¬ cies is shown in Fig. 102, a—c. Of these, a is a glyph carved on the back of Stela B at Copan, the first part of which gives the head of the blue macaw, while b is a sculpture in the full round representing the same bird and coming from the same city. Note the nostril at the top of the bill, the eyes surrounded by a cir¬ clet of small knobs as well as the hook-shaped appendage to the base of the eye, likewise composed of knobs. In c the short lower bill and the tongue are omitted and a more or less human ear with characteristic decoration is introduced at the side of the face. The upper bill is lengthened and enlarged. This last figure occurs twice on the front of Stela B at Copan and has often been explained as an elephant trunk. The true explanation has been worked out independently by a number of students. 1 Drawing d reproduces a bird with a similar head from one of the Mexican codices. Birds, either entire or in part, are frequently found on the more elaborate headdresses of the priests and warriors on the monuments. An interesting exam¬ ple from Palenque is given as Fig. 102, e, which represents a heron with a fish in its mouth. The same idea with only the upper part of the bird in view appears at Seibal on Stela 10 (Plate 25, fig. 2) and a second variant larly associated with death by the Maya, is characteristically represented in the drawing reproduced in Fig. 104. In this case the sprouting maize plant appears as a headdress, the association probably indicating a failure of the crops. The glyph of this bird occurs very frequently in the codices, usually with evil intent. The vulture and the harpy eagle are also represented with features 1 Parry, 1893, p. 166. Gordon, 1909, pp. 193-195. Tozzer and Allen, 1910, p. 343. 1 For example, pp. 16-18. 80 MAYA ART. drawn from the figure and ornamentation of man. The possible connection be¬ tween the Roman-nosed God and a bird of some sort has already been mentioned. Bird heads are prominent in the hieroglyphs, but the features are usually modified towards the serpent or the human type. In particular the higher period glyphs, including the cycle, katun and tun, commonly show birdlike hooked noses. As is well known, these glyphs are of two types, the face type and the so-called normal type. With the latter we have at present no concern. The face type is the more usual, but it is very difficult to determine the signifi- Fig. 105. — Highly modified birds as period glyphs: a, d and g, Stela D, Copan; b, e and h, Stela D, Quirigua; c, / and i, Altar B, Quirigua. cant feature from which the meaning of the face is drawn. Fortunately there are four examples of initial series which show bodies as well as heads. The earliest and clearest of these is on Stela D at Copan. On this monument each number is repre¬ sented as a human being who carries the period upon his back. Two other in¬ stances of full-form glyphs occur on Stela D at Quirigua and the fourth upon Altar B at the same city. These latter examples of picture writing show a contest be¬ tween the being who represents the number and the one who represents the period. The complexities and involutions of these sculptures are almost beyond solu¬ tion. In the series given here for comparison only the period forms are shown. The cycle according to the Copan example is represented by a parrotlike bird (Fig. 105, a), but in two Quirigua drawings (b and c) this resemblance is lost. In all three the lower jaw consists of an open hand with thumb pointing GENERAL CONSIDERATION OF MAYA ART. 81 forward, but in the first example this feature is somewhat disguised. In the Quirigua specimens the serpent head on the wing greatly complicates the design. It will be seen at once that this is the adventitious feature that has already been discussed under the caption, the Serpent Bird. Unfortunately nothing appears in these drawings to fix the species of bird unless we accept the suggestion of the parrot offered by the Copan example. In the simple face forms of the cycle glyph, as, may be seen in the series given by Mr. Bowditch, 1 the hand that replaces the lower jaw is a fairly constant feature. Most of the faces have beaked noses, but, aside from this and the hand just mentioned, they are ex¬ ceedingly divergent. Fig. 105, d, e and / give the bird figures that appear in the katun glyphs, and g, h and i those of the tun glyphs. Both these series are unintelligible as far as definite interpretation is concerned. The sub¬ ject of the former may indeed be an eagle, as suggested by Mr. Bow- ditch. The subject of the latter is in all cases an extremely grotesque bird. The bird beak is pretty clearly shown in many of the abbreviated glyphs, and in these the frequent presence of a peculiar ornament in front of the forehead and back of the ear suggests the Yucatan screech owl or Moan bird. This bird is closely connected with the idea of death. In harmony with the interpretation is the bone that appears on the lower jaw of many of the heads where the bird element is wanting. Perhaps the clearest and most consistent use of a bird head is in the hiero¬ glyph for the month Kayab. 2 This glyph has been explained as the head of a turtle, but a careful comparison of all the forms shows that it really represents a macaw. The short under bill and the tongue are clearly marked, as well as the nostrils at the base of the bill. Often the feathers at the back of the head can be easily distinguished. ■ 1910, pi. 12. Fig. 106. — Examples of sculptured feathers: a, b, d, e, f, g, Copan; c and i, Piedras Negras; h, Seibal; j, Uxmal. 2 Bowditch, 1910, pis. 8 and 10. 82 MAYA ART. Fig. 106 shows typical examples of the use of feathers in decoration. In a is given the first appearance of independent feather-work decoration on a Copan stela. Previous stelae in this city do not show feather drapery, although it subsequently was magnificently developed, as may be seen from details given in d and e. The handling is very free, and frequent use is made of a sort of rosette that loops or binds the feathers together. Long plumes with these cir¬ cular ornaments occur widely on headdresses (Copan, Tikal, Ixkun, Yaxchilan, Kabah, Chichen Itza, etc.) and in simplified form are used as motives for mould¬ ings on the facades of buildings from Copan to the cities of northern Yucatan. Feather drapery without such binding is likewise common. The flowing head¬ dresses of the Seibal stelae furnish good examples, as may be seen from h. This free feather-work is represented in architecture by the example figured in j that occurs, according to Stephens, 1 on the House of the Birds at Uxmal. Occasion¬ ally attempts were made to give the finer details of feathers, as appears in c, which may represent a turkey plume. Notched margins are likewise seen on some representations of long feathers occurring on the monuments. Feather cloaks and aprons are commonly worn by the elaborately attired figures on the stelae. Specially beautiful examples of these are found at Piedras Negras and Naranjo. 2 The apron of Stela D at Copan is reproduced in b, while in i is given the more elaborate one of Stela 7 at Piedras Negras. One of the most noteworthy features of feather-work is the service it pays to composi¬ tion. The Maya artists frequently balanced their designs by sweeping plumes. Fig. 107 illustrates the use of feathers in filling up cor¬ ners. The feather-work of the Maya never becomes stiff and heavy. The curves are those natural to drooping feathers and quite in con¬ trast to the rather tortuous curves derived from the serpent. Miscellaneous Animals. Besides the classes of figures that have already been considered, examples may be given of many other animals, of bats, of fish and shells and of plants. Some of these play very minor roles and are presented without elaboration. Others were apparently of some religious significance and show modifications towards anthropomorphism. It is possible that some of the latter were of ancient totemic importance. _ According to 1' orstemann, 3 the snail denoted the winter solstice and the tor¬ toise the corresponding period of the summer. Both are represented in the codi¬ ces and the sculptures, but under somewhat different conditions. The hiero¬ glyph for the month Kayab, in which the summer solstice falls, is explained by h orstemann as the head of the tortoise, but it seems almost certain that this head represents the blue macaw, as may be seen by the spiral hook under the eye, the dotted circle around the eye and the nasal opening at the upper part of the bill (compare the typical glyphs 4 with Fig. 102, a and b). The tortoise Fig. 107. — Head showing manipulations of feathers to fill out corners: Yaxchilan. 1 1843, I, p. 311. 1 Maler, 1901, pi. 21, seated figure, and Maler, 1908, b, pis. 30, fig. 2, and 35, fig. 2. 1 1902, p. 27, and 1906, p. 161, etc. 1 Bowditch, 1910, pis. 8 and 10. GENERAL CONSIDERATION OF MAYA ART. 83 clearly occurs in connection with astronomical signs in the codices, 1 but its exact significance is open to some doubt. The snail does not seem to have any clearly defined meaning of an astronomical nature. The Altar of Stela C at Copan is carved in the form of a tortoise or turtle, 2 but the so-called Great Turtle Altar at Quirigua is not properly a turtle but an ex¬ tremely elaborate exam¬ ple of the composite ani¬ mal aready described as the Two-headed Dragon. Small turtle carvings form a sort of frieze on the House of the Turtles at Uxmal. The most striking use of this reptile in art is seen at Chichen Itza, where it reaches the anthropomorphic stage. Here the carapace of the turtle incloses the middle part of the hu¬ man body. This figure may be studied on the fagade of the Iglesia, where it is represented in high relief, and on columns and jambs of the Castillo, the Temple of the Tables and other buildings, where it oc¬ curs as an atlantean or caryatid motive in low relief. The turtle is drawn realistically on the piers of the Lower Chamber of the Temple of the Jaguars. The snail, so called, is represented in combina¬ tion with human form Fig. 10S. — Shells and figures associated with them: a, Codex Peresianus; b, Chama; c and d, Dresden Codex; e, Tikal; /, Codex Nuttall; g, Codex Borgia; h, Palenque. much more often than the tortoise, and occurs not only on the buildings at Chichen Itza just referred to, but also in the codices and on objects of minor art such as pot¬ tery. The word “snail” is commonly used, but there are no means of telling whether the shell represented belongs to the snail or to some other mollusk. Ac¬ cording to Tozzer and Allen 3 the shell is probably that of the Fasciolaria gigantia, which is the largest known American shell and is found along the coast of Yucatan. In Fig. 108 are given a number of representations of the human form com- 1 For instance, Tro-Cortesianus, p. 37; Peresianus, pi. 24. 2 Gordon, 1896, p. 40. 3 1910, p. 296. 84 MAYA ART. Fig. 109.—God N: Dresden Codex. bined with a shell. The first example is from the Peresianus Codex, and shows the personage which Schellhas 1 calls God N, the God of the End of the Year. Seler, 2 however, names him the Old Bald-headed God, and suggests that he gov¬ erned the moon. He is probably related to God D, the principal Roman-nosed God. Usually, but not always, this God N wears a large shell from which the upper part of his body seems to emerge. A draw¬ ing of this god from the Dresden Codex without the shell append¬ age is given in Fig. 109. It is worthy of note that the tun glyph or year symbol often appears on the headdress. Mr. Dieseldorff excavated near Coban on the highlands of Guatemala several pieces of pottery that have painted or incised representations of the Shell God: one of these is presented in Fig. 108, b. Sculp¬ tured figures on buildings at Chichen Itza often have a shell attached to the body (Fig. 110). Differing somewhat from these is the small childlike figure sitting under water with his feet in a shell (Fig. 108, d ). Among the Nahua the snail was commonly associated with birth and death, as in the drawing from a Mexican codex that is reproduced in Fig. 108, g. In the lower, right-hand corner of the Tablet of the Foliated Cross at Palenque is a shell ( h ) in which is parti- _ ally concealed the Long-nosed God. From the hands of this god issues a plant amid the leaves of which is a face resembling that of the Maize God. The shell in this con¬ nection probably appears as an indication of water. The shell that is shown in top or profile view on the head¬ dress of the rear head of the Two-headed Dragon needs no further comment. It probably is a sign indicating water, and apparently has no connection with the so- called Snail God. So much for this conflicting evidence concerning shells combined with the human body. Representations of shells occur as details of dress on many of the stelae. A common girdle ornament is given in Fig. Ill, the principal part being a central group of three pendants probably cut from large shells. Smaller shells, resem¬ bling the olive shell, some¬ times form a fringe at either side, as in b and c. The head of the leaf-nosed bat makes the hieroglyph for the month Zotz, 3 which means bat in the Maya language. Upon the back of Stela D at Copan this month glyph (Fig. 112, a) is given its full form and the membranous wings clearly represented. The Bat God was probably a deity of considerable importance 4 both among the Maya and the surrounding nations. Elsewhere at Copan he is represented with a human body but with the same upturned nose (Fig. 112, b and c). The more or less humanized figure of the bat with wings outstretched occurs as a painted decoration upon pottery from the Uloa Fig. 110. — Man with shell at¬ tached: Chichen Itza. Fig. 111. — Shell girdle ornaments: a b, Ixkun; c and d, Copan. Palenque; 1904, pp. 37-38. 1 1902-1908, III, pp. 593-595. 1 Bowditch, 1910, pis. 7 and 9. 1 Seler, 1902-1903, pp. 112-115. GENERAL CONSIDERATION OF MAYA ART. 85 Valley 1 and from the highlands of Guatemala. 2 A fine example is also found on the remarkable stucco reliefs of Acanceh in northern Yucatan. Animals such as the deer, the dog, the peccary, etc., are usually represented with little variation from the natural form. The first-mentioned occurs rather frequently in both the sculptures and the manuscripts (Fig. 113). The dog is given a ceremonial importance in the codices. From the body markings it has been determined that a domesticated species is repre¬ sented. Other animals are not of enough interest to be presented in a detailed study at this time. The principal occurrence of vegetable life has already been noticed under the discussion of the fish and water-plant motive. Fla - 1 |, 3 ' — Realistic drawing of a deer: Dresden Codex. Bones and Death. The frequent representation of death and its attributes is responsible for the more gruesome aspects of Maya art. Symbols of death are found everywhere in the codices and sculptures. Bones and death’s heads even occur as motives for architectural embellishment. Maya religion seems to have been strongly dualistic and to have been concerned with the unceasing conflict between good and evil, life and death. If we may credit the evidence of the art, repeated many times over, the Death God rode supreme over all the other deities. Perhaps the explanation is that death and destruction were within the sphere of every god if he cared to extend his powers beyond a given point. The dual natures of the general divinities called the Long-nosed God and the Roman-nosed God have been explained by examples. The power of the Death God Fig. 114.—Represents over good crops and over women in childbirth is strikingly p?ed°L h N™grM aCnSoe: re P resen t e d in the codices. Propitiation against death was accomplished at the cost of life. Although human sacrifice was not so excessive as among the Mexicans, still it existed and is clearly represented on page 3 of the Dresden Codex as well as in Fig. 114 from Piedras Negras and in a number of cases at Chichen Itza. The usual representations of skulls, skeletons and separate bones show some curious and characteristic features. A typical collection is given in 1 Gordon, 1898, a, pis. 1, fig. 11, and 3, b. 3 Dieseldorff, 1894, b. 86 MAYA ART. Fig. 115. In n, b, d and h are shown long bones. As a rule, they have knobbed ends, the knobs being two or three in number. On the enlargement at each end of the bone are usually two circles or crescents, while a wavy line runs along the middle of the shaft. Most ear and nose plugs as they occur on the monuments of the southern Maya cities appear to represent bones. In g the ear plug of an old man (the Roman-nosed God?) upon a tablet at Palenque shows three bones Fig. 115. — Various representations of bones and death: a, Chichen Itza; b, c,f, i and j, Copan; d, e and n, Tikal; g, k and p, Palenque; l and o, Uxmal; m, Dresden Codex; q, Mexican codex. bones are somewhat conventionalized as follows. An encircling band is first placed be¬ low the enlargement at the end of the bone (h and e ), then the head of ' the bone, already pos¬ sessing the small circles or crescents, is modified into a simple face. Such little faces are given in / from the headdress of Stela H and in i from the nose plug of the large con¬ ventionalized face on the back of Stela B, both at Copan, while k and p represent similar devices from Palenque. Long bones as motives for architectural decor¬ ation are given in n and o, the first from Tikal and the second from the environs of Uxmal. The characteristic death head is usually grotesque rather than strictly realistic, as may be seen from c taken from Altar R at Copan. Here death is indicated principally by the presence of circles, ovals and crescents on the forehead, the jaws, etc. The eyes show little modification, while the nasal cavity of the skull is inadequately represented. The setting of the teeth in bone rather than in the gums is indicated by the use of double and triple outlines. Although the face seems intended to represent a bleached skull, yet the ears are drawn in full flesh. Similarly in many other delineations the hands and feet are represented entire, while the legs, arms and other parts are bare bone. The torso of a skeletal figure from Uxmal that is now in the American Museum of Natural History is GENERAL CONSIDERATION OF MAYA ART. 87 Fig. 116. — DeatL God: Dresden Codex. reproduced in l. Often bodies are represented practically entire yet with cer¬ tain symbols which indicate death. A device resembling a percentage sign and often called the maggot symbol surely indicates death. Dotted lines connecting small circles as well as black spots and closed eyes appear to do the same. The Death God has been called God A by Schellhas, 1 who gives full information concerning his attributes and associations. An elaborate representation of the Death God seated on a throne made of bones is given in m of Fig. 115. Note the use of lines or circles and dots along the limbs and the full-fleshed hands and feet. A frequent characteristic of the Death God is a spiny back, made so by projecting vertebrae (Fig. 116). The hieroglyphs of the Death God have been definitely determined. They are found in many places where the figure is absent and seem to indicate misfortune and failure. The attributes of the Death God appear in con¬ nection with many conceptions represented on the monuments. In particular the rear head of the Two-headed Dragon has, as we have seen, a bone for the lower jaw. Many other figures of the Long-nosed God are also characterized by this gruesome feature. On Stela I at Copan the double-headed serpent that forms the Ceremonial Bar is a thing of dry bones, as may be seen from Fig. 115, j, which reproduces part of it. The heads in the mouths of this object represent the Roman-nosed God. The importance of the death element in the hieroglyphs is easily illustrated. The day Cimi and the number ten are represented in the four ways shown in Fig. 117: a is the face of a dead person characterized by the closed eye, b is the so-called maggot sign, c is a face bearing this sign and d is a skull. The face glyphs for numbers from 11 to 19 are often merely the faces from 1 to 9 with added death sym¬ bols such as a bone for the lower jaw. The same feature usually occurs in the face forms of the tun glyph. Grotesque Figures. Figures which ac¬ cording to European standards would be termed grotesque occur frequently in Maya representative art. In some cases the gro¬ tesque character seems to have been taken seriously as a means of expressing a supernatural quality. Most of the representations of gods in the codices are grotesque in many of their features. The manikin figure derived from the serpent is certainly grotesque, and by this grotesqueness the reptilian nature of this god in human form is made evident to anyone. The methods used by Maya artists to produce grotesque figures and effects are much the same as prevail elsewhere. Some are true composites, while others show purely fanci¬ ful exaggeration. The elaborate bird forms in some of the initial series inscriptions have al¬ ready been commented upon. Very similar forms appear on various parts of the so-called Great Turtle Altar at Quirigua (Plate 2). At first it seemed possible that some cryptic inscription was contained in these figures, but the possibility of this is slight. Examples of these grotesques have been given in Fig. 32. 1 1904, pp. 10-15. Fm. 117.—Glyphs indicating death: a and b, Dresden Codex; c, Quiri¬ gua; d, Copan. 88 MAYA ART. Conscious manipulation on the part of the artist is shown in faces which might be described as reversible. Fig. 118, b, presents such a face from the Great Turtle Altar. As it stands, this represents a grotesque human face with a knob nose, a protruding tongue and a fringe of beard. When this face is inverted, it be¬ comes an elaborate but typical bird head. A small figurine of a fish attached Fia. 118. — Grotesques: a, head with three faces, Palenque; attached to shell': b, inverted bird’s head, Quirigua. Chajcar. to a shell is shown in Fig. 119. A very good likeness of the Long-nosed God may be seen along the back, facing in the opposite direction, the posterior dorsal fin of the fish forming the nose of the god. The breast ornament of Stela N of Copan is also a reversible face. Maudslay 1 has commented on the straps that hang from the girdles of stelae at Copan. In a number of cases the wrapping at the top of the straps forms part of an inverted head (Fig. 120). Facial expression is more highly developed in grotesque than in realistic sculpture. The elaborated initial series glyphs, already referred to, have many figures showing grimaces and other violent distortions of the face. Maize God. The Maize God which Schellhas 2 has termed God E usually bears upon his head the kan sign, which repre¬ sents a grain of maize. To the latter is ordinarily attached a growth of some sort that may either represent an ear of maize surrounded by leaves or a young sprout. The simple kan sign is basis of the glyph for the day, Kan (Fig. 121), and likewise occurs in many other situations. Offerings of maize cakes are indicated ^ IG 8trap wrap- m tlle codices by bowls containing kan signs. The sign is also pings modi- placed in juxtaposition with some of the gods to indicate powers vertedface. 111 " f avora ble or unfavorable to good crops. The kan sign in connection with the ear of maize or the sprout, whichever it may be, is shown in Fig. 122, d. This latter object is some¬ times represented as a curiously conven¬ tionalized serpent head of which the eye at the top is the most conspicuous feat¬ ure. According to Schellhas the head of the Maize God was itself evolved out of this object. In this statement, how¬ ever, he probably goes too far. Sometimes, as in Fig. 122, b, the face and the headdress are very closely combined or even fused together. Sometimes the face is divided into a forward and a backward part and the two divisions differ¬ entiated in color. But as often as not the face and form of the Maize God is 1 1889-1902,1, p. 37. 1 1904, pp. 24-25. .. — Glyphs of the day Kan. GENERAL CONSIDERATION OF MAYA ART. 89 -Various representations of the Maize God: a and b, Dresden Codex; c-c, Codex Peresianus. youthful, beautiful and of a purely human type (Fig. 122, a and b ). The Maize God, apparently represented as newly born, and with the umbili¬ cal cord still attached, may be seen on Plate 19 of the Peresia¬ nus Codex (Fig. 122, c). On the monuments the rep¬ resentation of this god may be discerned in the youthful figure with a leafy headdress, examples of which are given in Fig. 123. It occupies a secondary position on the monuments, but the char¬ acters are constant and are, moreover, consistent with those appearing on the figures in the codices. On Stela H at Copan several small human beings of this type, Fig. 123, e, may be seen climbing round and over the interwoven bodies of serpents. At Quirigua the occurrence is similar (/), while at Tikal the head shown in b thrusts itself out of the eye of a richly em¬ bellished serpent head, the upturned nose of which is shaped into the face of the Roman¬ nosed God (Fig. 96, g). In all these drawings the determining feature is the bunches of circles enclosed in leaflike ob¬ jects that may repre¬ sent the ear of maize or bursting seed pods. In an interesting stucco decoration in the Pal¬ ace at Palenque (Fig. 123, a) are shown com¬ parable circular details as well as maize ears rather realistically drawn, while the god himself appears at the top of the design. De¬ tails which seem to represent ears of maize or bursting pods are recorded in a drawing by Waldeck 1 of one of the now lost tablets of Palenque. The maize 1 Maudslay, 1889-1902, IV, pi. 86. - The Maize God in the sculptures: a and d, Palenque; b, Tikal; c and e, Copan; /, Quirigua. 90 MAYA ART. — God C and his hieroglyph: Dresden Codex. ears in this instance seem to depend from the inverted head of the Long-nosed God. The form of the Maize God in all these instances is distinctly human and in marked contrast to the other deities so far considered. The beautiful sculp¬ ture 1 from the facade of Temple 22 at Copan which Maudslay calls a “singing girl” may represent the youthful Maize God. Other comparable figures from the same building are in the Peabody Museum (Plate 26, fig. 3). The headdress resembles that of this deity as given in the codices. There is clear enough evidence that the faces and figures of the Long-nosed God, the Roman-nosed God and the Death God were used to decorate the facades of temples in this city, and the usage may have in¬ cluded other deities as well. Two sculptured stones from the terrace east of the Great Plaza at Copan doubtless bear representations of the Maize God. 2 The figures are human. The headdress has the usual sign of the growing plant surmounted by a small face of the Long-nosed God. The Maize God seems to have been a divinity with little absolute power. He is frequently shown in the codices under the protection of the benevolent gods, B and D, or under the malign control of the Death God whom Schellhas calls God A. The Maize God seems to bear some relation to the num¬ bers eight and eighteen, 3 because his face occurs in some in¬ stances as the glyph for these numbers. The higher number shows, of course, the usual death signs in addition to the natural features. Other Divinities. The following gods ac¬ cording to the system of Schellhas have already been considered in more or less detail: A, B, D, E, G, I, K and N. In addition to these many other forms that do not fall into this cate¬ gory have been taken up. There re¬ main a number of fairly well-defined . gods who deserve brief comment. God C, the god of the North Star or the northern sky, has very characteristic features. Figs. 124,125 and 126 present drawings of God C from each of the three Maya codices. The face of this god was found a number of times as a hieroglyph during the exploration of the Hieroglyphic Stairway at Copan, one of the original stones now being in the Peabody Museum. It also occurs in various other inscriptions (Fig. 127) and in astronomical bands on several of Fl0 . 127 _Hicrog!yph 9 containing the monuments. No extensive use of this god's figure faceofGodC:a,Paienque;&,Co- has come to light in the larger sculptures. pan; M ,Tro-Cortesianus Codex. Gods F and H have not been clearly identified in the sculptures. The first 1 Maudslay, 1889-1902, I, pi. 17, a and b. 2 Gordon, 1896, p. 2. 3 Bowditch, 1910, pis. 16 and 17. Fig. 125.—God C with a mott led green body: Codex Peresianus. Fig. 126.—God C: Tro- Cortesianus Codex. GENERAL CONSIDERATION OF MAYA ART. 91 is a god of war who has human form. It is possible, of course, this portrait ap¬ pears in some of the sculptures that deal with warfare and conquest. The sec¬ ond god is called the Chicchan God because he has certain markings on his face that resemble the markings on the serpent body and on the glyph for the day Chicchan. He also seems to be a warlike divinity. Seler calls this god the Young God. He may perhaps be identified with the head having purely human features that rarely appears in the serpent mouth (Fig. 24). As a rule, the head in the serpent mouth belongs to the Long-nosed God group or to the allied group of the Roman-nosed God. Gods L and M are represented in the manuscripts with black bodies. The latter according to Schellhas may be identified with Ekchuah, the black god of the traveling merchants. Goddess 0 represents an old woman; few represen¬ tations of her occur. God P has been called the Frog God. The frog is repre¬ sented with some frequency on the monuments and seems to be the original of the uinal glyph that represents the month. This connection appears to be purely phonetic, however, since “uo” means frog and “u” month. 1 Astronomical Signs. Bands of astronomical symbols have already been many times referred to and compared in a general way. There is considerable uncertainty concerning the exact significance of many of the individual symbols. It is probable that the sun and moon, the important planets, and the larger con¬ stellations were represented specifi¬ cally and that there were other signs that were general and inclusive in their meaning. It has already been demon¬ strated that one of the so-called astronomical symbols is the con¬ ventionalized face of the Roman¬ nosed God (Fig. 128, a). The sign probably signifies either the sun specifically or the more general idea of day or light. Its importance is indicated by the fact that in one case it occupies the entire space at the side of a monu¬ ment (Fig. 92) and in another alternates with the unusual Caban sign (Fig. 91, /) that may represent its opposite, which might be the moon, or perhaps darkness. The usual form for the sun is the normal kin sign (Fig. 128, 6), which consists of an oval or oblong with one or more marks extending inward from the middle of each side. Sometimes a circle occurs in the center. Dr. Seler 2 includes in this group of sun symbols the forms with dotted diag¬ onals (Fig. 128, c and d), which Forstemann considers to be symbols for the planet Mercury. Mr. Bowditch, 3 however, shows that the calculations in the Dresden Codex upon which the supposition is based do not agree very closely with the periods of revolution of this planet. The sign is of very frequent occur¬ rence, both in the sculptures and in the manuscripts. Another form which may represent the sun is less commonly encountered. It is more or less circular, with a normal kin sign in the center and a serpent head projecting outward at four points. As an astronomical symbol it may be seen on Stela 10 at Piedras Negras 4 and perhaps in the upper division of Stela 1 1 Bowditch, 1910, pp. 257-258. ’ 1901-1902, pp. 165-166. >1910, p.228. < Maler, 1901, pi. 19. Fig. 128. — Sun symbols: a, Palenque; b, Yaxchilan; c and d, Dresden Codex. 92 MAYA ART. - Sun disk represented in fresco: Chichen Itza. and in the upper left-hand corner of Stela 4 at Yaxchilan. 1 As a medallion decoration on piers it occurs at Palenque, 2 where it incloses a head. It like¬ wise is found in the Peresianus Codex. 3 In the frescos of Chichen Itza a sun disk is represented with a serpent head projecting on the four diagonals (Fig. 129). This sun disk is itself a Nahua and not a Maya concept, but the four serpent heads in connection with it may hark back to a genuinely native origin. A symbol which Forstemann considers a moon symbol (Fig. 130, a) occurs very frequently in the codices on terms of equality with the kin sign in the heraldic shields which are attached to the under sides of astronomical bands. It represents in a cursive and a demotic manner an ornamented eye or a partial face. This symbol is very similar to the sign for twenty in the codices as well as to the hieroglyph for God D, the principal phase of the Roman-nosed God in the manu¬ scripts. Dr. Seler 4 considers these signs to represent the bloody sockets of gouged- out eyes, the hieroglyphs of Itzamna, whom he considers Lord of Life and of the Milky Way. He is probably right in the general conclusion that the face represents a god who takes the form and features of an old man. But the points upon which the interpretation is made appear highly fanciful and hardly to be supported by objective study. Ornamented eyes are of very general occurrence, and in most cases where circles are drawn be¬ neath them there is no other evi¬ dence of “gouged sockets.” It seems probable that this cursive face should be correlated with the rectangular presentation of the Roman-nosed God that has just been con¬ sidered (Fig. 128, a). This symbol that occupies so prominent a place in the monuments does not appear in the manuscripts unless under the present guise. As for the crescent-shaped symbols (Fig. 130, b and c) which Dr. Seler in the same passage associates with the symbol of the ornamented eye, just considered, a comparison of other forms does not seem to support his conclusions. The form occurring on Stela 10 at Piedras Negras 5 clearly shows a head inclosed in the deep crescent. The most interesting example of this figure is seen in the upper right-hand figure of Stela 4 at Yaxchilan, where it is in a position of opposition to the circular sign with the four serpent heads. It is of course quite possible that these crescent symbols should represent the moon. They also resemble one of the glyphs of the so-called Supplementary Series. 6 The symbols of the planet Venus are pretty well ascertained. There are two principal forms, as - Symbols that may represent the moon: a, Dresden Codex; b and c, Palenque. 1 Maler, 1903, pla. 69 and 70. 1 Maudslay, 1S89-1902, IV, pi. 6. 1 De Rosny, 1887, pi. 21. 1 1901-1902, pp. 166-167. 5 Maler, 1901, pi. 19. 1 Bowditch, 1910, p. 244. GENERAL CONSIDERATION OF MAYA ART. 93 shown in Fig. 131, a and b. Also there are many irregular or unusual forms that include one of these simple signs combined with an animal or some other object. Venus symbols are of very common occurrence not only in astronomical bands but also as details on various sculptured figures and in the hieroglyphic inscrip¬ tions. In fact, some of the most extensive inscriptions may refer largely to the correlation of the solar and Venus years. An extended passage in the Dresden Codex is given over to this subject. 1 Some of the variant signs of the Venus symbol doubtless refer to different appearances and conjunc¬ tions of this planet. Gordon 2 3 b c derives the Venus symbol of F,G ' m — Ven “ C ° p,m: the first type from the con¬ ventionalized jaws of the serpent. But the stages of this development are not very clear if you omit the doubtful forms that occur in Nahua art. The second type seems, however, to have developed from the first by a simple folding over or reduplication. Another sign which is pretty well settled as to its significance represents the face of God C, the divinity who rules the North Star or the entire northern sky. This sign occurs more frequently in the codices than on the monuments. Fig. 132. — Planet symbols: a, Copan; c,/, g, j and k, Palenque; b, d, e and i, Dresden Codex. A symbol in the form of a cross (Fig. 132, a-c) is exceedingly common. It probably has some very general meaning such as the sky as a whole. This sign frequently occurs as a hieroglyph and upon headdresses. The symbol which resembles Akbal may mean night or it may have some more specific mean¬ ing. Comparable forms in the manuscripts and monuments are given in the text (Fig. 132, d-f). This symbol has been referred to the planet Jupiter on rather doubtful grounds. 3 The symbol which shows a serpent head arranged diagonally in the oblong panel (Fig. 132, g-i) has been similarly ascribed to Saturn. Other symbols of less frequent use are shown in Fig. 132, j and k, as well as in Figs. 5, 6, 83 and 89. Hieroglyphs. The Maya hieroglyphs have been so many times referred to in the text that it seems hardly necessary to accord them here more than a gen¬ eral treatment. Although at the present time few of the hieroglyphs have been deciphered, the task does not seem to be an insurmountable one. A large part of 1 Bowditch, 1909. 2 1905, pi. 6. 3 Bowditch, 1910, pp. 229-231. 94 MAYA ART. Fig. 133. — Composite mon¬ ster and its glyph: Dresden Codex. the inscriptions have to do with astronomical calculations which introduce cer¬ tain absolute factors. In fact, the glyphs connected with numbers and the calen¬ dar are now pretty well ascertained, including the so-called period glyphs and the numerals from one to twenty, and the signs for each of the twenty days and for each of the eighteen months. The hieroglyph for the extra period of five days used to complete the annual calendar is also known as well as the signs for zero or completion. Some of the hieroglyphs which refer to certain heavenly bodies and to certain gods have been isolated. The symbols for the four directions have been determined with considerable certainty as well as a few other signs of lesser importance. It may be remarked, en passant, that the results of Le Plongeon 1 and Brasseur de Bourbourg 2 are of very little value so far as the decipherings of inscriptions are concerned. Is or has the so-called alphabet of Landa 3 proved of much service, although it was evidently taken down in good faith. Most of the Maya hieroglyphics are probably ideographic and consist of abbreviated pictures of the thing intended or of some object connected with it. In Fig. 133 we see a representation of some mythological conception and the glyph which refers to it. The glyph that probably stands for the rear head of the Two-headed Dragon contains, as we have seen (Fig. 82), the three peculiar signs of the headdress and the sun sign on the forehead. It seems pretty clear that certain symbols have a phonetic value, probably of the syllabic rather than of the alphabetic type. This phonetic character is particularly demonstrable in words containing the syllable kin, which may be represented by the sun symbol. But it is extremely doubtful if there was a complete syllabary adapted to narrative texts. The general status of writing was probably much like that of the Valley of Mexico, which has been ex¬ plained by Brinton, Peflafiel and others, although it is pos¬ sible that the range of subjects was somewhat greater. The Nahua seem to have learned this valuable art from their southern neighbors. Many points of divergence in the two areas must be ascribed to the differences in organic structure in the two languages as well as to the different standards of art. In addition to what is now known we may expect to find in the Maya inscrip- 1 1886 and 1896. 3 1869-1870. 3 1864, pp. 316 et seq.\ Valentini, 1880; Brinton, 1894, b. Fig. 134. — Series showing modifications of glyphs: a and b, Cib; c-j, Caban; a, Peresianus Codex; b, Dresden Codex; c. Tro- Cortesianus Codex; d, Landa; e, Copan; f—j, Quirigua. GENERAL CONSIDERATION OF MAYA ART. 95 tions some hieroglyphs that give the names of individuals, cities and political divisions and others that represent feasts, sacrifices, tributes and common ob¬ jects of trade as well as signs referring to birth, death, establishment, conquest, destruction and other such fundamentals of individual and social existence. Juxtaposition of these hieroglyphs together with directive signs and dates would make possible records of considerable accuracy. On many of the monuments the small number of hieroglyphs left, once all the dates have been eliminated, sug¬ gests that such an abbreviated system of writing was in vogue. As to names of individuals or cities, it is worthy of note that several of the sculptures show hieroglyphs over the heads or upon the bodies of human beings. These might very well be names (Figs. 10 and 17). The hieroglyphs over the heads of the warriors in the processions shown on the walls of the Lower Temple of the Jaguars at Chichen Itza may represent cities rather than individuals. They are of Nahua rather than Maya type, however. The degree of variation in the Maya hieroglyphs that have been ascertained is very great, as may be seen from the series collected by Mr. Bowditch. 1 Fig. 134 presents a series of glyphs of the day Caban which are unusually consistent. The characteristic feature is the corkscrew curl. But even this feature reappears in other glyphs, particularly in the forms for the day, Cib, from the codices. An example of elaboration is given in Fig. 135 from Stela D at Copan. The essential features of the hieroglyph seem to be those that are carried in the arms of the two human figures. Other examples of complicated hieroglyphs have received comment. It is apparent from these examples that the study of the art is very necessary if one is to arrive at an understanding of the subject matter expressed in these abbreviated or elaborated pictures. Fig. 135. — Hieroglyph with entire figures: Stela D, Copan. 1910. i II. CONSIDERATION OF THE MATERIAL ARTS Architecture Assemblage. The mapping of the principal Maya ruins has disclosed several styles of assemblage of the city as a whole, each of which seems to have a fairly definite geographical distribution and at the same time a topographical explanation. The clear types will here be briefly described as well as some of the apparent transitions. Most of the Maya cities were built upon level ground, either extensive plains or valley floors. Where such was the case the assemblage was unhampered and followed the fashion of the region or perhaps of the period. But in some in¬ stances — particularly in the western part of the Maya area — the cities were hemmed in by hills and streams. Here the assemblage had to adapt itself to surroundings, but doubtless the builders attempted to preserve as much of the usual order as they could. Perhaps the most careful and elaborate grouping of the city as a whole is seen at Copan. 1 In this place there is a massive platform mound, a sort of arti¬ ficial acropolis, with terraces and sunken courts at various levels. Rising from the flat of this principal mound are small pyramids of the usual type crowned with temples. The great mound overlooks an extensive plaza in which are set up stelae. The plaza is surrounded by a stepped wall as if it were a sort of thea¬ ter. The so-called sunken courts are also inclosed by stepped walls and are drained by tunnels that pass under the walls. Most of the small mounds which mark the domiciliary structures lie on the opposite side of the great mound from the plaza. This elaborate mode of grouping may have been intended to obtain a broad architectural effect. While there is apparently no definite orientation, there is an orderly alignment of the buildings and terraces. At Quirigua 2 a very similar assemblage is found, both as regards the form of the artificial acropolis and the use of the plazas in which were set up stelae. The use of the great platform mound to serve as a base for a number of smaller substructures also characterizes the great ruins of the Peten region. But in these cities the artificial acropolis is perhaps not so conspicuous as in the cities just named. A plan of Ixkun by Maudslay 3 shows an artificial acropolis on which are several pyramids arranged around courts. At Naranjo, 4 near the western end of the city, is a large rectangular mound with a lower adjoining terrace. Upon this large mound are remains of six structures very much destroyed. At the eastern end of the city there is, in all probability, a low but extensive mound which serves as a foundation for the three principal courts with their inclosing structures. 1 Maudslay, 1889-1902, I, pi. 1; Gordon, 1896, pi. 1. 3 1889-1902, II, pi. 67. 2 Maudslay, 1889-1902, II, pi. 2. 4 Maler, 1908, b, p. 83. ) CONSIDERATION OF THE MATERIAL ARTS. 97 At Tikal there are, according to the recent explorations of Dr. Tozzer 1 and Mr. Merwin, no less than three great foundation mounds that mark out the civic and religious centers of this most important city. These mounds are of consider¬ able height, but do not compare with the great mound of Copan. Upon these mounds are many closely connected courts surrounded by temple and palace structures. The artificial acropolis also occurs at Nakum, La Hondradez and other sites in the Peten region. 2 The orientation of courts and buildings with strict regard for the four direc¬ tions prevails among the cities of southern Yucatan that have just been men¬ tioned as well as at Siebal 3 and doubtless other places. It does not occur at Copan and Quirigua, although at these sites there is an orderly alignment of walls and mounds. At Copan an east and west line passing directly across the city seems to have been surveyed with tolerable accuracy and marked by two stelae placed on hills on opposite sides of the valley. 4 Perhaps the most interesting point concerning the use of closely connected and carefully oriented courts is that each court with its associated buildings naturally served as a unit of city growth and that the sequence is more or less exactly indicated by position. The ruins of northern Yucatan, 5 although situated in a level country, show neither the artificial acropolis nor the careful orientation that distinguishes the cities of the south. Occasionally advantage was taken of slight natural eleva¬ tions, as at Labna. The irregular platform mounds that serve as foundations for the larger structures may perhaps be considered decadent examples of the early artificial acropolis. The assemblage may be termed haphazard. There are, as a rule, several independent groups of correlated buildings. These inde¬ pendent groups may represent different periods of city growth. In the corre¬ lations of buildings within these groups the principle of arrangement is very often that of the rectangular court with one or more buildings on each side. Yaxchilan, 6 Piedras Negras 7 and Palenque 8 are examples of cities situated in narrow valleys where the topography modified the assemblage. In the first two sites natural hills or ridges were leveled off and terraced. But the buildings erected upon these hills have little exact and premeditated grouping. The same may be said of Palenque. No natural or artificial acropolis occurs at this site, unless the mound that supports the Palace is considered one, but instead a nar¬ row valley, the sides of which were terraced to a considerable height. There is good reason to believe that the artificial acropolis would have prevailed at these cities if the topography had permitted. At Comalcalco, which is even farther west than Palenque, there is, according to Charnay, 9 a massive mound upon which are the ruins of several buildings. The same occurs at Ocosingo or Tonina. 10 Certain minor features of assemblage will now be presented principally in 1 Tozzer, 1911, pis. 29 and 30. 2 Sapper, 1897, p. 360 (Ixtinta) and p. 362 (S. Clemente). 3 Maler, 1908, a, p. 13. 4 Gordon, 1898, b, p. 4, gives a map of the envi¬ rons of Copan, as does Maudslay, 1886. 5 For plans of Chichen Itza see Maudslay, 18S9- 1902, III, pi. 2, and Holmes, 1S95-1897, pis. 17 and 18. For Uxmal see Holmes, 1895-1907, pis. 8 and 9, and Stephens, 1843, I, p. 165. For Labna see Stephens, 1843, II, frontispiece showing panorama, and an unpublished map made by the Peabody Museum under Mr. Thompson. For Kabah see Stephens, 1843, I, p. 385. For Ake see Charnay, 1885, p. 249. For Tuloom see Stephens, 1843, II, p. 396. 6 Maudslay, 1889-1902, II, pi. 76; Maler, 1903 pi. 39. 7 Maler, 1901, pi. 33. 8 Holmes, 1895—1897, pis. 24 and 25; Maudslav, 1889-1902, IV, pi. 1. 9 1885, p. 167. 10 Sapper, 1897, p. 361. 98 MAYA ART. connection with frontier ruins. Along the Uloa River 1 the influence of Copan and Quirigua may be seen in the arrangement of mounds and courts. At one of the ancient ruins a crude stela appears in front of the principal mound. The use of sunken courts and of plazas with stelae extends up the Motagua River and over the highlands of Guatemala as far as the Chiapas Valley. Many of the plans of ancient settlements given by Sapper 2 and Seler 3 show similarities to the lowland sites. Careful orientation was not observed. These outlying towns were doubtless provincial in character, and most of them seem to have flourished at a later period than the great cities of the Peten region. The elaborate arti¬ ficial acropolis is not seen, while the parallel walls of the so-called ball courts present a new and probably un-Maya feature. Many of the frontier settlements offer definite evidence of fortification that is lacking elsewhere. Tenampua 4 in central Honduras is described as occupying an impregnable position upon a lofty hill and as being further strengthened by surrounding walls. A number of towns in the highlands of Guatemala 6 were placed between barrancas and in other easily defended positions. Of course the artificial acropolis in the great cities may have been partly intended for defense. There is no doubt that warfare was highly and scientifically developed. Steph¬ ens 6 describes a wall at Tuloom that seems to have surrounded the city. Upon the frescos of Chichen Itza are represented earthworks behind which warriors are fighting, as well as what may be taken for scaling ladders. Function of Buildings. Little is known concerning the function of Maya buildings other than that they were largely of a religious nature. It is possible to distinguish between buildings for strictly ceremonial uses, such as the small temples on lofty pyramids, and other buildings, usually larger and situated on lower terraces, which may have served as dwellings for the priests and the nobility. This latter group includes the great rambling collec¬ tions of rooms, usually arranged around courts and commonly called palaces. Evidence concerning the differentiation and development of the temple and the palace will appear under various headings. Towers of several stories were sometimes built, but their use is unknown. The square tower at Palenque, four stories in height, has often been described . 7 A similar structure, not so high, is found at Comalcalco . 8 Maler 9 figures a tower¬ like structure, with a great stucco face on one side, that occurs at Nocuchich, as well as another tower of more slender dimensions. Round towers occur at Mayapan 10 and at Chichen Itza, the example from the latter city being the famous Carocol . 11 It seems pretty clear that these towers were not intended for observation. None of the towers have pyramidal substructures, and in each city where examples are found there are other buildings that exceed the tower in elevation. The round towers are said to have been associated with the wor¬ ship of Kukulcan. 1 Gordon, 1898, a, p. 11. 6 Maudslay, 1889-1902, II, pis. 70 (Rabinal), 3 1895, a. Map No. 5 (Hacienda Grande); No. 6 72 (Utatlan) and 73 (Iximch6). (Las Qdfebradas); No. 8 (Cakiha); No. 9 (Chacujal); 6 1843, II, pp. 395-396. No. 11 (Sacramento); No. 12 (Bolonchac); No. 13 7 Maudslay, 1889-1902, IV, pi. 39; Holmes, 1895- (Saculeu); No. 17 (Kalamt6); No. 18 (Comitan- 1897, pp. 179-186. cillo); No. 20 (Sajcabaja). 8 Charnay, 1885, p. 170. 3 1901, c, pp. 100 and 131 (Quen Santo). 9 1895, pp. 281 et seq. * Squier, 1858, pp. 133-138, and Bancroft, 1875- 10 Stephens, 1843, I, p. 136. 1876, IV, pp. 72-77. “ Maudslay, 1889-1902, III, pi. 20. CONSIDERATION OF THE MATERIAL ARTS. 99 The ball court or gymnasium seems to have a foreign origin and will be discussed in that connection, together with certain other imported ideas, archi¬ tectural, decorative and religious. This structure is found in none of the early Maya cities. The dwellings of the common people in ancient times were probably not essentially different from the huts still made and used by the natives of Yucatan. 1 Fig. 136. — Representations of These huts are generally oval in plan, containing a single room. The framework is of light poles skillfully bound together by withes. The walls are of wattle and mud, while the roofs are heavily thatched with the leaves of the Sabal and other palms. Mural paintings in the Temple of the Jaguars at Chichen Itza represent the ancient dwellings of the lower classes. Examples of these are reproduced in Fig. 136. Near most of the remains of stone temples and palaces there are, according to Mr. Thompson, 2 many evidences of these poorer dwellings. The outlines of the huts may be traced out by the uneven surface of the ground and the three-stone fireplaces uncovered by slight excavation. Ground-Plans. Some idea of the uses to which buildings were put may be obtained from a study of the ground-plans. The simple room, with the door in Fig. 137. — Series showing development of the sanctuary: a, Two-roomed Temple; b, Temple I, Tikal; c, Temple 22, Copan; d, Castillo, Chichen Itza; e, Temple of the Sun, Palenque. the center of one of its long sides, seems to have been the common starting-point for both the temple and the palace type of structures. In the case of the temple this simple room was modified by interior partitions until there was a clear development of the sanctuary or inner sacred chamber, while in the growth of the palace structures there was an agglutinative process by which one room, with or without interior divisions, was simply set up against another. Rarely indeed is there a series of interior doorways connecting the different room units of a large building. The development of the sanctuary in the temple is indicated in the series of temple ground-plans shown in Fig. 137. The simple two-chambered temple, a step in advance of the single room that served as a starting-point, is shown in a. Here 1 Thompson, 1892, a, p. 262. » 1892, a, pp. 263 et seq. >/ 100 MAYA ART. the inner room, getting only the diffused light from the outer door, might fittingly have been considered the holy of holies and the mysterious abode of divinity. Often the floor of this interior room is raised a foot or more above that of the outer room and the doorway and inner walls adorned with sculptures. The temples of Tikal have very massive walls and small cell-like rooms, sometimes three in number one behind the other (Fig. 137, b). The wooden lintels over the door¬ ways are in some cases splendidly carved. The ground-plan of Temple 22, at Copan, is given in c of this series. The ceremonial importance of the inner cham¬ ber of this temple is emphasized by elaborate carvings, representing the Two- headed Dragon supported by kneeling atlantean figures. This design enclosed the doorway. The entrance to the inner chamber of Temple 11, at Copan, is also elaborately ornamented by carvings. The highest form of the sanctuary is seen at Palenque (e), in the Temples of the Sun, the Cross and the Foliated Cross. The sanctuaries here are little temples in themselves, roofs and all, and are adorned with the most wonderful native bas-relief carvings of the New World. Some of the temples of Chichen Itza also have well-defined sanctuaries, as may be seen from d, the ground-plan of the Castillo. In many temples there is a built-up bench in the b a Fig. 138. — Structures showing the extreme . 188. — Structures showing the extreme , i, • i i 1 development of the portico: Group of the sanctuary which may have served as an altar. In some cases table altars have been found in position in the outer room directly Columns, Chichen Itza. in front of the door to the sanctuary. Closely paralleling the development of the inner division of the simple temple into a true sanctuary, the outer division becomes a portico. At Copan, Quiri- gua and Tikal, where the walls are exceedingly massive, the temple facades are broken by but one doorway. As one proceeds toward the north, the walls be¬ come much lighter, although at best cumbersome, and two or more doorways, symmetrically placed, give entrance to the outer chamber of the temple. At Yaxchilan nearly all the temples have several doorways. At Palenque the door¬ ways are placed so closely together that the portions of the wall remaining be¬ tween them are hardly more than piers. At Labna, Chichen Itza and other cities of northern Yucatan these pierlike portions of the wall are often actually replaced by square or round columns. With the extended use of such supports the front room of the temple becomes more and more open. Fig. 138 gives two ground-plans that illustrate this ultimate development of the outer chamber into a light and airy portico. In a the sanctuary is a small cell behind a wide open chamber and in b the portico has a double row of columns. The buildings show¬ ing double rows of columns are found only in northern Yucatan and are of late date. It will be shown in a later treatment that the development of the sanctu¬ ary and the portico, that has just been sketched out, is really historical and covers practically the entire chronological range of Maya art. Some structures, not properly of the one-room origin just described, but consisting of two or more independent rooms, were probably used as temples and not for civil or domiciliary purpose. The House of the Magician at Uxmal is an example. The main temple has three rooms in a row without connecting doorways. The middle chamber of this temple opens in the direction opposite t-lBK CONSIDERATION OF THE MATERIAL ARTS. 101 to the doorways of the end chambers and is, in fact, rendered inaccessible by the Annex built apparently at a later date. The steep and lofty substructure of the House of the Magician would have made it an inconvenient abode. But as a temple it is exceedingly impressive, looking down as it does upon the Nun¬ nery Quadrangle. Fiq. 139. — Partial ground-plan of the Group of the Monjas: Chichen Itza. Ground-level buildings in black; b, second range structures; a, third range. It is difficult, if not impossible, to draw any definite line between the temple and the palace; for it is likely that even in buildings clearly of the latter type there were certain rooms given over to religious rites. A partial plan of the Group of the Monjas at Chichen Itza, after Maudslay, is presented in Fig. 139. This group shows several distinct periods of growth. Moreover, some of the parts seem to show differentiation in use. In particular the small closed court marked A in the plan appears almost necessarily domiciliary. The elaborate frieze dec¬ oration of grotesque masks, that characterizes the north and east fagade of the eastern ground-level wing, stops abruptly after turning the corner on the south side. The rest is plain. The two buildings which completed the square were of unusually light construction and may even have had wooden roofs. They are in complete ruin and the Fl °' 140 ' — Ground 'u^ : a i HouBe ot Turtles ’ amount of debris is not great. This secluded court may well have been the abode of the temple attendants. On the other hand, the elaborate decoration of the eastern fagade indicates a religious significance for the end chamber. The third story of this building must also have had purely religious uses. A wide stairway leads up to a small single-room structure with an altar-like object in front. The small detached building known as the Iglesia was probably for purely religious uses. The other chambers of this group might have served as a religious college or monastery, or as a chief’s palace. The remarkable symmetry in plan as well as the agglutination of independent 102 MAYA ART. room-units of the larger structures is well brought out in the ground-plans of the House of the Turtles (Fig. 140) and of the Governor at Uxmal 1 and of the Akat’cib at Chichen Itza (Fig. 141). The rooms are either strung out or clustered. The palace structures at Palenque 2 and some of the southern cities often show several groups of rooms adjacent to each other but with disconnected walls. Elevation Plans. It does not seem necessary to enter into an extended Fig. 141. —Ground-plan: Akat’cib, Chichen itza. consideration of elevation plans, be¬ cause most of the important facts con¬ cerning them are brought out in the consideration of other subjects. However, the usual Maya method of erecting buildings of more than one story is both interesting and significant and is readily seen from elevation plans. Owing to the cumbersome construction it was ordinarily not deemed safe to put one room directly over another. Nevertheless this feat was accomplished as may be'seen from the four-storied tower at Palenque. The so-called Temple of the Five Stories (Structure 10) at Tikal 3 shows three stories, one above the other, and Fig. 142. — Ground-plans: Santa Rosa Xlabpak. two lower stories at the side of the foundation mound. As a rule, however, the second story was built upon a solid substructure immediately behind the room or rooms of the first story and on a level with its roof. As a development of this method a series of rooms was sometimes constructed entirely around or at either end of a solid mass of masonry. Upon this mass of masonry was built the second story, which might in turn have a smaller core of solid masonry to support the third range of rooms. The principal building of Santa Rosa Xlabpak is a most interesting example of symmetry and fine construction. In Fig. 142 are repro- 1 Holmes, 1S95-1897, pi. VIII. 2 Maudslay, 1889-1902, IV, pi. 3. 3 Tozzer, 1911, pp. 112-113. CONSIDERATION OF THE MATERIAL ARTS. 103 duced the careful drawings of Maler giving the floor plans of the three stories of the Palace-temple Tampak, as he calls it, and in Fig. 143 are three typical elevation cross-sections. The masses of solid masonry are shown in black. The first story is on a level with the ground. The rooms open out on all four sides of the buildings and are generally double, one chamber being behind the other. The second story covers a somewhat greater area than the solid core of the first story. The outer walls of the second range of rooms fall in some cases over the interior walls of the first. The third story faces the east and is approached by a broad flight of stairs, at the top of which stands a portal arch. The narrow winding stairway at the back also ascends to the third story (Fig. 142). The rooms at this high level are all single. Stephen 1 calls this building “ the grandest structure that now rears its ruined head in the forests of Yucatan.” Fig. 143. — Elevations: Santa Rosa Xlabpak. The Akat’cib at Chichen Itza (Fig. 141) may be compared in certain details to this building. Here the ground-level rooms are finished and the solid mass that was probably intended to support the upper stories is in place, but no sec¬ ond and third story was ever begun. The principal building at Sayil 2 shows all three stories, and is perhaps the most extensive single structure in the Maya area. Other examples of two or three stories built according to the same prin¬ ciple of an interior core might be named. A peculiar feature of many Maya structures that has frequently been com¬ mented upon is the occurrence of rooms that have been filled with earth and stone and sealed up. In almost all cases this seems to have been a preliminary to the construction of second-story rooms immediately above. Examples of such filled rooms are seen in the Monjas at Chichen Itza (ground-plan in Fig. 139). Apparently it was the purpose of the builders to erect a second range of rooms over the East Wing, but this intention was never carried out. The single room of the third story is directly over a filled-up chamber. Correlation of Buildings. It has been stated that assemblage in northern Yucatan is haphazard as far as the city as a whole is concerned, but that corre¬ lation is frequently shown between a number of structures. Of course this is also true in the south, where the cities are generally divided into courts. The grouping of structures around a court likewise occurs in northern Yucatan, but with certain differences. The ground-plan of the Monjas group at Chichen Itza has already been presented. In this group there are several buildings care¬ fully aligned that partially enclose two or more courts. The principal fagade is on the north side. At Uxmal correlation is shown in more unmistakable ways. In the case of the grand Nunnery Quadrangle the group as a whole faces the south. The stair- 1843, II, p. 162. 2 Stephens, 1843, II, p. 22. 104 MAYA ART. ways are on the south side and in the middle of the South Range of rooms is a portal arch. The North Range is on a higher substructure, and the walls of the building itself are carried higher than usual and are very richly decorated. As a. o 50 FEET CL Fig. 144. — Assemblage of Edifice 5, Chacmultun. a result, this building is visible over the top of the South Range from any point of vantage. The House of the Magician seems to be attached to the southeast corner of the Nunnery Quadrangle, the transition being affected according to CONSIDERATION OF THE MATERIAL ARTS. 105 Holmes’ plan 1 by two small buildings. It seems likely that the House of the Magician proper is an older building than the Nunnery, since it faces in the oppo¬ site direction, but that the Annex to it was built in connection with the latter structure. The Southwest Group at Uxmal has been clearly presented by Mr. Morley. 2 This group has a four-roomed temple at the southern end, toward which the terraces gradually rise in several levels. Most of the ranges of rooms are built against the side of the next higher terrace. Across the middle of the group ex¬ tends the House of the Pigeons, with its castellated roof comb and its portal arch (Plate 6, fig. 1). A still more striking development of this idea in a group of much smaller size is seen at Chacmultun (Fig. 144). The details of Maya construction have been so clearly and admirably de¬ scribed by Mr. Holmes, in his “Archaeological Studies among the Ancient Cities of Mexico,” that it is here only necessary to recapitulate the main features. Where, however, a somewhat different interpretation of accepted evidence may have important bearing upon the development of the architectural decoration or upon the connection of the Maya building art with an earlier, more primitive type, then the matter will be discussed more fully. Substructures. The stone buildings of the Maya, as we have seen, were seldom erected upon ground level, but instead upon artificial mounds. These substructures were apparently not built with an eye to defense (although more or less adaptable to such purposes), but seem to have been purely architectural in function. A large part of the Maya area is without much natural relief, and it might be imagined that the fact led to the use of lofty substructures. But it must be pointed out that these foundation mounds were used as much in hilly country as on the level plain, not only in the Maya area but also in the neigh¬ boring Zapotecan and Nahua areas. Often natural elevations were entirely neglected and enormous mounds built up directly from the valley floor, as at Copan. At other sites where the topography was an inevitable factor in the laying out of the city, as at those of the Usumacinta Valley, the natural hills were leveled off or terraced and then artificial substructures reared upon these platforms. Thus it seems clear that the substructure was considered an archi¬ tecturally valuable feature and one that made directly for grandeur and magni¬ ficence quite apart from the question of mere elevated outlook. The pyramid was part of the temple. The artificial mounds vary much in size, height and shape, but are fairly similar from the point of construction. They usually consist of a solid mass of rubble, mortar and earth faced with cut stone. In many cases it is evident that they were built up by levels. Buried walls and pavements are occasionally found, as in the great acropolis mound at Copan. These may be the remains of the sides and crowns of earlier mounds, which, as the city grew, were found inadequate and so were deeply buried under the new acropolis, or they may be evidences of the method of construction. It is possible to divide the mounds, as regards size and shape, into two gen¬ eral types, the platform mound and the pyramid. The platform mound includes a great range in contour and elevation, but is marked by the general presence of right-angled corners and irregular terracing. Usually rather low in elevation, 1 Holmes, 1895-1897, pi. 8. 1 Morley, 1910, a. 106 MAYA ART. they have often several levels, both on the main mounds and on the ells, which are of frequent occurrence. These platform mounds serve as foundations for the larger and more irregular structures. The artificial acropolis that has already been described may perhaps be thrown into this class, although it is a communal rather than an individual substructure. The pyramids are truncated, usually rectangular, although some have rounded corners, and rise in a series of either vertical steps or slanting terraces. Many variations in form are found. In height the pyramids run from twenty to one hundred feet. The pyramids of Tikal are the highest and the steepest in the Maya area. There are many lofty pyramids in northern Yucatan. The temples of the Usumacinta region are placed, as a rule, on low pyramids. A very effective decoration was obtained by the use of sunken panels in the stone casings of the various terraces, as may be seen on the pyramid of the Castillo 1 at Chichen Itza and on that Temple I at Tikal. 2 Ornamental stone¬ work of simple but agreeable character may likewise be studied on the very steep pyramid at Rio Beque which has been described by the Comte de P6rigny. 3 Sometimes each terrace wall was provided with a simple cornice. The upper part of the substructure of the Monjas at Chichen Itza bears a frieze of mask panels (Plate 28, figs. 3 and 4). Plastic decorations in stucco and painting upon 1 Maudslay, 1889-1902, III, pi. 56. 3 Maler, 1911, pl/2. 3 PSrigny, 1908. CONSIDERATION OF THE MATERIAL ARTS. 107 a plaster base may also have been used to adorn certain substructures. Scanty traces of such decorations are still to be seen. Stairways were constructed on one or more sides of the pyramids and plat¬ form mounds. They were very steep, but usually projected somewhat at the base, and so were not so steep as the mounds themselves. Occasionally, as in the case of the Hierolgyphic Stairway at Copan, the steps were carved with glyphs and monuments were placed at intervals. Low balustrades were rather frequent. Sometimes they received considerable attention. Fig. 145 shows a restoration of the balustrades of the Hieroglyphic Stairway at Copan. A number of so-called serpent balustrades are found at Chichen Itza. 1 The heads of the serpents extend outward at the base of the balustrade, which may be said to represent the simplified serpent body. Serpent stairways seem to be restricted to a definite class of pyramids, namely, those with stairways on all four sides, and to a def¬ inite period of Maya art. Walls. The ordinary wall construction resembles that of the Romans. It is not true masonry, but a rough concrete faced with cut stone. The uni¬ versal lime rock of the country was the material that was . broken up for rubble, burned for mortar and plaster, and cut with flint chisels for surfacing stones or ornamental sculptured details. Walls made entirely of cut stone are rare. Perhaps the nearest approach to true stone masonry occurs at Copan. Here rectangular blocks of fairly uniform size were laid in a neat and orderly manner. The joints were broken with fair regularity and the corner-stones were laid in a simple locking system. Plate 3 shows some of the best preserved walls at Copan in which these details are readily discerned. A heavy layer of mortar was used for floors and a thin coating for walls, but this material seems to have been seldom used to cement together the building stones. The most peculiar and significant feature of the wall construction of Copan has yet to be mentioned. The temple walls are not made entirely of stone, but have a core or hearting of pounded earth or clay mixed with broken stone (Fig. 146). This impermanent filling could serve no useful purpose, except to give greater mass and weight to the wall should that be needed. The stonework is thereby converted into retaining walls for an earth embankment. The collapse inward of the stone retaining walls, such as may be seen in Plate 3, fig. 2, is per¬ haps due to the washing away of the earthen filling, although Professor Saville ascribes the destruction of the walls at Copan to earthquake action. 2 This peculiar method of construction may indicate that the prototype of the Maya temple was a mud-walled structure and that stonework was added as a veneer. In the wall construction of Tikal, Yaxchilan, Palenque, etc., the use of irregu¬ lar slabs of limestone set in a plentiful supply of mortar predominates. The 1 Maudslay, 1889-1902, III, pi. 58. 1 1892, p. 273. - Cross-section of an interior wall, showing the hearting: Copan. 108 MAYA ART. stones on the outer surfaces are roughly trimmed to shape and minor irregulari¬ ties covered up under a liberal coating of stucco, which likewise formed the chief medium of ornamentation. The large stucco figures in high relief that served to ornament the roof combs were built up over stone skeletons, as shown by Mr. Holmes. 1 In northern Yucatan, stucco surfacing is employed to a much less extent than in the south and west. As a rule, the walls are finished off with excellently dressed stones neatly fitted together. But these facing stones are not rectan¬ gular blocks such as are found at Copan. The outer face of each block is rec¬ tangular, while the inner part is roughly shaped into a tenon and set into the mortar and rubble hearting. The blocks come in contact only along the outer edges. Although having the general appearance of stone masonry, these facing stones have no real structural value. Much fault has been found with the ancient builders for not breaking joints, but with stonework of this character it really makes no difference whether or not the joints are broken. The mosaic veneer character of the stone surfacing has an important bearing on any criticism of the architectural decoration. For if the stone facing had no structural character, and was frankly considered mere veneer, then the fagade decorations were not limited in any way by considerations of mechanical fitness. The ornamental stones could be applied as mosaic without the necessity of main¬ taining any structural lines. The whole surface of the building became a fair field for unlimited fancy. Such seems to have been the understanding, since even the cornices and string courses of Maya buildings had no real virtue as binding stones, and were apparently intended for adornment alone. The mosaic elements used in fagade decoration have roughly hewn tenons that were set into the walls. Vaults. The Maya vault has usually been described as a corbelled or false arch, built not upon the side-thrust principle of the keystone, but upon the down¬ ward thrust of a load upon over-stepping stones. The principle of such a vault was doubtless understood by the Maya builders. Plate 4, fig. 1, reproduces a photograph of a small chamber at Copan that was formerly vaulted. The long neatly cut roof stones may still be clearly made out. Such stones are long enough and broad enough to allow considerable purchase along the contact planes, and the arch of this chamber was doubtless of the corbelled variety. But although the corbelled arch was known and used to a slight extent, the typical Maya vault was monolithic in character through the liberal use of cement, and was intended to be so by the builders. Plate 4, fig. 2, shows an excellent natural cross-section of a typical Maya vault of northern Yucatan. It will be noted that the stones are fairly well cut on the outer surface, but that they have no pur¬ chase upon each other, since only the merest edges come in contact. The stones are held in place by the mortar of the filling and the vault is in effect monolithic. The mortar used by the Maya seems to have been rather variable in quality. Sometimes the hearting was made exceedingly tight and resistent, but in other cases the mortar was badly mixed with earth and, after the outer coating of plaster had fallen away, the roots of trees were often able to force their way into the chinks, scale off the veneer and even disrupt the walls. But it is ex¬ ceedingly doubtful if any kind of construction could have resisted better the » 1895-1897, p. 198. CONSIDERATION OF THE MATERIAL ARTS. 109 tropical conditions of heavy rainfall and luxurious vegetation than the mortar construction we have just examined. Certain other features which increase the stability of Maya vaults deserve mention. The center of gravity for each half of the typical vaults of northern Yucatan often falls within the limits of the supporting walls. As a result, one half of a vault is frequently found standing when the other half is in ruins. The weight over the capstones that bridge the five or six inches between the two halves of the vault is usually very light. When the building is two rooms deep the center wall is very stable, since the overhang is equal on the two sides of the wall. This device of reducing the strain to a minimum has been interpreted by some as an understanding of the principle of the cantilever, but a careful exam¬ ination- will show that such is not the truth. The fagades of the buildings of the Usumacinta region have a sloping upper zone, and as a result the front wall has nothing to balance the vault overhang. The vaults in all Maya cities show one constant feature. Always, or at least in the vast majority of cases, there is a projection of a few inches at the springing of the vault on the inside, that is, the widest part of the vault is per¬ haps six inches narrower than the width of the room. Now this persistent fact doubtless reflects a universal method of construction. It seems probable that the vault was built over a wooden form, and that the shoulder projection at the spring of the vault was to give a few inches leeway to permit the ready removal of the false work. The walls may also have been made inside a wooden frame which held the veneer in position till the slacked lime mortar had set. Wooden struts were often used. Sometimes these are still found in place. 1 The application of the monolithic arch to other uses than the roofing of rooms was rare, but a few interesting cases may be noted. It was not used for windows, but windows are an almost unknown feature in Maya buildings except in the roof structures. It was not ordinarily used in doorways. At Palenque, however, there are notable examples of vaults over interior doorways. Similar arched openings also occur in the medial walls above the spring of the vault. Several of these arches have a peculiar trifoil shape. The half arch built against a fagade was used in a number of sites to afford a narrow passage under a stairway. A well-preserved example is found at Chi- chen Itza under the stairway that ascends to the uppermost range of the Monjas. The arch was also used in aqueducts and small bridges. But perhaps the most interesting employment is in the independent portal or triumphal arch, examples of which occur at Santa Rosa Xlabpak (Fig. 143), Kabah 2 and Labna. One view of the famous portal arch at the latter city is pre¬ sented in Plate 15, fig. 2. Somewhat allied to this use of the arch are the vaulted passages in several long buildings at Uxmal, such as the House of the Pigeons, the South Range of the Nunnery and the House of the Governor. The arches in the last-named building (Plate 5, fig. 2) have been blocked up. It is important to note that the vault as described in the preceding pages is peculiar to the Maya culture and is found in all parts of the area. Although very narrow vaulted chambers occur at Monte Alban and at a few other sites outside of the Maya area, yet the fundamental principles do not seem to have 1 The interesting study of the Genesis of the 2 Stephens, 1843, I, pp. 399-400. Maya Arch by Mr. E. H. Thompson appeared too late to be of service in this discussion. 110 MAYA ART. been grasped by any of the neighboring peoples, although other details of con¬ struction were readily imitated or developed. Mrs. Nuttall 1 suggests that vaulted rooms of the Maya type may have been in use on the Island of Sacrifices in the harbor of Vera Cruz, but this point cannot be regarded as settled. The photographs of buildings which are reproduced in Plates 3 to 16 offer an abundance of proof on the principal points that have been made con¬ cerning wall and vault construction. It is obvious that most of the surfacing stones do not support any weight, because other stones beneath them have fallen away. Nearly all of these structures show scaling off of the facing stones, now in the lower zone of the facade, now in the upper zone, and now in the vaults themselves. The inevitable conclusion is that the plain and sculptured blocks were purely and simply a veneer and a mosaic, and that once the mortar had hardened the entire structure was as a single stone. Roof Structures. Not content with the amount of space for decorations afforded by the facade of the temple, the Maya builders greatly enlarged the area by raising a wall-like superstructure upon the roof. This superstructure is found ■ 1910, p. 269. CONSIDERATION OF THE MATERIAL ARTS. Ill in somewhat different forms in different cities, and from its most common type is often called a roof comb or roof crest. It will be described here in the probable order of the development. This device for overloading temples with ornament probably was not used at Copan and Quirigua. To be sure, the buildings in these two cities are so com¬ pletely dilapidated that little beside the floor plan can now be made out. But the amount of debris is not enough to justify the assumption that the buildings were of more than one story. Moreover, the number of sculptured stones, while considerable, does not demand more space than the upper walls of an ordinary fagade could give. As a rule, the sculptures on roof structures were of stucco, a material little used in these southern cities. At Tikal the principal temples, crowning very steep and high pyramids, were themselves topped by a lofty roof structure, which, like the pyramidal base, rose by a succession of narrow terraces. The back wall of the temple appears almost vertical and the greater part of the terrace recession is from the front. In order to support this massive superstructure the temple walls had to be made very thick indeed. The proportion of room space to wall space is much smaller in Tikal than in any other city. From its cumbersome nature we may reason¬ ably conclude that the Tikal roof structures represent the first attempt in this direction by the Maya. The zones of the roof structures appear to have been ornamented with mask panels. During the recent researches of Dr. Tozzer at Tikal he was fortunate enough to discover two sealed chambers in one of these roof structures. If the plan of the ancient builders was symmetrical, there must be two other similar chambers. A cross-section of this temple is shown in Fig. 147. The four rooms are in two stories and are entirely inclosed by the walls, so that no evidence of them appears on the outside. Their obvious purpose was to lighten the enormous load of the masonry. The temple where this discovery was made has only one very small open chamber and an almost unbelievable large proportion of solid wall. The other temples of the same type have two or three open chambers and a proportionally smaller volume of solid wall. It seems likely that these buildings are of somewhat later date and show an increase of skill in handling the mechanical difficulties. At Nakum the roof structure on one of the temples appears as three massive towers. In each one of these towers is a small sealed room. At either side of the room, but not connecting with it, are three rectangular perforations, one above the other, that pass completely through from the front to the back of the towers. At La Hondradez roof towers also appear as well as an example of a continuous roof structure. The latter has one feature of peculiar interest. The chamber of the temple is directly under it, and the side walls of this chamber run up to a very great height inside of this lofty roof structure. The roof structures of the Usumacinta region are typically of a much lighter and more airy construction. The simplest form is a vertical wall, pierced with windows, that rises from the center line of the roof. It seems possible to pos¬ tulate the course of development of the roof comb at Yaxchilan from a compari¬ son of the structural remains. Naturally the builders gained experience with each new attempt, and so the relative success with which the same problems of construction were met probably indicates time sequence in building. In the first place, there is the roof comb constructed over the single room. To bear 112 MAYA ART. the additional weight of the roof comb, interior buttresses were built which divided the room into a series of compartments, as may be seen in Fig. 148, a and c. In one group of such buildings it was apparently considered unsafe to put the roof comb directly over the ridgepole; it is therefore set somewhat back. In a second group the roof comb is over the center line, but the interior but¬ tresses are still necessary. A third group shows the temple divided by a longi¬ tudinal partition into two rooms. The roof comb then arises with perfect safety over this central wall. After the discovery of this simple and economical device the builders must surely have dropped the old clumsy method. But the develop¬ ment was not yet complete. Heretofore the roof combs have been narrow, ver¬ tical walls pierced by windows. 1 In the fourth group the roof comb is made by Fig. 148. — Plans of Yaxchilan Temples: a, Structure 25; b and c, Structure 33. bonding together two walls which incline towards each other. These walls are pierced by windows, as before. The greater breadth and increased stability permitted these structures to be raised to a height of at least two stories. The lower story of the roof comb is like a long corridor. Fig. 148, b, reproduces a cross-section of a Yaxchilan temple of the single-room type crowned by a very cumbersome two-walled roof structure. At a ruin on the Tzendales River there is a temple with a roof comb resembling an open corridor with six windows on each side. This structure and the ones at Yaxchilan that have just been considered are undoubtedly related to the roof structures of Tikal. The sealed rooms of the latter city in the course of develop¬ ment appear to have opened out into corridors with windows. The roof combs of Palenque have often been described. Maudslay gives complete plans of them. 2 They show the highest refinement in the use of the two walls inclining inward, the mass being reduced to a minimum, so that the whole structure may be described as stone trellis work. The weight of these roof combs is borne by the medial longitudinal partition. In every constructural feature, particularly in economy and efficiency of support, and in artistic refine¬ ment, the temples of Palenque are superior to anything else in the Maya area. 1 It must be admitted that Maler’s plans and 2 1889-1902, IV, pis. 65 and 85. See also Holmes, descriptions are not very definite in regard to the 1895-1897, p. 201. single-walled roof comb. CONSIDERATION OF THE MATERIAL ARTS. 113 The roof structure in the cities of northern Yucatan received a sort of dif¬ ferential development. The double-wall type apparently does not occur. The castellated roof comb of the House of the Pigeons at Uxmal (Plate 6, fig. 1) shows a modification of the single-wall type. Roof crests more like those of the south are seen at Sayil 1 and Hochob." A roof comb with rows of windows and cornice¬ like mouldings is seen upon the Casa Colorado at Chichen Itza (Fig. 149). Indeed, this building is somewhat unusual in that it has both a roof comb and a so-called flying fagade. The flying fagade, which is the most common form of roof structure in northern Yucatan, is really a vertical extension of the front wall of the temple, giving a false impression of the height of the building. Although not so beautiful as the roof comb of the Palenque type, yet the flying fagade served better to carry the mask panel decoration so common in northern Yucatan. Plate 15, fig. 1, pictures Fig. 149. — Plans of Casa Colorada, Chichen Itza. the Iglesia of Chichen Itza with its flying fagade which is decorated on the front with mask panels and on the rear with a simple lattice design. A graceful flying fagade of lattice work surviving upon a badly ruined temple at Sabacche (Plate 6, fig. 2) gives evidence of the excellent construction of this region. 3 In Plate 16, fig. 3, is shown a splendid building at Uxmal with tower-like elevations over the doorways instead of a continuous flying fagade. Columns. The development of the column is closely connected with the handling of doorways to allow for the admission of more light to the inner cham¬ bers. The column does not occur at all in the southern part of the Maya area and is none too common in the northern parts. At Palenque its prototype exists m the rectangular piers, all that remains of the front wall of the temple when three doorways are taken out. In northern Yucatan square columns occur, particularly at Chichen Itza, but round ones with a square capital are perhaps more common. The columns are made up of several drums or sections. Good examples of wide doorways with two or three columns occur at Labna, 4 Sayil, 5 Dsehkabtun, 6 Chacmultun (Plate 7, fig. 1) and Tuloom. 7 In many buildings the use of round or square columns really turns the outer chamber into a portico, as has been already explained. Plate 7, fig. 2, shows an intricate interior at Chichen Itza, with rows of drum columns and other unusual features. Columns for inte¬ rior roof support are rare, but occur at Chichen Itza if not at other cities. The highest development of the column is reached in the serpent columns of the Cas- 1 Stephens, 1843, II, p. 25. 2 Maler, 1895, p. 285, central building shown. 3 For other examples see Stephens, 1S43, II, pp. 50-53 (Labna), and Maler, 1895, p. 253 (Chunydx- nic) and p. 254 (Sabacche). 4 Stephens, 1843, II, frontispiece. 5 Stephens, 1843, II, p. 17. 6 Maler, 1902, p. 227. 7 Stephens, 1843, II, pp. 402-403. I 114 MAYA ART. tillo, the Temple of the Jaguars and the Temple of the Tables, all at Chichen Itza. These strongly resemble the serpent columns found by Charnay at Tula. In the Court of the Columns at Chichen Itza are hundreds of columns, four or more abreast, in long alignments. Their purpose is unknown, but it seems certain that they did not support vaults. It is possible that they supported flat roofs and that the whole group formed a sort of open marketplace. The presence of temples in the group is an Y argument against this hypo¬ thesis. A number of similar groups of smaller size exist in northern Yucatan and will be discussed later in another connection. The small plain or banded columns so much used as a fagade decoration in northern Yucatan must be clearly dis¬ tinguished from the columns that serve as supports. The former have no structural character, but are simply mosaic elements used in architectural embellishment. Cornices. The cornice is one of the most interesting features of the Maya build¬ ing. The term must be used somewhat widely. Not only are the terraces of the pyra¬ mid and the upper portions of platform mounds orna¬ mented by projecting tiers of stone which may be referred to by the term, but the build¬ ings themselves often show cornicelike projections at sev¬ eral levels. The moulding commonly called the medial cornice separates the upper and lower zones of the fagade. The cornice shows a refinement in form and an increase in variety from south to north. A series including the most important varieties of cornice forms is shown in Fig. 150. In the Copan substructures the cornices consist of a simple pro¬ jection of two courses of stone (Fig. 150, a) at the tops of the terrace walls. In the cities of the Usumacinta Valley the type is varied in that the lower part is beveled and projects farther than the upper part, as in b. In northern Yucatan the main type is the three-part cornice shown in c. Buildings with the sloping upper range have cornicelike projections at the eaves and also at the top of the sloping portion (Fig. 150, e to i). Fig. 150. — Cornice forms: a, southern area; b, c, northern area; d, h, i and j, Chichen Itza; que; k, Labna; l, Chacmultun. Usumacinta area; e, f and g, Palen- CONSIDERATION OF THE MATERIAL ARTS. 115 The eaves cornice becomes the medial cornice when the wall becomes vertical (j to l). The three-part cornice is much modified by the separation of its members and the introduction of a mosaic moulding of zigzags and other geometric ele¬ ments ( d , k and I). The Monjas and the Iglesia at Chichen Itza furnish exam¬ ples of these modified cornices (Plate 13, fig. 2, and Plate 15, fig. 1). In the Carocol at Chichen Itza the three-part cornice is developed into a five-part cor¬ nice (t). It is worth noticing that the three-part cornice is identical in profile with the wrapping of the banded columns so common in the fagade decorations of the buildings of northern Yucatan. Wooden poles used as vault struts some¬ times have similar bands carved upon them. Doorways and Wall Openings. The most interesting forms of doorways have already received an oblique presentation under the discussion of the develop¬ ment of the portico and the use of the vault and the column. Fundamentally, Maya doorways are of the simple post and lintel type, but after the concrete walls of the building had hardened there was very little weight upon the lintel. The stones that form the jambs are often of larger size than the usual run of building stones. The lintels are of the very durable zapote wood or of stone. The use of wooden lintels was a decided element of weakness, because after the decay of the lintel the mortar conglomerate over the doorway was often unable to bear up under its own weight. It is probable that no doors were hung in any of the wall openings, although curtains may have been used. Occasionally the original three faces of each door jamb (Fig. 151, a) were increased by a simple modification. At Uxmal there are examples of doorways with the jambs modified as in b, while at Chichen Itza a pilaster was sometimes set up to carry a shorter under lintel so that the cross-section of the door jamb is like c. This device allowed seven vertical panels for decorative purposes. With the extended use of piers and columns, as we have already seen, doorways are widened until often the whole front of the building resembles an open portico. Typical doorways are shown in many of the photographic plates, and clear drawings of the different types are given by Mr. Holmes. 1 The use of a vault over an interior doorway is found at Palenque. 2 Wall openings other than doorways are almost negligible when considering the Maya area as a whole. Small rectangular or tau-shaped windows occur. Perforations in the medial walls above the spring of the vaults are characteris¬ tic of Palenque, but are not found at most other sites. Application of Decoration. In all parts of the Maya area the fagades of buildings were richly decorated. The upper zone of the fagade, whether of the sloping or the vertical type, seems to have been the favorite place for applying decoration. This zone was turned into a wide frieze for designs of many sorts expressed in high-relief stone sculpture, in stucco modeling and in realistic or geometric mosaics. Occasionally the lower zone was also covered with ornament. 1 1895-1897, pp. 4044. ■ Maudslay, 1889-1902, IV, pi. 5. I 4 Fig. 151. — Modifications of door jambs; a, common form; b, Uxmal; c, Chichen Itza. 116 MAYA ART. The surfaces of the roof combs and flying fagades carried decoration of the same diverse character. In many cities these outside designs have been almost com¬ pletely destroyed by the elements. Some of the more complicated interior decorations were used to enhance the inner chamber which has been described as the sanctuary. Stone and wooden lintels over the outer and the inner doorways of temples were frequently carved with remarkable pictorial compositions. These sculptures were generally on the under side of the lintel and so directly overhead. They could not be viewed with comfort or accuracy. In a few cases the front of the lintel block was also ornamented. Carved lintels of wood or stone have been found at Tikal, Yax- chilan and other Usumacinta sites and at Kabah and Chichen Itza. The front wall of the inner chamber was sometimes elaborately ornamented at Copan. The door jambs and the narrow spaces either side of the inner door and the interior columns were also sculptured in some of the more splendid tem¬ ples, particularly at Chichen Itza. But the most successful and artistic decora¬ tions are those which in several instances were applied to the inner walls of sanctuaries. The famous tablets of Palenque belong to this type of architectural enrichment, as do the frescos of the Temple of the Jaguars at Chichen Itza. Stelae with altars were correlated in many cases with temples and should be considered as a secondary architectural feature. Pyramids and stairways also deserve mention in this connection, although they are treated separately. Realistic Decoration. Architectural decoration may be divided conveniently into two divisions. 1st, fagade decoration; 2nd, interior decoration. The designs employed in the second division by the Maya do not lend themselves to comparative study from an architectural point of view. Those that empha¬ size the importance of the sanctuary have already been commented upon. Sub¬ jectively the range of the interior designs is wide and the manner of presentation realistic. The most striking door jamb, lintel, and sanctuary decorations have been treated already under different headings and do not deserve further discussion. Fagade ornamentation offers a rich field for comparative study. It may be subdivided, according to manner, into realistic, conventional and geometric, as these terms are commonly applied. The markedly conventional will be studied under the captions of “mask panel” and “profile panel,” and the minor conventionalizations of the more realistic designs will not be considered. Few buildings of the Maya area that were decorated in a free manner with realistic designs are now in a well-preserved condition. The method seems to have been more characteristic of the south than of the north. At Copan the fagades were apparently decorated in a free manner with human and grotesque figures, the latter sometimes representing divinities, and with feather drapery. The arrangement of these sculptures on the walls is only known in part. The upper zone of the fagade apparently carried most of the ornamentation. Temple 22 had a frieze of splendidly carved busts that were possibly arranged in a line with equal intervals of blank wall. At each corner of this building were two great heads, one above the other, made of several stones neatly fitted together. Temple 32, which is situated on a low mound just south of the acropolis, was decorated by more or less realistic carvings representing skulls, heads of the Roman-nosed God, serpent heads in profile, human figures and feather mouldings. The larger sculptures were made in several pieces, each with a long tenon that could be CONSIDERATION OF THE MATERIAL ARTS. 117 set in the wall. On this building, as on others, the head and headdress of human figures were sometimes carved on one block, the bust or torso on another and the two legs on a third and fourth. It is worthy of note that true gargoyles, serving as water spouts, occur at Copan. An example is given in Fig. 152. Judging by the debris, the temple on Mound 26, which is the one approached by the famous Hieroglyphic Stairway, was embellished by feather drapery sculptured in the freest manner imaginable. There were also sculptured human bodies, as well as faces of both the Long-nosed God and the Roman-nosed God. At Tikal less elaborate decoration limited to bands or friezes is still to be seen on two or more buildings. 1 One frieze that is interesting is seen on the rear of the central temple of a row of seven temples (Structure 55). The design con- Fig. 152. — Gargoyle in form of serpent head: Copan. sists of five parts. At each end are two bones modified so as to form an over¬ looking pattern, next comes an object that probably represents a shield, while in the center is a much destroyed figure that may present some sort of face sur¬ rounded by feathers. At Palenque free decoration executed in stucco is seen on the piers of the Temple of the Inscriptions and of some of the Palace structures. The designs are presented in a most realistic manner in panels that are framed in by bands of astronomical and other symbols. Some of these designs have already been pre¬ sented and discussed in the General Consideration. Maudslay 2 gives photo¬ graphs and drawings of all these mural decorations. He also gives a drawing 3 of the much destroyed stucco frieze on the sloping upper zone of the Temple of the Cross. This frieze represents a dragon head in front view with a leg at each side and with fish attached to the headdress. The stucco ornaments on the roof combs of Palenque are too badly destroyed for reproduction. Waldeck 1 gives a drawing of one that shows atlantean figures supporting the cross beams. Mr. Holmes 6 shows in a drawing the stone skeletons that were used for the larger stucco sculptures and the method of attachment to the walls. In northern Yucatan the surviving use of the sculptures of human figures and other realistic motives on the facades of buildings is seen in a number of instances, although the general method of decoration here is very formal. The Iglesia atChichen Itza (Plate 15, fig. 1) has on its middle zone two panels each with two seated figures representing anthropomorphic gods. Bodies attached to the wall by tenons occur at Uxmal. On some of the flying facades at various sites in northern Yucatan the same style of sculptures seems to have been used. At intervals, in the geometric and conventionalized ornament that adorns the temples of Uxmal, are details that are strongly realistic. A sculpture of the sort 1 Maler, 1911, pi. 8. 2 1889-1902, IV, pis. 8-11, 27-28, 53-56. 3 1889-1902, IV, pi. 68. 4 1866, pi. 26. 6 1895-1897, p. 198. 32-37 and 118 MAYA ART. may be seen in Plate 8, fig. 2. It represents a personage with a lofty feather headdress seated over the open jaws of a serpent. The carving is executed on several blocks which are fitted together. Plate 9, fig. 1 shows a portion of an interesting fagade at Uxmal, where the intertwined bodies of two serpents overlie the geometric and mask panel decoration. Realistic designs which are too indefinite for study are shown in Cather- wood’s views of Tuloom. An interesting feature on many fagades in northern Yucatan is the niche which was intended to protect or embellish a seated individual repre¬ sented in stone or stucco. Slight remains of such seated figures can still be made out in some in¬ stances. The niches usually take the form of little houses, with two sides and a roof, and are Fl ° lo3 fa^SeTuxmS 1 mche m frequently decorated. An example after Cather- wood is given in Fig. 153. The finest development of the niche is seen at Labna, Chacmultun and Uxmal (Plates 7, fig. 1, and 16, fig. 3). The niche is often placed over the doorway, and it seems likely that the figure enclosed in it was that of some deity. Mask Panel. The use of the mask panel is the most noteworthy character¬ istic of Maya fagade decoration. The mask panel is essentially a highly conventionalized face, represented in front view, with its de¬ tails so modified as to fill an oblong panel. This Fig. 154. — Faces limited to rectangular spaces: Chichen Itza. panel either extends along the wall surface or folds around a corner. In the case of the corner masks the relief is ordinarily higher than in those on a flat wall. Before considering the mask panel as it is used in archi¬ tecture it might be well to examine the general application of faces to rectangular areas. Such areas frequently occur at the bases of stelae, on sculptured door jambs, etc. Fig. 154, a, shows a face occurring in a long decorative band. The sides of the face are not framed in, so the design as a result is not strictly rectangular. Another face with many similar features is given in b. This face is framed in on all four sides, and the parts extend into all the corners of the area. The changes which occur when a face in profile is turned into front view is illustrated in Figs. 155 and 156. The profile face is taken from the side of Stela B at Copan. Note the pendent nose, the curled object at the side of the mouth, the oval ear plug with inferior and superior ornaments, the feathered eye and the hair or feathers on the forehead. The front-view face (Fig. 156), occurring on the base of Stela 4 at Yaxchilan, retains most of these Fig. 155. — Face in profile: Copan. CONSIDERATION OF THE MATERIAL ARTS. 119 features modified to suit a rectangular area. The featherlike details of the forehead and the feathered eyes are easily seen. The nose hangs down over the mouth. The teeth are replaced by the tips of feathers, a rather unusual substitution. The curled object at Fig. 156. — Mask panel at base of stela: Yaxchilan. each side of the mouth occurs, how¬ ever, as well as the oval ear plugs with decorative appendages above and below. A somewhat similar panel at La Hondradez is given in Fig. 157. Others of the same type occur in this city, likewise on the bases of stelae. The rectangular panels on the bases of the stelae of Quirigua are often much more complicated than the examples just given. They present two or three superimposed heads, the upper ones being the headdresses of the lower ones. 1 In the southern portion of the Maya area the panel treatment of the fagades of temples held a certain vogue, but the details were applied for the most part in stucco and so have since crumbled away. At Copan there were apparently two superimposed mask panels on the corners of Temple 22. These are now very incomplete, and the upper one is represented only by the ear ornaments in the photograph taken by Maudslay. 2 Part of the lower mask on one of the other corners is shown in Plate 3, fig. 2. Aside from this example the architectural decoration at Copan seems to have been in another and freer style than of the mask panel. Portions of mask panels on the fagades and roof struc¬ tures of Tikal, Nakum, Yaxchilan and Palenque can still be made out. The details of mask panels at Nakum and La Hondradez have been furnished by Dr. Tozzer. Nearly all of these designs are incomplete, but they are decidedly interesting in showing a less trammeled hand than the anal¬ ogous designs of the north. Three of these panels are given in Figs. 158-160. The most complete one (Fig. 158) is a detail on one of the towers of the roof structure. The face is very simple, the most notice¬ able feature being the large nose plugs at the base. The second and third examples have the pupil of the eye represented by a spiral groove, while from the upper part of the ear plug ex- 1 Maudslay, 1SS9-1902, II, p. 10. * 1889-1902, I, pi. 17, c. Fig. 158. — Mask panel on tower: Nakum. Jr * 1 cvj r- All /p! M ~VM lgJ XT G3 Fia. 159. — Mask panel partly restored: Nakum. 120 MAYA ART. (ffffM oljjp P) SI v "4 L:i! to x 'oj ( ‘l Fig. 160. — Mask panel: Nakum. tends outward a bunch of feathers. These masks were built out of specially carved stones, but the finer details were expressed in stucco which was applied to the surface of the stones. The mask panels at Tikal, Yaxchi- lan and Palenque were largely con¬ structed of stucco. At the latter city, on the frieze of House C, of the Palace, a row of seven faces can still be made out. 1 These are less rectangular than the usual run of mask panels. On an inside wall of the same building are nine other faces in a much better state of preservation. Judging by the examples which Maudslay 2 gives, each one of these faces was different from the others. All were finely modeled in stucco. Realism is not a marked characteristic of the mask panels, yet it seems likely that a number of conventionalized representations must be included under this general heading. Ex¬ amples of two designs from Labna illustrate the most realistic panels, following the serpent model, en¬ countered in northern Yucatan. The first of these (Fig. 161) is a corner mask, built up mosaic fashion out of many carved stones. In the open jaws at the front appears a small human head. Above the upper jaw the nose rises in a scroll, and back of this is seen the eye decorated with feathers. The rest of the face is a hodgepodge of sculptured stones that do not seem to represent natural features. The next mask on the Fig. 161. — Corner mask built up mosaic fashion: Labna. same building (Fig. 162) is in front view. The jaws are much less prominent, although both protrude slightly. The nose is much enlarged. The ear ornaments at each side of the face are not complete, but the sug¬ gested forms are more in keeping with those on the more usual mask panels. A complicated mask lacking the lower jaw is presented in a some¬ what restored condition in Fig. 163. The nose projects hardly more than the teeth. The headdress is ornamented with checker-work. The ornaments at the side of the ear plugs are unusually elaborate for a mask with as much realism as this one shows. The mask is centered over a doorway which is the position first in importance to be filled by such designs. i Maudslay, 1899-1902, IV, pi. 20. 2 1S99-1902, IV, pi. 24. CONSIDERATION OF THE MATERIAL ARTS. 121 Fig. 163. — Mask panel: Xkichraook. Masks in which human features predominate are seen in the stucco panels of Izamal preserved in drawings by Catherwood 1 and Holmes. 2 A great human face modeled in stucco appears on the tower at Nocuchich, which has been de¬ scribed and photographed by Maler. 3 From the examples so far presented the mask panel appears to have had a diverse origin. But, as a rule, it seems pretty clear that the mask represents the feathered serpent. The eyes often show feathered lids. The projecting curl represents the nose of the serpent, which, as has been observed, was com¬ monly elongated. The serpen¬ tine head may of course have been intended for that of the Long-nosed God. Fig. 164 gives an ideal mask in its most simplified form but with all the parts. It may be well to run over the main features of the complete mask. The head band usually represents a series of threaded disks or a line of rosettes, the middle and terminal ones being more ornate than the others. This head band extends over the eyes and ends above the ear orna¬ ments. The eye consists normally of three parts, the upper and lower lids and the eyeball. There are two principal forms of the eye, one being round and the other rectangular. In the round eye the upper and lower lids are equal in height and shape, while in the rectangular eye the upper lid is a straight bar and the lower lid is trough-shaped and includes the sides. The eyeball is frequently repre¬ sented with a forward and backward part, the former rep¬ resenting drooping feathers. The nose is curled in nearly all cases, but presents a con- 1 Fia. 164. — Simplified maBk panel with all the siderable Variety in profile. J usual parts. The superior nose ornament =eL is usually either a roll-shaped body or a human face. The curious detail is the homologue of the nose scroll on the profile serpent head (Fig. 30, e). Through this object the nose plugs were thrust. Two of the masks on the flying facade of the Iglesia at Chichen Itza still show nose plugs (Plate 15, fig. 1), but they are usually omitted on mask panels. The mouth varies in many details. The lips are much reduced. The teeth are of two kinds; those at the side of the mouth correspond to molars and those on the front to incisors. These teeth have frequently been described as filed, but a comparison with those on the profile head of the serpent proves that the tradi¬ tional method of representing teeth was followed in these architectural designs. The lateral mouth ornament corresponds to the curled fang at the back of the 1 l 844 , pi. 25. * 1895-1897, p. 99; Charnay, 1885, p. 262. 3 1895, p. 289. 122 MAYA ART. mouth on the more realistic representations. The ear plug is usually square and fitted with a peg in the center. The inferior ear ornament represents a pendant, while the superior one varies widely in form. The lateral ear orna¬ ment is very important in emphasizing the formal quality of the mask. It usually consists of two frets turned in opposite directions and separated by a horizontal object. These frets possibly symbolize feathers. The general processes by which designs are modified, namely, simplification, elaboration, elimination and substitution, have already been explained. The general conceptions of the mask panel and its diverse origin have been touched upon, as well as the convergent results attained by the process of simplifying the original forms and throwing them into a geometric order. In the compara¬ tive study of the mask panel it is necessary to take strict account of the changes which take place in ho¬ mologous parts. The ideal simplified mask may be made to serve as a standard for this comparison. Some of the mask panels of Chi- chen Itza are very simple, and others show adornment of the different fea¬ tures. Examples may be examined in Plates 13, fig. 2; 15, fig. 1; 27, fig. 2, and 28, figs. 3-6. Pleasing elabora¬ tion is shown in the last illustration. Fio. 165. — Superimposed mask panels: Uxmai. Better examples of highly elaborated masks are seen at Uxmai (Fig. 165 and Plates 8, fig. 2; 9, fig. 1; 14; 16, fig. 3). From the ebb-tide mask of greatest simplicity, the use of subsidiary ornament in elaborating each separate element becomes more and more prominent. The eyelids are ornamented with circles or other figures. Similarly the head band becomes a row of rosettes instead of simple disks. The nose takes on adventitious details, such as crosses and swas¬ tikas. The various ear ornaments assume a great variety of shapes, the lateral ones sometimes developing into serpent heads. The same masks that show elaboration often show elimination as well. Elimination in the case of the mask panels seems to proceed by a pretty definite rule. The outer features are the ones that are cut off, but the process may con¬ tinue till only the eyes and nose remain. Space considerations have something to do with elimination in many instances. The lateral ear ornaments are the most elastic features of the mask panel. The arms of the frets can be length¬ ened or shortened according to the space to be filled. If the space is very short, the lateral ornaments are left off entirely. Since the first consideration in placing a mask was to get it centered over the door, this elasticity counted. On the east wing of the Monjas at Chichen Itza (Plate 27, fig. 2) are five masks of varying width, three of which are placed over doors. The spaces not filled by the masks are given over to geometric decoration. On the east front of the same building the lateral ear ornaments are omitted entirely. The masks that cover the fagade of Structure 1 at Kabah (Plate 8, fig. 1) evidence further elimination. Not only the lateral ear ornaments but also the head bands are omitted. In fact, except in CONSIDERATION OF THE MATERIAL ARTS. 123 the lower tier of masks, each ear plug is held in common by two masks. Exam¬ ples might be multiplied. About the last stages of elimination are shown in Fig. 166. — Mask panels over doorway, showing extreme elimination: Labna. Fig. 166 and Plates 9, fig. 2, and 10, fig. 1. In the last photograph referred to the ear plugs and lateral ear ornaments are present but much reduced, and the teeth simplified to a notched line at the bottom of the face. In the other examples only the eyes and nose survive. The process of substitution overlaps that of elab¬ oration and is even seen in designs where elimination has had full play. An example of the latter is pre¬ sented in Fig. 167. Here several features are wanting, and a simple geometric design consisting of a line of squares standing on their diagonals replaces the mouth. In Fig. 168 the mouth is replaced by a double Fm. 167. — Mask panel showing fret, turning inward. Both of these figures are taken S *" and Bubstitutio,l: from the drawings of Cat.herwood. A more complete example of substitution is that given in Plate 10, fig. 2. On the lower zone of the fagade of the building here depicted are two panels made, mosaic fashion, 1 p n 124 MAYA ART. Fig. 168. — Mask panel showing substitution: Uxmal. out of separately carved stones. The panels are on either side of a door¬ way, and while each is slightly asymmetrical the error is reversed from one panel to the other, so that the design as a whole is perfectly balanced. The motive that resembles a letter C recalls the shell beads of the more realistic head band. The fret at each side of the central part of the panel occupies the same position as the eyes, and the tau-shaped grouping of plain and banded columns cor¬ responds to the nose. The teeth and the lateral mouth ornaments are also suggested at the base of the panel. It should be noted that there is no actual survival of a single feature, but simply a survival of the old characteristic order and assemblage. In Plate 28, fig. 2, is given a geometric panel from the upper range of the Monjas at Chichen Itza. The double frets at each side of the panel suggest the lateral frets of the com¬ plete masks on the lower range of the same structure. The face, however, is replaced by an arrangement of squares and drum columns. Characteristic geometric panels that offer only a vague suggestion of the mask panel are given in Figs. 169 and 170, and appear likewise on many of the buildings shown in the plates. The purely geometric decoration will be treated elsewhere. Profile Panel. We have seen that the mask panel is merely a front-view face, of any sort, definitely limited to a rectangular space. Theoretically it is quite possible to develop artistically a profile face in much the same way. Examination proves that such designs actually occur in Maya art. They are much less common than the front-view panel, probably be¬ cause of their necessary asymmetry. This is of course overcome by the opposition of two similar designs where such an arrangement is possible. The rectangular spaces at the bases of stelae are sometimes decorated with profile faces. A particularly fine example is reproduced in Fig. 171. It repre¬ sents the much elaborated face of the familiar Long-nosed God, looking toward the left. From the feathered eye issue two strands which pass to the bottom of the panel and thence to either side, where each ends in an attractive vignette containing a small animal figure. In the circlet at the left is shown a rat, or some such animal, and in that at the right a deer. Much smaller and simpler profile heads may be seen on Stelae 2 and 3 at Naranjo. 1 On Stela A 2 at Quirigua the lower panel contains a face, looking upward, which has almost completely broken down into meaningless scroll-work. Upward-looking heads of a peculiar type are seen on the bottoms of Stelae 6 and 10 at Yaxha (Fig. 172). The human being above may be said to stand on the open jaws of the serpent below. The inverted face ' Maler, 1908, 6, pi. 20. s Maudslay, 1889-1902, II, pi. 8. Fig. 169. — Geometric panel: Dsibiltun. Fig. 170. — Geometric panel: Dsibiltun. CONSIDERATION OF THE MATERIAL ARTS. 125 and the hands are hard to explain. While these examples carry us afield, they are evidences of the suggestions for profile heads to be used architecturally. Fig. 171.—Profile panel: Yaxchilan. Upon the In Temple II at Copan is represented a conventionalized serpent head, marked with death symbols, arranged vertically at the side of a door. Fig. 173 presents this design as it appears in one of Maudslay’s photographs. It is prob¬ able that a similar head was placed at the opposite side of the door, fagades of buildings in the Peten region ele¬ ments taken from profile heads sometimes appear. A frieze on the so-called Palace of the Five Stories at Tikal (Structure 10) consists of a rectangular eye repeated with intervals of blank wall. At Nakum there are scanty re¬ mains of what may be regarded as profile heads, although the}' do not appear to have Fig. 172. — Panel at base of stela: Yaxha. been distinctly limited to rectangular spaces. The best preserved of these is given in Fig. 174, taken from photographs and drawings made in the field by Dr. Tozzer. It shows an open serpent mouth which contains a human head and an extended arm. The upper part of the serpent head has fallen away, but the rectangular ear plug with its pendant is still in position. A noteworthy fea¬ ture of this decoration is that it is modeled in the stucco covering of the wall instead of being built up out of carved blocks in the manner of a mosaic. However, a second fragment at Nakum has carved stones fitted together. Something of the uncertain genesis of the profile mask panel may be gathered from the pre¬ ceding examples. But the principal occurrences where the geometric mould is unmistakable are in northern Yucatan. Here two profile panels assembled in opposition, one on each side of a doorway, unite with a front-view panel placed above the doorway to form a striking scheme for the decoration of an entire fagade. Fig. 173.—Highly conventionalized ser¬ pent head in profile: Copan. 126 MAYA ART. Plate XI gives views of two portions of the Palace at Hochob which has been explored and described by Maler. 1 The upper picture represents a small fagade with the lower zone plain while the upper zone bears a somewhat elaborate mask Fig. 175. — Profile mask panel: Hochob. principal difference lies in the wavelike figures that represent hair above the two intertwined serpents of the headdress. Below this face and on either side of the door is an elab¬ orated serpent face in profile, the whole cast into a strikingly rectangular mould. One of these profile heads is given in Fig. 175. It is an interesting example of elaboration, since the top of the eye is formed by a small complete serpent whose tail constitutes the nose plug of the greater head. The note¬ worthy features in the present connection are the teeth that project inward at the side of the door and the peculiar right-angled turn of the jaw. It should also be stated that the design does not completely fill a four-sided area and that details from other masks in¬ trude into the open spaces. A drawing of a broken-down fagade at the same city which shows the same elements treated in a simpler manner is given in Fig. 176. Another similar fagade, admirably preserved, and from another site, is re¬ produced in Plate 12, fig. 1. The splendid temple shown in Plate 12, fig. 2, is perhaps the clearest example of any. The doorway has been sealed up, and as a consequence hardly a stone has fallen from its place. The upper face may represent the Sun God with the ornamented tooth. The profile faces are perhaps more complete than any we have yet seen, since the short under-jaw is shown as well as the forked tongue which hangs below it. Two other important fagades of this type remain for consideration — one at 1 1895, pp. 278-279. Fig. 176. — Assemblage of profile and front view mask panels: Hochob. CONSIDERATION OF THE MATERIAL ARTS. 127 Uxmal and the other at Chichen Itza. The fagade at Uxmal is that of the Annex to the House of the Magician (Plate 13, fig. 1). No near-by view of the building as a whole is obtainable. Catherwood 1 gives a drawing of the fagade, which is accurate as far as it goes, but the upper front-view mask is merely suggested. Seler 2 reproduces this upper mask with all the details that now remain, and Le Plongeon 3 presents detailed photographs of the profile panels with which we are most concerned at this time. The three principal faces are much complicated by bands of astronomical symbols which overlay the important features and fill in the blank spaces. The one above the doorway is readily made out on account of the eyes. The profile face at either side of the doorway harks back to a type of serpent head that has already been discussed, namely, the more or less modified head with the nose turned back into a fret. The eye is rectangular and is partly concealed by a small human figure attached to the wall by a tenon. This figure is now much destroyed. The teeth of the profile serpent head project inward toward the door¬ way. The lower jaw that appeared on the simpler examples is lacking. To sum up, the assemblage of three faces shows the effects of all four processes of modification, but especially those of elaboration and elimination. The last fagade to be examined is the justly famous one at the eastern end of the east wing of the Monjas at Chichen Itza (Plate 13, fig. 2), and here elimi¬ nation and substitution have been carried much further. The only features of the mask panel over the door that survive in their original form are the lateral ear ornaments (see other mask panels on the northern side of the east wing, Plate 27, fig. 2). The face proper has been replaced by a seated human figure, in front view, with a drooping feather headdress. This figure is inclosed in an arch made mosaic fashion. Over the doorway is a row of projecting teeth sep¬ arated by mouldings from the figure above. Of the original side-view faces even fewer traces remain. The teeth at either side of the door are of the same character as the ones over it, but are homologous with the teeth of the profile faces in the earlier and more intelligible assemblage. The other features of the two profile heads are replaced by two front-view faces one above the other. The general outlines of the earlier grouping are pretty well maintained in the scheme presented in this fagade. The devices that appear on the upper member of the cornice may survive from the earlier scroll work representing hair. Maudslay 4 has expressed the opinion that the doorway represents an open mouth, but the series just given shows that it represents the surviving elements of an old arrangement of three heads, one in front view and two in profile. There is good reason to believe that the last two examples of architectural decoration in this series are later in point of time than the ones with easily recognizable features. Geometric Decoration. Many motives which are purely geometric occur on the buildings of northern Yucatan. The geometric panels that show affilia¬ tions with the conventionalized faces through the process of substitution have already been considered in some detail. As a rule, geometric figures are not limited to panels, but are applied in string courses or in all-over patterns. Each geometric element is usually carved on a single stone and combined in different 1 1844, pi. 11. 2 1908, p. 162. 3 1896, pis. 71 and 73. * 1889-1902, III, p. 17. 128 MAYA ART. ways to form different designs. Typical design elements are given in Figs. 177 to 180. The element given in Fig. 177, a, is used to form zigzags or squares set on the diagonal (Plates 8, fig. 1; 9, fig. 2; 13, fig. 2; 15, fig. 2; 27, fig. 1, and Fig. 178, b). The ele¬ ment with a cross (Figs. 178, a, and 180, b ) is much used in imitation diagonal trellis work which often fills in the spaces between mask panels, but which is sometimes used as the sole motive, as in Plate 6, fig. 2. Squares with figures of different kinds carved on them, rosettes (Fig. 179) and stepped pyramids (Fig. 180, a) are of frequent occurrence as in- Fig. 177. — Mosaic elements: Labna. dependent elements. In other motives, such the figure may occur on the same stone. The fret is usually of large size and is built up out of many plain stones. As a rule, there is an outer and an inner fret, the latter more or less sunken below the former, but still in relief against the wall. An example of a fret with the planes dif¬ ferentiated by shading has already been given (Fig. 170). The plain and banded column motive (Fig. 181) will be treated in special detail be¬ cause of its importance and frequent use. Typical string courses follow the line of the medial cornice. Usually the design shows sev¬ eral motives in rows, one beneath the other. Fig. 182 presents an interesting combination of moulding on either side of a beautiful head that is now in the Museo Nacional at Mexico City. At the top are feathers, next comes a represen¬ tation of vertebrae. The third row consists of banded columns, the fourth of assorted geo¬ metric motives, and the fifth of the stepped pyramid or wall of Troy motive in an inverted position. Frequently the cornice mouldings are three repetitions of b 4 > < ► ► ◄ _A, T > < A mmmsm Fig. 178. — Mosaic elements with ex¬ amples of their development: Chichen Itza. modified by the introduction of geometric ornament, such as zigzags, guilloches and short columns between the different members. The rich combination of geometric, conventional and realistic elements that defies description is seen particularly on the buildings of the Nunnery Group at Uxmal and on the House of the Governor at the same city. Photographs of the structures are given in Plates 8, fig. 2; 9, fig. 1; 14, etc. The most striking single feature is the fret which does not form long meanders but is arranged singly or in groups of two or three. The use of engaged columns, either plain or with simple banded ornament, ? element: Labna. 1 CONSIDERATION OF THE MATERIAL ARTS. 129 Fiq. 180. — Mosaic elements used in facade decorations: Labna. is characteristic of a great many buildings in northern Yucatan. These columns are in sections, and each section may be considered a mosaic element. In fact the plain and banded sections of the decorative column are frequently intro¬ duced into cornice mould- _ ings or into mask panels that show substitution. One of the banded sections is shown in Fig. 181. The tenion at the back is not present in all instances. Sometimes the entire upper range of a building is ornamented with banded columns placed close together between the medial and the true cornice mould¬ ings. Examples of such buildings are given in Plates 7, fig. 1, and 16, fig. 1. The columns may appear also in the lower zone, usually in groups and not in a continuous distribution. The detail of a facade decorated with banded columns is given in Fig. 183. In facades of this type the banded sections of the columns may occur at several heights. Three or more banded columns are frequently used to flank or frame in mask panels and doorways, the former in the upper and the latter in the lower zone of the fagade. Examples of such uses are shown in Fig. 166 and Plates 9, fig. 2, and 10. Stelae. The great monolithic monuments of the Fig. 181. — Typical banded sec¬ tion of the banded column. Maya commonly called stelae may have served in some cases as grave monuments, but if so this was decidedly a minor pur¬ pose. Small cruciform chambers have been found under a few of them containing re¬ mains of what might have been a founda¬ tion offering. The prime purpose of the monuments is very uncertain. They may have been idols in the same sense that the representations of Buddha are idols. It seems unlikely that they were monuments to individuals, first, because they lack in in¬ dividuality; second, because most of them bear dates that fall on even, half, or quarter katuns which correspond to intervals of about five years. They may have been connected primarily with the completion of a time period and secondarily with the his¬ torical events that took place during that time period or the gods that governed it. Whatever their true significance, it seems clear that as objects of art they may be put into two groups. The first group includes those that are apparently independent of temple structures, and the second those that serve as auxiliary temple adornment. Fig. 182. — Richly ornamented wall with pro¬ jecting sculpture: Uxmal. 130 MAYA ART. The independent stelae prevail at Copan, Quirigua, Tikal and some other sites in the Peten region, where they were set up, as a rule, in a great paved court or plaza. Before each stela there was in most cases an altar. Stelae of the second group also occur at these cities. The dependent stelae in Copan were likewise mostly set up in the Great Plaza, but were definitely correlated with some mound and generally placed at the foot of stairways leading to the temples. Stelae 3, D, M and N are examples of monu¬ ments with a secondary architectural character. Stela 3 is correlated to a mound in the Great Plaza, Stela 4 stands before a minor hieroglyphic stairway, Stela M is directly in front of the famous Hieroglyphic Stair¬ way and Stela N is at the base of the wide stairway leading up to Temple 11. All the stelae at Naranjo, Seibal, Y^axchilan and Piedras Negras seem to have been correlated with temples. The arrangement is more complex than at Copan. The stelae were usually placed upon the ter¬ races in front of the temples in symmetrical order, which, however, is hardly the same in two cases. The question of grouping will be considered again under chronological sequence. Stelae occur at nearly all the ruins in the southern and western part of the Maya area. Only one has so far been noted at Palenque. In northern Yucatan stelae occur at a few sites, such as Sayil and Tabi, but they are very unusual and are rudely carved. At Sayil the three stelae described by Maler 1 were set up on a low platform. Crude sculptures as well as plain pillars are also found in Chiapas. 2 The frequent occurrence of perfectly plain stelae in the Peten region has been noted by Dr. Tozzer. It seems possible that these may have been painted with figures instead of carved. It is not improbable that great wooden sculp¬ tures, comparable to the stelae, preceded these laborious monuments. Altars. The most widespread type of altar is drum-shaped, either plain or sculptured. In many cities this is the only kind that occurs. In Copan and Quirigua the altar was especially developed. First came rectangular or drum- shaped altars with wrappings and knots sculptured upon them in addition to human figures and other designs. 3 Apparently this type of altar was intended to represent a bundle. A similar altar painted on a pottery vessel will be shown hereafter. The Altar of Stela 4, at Copan, represents a single knot. The animal altars have been described. Most of them fall into the series of the Two-headed Dragon, but the Altar of Stela C 4 represents a turtle, and the Altar of Stela F represents two jaguars bound to the sides of two grotesque heads placed back to back (Fig. 99). There are a number of sculptures at Copan independent of stelae that may be called altars. Some of these are rectangular blocks with beautiful carvings on the sides and tops representing seated figures in rows, grotesque faces of large Fig. 183. — Detail of a facade dec¬ oration of banded columns Tantah. 1 1895, pp.'277-278. 2 Brinton, 1897. 3 Gordon, 1902, b. 1 Gordon, 1896, p. 40. CONSIDERATION OF THE MATERIAL ARTS. 131 size and masses of hieroglyphs. Others are thin vertical slabs with sculptures of reptiles. The animal altars of Quirigua (see Plates 1 and 2) seem for the most part to be independent of the stelae. They mostly belong to the series of the Two- headed Dragon, but one represents a jaguar and another a reptile’s head. A drum-shaped altar is also to be seen at Quirigua. At Tikal most of the altars are plain drum-shaped blocks. Altar 5, however, is finely sculptured on its upper surface. At the Usumacinta sites the altars are also usually drum-shaped and unsculptured, but there are one or two notable exceptions to both of these qualifications. Examples of more or less rectangular altars supported on carved blocks or sculptured heads may be seen at Piedras Negras. In northern Yucatan the table altar is highly developed at Chichen Itza. Here a flat stone is supported by from two to fifteen small stone sculptures. The low platforms or benches that are sometimes built against the wall in sanctu¬ aries or at the heads of stairways may have served as altars. Portable incense- burners of pottery were much used and probably were placed on the altars. Color. Like the Greeks, the Maya painted their stone sculptures and their stone buildings. There are still many vestiges of color. In some cases an entire monument or building seems to have been painted over by a single tint. In other cases details of ornament were picked out in contrasting tones. The colors were usually applied in a fairly definite way, red for flesh tones, blue and green for ornaments and green for feathers. In the Peabody Museum are many examples of carved feather-work, gro¬ tesque figures, etc., which were used in architectural decoration at Copan. They appear to have been surfaced with smooth plaster and then painted red. Suc¬ cessive coatings seem to have become so thick that they may have seriously impaired the beauty of the original sculptures. Red apparently prevailed at Copan, for Stela 4 likewise shows traces of this pigment. Maler notes many traces of color on the stelae of Piedras Negras. In this city there was considerable variety in the coloring, with the result that the de¬ tails of the complex sculptures must have been rendered much more intelligible. Thus the color remains on Stela 1 showed: 1 face, arms and garment, bright red; background, dark red; edge of garment, blue; breast cape, blue; feathers in all cases, green. Stela 7 2 showed the following color scheme: flesh parts and interior of ser¬ pent mouth, bright red; disks of head and breast ornaments, sky blue; feather- work, green; captive’s body, red. The feathers were painted green to represent the plumage of the favorite quetzal bird, the sky-blue disks may have been intended for turquoise or jade, while red gave the body a more natural appear¬ ance. Maler could find no traces of black, yellow or white. Maudslay 3 reproduces a painted stucco ornament from one of the rooms of the palace at Palenque, which shows decorative skill of no mean order. Miss Breton has recovered many of the vanishing traces of color on the reliefs of the Lower Temple of the Jaguars at Chichen Itza. They show a large variety of tones, by which the ornaments were clearly contrasted. The whole effect is one of rich tapestry. The bewildering detail which confuses when presented in one tone becomes perfectly intelligible when worked out in color. The remark- 1 Maler, 1901, p. 46. ! Maler, 1901, p. 51. 1 1889-1902, IV, pi. 18. 132 MAYA ART. able stucco reliefs of Acanceh were brilliantly painted when first uncovered. For preserving a record of the form and color of these reliefs thanks are due to Mrs. James, of Merida, and to Miss Breton. Many instances of fresco paintings upon a flat base are given by Stephens. 1 The fragmentary frescos of Chacmultun have been preserved by Thompson. 2 These are probably purely Maya, while the better known frescos of the Upper Temple of the Jaguars at Chichen Itza may show some influences from Mexico, particularly in the use of speech scrolls. These remarkable paintings have been drawn in fac-simile by Miss Breton. They represent a variety of scenes from the unceremonial side of life. A large number of figures are painted with an aston¬ ishing brilliancy of coloring. There is no reproduction of light and shade, but the painting as a whole is in tone. The background is green or blue and thus makes an admirable contrast for the warmer flesh tints. From an examination of the tones, which are numerous but always laid on flat, it seems probable that the artist made color blends, mixing his paint fresh for each piece of color. 3 In the main range of the Monjas at the same city are a few fragmentary paintings done in the same manner. Frescos have also been discovered at Santa Rita, 4 near Corosal, in British Honduras. These are executed in a style rather similar to the well-known frescos of Mitla, and seem to show strong foreign influence both in the manner of drawing and in the subjects. Prototype of the Maya Temple. The question of the probable prototype of the Maya temple deserves brief consideration. Viollet-le-Duc 5 finds evidence in the ornamentation of some of the stone temples of an earlier wooden construction. As is well known, decorative or utilitarian features developed in one material are frequently imitated in other materials, as, for instance, in pottery that sometimes takes over the designs used on baskets or textiles or imitates the natural forms of gourd vessels. It is also in evidence in the higher arts, for the Greek temples constructed of marble retained the shapes of earlier wooden parts as ornaments. In the case of Maya architecture Viollet-le-Duc finds in the fagade given in Plate 14, fig. 2, evidence of log-cribbing and lattice work. The analogy is close enough, but the ornamentation of this particular fagade is unique. Lattice work made in stone is very widespread in northern Yucatan, where, however, the build¬ ings are of a much later date than in the southern part of the area. The fagades decorated with plain or banded columns (Plates 10 and 16, fig. 1) suggest a wooden construction of upright poles such as is still used in the huts of the natives. The bands might represent in an ideal way the withes which bind these poles together. Here, again, the established chronology interferes with the ready acceptance of this theory. It seems reasonable to suppose that the original wall construction was of adobe, which was later faced with cut stone. Adobe bricks are widely used in ancient and modern construction from the Pueblo region on the north to Peru on the south. In many instances pyramids and other structures made of them are surfaced with a veneer of cement. This practice was clearly employed in the 1 See, for instance, 1843,1, pp. 204-205, 409-410; 3 Thompson, 1902; Breton, 1906, a. II, pp. 73-75 and 92-93. 4 Gann, 1898-1899, pp. 655-673. 2 1904, pis. 8 and 9. Other examples recovered 6 Charnay and Viollet-le-Duc, 1863, pp. 64- by Mr. Thompson are from Tzuli, 1904, pi. 2, and 68. Xkichmook, 1898, pp. 226-227. CONSIDERATION OF THE MATERIAL ARTS. 133 case of the great pyramid of Cholula and is seen in many modern houses. The use of adobe is common in the less humid parts of the Maya area. The wide occurrence of this simple form of wall construction points to the likelihood of its being of very early origin. The earthen filling in the walls of Copan may be explained as the surviving indication of an adobe prototype. Copan was one of the earliest Maya cities. The earthen core was a decided element of weakness which was turned into an element of strength in later Maya cities when mortar was substituted for the clay. Yet building stone was plentiful at Copan and could be cut in any required size. Lime of the finest quality was also close at hand, so that even if it were considered necessary to increase the thickness of the wall to give an adequate support to the overstepping roof stones, this need not have been done at the cost of efficient construction. At the same time there is no reason why wooden architecture should not have been developed by the people living in the forested areas where no stone was to be had. There are many mounds in the Usumacinta area from which the superstructures have entirely disappeared. 1 Minor Arts Ceramics. In the portions of the New World having the highest culture ceramics often rises to the importance of a major art, and pottery remains fre¬ quently constitute the principal results of archaeological research. In the Maya area this art, although finely developed, sinks into comparative insignificance in view of the sculptured monuments. The art of making pottery had in pre-Columbian times a practically contin¬ uous distribution from central Argentina and Chili to southeastern Canada. But the materials and technique varied widely and there were many intensive devel¬ opments in form and ornament. The art was in some respects most highly devel¬ oped in the central portion of this vast area, particularly in Peru, Central America, Mexico and the Southwest. But even in these regions there were many fairly distinct ceramic provinces. Of course it is possible, and indeed probable, that there was an infiltration of culture from one province to another. Certain structural features have a very wide distribution, such as, for instance, the use of ring base and tripod supports for round-bottomed pottery. Three legs furnish the simplest means of stability possible and are employed for a variety of objects the world over. It is but natural that the mechanical economy should be carried a step farther and the legs made hollow rather than solid. The fundamental similarity in shape and construction, if not in decoration, be¬ tween the ceramic products of Costa Rica and the Maya area suggests some sort of cultural connection. The art of Nicaragua seems to be more or less intermediate, and shows certain similarities to the Maya vessels in decoration as well as in form. The ceramic remains found throughout the Maya area include a great variety of vessels for domestic and religious uses as well as figurines, whistles, moulds, stamps, etc. Certain forms are widespread, and possibly show commercial 1 Mr. E. H. Thompson in a recent article (1912) derives the stone structures of northern Yucatan from the common thatched hut of the present Maya Indian. His explanations do not fit the facts in the earlier Maya cities of the south, however accurately they agree with conditions in the north. The im¬ portance of an historical perspective in this discus¬ sion will be more apparent after the problem of chronological sequence of style shall have been presented. 134 MAYA ART. distribution from a definite center of manufacture. On the other hand, it is clear that some types of pottery can be referred to definite periods of time, so that the variety is in part explained by chronological sequence. The pottery was shaped by hand and not by the potter’s wheel. To be sure, in some places a block turned by heel and toe was used under the vessel while it was being formed. But this object cannot be called the potter’s wheel, because the essential character of the latter comes from the development of centrifugal force. The block in question is still in use in northern Yucatan. 1 A small dish answering the same purpose is used to-day by the Pueblo Indians of the Rio Grande. An effort will be made to consider briefly the different characteristic styles of Maya pottery, es¬ pecially as regards ornamentation, illus¬ trating each with a few noteworthy ex¬ amples. No attempt will be made to dis¬ tinguish the many variations in paste. Coarse heavy pottery for household pur¬ poses has usually no decoration and de¬ serves little consid¬ eration here. It is commonly black or red in color, probably depending on whether the burning was done in a smothered fire or an open one. The shapes are various, but flat-bottomed vessels with sides that flare outward seem to predominate. The more artistic pottery falls naturally into groups according to the method of decoration: 1st. Vessels with incised decorations. 2d. Vessels with moulded or stamped decorations. 3d. Vessels with modeled relief decorations. 4th. Vessels made in the forms of animals, fruits, etc. 5th. Vessels with painted decorations. 6th. Figurines, stamps, moulds, spindle whorls, whistles, etc. 1 Mercer, 1896, pp. 161-166, describes in de- 1907, pp. 62-63 and pi. 13, fig. 3. Mr. Thompson tail modem pottery making in Yucatan; Tozzer, has also collected material upon this subject. 135 CONSIDERATION OF THE MATERIAL ARTS. Incised pottery made of a fine black or red paste is very widespread. Some¬ times the designs are geometric patterns or simplified .hieroglyphs, incised in the soft clay with a sharp instrument, and sometimes they are elaborate draw¬ ings brought into relief by cutting away the background. Fig. 184, a, illustrates the simplest sort of incised black pottery. Such pottery is found throughout the Maya area. Some of the more graceful pieces of red or black pottery that come under this heading have vertical flutings (c) made probably by the finger while the clay is soft. Fig. ISo reproduces a bowl, found near Peto in Yucatan, that is now in the Peabody Museum. The base is a pale yellow and there are traces of a red sizing. The shape of the bowl is indicated diagrammatically in the drawing. The orna¬ mentation consists of a series of painted scrolls on one side and a carving in relief on the other. The carving which is shown in the drawing represents a jaguar seated within a closed ring that is made up of what probably represents a water-lily stem coiled and knotted. The stem has two buds or flowers branch¬ ing out at opposite sides of the circlet. The jaguar wears a cape tied round the shoulders and a loin cloth or skirt, as well as wrist and ankle bands, nose plugs and a headdress consisting of the well-known head of the Long-nosed God, in front of which is a small flower similar to the flowers at the side of the circlet.' Seven oval glyphs are carved around the top of the bowl, two of these being shown m the drawing. The lines which delineate the coiled stem and the flowers are deeply incised. The jaguar figure is brought into relief through the simple device of cutting away the background. The details of the dress upon the body of the animal are incised in delicate lines and there is little or no modeling. The spots of the jaguar are represented in black paint which has now largely disappeared. The sunken background is marked with incised cross lines which still retain traces 136 MAYA ART. of heavy red pigment. The carving or engraving of this remarkable piece ap¬ pears to have been done when the clay had become fairly hard and after the sur- __ face had been polished, but be- fore burning. It was certainly / °°/ 07 x n °t m °deled in soft clay. \ o(C)j &A somewhat similar style of decoration is shown in Fig. /yl/ftl ''vP^ 186. The bowl represented "Al^Jf /w ® here * s a fi ne pi ece of pottery 13/ k 3ISr7^9 -) ® I//Q) coming from northern Yucatan J 131 /\ 2 A-x an( l now the private collec- (S\vv(^ -J ^°S )) tion of Don Enrique Camara The drawing is of Merida. copied from one made at the ex¬ pense of Mr. E. H. Thompson. In an elaborate scroll medallion appear the head and left arm of a man who holds diagonally a flexible object. The upper end of this object is a simplified face and the lower end is a flower, possibly a water lily. The composition is very pleasing to the eye. But certain features as, for instance, the headdress, have lost something of their original form, perhaps owing to constant repeti¬ tion. As in the preceding vessel, the background is here cut away so that the figure stands out in flat relief. Other examples of engraved pottery are seen in Figs. 108, b and 187. Incised decoration was sometimes effectively modi¬ fied according to the following method. The outer surface to be decorated was smoothed and covered with a fine white or black sizing. The vessel then ap¬ pears to have been burned, after which the design, which was usually limited to a band or a panel, was incised with a sharp tool and the back¬ ground cut away. The lines of the design and the open spaces of the background thus show in dull red color, while the surface of the raised figures is white or black and more or less polished. In the Peabody Museum there are several interesting pieces of this ware from the environs of Santa Cruz Quiche, Guatemala. An example has been shown in Fig. 48, and another fragment of manner that more nearly imitates the appearance Fig. 186. — Bowl from Calcetok, Yucatan. Fig. 187. — Bowl from Island of Jaina. Fig. 188. — Engraved potsherd from Santa Cruz Quiche. CONSIDERATION OF THE MATERIAL ARTS. 137 Stamped pottery is somewhat unusual, but a number of interesting forms are encountered. A peculiar flask-shaped type of vessel made of a smooth white paste and bearing on both faces a stamped portrait of the Roman-nosed God and on each of the narrow sides a double column of stamped glyphs has already received comment (Fig. 94). Nearly identical examples of this ware have been found at Coban, Copan and in the Uloa Valley. 1 A rec¬ tangular bottle of the same paste and style of decoration is figured by Seler. 2 Fig. 189. — Potsherds showing applied relief decorations. Other examples of stamped ware of dull red or yellow color are to be seen in the Peabody Museum collections from the Uloa Valley and from Santa Cruz Quiche, Guatemala. From the latter site come two interesting pieces, cylindri¬ cal in shape and decorated with stamped designs that are repeated several times around the outside. The stamped designs on these and other bowls are mostly fanciful heads limited to rectangular panels. In some cases tripod legs bear stamped patterns. In Fig. 184, d and e, are given two potsherds with realistic designs modeled in relief. It is sometimes difficult to distinguish between this kind of decoration and the finer examples of stamped ware. Of course the arrangement of figures on the outside of a bowl of this sort is unrestrained, while stamped ware shows formal designs. The rounded character of the modeled relief is in marked contrast to the method of engraved or incised relief that has already been described. Reliefs made by appliqud work are much commoner and, as a rule, cruder. The designs and figures that decorate this class of pottery are laid on or built up. Rolls of clay, ribbon¬ like strips, flat or pointed nodules, and modeled faces are the objects used in making up the decoration. These are arranged on the sides and rims of plain vessels in a variety of ways, while the clay is yet soft, and they remain firmly at¬ tached after the firing. This process, although found widely, reached its highest development in the Maya area on the highlands of Guatemala. It seems to be related in a general way to the technique so finely exemplified in the Zapotecan funeral urns. Potsherds showing this method are given in Fig. 189. A beaut.i- Fig. 190. — Jaguar head vase from Copan. 1 Gordon, 1898, a, pp. 19-20; Seler, 1902-1908, III, pp. 685-686. *1 2 1902-1908, III, p. 682. 138 MAYA ART. fully made vessel of this style has already been figured (Fig. 1,6). Many others are described by Dr. Seler. 1 Occasionally color was used upon certain details. Vessels made in various natural forms are fairly common. A remarkable vessel in the form of a jaguar head is given in Fig. 190. A tripod vase with the body modified into a bird and a human face seen in the bird’s open mouth was excavated at Copan. Gourd-shaped pots are sometimes found. A splendid example of such a pot, found at Acanceh is in the Museo Yucateco in Merida. The tall-necked bottles found by Dr. Gordon 2 in the caverns of Copan remotely resemble gourds. The bodies are sometimes fluted. Incense burners present many different forms. Some of the more elaborate ones show a sitting or standing human figure attached to one side of the bowl. Others are modeled into the form of a head or have heads joined to the rim. Incense burners will be discussed more in detail at another time, be¬ cause they furnish important evidence of the last phases of Fig. 191. — Crude painted figures on food bowls: Copan. Maya art from an historical standpoint. There are many different kinds of painted pottery. Much of the common red or yellow ware evidently intended for domestic uses has painted designs running from crudely drawn monkeys and other animals (Fig. 191) to patterns of purely geometric bands. There is also a red ware of exceeding fineness with designs in black or white. The finest pottery of all is polychrome ware. The paste is very smooth and of light weight. The background sizing is usually a highly polished yellow-orange or red. The applied colors are very rich and permanent, and include white and black and various shades of red, yellow, orange and brown. Many of the finer pieces have the general appearance of lacquer ware, so glossy is the surface. But this glossy surface is really polished rather than glazed. In fact, glazing does not seem to have been understood by any of the potters of the New World, although in two or more regions they were on the verge of the discovery. A number of pieces from the Maya area show a thin and probably accidental glaze. This is seen on several vases from Finca Pompeya, Guatemala, that are now in the American Museum of Natural History, as well as on figurines from Jonuta, on the Usumacinta River, that are now in the Peabody Museum. The former specimens have a greenish hue, while the latter are jet black. It is interesting to note that a glazed paint was used by the natives in several parts of the Pueblo area, particularly in the valley of the Rio Grande. The glaze in this case probably came from borax or other salts, which formed a flux upon moderate heating. What may have originated by accident seems to have been developed purposely. The art, however, was short-lived and is unknown to the present-day Indians. It apparently came into general use shortly before the arrival of the Spaniards and may have lasted till the eighteenth century. 3 1 1901, c, pp. 139-184. de los Frijoles, all of which are believed to be pre- 2 1898, b, pi. 1. historic. Potsherds with this kind of paint are 3 Pottery with glazed paint is found in New found on nearly all the sites near the Rio Grande Mexican ruins near Ojo Caliente and in various sites that are known to have been occupied at the coming in Pajarito Park, including Puye, Tcherigi and Rito of the Spaniards, from Taos and Picuris on the north the figures is black. The quetzal birds have light red markings on wings and head. In the glyphs there are details in several tones of red and brown. The to Abo and Tabira on the south. The ware is particularly plentiful in the Galisteo Valley and in the Valley of the Jemez River. Occurrences have been noted as far west as the Little Colorado. Castaneda, the historian of the Coronado Ex¬ pedition, refers to the art as follows: “ Some very beautiful glazed earthen-ware with many figures and different shapes. Here they also found many bowls full of a carefully selected shining metal with which they glazed the earthen-ware. This shows that mines of silver would be found in that country.” The last statement indicates that he believed the flux to be antimony. In a later passage he says: “In all these provinces they have earthen-ware glazed with antimony and jars of extraordinary labor and workmanship which are well worth seeing. ” The glazed paint is usually black but thin appli¬ cations are more or less translucent and colored by the tone of the background. In some instances the color is brownish with a slight yellowish tinge. 140 MAYA ART. vase in question is cylindrical in shape, with a flat bottom and three short knobs instead of legs. It was excavated by the Peabody Museum Expedition at Copan. Several other vases of the same shape but with complicated designs represent¬ ing priests in ceremonial regalia were found by Dr. Gordon 1 in the Uloa Valley. The most interesting designs have been figured by him in his report. Many smaller pieces that may be classed as bowls have designs such as owls, bats and serpents. A number of remarkable painted vases from the highlands of Guatemala have been described in some detail. The finest specimens were excavated in the environs of Coban. Of these the two most famous were found at Chama by Mr. E. P. Dieseldorff. 2 One of these bears two representations of a bat with outspread wings. On each wing is a crescent-shaped marking. The body of Fia. 193. — Polychrome vessel: El Jecaro, Guatemala. the bat is grotesquely human. A flame-shaped speech scroll begins in front of the face and twining upward divides into a forward and a backward part. Be¬ tween the two figures are six excellently drawn glyphs. The background is a rich orange with bands of black, yellow and brown at top and bottom. The delineation is in black, and the interior areas are filled in with white, red and brown. The second vase represents seven human beings drawn in profile and with clear details of dress, upon a light yellow background. The flesh of five of the figures is yellow orange in color but in the case of the remaining two is painted black. The most interesting figure is striding forward with a spear in one hand and a fan in the other. The foreshortening of some of the bodies is rather poor, and the bodies themselves are gross. In connection with each figure are several glyphs that probably give the name of the individual or other infor¬ mation concerning him. Other vessels of the same general character might be mentioned. 3 But as a rule the variations from black are towards green suggesting the use of borax. In fact, in some cases, the glaze is quite green while in others there are green spots or bubbles. Sometimes there are thin washes of green stain extending beyond the paint. The vessels were turned upside down in firing and in many instances the paint ran badly. Since it is an invariable rule to let the clay and the paint dry thoroughly before firing, it seems likely that the paint fused during the firing. In some cases it seems to have boiled or bubbled consider¬ ably. Blisters are seen on many specimens. Glazed paint does not seem to have been limited to any one kind of ware, but to have been used on all wares that were burned in an open draft kiln (that is, all wares that are not black). It is com¬ monest on a thin hard red ware which shows the use of red and orange sizing. It is also found on terra cotta ware and on ware with a cream coloring sizing. 1 1898, o, pis. 4 and 5. 2 See collection of papers on this subject trans¬ lated in Bull. 28, Bur. Am. Ethnol., pp. 639-666. For the original color reproductions see Dieseldorff, 1894, a and b, pis. 8 and 13. 3 See Dieseldorff, 1893, b, pi. 14 (Chama); Seler, 1902-1908, III, pp. 629, 633 (Rio Hondo), 718 (Nebaj). CONSIDERATION OF THE MATERIAL ARTS. 141 A small vessel of this type is given in Fig. 193. The persons represented wear flowers in their hair, as do those on the vases just described. They are seated and apparently engaged in conversation rather than in ceremonies. An inter¬ esting feature on this vase is the representation of an altar which is apparently a bundle tied by broad bands. The altar recalls the wrapped and knotted stone altars of Copan, as has already been pointed out. Upon the top of the altar shown on the vase is a human head, indicating sacrifice. The glyphs are ap¬ parently merely decorative, for all are approximately the same. The colors are orange for background, ^ bbbto co coo zn n: m Hr- 1 i— J ~ la la la 13 TETL5TS - black for lines, and red, white and brown for masses. This vase was excavated at El Jicaro, near Zacapa, Guate¬ mala, during the construc¬ tion of the railroad, and was presented to the Peabody Museum by Dr. Lytle. In the private collection of Don Enrique Camara of Merida, Yucatan, there is a remarkable vase of the same class. This specimen was found at Calcetok, between Merida and Campeche. The paste is very light in both color and weight. The back¬ ground is a yellow orange sizing with a high polish. Upon this the figures were delineated in black, and cer¬ tain details filled in in white, red and brown. The shape is cylindrical. The pictures on this vase include three standing and two seated men, two deer, a jaguar, a tree, a serpent and a bird. A description of the remarkable collection of pottery with painted and plastic decoration from eastern Peten, collected by Mr. Merwin for the Peabody Museum, will add much to our knowledge of the ceramic masterpieces of the Maya. Many vessels that it is impossible to treat in detail show combinations of the different methods of decoration that have been described. A medallion decoration showing the combination of incised and painted design is given in Fig. 184, b. The geometric units used in ceramic decoration embrace nearly all the more common ones. The fret and the spiral are finely developed, as may be seen from Fig. 194. As a rule, the geometric elements are applied in bands around the neck or rim of the vessel. Small clay figurines occur very widely in Mexico and Central America, and in the Maya area especially are much diversified. Plate 17 shows human ^ Fig. 194. — Geometric motives used in pottery decorations. I 142 MAYA ART. figurines of markedly different types, which give an excellent idea of the range in form and finish. Many interesting figurines come from the Uloa Valley on the southern frontier (1 to 3). The figurines from Jonuta on the lower Usumacinta (4 and 7) are among the finest from the Maya area. The last four specimens come from the Island of Jaina near Campeche and belong to the private collection of Mrs. W. M. James of Merida, Yucatan, through whose kindness they are here reproduced. The two smaller pieces, 9 and 12, exemplify very common forms, but the two larger ones are exceptional in several ways. One of them is really a tableau and presents three small figures grouped about the legs of a much larger figure. The other is a very unusual piece of modeling. It represents a man, naked except for a loin cloth and breast ornament. The right hand is raised and the left one hangs at the side. The right knee is also raised. The feet are missing. It is possible that this figure is only part of a group. A cast from an ancient terra cotta mould is reproduced in 8. This mould was obtained in the region of the Rio Chixoy in northern Guatemala. The type of face re¬ calls the sculptures of Yaxchilan and Palenque. 1 Many figurines in human and animal form were used as whistles with three or four notes. Others may have been used as household gods. Pottery stamps with geometric and conventionalized designs and many small objects of clay with ornamentations of various sorts are found in all parts of the Maya area. Precious Stones. Jadeite and other semi-precious stones were much used for beads, ear plugs, nose plugs, amulets and other small carved objects whose use is unknown. Dr. G. F. Kunz, 2 an accepted authority on precious stones, writes as follows concerning the green stones of Mexico: “Chalchihuitl, a name celebrated in Mexican archaeology, was applied to certain green stones capable of high polish, which were carved in various ornamental forms and very highly valued. There has been much mystery and much discussion as to what this precious material really was, and whence it was obtained. It seems evident that several minerals were included under the name, among them a green quartz or prase, some of the deeper green varieties of techli or Mexican onyx (so called), and probably turquoise; but the precious chalchihuitl has now been proved to be jadeite, a stone which has possessed a singular charm for many aboriginal peo¬ ples in widely separated parts of the globe . . . ” Nephrite apparently does not occur in Mexico and Central America. Mrs. Nuttall 3 has been able to show, by the etymology of place names and by the tribute demanded from conquered cities by the Aztecs, that chalchihuitl was a product of definite regions in southern Mexico. Doubtless jadeite occurs in Guatemala and perhaps farther south. Hartmann describes jadeite objects from Costa Rica, which seem for the most part to be of local manufacture, and characteristic of the individualized art of Nicoya. A few specimens were probably acquired from the Maya region to the north. One piece in particular is almost surely Maya. 4 There is considerable similarity between the small human and animal figures crudely carved of green stone and possibly intended to be used as fetiches, that are common in western Guatemala 5 and in Costa 1 For other examples of figurines see Blackiston, crude jades from Oaxaca somewhat similar to many 1910, a and 6; Seler, 1895, d; Batres, 1888, pis. 1-3. from Guatemala and Costa Rica. 2 1907, pp. 20-21. 5 The localities represented in the American 3 1901, b. Museum of Natural History by these crude carvings 4 Hartmann, 1907, pi. 45, fig. 10. Plate 46 shows are Zacualpa, Joyabaj and Sajcabaja. CONSIDERATION OF THE MATERIAL ARTS. 143 Rica. This likeness may be due to similarity in technique which is characterized by the use of straight grooves to outline crudely either faces or entire figures. Of course this method is the simplest possible one. In Costa Rica, Guatemala and the State of Guerrero, Mexico, the shape which serves as the basis for most of these figures is that of the celt. Indeed, some of the ornaments must be re¬ garded simply as decorated celts. In other cases celts have been halved or quar¬ tered to get material for ornaments. The explanation of this may be that stones of the finest color and texture were hard to obtain, and that the celts which turned out after polishing to be of the desired material were simply transformed into figures with the least possible labor. It must be pointed out that the region in Guatemala from which these stones come is upon the western frontier, where the lapidary’s art was not so highly developed as in the great cities of the lowlands. In the working of jadeite and other stones 1 drills and cords were used for boring and cutting. The larger stones were sawn into flat slabs. Sometimes, as has been said, celts were neatly halved or quartered by sand and water grinding in grooves probably through the agency of a cord drawn back and Fra. 195 .-Carved jadeite amulet with wiption. forth. Frequently irregular but some¬ what flat pieces of stone were smoothed off or carved upon one side, while the other side was left in the original rough state. Indeed nearly all the carved jadeite objects from Mexico and Central America show portions of the original weathered surface, which is a good indication that the material was found only in small pieces. The cruder specimens have straight incised lines to mark out the details of the figures. Many of the finer objects, however, are freely carved m the graceful curves characteristic of Maya art, often with marked relief and with modeled surfaces. Perhaps the most famous piece of worked jadeite from the Maya area is the Leiden Plate." This is a thin oblong slab with rounded corners, having upon one side an incised drawing of an elaborately attired human being holding a Ceremonial Bar in his arms. Upon the other side is a column of hieroglyphics. Further discussion of this specimen is reserved for a later section: Another important though less artistic piece is the San Andres Tuxtla Statuette, 3 which will also be treated later. A well-known jadeite amulet belonging to the famous Bishop collection 3 of jades, now installed at the Metropolitan Museum of Fine Arts, New York, is reproduced in Fig. 195. Upon the front is represented a human face in front view, with a headdress consisting of the upper portion of an animal head. The relief is rather high. Upon the back, which shows the natural weathered surface, are a number of glyphs. Dr. Forstemann attempted a tentative interpretation of the inscription as follows: “From the day 4 Ahau, the 7th of the month Zip, 1 Sahagun, 1SS0, pp. 585-587; Seler, 1890, pp. 418-425; Holmes, 1895-1897, pp. 304-309; Saville, 1900, b, pp. 106-107. Valentini, 1881, b. 3 Holmes, 1907. * Bishop Collection, No. 309, p. 100. 2 Leemans, 1877; Holden, 1879-1880, pp. 229-230- lftisl ml 1 OOI t 7 144 MAYA ART. Fig. 196. — Jadeite amulet with red paint on front and inscription on back: Island of Jaina. when the god K ruled, lived (the person whose face is shown on the other side) until the year which begins with 8 Kan on the 13 of the month Zee.” According to this the deceased person lived fifty years, two months and six days. Although it is evident enough that jades were often buried with the dead or used as votive offerings in the temples, it is unsafe to consider them as examples of portraiture. A jadeite amulet comparable to this in having an inscription (Fig. 196) on the back was found by Maler on the Island of Jaina near the city of Campeche and is now in the American Museum of Natural History. The grooves on the front show traces of red paint and there are eight incised glyphs on the back. A number of finely carved ornaments, mostly obtained at Ocosingo, Guatemala, by E. G. Squier, have been described 1 by that early student of Central American archaeol¬ ogy. The larger pieces are pierced for sus¬ pension as amulets. A flat slab about four inches long and half as wide shows a human figure seated cross-legged, with the body in front view and the face in profile. Beneath the figure is a conventionalized face. The drawing is rather poor and the relief is flat. More rounded relief is shown in another specimen, roughly triangular in shape, which bears upon the front a human face in front view, with a simple headdress probably representing a convention¬ alized animal head. The ear plugs at the side of the human face have flaring featherlike appendages, that fill the two upper angles of the triangle. One of the most interesting pieces is reproduced in Fig. 95. This is a thin plate of translucent jadeite representing in stencil-like profile view a typical grotesque face of a Sun God. The carving is the same on both sides of the plate, and the four divisions of the “kin sign” on the forehead and the curve that indicates the pupil of the eye are cut through from one side to the other. Fig. 197 gives three diagrammatic views of a typical jade to show the perforations. One perforation traverses the entire length of the jade and another crosses from side to side. Six dowel holes enter from the side and issue at the back. These were probably used for the attachment of feathers. In Fig. 198 we have the design on the front of the specimen. In the Squier collection is a globular head with three inscribed glyphs and several kinds of ear plugs, such as are commonly represented in the sculptures. Fig. 199 shows a well-carved square ear plug and Fig. 200 reproduces a clay figurine of a woman found near Coban with ear plugs identical in shape and design. The excavations at Copan by the Peabody Museum revealed a number 1 1870. The collection is now in the American Museum of Natural History. Fig. 197. — Perfora¬ tions of a jadeite amulet from Chi- chen Itza. Fig. 198. — Jadeite amulet from Chi- chen Itza. CONSIDERATION OF THE MATERIAL ARTS. 145 of burials in which were found jadeite beads and other ornaments of stone and shell. The beads were of two kinds, globular and cylindrical. According to the necklaces represented on the stone sculptures these two kinds of beads were frequently strung alternately. Vessels of alabaster and marble are occasionally found. A fine cylindrical jar of marble is figured by Hamy 1 and described as coming from Honduras, on the southern frontier of the Maya area. Two small bowls of very similar appearance were obtained by Dr. Gordon 2 on the Uloa River in northern Hon- ^piug- duras and are now in the Peabody Museum. A small and a large bowl of the same type, the latter with a perforated ring base, belong to the Squier collection in the American Museum of Natural History. All these stone jars have a very characteristic decoration, consisting of short scrolls carved in low but rounded relief, and knob handles carved into animal heads. The large one in the Squier collection has a face on each side (Fig. 201) enclosed in spirals of the same character as those on the smaller pieces. A fine alabaster vase cylindrical in shape and with a simple but pleasing geometric decoration was found by Thompson Fig. 200. — Clay figurine in the so-called High Priest’s Grave at Chichen Itza ;taT 8 i : nE c^b e an 0,iadeitee0r This is now in the Field Museum at Chicago. Metal Working. A large part of the Maya area is very young geologically and does not con¬ tain ores of any sort. Upon the southern and western frontiers metals are found, in addition to which small quantities may have been obtained in trade. Owing to this lack of material, within the region of the highest culture, objects of metal are rare. Such as do occur are fully equal to the metal work of the Valley of Mexico or the more distant Costa Rica. Gold and copper and sometimes silver were worked by hollow casting and by hammering. The making of wire filigree is found from Colombia to Central Mexico, 3 and the distribution of this interesting technical process goes farther to show cultural con¬ tact than any number of fanciful resem¬ blances in decorative art. The objects to be cast were modeled in wax, pitch, or some such substance. If hollow, an inner form, probably of clay, was used. Decoration was added by rolling out a wire of the wax or pitch and applying 1 1896, pp. 16-11, and pi. 2. 3 References on metal working are McCurdy 1 190S, o, pp. 26-26 and pi. 12, e and /. Dr. 1911, pp. 189-226 (Chiriqui); Penafiel, 1890, pis! Gordon comments on the distribution of this type 109-113 (Nahua and Zapotecan); Seler, 1890,' pp. of object, which seems pretty definitely limited to 401-418 (Nahua); Lumholtz, 1902, II, pp '296 Honduras. 413-416 (Tarascan, etc.); Valentini,’ 1879, o. ' this to the surface in whatever design was desired. The whole was then inset in a clay mould and the wax or pitch melted out. In the metal object as cast the wire decoration has the appearance of having been attached by some metallic cement. Sometimes the entire object was built up by coiling and folding over the wire. Fi 'with’fnpiSdMora" for the sma11 percentage of gold in the metal. Other caches Chichen Itza. A few gold bells have been found in the Maya territory. Small ornaments of gold and copper are fairly common in Chiapas and Tabasco. Bernal Dias speaks of the scarcity of gold among f fc ~ ~ - —j narrative called the “Let- ing descriptions of objects made of precious stone (probably jadeite) and metals are given. While the general enthusiasm of these discoverers must be discounted, there is no reason to distrust their specific references. There is doubt, however, whether these objects were seen in Yucatan, as stated, or in southern Vera Cruz. In any case the list is instructive. “Two round disks, one of fine gold, the other of fine silver, handsomely worked with beautiful figures drawn with a free hand. . . . The former measures seven spans and the silver one is smaller by about a little finger. “Further, a head of a great serpent or dragon: a figure of very fine gold with golden teeth that are easily a span wide and three fingers thick; gather for yourself how large the head is; the eyes are of precious stones and are adorned with very costly feathers. “Further, a great disk of precious stone completely covered with tiger hide, the hide being highly valued. “Further, four necklaces with many precious stones set in gold. “Further, a horn of a seafish, of gold, two spans long and about two palms wide, all of gold. “Further, a head of gold and many other pieces of gold, silver and precious stone.” Basketry. A basket of simple twilled weaving is held in the arms of a kneel¬ ing supplicant shown on one of the lintels of Temple 21 at Yaxchilan. 4 A similar 1 1688, p. 181. 3 Muller, 1871, pp. 26-30. See also Valentini, 1879, a, p. 97. 2 1910, b. * Maudslay, 1889-1902, II, pi. 83. CONSIDERATION OF THE MATERIAL ARTS. 147 but more elaborate basket is shown in Fig. 203, a, which is also represented on a lintel at Yaxchilan. The upper portion of this basket is executed in twilled weaving; the middle portion shows a finer weave, with designs of stepped frets and small rectangles in groups of three. The bottom of the basket seems to be ornamented with feather-work. The shape of these two baskets is somewhat unusual on account of the straight sides and flaring rim. Fig. 203, b, apparently represents a basket with colored decoration consisting of ribbons and other adven¬ titious ornament. The drawing is taken from the sculptured wall of the Lower chamber of the Temple of the Jaguars, at Chichen Itza. Basket-work fans are shown in the hands of the figures on the Chama 1 vase. Braided bands are sometimes represented in headdresses, as in the case of Stela A at Copan and Stela 10 at Seibal. It seems probable that some sort of Fig. 204. — Imitation of basket weaves on painted pottery: Uloa Valley. basket foundation may have been used to support the elaborate feather-work. A type of ceremonial staff found at Yaxchilan on Lintels 6 and 43 consists of a woven object similar to an inverted waste-paper basket which is carried on a pole. The Long-nosed God with the serpent appendage is seated on top. Simple basket weavings appear as painted ornamentation on potsherds from the Uloa Valley (Fig. 204). Complicated braided patterns are common as the rim decoration on pottery from this region, and may have had their origin in the imitation of wicker-work basketry. It is probable that basketry was not of much importance as an art among the Maya, owing to the high development of ceramics. Textiles. The textile art of the ancient Maya must be studied mostly at second hand from designs sculptured or depicted on the garments of figures rep¬ resented on stelae and lintels or in mural decorations and codices. There is a strong probability of certain survivals in the modern art of the Indians of Yu¬ catan and Guatemala, but little information is available. The native women of northern Yucatan still embroider their dresses with floral and sometimes geo¬ metric patterns which may be pre-Spanish. In 1765 Lieutenant Cook, after¬ wards the famous captain who explored the Pacific, made an overland trip from Bacalar to Merida. His report contains an excellent description of the ap¬ pearance of the natives and the country. He comments on the comeliness of the women, who, he said, wore white cotton smocks embroidered with flowers in needlework at the bottom. 5 The present mode is then at least one hundred 1 Dieseldorff, 1894, a, pi. 8; Bulletin 28, pi. 48. - Cook, 1769, pp. 29-30. 148 MAYA ART. and fifty years old. The Quiche and other Indians living on the highlands of Guatemala 1 still make their own textiles, but European influence seems to be pretty strong. For instance, silk is widely used for embroidery. The designs are mostly stripes of simple geometric figures, although some pieces show birds and other life forms finely conventionalized. Careful field study among these peo¬ ples might result in the determination of many of the ancient designs and their meanings. The early explorers and historians 2 comment on the beautiful garments worn by the natives of Yucatan. Cogolludo 3 says the cotton cloth of Yucatan made in various colors was traded over all of New Spain. Aguilar 4 likewise com¬ ments on the extent of the textile industry. Much of the tribute 5 demanded by the Spaniards was in cloth. The every-day dress of the men was a sort of breech cloth that passed around the hips and had end flaps hanging down in front and behind. In the ancient sculptures these apron-like flaps are often embellished. The apron with a gro¬ tesque face between two serpent heads conventionalized in the form of frets (Fig. 15) may have been purely a ceremonial elaboration possible in sculptures but not used in real life. It has, however, a remarkably wide distribution among the southern cities of the Maya area. Often aprons have a sort of openwork 1 Stoll, 1889, pp. 96—101; Maudslay, A. C. and 4 1639, p. 94. A. P., 1899, pp. 41-43. 6 References to tribute are numerous; see Rela- 2 Landa, i864, pp. 117 and 182-184; Muller, 1871, ciones de Yucatan, 1900, II, pp. 57, 67-68, 150; p. 28; Relacioncs de Yucatan, II, pp. 29, 46-47, 104- Relacion de los Conquistadores, 1870, pp. 193-195, 105, 123, 154, 211-212, etc. For a modern discussion etc. Besides mantas, the tribute matter included see Schellhas, 1890, pp. 214-228. wax, cocoa, and, in ancient times, green stones, red 3 1688, p. 173. shell beads, etc. CONSIDERATION OF THE MATERIAL ARTS. 149 design in the center and a fringe at the bottom, as may be seen in Fig. 205, b. Aprons of greater width and more elaborate decoration occur, such as the one shown in a, drawn from one of the statues on the Hieroglyphic Stairway at Copan. A most beautiful apron is found on Stela 7 at Piedras Negras. 1 The design consists of a symmetrical arrangement of small frets around a Greek cross. a 6 c d Fig. 206. — Garments represented on sculptures and in codices: a and b, Chichen Itza; c-e, Dresden Codex. A garment in the form of a short skirt reaching half-way down the thigh is sometimes seen upon figures which evidently represent men. Often this skirt has the characteristic markings of a jaguar skin, as may be seen on Stelae P, 2, etc., at Copan. The belt at the top of this short skirt receives the greater part of the decoration, which is usually of a geometric style. Fig. 206, a, reproduces the skirt of one of the processional warriors from the Lower chamber of the Temple of the Jaguars at Chichen Itza. This short skirt is marked with crossbones, which may repre¬ sent painted rather than woven decoration. The most elaborate textile patterns are found on a sort of blanket which usually envelops the entire body, although in some cases it seems to have been bound around the waist so that the corners hang down on either side. Fig. 207, taken from one of the lintels of Yaxchilan, represents a kneeling supplicant. The entire body of this person is enveloped in a robe having an all-over geo¬ metric decoration in squares, a rich border at the bottom and a tasseled fringe along the edge. It is possible that this dress rep¬ resents the sacklike garment still worn by the Maya women. Similarly gowned figures occur on a number of the lintels of Yax¬ chilan and upon several stelae from Piedras Negras. As a rule, these figures are of smaller size than those on the same lintels which have the usual masculine dress. The geometric designs on these robes are applied in horizontal stripes, in diagonal stripes and in all-over patterns. Very often the border of the bottom is differentiated from the design on the blanket as a whole. Fig. 205, / and g, 1 Maler, 1901, pi. 16. Fig. 207. — Woman richly attired: Yaxchilan. 150 MAYA ART. shows two such border decorations, and h a fringe border marked with the com¬ mon planet symbols. Fig. 205, e, reproduces the design on a garment which hangs over the shoulder of a male figure. Robes somewhat like these which have just been described are represented in the codices upon figures which are clearly masculine (Fig. 206, c to e). Fig. 208. — Facial tattooing. What might be called an inset lace medallion often occurs as an evenly dis¬ tributed design in the body of a garment. The details of two of these inset medallions are given in Fig. 205, c and d. Tattooing. Dr. Schellhas 1 has already commented on the evidences of tattooing among the peoples of Central America. These evidences are found upon figurines and in the codices as well as on the sculptures. In Fig. 208, a to / are taken from the examples that Dr. Schellhas gives. Of these, the first (a) is a sketch of the head of a Yucatan figurine representing a man with the cheeks and chin tattooed with a drawing of a jaw bone. There is also a simple design in the center of the forehead. A simpler example of the jaw-bone design is seen in b. More often, however, the tattooing consists of simple markings made either around or at the side of the mouth, such as are shown in c, d and e. Fig. 209 presents a terra cotta head with raised designs on each cheek, and on the lips as well, that doubtless repre¬ sent tattooing. 2 The most elaborate tattooing seems to have consisted of spiral scrolls at each corner of the mouth; examples of this type of facial decoration are Fig. 209. — Pottery whistle seen on the lintels of Yaxchilan (Fig. 208, 6). The splen- showmg tattooing: Te- ' , _ coipa, Tabasco. did head formerly at Uxmal, and now in Mexico City (Fig. 182), has similar markings upon one side of the face. There are no certain objective evidences of tattooing upon the body. The markings which occur on the legs of many of the grotesque figures of gods may simply indicate reptilian affinities. The human beings represented in the fresco paintings of Santa Rita, 3 British Honduras, usually have all bare portions of the body covered with minute geometric markings. Circles surrounded with dots, short hooks and short parallel lines are the prevailing motives. It is doubtful whether these markings represent real body ornamentation or merely adventitious elaboration by the artist. Minor Carvings. A remarkable drawing made on the skull of a peccary is reproduced in Fig. 210. The engraved skull was found in a tomb at Copan, 1 1890, pp. 212-213; Landa, 1864, p. 120. 2 Other examples of tattooing on figurines might be given. See, for instance, Batres, 1888, pis. 1-3. * Gann, 189S-1S99, pis. 29-31. along with a number of pottery and bone objects, during the excavations con¬ ducted by the Peabody Museum. The drawing is here shown as if spread out flat and no account taken of the uneven surfaces. In the upper left-hand corner are shown three running peccaries, drawn with a considerable degree of natural¬ ness. The hair is represented by short lines not very close together, which give an excellent idea of the general grizzled appearance of the animals. In the cor¬ responding right-hand corner are represented a jaguar and a monkey in profile, back to back. The jaguar is well drawn without any marked divergence from the natural form. The monkey wears a breech cloth, a necklace, ear plugs and a simple headdress. In his right hand he holds a rattle, apparently made from a gourd, marked with three small crosses. The ex¬ pression on the face of the monkey is very realistic. In the center of the skull within a quadrifoil medallion are two men seated, facing each other, in easy conversational attitudes, one with the face in profile and the body in front view, and the other with both face and body in profile. Both these men wear rather elaborate headdresses. Their breasts are bare and their belts have for decoration a simple face. Between the men and arranged one above the other appear at the bottom a grotesque head, at the top a group of four glyphs and in the middle an oblong object with interlacing bands which may represent an altar. In the lower left-hand corner of the skull is drawn a deer wearing a loin cloth and standing in erect human attitude. In front of this deer is a stooping figure with a death’s head. He carries on his back, by means of a tump line, a bundle tied by a number of broad bands. Over this bundle hovers a bird with wings outstretched. Besides the drawings of men and animals which have here been briefly enumerated, there are a number of glyphs scattered about. These glyphs may be name glyphs referring to the nearest figure. This drawing is remarkable for its freedom, its naturalness and its lack of conventionality. A fragment of a carved animal skull picked up on the river front at Copan is reproduced in Fig. 60. It has already been commented upon at some length. Shells carved with glyphs have been found at a number of sites. Some inter¬ esting examples come from the Island of Jaina. A large shell with a fine inscrip¬ tion has been figured by Thomas, 1 with the statement that it was secured in British Honduras. A beautiful carved shell of Maya manufacture was found during the excavations of Tula. A seated human figure is upon one side and > 1894-1895, pi. 69. - Peccary skull with incised drawings: Copan. 152 MAYA ART. four Maya glyphs upon the other. This specimen has been reproduced by Penafiel. 1 It is now in the Field Museum at Chicago. Illuminated Manuscripts. The three Maya codices 2 are of unequal value as objects of art. The Dresden is by far the best. The drawings of some parts are of exceeding delicacy and smoothness, while other pages show much rougher work. This difference in finish may be explained by difference in paper or in authorship. The codex treats of a variety of subjects, and it is possible that each subject was written out by a different hand. The glyphs of this codex are rich in detail and similar in most features to the glyphs of the sculptures. The cursive style of representation has caused some changes. In the first place, the glyphs are tilted slightly toward the right. Ex¬ amination makes clear that the stroke for outlining faces and other oval or circular bodies was begun at the lower left-hand corner and carried over and around to the right, ending a slight reverse curve that produces a sort of beak. Details were doubtless put in after the outline had been made. The columns of glyphs were marked out in faint lines to guide the scribe, and the numbers to be set down were marked on lightly. Certain erasures are in evidence. Mis¬ takes of more serious nature were made in some of the calculations, according to Forstemann. It seems possible that there were professional scribes or copiers who did not understand very well what they were transcribing. The oblong pages of the Dresden Codex are usually divided horizontally into three zones, and each of these zones subdivided vertically into three sections. Frequently a drawing of one or more figures is shown in each of the nine sections, associated with glyphs and numbers. The glyphs often show an intelligible abbreviation of the main figure, and by this means the glyphs of nearly all the gods have been determined. In these uniform spaces all the men and women are of a certain height, no matter whether standing or sitting. The figures are drawn in lively attitudes. Among the poses shown may be mentioned sitting in side view so that of the legs only the bottom of one foot and the foreshortened thigh of the other leg are in view, sitting with legs in front view but with the head turned in profile; sitting with the knees up, squatting and stooping, walking, reclining, and even falling headlong with the body curiously twisted. Objects held in the hand or placed before the figure indicate the attributes and powers. While most of the more finished drawings are in simple black and white, many others show the use of color for the background, and a few resemble pic¬ tures in that several colors are combined to mark out the details of dress and ornament. The probable connection of some of the gods shown in this and other codices, with those represented by other means, has already been discussed. From the prevalence of glyphs of the southern type, and especially the occurrence of the period glyphs, this codex has been referred to the southern or western part of the Maya area. Forstemann 3 believes that the codex comes from the region of Palenque. It is significant that neither the Ceremonial Bar nor the Manikin Scepter, the two principal ceremonial objects in the Maya cities of the great period, is seen in this codex. The dates given in the Dresden Codex will be considered in another 1 1899, pi. 80, and 1890, pi. 169. See also Charnay, 1885, p. 74. 3 A revised pagination for these documents is given on page 260. 3 1897, p. 48. CONSIDERATION OF THE MATERIAL ARTS. 153 section. If the last of these dates is historical, the Dresden Codex comes down from the time of the Transition, and may have originated in the region south of Uxmal or perhaps in Tabasco. There are a few remarkable similarities be¬ tween this codex and others that are ascribed to the Zapotecans. An instance in point is the staff with a hand 1 at the upper end that occurs likewise in the Codex Borgia (Fig. 211). The Peresianus Codex, now, unfortunately, in a very fragmentary condition, 2 seems also to be an old manuscript that may antedate the coming of the Spaniards by several centuries. The drawing is very fine, and there is reason to believe that in its prime it was equal to the Dresden Codex. The details of the codex show certain rather definite similarities to the sculptures of Naranjo, Quirigua and Piedras Negras, and it seems likely that the manuscript originated in the general region of these cities or was copied from an earlier manuscript having such an origin. A croco¬ dile-like animal with its head hanging over the edge of a throne and with its feet bound to the body is seen on Plates 3-11 of the de Rosny reproduction. An apparently human figure is seated on this throne, and a god, that varies from page to page, stands be¬ fore the throne holding a head in his hands, which usually appears to be that of God K, while before his face flutters a bird and before his feet is an offering of maize. The idea of a figure seated on a throne is common on the stelae of the Peten region. Here also the figure is usually human in appearance. One or more other figures, also in human guise, are sometimes F, °- 21 J- — '- ta£f ™ d “g in hand: .. „ , , , , a, Dresden Codex; b, Codex arranged before this person, and these frequently hold Borgia. the Manikin Scepter or some other ceremonial object in their hands. In other cases the seated figure himself holds the Manikin Scepter or the Ceremonial Bar. The throne or niche in almost all instances has a bound animal placed across it. Sometimes the animal assumes the characteristics of the Two-headed Dragon. Sometimes two animals, one above the other, are rep¬ resented, the upper one being highly conventionalized. Illustrations of the thrones in the Peresianus Codex and on Stela 32 of Naranjo have been already presented in Fig. 6, d, f and g. For comparative study reference is made to Stelae 5, G, 11. 14, 25 and 33 at Piedras Negras, and Stelae 22 and 32 at Naranjo. The technique of the sculptures is very different from that of the manuscripts, and this fact should be taken into account. The form of the astronomical band on Plate 22 of the Codex is almost identical with the astronomical band on the back of Stela H at Quirigua (Fig. 6, a and c). The face of the Roman-nosed God attached to the body of a bird appears on Plates 4 and 8 of the manuscript and on the top of Stela 5 at Piedras Negras. Although the resemblances which have just been pointed out are not conclusive, yet they are more so than those shown by either of the other codices for any particular region. The Tro-Cortesianus Codex is much inferior to the other two in artistic skill, and may be a late work or at least a late copy. The difference in style is 1 This object also is seen in the Tro-Cortesianus Codex, p. 89. 2 For the romantic history of this manuscript see L 6011 de Rosny, 1876, p. 6 . 154 MAYA ART. well shown by the drawings given in Figs. 77, 78, etc. The most striking simi¬ larity of the Tro-Cortesianus to the sculptures is in the curious representation of the two birds with interlocked necks (Fig. 102, i), which resembles in subject a rather crude carving at Labna. The glyphs are cruder and simpler than those of the other two codices, and are inclined to be angular. The glyphs of the Books of Chilan Balam, which belong to the Spanish epoch, are very much more an¬ gular. The Tro-Cortesianus Codex may with some assurance be assigned to northern Yucatan and to a date not much later than 1200 a.d It contains no representations of the sun disk and other Nahua and Zapotecan features which appear in the late sculptures of Chichen Itza. The forms of the gods are similar in detail to those of the other two codices. All three of the Maya codices are quite clearly marked off from the ancient books of the neighboring peoples, although random resemblances occur, as has already been noted. A word or two may be given to the subject matter embodied in these codices and the attempts at decipherment. Since the days of Ivingsborough 1 attention has been directed to this field of study. Most of the early attempts at, elucida¬ tion are practically worthless and even in modern times much has been written that is of little value. Brasseur de Bourbourg, 2 for instance, began his transla¬ tion of the Codex Troano at the wrong end of the manuscript, and used the so- called alphabet of Landa, 3 which is now known to be no alphabet at all. Charen- cey and Leon de Rosny worked in a careful and painstaking manner and their results deserve credit as pioneer efforts. The most important contribution of Thomas, in his Study of the Manuscript Troano, was his demonstration that pages 34 to 37 of the Tro-Cortesianus Codex referred to the ceremonies of the new year as described by Landa, and his comparison with these pages of an analo¬ gous passage in the Dresden Codex (pages 25-28). Undoubtedly the greatest single contribution to the subject in hand is Forstemann’s Commentary on the Dresden Codex. Many valuable papers on definite subjects relating to the codices have been contributed by Bowditch, Schellhas, Seler, Fewkes, Gates, etc. Many of these have already been referred to in this text. The contents of the three manuscripts are largely religious and astronomical, although the Tro-Cortesianus Codex also casts considerable light on the every¬ day life of the Maya. The Tonalamatl 1 or 260-day period is frequently indicated, usually in a much more abbreviated form than in Nahua and Zapotecan codices. This time period, with its varying divisions, is employed in connection with gods, ceremonies, avocations and events. The representation of the new year ceremonies in the Tro-Cortesianus and Dresden Codices has already been men¬ tioned. Possibly the same subject is contained in Plates 19 and 20 of the Codex Peresianus. Several pages of the Dresden Codex are devoted to intricate astro¬ nomical calculations in which the lunar, solar and Venus calendars are correlated in a wonderful manner. It is possible that some historical references are con¬ tained in the Dresden and Peresianus Codices, although this is not very likely. In the former there are a number of dates that can be expressed in the same system that was used on the ancient monuments. It is probable, however, that the true historical records were among those destroyed by Landa and other Spanish priests. 1 The Dresden Codex was first reproduced by 3 1S64, pp. 316-322. Kingsborough, 1831-1848, III. 4 Bowditch, 1910, pp. 266-274; Forstemann, 1 1869-1870. 1895. III. CHRONOLOGICAL SEQUENCE. First Epoch. Statement of the Problem. In the analysis of any great national art the determination of the chronological sequence of forms is of first importance. How difficult a problem this may become, even in the full light of a civil¬ ization continuous to our own times, is seen in the years of labor that were necessary to arrange in such an order the remains of Greek art. Although the pre-Columbian peoples of Central America had reached what may properly be called the historic stage of civilization, yet their history is un¬ known to us because we cannot decipher the inscriptions on their monuments. Only in so far as these inscriptions deal with the absolute relationships of numbers have they been satisfactorily explained. It has long been thought that the many glyphs which contain no numbers were used to carry the historical narrative, to give the names and attributes of chiefs and deities, or to make clear the exact nature of the ceremonies connected with each particular monument or temple. The glyphs which do contain numbers are found in series that express one or more dates in the wonderful system of the Maya calendar. But unfortunately these dates were measured in cycles from an imaginary beginning of time in the distant past. Mr. Bowditch, 1 speaking of this beginning date, says: : 1 “ If it were possible to connect with certainty the date 4 Ahau 8 Cumhu, from which all these other dates are counted, with our own chronology, we could easily reach a clear knowledge of the dates on which these monuments were erected and these inscriptions were carved, always provided, however, that the dates so given are records of the dates of the erection of the monuments, or at least of the buildings in which the inscriptions are found; and this, I think, is now generally conceded to be the case in almost all instances.” It must be admitted, however, that the attempts which have been made, upon the sole basis of the recorded dates to reconstruct the ancient history of the Maya have not proved over successful. We will now consider the evidence indicating the historical sequence of monuments and cities which a careful study of the art is able to furnish. Such a sequence, even after it has been determined, cannot be put on a basis of actual years until the historical character of at least some of the dates in the inscriptions has been established. The easiest method of presentation is to take up one city after another, beginning with those that show the most archaic forms. Copan. In western Honduras is situated the ancient city of Copan. 2 The principal monuments of this pre-Columbian capital were first made known to the world through the descriptions of Stephens and the drawings of Catherwood. More recently the splendid plates that illustrate Maudslay’s work and the reports 1 1903, b, p. 3. 1 See Table of Nomenclature, p. 251. I 156 MAYA ART. of the detailed explorations of the Peabody Museum, unfortunately prevented from being carried to completion, have furnished additional information. The hundreds of unpublished photographs taken by the Museum expeditions, as well as the fine collection of original sculptures which are now available for study at this institution, were indispensable aids in the present work. At Copan there are about twenty-five stelae. At least fifteen of these fall into a remarkably homogeneous series presenting the human figure in ceremonial attire. In connection with some of the stelae occur altars also decorated with carvings. There are, in addition, a considerable number of altars that are independent of the stelae. There are abundant remains of temples which had elaborate fagade and interior decorations. An attempt will be made to throw this mass of sculpture into its proper chronological sequence. The stelae of the homogeneous series, already mentioned, will now be consid¬ ered. Each monolith shows on the front and sometimes on the back also a human figure considerably larger than natural size and richly attired that holds in its arms an object which has already been described as the Ceremonial Bar. This human being stands in a perfectly symmetrical pose, with the heels together and both arms held at the same angle. The figure wears an elaborate headdress, fre¬ quently consisting of one animal head over another. At the side of the face are ear plugs with ornaments attached; upon the wrists and ankles are decorative bands, usually carved to represent serpent heads; about the waist is a heavy girdle to which are attached small human or grotesque faces and a fringe of sea shells. From the center of the girdle hangs an apron which is also elaborately ornamented. In the course of this study the stelae were first arranged in groups according to the proportions of the human figures carved upon them. It was found that this simple method threw them into a definite series in which other progressive variations were easily noted. For instance, the details of dress, of pose, and the degree of relief, all pass through a similar harmonious modification. A brief description of a number of typical stelae will serve to indicate some of the progressive changes. The drawings in Maudslay’s great work may be used to advantage in following this description, because they show the details of dress and ornament so clearly, but it must be stated in advance that the delineation has been more or less standardized. The cruder sculptures are frequently over¬ drawn in the style of the better ones. The face of Stela P, for instance, is very much overdrawn. Stela P shows a tall slender figure wearing a jaguar skin skirt and standing with the heels together and the feet turned straight outward. A portion of the torso is distinctly visible above the girdle. In the arms, which are held with the elbows close to the body and the forearms nearly vertical, there is supported the Ceremonial Bar, which in this case has a pendent body ( Fig. 46, a). About the neck of the personage represented on this stela is a collar consisting of a grotesque face, and upon the breast is a small face which probably represents a stone or shell pendant. The ear plugs are circular objects to which are con¬ nected a number of serpent heads. The purely ornamental details of this monu¬ ment show very neat and careful work that does not seem at all archaic. But the carving of the legs, arms and face is flat and crude, with sharp edges. The face in particular is very badly done, and the eyes protrude in the same way as on the archaic Greek sculptures. CHRONOLOGICAL SEQUENCE. 157 Ste/a P Stela 2 Ste/a I Stela 2 is another slender stela, although the slenderness is not so marked as in the preceding one. The body maintains the same pose, except that the fore¬ arms are not held nearly so vertical. The details of dress are remarkably similar, although in general somewhat more complicated. The torso is broader and the legs shorter and more muscular. The relief is somewhat higher than on Stela P, but is hardly less angular. Both of these stelae show the outlines of the shoul¬ ders and waist clearly. The carvings on Stela I are much more complicated. The figure wears a mask over the face. The Ceremonial Bar is of the pendent type, but represents a dead snake. The girdle is heavier than heretofore, and the outlines of the upper portion of the body are entirely concealed. The figure is decidedly stocky and is sculptured in rather high but angular relief. Certain details of dress and orna¬ ment are similar to those of Stelae P and 2, particularly the ear ornaments with attached serpent heads (Fig. 41) and the grotesque face under the chin. In Stela B a great change is to be noted. Many details of dress and ornament are new or modified. The pose is in general the same, but the feet are turned slightly inward in a more comfortable position and the forearms are held hori¬ zontally. The Cere¬ monial Bar no longer has a pendent body, but consists of a straight panel termi¬ nated by much-elong¬ ated serpent heads (Fig. 46, b). The gir¬ dle is exceedingly cumbersome. No part of the torso is visible. The face is visibly larger in proportion to the rest of the body than in the preceding stelae. The legs also have increased in length and the middle part of the body has been much reduced. The face is carved with a fair degree of modeling in almost natural relief. The arms and legs are also well rounded and stand out in high relief from the background. Stela N presents a confused mass of superficial ornament very deeply under¬ cut, but shows much the same pose as Stela B. The face is carved with a fair degree of naturalness, so that the cheeks show delicately rounded contours. The eyes do not protrude, although they are not very deeply sunken. The nose is carved in good relief. The legs and arms are in the full round and are carved almost free from the block. The feet are turned out, but not so much as in the case of Stela B. The marked increase in relief is illustrated in Fig. 212, which gives the cross- sections of five stelae taken at the height of the thighs. In each case the central projection is that of the apron, while on either side of this is seen the projec¬ tion of the legs (except in Stela H, which represents a skirted figure). It is readily Fig. 212. — Cross-sections of Copan stelae. 158 MAYA ART. seen from the cross-sections of Stelae P and 2 that the relief of these sculptures is low and angular and practically confined to one face of the quarried block. The relief on Stela I is higher, but still angular. On Stela A (which is very similar to Stela B, already described) the legs are not only pretty well rounded, but are carved almost free from the block. The sculpture requires about one-half of the plinth. The personage on Stela H is carved in such high relief that the greater part of a very massive block is required. In Fig. 213 is given a table of proportions for all the figures on stelae of which reasonably complete measurements could be secured. Tiie headdress is such a variable feature that no account is taken of it. It will be observed that the face is at first about twelve per cent of the length of the body, but later increases Group z Groupn Group nr £ P 2 I 3 3 GB ADNS/TH 100,---[-----.------- 9o 60 Lent M c f ft ce Ch/f ' fo fop of t irch e SO G/ri i/e . L o fr 'npe of ihir so Ski ft ; ' o it 1 tf~o/ 7 of fee t 10 Fiq. 213. — Proportions of the body on Copan stelae. to about twenty per cent. The distance from the chin to the top of the girdle is considerably greater in the case of the first three stelae than in the succeeding ones. The distance from the top of the girdle to the fringe of the skirt decreases considerably. But as a matter of fact the skirt disappears entirely in the later sculptures, and the measurement represents the width of the girdle, which shows decided increase. The distance from the fringe of the skirt or girdle to the bot¬ tom of the feet, which represents the visible length of leg, increases nearly one- half over the proportions shown in Stelae E, P and 2. So, instead of an increase of accuracy in the representation of the natural pro¬ portions of the human form, there is a marked falling off. This is due undoubt¬ edly to the heavy ornaments which overlie the body. In the later stelae these are given such high relief that they distract attention from the human form beneath; all the hidden portions of the body are dwarfed. In contrast, the bare portions have an exaggerated importance and are carefully treated. It will be CHRONOLOGICAL SEQUENCE. 159 remembered that the face, arms and legs of the later badly proportioned stelae show much more skillful modeling than do those of the earlier and more properly proportioned ones. Photographic reproductions are perhaps more convincing than drawings to indicate the range of changes which have been noted. In Plates 18 and 19 are given eight examples of sculptured stelae at Copan. Stela 7 (Plate 18, fig. 1) is probably the earliest stela at this city which bears a sculptured represen¬ tation of a human figure. 1 The carving is very low, and much of it has disappeared. Stela E (Plate 18, fig. 2) is in higher relief. The forearms are almost vertical. There seems to be no doubt that both monuments antedate Stela P. Next in order is Stela P (Plate 18, fig. 3), which has already been described. Note the flat angular face with the protruding eyes. Stelae 2, 1 and I proceed in order, but are not shown in the series of reproductions. Of these Stela 1 is interesting because it introduces a different type of subject characterized by a turban head¬ dress and a body with little elaboration. This new type is developed in Stela 6 (Plate 18, fig. 4). Two fragments of Stela 5 are given (Plate 19, figs. 1 and 2), one presenting a human face and the other a grotesque. The sculpture is considerably more advanced than in the preceding examples. Stela 3, like Stela 5, has a figure upon both front and back, the best preserved one being shown in the reproduction (Plate 19, fig. 3). This stela probably is the latest one having a Ceremonial Bar of the pendent type. Stela J may be placed with these two sculptures. It has no full-length figure, but instead a grotesque face curiously conventionalized. The new style is ushered in with Stelae A, B and D in the order named. The minor criteria of sequence in the sculptures of the last group are, 1st, the placing of the feet; 2d, the increasing use of feather drapery; 3d, the shape of the eye. The feet of Stela A are turned straight outward, although the relief of the heavy apron would have permitted the more natural pose seen in Stelae B and D. The feather trimming on Stela A is insignificant, but in D and the later monuments it plays a much more important part (Fig. 106, a, d, e and/). The eyes likewise change, although those of these three stelae are almost the same. The most excellently sculptured stelae at Copan are M, N, F, H, C and 4. The feet are in all cases turned in as far as practicable. The use of feather drapery is developed into a splendid decoration. The eyes assume more and more the form with the nearly straight top. The eyes of Stela F have circular markings to indi¬ cate the iris. Stelae N and H are reproduced in Plate 19, figs. 4 and 5, and the re¬ maining ones may be examined in Maudslay’s photographs and drawings. Even in the latest stelae the Copan sculptors did not correct the inaccuracies in proportions that have been noted. To the last the original plinth shape of the quarried block makes itself seen in the finished monument. The face pro¬ jects as far outward as does the chest (except in Stela F). These anatomical errors are not apparent in the small nude busts from the frieze of Temple 22. Besides the artistic development which has been indicated, there are other reasons for placing these monuments in the order named. Most of the early stelae lie outside of the limits of the great plaza of the city, while the later ones are set up in this plaza. But the strongest proof is furnished by the inscriptions on the sides and back of the monuments. Many of these inscriptions present 1 Fragments of a stela showing a still earlier January, 1912. A low relief headdress of a large style were found by the writer in Copan village in figure remains on one block. 160 MAYA ART. decipherable dates in the Maya system. The sequence of the dates, when they occur, is the same as the sequence that has already been indicated by the style of sculpture. This was an unexpected confirmation, and permits the time rate of change to be accurately measured. The interval between Stela 7 and Stela N is about one hundred and fifty years. This interval of time is approximately the same as that which has been marked in the course of Greek art, between the crude metopes of the Temple of Selinus and the magnificent frieze of the Parthenon. Further reference to the dates will soon be made. A study of the glyphs carved upon the backs of the stelae shows that the style of drawing in those glyphs goes through a series of changes parallel to that of the greater sculptures. We are able by a comparison of the glyphs to place a number of stelae which have no large human beings carved upon them. Some of these stelae having only glyphs are undoubtedly earlier than the stelae which have sculptured figures. Stela 9 bears the earliest definitely settled date at Copan, and the style of carving of the glyphs is very similar to that of Stela 7, which is the earliest stelae with a sculptured body. One side of Stela 9 is now plain, but this may once have been decorated with a painting or a very low relief sculpture. 1 The early glyphs of Copan are flat and rectangular, with many details given in fine incised lines. The drawing is often more vigorous and lifelike than in the later forms. Each glyph is treated as a design and made to fill exactly a rec¬ tangular space. The bars that stand for the numeral five are sharp-cornered and often have diagonal markings. There is a marked use of ornamental double outlines on faces. Gradually the glyphs lose their sharpness of outline and rich¬ ness of detail and become more rounded and simplified. The relief becomes much greater, so that there is often considerable modeling. Perhaps the most advanced carving of glyphs is seen on the Hieroglyphic Stairway, where the forms are very well modeled in high relief. In Plate 20 is given a series of inscriptions from the earliest to the latest. The sequence of the examples is as follows: 1, 5, 6, 2, 7, 8, 3, 4, 9 and 10. The last two examples are from Quirigua. Probably the earliest stela at Copan is that which has been numbered 15 (Plate 23, fig. 2). This valuable monument has been broken in two pieces in recent years, and now adorns the entrance to a pig-pen in the modern village of Copan. It bears glyphs on all four sides, and on one side is what appears to be a date. Stela 12, the eastern Piedra Pintada, has glyphs which are in very low relief. In fact, they are scarcely more than incised. They have, however, rounded outlines and so may not be so very early. Stela 10, the western Piedra Pintada, has very well-carved glyphs on all four sides. It also has a date which has so far escaped decipherment. Although the relief is rather low, the details of the glyphs are fairly -well rounded, and the monument may fall about the time of Stela I or even later. Stela 13 also is more or less indeterminate and probably belongs to the middle period. It has, however, a very primitive drum-shaped altar. Stela 8 appears to be rather late. From this it seems clear that the stelae without full-length sculptures of human beings do not as a group precede the sculptured ones, although most of them are early. In connection with each stela there was probably an altar, although most of the earlier stelae do not at present show such altars in position. Stela 13 has a 1 Unfortunately this important monument along January, 1912, and used in the foundation of an with Stela 8 was broken into a thousand pieces in adobe wall around the cemetery of Copan village. CHRONOLOGICAL SEQUENCE. 161 very crude drum-shaped altar with two glyphs carved upon the side. Altar 14 is also a drum-shaped object (Fig. 214). A knotted band encircles it, to which are attached a number of water symbols or “cloud balls,” and a crudely drawn bird. From available information it is impossible to tell with what stela this altar was correlated. Stela I has an altar (Fig. 215) still in position. It is drum¬ shaped and marked with broad knotted bands that pass around each rim of the circumference and over the top and bottom in two directions. The whole is Fig. 215. — Altar of Stela I: Copan. doubtless intended to represent a bundle. Glyphs are placed around the cir¬ cumference between the bands. These glyphs are of the same type as those on the stela with which the altar is associated. Broken altars of the same kind but somewhat simpler decoration are found in connection with Stelae E and 1. These are doubtless still earlier than the one just considered. Two drum-shaped altars are connected with Stela 5, one for each sculptured face. It is probable that the rectangular sculptured blocks X and Y found buried under Stela 5 and Stela 4, respectively, are altars of earlier stelae which were - placed under later ones for some ceremonial pur¬ pose. The carving upon these rectangular altars (Fig. 216) is very similar to that of the glyphs of the earliest period, and knotted bands are present as in the circular altars. A fragment of a similar rectangular altar (Plate 20, fig. 1) was used as one of the steps in the Hieroglyphic Stairway. 1 The old carving was turned in and so preserved. The altars of the later stelae are of diverse form, and most of them have already been described. There are a number of independent altars that are richly sculptured. Most of these can be placed pretty accurately in their proper sequence by the technique of the carving as illustrated by the stelae. It has already been noted that most of the early stelae are no longer in posi¬ tion. Stela E, placed on the platform west of the Great Plaza, was probably in conjunction with Mound 1. If the monument was originally set up here, the mound must have been one of the first large structures. Stela I was apparently in position before the eastern wall of the Great Plaza was begun, because the wall was continued around and beyond the monument and a niche made for it. The northern end of the Great Plaza was probably only completed when Mound 2 (with the correlated Stela D) was erected. The minor hieroglyphic stairway 1 Gordon, 1902, a, p. 19. 162 MAYA ART. on the front of this mound may have served as the original suggestion for the great Hieroglyphic Stairway of Mound 26. Most of the monuments and sculptures found in connection with the Great Mound or the Acropolis seem to date from the latest period of the city, following Stela A. An exception is Stela P, which seems to have been reset in the western court without its altar. The altar of this stela may be the one found under Stela 4 in the Great Plaza (Sculpture Y). Stela M is correlated with the Hieroglyphic Stairway and probably dates from the same time. It is set up directly opposite the base of the stairway. Stela N is correlated with Temple 11 in a similar manner. The carvings on the interior step of this building are of the same style as those on some of the inde¬ pendent altars, notably Altars T and Q, and are probably the work of the same sculptor. The most beautiful and perfect sculptures at Copan are those that served to decorate the facade of Temple 22. A splendid example now in the Peabody Museum is given in Plate 26, fig. 3. Other excellent sculptures, not quite so perfect (Plate 26, fig. 1), occur in connection with Structures 21, 26,32 and 36, while the rather crude faces (Plate 26, fig. 2) found in the debris of Structure 41 are apparently considerably earlier. A table that shows in a somewhat graphic manner the sequence and relation¬ ships that have just been discussed is herewith presented. Nothing seems to need further explanation except the column that gives the dates in the Maya system. Since frequent reference will be made to the dates on the monuments, it seems best to give a brief description. The subject is complex, and only a few points will be considered. A complete treatment of this difficult subject may be had in a recent work by Mr. C. P. Bowditch entitled "The Numeration, Calendar Systems and Astronomical Knowledge of the Mayas.” The periods in the first column of the table are indicated by position, the digits increasing in value toward the left. Thus the date on Altar K, written 9-12-16-7-8, may be read 9 cycles 12 katuns 16 tuns 7 uinals 8 kins. The system is based on twenty, except in the third position (tun), which is only 18 times the preceding period (uinal), so as to approximate the length of the year. Expressed in days and arranged in ascending values, the periods are as follows: Kin = 1 day. Uinal = 20 x 1 = 20 days. Tun = 18 x 20 x 1 = 360 days. Katun - 20 x 18 X 20 X 1 = 7200 days. Cycle = 20 x 20 x 18 x 20 x 1 = 144,000 days. In the inscriptions the periods are indicated by glyphs, usually grotesque faces, and the numbers that tell how many periods are taken are indicated either by bars and dots, which usually precede the period glyphs, each bar meaning 5 and each dot 1, or by definite face glyphs representing numbers from 1 to 20. Expressed in days, the date of Altar K, given above, is : 9 X 144000 = 1,296,000 12 X 7200 = 86,400 16 X 360 = 5,760 7 X 20 = 140 8 x 1 = 8 1,388,308 day l CHKONOLOGICAL SEQUENCE. 163 This vast number of days is added to a constant beginning day which is named 4 Ahau 8 Cumhu and leads to a resulting day which in this case is 3 Lamat 16 Yax. These resulting days are omitted in this table and in most of the other dates which are used in this paper. The quadrinomial system of fixing days which is seen in the names 4 Ahau 8 Cumhu and 3 Lamat 16 Yax is briefly as follows. The Maya year consisted of 18 months of 20 days each plus 5 days that did not fall in any month but which were added to fill out the year. Each of the 20 days was given a number which ran from 1 to 13, and then re¬ peated, so that the same day and number, for instance 4 Ahau, recurred every 260 days. But in addition each numbered day was further specified as occupying a certain place in a certain month. Thus 4 Ahau 8 Cumhu means, in full, the day Ahau with the number 4 falls on the 8th day of the month Cumhu. Of course each named day would occupy a fixed place in each month were it not for the 5 days added to fill out the year. This addition advances the list of days 5 days each year, with the result that each day can occupy on different years 4 places in each month. With the permutation system thus established any day whose number is given and whose place in a month is stated must occur once every 52 years, neither more nor less. This period of 52 years is known as the calendar round. Dates which have only the quadrinomial designation are referred to as being in the short count. They are accurate enough if one can be sure of the particular cycle of 52 years in which they fall. Most of the dates given in the first column belong to the class known as initial series and are definitely fixed for a very long period of time. Nearly all such dates fall in the ninth cycle, but a few will be considered later that fall in the eighth and tenth cycles. The ninth cycle is approximately 400 years long and is divided into 20 katuns each of about 20 years. The katun will be used more than any other period in comparing dates, because monuments falling about 20 years apart usually show differences that may be ascribed to the advance of culture. The tun period of about one year will be used for more specific datings. Still other methods than those given were used by the Maya. Sometimes a certain day is declared to fall on the first day of a tun or katun with a certain number and is thus fixed in the long count without a very large element of doubt. The method used in the Books of Chilan Balam and at the time of the Spanish Conquest will be described later. The second column gives the lapse of years between the first accurately dated monument and the monument before which the date in question appears. A few notes on doubtful dates are perhaps in order. Stela 15, which from artistic criteria has been placed first in the list of monuments, seems to have an initial series on one side (Plate 23, fig. 2). The inscription, however, is incomplete. Face numerals are used instead of bars and dots, and these occur in the left-hand column, while the period glyphs occupy the right-hand one. The period glyphs are indistinct except the uinal, which represents an entire frog, as on Stela D. The cycle number may be the usual 9. The face for the katun number, which is the significant number, if this is a ninth cycle date, strongly resembles the face for 4 which represents the Roman-nosed God with a kin sign in front of the ear plug. The number before the tun clearly shows a bleached bone and so must be 10 or above. The uinal is preceded by a face with a hand for the lower jaw, 164 MAYA ART. one of the regular signs for zero. The kin glyph and number are wanting. The inscription, then, may be read 9-4-10 (or over)-0-? Since there is a strong tendency to express only even, half and quarter katuns, the most probable date is 9-4-10-0-0. This would be about 40 years earlier than the date of Stela 9 and would agree with the style of the carving. Stela 13 bears an initial series which, according to a rather poor photograph, seems to be 9-11-0-0-0. Stela 10 also probably has an initial series inscription, but the glyphs are partly defaced and Maudslay’s drawing may be slightly inac¬ curate. The upper part of the first glyph should probably be the face for 9 with the dots around the mouth rather than the face given with the hand for a lower jaw. The katun seems to bear the number 15, the tun the face for 10 (?) the uinal the number 16 or 17 (the 18 is impossible) and the kin 0. But any date in the fifteenth katun is 40 years or more too late for glyphs of the style shown on this monument. The date on Stela 7 is given in face numerals and was deciphered by Mr. Morley. The early use of face numerals at Copan is worthy of note; it shows that even during the archaic period the calendar and the glyphs were already highly developed. The inscription on Stela 2 is partly defaced, but the top of the katun glyph shows a bleached bone, the usual indication of 10, and in view of the place that this monument occupies in the artistic sequence, it seems pretty certain that the date falls in the tenth katun. There is an undeciphered initial series (owing to the imperfect photographs) on the Altar of Stela 1. The dates of Stelae 3 and 5 have not yet been deciphered. The initial series on the former is largely destroyed. An important fragment showing the beginning of the initial series of Stela 5 was discovered by the writer on a recent visit to Copan. The glyphs are much worn, but with the aid of the inscriptions on the two circular altars the date of this monument stands in a fair way to be deciphered. In the stylistic sequence these monuments lie between Stelae 1 and 6 on the one hand and A and B on the other. The date of Stela 5 probably falls in the thir¬ teenth katun and that of Stela 3 in the early part of the fourteenth. The inscrip¬ tions on the two early rectangular altars, X and Y, seem to be too incomplete to admit of an exact placing in the chronology. There are, or rather were, a number of initial series on the Hieroglyphic Stairway, perhaps recording the history of the city or the lives of its rulers. The first three that are given by Gordon 1 seem to be trustworthy. They are as follows : 9-5-19-12-0. 9-8- 8- 6-5. 9-9-14-17-4. The last definite date at Copan is 9-16-10-0-0. To be sure the beginning of the tenth cycle, 70 years distant from this date, is declared on Altar S and possi¬ bly also on Stela 8. The style of the former monument is much more nearly in accord with its own initial series, w'hich marks the beginning of the fifteenth katun. As for Stela 8, the style of sculpture would place this still earlier. These two declarations may have had some prophetic significance. The dates 2 on Altars 1902, a, pp. 21-25. 1 Bowditch, 1910, Table 29. CHRONOLOGICAL SEQUENCE. 165 Gl, G2, G3, and Q are rather doubtful. They are given only in the short count, but each date falls in one of its possible positions at the beginning of a quarter katun. The dates of Stela H, F, C and 4 are as yet undeciphered. The last three have long inscriptions with an introducing glyph that resembles that of the regular initial series. 1 Stela H simply states a day in the short count which recurs every 52 years. The chronology of Copan may be summed up as follows. The earliest monu¬ ments are very crude and archaic particularly in regard to the carving of the human face. A steady improvement is noted extending from the ninth to the fifteenth katun. 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 and possibly somewhat longer. Tikal. The chronological sequence of the sculptures of Copan has been pretty definitely established. That of Tikal, a city at a considerable distance from Copan and in a different environment, will be presented in much the same way. Differences dependent upon the physical nature of available material account for only a small part of the dissimilarity between the monuments of these two cities. Both inherited the same culture, but each developed it along individual lines. Each Maya city has what may be called its personal equation. In each the same traditional ideas are presented in a somewhat different way and with a different emphasis upon details. New ideas radiated from the points of origin, becoming more or less modified in transit. Undoubtedly some cities were more progressive than others of the same period, or more fortunate in possessing artists of greater originality and builders of greater daring. Some cities were regal and others provincial. Some were great centers of wealth lying in fruitful lands, while others were poor in resources and perhaps held in tribute. Some were creators in fashion, and others mere imitators. Thus, at the same point of time a number of cities might show an unequal advance in technical skill. Some might be found still clinging to old fashions which had passed away in the more pro¬ gressive centers of art. A personal equation of time must be added to, or sub¬ tracted from the apparent time of the styles of sculpture. Besides this inequality of cities really contemporaneous, there is the further perplexing problem of real sequence in cities not contemporaneous; for many settlements were doubtless colonial offshoots of earlier centers of population or were new establishments of older cities, whose people had migrated en masse. Tikal, like Copan, furnishes examples of archaic workmanship in many pieces of sculpture. Here, however, there is no long homogeneous series of monuments, but instead a number of small groups. It is somewhat difficult to argue the line of development from one of these groups to another, because the groups, as such, were probably not strictly successive. When results are cast up, it is found that out of a mass of surmise and conjecture a small number of facts have been definitely established. These seem distinctly worth the trouble. 1 For attempts at decipherment see for Stela tant to note that the late monuments often show F, Goodman, 1897, p. 131; for Stela C, Goodman, calculations much more complicated than the early 1897, p. 130; Seler, 1899, p. 708; Thomas, 1897- ones; for instance, Stela N, Bowditch 1910 pp. 1898, pp. 776-777; Bowditch, 1910, pp. 134, 195- 186, 320-321, and Altar U, Bowditch’ 191o’ pp' 196; for Stela 4, Bowditch, 1910, p. 135. It is impor- 206-207, Altar Q, Bowditch, 1910, pp. 135 and 185. 166 MAYA ART. The stelae 1 may be arranged stylistically in the following groups: Group 1 — Stelae 3, 7, 8, 9, 13. Group 2 — Stelae 1,2. Group 3 — Stelae 4, 10, 12. Group 4 — Stelae 5, 11, 16. Group 1. In this group the human figure is represented in a style strikingly different from that which has been noted in Copan. The entire figure is shown in profile and in low relief. The proportions of the body are slender, and there is usually no cumbrous mass of dress and ornament. The pose is natural, one foot being placed slightly in advance of the other, so that the rather slender legs are somewhat separated below the short garment. The figure faces either right or left, and one hand grasps a ceremonial staff which rests upon the ground, while the other hand carries a decorated pouch. The blocks of stone upon which the figures are carved do not show careful quarrying, since in all cases the tops are roughly rounded off and the sides more or less irregular. Within this group the stelae may be arranged tentatively in the following chronological sequence: 7, 8, 13, 9, 3. Stela 9 is in the finest state of preservation and is really a very delicate and graceful piece of sculpture. It seems later than the other stelae of the same general style, but the glyphs are certainly less perfect than those of Stela 3. Of this series of stelae numbers 7 and 9 are reproduced in Plate 21, figs. 1 and 2. Group 2. Stelae 1 and 2 may be put in a class by themselves, although in point of time they probably fall within the limits of Group 1. Both show the same style of carving and both are broken and partly destroyed. The top of Stela 1 is missing and the bottom of Stela 2; between the two, however, the design can be made out very nicely. The two stones stood side by side before a small temple somewhat apart from the main plaza 2 in which were set up most of the other Tikal monoliths. There is one feature that seems to indicate that these sculptures were broken rejects. The lowermost hieroglyphs on the back of Stela 1 were never finished, but were merely blocked out in the rough. The carving of the principal figure on each monument fills the two sides as well as the front, the design being simply bent round the corners of the rectangular block. The pose as far as the head and feet are concerned is the same as in the previous group. But the shoulders and the breast are by necessity shown in front view since, instead of staff and pouch, the Ceremonial Bar is held in the arms after the manner of Copan (Fig. 58). This Ceremonial Bar has a straight central panel, and each serpent head shows the complete manikin god, ventral append¬ age and all, sitting upon the lower jaw. The sides of the stelae present a con¬ fused mass of supernumerary heads attached to a chainlike object that may represent a serpent body. The front of Stela 1 is reproduced in Plate 21, fig. 3. Group 3. Stela 4 in Group 3 is an irregular stone upon which a figure is rather crudely carved with the face in front view and in very low relief. It is different from any other stela and so cannot be accurately placed, but the style of the glyphs is that of the early period. Stelae 10 and 12 show figures carved in high relief, with the body in front view and the face in profile. In general style these two stelae are comparable to the first group of stelae at Copan, except that the 1 The stelae referred to are all reproduced by 2 The location of the stelae in this plaza is shown Maler, 1911. For a correlation of nomenclature by Tozzer, 1911, p. 119, and outside of it in pi. 29 see page 256. of same memoir. CHRONOLOGICAL SEQUENCE. 167 relief is much higher. The legs and other parts of the body have the blocklike character and the dominant angularity already noted in the archaic sculptures of Copan. The feet are turned straight out with the heels together, and the waist is surrounded by a very heavy belt. A new feature that will be frequently met with in other cities not yet considered is the captive who lies on the ground just behind the standing figure. Group 4. The three stelae in this group are carved upon large blocks of stone with straight sides and rounded tops, which have been carefully trimmed to shape. With a certain degree of assurance, these stelae may be placed in the following order: 16, 11, 5. Stela 16 (Plate 21, fig. 4) represents an elaborately dressed figure standing, with the body in front view and the head in profile and supporting horizontally in the two hands a Ceremonial Bar of unusual type (Fig. 65). The figure stands out clearly against the plain sunken background which is broken only by three short columns of raised glyphs. The carving is very flat, but the space relations of the third dimension are indicated by slightly differentiated planes. This stela is set up in the western part of the city at some distance from the main plaza, and before it lies the beautiful Altar 5. The style of sculpture on the altar is very similar to that on the stela, although the subjects are quite distinct. Stela 11 presents a pose somewhat similar to that of Stela 16, except that the Ceremonial Bar, which is of the same type, is held by one hand diagonally across the body, while the other hand is outstretched. The relief is low and the detail rich. The headdress shows a fine free use of feathers drawn in sweeping curves. The raised margin of the stela has a simple pattern decoration. On Stela 5 (Plate 22, fig. 1) the figure stands in pure profile, with the feet close together, one behind the other. The nearer hand hangs at the side and holds a decorated pouch, the farther hand holds up a Manikin Scepter by its usual serpentine appendage. The carving of the figure is in fairly high relief, with a considerable degree of careful modeling. Details of the dress are under¬ cut. In particular the elaborate feather ornament which hangs down at the back shows a skillful foreshortening of the feathers. There seems no room for doubting that this stela dates from the same period as the splendidly carved wooden lintels of Tikal. 1 The glyphs carved on the sides of this stela are identical, in style and the handling of decorative details, with the glyphs carved on the lintels. There is also a remarkable similarity between this stela and several at Yaxchilan, particularly in matter of dress. Concerning the decipherable dates of Tikal the following detailed information is contained in a letter from Mr. C. P. Bowditch under date of August 3, 1910. “Stela 3 is surely 9-2-13-0-0, 4 Ahau 13 Ivayab. If this is a historical date, it is one of the earliest known and gives evidence of Tikal having been occupied before the other cities and perhaps being the center of Central American civili¬ zation. “On Stela 10, lower part, we find 9-3-6(or 11)—2—? 9-3-6-2-0 would be 6 Ahau 8 Pax. 9-3-11-2-0 would be 11 Ahau 3 Muan. The 0-11-19 over this series are not, I think, ‘period’ glyphs. “On Stela 17 we read by means of the two dates 9-6-3-9-15, 10 Men 18 Chen. 1 Maudslay, 1889-1902, III, pis. 71-74, 77 and 78. 168 MAYA ART. “ On Stela 16 we have 7 Ahau 13 ? and on A4 we have what looks like the 1 end of tun 14,’ but no 7 Ahau appears at the end of a tun 14 in cycle 9 except 9—11— 14-0-0, 8 Ahau 18 Mol, and the month number in the inscription is 13. If, however, A4 is ‘end of uinal 14,’ we should have9-2-4-14-0, 8 Ahau 13 Muan, and the month looks more like Muan than anything else. “The Altar 5 has no means of determining its place in the long count. The dates run I Muluc 2 Muan 11-11-18 13 Manik 0 Xul 8-9-19 II Cimi 19 Mac No distance number, — 3 is needed. 1 Muluc 2 Kankin. From 1 Muluc 2 Muan to 1 Muluc 2 Kankin is 20 x 364 days = 7280 days =20 years less 20 days = 1-0-4-0 = 4(5-l-0). “So far the dates are early. But on Stela 5, if the reading of A5 is ‘end of tun 13/ as it seems to be, there can be but one date in Cycle 9, namely, 9-15- 13-0-0, 4 Ahau 8 Yaxkin, though possibly the month may be something else than Yaxkin.” From this conservative account it is evident that the dates of Tikal are few in number and difficult to place definitely. According to style of carving we have already seen that Stela 3 probably belongs to the end of Group 1 and that there are a number of similar monuments of cruder workmanship and apparently earlier date. If the initial series of Stela 3 is accepted as historical, it must follow that the art of Tikal in its beginnings antedates that of Copan. The date on Stela 3 is about 76 years previous to that on Stela 9 at Copan and about 125 years before that on Stela 7, which is the earliest Copan stela, yet discovered, having a sculptured human figure. These unexpectedly early dates at Tikal need not militate against the se¬ quence that has been established at Copan. We have seen that on the first mon¬ uments of Copan the hieroglyphs were excellently carved and in point of design were even better than on the later ones. They were sharply defined, and each was made to fill exactly a given rectangular space. The line of change was not toward more excellent glyphs but simply toward more rounded lines, higher relief and a greater amount of modeling. We have also seen that the earliest Copan subjects were elaborately attired. No figures showed simple dress ex¬ cept Stelae 1 and 6, and these represented a new departure during the middle period. Now at Tikal we find what was lacking at Copan; namely, glyphs that seem truly archaic, and figures with simple attire. The method of carving in profile is much easier than carving in front view. It is to be noted, however, that the only stela at Tikal which exhibits a mastery of modeling and foreshortening is Stela 5, that bears the date 9-15-13-0-0 and is contemporaneous with the best period of Copan. The stylistic development of the hieroglyphs of Tikal should be presented at this time, because so much of the tentative chronological arrangement of the monuments and buildings depends on the evidence they furnish. The crudest form of glyph may be seen on Stelae 8, 13 and 9. The gtyphs are placed in a single vertical column, although this fact may have no special significance. They CHRONOLOGICAL SEQUENCE. 169 are not evenly spaced and are not of uniform size. The contours (see Plate 22, figs. 2 and 3, for examples of glyphs from Stelae 13 and 9) show a lack of refinement and artistic quality, and the masses do not neatly fill out a rectangular space. On Stela 3 and on one side of Stela 10 the hieroglyphs are in double columns and exhibit more order in sizing and spacing and are better drawn. Some of the glyphs on Stela 10 have decorative detail expressed in fine, incised lines. This use of decorative etching is still more evident on Stela 12. In fact, many of the glyphs on this stela resemble markedly the earlier glyphs of Copan, but, in general, do not show such angular blocking as is seen at the latter city. The glyphs at the top of Stela 1 are well blocked out and show considerable detail. Those at the bottom were apparently never finished (Plate 22, fig. 4). The inscriptions on Stela 16 and Altar 5 have well-chosen detail and better qualities of design than any examples so far described. They have, however, the same flat surface as all the forms that have preceded. On these latter monuments we see the first use of raised strips of glyphs introduced into the field of the composition itself. Previously the glyphs have occurred only on the sides or the back of the monu¬ ment. The most skillful use of blocks of glyphs to vary the design and balance the composition will be seen at Yaxchilan and Palenque. From this consideration it becomes pretty evident that the suggested date of 9-2-4-14-0 is altogether too early for so well-sculptured a monument as Stela 16. The date 9-11-14-0-0 seems much more credible, although still very early, and the simple mistake of leaving out a bar over the month sign glyph is one that might easily occur. A reading more in agreement with the apparent date of the sculpture is offered by Mr. Morley, namely, that the inscription declares katun 14 rather than tun 14. Katun 14 begins with the day 6 Ahau 13 Muan. Owing to the use of ornamental dots to fill out glyphs, it is very easy to mistake 6 for 8. The lowest dot in A1 appears to be crescent-shaped rather than round, and if this is the case the upper dot must be the same, so that the bar and three dots are cut down to a bar and one dot. This emendation 1 leads to the date 9-14-0-0-0. The dates on Altar 5 might well fall within the 52 year period that includes the date on the correlated Stela 16. In fact, the final date on this monument would then fall just 31 days short of the date on the stela. We now come to the final and finest stage of glyph making at Tikal, — the inscriptions on the sides of Stela 5 (Plate 22, fig. 5) and on the beautifully carved temple lintels. Although not carved in high relief, the glyphs are deli¬ cately modeled and escape the dead flatness of the earlier forms. The ornamental detail is extremely rich and consists in the skillful use of double-lining, cross- hatching and beading. Under all this, however, there is a strong note of sug¬ gestive realism. It seems probable that the dates in the short count on the wooden lintels fall near the date of Stela 5, which shows the same type of carving. The reading suggested by Mr. Bowditch 2 for two of these lintels is 9-15-10-0-0, which is within three years of the date on the stela. 1 Mr. Bowditch also seems to concur in this; 1910, seems the better one. The lintel of Temple I P- 184. (Maudslay, 1889-1902, III, pi. 71) begins with 9 2 1910, Table 29. Elsewhere (p. 295) he suggests Ahau 13 Pop, which may be 9-13-3-0-0. the position to be 10-0-15-8-0. The earlier date 170 MAYA ART. The dates of Tikal seem to agree with the sequence as indicated by the style of carving. The summary of the known dates covers a period of exactly 13 kat- uns or about 257 years, as follows: Stela 3 9-2-13-0-0 Stela 16 9-14-0-0-0 Stela 10 9-3-6 (or 11)- 2- ? Lintel, Temple II 9 - 15 - 10 - 0 - 0 Stela 17 9-6-3-9-15 Lintel,TempleIV 9-15-10-0-0 Altar 5 9-13-19-16-19. Stela 5 9-15-13-0-0. Mr. Maudslay 1 makes the suggestion that the location of the principal temples of Tikal may indicate their sequence in construction. New temples may have been erected when the fairway of the old ones had become obstructed. A safer method of chronological classification concerns the methods of construc¬ tion. All the temples of Tikal have a definite type of ground-plan with very little variation in the essential parts. In all cases the room space is a very small proportion of the area covered by the walls. The roof structures are responsible for the heavy construction, but the study of these is incomplete. Dr. Tozzer found inclosed rooms in one of them. A careful plan of every part of all the temples would doubtless give data upon which the structures could be arranged in their proper sequence. Only a suggestion can be made at the present time; namely, that the temples with the largest proportion of room space are the latest in construction. The width of the rooms is especially significant. However, the size and character of the roof structures may explain the differences in floor space in the various temples rather than real advance in the building art. Upon the basis of comparative floor space the five principal temples fall into the following order of construction: V, IV, III, I and II. The difference in floor space between the two extremes of this list is well marked. 2 The well- known carved lintels which show close technical resemblances to Stela 5 prob¬ ably were taken from Temples I, II and IV. There are other temple structures besides the five upon the lofty pyramids. Most of these have ground plans very closely resembling those of the principal temples, but executed on a smaller scale. As regards the residential buildings at Tikal it may be noted that the rooms are very narrow, seldom more than six feet in width, and that the walls are thicker than the walls of similar structures at Palenque and in northern Yucatan. The narrowness may be in part due to the frequent use of a second story, which in northern Yucatan was rarely built over the room beneath but over a solid core. It seems certain that Tikal was one of the first Maya cities to become a center of art and culture. Its monuments illustrate the archaic period as well as the period of greatness. The dates, however, are few and the latest ones so far known do not extend into the sixteenth katun of the ninth cycle. It cannot safely be said that Tikal was abandoned at this time, but it is very significant that such a large city does not show the structures with the superior construction that will presently be described in neighboring cities with later dates. Earliest dated Objects. The earliest remains of Maya art which bear dates in the long-time count are two small objects of jadeite — the Tuxtla Statu¬ ette 3 and the Leiden Plate. 4 The former piece of carving was dug up near 1 1889-1902, III, p. 48. 3 See Holmes, 1910, for symposium. 3 Maler gives a table of heights of these temples, 4 Leemans, 1877, p. 299; Holden, 1879-1880, 1911, p. 50. p. 229; Valentini, 1881, b; Bowditch, 1910, p. 121. I CHRONOLOGICAL SEQUENCE. 171 San Andres Tuxtla in the southern part of the State of Vera Cruz, Mexico, and the latter was found during the excavation of a drainage ditch near San Filippo and the Gracioza River on the frontier of British Honduras. Unfortunately this town and river are not shown on any maps that have come to hand. Upon the back of the Tuxtla Statuette is a somewhat imperfect inscription which has been carefully examined by a number of authorities. The date seems to be 8-6-2- 4-17, which, if contemporaneous with the carving of the statuette, would make this object of art 403 years earlier than Stela 9 at Copan. The inscription on the Leiden Plate is 8-14-3-1-12 which is 160 years after the date on the Tuxtla Statuette and 243 years before that of Stela 9. As has been seen, the date on Stela 3 at Tikal is 9-2-13-0-0, which is 76 years earlier than the first certain date at Copan. 1 Moreover, there are a number of stelae at Tikal which are apparently earlier than Stela 3, thus reducing considerably the time to be accounted for. It must be accepted as self-evident that the Maya calendar could not have sprung suddenly into being, based as it is upon exact astronomical facts and intricate mathematical calcula¬ tions. There was no earlier civilization in the American field sufficient to furnish even the fundamental concepts of the calendar. No one can tell how long a period of observing, recording and correcting was necessary before the Maya year count was made nearly as accurate as our own, and far superior to the best that the classical culture of Greece and Rome could offer. Furthermore, other features of Maya culture must have passed through a long process of selection and evolution before the beginning of the period of recorded history. The simple pictographs of the American Indian, the only prototype that research has offered, could not in a moment have developed into a complicated hieroglyphic system. Government and religion must also have had time slowly to muster its control over the masses of the people before the great pyramids, some of which probably antedate even the most archaic monuments, could have been attempted. On the Tuxtla Statuette the initial glyph is of a very simple form, with a trifoil at the top. There are no period glyphs, the periods being indicated by position as in the Dresden Codex. If this inscription is really an initial series, calculated from the normal 4 Ahau 8 Cumhu, the resulting day and month would be 8 Caban 0 Kankin. There is a glyph at the bottom of the column which has before it the number 8, but this glyph is different from any known form of Caban. The other glyphs on the Tuxtla Statuette are exceptionally angular and lack the usual rich ornamental detail. Examples of these glyphs are shown in Fig. 217. Somewhat similar glyphs were made in northern Yucatan at the time of the Spanish conquest. 2 But it is often difficult to distinguish the crudity of first and last attempts, which in the one case arises from inexperience and in the other from decadence. Fig. 217. — Glyphs from San Andres Tuxtla Statuette. Fig. 218. — Introduc¬ ing glyph on Lei¬ den Plate. 1 Or 36 years earlier than Stela 15 if the sug¬ gested date is correct. 2 See Bowditch, 1910, for plates giving the range of day, month and period glyphs; also Brinton, 1882, d. The late demotic forms given by Landa show no angular treatment. 172 MAYA ART. The initial glyph of the Leiden Plate (Fig. 218) is comparable to the usual run of initial glyphs in the inscriptions. It shows the common hassock-shaped figure at the base, which seems to be a spread-out tun sign, as well as a head with a kin sign for the ear plug and the ribbon ornament at the top which corresponds to the trifoil of the Tuxtla Statuette. The comb-shaped figures which commonly occur at the side of initial glyphs are wanting on that of the Leiden Plate. In minor details the initial glyph under discussion seems most to resemble the initial glyph of Stela 9 at Copan. Period glyphs occur on the Leiden Plate, but they differ from those in the inscriptions. The cycle and katun glyphs seem to be turned about, since the latter rather than the former shows a hand for the lower jaw of the grotesque face. The tun glyph is of unusual form and seems to represent a fish, judging by the tail-like appendage. A very similar tun glyph appears on Stela 3 at Tikal, which is nearest to the Leiden Plate in point of time. The uinal glyph shows the characteristic curled fang at the back of the mouth. The differences in the form of glyphs from those on the stelae are such as might naturally be expected in early specimens of a complicated art. After all, the method of indicating periods of numera¬ tion by position was entirely sufficient for the needs of the time count. The period glyphs simply gave an extra artistic flourish. The same marked fondness of the Maya for unnecessary complexity led to the use of face numerals instead of bars and dots, and to cryptograms such as occur on Stela J at Copan and Stela H at Quirigua. From this examination it seems justifiable to consider the Tuxtla Statuette and the Leiden Plate as very early examples of Maya art. The dates may tentatively be considered contemporaneous with the making of the objects. The character of the drawings upon these objects deserves some slight atten¬ tion. The Tuxtla Statuette is thus described by Holmes: 1 Fig. 219. — Head of the San An- drda Tuxtla Statuette. “The upper part represents a human head with somewhat pointed crown, and with features well defined but primitive in treatment. The lower part of the face is masked with the beak of a bird, suggesting that of a duck or other water-fowl, carved in relief and extending like a beard down over the chest; while covering the cheeks and passing half-way down the sides of the beak are two mustache-like devices in low relief. The idea of the bird suggested by the beak is further carried out by wings covering the sides of the figure, the lower margins of which are engraved with alternating lines and rectangles to represent feathers. Beneath the wings in incised outline are the legs and feet of the bird.” The question might be raised whether the “bird-beak” on the lower part of the face (Fig. 219) may not have been intended to represent the nose of a serpent. There is a narrow tongue-like projection at the end. The statuette would then represent a complex of human, bird and serpent elements quite in keeping with the later developments of Maya art. The drawing on the Leiden Plate is of the utmost interest, and certain fea¬ tures have already been repeatedly referred to. The drawing represents a richly dressed figure standing with the head and lower part of the body in profile and 1 1910, p. 692. CHRONOLOGICAL SEQUENCE. 173 the breast turned nearly in front view. The feet are placed one behind the other. In the arms is held a Ceremonial Bar (Fig. 45), with pendent body such as is seen on the early stelae at Copan (Fig. 46, a). The grotesque heads in the serpent mouth at each end of the bar show the characteristic features of the Sun God. In many details of dress a close connection is shown between the drawing on the Leiden Plate and the monumental sculptures. The headdress has several heads, one above the other; the ear plugs have serpentine ornaments; the belt is adorned by small faces and large shells arranged in threes; a decorated apron hangs down from the waist, and the ankle bands show the common ser- Fig. 220. — Foot of figure on Leiden Plate showing ser¬ pent-head ankle ornaments. Fig. 221. — Feet of Copan stelae showing ser¬ pent-head ankle ornaments. pent form. Compare, for instance, the ankle ornament in Fig. 220 with those from Copan in Fig. 221. A prostrate figure with the hands tied lies on the ground behind the principal figure. It seems hardly likely that the heroic figures on the stelae could ever have been attempted without a preliminary development of the designs upon a smaller scale. The Leiden Plate is valuable as showing such an early development. Quirigua. Quirigua, distant about twenty-five miles from Copan as the crow flies, naturally shows remarkable similarity in assemblage and in monumental remains to the latter city. The agreement at Copan between the dates on the monuments and the stylistic development of the carving encourages the trial use of the dates in arranging and studying the sculptures of Quirigua. Omitting a few dates which are so far removed from the historic period that they probably had merely a traditional or mythological significance, and taking usually the latest date on the monument when there is a choice to be made, the list is as follows: 9-16-0-0-0 . . . Stela H 1 9-17-10-0-0 . . . . Altar B 9-16-5-0-0 . . . Stela J 9-17-15-0-0 . . . . Altar G 9-16-10-0-0 . . . . Stela F 9-18-0-0-0 . . . . Altar O 9-16-15-0-0 . . . . Stela D 9-18-5-0-0 . . . . Altar P 9-17-0-0-0 . . . . Stela E 9-18-10-0-0 . . . . Stela I 9-17-5-0-0 . . . . Stela A 9-18-15-0-0 . . . . Stela K 9-17-5-0-0 . . . . Stela C The choice of the quarter katun dates as the historical ones is admittedly arbitrary. Fuller details are given below. Stela J. The initial series is 9-16-5-0-0, 8 Ahau 8 Zotz. From this is made a subtraction of 1-11-13-3 leading back to 9-14-13-4-17, 12 Caban 5 Kayab. The katun coefficient of the subtrahend is apparently zero rather than 1, but if this were really the case the glyph would have been omitted entirely. 1 The date of Stela H is given on the verbal authority of Mr. S. G. Morley, who recently examined the original monument. 174 MAYA ART. Seemingly independent of this is a second subtraction from the same 9-16-5- 0-0, 8 Ahau 8 Zotz, of 18-3-14 leading to 9-15-6-14-6, 6 Cimi 4 Tzec. Both of these resultant dates appear on Stelae E and F and on Altar G. Stela F. The initial series on the west side is 9-14-13-4-17, 12 Caban 5 Kayab. To this is added 13-9-9 leading to 9-15-6-14-6, 6 Cimi 4 Tzec. The date 3 Ahau 3 Mol, which falls on 9-15-10-0-0 is also stated but apparently is not directly reached by addition or subtraction. 4 Ahau 13 Yax which falls on 9-15-0-0-0 is also declared. Then comes the distance number 1-16-13-3 which when added to 9-14-13-4-17, 12 Caban 5 Kayab, carries us to the concluding date 9-16-10-0-0, 1 Ahau 3 Zip. On the east side this last date is declared in the initial series. Stela D. The initial series on the west side is 9-16-13-4-17, 8 Caban 5 Yaxkin. Near the bottom is a secondary series of which the last two digits, 13-3, are clear. These are sufficient to raise the date to an even tun whatever the rest may be. The initial series on the east side is 9-16-15-0-0, 7 Ahau 18 Pop. Stela E. The initial series on the west side is 9-14-13-4-17, 12 Caban 5 Kayab, a date we have seen twice before. To this several additions are made. There are a number of manifest errors which are overcome by the double check of distance numbers and quadrinomial dates as pointed out by Goodman, 1897, pp. 125-127. The first addition is of 6-13-3 and leads to 9-15-0-0-0, 4 Ahau 13 Yax. A second addition of 6-14-6 carries us to the familiar date 6 Cimi 4 Tzec. A third addition of 1^1-16-15 brings us to 9-16-11-13-1, 11 Imix 19 Muan, and a fourth of 8-4-19 to the concluding date 9-17-0-0-0, 13 Ahau 18 Cumhu. On the east side the initial series gives us this last date in full. The date 13 Ahau 13 Uo is also declared, but its position is not stated. This date falls at the end of a quarter katun in 10-0-5-0-0. Stela A. The initial series is 9-17-5-0-0, 6 Ahau 13 Kayab. In another place is a quadrinomial date, 6 Ahau 13 Zac which may fall at 9-7-10-0-0. No subtraction is in evidence, although the date in this position would hark back nearly 200 years. Stela C. On the east side the initial series declares 13-0-0-0-0, 4 Ahau 8 Cumhu. This date marks the beginning of the grand cycle and is over 3,000 years earlier than 9-15-0-0-0 around which the really historical dates cluster. On the west side the initial series is 9-1-0-0-0, 6 Ahau 13 Yaxkin. Later an addition of 17-5-0-0 is declared to lead to 6 Ahau 13 Kayab. This date actu¬ ally occurs at 9-17-5-0-0 rather than at 9-18-5-0-0, showing that the secondary series was either added to 9-0-0-0-0 or that the katun value was intended for 16 rather than 17. Altar B. The hieroglyphs on this monument are very difficult to read be¬ cause they represent entire figures. Mr. Bowditch makes the initial series 9-10-0-0-0, 1 Ahau 8 Kayab, but Dr. Seler and Mr. Morley offer the reading given above (9-17-10-0-0, 12 Ahau 8 Pax). The katun glyph seems to show the Roman-nosed God with the twisted nose ornament. This head is character¬ istically used for 7 and 17. The declaration of the day and month is partly destroyed. Altar G. The initial series is clearly 9-17-15-0-0, 5 Ahau 3 Muan. The calculations that follow are complicated and the glyphs partly destroyed. It seems indisputable, however, that the date 10-0-0-0-0, 7 Ahau 18 Zip, is declared. CHRONOLOGICAL SEQUENCE. 175 This might be considered to refer to the future rather than to the past. Mr. Bowditch suggests that 10-1-0-0-0, 5 Ahau 3 Kayab, might be intended in another glyph. On the other hand the familiar dates 12 Caban 5 Kayab and 6 Cimi 4 Tzec that on Stelae J, F and E occupied the positions 9-14-13-4-17 and 9-15-6-14-6, respectively, also occur although the long distance numbers given do not seem to lead to them directly. Altar P. The initial series is clear but the succeeding calculations which may run forward into the future or backward into the past are much destroyed. The remaining monuments listed, Stelae H, I and K and Altar 0, bear initial series dates with little or nothing in the way of addition or subtraction. Altar L may have a partial initial series, but the forms are very unusual. Altar M has a distance number, 3-2-0 running from 4 Ahau 13 Yax to 6 Ahau 18 Zac. The former date falls at 9-15-0-0-0. From this detailed account it is seen that the choice of an even quarter katun as the date of erection for some of the monuments rests upon a rather slender basis. Calculations run forward and backward. The dates which might have a real historical value may be those which do not fall on an even quarter katun but are reached by calculations. Two of these occur, as we have seen, at least four times. But dates which are important in the city’s history may, after all, have no direct bearing upon the erection of the monuments. From this list it is seen at once that with a few exceptions the dates at Quirigua are later than those at Copan; furthermore, that they occur at quarter katuns or intervals of about five years. Quirigua was apparently founded well along in the historic epoch, possibly by a colony from Copan, and it may have been the place of refuge for the people of Copan if that city was really abandoned, as seems to have been the case. The course of development of the stelae and altars may be said to begin at Quirigua where it leaves off at Copan. None of the sculptures of Quirigua shows the flat archaic carving of the face that characterizes the early stelae at Copan. Instead the faces of the principal figures are carved in the full round, with eyes well sunken and noses in marked relief. The stone at Quirigua is much harder than at Copan. There is, ex¬ cept for the face, an evident reversion to the less laborious method of low re¬ lief. A recession at the shoulders, which frequently extends to the top of the stela, throws the face and the central portion of the headdress into full relief, but the arms, the legs and the details of body ornament follow the plinth-like outlines of the quarried block and have neither the high, rounded relief nor the deep undercutting of the later stelae of Copan. This reversion to flat relief occurs also at other late cities, and may be called archaistic to distinguish it from the truly archaic. While the stelae are, as a rule, taller than those of Copan, yet the proportions of the human body, as represented by the heroic figure, show the same defects or dwarfing the parts that happen to be covered with clothing or ornament. In¬ deed, the dwarfing is carried much farther than at Copan. The headdress is much elongated, and a decorated panel is placed beneath the feet so that the design as a whole is lengthened. The poses have greater freedom and variety, frequently departing from the strict observance of bilateral symmetry in the disposition of the limbs. Instead of the Ceremonial Bar the Manikin Scepter is often the principal religious object. This is held in one hand by the appendage 176 MAYA ART. so that it extends diagonally across the body. Feather drapery, skillfully and freely applied, adorns the top of most of the stelae. It must not be imagined that the artists of Quirigua drew all their ideas from Copan. There are features found there which occur at Tikal, Piedras Negras and other cities, but not at Copan. One of these is a peculiar ornament placed over the ankle (Fig. 222). This occurs widely in the Peten and Usumacinta regions. Another is the Manikin Scepter (Fig. 42, 6) with the characteristic ventral appendage in the form of a serpent. This was doubtless known to the artists of Copan, but it does not appear on any of the monolithic sculptures. The method of representing a figure sitting in a niche, which characterizes the sculptures of Piedras Negras, is seen on the back of Stela I at Quirigua (Plate 23, fig. 2). The band of planet symbols ornament: Quirigua. that arches over the niche is also a feature prevalent at Piedras Negras. The apron-like ornament below the figure in the niche finds its closest analogy on the back of Stela H at Copan. 1 So much for the general features of the stelae of Quirigua. Chronological sequence at Quirigua is difficult to determine from the art alone. The best evi¬ dence is that of increasing complexity, and this is seen in the altars rather than in the stelae. The earliest stela seems to be Stela H, which shows on the front a human figure standing upon a grotesque head and holding a Ceremonial Bar (Fig. 84, a) in the same manner as is seen at Copan, and on the back an inscrip¬ tion in a braided cryptogram somewhat like that on the back of Stela J at the latter city. The heads of the bar are represented on the sides of the monument. After its occurrence on this monument the Ceremonial Bar is seen only in decadent forms at Quirigua. No significant difference in style of carving between Stela H and Stela J is noted. The latter appears to be in somewhat higher relief and to have somewhat richer feather drapery. Stelae F, D and E are the finest monuments of this sort at Quirigua. All three are over twenty-five feet in height and are characterized by extreme elabo¬ ration of dress and by splendid use of feather drapery at each side of the head¬ dress. Each stela has a full-length figure on both front and back. The figures on these three stelae as well as those on the fronts of the Stelae A and C wear a small beard. This is likewise seen on some of the later monuments of Copan (Stelae B, C, D, etc.). These three stelae are given in the order of the dates carved on the sides. Except for a slight increase in height there seem to be no features indicating any advance from one monument to the other. The lapse of time represents only ten years, and much change is not to be expected. Stelae A and C are almost identical in style and subject. Both show a rather simply attired figure on the front and a complicated low-relief design on the back representing a figure with the face turned in profile. They were doubtless carved by a different sculptor than the three stelae just considered. The quarter katun monuments for the next five periods are monolithic altars. After these come two more stelae, I and K, both being much dwarfed in their proportions. The first of these has already been commented upon. The second is often called the Dwarf. The face is large and the body broad and short. It seems pretty clear that no real dwarf is represented, and that the bad proportions 1 Maudslay, 1889-1902, I, pi. 61. CHRONOLOGICAL SEQUENCE. 177 Fig. 223. — Altar N: Quirigua. are to be explained by the overlying ornamentation of the body which caused a similar distortion at Copan and elsewhere. The earliest altar is doubtless Altar L, which is of the circular type, with a figure sitting cross-legged in front view but with the face in profile carved upon the top. This altar seems to be the earliest monument so far found at Quirigua. The style of carving might almost be called archaic. There is a curious and apparently incomplete inscription on this monument which, according to Mr. Bowdit.ch, may be 9-14-10-?-?. Another sculpture at Quirigua which may be an altar is described as an alligator’s head (Altar M). It appears to be much later than the circular altar, although the date inscribed upon it may be 9-15-0-0-0. The remaining altars of Quirigua (Plates 1 and 2) are all large and impor¬ tant sculptures that have already been described in some detail. One of them, Altar G, represents a jaguar with a greatly modified body, the other altars present the much elaborated body of the Two-headed Dragon. Animal altars, as these may be called, are known to occur only at Copan and Quirigua, 'and so offer very strong evi¬ dence concerning the connection between these two cities. The last of the altars and by far the most complex, although all are complex enough, is Altar P. The simplest animal altar at Quirigua and the one nearest to those of Copan in style is Altar N (Fig. 223). Practically nothing is known concerning the temples of Quirigua, but it is presumed that they were of the same character as those of Copan with much less decoration. From this survey of the dates and the monuments it becomes evident that Quirigua flourished after the archaic period had passed. The changes which have been recorded witness the struggles of the artists for new effects which they hoped to obtain by complexity of form and ornament. Naranjo. The important ruins of Naranjo lying east of Tikal and near the boundary of British Honduras only recently have been made known to arch¬ aeologists through the descriptions and photographs of Maler. 1 On account of the large number of well-preserved stelae, most of which bear decipherable dates in the native reckoning, the remains of this ancient city are of especial value in the study of the history of Maya art. The buildings of Naranjo are in such an advanced state of ruin that they fur¬ nish little evidence on questions of sequence of construction. The general fea¬ tures of ground-plan and elevation are determined with difficulty, while nothing is known concerning the interior and facade decorations. The sculptured stelae set up before the temples must, however, have been intended to serve a secondary decorative function. It is the distribution and character of these monuments that demand attention. As has been explained, the chief structures of Naranjo are assembled around courts or plazas and orientated according to the four directions. Each court thus constitutes a natural unit, and with its associated temples and stelae might be expected to correspond to a definite period in the * 1908, b, pp. 80-127. 178 MAYA ART. growth of the city. At least three courts inclosed by important secular or re¬ ligious buildings appear upon Maler’s map of Naranjo. 1 In the western court are Stelae 6-11, in the middle court Stelae 12-19 and in the eastern court Stelae 20-32. Stelae 1-5 lie in the western part of the city near the acropolis and are not comprised in a regular court. As a whole, the sculptures on the stelae of Naranjo bear a greater resemblance to those of Tikal than to those of Copan and Quirigua, but in the character of subject, dress and ceremonial regalia they serve to emphasize the common basis of the culture of all these cities. The figures are carved in very low relief rather than in high relief or full round. As a consequence of the method of low relief the face and headdress are always turned in profile, although the rest of the body is shown in front view. Small faces on girdles are alone represented in full view. The usual subject portrayed is an heroic figure, standing with his feet somewhat apart and turned out, who holds in his arms the Ceremonial Bar. The distortion of the feet and the frequency of the Ceremonial Bar recall the sculptured figures of Copan. But in a number of cases the Manikin Scepter or some sort of ornamented staff or pouch replaces the Ceremonial Bar. The apron with a ^ '. .. grotesque face between two serpent heads conven- anjo, showing placing of the stelae. tionalized in the form of a fret occurs here, as in most of the cities of the southern Maya area. A number of figures, possibly intended to represent women, wear the long network skirt such as is seen on Stela H at Copan. An important feature at Naranjo, which has been observed at Tikal and on the Leiden Plate but not at Copan or Quirigua, is the presence of a bound captive beneath the principal figure. At Naranjo the principal figure stands on the back of the bound captive, while at Tikal he stands before it. These bound captives have been taken by Maler as conclusive and harrowing evidence of human sacrifice. But they may as well symbolize success in war, even as the foot of the king on the neck of the captive stands for conquest on the ancient monuments of the Far East. At the eastern end of the city stands the structure that Maler calls the Main Temple. On two terraces before this temple (Fig. 224) are arranged eight stelae. Nearest the temple are Stelae 25, 26 and 27, while in front of these on the same terrace are Stelae 28, 29, 30 and 31. Stela 32 occupies the medial position on a lower terrace, which seems to have been specially constructed to support this monument. It is pretty clear, from an examination of the sculptures, that before this one building is displayed the full chronological range of sculptural art at Naranjo. Of the three stelae in the upper row, namely, 25, 26 and 27, only the first was found in condition to be photographed. This stela (Plate 24, fig. 1) is by all odds the crudest and most archaic in the city. The figure represented upon it holds in an almost vertical position a straight Ceremonial Bar. The figure is carved very simply and there is a noticeable lack of ornamen¬ tation. The relief is very low and flat. Stelae 25 and 27 were probably similar in style to this, and the three may well have been taken from some earlier temple 1 1908, b, p. 83. See also Morley’s map, 1909, p. 544, on which the principal structures are num¬ bered as in this text. CHRONOLOGICAL SEQUENCE. 179 to be set up again in front of this one. Stelae 28, 29, 30 and 31 are much more elaborate. Of these four monuments, Stelae 28 and 29 seem to show the least ad¬ vance in sculptural art. Stela 30 (Plate 24, fig. 2) is admirably preserved, and many of the incised details of the dress come out clearly in the photograph. This stela, however, presents no real advance in the representation of the human form over the sculptures of Group I at Copan or at Tikal. Stela 32, which stands on the lower terrace, has been so much destroyed by the flaking off of the sculpture that little of the design can be made out. Apparently a figure was represented seated upon an elaborate throne and holding diagonally a Ceremonial Bar. It is still possible to make out the end of this Ceremonial Bar (Fig. 225), which consists of a very complicated scroll-work representing the highest elaboration of the serpent head. This stela is an extreme example of the general process of change leading toward flamboyant curves and complicated detail. It is undoubtedly the latest work of art that has yet come to light at Naranjo. On the northern side of the same Eastern Court is a temple before which are set up Stelae 21, 22 and 23. Stela 22 shows a style of sculpture apparently much later than the ones that flank it. As on Stela 32, the sculpture repre¬ sents a figure seated on a throne and holding a Ceremonial Bar. This object does not show the exaggerated scroll¬ work seen on Stela 32, but the carving of the throne, the lower part of which represents a complicated grotesque profile, is in an advanced style. In the Western Court before a temple facing the south are alligned Stelae 6, 7 and 8. From a stylistic comparison it seems perfectly clear that Stela 6 is earlier than Stela 8 and the latter, in turn, earlier than Stela 7. The increasing use of flamboyant lines in de¬ lineation from the date of Stela 6 to that of Stela 32 may be readily seen by comparing the Ceremonial Bar on the earlier stela (Fig. 226) with that on the later (Fig. 225). Intermediate stelae (numbers 7, 12, etc.) present an intermediate elaboration. The evidence furnished by the monuments attached to these three temples shows that correlated monuments in Naranjo, at least, are of doubtful value in determining the time of construction of the temples themselves. The placing of a number of stelae before a building seems in itself to have been a rather late development. In each of the cases so far considered the middle stela appears to be the latest. Owing to the lack of homogeneous series, it would be unwise to attempt to arrange all the stelae of Naranjo in their chronological sequence. Many are so badly weathered that the style of sculpture can no longer be determined with assurance. Stelae 2, 3 and 5 would be early monuments at Tikal, but here they are rather late. Stelae 12 (Plate 24, fig. 3), 13 and 14 before a temple in the Middle Court apparently belong to the later period of the city. Stela 10, located in the same court, has glyphs of late form, but no sculptured figure. This court as a Fig. 226. — Ceremonial Bar, Stela 6: Naranjo. Fig. 225. — One end of Ceremonial Bar of Stela 32: Naranjo. 180 MAYA ART. whole seems to be the latest at Naranjo, although the latest single monument is found in the Eastern Court. Dominating the Western Court is a structure that Maler terms the Palace of the Tiger Head Stairway. This large building has at its base a hieroglyphic stairway. The glyphs on this stairway are beautifully carved in the most ad¬ vanced style. One of the blocks is broken and half is missing. In its place is a fragment of a lintel covered with glyphs of a much earlier style. Although a large and important city, Naranjo does not deserve to be placed in the same class as Copan, Quirigua, Tikal and certain other cities which remain to be considered. The art of this city has a provincial character. The earliest examples are crude, but their crudity lacks the vital quality which distinguished the early art of Copan and Tikal. None of the monuments of Naranjo bear dates that are very early. In nearly all cases the calculations show many additions or step-ups. These step-up dates do not seem to occur on the very early monuments of the Maya. Over a long period the sculptures of Naranjo show a dead level with a few signs of progress. Toward the end there is a rapid development to¬ ward flamboyant exaggeration that in itself was a type of degeneration. The dates of Naranjo have been deciphered by Mr. Bowditch 1 and by Mr. Morley. 2 So far as known there are eight initial series dates and a few additional dates which are fixed in the long count by the declaration of a definite katun. In all other cases the dates are given in the short count and may recur at inter¬ vals of 2-12-13-0(52 years). When such recurring dates are encountered, the choice of the most probable positions in the long count are made according to two methods. First the date is chosen on which the named day marks the begin¬ ning of a whole, half or quarter katun. Failing to find such a one, the second method is to accept the date nearest the ascertained date of related monuments. The dates in the long count are as follows: Hieroglyphic Stairway Old lintel in the Stairway Stela 24 Stela 29 Stela 22 Stela 23 Stela 30 Stela 13 Stela 14 Stela 8 Stela 7 9-10-10-0-0. Katun 10 declared. 9-12-10-5-12. 9-12-10-5-12. 9-12-15-13-7. Katun 14 declared. Katun 14 declared. 9-17-10-0-0. 9-17-14-4-3. 9-18-10-0-0. Katun 19 declared. The final or latest dates on each of the stelae of Naranjo, grouped according to the structures before which the monuments are erected, will now be given. Main Temple (Structure 29). Stela 28 Stela 29 Stela 30 Stela 31 Stela 32 9-12-19-0-0. 9-14-3-0-0. 9-14-3-0-0. 9-14-10-0-0. 9-19-10-0-0. The dates of the three crude stelae that form the upper row are unknown. The next four stelae date from what corresponds to the last portion of the archaic 1 1910, pp. 102, 118—119, 129, 143, etc. Tables 29 and 31. 1 1909, pp. 545-550. CHRONOLOGICAL SEQUENCE. 181 period at Copan. They are much later than the stelae of corresponding style at Tikal, but show scarcely any more advance in sculpture. Stela 32 is the latest monument of any city so far considered. The style is the most advanced at Naranjo, but not nearly so remarkable as the sculptures of Quirigua. The date on this stela was decided by Mr. Morley 1 largely upon a consideration of the sequence of style as presented in this paper, but, apart from this line of argument, the fact that a half katun is reached in the calculation is much in its favor. Structure 27. Stela 24 Structure 26. Stela 21 Stela 22 Stela 23 Structure 23. Stela 20 9-13-10-0-0. 9-13-9-3-2. 9-13-10-0-0. 9-14-0-0-0. 9-13-2-8-16. The style of Stela 22, as has been already stated, is more advanced than that of Stelae 21 and 23, and its date seems to be altogether too early when the sculptures of this city are taken as a whole. In fact, all the early dated monu¬ ments of the Eastern Court are better than might be expected in a city whose later sculpture is so mediocre. Structure 21. Stela 19 Structure 17. Stela 12 Stela 13 Stela 14 9-17-10-0-0. 9-18-10-0-0. 9-18-0-3-0. 9-18-0-0-0. These dated monuments of the Middle Court are seventy years or more later than those of the Eastern Court, but some of them show little if any advance. Structure 15. Stela 10 Stela 11 Structure 14. Stela 6 Stela 7 Stela 8 9-19-0-3-0. 9-17-18-0-0. 9-17-1-0-0. 9-19-0-3-0. 9-18-13-0-0. The monuments of the Western Court bear uniformly late dates. Stela 10 has no sculptures except a double column of glyphs. These are of a well-rounded type and justify the extremely late date. Stela 11, however, is a reversion in style and subject. The style is not very different from that of Stela 30 (Plate 24, fig. 2) and the subject is close to that of Stela 21, which is dated about ninety years earlier. Judging by this evidence, the increase of skill during this period was almost nil. The very late character of the beautiful hieroglyphs of the Tiger Head Stairway is evident at a glance. The early date that appears in the inscription must have a memorial significance. The piece of a lintel which replaces part of one of the sculptured steps is important for several reasons. In the first place, it is valuable as an early fragment, although it may not be so very early after all. In the second place, its presence in the step to com- 182 MAYA ART. plete a broken sculpture may indicate occupation and use of the building for a long time after it was finished. Why did not the builders carve a new block and put it in place of the broken one? Stela 9 has no decipherable date. It is located on the north side of the Tiger Head Stairway. Subjectively it is one of the most interesting monoliths at this city, since it shows five figures, one larger than the others. The serpent heads of the Ceremonial Bar are decidedly flamboyant, so the stela probably dates from the eighteenth or nineteenth katun. As for the stelae belonging to Structure 14, the marked difference in style, which corresponds to the considerable difference in the dates, has already been pointed out. Stela 6 is very flat and angular. Stela 7 is, next to Stela 32, the most com¬ plicated and flamboyant sculpture in the city. Stela 8 has more rounded contours than Stela 6, but is far behind Stela 7. Structure 9. Stela 5 9-17-13-2-8 This monument in the western part of the city shows the earlier style before flamboyancy came into vogue. Stelae 1, 2, 3, and 4 of Structure 8 have no de¬ cipherable dates. Judging by the style of Stelae 2 and 3, the date may be close to that of Stela 5. This ends the list of dated monuments. The most significant fact to be gathered from this rather tedious survey is that while Naranjo started well it remained stagnant during the period from the fifteenth to the eighteenth katun, which was the most brilliant period in Copan and Quirigua. There seems to have been a sudden development during the eighteenth katun that was directed to¬ ward complicated curvilinear effects. Taken by and large, the latest dalles on the monuments agree very well with the artistic sequence. WS Fia. 227. — Ceremonial Bar, Stela 7 Naranjo. representing a bound captive is given in Fig. 228. The relief is low, but the carving is spirited and seemingly well along toward the great period. The recently illustrated stela of Motul de San Jose 4 is apparently an¬ other late piece of work. In regard to correlating the scattered monuments good 1 Maler, 1908, 6, pis. 15, 16, 17 and 18, fig. 1. 1 Maler, 1908, b, pi. 19 ' Maudslay, 18S9-1902, II, pis. 6S and 69. ‘ Maler, 1910, pi. 45. CHRONOLOGICAL SEQUENCE. 183 use can be made of identities in ceremonial regalia such as staffs. Some of the more common staffs are given in Fig. 229. Seibal. At Seibal there are a number of interesting monuments that have been figured and briefly described by Maler. 1 The forest growth is very heavy over the ruins of this city, and only a meager plan of the principal temple groups was obtained. The stelae are set up in definite relationship to mounds upon which temples formerly stood. An interesting example of correlation is seen in the case of a square mound with a splendid stela opposite the center of each side (Stelae 8-11). Before an oblong mound are arranged Stelae 5 and 7, while the shattered remains of Stela 6 lie be¬ tween. All but one of the sculptures repre¬ sented by photographs are carved in low, delicate relief with the faces in profile. The exception (Stela 2) is a rather clumsy figure in front view. The artistic quality of the Seibal monuments varies widely. Stelae 1, 3, 8, 9, 10 and 11 are among the most beautiful examples of art in the Maya area, while Stelae 2, 5 and 7 are notably crude. A careful examination of the later two monuments seems to indicate that their crudity must be explained by provincial inefficiency rather than by truly archaic ignorance. The drawing is bad; the eyes, however, are of the late form and the 229. - Ceremonial spears: a and b, Chichen ltza; c, ralenque. glyphs are rounded. Stelae 6 and 7 (Plate 25, fig. 1) bear, according to Mr. Bowditch, 2 the comparatively late date 9-17-0-0-0. It seems possible that Siebal was a city that came into power long after Tikal and Copan. Mr. Morley obtained interesting results, as yet unpublished, from an examina¬ tion of moulded fragments of Seibal monuments, some of which are reproduced by Maler as fragments of Stela 6 and of Stelae 12-15. He was able to piece to¬ gether several parts of a calculation running from 9-15-15-0-0, 9 Ahau 18 Xul, to well up towards the beginning of the eighteenth katun. Stela 1, which stands alone at the end of a ruined oblong structure, is a splen¬ didly preserved and exquisitely carved monument. The relief is low with an archaistic flatness, but the details of dress are shown on different planes. The first two glyphs are apparently 3 Ahau 3 Yax, corresponding to the quarter katun at 9-18-15-0-0-0. The four stelae numbered 8 to 11 are very similar in technical skill. Of these Stela 10 (Plate 25, fig. 2) is in almost perfect state of preservation. The sculp¬ tures all exhibit the flamboyancy that was noted in the late carvings of Naranjo and show moreover a number of definite points of resemblance to the latter city in details represented. In two cases a markedly decadent form of the Ceremonial Bar is given. In the second line of glyphs at the top of Stela 11 we see the date 7 Ahau 18 Zip, tvhich ushers in the tenth cycle declared in the following glyph. 1 190S, a, pp. 10-27 and pis. 3-10. 1 1910, table 29. 184 MAYA ART. Next comes the sign for 1 katun, and in the column of glyphs in front of the human figure is the date 5 Ahau 3 Kayab followed by the ending sign with 1 katun. This 5 Ahau 3 Kayab is doubtless 10-1-0-0-0. The same date introduces the calculation on the remaining three monuments of this very important group. It may be stated here that these sculptures are the latest accurately dated ex¬ amples of high art, that have so far come to light. There are two or three other tenth cycle dates that are a few years later than the ones above recorded, but they are not associated with sculptures of great merit. Yaxchilan. Yaxchilan, situated on the western bank of the Usumacinta River at about 17° north latitude, has been visited and described by a number of explorers. The chief works of art are sculptured stone lintels and stelae. 1 Many of the temples of this city are in a fair state of preservation. Few of the sculptures of Yaxchilan offer evidence of archaism. Many, on the other hand, show an advance in the representation of the human figure and a knowledge of grouping, perspective, and foreshortening beyond anything seen in the southern cities so far treated. The principal criteria of chronological se¬ quence at Yaxchilan are three: first, development of rounded relief out of fiat relief, such as has been already studied at Copan, Quirigua, Tikal, etc.; second, the increase of skill in perspective and foreshortening, briefly noted in the later stelae of Tikal; third, the invention of safer and lighter methods of construction. The first two criteria concern the sculptures, and the last the buildings. The lintels at Yaxchilan are usually carved on the under side, the space reserved for the carving being approximately square. In this space two or more human figures are represented in low relief. Blocks of glyphs commonly fill the corners and occupy most of the open space between the figures. One of the persons represented is, as a rule, somewhat larger than the other. They com¬ monly face each other, one being drawn in pure profile, while the other has the body in front view and the face in profile. The feet of the person shown in front view are turned directly outward in the awkward pose so frequently noted in other cities, while the figure in profile stands in a soldierly attitude with one leg con¬ cealed behind the other. The attempt to represent the human body in side view, after the front-view method had been established, seems to have led to a fair understanding of the difficult feat of foreshortening, especially in the details of breast ornaments and aprons. It should also be noted that lintel sculptures have the character of a design limited to a given space, and as a natural development of this limitation there results a feeling and an expression of that subtle balance in the grouping of points of interest which is commonly called composition. The earliest lintels of Yaxchilan seem to be those which show the carvings in low, flat relief, with details of dress and of glyphs simply incised. In the more advanced sculptures the relief is considerably higher and there is more of a feel¬ ing for well-rounded-out contours. Also there is a marked increase in artistic quality. As a rule, there are several lintels in each temple, and these commonly show a similar style of carving. The tentative order of some of the principal temples of Yaxchilan on the basis of lintel classification will now be given. There is really little choice among the first five or six positions. 1 For the notation and nomenclature see p. 259. CHRONOLOGICAL SEQUENCE. 185 Structure 1 Structure 20 Structure 33 Structure 10 Structure 16 Lintels, 5, 6, 7, and 8. Lintels 12, 13 and 14. Lintels 1, 2 and 3 Lintels 32 and 33. Lintels 38, 39 and 40. Structure 21 Structure 2 Structure 42 Structure 44 Structure 23 Lintels 15, 16, and 17. Lintel 9 Lintels 41, 42, 43. Lintels 44, 45 and 46. Lintels 24, 25 and 26. In addition to these there are some lintels with only glyphs which are not easily placed. Lintel 10 of Structure 3 seems to show rather archaic carving. Lintels 18, 19, 20, 21,22 and 23 of Structure 22 are not uniform in style of carving or in appearance. Lintel 18 has incised glyphs; Lintel 21 has glyphs in low relief; Lintel 22 has glyphs in rather high relief, hardly a single one of which is recog¬ nizable. The glyphs on this stone resemble somewhat those on Lintels 35 and 37 from Structure 12, but the latter have much more artistic quality. Lintels 27 and 28 of Structure 24 are so badly weathered that it is difficult to judge their style. They resemble the lintel which is now in the Berlin Museum. 1 It is possible that this lintel was taken from the same much destroyed structure. The stelae of Yaxchilan are arranged before temples much after the manner of those at Naranjo. As a rule, the temples which have carved lintels do not have associated stelae. Stelae 1 and 2 are apparently correlated with Temple 33, which is built upon the greater acropolis while they stand below upon the river bench. This temple also has Lintels 1, 2 and 3 and is one of the largest and best preserved buildings in the city. Stelae 3, 4, 5, 6 and 7 are aligned in front of Temple 20. Stela 3 is at a considerable distance opposite the center of the stairway, while the other four stelae are set up on the lower terrace of the temple. This important temple likewise has three carved lintels. The three Temples 39, 40 and 41 are situated upon the back portion of the greater acrop¬ olis. Before Temple 39 is a single stela, No. 10, that is placed directly in front of the doorway. Before Temple 40 are Stelae 11, 12, 13 and 14, arranged symmetri¬ cally. Temple 41 has Stelae 15, 16,17,18, 19 and 20, three being placed upon the upper terrace and the other three upon the middle terrace of the temple. The two remaining stelae are placed before two of the so-called sepulchral pyramids. As a rule, the stelae of Yaxchilan have sculptures upon both front and back faces, and before each face is a drum-shaped altar. According to Maler, 2 the side which faces the temple has a religious significance and the side away from the temple a secular one. It will be remembered that when several stelae are arranged before a building at Naranjo they are not all of the same style and period. The same situation exists at Yaxchilan. Stelae 1 and 2 are correlated with the central axis of Struc¬ ture 33, but are some distance from the structure and on lower levels. Stela 2 is nearer the temple, while Stela 1 occupies the commanding position. The latter monument is splendidly carved in a style far superior to that of the former stela and of the three lintels in the temple itself. Similarly in the case of Structure 20, the monument that occupies the position of honor, namely, Stela 3, is un¬ doubtedly a late work. There are four other stelae before this structure. The two flanking monuments, Stelae 4 and 7, are excellent pieces, and may belong to the same period as Stela 3, but the two middle sculptures are much inferior in design and are carved in lower, flatter relief. Curiously enough, the fragment that according to Maler is the upper part of Stela 5 has 1 Maudslay, 1889-1902, II, pi. 98. 1 1903, p. 126. J 186 MAYA ART. been published by Maudslay as a part of a lintel from Structure 44 (House M). Probably some confusion in notes occurred, because it is pretty evident that the fragment did not form part of a lintel and the two buildings in question are at opposite sides of the city. The three lintels of Structure 20 seem to be earlier in style than the three fine stelae, but may belong to the same period as the two crude ones. In front of Temple 39 is a single monument showing careful carvings in inter¬ mediate relief. Before Temple 40 were four (or perhaps only three) stelae. Of these Stela 11, which occupies the important position, has splendidly carved and excellently preserved designs upon front and back. The sculptures are in high but somewhat flat relief and show excellent composition. The remaining monuments, judging by Stela 13, are of a much earlier period. None of the six stelae in front of Structure 41 is stylistically of the latest period. The sculpture in most cases seems to be very low. Stela 16 is something of an exception, but the carving is much inferior to that of Stela 11 before the neighboring temple. Stelae 8 and 9 remain to be considered. No photographs of the former could be secured. The latter is a carefully executed piece with much grace. The relief is low, but the finish is smooth. The placing of these monuments in a definite order cannot safely be attempted at this time. Suffice it to say that Stelae 1, 3, 4, 7 and 11 represent the latest and best work, next in order appear to be Stelae 2, 5, 9, 10 and 16, while the remaining known monuments, including Stelae 6, 13, 15, 18, 19 and 20, are in the earliest group. Under the previous section devoted to architecture many progressive changes in construction were pointed out. As a rule, the crude beginnings were seen in the southern cities of the Maya area and the finished products in the western and northern ones. At the time the statement was made that such structural developments probably indicated chronological sequence. It is almost axiomatic that a sound principle of construction once thoroughly mastered is seldom for¬ gotten. Esthetic art ebbs and flows, but utilitarian art rises steadily and con¬ serves its positive gains. This is particularly true of architecture, as may be seen from the long history of this art in Europe. The development of roof structures has been explained in some detail (see page 110), from the cumbersome first attempts at Tikal to the airy superstructures at Palenque and in northern Yucatan. At Yaxchilan three or four stages are shown in as many groups of buildings. The simplest examples show the roof structure, in the form of a narrow wall perforated by windows, placed over the ridge pole of a one-roomed building. The weight is supported for the most part by heavy interior buttresses which divide the long narrow room into a number of compartments and necessitate a number of doorways in the outer walls. Structure 39 is an example with very heavy walls and a heavy roof crest. The room in this temple is very narrow, resembling the rooms in the temples of Tikal. The roof comb is lightened, and the proportion of wall space to room space is reduced in Structures 25, 40 and possibly 41. The attempt to lighten the load by throwing the roof comb off center is seen in Structures 20, 42 and 44. In these buildings the width of the room is increased considerably over the structures of the earlier group. The I CHRONOLOGICAL SEQUENCE. 187 buttresses become more prominent. According to Maler’s plan, the roof wall in Structure 21 was built directly over the front wall of the building in the form of a flying facade. In the next group the roof structure rises over the medial partition of a two- roomed building. There are at least two examples of this stage, namely, Struc¬ tures 23 and 30. In the first of these the interior buttresses are still seen in one of the rooms in spite of the direct support that the roof crest receives. The superstructure is still a single wall with perforations. 1 The final stage of development shows a roof structure consisting of two walls sloping inward and bonded by cross beams of stone. Each of these walls con¬ tains rows of windows. In the case of Structure 33 this double roof wall is placed over a single room and the old interior buttresses are again called into play. It may be remarked, however, that the outer walls of the temple support most of the weight and that this roof crest is much grander and more substantial than any that preceded it. According to Maler’s diagram the roof wall of this temple consists of but one wall, but the photographs and Maudslay’s sketch prove the opposite to be the case (Fig. 148, b and c). Of course it is uncertain whether this stage came before or after the stage just given showing the mechani¬ cal use of a medial wall. It is important to note that the temples of Palenque present a combination of the double-walled roof crest with the medial wall support. Another example of the double roof structure is seen in Structure 6. Here the two walls rise above the two longitudinal partitions of a three-roomed build¬ ing. The roof structure may be said to straddle the narrow interior room. There are no interior buttresses. A third example is seen in Structure 19. The correlation of these different lines of evidence with each other and with the dates given on lintels and stelae is difficult, and the conclusions are far from satisfactory. In many cases the inscriptions are incomplete, and there are no means of knowing whether or not the latest date has been deciphered. Frequently the calculations run from one lintel to another. It has been shown that the different stelae before a single building were probably set up at different times; hence the dates on these stelae are of doubtful value in determining the age of the structure. Mr. Bowditch 2 has carefully worked over the inscriptions of Yaxchilan, and the following list of dates is compiled from his results. Only the latest date in the inscriptions connected with each building is taken. The arrangement is chronological. Structure Names of sculptures Latest date 24 Lintels 27, 28 9-10-18-16-17. 44 Altar, Lintels 44, 45, 46 9-12-9-8-1. 20 Lintels 12, 13,14 1 Stelae 3, 4, 5, 6, 7 j 9-15-10-0-1. 21 Lintels 15, 16,17 9-16-?-?-?. 22 Lintels 18, 19, 20, 21, 22, 23 9-16-1-0-9. 1 Lintels 5, 6, 7, 8 9-16-1-8-6. 16 Lintels 38, 39, 40 9-16-3-3-6. 42 Lintels 41, 42, 43 9-16-4-1-1. 1 It must be confessed that Maler’s plans are Usumacinta Valley can be made stable and satis- hardly convincing on this point. Further field work factory. Single wall roof combs are found in Peten. must be carried on before the art history of the J 1903, a, pp. 27-29. / 188 MAYA ART. Structure Names of sculptures Latest date 33 Lintels 1, 2, 3. Stelae 1, 2 9-16-6-0-0. 23 Lintels 24, 25, 26 9-17-?-?-?. 10 Lintels 29,30, 31, 32,33 9-18-0-0-0. 39 Stela 10 9-18-9-12-1. 40 Stelae 11, 12, 13,14 9-18-13-13-0. 41 Stelae 15, 16, 17, 18, 19, 20 9-18-17-17-6. Structure 24, which according to this list has the earliest final date, has two lintels carved on the outer edge but not on the under surface. There are no sculp¬ tures on these lintels, except a double row of weathered glyphs. The style, however, seems to be reasonably advanced and the true date of the building may fall in the fifteenth or sixteenth katun. The temple itself is in utter ruin. In the case of Structure 44 the inscriptions on the three splendid lintels, which are of the last type of lintel carving, are undecipherable. The date given is obtained from a rectangular sculptured block which may have served as a sort of altar. The initial series date is clear enough, but certainly does not give the true date of the monument. With these two exceptions the dates on the lintels agree in the main with the grouping of these works of art upon a basis of the style of sculpture. It seems that the low-relief style flourished in the first part of the sixteenth katun. In the case of Structure 10 the lintels are accredited with a later date than the style of sculpture warrants. The calculations on these lin¬ tels are apparently accurate and specific, and lead to Fio. 2 ^ intel I ^ d Ya X cbiian Cnt ° f the beginning of the seventeenth katun if not to the eighteenth. Maler considers that the building has an older and a newer part. The first three lintels have only glyphs carved in low relief. The outlines of the glyphs are fairly well rounded, but the style does not seem at all advanced. The other two lintels are carved with figure com¬ positions in low relief. These lintels, which are situated in an L-shaped addition to the building, have gl 3 r phs that are apparently of a later type than those of the first three instances. The dates which are given to Structures 39, 40 and 41 seem to be too late, particularly in the case of the last, where the sculptures are all rather poor and in the earlier style. The interpretation of these dates in the eighteenth katun instead of the sixteenth depends upon the value of a glyph which occurs with unusual frequency at Yaxchilan. This is the katun sign, surmounted by the Ben-Ik sign and preceded by a bar. This sign has been read as meaning 18 katuns, but this reading is admitted to be more or less of a moot point. After the late dates encountered at Quirigua, Naranjo and Seibal the dates of Yax¬ chilan are fairly early. But probably they all occur after the close of the archaic period. A close resemblance in style is to be noticed between Stela 5 of Tikal and the best group of stelae at Yaxchilan. The method of represent¬ ing the eye at Yaxchilan (Fig. 230) is the same as seen on the later stelae of Copan. The dates of Yaxchilan are noteworthy because of the comparative scarcity CHRONOLOGICAL SEQUENCE. 189 of even quarter katuns. The declarations of odd days are obviously more apt to refer to definite events than those which fall at the end of a recurring five-year period. The earliest date at Yaxchilan is an initial series on Lintel 21 that regis¬ ters 9-0-19-2-4, 2 Kan 2 Yax. This date precedes any known date at Copan or Tikal, and while it may refer to the traditional history of the people of Yaxchilan it is not possible to associate it with any archaic sculptures. A secondary series of 15-1-16-5 brings the final count of this lintel into the sixteenth katun. Stela 1 may read 9-11-12-0-0, 3 Ahau 8 Chen, and may likewise advance into the six¬ teenth katun. An early initial series is found on an altar near Structure 44. The date reads 9-12-8-14-1, 12 Imix 4 Pop. Although this date does not count for¬ ward more than a few months in the secondary series, there is good reason to believe that the true date of the carving is at least fifty years later. Most of the increase in the count on the other monuments covers a comparatively short period and usually falls in or after the sixteenth katun. Maler mentions a hieroglyphic stairway in connection with Structure 5 but gives no photographs of the inscrip¬ tions. From this inconclusive survey it is evident that Yaxchilan promises splen¬ did results to the archaeologist of the future. Piedras Negras. Piedras Negras 1 is situated on the Guatemalan side of the Usumacinta River, about half-way between Yaxchilan and Tenosique. At this city Maler photographed a considerable number of stelae and a few lintels and large table altars. The stelae vary widely in subject and appearance, but are remarkable for the common use of high-relief sculpture showing the face in front view. In a number of cases a small seated figure is carved in high relief in a sunken niche. The sides of the niche are decorated in delicate low-relief sculpture. Most of the stelae bear figures on both faces, but usually one side is almost destroyed. These monuments are placed in front of buildings, as at Naranjo and Yaxchilan. The buildings, however, are mostly in utter ruin. The dates which have been deciphered at Piedras Negras are all rather early. The inscriptions in most cases are incomplete, however, and it is possible that there are much later dates than any so far discovered. The latest certain date is that of Stela 3, which registers the beginning of the fourteenth katun. It is possible that 4 Ahau 13 Yax given on Stela 6 may announce the beginning of the fifteenth katun, although Mr. Bowditch prefers the reading 4 Ahau 13 Uo which falls on 9-2-0-0-0. The problem that presents itself at Piedras Negras is the same that we shall find at Palenque. The known dates are much too early to accord with the advanced style of the art. In each case there is urgent need of further exploration. The earliest and latest dates on the monuments of Piedras Negras that have been deciphered with a degree of assurance are: Altar 1 13-0-0-0-0, 4Ahau8Cumhu, Stela 25 9-8-10-6-16, 10 Gib 9 Mac to 9-8-15-0-0, 10 Ahau 8 Tzec. Stela 36 9-10-6-5-9, 8 Muluc 2 Zip to 9-11-15-0-0, 4 Ahau 13 Mol. Lintel 2 9-11-6-2-1, 3 Imix 19 Ceh to 9-11-15-0-0, 4 Ahau 13 Mol. Stela 1 9-12-2-0-16, 5 Cib 14 Yaxkin to 9—13—14—13—1, 5 Imix 19 Zac. Stela 3 9-12-2-0-16, 5 Cib 14 Yaxkin to 9-14-0-0-0, 6 Ahau 13 Muan. Stela 6 9-15-0-0-0, 4 Ahau 13 Yax or 9-2-0-0-0, 4 Ahau 13 Uo. 1 Maler, 1901, is the only original authority on the monuments of this city. 190 MAYA ART. The following dates appear twice and may have some special significance. Stela 36 and Lintel 2 9-11-15-0-0, 4 Ahau 13 Mol. Stela 1 and Stela 3 9-12-2-0-16, 5 Cib 14 Yaxkin. Stela 1 and Stela 3 9-13-14-13-1, 5 Imix 19 Zac. It has been suggested that the series of dates on Stelae 1 and 3 might very well refer to the life of some individual. Before what was probably the principal temple at Piedras Negras are eight splendid stelae (Nos. 1-8), each with one well-preserved face. No two of these are alike. The first one is artistically of less interest than the others, but even here it is seen that the face and headdress are excellently carved in the full round, although the body is given in low relief. This recalls the reversion from the late method of Copan that was noted at Quirigua. The figure is that of a woman wearing a skirt with an all-over decoration of lace insertions in the form of a Greek cross. The glyphs are carved in low, delicate relief, but with well-rounded outlines and many details of enrichment. In the cases of Stelae 2,4 and 5 the face of the principal figure is turned in profile. Stela 4 is a splendid example of flat, sharp-cornered relief with much fine detail. The headdress is sculptured on several differentiated planes, so that the overlay of one detail by another is clearly indicated. Stela 5 is sculptured in somewhat higher and much more rounded relief. The subject is a man seated on a canopied throne. The canopy is a grotesque head upon the top of which sits a bird, while from the eye issues a grotesque figure that probably represents a god. Other grotesque figures are seen at the back. The personage on the throne holds in one hand a staff bearing the head of the Long-nosed God. A human being in ordinary dress stands facing him. Stelae 3 and 6 present seated figures in fairly high relief. All the features are given in low relief. The latter stela furnishes an excellent exam¬ ple of the figure in a niche. A strip of astronomical symbols combined with the Two-headed Dragon and the Serpent Bird frames in the seated person. A similar design from this city has already received comment (Fig. 57, d). Stelae 7 and 8 show standing warriors in front view. The relief is rather high, and certain de¬ tails are treated in the full round, while certain others are treated in low relief. The enrichment of the dress is remarkable. Space forbids a complete survey of the wonderful monuments of this little- known city. Careful study of the sculptures available for study fails to disclose any truly archaic specimens unless Stela 29 should be such a one. The glyphs on this broken stone resemble somewhat those on the earlier stelae of Copan. The artistic evidences indicate that Piedras Negras flourished after the fifteenth katun, which may be taken as marking the end of the archaic period. The mastery of the full round seen here is comparable to that of Copan and Quirigua. The developed form of eye is found here as at Yaxchilan, Quirigua and Copan. The course of development of this feature may be studied on the later monuments of the last-named city and thus pretty accurately dated. Another detail, the development of which may be studied in the light of a known chronology, is the placing of the feet. On Stelae 7 and 8 the feet are turned outward, but the heels are placed as far back as possible, so that the outer angle is less than a straight angle. This is likewise seen on the stelae at Copan erected after 9-15-0-0-0 and on Stela K at Quirigua. Often the relief is so low at Quirigua CHRONOLOGICAL SEQUENCE. 191 and Piedras Negras as not to permit this adaptation, but the sculptors took advan¬ tage when given the chance. It was noted at Copan and Quirigua that the elabor¬ ation of dress tended to destroy the proportions of the body. At Piedras Negras this malformation is very little in evidence, although the dresses are extremely ornate. The close resemblance between the little seated figure on the back of Stela I of Quirigua and figures in niches at Piedras Negras is another bit of oblique evidence on the lateness of this city. The date of Stela I is 9-18-10-0-0. It will be remembered that the poses on the larger monuments at Copan are stiffly symmetrical, while at Quirigua this symmetry is more or less broken up. The profile sculptures of Tikal and Naranjo are also formal, although a pose showing bilateral symmetry is naturally impossible. The grouping of two or more human figures is seen on some of the stelae at Yaxchilan as well as on most of the lintels. In the earlier cities a single human figure is represented upon each monument or upon each sculptured side. Now at Piedras Negras not only are the poses greatly relaxed in many cases, but there are also excellent examples of compositions containing several figures. The pose of the figure sculptured on Stelae 13 (Plate 25, fig. 3) is a remarkable exhibition of ease. The turning out of the feet is the only awkward feature. Real action is indicated by seeds or other objects that are thrown downward from the open right hand. The rich details of the dress, illustrating feather-work, beadwork, carved faces, sea-shell fringes and jaguar-hide garments, come out with the utmost sharpness and fidelity. Note, for instance, the plain inside foundation of the feather cloak that hangs down the back. In its triumphs over traditional defects this monument is far beyond anything yet presented in this historical consideration. Stela 14 shows a human figure, apparently a woman,- standing before another figure seated in a niche upon a high throne. To combine a richly attired person in low relief with another in high relief so that the effect is harmonious is no easy problem. The ornate apron of the seated figure lies loosely and naturally across the knees and hangs down in front. Stela 12 is perhaps not so effective, but is an even more ambitious attempt. A chieftain richly attired and holding a deco¬ rated spear in one hand sits looking downward in an easy position on a lofty throne. Below him two soldiers, one at either side, keep guard over nine miserable captives bound with ropes. The soldier on the right-hand side is excellently carved, with the torso in three-quarters view. The new desire for realism appears in the graceful disarray of the girdle fringes. It is interesting to note that Stela 1 at El Cayo, 1 an ancient city situated between Yaxchilan and Piedras Negras, is almost identical in pose with Stela 13, that has just been described and figured. A sculptured lintel from the same building at El Cayo before which this and another stela are placed bears an in¬ scription that clearly runs up into the seventeenth katun. 2 In the same connection it should be stated that the remarkable Stelae 1 and 2 of La Mar, 3 which pre¬ sent the closest analogies in grouping and freedom of action to Stela 12 of Piedras Negras, date from 9-17-15-0-0 and perhaps later. Other evidence pointing to the same conclusion of a late date for Piedras Negras is seen in the nature of the objects portrayed or omitted. It is significant that the Ceremonial Bar, which appears upon some of the earliest monuments of 1 Malcr, 1903, pp. 83-89 and pis. 34 and 35. 1 Maler, 1903, pp. 93-96 and pi 36- Bowditch 1 Bowditch, 1903,