1817
SCIENTIA
ARTES
VERITAS
LIBRARY
OF THE
UNIVERSITY OF MICHIGAN
20
TUDBOR
SL-QURRIS PENINSULAM-AMOENAM
CIRCUMSPICE
ASINING
MUSEUM

Museum
F
3429.
P2
U3
C D
University of Michigan
Benual Libray
&
UNIVERSITY OF PENNSYLVANIA
DEPARTMENT OF ARCHÆOLOGY
PACHACAMAC
REPORT OF THE WILLIAM PEPPER, M.D., LL.D.,
PERUVIAN EXPEDITION OF 1896
BY
DR. MAX UHLE
PLAN OF THE CITY AND TWENTY-ONE PLATES IN PHOTOTYPE
PHILADELPHIA, PA.:
PUBLISHED BY
THE DEPARTMENT OF ARCHÆOLOGY OF THE
UNIVERSITY OF PENNSYLVANIA
1903


UNIVERSITY OF PENNSYLVANIA
DEPARTMENT OF ARCHÆOLOGY
PACHACAMAC
REPORT OF THE WILLIAM PEPPER, M.D., LL.D.,
PERUVIAN EXPEDITION OF 1896
BY
DR. MAX UHLE
UM
PLAN OF THE CITY AND TWENTY-ONE PLATES IN PHOTOTYPE
PHILADELPHIA, PA.:
PUBLISHED BY
THE DEPARTMENT OF ARCHÆOLOGY OF THE
UNIVERSITY OF PENNSYLVANIA
1903

TRANSLATED
BY
C. GROSSE
PRINTED BY
MACCALLA & CO. INC., 237-239 DOCK STREET, PHILADELPHIA
PHOTOTYPES BY E. MOEBIUS, CAMDEN, N. J.
1903

anthropology (musi)
Pa, uni
5-18-31
24012
TABLE OF CONTENTS
PAGE
LIST OF PLATES
V
LIST OF ILLUSTRATIONS IN THE TEXT
VII
BIBLIOGRAPHY
IX
PREFACE
XI
CHAPTER I–Geographical Situation
1-3
CHAPTER II-Ethnical and Political Conditions of the District
4-5
6-7
8-9
10–11
.
12
13-15
CHAPTER III—Recent History of the Valley
CHAPTER IV-Former Explorations of the Ruins
CHAPTER V—The Ruins—General Topography of the City
CHAPTER VI–General Survey of the Graves in the City
CHAPTER VII–The Temple of Pachaka maj
CHAPTER VIII–The Cemetery in the Temple Grounds—General Remarks
CHAPTER IX—Discovery of Earlier Graves under the Base of the Temple
CHAPTER X-Grave-finds
Graves of the Ancient Cemetery under the Base of the Temple, and in Front of it under the New Soil
Finds in the Style of Tiahuanaco
16
.
19-21
22
22
22-25
26
. 26-29
29–33
CHAPTER XI—Finds in the Epigone Style
A. Pottery
B. Textiles
C. Woodcarvings
D. Ornaments of Shell, Bone and Teeth
E. Spinning-whorls
33-34
34
34
CHAPTER XII–The Graves in the New Soil
35
35
35
36
37
39–40
A. Pottery
B. Textiles
C. Metal Implements
CHAPTER XIII–Graves in a Second Stratum of the New Soil in Front of the Temple
Marks of the Inca Nobility
CHAPTER XIV—The Graves in the Uncovered Section of Cemetery I
A. Pottery
B. Textiles
Description of Objects
General Review of the Period
41
OM
41
42
43
44-45
46
47-48
49–52
52-53
CHAPTER XV—Remarks on the History of the Temple.
The Divinity of the Temple-Its Historical Figure
The Divinity of the Temple According to the Myths
The God as the Centre of an Organized Nature Cult
Conquest of the Sanctuary by the Incas
CHAPTER XVI—The City within the Inner Walls
Types of Buildings
Constructive Details
Conclusions
53-55
56
57-59
59-61
61
(iii)

PAGE
62
63
64-66
CHAPTER XVII- The Outskirts of the Town
Objects Recovered from the Outer Zone of the City
A. Pottery
B. Textiles
C. Implements of Metal
D. Utensils of Wood
. 66-68
69
69
E. Gourds
70
CHAPTER XVIII—Discoveries in Other Parts of the Valley.
The Cemetery of La Centinela
The Graves of Venturosa
71
. 71-72
72
73
73
74-76
. 76–77
77
78–79
. 80-81
. 81-82
82-83
84
CHAPTER XIX--The Sun Temple
Situation and General Remarks
The Northwest, Northeast and Southeast Fronts of the Temple
The Temple Entrances
The Southwest Front of the Temple
The Temple Plateau
Details of the Exterior of the Structure
The Architecture of the Temple and Its Builders
Interpretation of the Temple .
CHAPTER XX—The Cemetery of the First Southeast Terrace of the Temple
General Remarks ..
Cuzco Character of the Cemetery
CHAPTER XXI—Objects in the Cemetery of the Sacrificed Women
Clothing
A. Textiles
B. Pottery
C. Vessels and Household Utensils of Wood
D. Ornaments
E. Combs
F. Implements for Spinning, Weaving, etc.
84
84
89
89
90
93
94
95
95
96
CHAPTER XXII–The Convent
97
100
CHAPTER XXIII—Interpretation of the Convent .
CHAPTER XXIV–Technical Points of Buildings
102
(iv)

LIST OF PLATES
Plate 1. Panorama of Pachacamac. Seen from the Temple of Pachakamaj.
Plate 2. Fig. 1. The Sun Temple and the Temple of Pachakamaj. From the north.
Fig. 2. The Convent, Mamacona. Seen from the sunfields.
Plate 3. Plan of the Temple of Pachakamaj. Sc. 1 : 500.
Plate 4. Cemetery under the Temple of Pachakamaj. Objects in the style of the monuments of Tiahuanaco.
Plate 5. Cemetery under the Temple of Pachakamaj. Objects in the style of the Epigone period.
Plate 6. Cemetery under the Temple of Pachakamaj. Textiles of the Epigone period.
Plate 7. Figs. 1 to 9. Graves in débris of buildings under the Temple of Pachakamaj, above the original Cemetery. Objects
of later pre-Inca period.
Figs. 10 to 21. Objects from Inca graves in front of the Temple of Pachakamaj.
Figs. 22 to 236. Objects from the Cemetery of La Centinela near Lurin.
Plate 8. Cemetery surrounding the Temple of Pachakamaj. Objects of the later pre-Inca period of Pachacamac.
Plate 9. Fig. 1. View of the oldest part of the town. From the north.
Fig. 2. View of the southwestern part of the town. From the north.
Plate 10. Fig. 1. The eastern street of Pachacamac. From the east.
Fig. 2. Entrance to the principal palace of the town.
Fig. 3. Ruins of a typical palace of the town.
Fig. 4. Ruins of a palace without terraced foundations.
Plate 11. Fig. 1. View of the northern part of the town. From the west.
Fig. 2. Terraced slope of the hill with watch tower in the northwestern part of the town.
Fig. 3. Palace with sunken chambers in the eastern street of Pachacamac.
Fig. 4. Sunken chambers in the southwestern part of the town.
Plate 12. Fig. 1. View of the Sun Temple. From northeast.
Fig. 2. View of the Sun Temple. From southeast.
Plate 13. Cemetery of the outer city. Towards the north. Objects in the style of the Inca period of the town.
Plate 14. Views of the Sun Temple.
Fig. 1. Terrace of the large niches. From the west.
Fig. 2. The southwest front. From the south.
Fig. 3. Niched walls of the southwest front. Seen from the plateau.
Fig. 4. Entrance to first terrace.
Plate 15. Views of the Sun Temple.
Fig. 1. Northeast front with the ruins of the terrace passageways. Seen from the third terrace.
Fig. 2. Terraces of the southeast front with the Cemetery of the sacrificed women.
Fig. 3. Mound built of adobes. To the north from the Temple.
Plate 16. Plan of the Sun Temple. Sc. 1: 1000.
Plate 17. Figs. 1 to 4. Elevations of the Sun Temple at Pachacamac.
Plate 18. Cemetery of the sacrificed women. Sun Temple. Objects of pottery, wood, etc.
Plate 19. Cemetery of the sacrificed women. Articles of dress.
Plate 20. Fig. 1. View of the northern and eastern parts of the ruins of the Convent.
Fig. 2. Front with large niches. Convent. Seen from the court.
Fig. 3. Wall twenty feet wide in the north of the extended town.
Plate 21. Plan of the ruins of the Convent. Pachacamac. Sc. 1 : 500.
(v)

!

LIST OF ILLUSTRATIONS IN THE TEXT
FIGURE
PAGE
15
15
a
3
7
1
1. Copper knife. Spanish period. From an ancient house in Pachacamac. } n. s.
7
2. Plan of cemetery at the foot and under the base of the temple. Sc. 1 : 240 (site of excavations)
19
3. Sectional views of the cemetery at the foot and under the base of the temple of Pachakamaj. Sc. 1: 120. . 19
4. Section of temple-terraces at the site of explorations of fig. 2. Sc. 1: 120.
20
5. Similar section. Sc. 1:120.
20
6. Mural painting, exposed by excavations. Ab. 32 n. S.
20
7. Fragments of painting on the wall of a temple-terrace. Ab. 15 n. S.
21
8. Traces of mural painting on another part of the same wall. Ab. 1. n. S.
21
9. Mural painting, fishes, on another terrace-wall. Ab. 1. n. s.
21
10–13. Fragments of two or three vessels with a similar design, mythological figures. Tiahuanaco style. in. s.
24
14. Mythological winged animal. Design on ancient cup from Tiahuanaco. } n. s.
25
15. Fragment of a cup. Epigone period. } n. S.
26
16. Complete design of cup, pl. 5, fig. 2. } n. S.
26
17a. Cup. Ab. 7 n. S. Mus. 26729.
27
176. Complete design of cup, pl. 5, fig. 3. Epigone period. }n. s.
27
18a. Cup. Ab. 7 n. S. Mus. of Sc. and Art, 26710.
27
186. Design of cup. Epigone period. } n. S.
27
19. Design of cup. Epigone period. } n. s. Mus. 26728.
27
20. Decorated cup. Epigone period. {n. s. Mus. 26733.
27
21. Design of vase, pl. 5, fig. 5. } n. S. Mus. 26747 .
27
22. Skull-shaped vessel from Chimbote. } n. S. Mus. 34334.
28
23–25. Three stages of degeneration of a bird design on objects of later period. } n. S.
28
26. Fragments and meander border of a vessel of Epigone period. } n. S. Mus. 26748.
29
27–28. Fragments of vessels of Epigone period, painted with textile patterns. n. s. Mus. 27596 g. Fig. 28, Mus.
27596 f. .
* 29
29. Neck of vessel, painted like former. } n. S. Mus. 26863 a.
29
30. Complete design of embroidered fragments of cloth, like pl. 6, fig. 1. } n. S.
..
30
31. Stuff composed of pieces, woven and dyed separately. Epigone period. i n. S. Mus. 29782.
32
32. Embroidered human figure. Epigone period. n. S.
32
33. Design of a thick braid-like stuff. Epigone period. n. S.
32
34. Decorative design of animal on cup. Tiahuanaco style. } n. S.
33
35–36. Decorative heads of animal (dog) on vessels. Tiahuanaco style. } n. S.
33
37a, b. Typical forms of animal heads in textile patterns. Epigone period.
33
38–48. Spinning whorls (figs. 38–40, of bone; fig. 41, of shell ; figs. 42–44, of stone; figs. 45–48, of clay) .
34
49. Design of tapestry border, garment of brown gauze. Later pre-Inca period. n. s.
35
50. Design of an embroidered border of a poncho. Inca period, coast style. n. s.
38
51. Design of a border of a poncho, pl. 7, fig. 19. From Inca grave. n. S.
38
52. Design of a border in raised embroidery on large cotton cloth. Inca grave, coast style. } n. S.
38
53. Border like the preceding. I n. S..
38
54. Design of border of a large cloth. Similar to the preceding numbers. } n. s.
39
55. Colored braid, llaut'u. From a grave of Inca period. } n. s.
40
56. Tapestry with mythological design. From later pre-Inca period. n. S.
43
57. Three-ply weaving of pouch, pl. 8, fig. 14.
44
58. Ruins on the Mama river, above Chosica.
51
59. Sectional view of a terraced house at Pachacamac. 57o n. S.
57
60. Mode of entrance to rooms from above. Ab. o n. S.
58
n
3
1
3
1
250
n.
(vii)

FIGURE
PAGE
1
20
1
50
16
4
a
3
68
1
18
n
61. Lintel of a door, made of cane. Ab. 20 n. s.
60
62. Same lintel, from below. Ab. 21 n. .
60
63. Typical form of niches in the buildings of the town.
60
64. Wall decorated with niches. Ab. 50 n. S.
. 60
65. Cross-section of the same wall. Ab. z. n. S.
60
66. Large vessel from grave in an ancient house, outer zone of city. Inca period. 1o n. s.
63
67–76. Types of black pottery from graves of the outer city. Inca period of coastland. } n. s.
63-64
77. Black vessel from graves of the outer city. Inca period of coastland. } n. S.
64
78. Typical flask of black pottery. From graves in the outer city. Inca period of coastland. } n. S.
65
79–80. Flasks with flattened sides. From graves in the outer city. Inca period of the coastland. i n. s.
65
81–83. Types of vases. From graves of the outer city. Inca period of the coastland. } n. S.
65
84–85. Types of cooking utensils. From graves of the outer city. Inca period of the coastland.
66
86. Typical flask of red clay. From graves of the outer city. Inca period of the coastland. } n. s.
66
87. Clay figure. From a grave of the outer city. Inca period of the coastland. } n. s.
66
88. Cross-shaped ornament of a necklace. From a grave of Inca period. Outer city. n. S.
67
89. Decorative design (trees) in a tapestry. From a grave of the outer city. Inca period of coastland. } n. s. 67
90. Section of embroidered decorative design. From a grave of outer city. Inca period of coastland. n. S. . 67
91. Leaf design. Tapestry border. From a grave of outer city. Inca period of the coastland. į n. s.
68
92. Three-ply weaving of ancient Peruvian textiles. From graves of outer city. Inca period of coastland.
93. More developed three-ply Peruvian textures. From graves of outer city. Inca period of the coastland.
68
94. Woman's garment with knife-plaited trimming. From graves of the outer city. Inca period of the coastland.
is n. s.
68
95a, b. Bowl of carved wood. Shape of fish. From the graves of the outer city. } n. S.
. 69
96. End pieces of a weaver's beam. From the graves of the outer city. Inca period of the coastland. i n. S. 69
97. Small weaver's knife. Carved wood. From graves of the outer city. Inca period of the coastland. } n. s. 69
98. Knitting needle with carved ends (fish). From graves of outer city. Inca period of coastland. } n. S.
69
99. Lime-flask for coca-chewing. From graves of the outer city. Inca period of the coastland. } n. s.
70
100. Large cloth from the grave of a chieftain. Graves of La Centinela. Inca period of the coastland. so n. S. 72
101. Decorated vessel from graves of La Venturosa. Inca period of coastland. 1 n. S.
. 72
102. Elevation of north wall adjoining the two upper terraces of the southwest front of the Sun temple. To n. s. 79
103. Cross-section of same wall. T20 n. S.
79
104a-c. Engraved marks of ownership on vessels of Cuzco type. Graves of Sun temple. n. s.
. 87
105. Woman's garment, square ajsu, folded. From graves of Sun temple. 1o n. s.
90
106a-f. Various designs of women's girdles, chumpi, of the Cuzco costume. Graves of the Sun temple. ß n. S.
91
107. Woman's shoulder-cloth, llijlla. Graves of Sun temple. 20 n. S.
91
108. Typical stripes on woven pouches and other fabrics. Cuzco style. Graves of Sun temple. } n. s.
91
109. Various designs on pouches. Cuzco style. Graves of Sun temple. n. S.
92
110–111. Spade-like utensils of wood. From the graves of the Sun temple. n. s.
94
112. Pendant ornament of shell beads. Cuzco. From graves of Sun temple. } 11. s.
113. Necklace. Cuzco. From graves of Sun temple. n. S.
. 95
114-115. Copper bells. From graves of Sun temple. n. s.
95
116–118. Typical combs. Cuzco. Graves of Sun temple. n. S. .
. 96
119. Weaving staff. Cuzco type. Graves of the Sun temple. n. S.
96
120-121. Spinning whorls of wood. Cuzco type. Graves of Sun temple. n. S. .
96
122. Cuzco types of large niches in the convent. to n. S.
99
a
50
1
120
nS.
10
. s
2
95
1
60
(viii)

BIBLIOGRAPHY
SIXTEENTH CENTURY
MIGUEL ESTETE. La Relacion del Viaje que hizo el Señor Capitan Hernando Pizarro por mandado del Señor Gobernador su
Hermano, desde el Pueblo de Caxamalca á parcama' y de allí á Jauja (1533). In Francisco de Xerez, Verdadera
Relacion, pp. 119–149. Madrid, 1891.
Letter of HERNANDO PIZARRO to the Royal Audience of Santo Domingo, November, 1533, in Clements R. Markham, Reports
on the Discovery, 1872, III, pp. 111-127.
FRANCISCO DE XEREZ. Verdadera relacion de la conquista del Peru y Provincia del Cuzco. Segun la primera edicion impresa
en Sevilla, 1534. Madrid, 1891.
PEDRO DE CIEZA. La Chronica del Peru. Anvers, 1554.
PIETRO DI CIEZA. Crónica del gran Regno del Peru. Tradotta nella Italiana per Agostino de Cravaliz. Venetica, 1576.
PEDRO DE CIEZA. Segunda parte de la Crónica del Perú. La publica M. Jimenez de la Espada. Madrid, 1880 (Biblioteca
Hispano-Ultramarina, Tomo V).
AUGUSTIN DE ZARATE. Histoire de la Découverte et de la conquête du Perou, traduite de l'Espagnol. Paris, 1742, tome I.
JUAN DE BETANZOS. Suma y Narracion de los Incas. Publícala M. Jimenez de la Espada. Madrid, 1880 (Biblioteca
Ultramarina, Tomo V).
Informacion de las idolatrias de los Incas é indios y de cómo se enterraban, 1571. In Coleccion de Documentos Inéditos,
Vol. 21, 1874, pp. 131-220.
MIGUEL BALBOA. Histoire du Pérou, traduite par H. Ternaux Compans. Paris, 1840.
Relaciones Geográficas de Indias. Publícalas el Ministerio del Fomento. Peru. Tomo I, II. Madrid, 1881, 1885.
I, pp. 61–78. Descripcion y Relacion de la Provincia de los Yauyos hecha por Diego Davila Brizeño.
I, pp. 197–216. Descripcion de la tierra del Repartimiento de los Rucanas Antamarcas. Año de 1586.
II, pp. 29–37. Relacion fecha por el Corregidor de Chunbibilcas Don Francisco de Acuña.
JOSEPH DE ACOSTA. Historia Natural y Moral de las Indias. Sevilla, 1590.
F. HIERONYMO ROMAN. Republicas del Mundo. Salamanca, 1595. Especially Book I, chap. V: De los templos y lugares
sagrados que tuvieron los del Reyno del Peru.
BLAS VALERA. Extracts from his lost work by Garcilaso de la Vega.
SEVENTEENTH CENTURY
w
FRANCISCO DE AVILA. A Narrative of the Errors, False Gods and Other Superstitions and Diabolical Rites in which the
Indians of the Provinces of Huarochiri, Mama and Chaclla lived in Ancient Times, 1608. In Clements R. Markham,
Narratives, pp. 121–147.
Inca GARCILASO DE LA VEGA. Primera Parte de los Comentarios Reales. Madrid, 1723.
Inca GARCILASO DE LA VEGA. Historia General del Perù. Segunda Parte de los Comentarios Reales. Madrid, 1722.
(ONDEGARDO.) Relacion de los fundamentos acerca del notable daño que resulta de no guardar á los Indios sus fueros.
(POLO DE ONDEGARDO.) Del órden que los indios tenian en dividir los tributos y distribuirlos entre sí ; both in Coleccion
de Documentos Inéditos relativos al descubr. conqu. y organiz. de las antiguas posesiones españolas de América y
Oceania. Madrid, 1872, Vol. 17, pp. 1–177.
POLO DE ONDEGARDO. Report in Clements R. Markham, Narratives of the Rites and Laws of the Yncas, translated from the
original Spanish Manuscripts. London, Hakluyt Society, No. 48, 1873, pp. 149–171.2
Relacion de la Religion y Ritos del Peru hecha por los primeros Religiosos Agustinos. In Coleccion de Documentos Inéditos.
Vol. 3, 1865, pp. 5–58.
FERNANDO DE SANTILLAN. Relacion del Orígen, Descendencia, Politica y Gobierno de los Incas. In Tres Relaciones de
Antiguedades Peruanas. Publícalas el Ministerio del Fomento. Madrid, 1879, pp. 1-133.
CHRISTOVAL DE MOLINA. The Fables and Rites of the Yncas. Clements R. Markham, Narratives, pp. 1–64.
1 That is Pachacama.
2 The text here published corresponds to pp. 9–78 of the Spanish edition of the manuscript in the before-enumerated Documentos Inéditos. Madrid, 1872, Vol. 17.
(ix)

LUDOVICO BERTONIO. Vocabulario de la Lengua Aymara. Publicado de nuevo por Julio Platzmann. Segunda Parte
Aymara-Español. Leipzig, 1879.
JOAN DE SANTACRUZ PACHACUTI YAMQUI. Relacion de Antiguedades deste Reyno del Pirú. Tres Relaciones de Antiguedades,
pp. 229–328.
Relacion Anónima de las Costumbres Antiguas de los Naturales del Pirú. Tres Relaciones de Antiguedades, pp. 135–227.
BERNABÉ Cobo. Historia del Nuevo Mundo. Publicada por M. Jimenez de la Espada. Sevilla, tom. IV, 1893. Especially
chap. 17: Del famoso templo de Pachacama.
FERNANDO MONTESINOS. Memorias Antiguas Historiales y politicas del Perú. Madrid, 1882.
ANTONIO DE CALANCHA. Chronica Moralizada del orden de San Augustin en el Perú, 1638. Especially Vol. I, Book II,
chap. 19: Del celebrado adoratorio de Pachacamac.
EIGHTEENTH CENTURY
JORGE JUAN ULLOA Y ANTONIO ULLOA. Relacion historica del viage á la America Meridional. Madrid, 1748.
Y
NINETEENTH CENTURY
W. B. STEVENSON. A Historical and Descriptive Narrative of Twenty Years' Residence in South America. London,
1825, Vol. II.
CHARLES WILKES. Narrative of the U. S. Exploring Expedition during the years 1838–1842. Philadelphia, 1845, I.
J. J. von TSCHUDI. Reiseskizzen aus den Jahren 1838–1842. St. Gallen, 1846, Vols. 1 and 2.
WILLIAM H. PRESCOTT. History of the Conquest of Peru. London, 1855, Vols. 1 and 2.
FRANCIS DE CASTELNAU. Antiquités des Incas et Autres peuples anciens, Recueillis pendant l'Expédition dans les Parties
Centrales de l'Amérique du Sud. Paris, 1854.
Mar. ED. DE RIVERO Y J. D. DE TSCHUDI. Antigüedades peruanas. Atlas. Vienna, 1851.
MAR. EDWARD RIVERO Y JOHN JAMES von TschudI. Peruvian Antiquities, Translated into English by Francis L. Hawks.
New York, 1855.
J. J. von TSCHUDI. Die Kechua Sprache. Part 3. Wörterbuch. Wien, 1853.
THOMAS EWBANK. A Description of the Indian Antiquities brought from Chile and Peru in U. S. Naval Astronomical
Expedition to the Southern Hemisphere, 1849–1852. Washington, 1855. Appendix E.
CLEMENTS R. MARKHAM. The Travels of Pedro de Cieza de Leon, 1532–1550, contained in the first Part of his Chronicle of
Peru. London, Hakluyt Society, No. 33, 1864.
CLEMENTS R. MARKHAM. Reports on the Discovery of Peru, Translated and Edited with Notes and an Introduction. London,
Hakluyt Society, No. 42, 1872.
THOMAS J. HUTCHINSON. Two Years in Peru. London, 1873, Vols. 1 and 2.
ANTONIO RAIMONDI. El Perú. 3 volumes. Lima, 1874–1879.
GUSTAV Brühl. Die Culturvölker Alt-Amerikas. New York, Cincinnati and St. Louis, 1875–87.
MARIANO FELIPE Paz SOLDAN. Diccionario Geográfico Estadístico del Perú. Lima, 1877.
E. GEORGE SQUIER. Peru. Incidents of Travel and Exploration in the Land of the Incas. New York, 1877.
A. BASTIAN. Die Culturländer des Alten America. Vol. 1: Ein Jahr auf Reisen. Berlin, 1878.
Vol. 3, pt. 2: Nachträge und Ergänzungen aus den Sammlungen des Ethnologischen Museums. Berlin, 1889.
Congrès International des Américanistes. Comptes Rendus de la III, VIII, X Session. Bruxelles, Berlin, Stockholm, 1879,
1888, 1894.
W. Reiss und A. STÜBEL. The Necropolis of Ancon in Peru. Translated by A. H. Keane. 3 volumes. Berlin, 1880–87.
CHARLES WIENER. Pérou et Bolivie, Récit du Voyage. Paris, 1880.
DANIEL G. BRINTON. American Hero-Myths. Philadelphia, 1882.
J. J. von TSCHUDI. Organismus der Khetšua Sprache. Leipzig, 1884.
A. STÜBEL, W. Reiss und B. KOPPEL. Kultur und Industrie südamerikanischer Völker. Berlin. Vol. 1: Vorspanische
Zeit, 1889.
Vol. 2: Neue Zeit. 1890.
E. W. MIDDENDORF. Die Einheimischen Sprachen Perus. Vol. 2: Wörterbuch des Runa Simi oder der Keshua Sprache.
Leipzig, 1890.
Vol. 3: Ollanta, Ein Drama der Keshua Sprache.
J. J. Von TSCHUDI. Culturhistorische und Sprachliche Beiträge zur Kenntniss des Alten Peru. In Denkschriften d. k.
Akad. d. W., Wien. Philosophisch-Historische Classe. Vol. 39, 1891, I.
DANIEL G. BRINTON. The American Race. New York, 1891.
A. STÜBEL und M. UHLE. Die Ruinenstätte von Tiahuanaco im Hochlande des Alten Peru. Breslau, 1892.
E. W. MIDDENDORF. Peru. Beobachtungen und Studien. Vol. 2, Das Küstenland von Peru. Berlin, 1894.
E. T. Hamy. Galerie Américaine du Musée d'Ethnographie du Trocadéro. Choix de Pièces Archéologiques et Ethnographiques.
Paris, parts 1 and 2 (ab. 1897).

PREFACE
awe.
I
the course of early history in ancient Peru, several sanctuaries successively arose in importance above all others. Such
were the unfinished temple of Tiahuanaco, the marvelous structures of which, erected near Lake Titicaca at an altitude
of thirteen thousand feet, were so famous in myth and legend ; the sanctuary situated in the island of Titicaca in the lake
of that name; the Qorikancha, the magnificent temple of Cuzco, and, lastly, the temple of Pachakamaj, the creator god, at
Pachacamac. Throughout the period immediately preceding the Spanish conquest, the three last-mentioned temples rivaled
one another in importance. But of these, and in spite of the splendors of the Qorikancha, the “Golden Court” of Cuzco,
or the well-established fame of the island temple of Titicaca, the shrine of the creator was looked upon with greatest
Reports of the glories of this temple had reached Francisco Pizarro when, at Cajamarca, on his expedition of
conquest, Atahualpa informed him that at a ten days' journey on the road to Cuzco there was a wonderful “ mosque,”
regarded by the entire people as the chief sanctuary of the land, and held in highest reverence by himself and by
his father, Huaina Capac, before him. He added, that although each town had a special “ mosque ” for the worship of
the local deity, the shrine of the chief divinity of the land, to which rich offerings of gold and silver were brought
by the people of the entire country, possessed vast treasure.
According to Estete : “ Pilgrims coming from all parts of Peru, three hundred leagues or more, flocked to this
shrine, as the Turks and Moors flock to Mecca”.
For these reasons the writer, upon his arrival at Lima in January, 1896, decided to visit and explore the ruins
of Pachacamac, which lie in the vicinity of Lima. He found the most hospitable welcome in the house of Don
Vicente Silva, the proprietor of the hacienda San Pedro, on the territory of which the ruins are situated, and during
his sojourn of many months he received the most friendly help in his labors, both from the proprietor and his son,
Don Ricardo Silva.
The short visit planned at first was soon found to be inadequate, as it became more and more evident that the
historical problem of the ancient city could not be solved in such a limited period. Some of the unforeseen diffi-
culties met with may be mentioned here. The first graves opened, instead of clearing up former problems, presented new
Their contents could not be classified, and accordingly the cultural and historical position of the ancient sanc-
tuary became more obscure than before. Scattered over the site of the ruined city were found numerous fragments of
pottery of ancient date, corresponding to none of the vessels taken from the newly-opened graves. The search, there-
fore, had to be continued until burial-places could be discovered furnishing material dating of the same period as those
fragments. By this method only might the history of the town ultimately be cleared up, and the objects recovered
from the first graves be assigned to their proper period.
It may be added that no adequate plans or views of the city had ever been published, hence the first duty that
presented itself was the preparation of a plan of the city, in addition to the taking of a number of photographic views.
The writer begs to express here sincere gratitude to the late Dr. William Pepper for his liberal support, and to his
Peruvian friends, to whose kindness, indulgence and assistance he is indebted for much of his success in his undertaking.
a
ones.
PHILADELPHIA, 1898.
THE AUTHOR.
(xi)


CHAPTER I.
GEOGRAPHICAL SITUATION.
3
THE peculiar character of the Peruvian coastland consists
in a succession of vast, barren deserts and small, ex-
tremely fertile valleys. The former are the natural result of
the almost entire absence of rains along this coast, while the
fruitfulness of the valleys depends on the shallow coastal
rivers, which run their short, rapid course down from the
Cordillera to the ocean. By a wise provision of nature these
rivers are filled to overflowing during the heat of the sum-
mer, when it is the rainy season in the highlands, and so
the vegetation of the valleys is saved from drought. In
winter the water-supply decreases with the end of the rainy
season in the mountains, until there is barely enough to keep
the arid interior of the valleys from parching. Wherever
the water-supply is not sufficient for the purposes of irriga-
tion, which is the case near the mouth of these valleys, there
dense fogs from the ocean form a zone and protecting cover
over ten miles wide, stretching along the entire Peruvian
coast, and descending as rain upon the nearest headlands of
the interior. The slopes of these foothills then clothe them-
selves with fresh verdure, and the Indian herdsmen with
thousands of cattle come down to these 'lomas' and stay
until the beginning of the hot season, when again they seek
the higher altitudes.
A happy combination of human skill and the resources
of nature have brought about the flourishing condition of
these valleys. An admirable system of artificial irrigation,
begun thousands of years ago, carries and distributes the
water in numerous branches and side-arms, 'mother' and
daughter' azequias, from points in the upper valley
throughout the fields of the plain. The watered arable land
may be seen contrasting in a distinct horizontal line of ver-
dure against the yellow sand of the desert just above it.
These irrigation works were as complicated and perfect as they
are now before the coming of the Spaniards, who became the
heirs and successors to all that the industry of the Indians had
created.
The ruins of the ancient city of Pachacamac are situated
in the valley of Lurin, the loveliest of these valleys, to the
south of the valley of Lima and about seventeen or eighteen
leagues? from the present capital. The stream which flows
through the Lurin valley is smaller than the Rimac of the
Lima valley, but its water-supply is abundant and steady,
while the Rimac nearly runs dry in winter, since the latter
springs from the snow-fields of the Cordillera, and the Lurin
has its headwaters only in the foothills of the Cordillera, near
Huarochiri ; yet even this small watercourse may become
an impassable river during heavy rainfalls. The river is
not known to have any name of its own. In early records,
such as maps 2 and reports, it is indicated as the Rio de
Pachacamac, and it was known under this designation in its
upper course in the Huarochiri province. The river, the
sanctuary and the town bear their present common name of
Pachacamac only since the period of the Inca conquest,
about 170 years previous to the Spanish invasion, and so far
it could not be learned what the older name may have been.
The Lurin valley, owing to its smaller river, is not as
wide as the Lima valley. It forms a strip of land which near
the coast is two miles and a half wide and stretches inland for
a mile and a half, where it is narrowed in by projecting rocks
from the south side to only one mile and a half. At ten miles
from the coast it measures barely two-thirds of a mile; at six-
teen miles, near Cieneguillas, it is about a quarter of a mile
wide, and at twenty miles from the sea the Quebradas of the
mountains begin. The valley of Lurin is most charming.
Cieza4 calls it “ delightful and abounding in fruit,” and it has
probably not changed in that respect since the time of the
ancient Peruvians. Lines of tall trees, mostly willows (salix),
or thickets of cane (canillo) mark the courses of the canals
which intersect the fields. Small groves of waranga (Legu-
minosa sp.) and other trees may be seen here and there; also
fruit-trees, such as the Lucuma, Pakai, Palta and Chirimoya,
are numerous. Two small villages are situated in this valley
at the present day, Lurin and Pachacamac, and several haci-
endas, the largest of which is San Pedro. A large number
of the mestizos live in airy cane huts among the fields. The
modern village of Pachacamac must not be confounded with
the ruins or the site of the ancient city of the same name ;?
2 Compare the map in Descripcion y Relacion de la Provincia de Yauyos,
in the Relac. Geogr., I, p. 61.
3 Francisco de Avila, p. 142.
4 I, chap. 72.
5 The early agricultural products must have been mainly maize, camote,
cotton, etc. Seed-pods of the cotton plant are often found in graves. In modern
times the cultivation of sugar-cane was carried on for some time, but had to be
given up
See illustration in George Squier, Peru, p. 66.
7 The word Pachacamac will be used in the present report in three different
meanings, in each of which it is of general use :
1. For the ancient town on the right bank of the river.
2. For the modern village, founded in Spanish time, on the left bank of
the river, three miles inland.
3. For the divinity which was specially worshiped at Pachacamac and
which gave its name to the ancient city. Fernando de Santillan, p. 33: “The
. waka’ said her name was Pachacamac.” Cieza, I, chap. 72: “The name of this
devil means Creator of the world.'”
Pachacamac in Kechua stands for the Creator of the world,' 'He who
animates the universe,' 'the world-adjuster.'
The word Pachacamac is written in various spellings. The form now gen-
erally accepted for the geographical name is also found in the works of Garcilaso,
1 Latitude and longitude of Lima, according to Paz-Soldan, Diccionario, p.
513, are: Long. 770 16' 20'', lat. 120 2/ 34"' ; latitude and longitude of the ruins
of Pachacamac as given by A. Raimondi, III, p. 96: Long. 760 59' 27!!, lat.
12° 15' 19''. Assuming these figures to be approximately correct, the distance
between Lima and the ruins would be from seventeen to eighteen miles or about
six and a half Peruvian leagues, which nearly agrees with the seven leagues usu-
ally accepted in the neighborhood as the distance between the valley of Lurin
and the city of Lima.
(1)

a
it lies about three miles inland from the ruins on the oppo- *
site bank of the river, a mile away from it. The site of
Lurin is, like that of the ancient city, on the road which
skirts the coast, the “camino real.' It is on the same side
of the river as the ruins, but at a distance of a mile and a
half from either. Between these lies the hacienda San
Pedro, at about two-thirds of the distance from the river to
the village. Lurin has 500 inhabitants and the entire parish
about 2000; the modern village of Pachacamac has 300 and
the entire parish about 3000 souls; the district of the haci-
enda Cieneguillas in the upper valley comprises circa 5000
persons. The inhabitants of these villages are poor and are
nearly all economically dependent on the haciendas, mainly
on San Pedro.
The ruins of the ancient town are situated on the north-
ern limit of the valley, upon the right bank of the river,
near its mouth and only 600 yards from the ocean. The
desert Tablada de Lurin lies between the valleys of Lima
and of Lurin, the shallow beach and the towering mountain-
range of the interior;2 southward it slopes down to the val-
ley and it is bordered by the green fields of the plain on its
southeastern- and southwestern-, the valley- and the ocean-
sides, while on the bluffs which rise above the valley it is
from fifty to eighty feet higher than the plain ; several hills 3
rise from 150 to 250 feet above the level of the desert.
The territory between and around these hills comprised the
site of the ancient city. On the other side the ocean rolls its
mighty waves upon the sandy shore with an unceasing thun-
dering roar, and two or three miles from shore the pictur-
esque barren islands of Pachacamac may be seen with clouds
of pelicans and other sea-fowl above them. To the south, the
glance sweeps along the gently curving coastline as far as
Cape Bubulusa, barely discernible in the blue haze at a
distance of about twenty miles. Swept by the sea winds, the
sand of the desert drifts over the ruins. A few Tillandsia
plants, green only in winter, represent the scanty, almost
invisible vegetation; a hooting owl, a lizard sunning himself,
or perhaps a small viper, are the only signs of animal life
in this solitude. The sun burns fiercely upon the parched
soil, still a cooling breeze from the near-by ocean moderates
the heat, so that ninety and fifty-five degrees are the limits
of temperature rarely exceeded in winter and summer.
There is a charming view across the valley from several
points of the ruins. The church-spires of the modern vil-
lages of Pachacamac and of Lurin may be seen in the dis-
tance. The manor-houses of San Pedro and of the hacienda
of Buena Vista stand out upon a picturesque bluff above the
green groves of the plain. Farther beyond stretches the roll-
ing desert to the southward in the direction of Chilca ; to
the left, at a distance of forty-five miles, the foothills of the
Andes rise gradually to a height of 9000 feet and reflect at
sunset the rays of the evening sun. In summer one may
watch the course of the rainy season high up in the moun-
tains from the valley here below, and in winter the snow-clad
peaks may be seen looming up above this subtropical plain.
The presumable age of the present coastline is a question
of some importance for Pachacamac. At one time these
islands were doubtless connected with the mainland, as the
ocean-channel is very shallow at this point. J. J. von
Tschudi asserts that these islands were torn from the main-
land as recently as the Spanish time by the earthquake of
1586,4 which caused great changes along the Peruvian coast.
Had this occurrence really taken place at that time, then it
would have been impossible for Francisco de Avila to tell the
myths of the province of Huarochiri in the year 1608, when
he speaks of the beautiful Cavillaca, who fled out to sea with
her child, where both were changed into rocks and were
visited by another divinity. The mother and child turned
to stone must be the larger and one of the smaller islands of
the Pachacamac group which can be seen from shore.
It is claimed by Rivera and Tschudi that stone founda-
tions of buildings may still be found on these islands," which
is denied by others; but even if this statement were founded
upon truth, it would be but a slim proof for the existence
of a connection with the mainland in the earlier period of the
history of this valley.
It is another question to determine whether the coast-
line is now the same as it was at the time of the Spanish
conquest, which seems doubtful. There is no tradition in
the valley of any great disastrous flood, still it is apparent
that the ocean with its heavy surf is constantly washing into
the shore. It is asserted that near the village of Lurin at
very low tide the foundations of ancient buildings are to be
seen on the shallow bottom of the sea. Señor Silva who is
thoroughly familiar with that region states, moreover, that
one of the islands which lies a little farther to the south,
could formerly at low tide be reached by mules dry-shod,
when hauling guano, which would be impossible to-day. In
early legal documents, dating from the forties of the seven-
teenth century, which are to be found in the National Library
at Lima and also in the deeds of the hacienda San Pedro, a
place by the name of Quilcay is mentioned, which no longer
exists in the valley and there is a legend that it has sunk to
the bottom of the sea.
The coastline near the ancient town has undoubtedly
5
7
Santillan, p. 32, has 'Pachahccamahc.' Most authors (Cieza, Herrera, Zarate,
Miguel Estete, p. 123, etc.) wrote Pachacama, dropping the final guttural,
which is grammatically required. Secondary forms like Pacalcami, Hachacama,
Pacama, which occur in Estete's printed report, must be attributed to errors in
reading the author's manuscript. The editor of the Relacion de la Religion de
los Indios del Peru, in the Col. de Doc. ined. III, p. 10, by mistake changed the
correct spelling of Pachacamac in the author's MS. into Pachachaca.
According to the linguistic law of the Kechua, the accent lay originally
on the penult of the word Pachacamac. Now, probably through the influence
of the Spanish law of accentuation, the accent is on the ultimate and the word
is at present pronounced Pachacamac'. This accent, therefore, is wrong, both
historically and linguistically. In the form of Pachacamain early editions of
the works of Herrera and Zarate, the Spanish law of accent seems to have been
followed, yet the ultimate guttural which caused its application was suppressed.
For scientific reasons it seems best to introduce a phonetic spelling of all
ancient names, outside of names of places for which the Spanish spelling must
be accepted. A phonetic spelling-system, to be of lasting value, must be easily
read, therefore all modern designations have to be avoided. For this reason
the phonetic spelling which Hr. Middendorf follows in his grammatical works
on the Andine languages deserves the preference over that followed in the lin-
guistic works of J. J. v. Tschudi, Pacheco, Zegara (Ollantai, Paris, 1878), C. F.
Beltram, Ludovico Bertonio and others. The Middendorf system, however, needs
some correction. He spells the palatal k sound before a, o and u with c; before
e and i in the Spanish way with qu; the deep guttural with k. From reasons
better discussed elsewhere, the higher k sound will always be rendered by k
(k, kh, k'), the deeper by q (q, qh, q'); aspiration by h, the short mute (Mid-
dendorf 'c- Bertonio cc, kk, pp, tt) by — (k', q', t, p'). Jsounds here like
ch in German, similar to the German ach ; ll as in Spanish ; w like the English
w. For the Spanish and English ch it would be well to introduce the familiar
sign ě (also series: č, čh, č) if types were available. It must be admitted that
the introduction of this system of spelling for designations not geographical
involves an apparent inconsistency in writing the name Pachacamac in its
different meanings. In its geographical sense, as the name of the ancient ruins
and the modern village, it must be spelled according to the accepted Spanish
form; as name of the divinity it is phonetically Pachakamaj (corresponding to
this is the phonetic spelling in J. J. v. Tschudi, Beiträge, p. 121). The admis-
sion of this inconsistency seems to be of less harm than would be the avoidance
of the phonetic spelling throughout.
V
v
1 The relative distances of the ruins and of the village of Lurin from the
river and that of the ruins from the sea are not quite correctly marked on the
small map which accompanies the English translation of F. de Avila's Report
on the Errors, etc., of Huarochiri, ed. by Markham.
2 Compare panorama of the town, pl. 1.
3 In the plan of the town, W, X, Y, Z.
4 Peru, p. 245.
5 P., p. 129.
6 Cf. pl. 15, fig. 1, and pl. 16, fig. 3.
?P. 291. Waitz, Anthrop. der Naturw., N, p. 235, also erroneously attri-
butes to Wilkes the statement that there are still some ruins to be found on
these islands.
8 If any destructive flood like those which occurred at Callao and in the
Lima valley had visited this coast within the last two centuries, it would be
impossible for the ruins of Mamacona to have escaped destruction (see plan), the
base of which is only twenty-two feet above sea-level and the terraced chamber
not more than thirty-three feet.
(2)

changed since the time before the Spaniards arrived, for the
fields which extend between the town and the beach are
sometimes inundated by heavy tides. Midway between these
and the ocean, about 100 yards from it, the ruins of a strong
dike or some primitive masonry may be seen, which date from
the period of the Incas, as will be shown later. If this shore
had been subject to occasional high tides at that early period,
this dike would probably not have been built. Heavy spring
tides are also washing away pieces of the bluff with which the
desert, the Tablada de Lurin, abuts toward the coast, which
at that point is only from two to three hundred yards wide.
If Estete's account of Hernando Pizarro's march from
Cajamarca to Pachacamac may be credited, it must be
assumed that at the time of his visit the connection between
the two valleys, that of Lima and that of Pachacamac, was
still unbroken. At the present day they are separated by a
barren strip of land, which is three miles long at the nearest
point. Miguel Estete writes that H. Pizarro proceeded from
the Lima valley to the town of the god, “without ever leav-
ing the cool shade of groves and villages.” However inaccu-
rate this statement may be, there remains enough ground for
supposition that since the time of the Incas the coastline
has receded farther inland, although not the entire distance
from the rocky islands.
(3)

CHAPTER II.
ETHNICAL AND POLITICAL CONDITIONS OF THE DISTRICT.
IT
T may be assumed that the geographical, climatic, ethnical
and political limits between the coast-district and the
Highland generally corresponded. In the geographical, cli-
matic and political sense this was certainly the case. The
province of Yauyos of Spanish time, which is the adjacent
district in the mountains, must have existed, according to all
sources, as a distinct conception at the earlier period. In
Spanish time it extended down the slope of the mountain toº
the hot coastland, the Yuncas, and thus had its limits at
the climatic dividing line;" and since former boundaries are
usually accepted by the conquerors, it is almost certain that
the older province of Yauyos had the same boundaries. It
must, therefore, have included Sisicaya on the Pachacamac
river and Chosica, with its altitude of 2800 feet as well as the
neighboring Chechima, although the former lies but thirty-
two miles and the latter forty miles from the coast. From
this it may be seen that the political territory of Pachacamac
was as small as its climatic district.
It is likely that the ethnical distinction was quite as
sharply defined. Exhaustive anthropological investigations
of this district have not been made so far, nor would it be
advisable to select the neighborhood of the village of Lurin
for studies of this kind, as the population of the district is
a very mixed one. The mestizos of the modern village of
Pachacamac, who live further inland, may still possess some
of the old blood of the ancient inhabitants of the town of
pre-Spanish times. Among these people a type of face pre-
vails, with a long narrow chin and a somewhat aquiline nose,
similar to the type on the Chimu vase shown by Squier, and
which is still seen in modern Pachacamac.
It is not quite clear which language was spoken by the
inhabitants of this valley, but it seems a mistake to classify
it in the same linguistic group with the Chimu or their kin-
dred, the Mochica, who lived southward from them, as was
done by J. J. v. Tschudi. They more properly may be
said to belong to the Tallanes, who occupied a district between
the Chimu and the Mochica. P. Bernabé Cobo distinctly
says that the Chimu was spoken as far as Caraguaillo in the
Chillon valley, while the language spoken in the valleys of
the Rimac and of Pachacamac was different, and by some
individuals the two languages were still used as late as 1630.
Calancha also confirms the existence of a linguistic line of
division in the neighborhood of Lima. The Mochica, living
in the valleys of Cañete and Chincha, were the nearest neigh-
bors of the inhabitants of Pachacamac to the south ; so there
seems to be no doubt of the identity of the people of the
Rimac and Pachačamac valleys with the Tallanes of Zarate.
It was not possible, so far, to classify the language of the Tal-
lanes, and it will need thorough local investigations to do so.
It might be assumed as being of the same linguistic group
as the tongue of the Chimu and the Mochica, since the Tal-
lanes lived between the two, still, there is no proof for this
supposition as no traces of the early language are to be found
at present. The older names of places are mostly forgotten,
which is different in the highland. Names of persons in the
church registers might furnish one of the best sources, but
studies of this nature require more time than may well be
spared from the work of surveying and exploring of ruins.
A few stray words only of the Kechua are still used by the
people, although this tongue was generally spoken in all the
provinces during the Inca period as well as in early Spanish
times; but not the least relationship can be traced in these
words with the ancient language of the adjoining province of
Yauyos, the Aimará, probably the basis of the Kauki.?
The political conditions are not as complicated as the lin-
guistic. Aceording to Garcilaso, the valleys of Pachacamac,
Lima, Chancay, Huacho, Supe and Huaman (Barranca), dur-
ing the period preceding the conquest of this territory by the
Incas, were ruled by a chief who had his seat at Pachaca-
mac, and who, like the Inca at Cuzco,10 had the right to enter
the temple of the chief deity, although not in the character
of a high priest. 11 The dominion of this ruler of Pachaca-
mac (if Garcilaso is correct and not Calancha, who gives the
three northern valleys to the Chimu), may be called quite
extensive, as it is 120 miles long. But it must be remem-
bered that this entire district comprised only a few narrow
valleys between broad deserts. Also, if Garcilaso is right,
the connection between the different parts of the realm was
a very loose one for mere geographical reasons, even disre-
garding the very loose nature of ancient political systems.
One fact points to closer political relations between the four
9
12
5
1 Cf. the map in Descripcion de la Provincia de Yauyos, Rel. Geogr., I,
p. 61.
? The language of the province of Yauyos was, according to v. Tschudi,
Beitr., p. 74, the Kauki, which he thinks an independent language now extinct.
Other authors find in it relationship to the Aimará, such as Brinton, The Ameri-
can Race, p. 216, although here it is mentioned among the dialects of the Kechua
families; and Middendorf, Die Aimará Sprache, p. 15. Numerous names of
places indicate close kinship with the Aimará, such as Ayaviri, Chosica, Omas
(quoted by Middendorf), Huancané (in the district of the Mala river), Huaqui
(Desc. de la Prov. de Yauyos, p. 66), Tanqui, Huanavi, Calahuaya ; names of
places ending in co, like Surco, Chillaco (cf. Taraco, Tiahuanaco and many
others); Parará (= millstone, see Bertonio, II, p. 250), name of a ruin near Cho-
sica with a great many millstones near it (Hutchinson, II, p. 47).
8 Book VI, chap. 30.
9 Calancha states that the chiefs of the side valleys were dependent of the
priests of Pachacamac.
10 Garcilaso, VI, chap. 31.
11 Jerez, p. 115.
12 III, chap. 1.
2 See below.
3 P. 184.
*Organismus der Khetschua Sprache, p. 85; cf, also Brinton, The Ameri-
can Race, p. 225.
6 Zarate, I, chap. 6. Also Cieza, I, chap. 61, speaks of “tres o cuatro lina-
jes de generaciones,” in which the Yunkas were divided
6 Extracts of his MS. work : Fundacion de Lima, by Jiminez de la Espada,
Rel. Geogr., I, App. num. I, p. xxviii.
(4)

2
10
northern valleys and the adjacent Chimu country, that is :
the very peculiar language of this district which was spoken
as far as Lima.' The sanctuary of Pachacamac must have
enjoyed a certain political neutrality, according to Midden-
dorf, who states that the Chimu had never touched it. Even
if this assertion is founded upon an ancient source, it may
be
erroneous, since the views on this subject varied greatly from
the earliest times. Garcilaso tells of wars which the rulers of
Pachacamac carried on with the Chimu previous to the Inca
period and of defeats suffered by them. Various other dis-
tricts of the realm were at strife with their neighbors. So
the Descripcion de la Provincia de Yauyos relates that the
inhabitants of the Lima valley had pushed forward as far as
the Pariacaca mountains, and there had worshiped an idol';
but there is no record that the people of Pachacamac were
connected with this campaign.
The supremacy of the ruler of Pachacamac over the
Lima valley still continued during the Inca period, since
Francisco Pizarro, when he founded the city of Lima, enume-
rated the Rimac valley as a part of the Pachacamac district."
Only the names of four of these rulers are known at present:
Kuismanqu, who surrendered to the Incase and whose name
appears to be of foreign origin; Taurichumbi, who ruled
Pachacamac at the time of Hernando Pizarro's advent, and
whose name sounds Kechuan ;8 Saba, baptized by the Augus-
tinian friars in the fifties, and Luyan, who soon succeeded
him when the former did not prove subservient enough to
the new masters.
The valley was very populous when the Spaniards came,
a fact attested by numerous ruins and burial-grounds. A
large number of ruined buildings surround the rocky head-
land of ‘La Centinela. Opposite the valley, in the Rincon-
'
ada, on the sandy plain, are the ruins of large courts and
halls, and across from Lurin, on a barren hill, extensive
remains of buildings can be seen. Even on the brow of
La Centinela, 130 feet above the level of the valley, may
be traced the foundations of a large building surrounded by
a number of graves. Lines of masonry may be found all
over the valley, which by the size of the adobe, “tapia,' the
kind made on the spot, betray their pre-Spanish origin. A A
long narrow strip of ground belonging to the hacienda
Huarangal, a part of San Pedro, is covered with graves
bordering closely against the eastern side of the Centinela
rock. Some ancient graves on the Venturosa hacienda
show that the northern edge of the valley above Pachaca-
mac also had its settlements. At a distance of from nine to
twenty-three miles from the sea, on the right-hand edge of
the valley, may be seen the ruins of four villages, and on
the left the débris of two smaller ones, and in a narrow
side-valley the ruins of the larger village, Huaycan, which
stretch along a stony ravine (Kechua 'waiq'o ') and for
about half a mile up the side of the hill. The two small
villages were built against the hillside, which in places is
quite steep. This was done to utilize every spot of the arable
ground for the raising of crops, since the valley at this
place affords only a very narrow strip of fertile soil. The
walls of these houses were built of stone, with adobe used as
mortar, and tapered toward the top; owing to the dry climate
they are in an almost perfect state of preservation.
This district possessed a number of good roads, one of
which skirted the coast l' in the early days. When Hernando
Pizarro failed to obtain any information concerning the road
to Pachacamac from the people of the coastland (which they
were forbidden to give under penalty of death), he discovered
soon that he had only to follow the southern road along the
coast, from Paramonga, 10° 45' S. lat., and that the town
must be situated on or near it.11 Other roads followed the
northern or the southern slope of the valley up the hill to
Cienequillas. A bridle-path, now in a neglected state, but
still known as the 'camino del Inca,' skirted the southern
declivity and connected the ancient settlements along this line,
one of which still bears the name of Tambuinqa.
Contrary to views held by others, the writer thinks that
names which connect localities with Inca institutions were
usually proved to be of such an origin after some archæo-
logical investigations. This path was probably laid out by
the Incas, and a “tambo,' a wayside refuge, was located on
this spot. At Cienequillas the road from Lima strikes the
road, which continues through the valley as far as Huaro-
chiri; and in order to make a further connection between this
part of the province and the highland, after the latter had
been added to the realm, Tupak Inca had a bridle-path con-
structed from Huarochiri across the Pariacaca chain, one of
the highest and wildest mountain ranges of Peru, 16,000'
high, and extended it as far as Jauja. J. J. v. Tschudi 12 calls
it a miserable road, while Cieza 13 thought it well worth seeing
for the grandeur of the scenery in those regions.
1 According to Calancha.
2 Die Chimu Sprache, p. 21.
3 Garcilaso, VI, chap. 32.
4P. 71.
5 Order of Pizarro, issued at Pachacamac under date of January 8, 1535,
extracts of the MS. work of the Padre Cobo, p. xi; cf. also Calancha, II, chap.
29: “anbos valles de un mesmo cazique."
6 Garcilaso, VI, chap. 30.
7 Chumpi, in Kechua girdle, tarhui= lupine. Chumpi in some other
combinations: Ahuachumbi and Ninachumbi, two islands under 14° southern
lat. (letter of Pedro Sarmiento de Gamba, reprinted from M. Jiminez de la
Espada, Tres Relac. de Antig. Introduction, p. xxiv), Suculacumbi (probably for
Suculachumbi), name of Chancay (or of the cazique of that place; see below),
Estete, p. 130. Tauri suggests : Tauriqui, title of a cazique in northern Peru
(Sarmiento, l. c., p. xxv).
8 Estete, p. 133.
9 Wiener, p. 61, gives a vague plan of a ruin, said to be near Manchay ;
possibly he is referring to the ruins of the Rinconada.
10 Traces of ancient roads, which may be known by their enclosures of
tapia walls, can still be seen in the Lima valley and deserve to be more closely
studied. Quite well preserved sections of roads are found along the northern coast
of Peru, in the Santa valley and near Casma. The road there has a dividing
wall in the centre (Hernando Pizarro, p. 121, tells of bridges so divided), one
side being reserved for the ruler and his retinue, the other for the common peo-
ple, each side is about sixteen feet wide. It might be of interest to trace the
entire course of this road.
11 Letter of Hernando Pizarro, p. 122.
12 y. Tschudi, Beiträge, p. 75; p. 119.
13 Cronica, II, chap. 58.
(5)

CHAPTER III.
RECENT HISTORY OF THE VALLEY.
THE modern history of the valley begins with the journey
of Hernando Pizarro from Cajamarca to Pachacamac,
where he was ordered by his brother Francisco to capture
the temple treasure, early in the year 1533. From his own
report, he started from Andamarca, circa thirty-two miles to
the south of Cajamarca, about January 9, Julian cal., and
arriving at Pachacamac on January 30, he must have traveled
over the distance of 340 miles in twenty-two days, a daily
average of about sixteen miles only. Considering the swift-
ness of Indian messengers, it is small wonder that Pizarro
on his arrival discovered that all the gold treasure had dis-
appeared. Including the contributions of gold and silver
sent by the chieftains of Mala, Huarcu (Cañete), Chincha and
others, his booty amounted only to 85,000 pesos of gold and
3000 marks of silver, 90,000 pesos altogether, according to
Miguel Estete.
If it is considered that the ransom which was amassed
at Cajamarca for the delivery of Atahualpa amounted to
more than 1,326,000 pesos of gold, the booty of Pachacamac
seems quite insignificant, being only one-fifteenth of the
above amount.
4
At Pachacamac, Hernando Pizarro forced an entrance
into the temple of Pachakamaj,º and first of all demolished
the idol and the shrine in the presence of all the people. A
cross was raised upon the débris to symbolize the beginning
of a new era. From this day the decline of the city began,
as its prosperity had mainly depended upon the prestige of
the sanctuary.
Toward the close of the year 1534, Francisco Pizarro
reached the town by way of Jauja after having taken Cuzco.
He held wild orgies º here with Diego de Alvarado and made
a treaty with him, according to which the latter waived all
future claims in the Peruvian conquest. From this place
Francisco Pizarro decreed, in January, 1535, the founding of
a new capital, Lima, in the Rimac valley, a former plan of
locating the capital in Jauja having been abandoned for
climatic and economic reasons, while Pachacamac could not
be taken into consideration for its want of a harbor.10
The Spaniards, on their arrival, found the city and the
whole valley well populated,11 and it is described as a desolate
pile of ruins, as it appears to-day,"? in the first part of the
seventeenth century, with only a few Indians remaining in the
valley. The depopulation of the city was hastened by the
wars which ravaged Peru on the one hand and by the conver-
sion of the population through the religious orders on the
other, Franciscan and Augustinian friars and Jesuits appear-
ing on the scene in succession. The beginning in this process
of depopulation was made by the wars of Manco Inca, who,
in the year 1535, gathered about him all the Indian forces for
the death-struggle against the Spanish invasion; the next
step was, that the Franciscan monks established a new settle-
ment near the bridge in the valley and drew there the inhabi-
5
12
<
1 P. 121. Estete gives the 14th of January as the day of starting. He incor-
rectly fixes the 9th of January for the departure from Huaman mayo (Barranca).
The same day is given by Raimondi, II, p. 41, while apparently it was the 24th.
2 The writer saw in La Paz, Bolivia, an old Indian who had once covered
the distance of 190 miles from La Paz to Tacna in three days.
3 Letter of Hernando Pizarro, p. 123 ; Miguel Estete, p. 134.
* The wealth of the Incas was without doubt grossly exaggerated in early
accounts. Cieza, Cronica, II, chap. 65, tells of an offering of more than 100
'arrobas' of gold which one of the Incas had given to the temple in Pachacamac.
100 arrobas would have had a value in refined gold of $75,000,000; or in a strong
alloy of half the weight only in pure gold, it would have represented a value of
$38,000,000, which would be ten times the amount for Pachacamac alone more
than the ransom of Atahualpa, to which the entire realm of the Inca at Cayamarca
had contributed ; 85,000 or 90,000 castellanos or pesos de oro, according to Pres-
cott, I, p. 305, amounted to between $261,000 and $276,000, or if the gold value
of the dollar is estimated at 1.5406 grammes, this would have had a value of 402
to 416 pounds of fine gold ; 3000 marks of silver, a mark = 233 grammes, would
have been worth 1550 pounds of silver, and by the former standard of value
would be equal to $30,000, the present commercial value would be $1,116,000.
For the transportation of this booty llamas were used, each of which can carry a
load of 100 pounds. If the load of these animals depended merely on the weight,
which was the case with gold in the form of gold-dust, twenty animals would
have been sufficient. It is stated by Xerez, p. 118, that the spoils of Pacha-
camac consisted of twenty-seven loads of gold and 2000 marks of silver. Accept-
ing this statement as correct, each llama carried $10,000 or thirty-three pounds
in fine gold, or, if it was alloyed, about two arrobas.
a pight's march, and the camp of which Estete speaks was only that of the
next day, the 29th of January. Night marches through the desert are fre-
quent during the hot season on the Peruvian coast, and there was another
desert, twenty-eight miles wide, between Chancay and the Chillon valley. The
inhabitants of Chancay supplied the travelers with provisions for a night's
march, “dieron todo lo necesario de comida para aquel dia.” When, there-
fore, Pizarro left Chancay on the 28th in the evening, intending to reach Pacha-
camac the next day, “por allegar otro dio al pueblo donde estaba lax mezquita,'
he meant the day following the next night's quarters, the 29th of January. On
the 30th Pizarro made a short march only. But since he dined midway in an
Indian village, and the road passed through groves of trees and villages, the last
night's camp must have been within the Rimac valley, probably in the old
village of Huatica, near the ancient Rimac temple; and the interruption of the
march for a repast must have been at the ancient Surco, the present ruins of
Armatambo, near Chorillos, on the southern extremity of the Rimac valley. No
matter how this rather obscure passage of Estete's is read, the length of the
march on the last day, when Pizarro arrived at Pachacamac, cannot have been
more than seven miles.
6 Prescott, I, p. 292.
7 Estete, p. 133.
8 Prescott, p. 293.
9 Garcilaso, pt. II, book II, chap. 17; Prescott, II, p. 13.
6 Raimondi, Peru, II, p. 29, carefully traces the march of Hernando
Pizarro from Andamarca to Pachacainac, but without solving the difficulties in
the description of the latter part of the road. Assuming the 30th of January as
the day of the arrival in Pachacamac, Pizarro would have reached Supe on the
25th, Huaura on the 26th, while on the 27th he would only have traveled two
leagues to Huachu. Here he spent the rest of the day carousing with the
Indians. As the next day's march through a broad desert to Chancay would
have been too long for that day, forty miles, Estete, through some mistake,
gives only five leagues as the distance. Pizarro, therefore, started early in the
morning of the 28th of January and arrived at Chancay in the afternoon. This
place is unmistakable from Estete's description, although he gives a different
name (Suculacumbi). It is possible that this was the name of the chieftain of
that locality, which is frequently given instead of the name of the place. From
Estete's account it is clear that Pizarro and his men only stopped at Chancay
for supper and continued their way to the south immediately after. Here
Raimondi's misconstructions begin, p. 42. He cannot understand that the follow-
ing night's camp of Pizarro's troop was in the Rimac valley, while the distance
to that point was thirty-seven miles. It is evident that Pizarro undertook
10 Extracts of P. Bernabé Cobo's MS. work, Fundacion de Lima in Relac.
Geogr., I, App., pp. vi-xv. A foundation of the new city at Pachacamac, toward
the end of the year 1534, before the present site of Lima was chosen, of which
Calancha writes, II, chap. 29, probably never occurred.
11 Garcilaso, pt. II, book 2, chap. 17; Calancha, l. c.; Cobo, Historia, IV,
chap. 17.
12 Calancha and Cobo, l. c.
8
(6)

2
cona, and therefore convenient for the use of the hacienda.
Alterations of this kind were not found in any of the other
buildings of the old town, and only a few insignificant finds in-
dicate that the city was still inhabited when the Spaniards en-
tered it. A sickle-shaped knife of rude workmanship, sharpened
3
Fig. 1. Copper knife, Spanish period. From an ancient house at Pachacamac.
tants of the old town. In 1553, Francisco Hernandez Giron,
during his rebellion against the Spanish régime, sought
refuge at Pachacamac, and a garrison was quartered there
by the Government to keep him in check. These circum-
stances were considerable factors in the decrease of the popu-
lation of the town, so that at last only a few families
remained. When in the early fifties of the sixteenth cen-
tury the Augustinian monks arrived, they transferred the
town to its next site in the valley, abandoning the old place
entirely. Thus in the Descripcion de la Provincia de
Yauyos, 1586, it is described as being situated on the val-
ley road upon the left bank of the river; but here the city
was not allowed to remain either. When the Augustinian
friars left the valley, their successors changed the site of
the town once more to a point several miles inland, where
it stands at the present day. This succession of changes
interrupted the connection with the original place.
A remnant of an Indian tribe ruled by a cazique and
slightly differing from the surrounding people in appearance
as well as in customs, seems to have led a separate existence
in this valley almost down to our own time, from informa-
tion gathered in that neighborhood; but no trace of these
people can be found at present.
It is small wonder that so few indications of Spanish
influence can be descried in the old town, if the rapidity of
its depopulation is considered as well as that of its Christiani-
zation. A few changes of more recent date may be traced in
the buildings of the ancient convent, due to the fact that the
buildings were near the fields of the former hacienda Mama-
on its outer edge (fig. 1), found in a house during our explora-
tions, may be regarded as dating from the early Spanish time.
A few sea-green glass beads in a jar of the Inca period were
found in the ruins of a house of the outer city, and a cylindri-
cal blue glass bead of older appearance was picked up in the
inner town. These were the only signs noticed by the writer
that the city was inhabited during early Spanish times.
One of the more important settlements in the valley is
San Pedro, a Jesuit foundation. The ruins of a chapel
dating from that period are still in existence. It is better
not to mention here the village of Lurin in any historical
connection with the other places, since it is barely 100 years
old. It was founded as a new station on the “camino real,'
after the transfer of the old city of Pachacamac to its last
site farther up the valley. The name Lurin is of Kechua
origin, a language which was introduced here by the Incas and
gained a complete dominion in this locality during the early
Spanish period.
Calancha, 1. C.
2 Col. de Doc, ined. III, p. 11.
3 Calancha, i.c.
4 Calancha
5 Middendorf, Die Chimu Sprache, p. 20.
6 Lurin, or hurin = below, lowland. Comp. also the ancient classification
of the village population in the highland into two groups, as it was made by the
Incas, in Hanansaya and Hurinsaya.
)
(7)

CHAPTER IV.
FORMER EXPLORATIONS OF THE RUINS.
THE ruins of Pachacamae have been more frequently
visited and described by travelers than any of the
other monuments of ancient Peru, owing to their convenient
situation so near the capital, yet few ruins of the importance
of Pachacamac have been described so unsatisfactorily. Vari-
ous reasons may be given for this neglect. The majority of
the expeditions were undertaken for sightseeing rather than
for serious scientific exploration, and the time spent on the
ground was usually very limited. The ruins taken as a
whole are more attractive than a study of the details would
promise, for they seem more insignificant here than in many
other ruins; the larger buildings are very much decayed
and but few fairly well preserved ones remain. Prior to
the expedition here reported only one is known to have
made a stay here of two months, that of Bandelier in the year
1892 or 1893. This explorer took measurements of the more
important buildings and determined the purposes for which
they were constructed. He explored the eastern section of
the vast cemetery which surrounds the temple of Pachaka-
maj.' The finds of these explorations are at present in the
American Museum of Natural History at New York, and it
is to be regretted that the report of Bandelier was not pub-
lished; only a few notes on his discoveries in Peru are con-
tained in an article of F. Webb Hodge.?
The earlier descriptions of the ruins vary in character
and merit according to the time when they were written;
those of the sixteenth century claim the value of documentary
evidence, coming from the pen of contemporaries who saw the
city in its glory, or who at least had heard of its wonders
from the Indians who lived there in the old days. The
reports of Miguel Estete and of Hernando Pizarro are of this
character; they were the first Spaniards who entered the
old town and destroyed the famous temple. Their reports,
therefore, are those of eye-witnesses. Estete wrote two
accounts, one of which in diary form was printed in Xerez
Relacion de la Conquista, and with it was reprinted many
times; the other copy is known from a few stray notes of
William H. Prescott, who had the manuscript before him
when writing his work.
Estete's report is one of the most valuable documents
ever published on any ancient Peruvian town. Similarly
important is the letter of Hernando Pizarro to the Royal
Audience of Domingo, wherein this expedition to Pachaca-
mac is described. The letter having been written ten months
after the event, the description is more condensed in form, yet
gives the impression that the writer had a clear remembrance
of all the events. We owe to Markham the publication in
English of this important document.
We possess descriptions dating from the sixteenth cen-
tury by Pedro de Cieza, the great traveler and faithful
chronicler, and by P. Hieronimus Roman. Cieza saw the
town about one decade after the extinction of its religious life,
when the inhabitants were converted to Christianity or, if
not, they were at least quite intimidated by the invaders.
We find in the work of P. Hieronimus Roman, Las Repub-
licas del Mundo,4 an attractive description of the sun temples
in the empire of the Incas, apparently founded upon a close
study of the sun temple of Pachacamac, which he mentions
in his account; he deserves the more to be credited as he
gathered his information prior to 1595, the year of publication
of his work, and from eye-witnesses who inhabited the town
before its decline.
From the first quarter of the seventeenth century date
the accounts of the two distinguished chroniclers, Bernabé
Cobo and Antonio de Calancha; both knew the ruins well.
The former, a Jesuit, gives an admirable description of the
temple on the hill, which has the additional value of minutely
describing important details, then existing but now lost.
Calancha, an Augustinian friar, gives a splendid résumé
of all the facts he could gather concerning the temple and
its history. Unfortunately, when these men wrote their
accounts, all original knowledge concerning the former use of
the temple had died out; their information was from hearsay
and local tradition, which in the course of a few decades
distorted the facts; therefore their statements cannot be
credited without reserve, in many cases they are entirely
false. At the time when these men lived in this region, a
vague local tradition, entirely unsupported by facts, had
already begun to take the shape in which we find it at this
day and whereby even in our nineteenth century many of the
explorers were deceived. These false and misleading tales
must therefore be entirely disposed of in order to establish a
strictly historical account of the former conditions of the town.
The notes on these ruins, written by the two Señ. Ulloa,
early in the eighteenth century, reveal only their ignorance
and indifference concerning the history of the ancient temple.
The only fact we get from them is that the appearance of the
ruined city was just as desolate at the time of their writing as
it is now.
It is true that early in the nineteenth century a
scientific interest in the ruins was awakened, but without lead-
ing to any immediate results. The ruins since then were
visited and described by travelers as follows: In 1817 by
W. B. Stevenson; on June 28, 1839, by Wilkes (of the
U. S. Exploring expedition); by J. J. v. Tschudi, between
1838 and 1842; E. M. de Rivero and Clements R. Mark-
a
1 See plan of the city.
2 American Anthropologist, 1897, p. 393ff.
3 Crónica, I, chap. 72.
4 Quoted by Garcilaso, book II, chap. 2.
(8)

2
ham saw the ruins in 1853, 1854 and 1860; George Squier,
in the sixties; Hutchinson, between 1871 and 1873; Bastian,
July, 1875; Wiener, 1876; Middendorf, in the eighties, etc.
In the report of W. B. Stevenson we find no description
of the ruins; he only gives a few general observations.
Wilkes could, after his visit of a few hours at Pachacamac,
not give more than a few superficial remarks and two
indifferent views. The notes of J. J. v. Tschudi are also
too short to be of any real value. The description of the
ruins in the work of E. M. de Rivero and J. J. de Tschudi? is
quite uncritical, although often quoted; nor are the plan and
views of the town in the supplementary atlas to be relied
upon.
Markham often mentions the ruins in his works; he
also prepared a ground plan of the sun temple, which unfor-
tunately was not published. George Squier visited this
locality for a week 36 but he gave only a very general descrip-
tion of the place; still his plans of the two principal monu-
ments, the sun temple and the convent, are the best so far
published and the only ones worthy of any scientific notice,
although they need a few corrections. Hutchinson's notes?
on the ruins give few details and are unreliable; the views
are only partly true to nature. Bastian, in publishing his
notes on the ruins, merely wished to arouse an interest in
them, and this present report may be considered as an attempt
to meet his wishes. Wiener's' report on the ruins, together
with the views, is of no value, although his observations were
based on a visit of eleven days. Middendorf's 10 description of
the ruins has the advantage over the notes of most of his
predecessors in the great seriousness of observation and the
careful attention to detail shown by him. The views given
are reproductions from photographs, the first of the kind in
this field.
That the present report can be offered in much greater
detail than any of the foregoing, the writer owes mainly to the
fortunate circumstance of having been able to spend at this
one place a much longer period than former explorers, and
devoting his time exclusively to the exploration of this one
5
group of ruins.
1 Wilkes, I, p. 278. On January 26, d’Osery passed the ruins of Pacha-
camac. In the book of travels of F. Castelnau, Expéd. dans le part. centr. de
l'Amérique du Sud, Histoire du Voyage, Paris, 1851, IV, p. 179, there is only a
short allusion made to this visit.
2 J. J. v. Tschudi, Peru, I, pp. 289-291.
3 Rivero y Tschudi, Peruvian Antiquities, pp. 288-291.
4 Atlas, pl. LIV-LV.
Reports of the Discovery, Introd., p. xxi.
6 Squier, Peru, pp. 62-81.
7 Hutchinson, Two Years in Peru, I, pp. 147-176.
8 Bastian, Die Culturländer des Alten Amerika, I, pp. 51-53.
9 Wiener, Pérou et Bolivie, pp. 56–70.
10 Peru, II, pp. 110–126.
2
(9)

CHAPTER V.
THE RUINS.
GENERAL TOPOGRAPHY OF THE CITY.
FOUR hills may be noticed in the city plan, which partly
measures two miles and a half in length to one mile and a
third in width ; a few more hills were included in the north-
eastern section.
The structure which crowns the hill, W, in the centre of
the town is the Sun temple, erected by the Incas after their
conquest of this valley. The original sanctuary of the town,
,
however, was the temple of Pachakamaj, the ruins of which
lie at the foot of the hill, on the northern side, and on the
level of the city. Pl. 2, fig. 1, shows both these monuments
in a perspective view, the latter surrounded by a vast ceme-
tery, No. I of the plan.
Although the designation given here of these two build-
ings is quite beyond doubt, it has rarely been attempted. It
has become the custom to take the monument on the hill for
the temple of Pachakamaj, and no attention was paid to the
temple below or it was taken for the sun temple, as for in-
stance by Rivero and Tschudi.'
encircle the town or are covered by it, W, X, Y, Z.
The one at the extreme south, W, is the highest, about 250
feet above the level of the valley, and the most extensive of
all;" its southern side projects into the valley like an out-
post; surrounded by fields on three sides, it is on the
fourth connected with a strip of land fifty to eighty feet
high, upon which lies the ruined city, and on account of its
commanding position it may be seen from great distances with
the structures on its crest. From the northwest it is to be
seen from the coast at a distance of six miles; from the inner
valley at a range of four miles; from the south coast at a
much greater distance. The second hill, X, and the third, Y,
confine the city on the eastern side toward the valley. While
the latter is insignificant, the former? extends in a long ridge
in a southeastern direction into the valley and turns the river
away from the southern part of the town from 600 to 800 feet
out of its course.
The northeastern side of X slopes abruptly toward the
valley. The summit in its northeastern section is broad and
covered with innumerable craterlike hollows, which show
that the stone was quarried here to be used in some of the
buildings in the town. This stone is a sort of argillaceous
slate, the fragments generally having a parallelepiped form.3
Along the hillside toward the city traces of former terraces may
be seen encircling the hill and following its curves. On the
southwestern slope4 small stall-like cells were found, three feet
high and built of stones loosely laid ; they were in groups and
enclosed human remains, and in a few instances some broken
pottery of a very crude kind. These burial-places do not
resemble any others in the town, as they are far inferior to the
degree of culture shown in the antiquities of the city in gen-
eral, but it was impossible to determine their place in the
cultural history of the town since no objects were found in
them which could serve as guides.
The fourth hill, Z, stands in the centre of the town and
is covered with buildings. A narrow ridge branches off from
here in an eastern direction and divides the southeastern sec-
tion, B, from the northeastern, E. The city-plan shows the
inner town mainly, that section which is covered by high
adobe structures and which is a mile long from northeast to
southwest and two-thirds of a mile from northwest to south-
east. The entire area of the city within the outer walls
a
1 It may be seen very distinctly, pl. 2, fig. 1; also pl. 9, fig. 2, its western
declivity.
2 Pl. 9, fig. 1, as seen from the town.
Scherzer, Reise der Novara, III, p. 340, speaks of it as being cut in the
form of bricks.
6 Wilkes erroneously gives the area of the town as one-fifth by one-third
of a mile, which Brühl by a slip in writing changed to 800 + 500 feet (instead
of meters).
7 See plan.
8 By Middendorf and Markham, 1872, in the Introduc. to the Reports of
the Disc., p. xix.
9 Balboa wrote his work on the Inca Empire at Quito, between 1576 and
1586 ; it was based upon an account written by Cristóbal de Molina, who lived
at Cuzco, far from Pachacamac (see Markham, Narratives, Introd., p. vii);
neither of these authors had seen Pachacamac, so the latter simply repeats the
error of the former in stating that the temple on the hill, which had been
newly erected by the Incas, gave its name to the city. There may have been
certain relations between the cult of the Creator god Pachakamaj and that of the
Sun temple on the hill, as according to Betanzos, chap. 11, there were between
the cult of the Wiraqocha and the Sun temple at Cuzco; still it must be main-
tained from historical and other reasons that the temple on the hill W (see pl.
13 for ground plan) was the temple of the Sun, and the other at the foot of the
hill was the old sanctuary of the Creator god, named Pachakamaj from the time
of the Inca conquest. Fernando de Santillan, too, refers to the temple on the hill
as the Pachakamaj temple. His account of the organization of the Inca empire
dates from the time between 1651 and 1721 (Jimenez de la Espada, Introduc. to
Tres Relaciones de antiqu, p. xlii), it is therefore to be assumed that he followed
the inaccurate local tradition, which by that time had taken its present shape.
PP. Bernabé and Calancha (about 1620) speak of the temple on the hill distinctly
as the Pachakamaj temple. Their opinion does not bear much weight, however,
as it is easily seen in their statements that neither of them knew exactly which of
the buildings was Mamacona. The original meaning was quite unknown to P.
Calancha, so that he identified the name Mamacona with the temple on the hill
(plan, pl. 13). Cobo repeated the statements of Indians, that the temple on the
hill antedated the Inca conquest of the valley, which is impossible from inward
and outward reasons. The tradition that the upper temple was that of Pacha -
kamaj could probably arise from the fact that the lower, the true Pachakamaj
temple, was completely ruined. The worship of the Sun in the upper temple was
introduced by the conquerors, not a native cult, and later tradition connected
the cult of the ancient local deity with the imposing temple structure on the
hill. This error was adopted among modern authors by Stevenson, Archibald
Smith (Peru as it is, 1839, II, p. 308), Wilkes, de Castelnau, Rivero and Tschudi,
Prescott and Squier. J. J. v. Tschudi's opinion is that the temple was dedi-
cated to Pachakamaj and by the Incas turned into a Sun temple (Peru, I,
p. 290 ; also Beitr., p. 126), which contradicts Estete's clear statement that both
temples existed side by side. Rivero and Tschudi claim to have the testimony of
contemporaneous authors on their side, especially that of Hernando Pizarro.
How little this could refer to Estete and Cieza can be seen in the text above.
Pizarro speaks of two courts of the sanctuary of Pachakamaj, an upper and
a lower one; by the lower one he may have referred to the lower temple, by
upper to the one upon the hill. Still, since the cell in which the god spoke
to the priests in oracles was, according to Hernando Pizarro, “adjoining the
upper court,'' and not in the temple upon the hill, it can only have been in the
lower, the old temple; while the upper court could not have been any other
* Pl. 9, fig. 1.
5 Comp. Panorama, pl. 1, to the right; pl. 11, fig. 1, western view, etc.
3
(10)

a
1
The testimony of several authors of the sixteenth century
is sufficient for us, of men who saw the city at the time of its
discovery, like Estete, or who got their information from
persons acquainted with the city in its better days, as Cieza
and Padre Roman. Estete tells us with convincing clear-
ness: “The city has close to the mosque (i. l., the temple
of Pachakamaj) a house of the Sun, which stands upon a hill
and is well constructed with five terraces,” and further on:
“the image (of Pachakamaj) stood in a goodly house, which
was well painted.” 1 The only well painted house of the
city was the temple of Pachakamaj, the temple on the hill
with its monochrome coat of paint could never have been so
described.
Estete is corroborated by Cieza, who says: “The temple
of Pachakamaj was built upon an artificial mound of adobes.3
When the Incas realized that it would be difficult to sup-
plant the cult of Pachakamaj, they made the agreement with
the native caziques and the priests that in this temple of
Pachakamaj the ancient cult should be continued with the
condition that a larger temple for the Sun worship be built
upou the most prominent place. 124 Also Padre Hieronymo
Roman in his book refers to the upper temple as the temple
of the Sun.
To the grounds of the temple of Pachakamaj also
belonged the hill against which it was built, while the north-
eastern, northwestern and southwestern sides of the grounds
were occupied by the cemetery of the pilgrims.
In the plan of the city a line of ruined masonry may be
traced which encloses in a rectangle, and at an even distance
of several hundred feet, the temple and the cemetery. Within
its area stands one building to the northeast from the tem-
ple, which may have served some religious purpose, and the
Sun temple upon the hill. The ancient convent lies on the
outside of the older city wall opposite its western angle. The
ruins may be seen on the panorama, pl. 1, to the left in the
background. The name, Mamacona, whereby this pile of
ruins is still known, its secluded situation and its peculiar
style of architecture, that of the Incas, seem strong proofs for
the fact that this was the convent. Hernando Pizarro dis-
tinctly mentions the situation of the building in his letter:
“This city of the mosque is very large and has vast edifices
and courts. Outside of it is another large space enclosed
by walls with gates opening into the mosque. In this enclo-
sure are the houses of the women, who are the wives of the
devil.”
The ancient 'fields of the Sun,' 'inti chajra,' and in later
days the hacienda Mamacona, stretch toward the west, south
and southeast to the foot of the hill upon which the Sun tem-
ple stands. These are the fields from which the inmates of
the convent drew their sustenance. The plan shows that there
was but one gate which might correspond to that mentioned
by Hernando Pizarro, at the end of the western street; farther
south seem to be no indications of other gates in the walls.
Thus the entire enclosed area south of entrance A is secured
as the outer court of the temple of Pachakamaj.
Hernando Pizarro in his report continues: “Here are
also the storehouses where the treasure of gold is hidden.
There is none at the place where the women are kept.”
This seems to indicate that Pizarro thought the gold stores to
be outside the city and at a distance from the convent. The
buildings, the ruins of which lie to the southwest from the
temple precinct, enclosed by the latter, the Sun temple and the
hacienda Mamakona (the Inti chajra), must have had some
religious character on account of their isolated position, among
buildings of a religious nature only; here might be located
the ancient gold magazines, of which Pizarro º speaks, although
it seems unreasonable to seek them in these insignificant ruins.
It would be more plausible to locate them on the square piece
of ground to the north from the temple, which shows the rav-
ages of unceasing digging for treasure.
The town (see panorama, pl. 1, taken from the temple of
Pachakamaj) extended to the northeast, north and northwest
from the temple precinct. The oldest quarter of the town
.
presumably lay between the hills X, Y and Z, as this would
be a likely position, sheltered by the hills and near the river,
and our finds entirely confirmed this supposition; the built
up part of the town situated within the horseshoe shaped
inner city wall10 is a later addition. It is still possible to dis-
tinguish a street crossing which divided the town into quar-
ters.11 The northwestern and northeastern parts of the city
wall are similar to those which flank the northeastern and
northwestern streets. These walls, the western one of which
extends for 1000 yards in a straight line, seem to be parts of a
well planned scheme of enlarging the city, the same plan
was followed in the construction of the street crossings and of
these walls, all at one period and previous to the erection of
the buildings. This, however, happened at a later date than
the laying out of the temple precincts, for the streets adjacent
to these, the southeastern and the south western, are less
regular than the others.
For a second time the city was enlarged, this extension
taking in twice as much additional ground in the northwest
and northeast than before, but only 500 yards more toward
5
the sea.
It has been asserted that Pachacamac 12 at the time of the
Spanish conquest was partly deserted. Miguel Estete speaks
of ruined buildings and of the dilapidated condition of
most of the walls. Still it must not be forgotten that Her-
nando Pizarro and Miguel Estete in every line of their
accounts dwell on the size and importance of the town, and
that the latter mentions the ruined condition of some build-
ings and walls only as a proof for the antiquity of the
city. The dilapidated wall, of which Estete speaks, is the
inner, older one, as the outer city wall is, even to this day,
almost intact. All indications as well as the results of explo-
rations tend to prove that the new territory which was added
to the city at the time of its extension belongs exclusively to
the Inca period. Our view, therefore, that the city at the
time of the Spanish conquest was not a dilapidated place, but
still a flourishing town, seems to be well supported.
a
place than the plateau of the lower temple, and a different interpretation must
be found for the lower court. Markham (Cuzco, p. 15) changed his originally
correct view concerning the designation of the temple. Hutchinson's theories
on the purpose of the upper temple and the time of its destruction need not be
discussed here.
1 Estete, p. 133. “Este pueblo tiene junto á esta mezquita una casa del sol,
puesto en un cerro bien labrada con cinco cercas.” According to the way of
counting, the terraces of the Sun temple may be numbered as 3 or 6.
2 A few frescoed rooms, apparently chapels, cannot be mentioned for com-
parison.
3 In some respects this description fits the lower temple more exactly than
the upper, in others it applies to both.
4 Cronica, I, chap. 72.
5 See plan, pl. 2, fig. 1, and pl. 17, fig. 3.
6 See plan and pl. 2, fig. 1.
7 Pizarro speaks of "several houses,” probably he meant the entire group
with its several roofs.
8 At present a dependency of the hacienda San Pedro.
º Pizarro probably expected to find storehouses of gold near the convent,
since those buildings showed the Inca character in their architecture, and also
otherwise bore a strong resemblance to the convent at Cuzco.
10 This city wall has often been mistaken for the temple enclosure ; by
Rivero and Tschudi, p. 288 (and by others who quoted their statements, Brühl,
p. 140, Bancroft, IV, p. 796, etc.).
11 This division resembles that of the ancient city of Cuzco, where a street-
cross divides the place into four quarters, which were a microcosmic repetition
of the realm of Jawantinsuyu (Garcilaso, book VII, chaps. 8, 9), and it is possible
that the division of the city of Pachacamac had a similar significance.
12 Bandelier, in an article by F. Webb Hodge, American Anthrop., 1897,
p. 303.
(11)

CHAPTER VI.
GENERAL SURVEY OF THE GRAVES IN THE CITY.
QUITI
2
a
5
UITE in proportion to the great antiquity of the city, in
which all authors concur, is the enormous number of its
graves. Skulls and bones are scattered all over the ground,
as they were dug up and thrown aside. The graves may be
divided into three classes : those in open cemeteries, those in
ancient dwellings 2 and those inside of temples ; the first class
are the most numerous, of the third there are but few. In
the oldest section of the city, the southern, the number of
graves is of course the largest. This part of the town, proba-
bly, passed through a long history before its enlargement. In
the newer part of the town, there are likewise a large number
of burial places scattered here and there under the desert
sands, but they are not so overcrowded as some of the older
cemeteries of the inner town.
On the city plan the sites of graves are indicated by
crosses and some of the more important cemeteries are num-
bered, I to VI. The writer explored most extensively grave-
yards No. I and No. VI; little work was done in Nos. II and
III. Explorations were attempted in every part of the town,
generally with small results, yet sufficient to ascertain the cul-
tural periods to which each section belongs. These experi-
mental spots are marked in the plan by red crosses and small
letters. Judging from our experiences in exploring the ceme-
teries, we may estimate the graves at Pachacamac to have been
between 60,000 and 80,000 in number. Operations in ceme-
tery III were abandoned, the graves being much scattered and
all articles in them of perishable nature poorly preserved; still
it was ascertained that in this burial ground to a space of sixty
square feet about three graves were untouched and ten ruined,
disturbed by former explorations, yielding one body to five
square feet of ground. Judging from the size of this ceme-
tery of 250 by 100 feet, which is a low estimate, and the aver-
age closeness of graves as given above, this comparatively
small cemetery must have held circa 5000 dead. Originally
this cemetery extended 300 feet farther to the east with a
width of 150 feet, but it was probably built over at a later
period, therefore this one plot may have held 10,000 graves.
The cemetery at the foot of the Pachakamaj temple is
about 750 feet long by a general width of 200 feet. Thou-
sands of bleaching skulls and bones are scattered over its
rified surface. Estimating the average closeness of graves, the
same as in cemetery III, there may have been 30,000 burials
in this one place; the size of cemetery II is 230 by 100 feet,
and it must have held at least 5000 dead; the two burial
places IV and V, estimated in the same manner, would have
held 10,000 bodies, and an estimate of 10,000 more for the
rest of the city will surely be a low one.
Most of the graves were disturbed, especially those in
dwelling houses, so that it was impossible to find any
unopened ones there; cemetery II had been rifled to the
utmost ; cemeteries IV and V, after a cursory examination,
yielded no results, likewise had graveyard I been rifled to its
entire length and breadth. This fact is not surprising if it
is remembered with what avidity the digging for treasure
was continued ever since the Spaniards took possession of the
country. Cieza, who visited Peru in the first decade after its
discovery, says:4 “Many of these graves have been opened and
are still explored, for ever since the Spanish conquest of
the empire much gold and silver was found in this way;"
and, the captains Rodrigo Ordoguez, Francisco de Godoy
and others unearthed vast quantities of gold and silver
from graves at Pachacamac.” Ulloa, who visited the place
early in the eighteenth century, says:6 “No idea can be
formed of the topography of the ancient city and of its
houses, as large mounds of adobe cover every part of it, débris
of walls pulled down in the search for tombs and the treasures
supposed to be concealed therein, since wild tales of some
earlier finds spurred the people on to do this.” It appears,
therefore, that early in the eighteenth century the place was in
about the same condition as it is now. A good deal remains
yet to be accomplished for the explorer at Pachacamac, espe-
cially if systematic work is done in this field, as all former
explorations here were carried on in a desultory manner.
The graves still remaining are, fortunately for us, entirely
devoid of objects of gold or silver, or contain but few of these.
If they were of richer yield they, too, would long since have
fallen a prey to treasure hunters and none would be left now.
Middendorf's view that the yield of graves decreases in
proportion to the increase of exploration seems to be correct.?
Our own experience may serve as an example, since the only
burial plot which yielded a few silver vessels was the ceme-
tery of Huarangal, which had not been explored quite so
thoroughly, as it lies at some distance from Pachacamac.
The
Indians of old were not unacquainted with the custom of
reopening the graves after some time and removing provisions
and valuables. It must be remembered, too, that many of
the graves, especially older ones, which were as a rule the
objects of our researches, entirely lacked utensils of metal. It
appears, therefore, that the use of such implements for domes-
tic purposes was not so general yet to find them largely
represented in the graves of this period.
a
1 Compare Hutchinson, I, p. 165.
2 In ancient village ruins in the interior of the valley, at Huaycan, for
instance, the houses are provided with cellars or vaults under the rooms or in the
fore courts, which are lined and covered with stones and were used for the burial
of the dead. They were not filled with sand in the course of the centuries
like those at Pachacamac. Cieza, Cron., I, chap. 63, gives an interesting piece
of information regarding these vaults: “In the province of Jauja and quite gen-
erally in Peru the dead for burial are wrapped into a fresh llama-skin, that is
sewed together and a face attached to it with nose, mouth, etc. The bodies
are kept in the houses, and those who in life were chieftains and caziques are on
certain occasions during the year disinterred by their sons and with great cere-
mony carried about on open biers through the fields, and the customary sacrifices
are made to them which consist in young llamas, even children and women."
Doubtless the same customs obtained here in the Pachacamac valley.
3 Wiener, p. 66, speaks of three principal cemeteries, which he erroneously
assigns as belonging to the temple, the town and the pilgrims.
8 Cieza, I, chap. 13.
* I, chap. 63.
6 Chap 72.
6 P. 301.
Peru, II, p. 118.
(12)

CHAPTER VII.
THE TEMPLE OF PACHAKAMAJ.
2
THE ancient temple (see plan, pl. 3), the most famous sane-
tuary of the city, was so entirely destroyed that its ruins
are generally overlooked by visitors, and the imposing temple
ruins on the hill are taken for those of the temple of Pacha-
kamaj. Only Rivero and Tschudi, Markham? and Midden-
dorf distinctly mention these ruins, and only the two last
named assume them to be those of the Pachakamaj temple ;
Middendorf in his work attempts a description of these ruins.
This temple was built against the northwestern slope of
the hill W (see city plan), on the summit of which, farther
to the south, the ruins of the Sun temple are to be seen. The
structure faced the north west, probably an older part of the
town, and looked toward the Peruvian coastland, while the
valley extends back of the hill. The temple is 400 feet long
by 180 feet wide and covers an area of two-thirds of an acre;
by the projection of its northwest front, at h, in a western
direction, an angle was formed in the centre of the façade,
whereby the almost rectangular outline of the building was
interrupted. Three sides of the temple were terraced, each
of the grades being about twenty feet high, and the top of the
temple formed a plateau, 330 by 130 feet, the nearly rectangu-
lar form of which closely followed the general outlines of the
building. The fourth side or rear of the building, b, merged
horizontally into the summit of the hill against which the
temple was built.
Only on one side, i, the terraces are still distinguishable
Five low and narrow grades, each three and one-third feet
high by six and one-half feet wide, and covered with poly-
chrome frescoing, formed the approaches to the temple on this
side. The terraces on the other two sides are so deeply buried
under sand and débris that the structure here looks like the
natural hill, and only after excavating near h the former
terraces were exposed to view. In this locality, near the
northern angle, may have been the passageway up to the
temple, but it is almost obliterated at present. Of the two
galleries found here, k extends halfway up the temple, leading
past its once richly frescoed northeastern and southeastern
fronts, and here may have been the main entrance of the
temple. On the plateau at the top of the structure only few
details can now be traced. In the centre of the northwest
front, near a, there is a deep trench, fifty by twenty feet, the
earth and stones of which are heaped up in the mounds, b,
as indicated on pl. 2, fig. 1. The shrine of the temple must
have been at this place, the 'boveda,' as Estete calls it, which
was torn down by Hernando Pizarro, and on this site many
treasure hunters were at work in later days. The remaining
area of the temple plateau was by a partition wall divided
into an
eastern and a western section; the former measured
about 170 by 130 feet and must have been a large hall. A
few rooms, apparently sunk into the ground or excavated into
the rock, but now completely ruined, were in the northeastern
part. The walls of the large hall are ruined, but one small
piece of masonry three feet high remaining of the south-
eastern wall, c, formerly eleven feet high. This may have
been the hall where “the old bishop' stood and received
the envoys of chieftains after they had gone through the pre-
scribed fast of twelve months before being admitted to his
presence, when he would receive them with his head covered,
and after having heard the messages from their princes, pray-
ing for good harvests for the ensuing year, the devil's priest
entered the adjoining cell, where he claimed to have speech
with their devil,” according to Pizarro.
The western section of the plateau appears to have been
enclosed by a wall of inferior strength ; along dividing wall
d several rooms were sunk into the floor about six feet deep,
while two terraces, g, g, from six to seven feet high, were
built up against the rear wall. These grades are similar to
those on the uppermost terrace of the Sun temple (pl. 13, No.
20) and were probably used for some sacrificial purpose.
At the time of the conquest the temple must have been
as gorgeous an object to behold as it is now one of desolation.
Smaller in all its dimensions than the younger Sun temple, it
must have eclipsed it in splendor. Miguel Estete speaks of it
as “well painted.” 4 While the Sun temple was painted a
”
monochrome red color, the former was frescoed with bird 5
and animal designs, probably also with plant-ornaments, to
judge from finds and preserved fragments. On the small
remnant of wall c on the south side of the eastern hall a
yellow and red scroll pattern can still be seen. Miguel
Estete in his second unpublished report says that the door
of the cella was incrusted with “coral, turquoises, crystals
and other stones.” The truth of this statement is strength-
ened by the fact that in the Sun temple at Cuzco the
frames of several large niches were set with emeralds and tur-
quoises 8 in a similar manner. It is probable that the temple
was decorated with ornaments of gold, presented to the sanct-
uary by the Inca Hvaina Qhapaj ;9 we do not find any
reference to this, however, in the reports of the first discover-
ers, the Inca priests, warned of the approach of the Spaniards,
having removed everything of value, while the incrustations
upon the doors could not be taken off.10 Thus it happened
7
a
4 Cieza, I, chap. 72, tells that walls and doors were covered with designs of
wild animals.
5 See fig. 9.
6 Fig. 8.
Fig. 7.
8 Garcilaso, book III, chap. 22.
9 Cieza, II, chap. 65.
10 The story that a pilot in the service of Hernando Pizarro begged to be
given, for his share in the booty, the silver nails used to fasten the plates of ham-
mered gold to the walls of the temple of Pachakamaj, and that this apparently
small share in the spoils alone amounted to 4000 marks silver, was rejected by
J.J. v. Tschudi, 1. c., p. 87ff. In his opinion, which is secondary to the question
itself, the gold plates were not fastened to the walls by metal nails but by
wooden ones, an error on his part. In the Museum of La Paz the writer saw a
1 L. c., p. 290.
2 Introduction to Reports, p. xix.
3 Peru, II, p. 115.
(13)

that no gold worth mentioning was found in the temple by
Hernando Pizarro, only a small quantity of gold dust. The
great splendor and prestige of this place of worship, as well as
the high reverence in which its deity was held by the Indians,
made it the object of the most fanatic destruction by the Span-
iards, therefore the ruins of this temple appear so much more
desolate at present than those of the Sun temple.2
The wall in the rear of the temple, the southeastern, does
not stand closely against the hillside, but it is connected with
the latter by a sort of bridge, while the space between, m, is
filled up with adobes to a height of two meters above the floor
of the hall. An excavation near 1 did not result in finding
the original soil at a depth of twenty feet. This filling up as
well as the absence of any enclosing walls farther back, and
the line of masonry, p, at a projected point from the southwest-
ern edge of the temple plateau, suggest that these rear por-
tions of the structure must have had some particularly close
connection with the temple. These portions consist of the
plateau of masonry, m; of the natural ridge of the hill with
a northwestward slope, n, and the artificial mound continuing
the latter, o, as is separately shown in view 3, pl. 17, and can
also be seen on the extreme right-hand side of fig. 1, pl. 19.
A peculiarity of these parts of the structure is the small
amount of architectural detail in proportion to their massive-
ness, while the contrary must be said of the temple itself.
The two sections m and n extend from northwest to southeast
for about 360 feet, from southwest to northeast for 400 feet.
The mound a, mainly an artificial one, is 450 feet long by 400
feet wide and 60 feet high. The ruined sides of this mound
only show the construction of its nucleus, while no trace of an
exterior finish can be found. These facts, together with a
rude terrace grade, thirteen feet high, projecting on the north-
eastern side, give the impression that the structure was left
in an unfinished state. However, it must be assumed that
sections m, n, o were in use for a time. In section n founda-
tions of buildings were struck; in section o, which slopes
from n near » for about ten feet, a thick wall, s, was found;
the remains of stone masonry and on the southeastern edge a
large cremation ground were located near q, judging from
the red color of the clay soil, which is caused by fire. In the
entire section of n and o only one fragment of decorated pot-
tery was found, while on the western slope of the mound in
several spots large quantities of sea-shells (Concholepas peru-
vianus 3) were brought to the surface.
The temple stood within the enclosure as described above,
surrounded by an area extending to the northwest for 230
feet, to the southwest for about 260 feet and to the north,
near the ruined building within the enclosure, for 210 feet.
The enclosure itself is of a trapezoid form, its sides 1000 to
1370 feet in length seem to date from a very early period,
as only a few yards of the former wall remain, and merely
a sort of a dam enclosing a square space is all that is left.
Sections of its former course can be traced on pl. 2, fig.
1, on the northwestern slope of the hill W, below the Sun
temple; likewise in pl. 15, fig. 1, on the southern declivity
of the same hill; then again fragments of its eastern course
can be seen in the plain, partly winding up the hill X in
pl. 9, fig. 1, and traces of its northwestern line in the fore-
ground of the panorama of the town on pl. 1. At this point
the wall was not quite as thick as elsewhere, probably on
account of the structures adjoining it. This ancient wall
was doubtless the inner enclosure of the temple, as is clearly
shown by the way the enclosed area was laid out, as well as
by the structures outside of it. The dwelling house stood
with its rear against this wall; the two walks encircling the
square, at a small elevation, terminate abruptly before this
enclosure and make a turn alongside of it. The western
end winds about the structure in a semicircle and seems to
terminate between the houses to which it may have been the
approach ; these being within the outer temple precincts, this
walk must be considered as a second entrance to it.
Only one small terraced building stands closely against
the inner wall, opposite the northwest front of the temple, and
a winding path leads up to it from the inner temple. In the
panorama of the town this structure stands in the foreground;
its situation suggests a porter's lodge so clearly that Estete's
words may be referring to it when he says: “To this devil's
house the Indian pilgrims resort from remote parts of the
land, some of them coming 300 leagues or more with their
offerings of gold, silver and garments, and upon their arrival
they hand their gifts to the gatekeeper, who then goes into
the shrine and has speech with the idol and hears his
answers.”
As before stated, the outer precinct of the temple
extended as far as the western street of the town and to the
gate near it. The larger part of this area, marked as fore
courts 2 and 3, must for good reasons be regarded as outer
courts of the temple. Court 3 is a peculiarly designed area
near the temple gate, and can only be a part of a grand
and original scheme; it forms a level sandy ground of 1050
by 235 feet, and where the structures, standing in front of
the temple enclosure, have now disappeared, it is 350 feet
wide; two long dam-like lines, much broken into, u (probably
remains of a structure that was divided transversely on its
eastern end), extend in an even distance of twenty-one
feet along the entire north side of this large court. Two
parallel lines, formed of small mounds of earth, n and x, run
from the western side of the wall line through the entire
length of the large court. These mounds terminate near the
gatekeeper's lodge, the mounds x ending farther to the east-
ward. Near w the parallel rows are twenty-three feet apart;
near x only eighteen feet nine inches, and all the mounds
stand in pairs, facing each other. Examination of some of
these mounds proved them to be remains of pillars of three
feet three inches by three feet one inch thick and one foot
high, perhaps originally a little more. These were set upon a
path bordered at ten inches from the pillars with a single row
of stones, and which near x was twenty-three feet six inches
wide. These seem to be the remains of two covered walks and
the pillars may have served as pedestals or posts to support
a roof as a protection against the fierce rays of the sun. If
this court 3 with its peculiar design is claimed as an outer
fore court of the temple, the purpose of court 2, only indis-
tinctly separated from the former, is likewise established. The
significance of court 1, separated from court 2 by a thicker
r
,
a
stone from a wall at Tiahuanaco, which still held in drilled holes some copper
nails that had held the gold plates. On the site of the ruins of Pumapunku (Tiahu-
anaco) are the ruins of a monolithic gateway with a relief frieze where the holes
may be still seen in many spots like the ones above, and which have served a
similar purpose. Photographic views of the gate-ruins can be seen in the Museum
of Science and Art; also a fragment of a stone, ornamented with sculpture in
relief, on which these holes are to be seen.
1 Letter, p. 123.
2 When Cieza came to Pachacamac, scarcely a decade after Hernando
Pizarro, he found the temple partly ruined (I, chap. 72). The statement by
Prescott, I, p. 293, that the Spanish colonists had destroyed the temple and used
the adobe bricks as building material, seems to be unfounded and unlikely since
the Spaniards did not build any houses in that neighborhood, and for buildings at
a distance they could find material more conveniently in other ancient buildings.
* This sea-shell, from the Peruvian Sea, was found in several specimens in
graves belonging to the earliest period of Pachacamac, and in a few instances the
shells were colored on the inner side with a purple or an oxide green earthen
pigment; from a communication of Pilsbury this shell yields a purple pigment,
similar to the oriental color; it is probable that this color was used in Peru in the
early times.
4 P. 131.
5 Seen in panorama of the city, pl. 1, and in fig. 1 of pl. 17 from the distance.
6 Stübel u. Uhle, Ruinenstätte, p. 2.
(14)

wall, but opening into it by a wide gateway, is thereby also
guaranteed as the outer court of the temple. The houses, too,
may have had some particular relation to the temple. In
the court of the house near Z were found traces of large
amphores of the regular type used by the Incas for chicha.'
Hernando Pizarro speaks of these courts? as follows: “Before
entering the first court of the mosque a man was obliged to
fast for twenty days, but a fast of a whole year was imposed
before a person had the privilege of ascending to the upper
court.” We may consider the inner enclosure of the temple
1
66
or the outer court 2, or 2 and 3, as the first court in Pizarro's
sense ; since the latter saw the city in its old condition, it is
likely that he took for the first court what in reality was but
a fore court, and at that time must have been regarded
as such. If this be the case, then it is true that to enter
court 3 required a preliminary fast of twenty days, while
the threshold of the inner court could only be crossed after a
year's fasting. Among the buildings within the outer courts
2 and 3, we would therefore find the houses where the pilgrims
were lodged. All these facts taken together lead us to sur-
prising conclusions concerning the inaccessibility and the
extraordinary sanctity of this place of worship.
1 Letter, p. 123.
(15)

CHAPTER VIII.
THE CEMETERY IN THE TEMPLE GROUNDS.
1
THIS
THIS cemetery is one of the largest in the town and the
most crowded. So many human beings were interred
here that the soil is completely choked with remains of mum-
mies, skulls, bones, fragments of stuffs, ropes, slings, and
still it is full of unbroken mummies, which were buried at
some later period. In point of crowding this cemetery sur-
passes even burial ground No. II of the inner town, where the
graves are exceedingly close together. This crowding of
graves in the cemetery is explained by the great desire of
the Indians of old to find their last resting place near the
shrine of their deity. According to Cieza,2 this was a
privilege reserved to princes, priests and those pilgrims who
could bring rich offerings to the sanctuary. Zarate also states
that the chieftains of the entire community strove to get their
sepulture in the hallowed ground near the temple at Pacha-
camac.
3
was held and of its power to attract pilgrims from remote
provinces, yet on the other hand we are surprised to discover
in these finds proofs of an exceedingly extended use of this
cemetery, accordingly of an equally long history of the sanc-
tuary itself. With these finds as guides the history of the
temple may now be traced back to the earliest periods of
Peruvian civilization of which there is any knowledge at
present and down to the point of the Spanish conquest.
The time has now come for a more serious and scholarly
study and classification of Peruvian grave finds than that
devoted to this subject formerly; it is no longer sufficient to
merely explore in the old unsystematic way, without paying
any attention to the various historical periods, and leaving all
studies of this kind to future scholars in this field.
The extensive operations which are now undertaken in
Peruvian exploration must have the aim of extending our
knowledge of the great cultured types of Peru, of their devel-
opment and successive periods. Only by methodical work
and guided by scientific theories will these explorations lead
to practical and instructive results. Sufficient material is
now at hand and more will appear in time to enable the
scholar to connect genetically all the main types of specimens
of ancient Peruvian civilization in their historical relations to
one another. Local peculiarities of style may thus gradually
gain a typical importance in the general development of
ancient Peruvian cultural forms, as they will present them-
selves interwoven with great historical movements or
derived therefrom, and so at some future time the history
of all the various phases of Peruvian culture may be traced
to its source. Types which were generally regarded as merely
local now gain much greater importance. This may serve as
a warning against the careless grouping of specimens of
unknown extraction together with objects of similar appear-
On the other hand serious research will be rewarded,
as these types may serve as guides in establishing the dates of
certain periods, or even in settling the doubted identity of
culture in localities far distant from each other. Great care
must be observed not to consider rashly the finds of a single
operation as representative for a certain locality.
All observations unite to show that the course of devel-
opment of Peruvian civilization has been a very long one,
passing through many different periods, the deposits of which
we find side by side in the graves and cemeteries of all the
valleys. The cultural and historical character of a locality
may only be determined by examination of all the cemeteries
to be found there. Occasionally graves of different periods
are found intermingled. In such cases the contents of the
graves must be carefully distinguished and separated from
one another. It also happens that the graves are found
lying in strata, as shall be seen in one example below, which
The ancient Peruvians had the custom of deforming the
human skull from the time of birth on, the manner varying
in the different provinces. It might be expected, therefore,
that a great variety of types of cranial deformation could be
found among the skulls of this burial place. This opinion
was also entertained by Markham, but it seems that it
was the case to a very small degree only. The skulls taken
from graves about the Sun temple in this respect differ greatly
from the general class of skulls found at Pachacamac, for
they really show a greater variety in the types of deforma-
tion than those of other cemeteries, such as No. I for instance
There, in more or less pronounced form, prevails the common
type of deforming which was practiced along the coast, the
Chincha type, as it is called by J. J. v. Tschudi; exceptions
are few and generally limited to skulls taken from graves
closely adjacent to the front terrace wall of the temple, which
for various reasons must be considered as exceptional in the
cemetery. This proves that the people interred here did
not come from all parts of Peru, and not from such districts
where the deformations were different from those in the
Peruvian coast regions. Even the ancient provinces of the
coast are not fully represented among these graves, since the
well known ceramic type of the Chimu is entirely absent
from among the specimens, although it is contemporaneous
with this cemetery. If from these indications the graves do
not represent as great a variety of provincial types as was
expected, it is on the other hand surprising to see the differ-
ence among the graves in regard to the periods from which
they date. The finds in the cemetery do not give much addi-
tional proof of the high reverence in which the sanctuary
as
ance.
a
1 Cronica, I, chap. 72.
2 I, chap. 10.
* Markham, Engl. edition of Cieza, note to chap. 72.
4 Virchow, Crania ethnica americana, 1892, p. 14.
(16)

is an especially fortunate circumstance whereby chronological
classification is greatly facilitated. In these cases great care is
needed to distinguish all the signs of such a stratification in
territory outwardly identical.
One special ceramic type, a vase or flask with a long and
thin neck and conical bottom (pl. 18, fig. 1), was named
‘arybal’ by Hamy, from its resemblance to certain ancient
Greek types. This type is found all over Peru, Ecuador,
Bolivia, in a part of Chile and Argentina, where it was car-
ried by the conquest of the Incas. The same may be said
of some other ceramic types related to the former in form and
decoration, such as plates, jars, various kinds of jugs, of the
Cuzco type, concerning which there has long been a general,
if tacit, agreement.
Another prominent decorative type of ancient Peru must
be mentioned here, which is best known on account of its
peculiarity and its great antiquity. Stübel and the present
writer, in their work on The Ruins of Tiahuanaco, proved
that those ancient monumental stone structures date from an
early pre-Inca period, ignoring all legends of doubtful source,
but entirely for serious historical reasons. Numerous objects
are found, possessing a surprisingly strong resemblance to
2
4
1 Philippi, Verhand. d. Berl. Ges. f. Anthrop, 1885, p. 269. Bastian, l. c,
1888, p. 532. The writer: Kultur u. Ind. südam. Völker, pp. 24, 32 and 35.
Hamy, Galérie américaine, text to Nos. 112, 113.
2 Comp. pl. 18, fig. 6.
Comp. pl. 18, fig. 3.
* Comp. pl. 18, fig. 2.
the style of these monuments in all their characteristics.?
Some were shown in the above mentioned work, pl. 4,8 with
distinct reference to this similarity of style. The authors
then ventured to suggest that one of the vases might serve
as a proof of the occurrence of the same cultural forms on
the Island of Titicaca ;' moreover, its characteristic resem-
blance was pointed out to the much discussed sculptured
stone slab from the distant Chavin de Huantar in the north of
Peru.10 At that period it seemed surprising that one particu-
lar decorative style, peculiar to a small district on the south-
ern end of Lake Titicaca in Bolivia, should appear in an
entirely different region of ancient Peru. New discoveries
have since modified these views considerably. Rich col-
lections of specimens in the Museum of Berlin prove that
the decorative style, which, from lack of any other name,
may as well be named after the ancient monuments of Tia-
huanaco, is represented along all the shores of the big lake
as well as on the islands. The writer found to his surprise
that in Bolivia, to the south from the lake, the type abruptly
ceased, its last traces were met with at Ancoaqui on the
Desaguadero, ten miles south of the border town of Desagua-
dero. This important type must accordingly have been more
closely connected with the north than with the south. The
presence of objects belonging to this particular style all about
the lake formed most convincing links between the ruins of
Tiahuanaco and many objects in the museums from southern
Peru which bore a striking resemblance to them.11
If this conjecture proves to be correct the great difficulty
will be removed in establishing the connection of the stone-
slab of Chavin de Huantar with the ancient monuments of
Tiahuanaco and the resemblance in a richly ornamented
poncho found at Ancon. It becomes more and more evident
that the type of the Tiahuanaco culture is not an isolated one,
standing at the extreme southern border line of Peruvian
civilization, but that it may have been a great type of histo-
rical importance for all Peru, which by some accident hap-
pened to come into our notice before any of the others at the
extreme end of the Peruvian cultural territory. We have now
two great starting points for the study of the cultural develop-
ment of ancient Peru : the style of Cuzco, spread over a vast
region of ancient Peru by the Inca conquests, and the style of
the monuments of Tiahuanaco, the original seat of which can-
not be determined for the present. The style was named after
Tiahuanaco, as it found in these works its most pregnant
expression, and since the works were created on the spot it is
a proof that at some particular period this style belonged to
that region.
The founding and growth of the Inca empire was one of
the last great events in the pre-Spanish history of ancient
America. No particular style was discovered, so far, from
which the art of Tiahuanaco might have been derived ; it
therefore stands historically, and on account of its peculiar
archaic character, quite isolated at the beginning of the cul-
tural history of Peru, while the Cuzco style may be said to
form the culmination of the same artistic development. So
the task now presenting itself to us is, to fill in the wide gap
5 Also the type of vases in cup-shape (pl. 18, fig. 4) is Cuzco and must not
be classed with the ceramic products of the people around Lake Titicaca. Hamy,
Galérie améric., suppl. to No. 108.
Below is given a list of some of the places of origin of ceramics belonging
to the type of Cuzco, which will show the vast territory over which they were
spread.
a. ARYBALS.
Ecuador : Quito, Kultur u. Ind., pl. 7, fig. 3.
Achupallas, from Cañar: l. c., fig. 1.
Quinjeo near Cuenca: Congrès des amér. Bruxelles, Atlas, pl. VI,
fig. x.
Chordeleg near Cuenca: 1. c., pl. IV, fig. 4; VIII, fig. 6; XV, fig. 5;
VIII; pl. XIV, figs. 1, 27.
From Cuenca: 1. c., pl. IV, fig. 6.
Malachi, from Quito.
Peru : Huarmey : Kult. und Industr., I, pl. 10, fig. 9.
Pueblo nuevo : 1. C., I, fig. 10.
Ancon and Trujillo : Hamy, Gal. amér., note to Nos. 112, 113 ; Reiss
und Stübel, Necropolis of Ancon.
Pacasmayo: Congrès des amér., l. C., I, p. 476, fig. 14; besides places
about Lima and in the upper Rimac valley.
Bolivia : Corocoro: Kultur u. Ind., I, pl. 11, fig. 11.
Tiahuanaco, Copacabana and all islands of Lake Titicaca : Collections
at Leipsic, Philadelphia, Paris, etc.; comp. also Kult. u. Ind., I,
pl. 12.
Escoma : Collected by the writer, Mus. of Berlin.
Truitu on the Desaguadero river: Specimens collected by the writer,
Mus. of Philadelphia.
San Andres de Machaca : The same.
Turco and Chiriscalla, prov. of Carangas: Collected by the writer,
Mus. of Berlin.
Argentine : Loma Rica, valley of the Calchaquis : Verh. der Berl. Ges. f.
Anthrop., 1884, pl. 7, fig. 1.
El Paso, valley of the Calchaquis : Coll. by the writer, Mus. of Berlin.
Tinogasta, prov. of Catamarca, and neighborhood : Coll. by the writer,
Mus. of Berlin.
Ruins of an old town near Chilecito, prov. of Rioja : Fragments of
large Cuzco ary bals, coll. by the writer, Mus, of Berlin.
Pucara, between Andalgala, prov. of Catam and Tucuman : Sample-
specimens coll. by the writer, Mus. of Berlin.
Chile : Medina, Aborigenes de Chile, No. 211, and Philippi, see above.
6. PLATES
Ecuador : Datacunga : Congr. des amér. Brux., pl. VI, fig. 6.
Peru : Zepita on Lake Titicaca : Coll. by the writer.
Bolivia : Tiahuanaco, all western, southern and eastern (Omasuqu) shores
and all islands of Lake Titicaca : Coll. of Leipsic, Philadelphia,
Berlin, etc.
Truitu on Desaguadero river, Hachiri : Coll. by the writer, Mus. of
Philadelphia
Prov. of Carangas : Coll. by the writer, Mus. of Berlin.
Chile : Vallenar: Medina, l. C., No. 162.
c. VASES AND JARS.
Bolivia : Tiahuanaco and all shores and islands of Lake Titicaca, Mus. of
Philadelphia, Berlin.
Desaguadero: Mus. of Philadelphia.
Chile: San Jose de Maipo: Medina, l. C., No. 183.
d. Bowls.
Ecuador : Infantas: Congr. des amér. Brux., pl. VIII, fig. 2.
Prov. of Imbabura : 1. c., pl. XVI, fig. 5.
Bolivia : Shore of Lake Titicaca : Hamy, Galérie amér., No. 108.
Chile : Freirina : Medina, 1. c., No. 182.
7 Larger series of these are to be seen in the Museums of Berlin, Leipsic,
Philadelphia, Paris, etc.
8 The stylistic identity of specimens preserved at Leipsic and shown in Kul-
tur u. Industrie, I, pl. 11, figs. 1-10, 17-20; pl. 12, was pointed out, p. 34ff, ibid.
9 L. C., pl. 41, fig. 6.
10 L. c., p. 48. A copy and a cast of the above mentioned stone-slab, brought
over by Stübel, are in the Museums of Berlin and of Dresden. The first fairly
good view of this object of so great an archæological importance was given by
Middendorf in his work, Peru, I, p. 631.
11 Several objects of this class from the region of Cuzco, Ollantaitambo, etc.,
are in the museum at Berlin. Among others the vase shown by Squier, p. 458,
drawn by him at Cuzco at some earlier period.
12 Reiss und Stübel, The Necropolis of Ancon, pl. 49; Stübel und Uhle,
Die Ruinenstätte von Tiahuanaco, pl. 21, fig. 3.
6 p. 54.
3
(17
)

between these two extremes and to write the history of the
development of the Peruvian civilization into it.
Returning to the cemetery of the temple of Pachakamaj,
it is apparent that it was used from the period of the Tiahu-
anaco style down to the Inca time. However, it seems to have
been used mainly during the various periods preceding that
of the Incas. Only a few of the graves belong to the last
epoch, and these, judging from the manner of construction
and their individual importance, show that the religious pres-
tige of the temple had not suffered any decline during the
rule of the Incas.
The extremely early date of the first use of this ceme-
tery to which the finds must be assigned, and which are of
such conspicuously pre-Inca character, are a remarkable con-
firmation of the conclusions drawn above from similar proofs,
i. e., that this temple was beyond all doubt that of the Creator-
god Pachakamaj, and had no connection with the Sun cult of
the Incas, as was supposed by Rivero and Tschudi. The
statements of the early historians concerning the great
antiquity of the sanctuary of Pachakamaj are remarkably
confirmed, a fact which has so much more significance in view
of the ever growing tendency among archæologists to discredit
the reports of early chroniclers and to consider them as more
or less willful fabrications, although they are obliged to depend
almost entirely upon their testimony, and are in this case, for-
tunately, quite safe in doing so.
a
(18)

CHAPTER IX.
DISCOVERY OF EARLIER GRAVES UNDER THE BASE OF THE TEMPLE.
THE open ground which surrounds the temple furnishes a
vast field for exploration, and was made the object of
research by many former explorers as well as by the writer
of the present report. The terraces 6,1 dating from explora-
tions in early Spanish times, covered a great many graves of
ground before it was so much disturbed. The cemetery was
found to extend partly under the wall of the temple, and as
this must have been covered and built over a long time
before the Spaniards came, explorations on this spot seemed to
be promising of interesting results, since the covering up of this
P
P
Eigure 4
3
y
b
S
χ
y
figure 5.
Em
y
7
χ
72
7
y
72
Figure 3
Fig. 2. Plan of cemetery at the foot and under the base of the temple. Sc. 1.240 (site of explorations).
gus
1000000
09
t
0
Po
O
00
n
CD
2
5- DOO
87.pojat
08
lo
Holecka
a
a
B B
DIZA
Figs. 3-5. Sectional views of the cemetery at the foot and under the base of the temple of Pachakamaj. Sc. 1.120.
the ancient cemetery, thereby saving them from the ravages of
treasure-seekers. These graves date from the same period as
the rest of the burial ground, since the soil was piled up above
them long after this field had ceased to be used as a cemetery,
and from them we may judge of the condition of this burial
territory at some early pre-Spanish period formed, to a certain
degree, a preclusive time-limit for the construction of those
graves, and it seemed probable that they would shed some
light on the time of erection of those parts of the temple as
well as give us some new clues concerning the relative age of
the various classes of graves found here. The trenches made
1 Plan, pl. 4.
(19)

for the exploration of this section were cut through several
terraces and substructures of the temple, then continued in a
horizontal line until a natural obstruction was reached in the
front part. The local conditions are reproduced on the diagram
of these explorations, fig. 2, and in a transverse section of the
same, figs. 3–5.
Fig. 2 ; scale 1.240.
x, site of the former excavation of a treasure digger;
y, territory of writer's explorations; n, n, n, foundations of
pre-Spanish buildings in front of the temple façade; ſ, p, q,
thickness had apparently been spread over the soil under-
neath as a foundation for the building which was
to be
erected there. In addition to a wall, f, which seemed to rest
upon some point above this stratum, there were built upon
it the walls of terraces, p, q, 1", s, t, see figs. 3-5, of various
heights. During the centuries since the conquest these ter-
races were covered by sand, as indicated near o in fig. 3,
while fig. 4 shows them in their present uncovered state.
The wall p, fig. 4, was built against f, concealing it com-
pletely. During the progress of excavations in the direction
f
8
r
S
1)
Ji
it
000
2009
0
Fig. 4. Section of temple-terraces at the site of explorations, of fig. 2.
Fig. 5. Similar section. Sc. 1.120.
r, s, t, terrace walls; b, a covered stone wall in the interior of
the temple.
Figs. 4-5; scale 1.120.
The outlines of the transverse section of figs. 3 to 5 are
explained in the diagram, fig. 2. Figs. 3 to 5 have the same
marking-letters as fig. 2. The original soil, y, figs. 2, 3, sloped
y
down to the ancient stone wall b, in the interior of the temple,
figs. 2–5. The latter, built of broken stones without mortar,
was sunk about one foot deep into the original soil and rested
upon a stratum of clay two feet in thickness, c, figs. 3–5,
which extended a foot and a half beyond the wall. The
exposed section of this wall decreased irregularly from the
northern to the southern end from a height of ten feet
to four feet. Also the filling of the wall in the rear had an
irregular appearance, clearly it was not in its natural condi-
tion, but a ruin, and a very ancient one, too. In front of this
of the stone wall b, a fragment of masonry, p, fell and unex-
pectedly disclosed a section of wall which was covered with
frescoing. The design was faint and barely traceable in its
general outlines; it showed human figures in yellow, white or
a greenish tint with black contours painted upon a back-
ground of red. The further uncovering of the wall, as far as
it was still in tolerable preservation, exposed to view eight
human figures, arranged as if walking in a procession, some
of them connected by a line, as by a rope; there was also
some oval object indicated (see fig. 6). This mural painting,
in point of style, has no relation to any other find made at
Pachacamac, nor to any object of Peruvian antiquity, except,
perhaps, to the designs upon some of the most important of
the Chimu vases from the region of Chimbote, Trujillo, etc., a
number of interesting examples of which can be seen in the
Museum of Berlin. It would be rash to insist upon any close
I
Aud Try
Illy
walila
Fig. 6. Mural painting exposed by excavations. Ab. 3. n. s.
a
wall was found a stratum of adobe all broken and of irregular
sizes, some of them decorated on their sides, d, figs. 2–5; this
layer rested upon the natural soil of the burial-ground. A
stratum of earth, e, figs. 3-4, in which only a few broken
adobe bricks were found, extended to the upper edge of the
ruined wall, b. A stratum of grave-soil, figs. 3–5, full of
fragments of all kinds, was found above the former ; it was
intermingled with pieces of cane, llama bones, skulls, braids
of human hair, pieces of rush ropes, scraps of cloth, a few
fragments of crude pottery, etc. This layer of uneven
relation between these ceramic designs and the mural painting,
especially since the latter is so very poorly preserved; still, in
both of them there appears the processional arrangement of
the figures and the same way of expressing motion, as, for
instance, on the Chimu vase with the dance of the dead in
the Berlin Museum. This mural painting, therefore, may bear
some relation to the art of the Chimu.
The walls, figs. 7 and 8 and fig. 9, also showed some
well preserved frescoing in bluish green, red and yellow.
The design of fig. 7 cannot be interpreted. Fig. 8 seems
(20)

to have figures somewhat resembling birds in the conventional
style of the textile designs of the coastland. Fig. 9 shows
distinctly the form of fishes, probably sharks; there seems to
be a striking similarity of design in the type of this pattern
with that of a carved wooden bowl, fig. 95, of the time of the
Incas.
The original soil in front of the temple terrace, figs. 2, 3,
near n, n, n, was at some early period covered over in order
to raise the grade of the ground about the terraces, where it
was too sloping. A thin layer of leaves of the Pakai tree,
mummies indicated near u, fig. 3, nearest the temple. These
graves are marked by the objects found therein as belonging
to the period of the style of Tiahuanaco and the one imme-
diately following
2. Graves of the oldest superposed soil, d, e, over the
original cemetery, which were made previous to the construc-
tion of the temple terraces upon this ground; these graves
are marked v, fig. 3. The cultural character belonging to
these burials differs widely from that of the first stratum ; a
faint resemblance to the earlier style can be traced in these
;
2
Fig. 7. Fragments of painting on the wall of a temple terrace. Ab. i n. S.
Fig. 8. Traces of mural painting on another spot of the same wall. Ab, 15 n. s.
i, fig. 3, which may have been taken out of the fillings of
old mummy packings, were the foundations; over this was an
even stratum of one foot eight inches of grave soil, k, inter-
mingled with scraps of cotton, cloth, pieces of straw-rope,
slings, etc. Alayer of common dirt was met near I, and the
stratum m had finally been laid upon it for grading before
the walls n, n, n, were set upon this soil. In this manner the
section, y, fig. 2, immediately in front of the temple terrace,
was in pre-Spanish time raised from two feet six inches to ten
feet above its original level at the time of construction of
objects, but the technique, form and ornamentation show a
new type, artistically inferior to the former.
3. A few graves situated in front of the temple, dating
from a time after the erection of structures over the original
cemetery, y, and of the graves, v, as well as the terraces, q, r,
s, t; these graves are marked w on the cross section of this
burial ground.
These graves are marked by objects of the Cuzco type or
of that of the coastland, differing from the objects found in
the open sections of cemetery I, which belong to an earlier
S
2
DICIT
Fig. 9. Mural painting-Fishes, on another terrace wall. Ab. 15 n. s
t, and this was the reason that no more burials were made in
this locality.
It must be observed that to the three stages of soil, as it
was demonstrated here, correspond three stages of burials,
each of a different cultural character.
1. The graves in the original burial ground, y; the
type of the Peruvian coast-style. Below n, near 1 and k, fig.
3, some few unimportant mummies of children were found.
It would be a mistake to speak of a regular and intentional
stratification of these graves, since the relative depth of the
graves entirely depends on the thickness of the soil-stratum in
which they happen to be.3
1Cf. Reiss u. Stübel, The Necropolis of Ancon, pl. 64, fig. 2.
2 Inga feuilli D. C. Cf. Wittmark, Congrès des amér., Berlin, 1888, p. 328.
The fruit of this tree is often found in graves of this cemetery, together with its
leaves, which were generally used as stuffing for the mummy bales.
3 Squier, p. 72, speaks of strata in which the dead were buried, wherefrom
it might appear that he observed stratifications of graves in the open parts of the
cemetery, which would be erroneous, as there are none to be found there.
(21)

CHAPTER X.
GRAVE-FINDS.
GRAVES OF THE ANCIENT CEMETERY UNDER THE BASE OF THE TEMPLE AND IN
FRONT OF IT UNDER THE NEW SOIL.
THESE
THESE graves were found at various depths below the
original surface, but without any distinguishing marks
whereby their respective ages might be established. The ver-
tical plane of the uppermost graves was barely one foot six
inches below the surface, while others were as deep as ten feet
underground. These tombs generally were constructed in a
very solid manner, built of stone or adobe in the form of
chambers of a conical or cylindrical form, rarely square; a
few were roofed over with stone, but the majority had a cover-
ing of cane, matting or some similar material. In some
instances the roof was supported by a pole in the centre or it
was covered with wood. The mummies generally are in the
shape of bales or packages, with a false human head attached,
the face of which is carved of wood or is merely painted
upon a stuffed cushion, or occasionally made of burned
clay and crudely tinted. The types were of the general
class as given by Reiss u. Stübel in The Necropolis of Ancon,
pls. 14–19. The position of the mummies was found to be
similar in most cases, facing east and the temple at the
same time.3
tery it must not be regarded as exclusively ancient. The mummy, Reiss u. Stü-
bel, 1. c., pls. 12, 13, belongs to the Inca period, to judge by the stuffs with which
it is covered ; also, Cieza, Cron., I, chap. 62, states that in his time similar mum-
mies were still prepared at Jauja. The mummy types as given by Squier, p. 73,
and Middendorf, II, p. 117, in the description of Pachacamac, have no resem-
blance with any actually found there.
2 Peru, I, 398.
3 Tschudi's statement that the mummies faced west must therefore be cor-
rected.
1 Although this type is quite rare among the mummies of the open ceme-
FINDS IN THE STYLE OF TIAHUANACO.
10
Only a small number of specimens were found and those
intermingled with objects of younger, kindred or alien styles ;
objects shown in pl. 4, figs. 1 and 3, were taken from the same
grave with pieces of pottery, pl. 5, figs. 1, 9, 10, 11. The
association with articles of more recent date seems to prove
that the specimens in the Tiahuanaco style were ancient ones
at the time when these burials took place, since a grave must
be dated by the youngest objects found in it. None of these
graves, therefore, dated further back than the cultural period
subsequent to and developing from the Tiahuanaco style.
This conclusion is in harmony with another indication which
points to the time when the temple was first built: a few
decorated fragments of pottery were found during our opera-
tions near I, pl. 3, clearly belonging together, yet picked up
at various depths. From their decoration they were assigned
to the period described, p. 24, and seem to indicate that
at least a part of the oldest temple structure was built
during that period. The presence of articles in the style
of Tiahuanaco in the graves is a proof that this period
was still within the remembrance and feeling of the living
when these graves were constructed. The graves containing
objects of this class were the oldest in the cemetery and
therefore most damaged by subsequent burials; for the older
a grave is, the greater the danger of such a destruction. One
object preserved of an ancient period must be regarded as a
representative specimen of a larger number of similar ones,
than would be one object of a relatively younger epoch. We
may, therefore, assume that originally this oldest period was
far more numerously represented in these graves than it
would appear at the present moment; many fragments of
pottery, decorated in the Tiahuanaco style, dating from early
graves and destroyed by later burials, were picked up during
our explorations.
Pl. 4, figs. 1 a-1c: Pieces of cloth painted with mytho-
logical figures; 26721, Mus. of Sc. and Art; 1o n. S. This
stuff was found, with the pieces of pottery above mentioned,
near the mummy marked u, fig. 3. By its favorable position,
being closely set against the stone wall b, fig. 3, this mummy
escaped destruction. The stuff was folded double down the
middle of the central figure, the painted surface turned out-
side, and was used to envelope a mummy bale and the
back of its false head; in unwrapping the latter, the cloth
was found to be in five pieces, three larger and two smaller
ones, the edges of which were much rotten. The stuff may
have been in fragments when the mummy bale was made; but
it is also possible that the stuff, when folded, was in one piece,
for the damaged spots which appeared, when unwrapped, were
in each case symmetrical, figs. a, b and c, d and e;1 some
rents in the cloth had been mended before it was used to wrap
the mummy. The original length of this piece of stuff cannot
1 A few rotten fragments of the cloth dropped off since it was found, but
fortunately the design bad been copied previously, so of the lower parts of the
central fig. a. The design of the latter in pl. 4 is completed with dotted lines.
The texture of the cloth is still much firmer than it might be supposed from the
decayed condition of the edges.
(22)

1
2
4
9
be estimated; it was woven in one width of three feet eleven
inches; the material is a white cotton fabric, the warp of
double threads and the woof thicker than the single warp-
thread; it had twelve threads of woof to twenty-six of warp
to the half inch.
The design upon this cloth is outlined in brown, filled in
with a light yellowish brown, blue and red; the red color
is of an ochrous substance, the others seem to be some organic
pigments. Only a few colors were used, but each of these
appears to have some special meaning; red possibly stood for
gold, since the head ornaments are in that color; blue may
have been used for silver; the yellowish brown for copper.
In the three largest fragments of the fabric three figures may
be traced, which stand full front, while two outer figures
stand facing the central one. The grouping here is the same
as in the sculptured relief upon the large gateway of Tia-
huanaco. There are some additional points of stylistic resem-
blance with the former design: the division of the space into
squares, each of which is filled with one figure; also the
hands, the peculiar ring-shaped end of the nosess of some
figures, and the heads of animals attached to various parts of
these figures, to their feet, etc.
The central figure is three feet eleven inches high, and
if the same space were allowed for the two side figures, figs.
1 b and 1 c, the stuff would have been fully eleven feet six
inches long; however, the three fragments, figs. 1 a-c, do not
fit into one another in their details, and the two smallest
pieces, figs. 1 d and e, show one right and one left hand with
adhering parts which would not fit into the design anywhere if
it had consisted of these three sections only. It seems, there-
fore, as if one section of each had disappeared between 1 a and
1 b and 1 c; at this rate the design may originally have had
five sections and its full length may have been fully twenty
feet. The central figure has an ornamental belt, armlets and
anklets, ear and head-ornaments, the latter with long-necked
snakeheads drooping from it on both sides ; hooked lines are
indicated at the corners of the mouth. One hand of the
figure holds a broad knife of the usual shape, while the
other is grasping a human head by the hair. The figure is
represented apparently after having decapitated a person ;
and more severed heads are suspended like trophies on the
breast of the figure.
The awe- and terror-inspiring nature of the deity is fur-
ther indicated by the snakeheads upon long necks which pro-
trude from all parts of the body and from the headdress,
from the hand, shoulder, elbow, breast, hips, etc., even from
the knife-handle, from the topknot 6 and from the severed
neck of the head held by the figure in its hands. Nearly all
of them are stretching sideways and at the same time sym-
metrically filling the space about the figure. They are dotted
all over with red, blue and brown spots, which are probably
intended to indicate metals or precious stones. The snake-
heads are crowned with a square ornament like the feathered
serpent of Quetzalcoatl of Mexican myths; and all of them
appear in profile and have ears, while the half-opened mouth
shows its fangs; the end of the nose seems curved backward.
The heads of the side figures are in profile, while the rest
of the figure appears in full front; the lower parts are miss-
ing, also an arm of each and the greater part of the head-
dress; both have some ornament near the eye and the tongue
is protruding, the same as with the trophy-heads; the nose is
shaped differently, which shows an intentional mythological
individualization ; the objects which they hold in their hands
are different; the right hand of the figure to the left holds
only serpents, the left hand of the other figure a large knife;
the missing figure, next to the central one, to the left, held
a detached human head with its left hand, fig. 1 d. The
filling of the remaining space with serpents' heads is the same
as with the central fig. 1 a. The interesting design which
covers the remaining space we owe, perhaps, to the accident
that the entire right-hand side was left free. These figures
therefore were added to cover the space, as if to indicate that
they do not belong to the central figure and perhaps also
to mark the terror which the serpents inspire, they are all
turned away from the deity. There are four figures, the
upper ones of which, for lack of space, are only indicated by
their crowned heads. Nothing terrible can be noticed upon
them, the only feature which might be called mythological
the ring-shaped nose of the lowest one. The latter resem-
bles the two dwarf-like figures on the relief of the Gate at
Tiahuanaco,' and the tufted crown, the head and the entire
body, each occupy the same amount of space. A poncho
and a girdle, anklets and armlets are the ornaments of the
figure; on its breast is marked a crescent-shaped shield, the
left hand is raised to the mouth, while the drooping right
hand seems to hold a doll. The figure above is bent in a
semicircle, while its feet are resting against the right-
hand edge, probably for lack of another point of support
in the space. It wears only a red garment, poncho, and an
indented crown, similar to that worn by the severed heads and
it
may also be supposed to wear a girdle. This piece of cloth
which, judging from its size and its peculiar ornamentation,
probably came to be used as a mummy-wrapper in some
accidental way, originally must have been used to decorate
a temple wall, as an object of worship in itself.
Pl. 4, fig. 2: A figure with head and wings of a bird
from a tapestry poncho. The present figure was completely
reconstructed from several fragments of stuff, each of which
showed the figure imperfectly. The original stuff was a
black tapestry poncho, the pattern of which showed regular
squares with the design of the above figure, repeated in
rows, similar to the pattern in the tapestry poncho found at
Ancon.10 The rep-like fabric is of the finest wool, of 114
threads of the woolen woof to twenty threads of cotton warp
to the half inch. The colors are bright, as in all stuffs
of this period, green with white, red and yellow, blue with
yellow, red and brown. The figure is an exact repetition of
the sceptre-holding winged figures of the second row upon the
Gate of Tiahuanaco,11 with the only difference that the rectan-
gular spaces are more closely filled and that the figures are
stiffer. The heads of animals on the crown are also changed ;
the heads set upon the wings are simpler in form and a small
head, like that of a fish, was added to fill the space near the
sole of the hindfoot.12
The same figure as the above is shown on another frag-
ment of a much finer material, which has 80 by 116 threads
to the half inch and belongs also to a tapestry poncho. The
figure here is on a yellow ground of brighter coloring and
5
a
а.
1 Stübel u. Uhle, Die Ruinenstätte von Tiahuanaco, pl. 5, fig. 2.
2 L. C., pls. 8, 11, 15, 16, figs. 1-3.
3 Comp. with heads of puma, l. c., pls. 8, 10, 15, 16, fig. 3, and notes to pl.
8, No. 2,
* Comp. 1. c., pl. 16, figs. 1–3.
5 For the band-like pendants of the headdress compare the head ornaments
of various stone figures in the Ruins of Tiahuanaco, Stübel u. Uhle, pl. 31.
6 Cf. Ruinenstätte of Tiahuanaco, pls. 8, 9.
? A feathered serpent of a different variety is on a vessel in the Berlin
Museum.
8 A serpent with ears appears on a stone sculpture of Tiahuanaco, also on
serpents of gold of the Chilchas, Colombia, Kult. u. Indust., I, pl. 22.
9 Stübel u. Uhle, 1. c., pls. 17 and 21, figs. 1, 2.
10 Reiss u. Stübel, 1. c., pl. 49 ; cf. also Stübel u. Uhle, pl. 21, fig. 3.
11 Stübel u. Uhle, 1. c., pls. 13-15.
Comp. the latter ornament to the heads and necks of animals attached to
the points of the feet, fig. 1 a; also Stübel u. Uhle, pl. 21, figs. 1-3.
12
(23)

narrower.
a
The poncho of which this piece is a fragment
might have been a copy of the former, as the figures in both
have the same height, six and five-eighths inches, and the
design may originally have been of the same width, five and
five-eighths instead of four and one-fourth inches, only it was
more drawn in in weaving, the warp threads being contracted
by the woof. The greater stiffness in the design of both these
stuffs compared to the sculptured figures at Tiahuanaco was
not caused by the weaving alone, as the figures of the poncho
of Ancon show a much easier treatment of form, therefore it
may be explained by a somewhat later date of origin. It is
unlikely, too, that the very artistic arrangement of the bird
in the square spaces of these stuffs could be older than the
much easier arrangement of these figures in the stone sculp-
ture. The exact repetition of the figures from the Gate of
Tiahuanaco in these textiles is an interesting evidence of the
great importance they must have had to the contemporaries.
Even if the designs should have been entirely original with
the makers, which is improbable, it would only show that
the same ideas which caused the creation of the works of
Tiahuanaco were at work in an entire people.
Pl. 4, fig. 3: Cup with two faces, of clay. 26707 Mus. of
Sc. and Art. This cup was taken from the same grave, u,
fig. 3, p. 19, as pl. 4, fig. 1. The object possesses all the
technical perfections of the pottery of the Tiahuanaco style,
and which was only matched, or in some respects excelled,
by the art of the Chimu.
The design inside represented three figures facing front, like
the central figure of the Gate in Tiahuanaco, each of them
holding coiled serpents instead of a sceptre; they are clothed
in a poncho with bands encircling wrists and ankles, the same
as the figures on the right-hand side on pl. 4, fig. 1 c. The
head may have been turned sideways as in the fragment of
fig. 13, which, the same as fig. 12, belongs to a smaller vessel
of a similar kind; this head resembles a dog's head and
wears a square headdress. The linear design of pl. 4, fig. 4,
is repeated on the rim ; a ring-shaped animal's figure in out-
line runs around the outside of the vessel.
Pl. 4, fig. 5: Ear ornaments of wood, inlaid with pieces
of shell. Mus. of Sc. and Art 26720. This object is in the
shape of a pipe bent in an angle; it consists of a head with a
piece attached to it like a stem, and which is also hollowed
out like a pipe; but the vertical stem is so narrow that it may
have been used as a holder of a small tuft of feathers. The
face shows the fine relief which is peculiar to this period and
may be noticed on ceramics, stone sculptures, etc. The nat-
ural form of the eye with the pupil is surprising for this early
period. A diadem encircles the forehead and triangular bits
of shell were inlaid on the cheeks and above the diadem.6
Conclusions : The above described specimens in the style
of Tiahuanaco are closely related to the objects known as
coming from the ruins of Tiahuanaco; the flying and run-
ning bird on vessel, fig. 4, also harmonizes well with these
designs, although nothing exactly similar has so far been
5
B
= 0
Figs. 10–13. Fragments of two or three vessels with a similar design, mythological figures, Tiahuanaco style. i n. s.
The pottery of this period shows a greater variety of
colors than any of the later styles; the same may be said of
the textiles of this period. This cup, for instance, has red,
yellow, gray, white and black in its decoration. The mouth
shows four tusk-like teeth and scroll ornaments and wings are
joined on to the eyes ; lines of yellow scallops border the face
above and below. The nose of one of the faces is gone, which
helps to prove that these specimens of the Tiahuanaco style
were old when put into these graves.
Pl. 4, fig. 4: Flask with bent neck, decorated with two
birds. 26709 Mus. of Sc. and Art. The design on a red
ground, as in fig. 3, is also bright in color; the two birds
seem to be both in rapid flight and running; they must have
some mythological significance, as each has a hand and a foot
and some object hangs by a thread from their beaks, the
meaning of which has not yet been guessed. The figures are
separated by a linear design in black and white.
Among fragments of pottery belonging to this style
worthy of notice are figs. 10–13, 26712 a, Mus. of Sc. and
Art. These belonged to decorated vessels of a flaring shape,
thick and well baked and not very porous, the decoration of
the inside consisted of mythological figures and some had
a simpler design on the outside. Figs. 10 and 11 must be
parts of the same vessel, of about six inches in diameter at
the bottom by fourteen inches at the rim and seven inches
in height, with five-sixteenths of an inch in thickness.
found there; it is related to the kneeling winged figures in
the relief of the large Gate, as well as to a seated monster
with wings like fig. 14, and appeared also upon a vessel found
at Tiahuanaco.
Then, on the other hand, there are features of less con-
vincing nature. Bottle-shaped specimens, such as pl. 4, fig. 4,
have, to our knowledge, not been found yet at Tiahuanaco,
and although this object cannot be assigned to the northern
Peruvian coast, yet forms similar to this one are rather fre-
quent there; it may, therefore, be assumed that this flask was
produced in Peru and not in Tiahuanaco, south of the Lake
Titicaca. A decorative use of serpents, as seen in pl. 4, fig.
1, and in figs. 10–13 of the text, above, is foreign to the
style of Tiahuanaco. It may be said to bear relation to the
remains of the style of Chavin de Huantar, in the province
of Ancacho," although they cannot have been quite unknown
in southern Peru. Tusks in human faces, as in pl. 4, fig. 3,
are not confirmed in Tiahuanaco, but seem sufficiently sup-
ported in the stone sculpture of Chavin. The dog's head
a
5
1 Cf. fig. 11.
2 Cf. thread ornament of main figure, pl. 4, fig. 1 a.
3 Cf. besides pl. 4, fig. 3, Stübel u. Uhle, l. C., pl. 4, fig. 4; pl. 8, etc.
4 Cf. ear ornament, pl. 4. fig. 1 a-c.
5 The eyes on cups, pl. 4, fig. 3, are indicated in a similar manner.
6 Cf. pl. 90, fig. 13.
7 The sculptured relief of Chavin, reproduced by Middendorf, Peru, I, p.
631, and a carved column, reproduced by Wiener, p. 575, and differently by Mid-
dendorf, 1. c., III, p. 99.
8 From a stone sculpture from Cuzco, Mus, of Sc. and Art.
(24)

come mainly from the latter, as they were not produced at
Pachacamac and only a few of them might possibly have been
brought from the shores of the Titicaca. The examples in pl.
upon a human body, as well as the manner in which the
detached human heads are held by the hair, have a strange
appearance, as in pl. 4, fig. 1 a, this has a greater resemblance
to the design upon the central vase, Squier, p. 458, which
belongs to this style and was found near Cuzco, than to the
suspended head trophies as seen on the main figure of the
relief of the large Gate. All these details taken together
help to prove that, in addition to the district of Lake Titicaca,
there must have been another important locality to which this
style was peculiar. The specimens from Pachacamac seem to
Fig. 14. Mythological winged animal. Design on ancient cup from Tiahuanaco.
ž n. s.
4, figs. 2–3, sufficiently support the proof of a direct, inward
and cultural connection with the most important monument
of his style.
1 Stübel u. Uhle, pl. 8.
4
(25)

CHAPTER XI.
FINDS IN THE EPIGONE STYLE.
FOR
TOR lack of a more fitting term, we designated as the
Epigone style that cultural type which, although closely
related to the style of Tiahuanaco, is inferior to its famous
prototype in almost every respect. Objects of this class pre-
dominate in the graves under the covered section a of the
original cemetery. Articles of utility are especially numer-
ous among these; some classes of these specimens appear
at all periods alike, workbaskets for instance, while others
are peculiar to this particular period, among which are cer-
tain types of pottery, textiles, some carved sticks of wood,
spinning whorls and ornaments of shell, bone or tooth.
cat, pl. 5, figs. 6 a and b. 5. An important ornamental
detail, derived from Tiahuanaco, where it occurs frequently,
the triple, seldom quadruple, tufted headdress.
The decorative designs of numerous vessels doubtless
have some religious significance. The sceptre-holding figure
is identical with the image of the divinity pictured in the
Gate relief and elsewhere; the condor has a mythological
meaning in Peru and the running and flying condor much
more so, as it is always pictured in connection with mytho-
logical figures, as in the design pl. 4, fig. 4; concerning the
mythological nature of the cat, pl. 5, fig. 6, compare the
description of the same. All this is easily explained, when
it is remembered that these vessels were mostly used with the
chicha' drinking at religious festivals, and were for that
reason decorated with designs having some religious signifi-
cance. During this Epigone period religious ideas still found
expression in the typical mythological figures which the
classic period of the Tiahuanaco style bequeathed to the
following generations, and we are indebted to the religious
customs of these Peruvians for many specimens which demon-
A. POTTERY.
The shapes of the pottery of this period show in general
the same types as those of the former period. 1. Cups, see
pl. 5, figs. 1 to 4 and 7. 2. Bowls, see below, fig. 17;
these resemble vessels of which fragments figs. 10–13 are
parts. 3. Vases with bottle-necks, see pl. 5, fig. 5. 4. Ves-
sels shaped like animals, having the opening in the back,
FH
E
Fig. 15. Fragment of cup, Epigone period. i n. s.
Fig. 16. Complete design of cup, pl. 5, fig. 2.
4
2
such as pl. 5, figs. 6 a and b; all these ceramic types are
familiar ones from Tiahuanaco. 5. A number of pieces of
this period resemble the vessel, pl. 4, fig. 4. These objects
are technically inferior to the products of the earlier period,
both in shape and color,- although the decorations upon the
earlier specimens bear quite a strong resemblance to those of
the classic epoch. The most important decorative types are
these : 1. A figure bearing a sceptre and seen full front, like
the central figure of the Tiahuanaco Gate or of the relief of
Chavin. 2. A condor's head, as it appears upon the Gate
and upon many other objects from Tiahuanaco ; compare pl.
5, fig. 3. 3. The condor in flying and running motion; com-
pare the figures upon vessel, pl. 4, fig. 4; also fig. 19 below.
4. An animal type resembling a cat, apparently a mountain
strate the close connection between this period and the pre-
ceding one.
Pl. 5, fig. 4: Cup with square face, imperfectly pre-
served. 26724, Mus.; } n. S. The face only with a head-
dress is to be seen here and with some decorations to indi-
cate the clothing. The descent of this design from the type
of Tiahuanaco is proved by many features; for instance, by
the square shape of the face, its frontal position, the form
of the eyes and the ornaments near them, the interlocking
hooks in the headdress, 5 and the tufted effect similar to a
feather crown. Other cups of the same style and period have
a more flaring rim and a design in red on a white ground,
like fig. 15, 26723 a, Mus. The animal heads seem to be of
1 See below, figs. 23-25. Vases with wide necks, as shown in Kultur u.
Industrie, I, pl. 11, figs. 8-10, so typical for Tiahuanaco, were never found at
Pachacamac.
2 Many vessels, such as cups, bowls, flasks, are peculiarly dull in color ; the
cause of this may be that these articles served as drinking vessels for 'chicha.'
3 Middendorf, Peru, I, p. 631.
4 Therefore the local Wiracocha legend of Carabuco (Santacruz, p. 239) was
reproduced upon wooden cups received by the writer at Carabuco and now in
the Berlin Museum.
5 Comp. the headdress of the figure on the Gate of Tiahuanaco, Stübel und
Uhle, pl. 8 ; Kult. u. Ind., I, pl. 11, fig. 6.
6 Comp. I. c., pl. 11, fig. 6; also head ornaments of various sculptured
figures in the Ruins of Tiahuanaco, pl. 31.
(26)

the type of foxes ; the severed arms and legs in this design
have a peculiarly savage appearance suggesting cannibalism.
Pl. 5, fig. 2: Cup with the figure of a divinity. 26726,
Mus.; } n. S. The design represents a sceptre-holding god, as
in fig. 16, and in its general outlines as well as in some minor
points it is similar to the famous figure of the Gate relief.
Among the analogous details are: the indented headdress,
with a change of the centre to the form of rays; the vertical
stripes in the poncho, and the girdle with tassels ; likewise the
geometrical division of the sceptres and the shape of the
hands. From the close resemblance between the two designs
it is supposed that both represent the same idol, and that the
design upon the cup was derived from the older one. Four
white teeth, distinctly visible in the mouth of the figure, bear
a greater resemblance to the faces on cup, pl. 4, fig. 3, and to
the figure in the stone relief of Chavin.
forms the centre; the headdress, with band ends terminating
in condor heads and a plume set into it, has numerous paral-
lels in the relief of the Gate; the pendants at both sides of
the head, ending in a scroll and a short plume, may be parts
of a headdress.
The condor, running and flying, appears twice repeated
as the decoration of a small cup, 26728, Mus., fig. 19.
Another cup, from cemetery III, 26733, Mus., is orna-
mented with several conventional animal heads, fig. 20, resem-
bling those of fig. 15 above. Both are doubtless derived from
the style of Tiahuanaco, judging from the hooked ornament
on the neck. A final interpretation of this head may be
attempted only after some more specimens will be obtained,
showing this figure in a more complete design.
Pl. 5, fig. 5: Vessel in the form of a figure playing upon
a Pan’s flute, two monkeys at its sides listening. 26747, Mus.
Do
og
Fig. 17 a. Ca. 1
Fig. 17 6. Complete design of cup, pl. 5, fig. 3, Epigone
period. i n. s.
Fig. 18 a. Ca. 7 n. S.
Fig. 18 b. Design of cup, Epigone period. § n. s.
n. S.
§ n. s.
Pl. 5, fig. 3: Cup with two condor heads. 26729, Mus.
The stiffly conventionalized heads are quite in keeping with
the general stylistic degeneration of this period; the head
ornaments encircle the head like bands and end in feathery
pendants (above, No. 5) which as a rule seem to be flying
backward as if caused by the rapid motion of the creature.
The bowl, fig. 17 a, with its decoration, fig. 17 b (26730
Mus.), resembles the foregoing specimen, but is of superior
quality.
Pl. 5, fig. 1: Cup with two faces. 26734, Mus.; } n. s.
This object was found in the same grave with the cloth, pl. 4,
fig. 1, and the vessels, pl. 4, fig. 3, and pl. 5, figs. 9–11. The
faces upon the cup2 are similar to those of the cup, fig. 2, and
of the famous Gate; the curved line indicating the chin must
The head and arms of the figure are modeled; the
body of the vessel seems drawn in about the middle, and both
the upper and the lower parts are indented like a melon; the
face was painted red like the faces of mummies. The mytho-
logical nature of this figure is shown by the decoration on
the body of the vessel, fig. 21; in the upper part of the
design an animal is to be seen, with the head of a lizard ; its
tail is coiled upward and it is in rapid motion like the con-
dor, pl. 4, fig. 4. The decoration on the lower part of the
bowl consists of a crowned head, six times repeated. The
pose of this figure and the hooked design of the border point
to a close relation with the style of Tiahuanaco.
Pl. 5, fig. 6 a and b: Vessel shaped like a mountain cat,
front and side views. Mus., 26745; } n. S. The animal is
O O
29)
SOLO
m
SSSSSSS
Fig. 19. Design of cup, Epigone period. i n. s.
Fig. 20. Decorated cup, Epigone period. j n. s.
Fig. 21. Design of vase, pl. 5, fig. 5. i n. s.
be noticed; it suggests that the artist had in his mind a face
in relief when he painted this, such as are shown in Kult. u.
Ind., I, pl. 11, figs. 1, 2, from Tiahuanaco.
Another cup, 26710, Mus., found with the vessels above,
is of a somewhat flaring shape, fig. 18 a. The design, like
fig. 18 b, is twice repeated. The same head as that in fig. 16
quite realistic in modeling and coloring. The spots marked
by curves and the two colors of the tail leave no doubt that
the animal is meant for a mountain cat. Its name in Aimarà
and in Kechua is titi’; and the island of Titicaca, “Titi
qaqa,' rock of the mountain cat, derives its name from this
animal, which may to some extent explain its mythological
importance. The coloring of this vessel presents several
peculiarities: the pelt of the cat does not appear to be joined
in the back, so that the tail was painted on each side of the
animal, both sides meeting at the back; the animal stands on
four legs of crude shape and has two arms in addition thereto
:
1 Compare the rapid motion of the bird in pl. 4, fig. 4 and fig. 19 below;
also the mantles of the side figures in the relief of the Gate, Stübel u. Uhle, pls.
11-15, which appear to be blown backward by the wind. It is characteristic for
the divinity to be always pictured in repose, while its ministering servants are in
violent motion. It must be assumed that the relief of Tiahuanaco formed the
prototype for all these designs, or that the position of the divinity was repre-
sented according to a general conception entertained of it by the entire people
alike; in this case the mythological significance of the relief on the Gate would
gain considerably in historical perspective.
2 A vessel of bowl shape of the same period with similar faces is shown by
Reiss und Stübel, pl. 96, fig. 7.
3 Comp. Stübel u. Uhle, pl. 8.
4 Stübel u. Uhle, remarks pl. 8, No. 1; above, pl. 4, fig. 4.
(27)

on its forepart. The peculiarities of this object are to be
explained thus: the maker had two models in his mind—one
painted and the other modeled; the former, an animal resem-
bling a cat, with hands at its forepart instead of feet, in
accordance with its mythological character, and similar to
the condor on vessel, pl. 4, fig. 4. Among the objects from
Tiahuanaco, too, are found some with the cat type in their
Pl. 5, fig. 11: Vessel in the form of a seated animal
found like fig. 1. Mus. 26750; } n. s. The animal is repre-
sented in relief, all but the plastic head, and is colored dark
red, wherever it is not engraved, as on the face, for instance,
while the connecting parts on the body of the vessel show
the undecorated surface of the clay. This specimen cannot
be exactly placed in any of our rubrics, although it was
found together with a number of objects of the Epigone
style. It grows more and more apparent that contemporane-
ously with the development of the archaic style of Tia-
huanaco in the highlands, another style may have developed
elsewhere in Peru of which this present object may be a speci-
It may be said that the familiar Chimu pottery pos-
sesses some suggestive resemblances to this object, but at the
same time in other respects very strong deviations.
Pl. 5, fig. 8: Vessel with conventional design.3 Mus.
26755; } n. S. The design, divided into sections, shows a set
pattern composed of rings. Objects with a similar design
and with an additional decoration of some plastic small ani-
mal or human figure were numerous in the graves of this
period; technically, these designs have some resemblance to
those of the Epigone type. Although the character of this
period is that of a general deterioration from the style of
Tiahuanaco, there are some variations in the stylistic changes
that may be defined specially: the figure designs in pottery
Fig. 22. Skull-shaped vessel from Chimbote. n. S.
;
decorations, with hands instead of forefeet, where the maker
had sought to express the mythological character of the ani-
mal by supplying it with human hands. The peculiar two-
sided painting of the pelt with two tails originated from the
attempt to make the flat painted picture of the animal distinct
on both sides. Additional features clearly pointing to a
dependence upon the Tiahuanaco style are to be seen in the
indication of joints on the arms, and in the scroll pattern
VIII
3
O
O
Fig. 23
Fig. 25.
Fig. 24.
Three stages of degeneration of a bird design on objects of later period. n. s.
a
a
bordering the rim of the vessel. This specimen was found in
a grave, together with an exact counterpart.
Pl. 5, fig. 10: Vessel with two openings, a serpent; found
with fig. 1, etc. Mus. 26571; } n. S.
를 ​An object bent like a
horn; its open mouth shows fangs, and its body is decorated
with white dots upon dark lines on both sides of the creature.
This characteristic pattern occurs on various plainer vessels of
the period, and may be supposed in most cases to represent
the design of the snake skin.
Pl. 5, fig. 9: Vessel in the shape of a cat.
Found
together with fig. 1, etc. Mus. 26752; } n. s. The head
shaped like a ball has a shallow square mouth, a rather flat
nose, and deep cavities for the eyes, which like the mouth
have lines in them to indicate teeth and eyelashes. In its gen-
eral character, and especially in the deep-set eyes, the vessel
resembles those in the shape of skulls, one of which the
writer found in fragments in a cemetery near Chimbote, fig.
22, Mus. 34334. The maker of the above vessel had prob-
ably a model of this kind before him, although the shape of
mouth and ears shows that he intended to model a cat's head,
and by the details of decoration the resemblance to a skull
was lessened still more.
degenerate in various ways, mostly by shrinking into geomet-
rical patterns. The style of Tiabuanaco, among others, uses
the figure of a bird spreading about the neck of a vase. This
figure by progressive degeneration changed to patterns like
those of pl. 4, fig. 4, or figs. 23–25. Objects with this deco-
rative type were numerous in the above-mentioned graves.
Another class of vessels is decorated with textile pat-
terns, wherein the variation of the figure pattern of an earlier
period is still traceable in the main outlines. To this class
belongs:
Pl. 5, fig. 7: Cup with polychrome decoration. Mus.
26792; } n. s. The clay of this object is granular and brittle,
and the shape is inferior ; the decoration is arranged in sec-
tions, each of which is filled with an irregular design and a
textile pattern, the figures of which were changed into scrolls ;
the design is the same as that of pl. 6, fig. 9. Many vessels
are decorated with typical textile patterns without any mate-
rial change; meander borders are frequent ; fig. 26 shows
one of these which runs like a band around a plain vessel.
A large number of objects decorated with textile patterns are
only represented by many fragments, having been destroyed
by later burials; among these are parts of bowls and vases
with cylindrical necks, the lower parts of which had the shape
of birds, and terminated in a point shaped like a bird's tail.
a
a
1 Comp. heads resembling the puma in the Gate relief, the mythological
significance of which may be similar to that of the mountain cat. Comp. Fran-
cisco de Avila, p. 128.
2 Cf. Stübel u. Uhle, pl. 21, fig. 3 ; also pls. 11-15 and present report, pl. 4,
fig. 2.
3 Attached to this vessel is a thin string with which it was found.
4 Kult. u. Ind., I, pl. 11, figs. 9 and 9 a.
(28)

a
a
Some of the fragments with these typical decorations are
shown in figs. 27 to 29; another ornamental type to be found
among these vessels is an inverted rhomboid figure, as in figs.
27 and 29, while fig. 30 shows the same design applied to
textiles. The two vessels, pl. 5, figs. 12 and 13, were, for this
cemetery, in a certain sense, the forerunners of the subsequent
period. Their light ground, which allows of no coloring
except in the three tints, white, light red, black, as in fig. 12,
also the technique, justifies their classification with objects of
the next period, yet certain points connect these vessels with
the group found in the graves here described. The locality
in which each object was found is a point of importance,
and helps to interpret the same. Fig. 12 was found at a great
depth ; fig. 13 was taken from the same grave with the vessel
of the Epigone period, pl. 5, fig. 5. The same degree of
excellence of technique, as shown by figs. 12 and 13, was
never seen in objects from graves of subsequent periods.
Both these specimens show a textile design with a conven-
tional animal motive, similar to that of pl. 5, fig. 7, which is
quite general in the textiles of that period, while only in a
single instance did it appear upon an object of the later period.
Pl. 5, figs. 12 a and 12 b: Large object, a cat ready to
jump. Mus. 26744; \ n. s. The head of the animal, show-
ing mouth, teeth and tongue, is quite naturally formed, while
the body is of pear shape; the position of the feet is charac-
teristic, and the paws are very well formed. The ground
color of this specimen is white, with a textile pattern in red
and black upon it.
Pl. 5, fig. 13: Vase decorated with a human face upon
us the different technical procedures nor the treatment of the
decorative elements. Even now we would know but little
more, as the only specimen, a cotton cloth with a design
painted upon it, gives little additional information ; but as
the pottery of the classical period gave clues for the inter-
pretation of the pottery of the Epigone style, so we may
interpret in a similar way, but in inverted order, the tech-
nique and decorative types of the textiles of the earlier period
by that of the subsequent epoch, and we find that these
fabrics are really the oldest existing specimens of the tex-
tile art of ancient Peru, of which we are able to fix the
period, not including the tapestry poncho among woven
fabrics. A number of stuffs were found at Ancon, all of the
same class and period, but the circumstances of their discov-
ery were not as favorable as those of Pachacamac to the deter-
mining of their age, which we now know as being that of the
period of transition between the style of Tiahuanaco and that
of the coast, and they were at first not valued as much as
they are now that their high antiquity is recognized. The
high degree of perfection of the textile art during the
Epigone period, which is only equaled by that of the later
period, the coast style, permits us to form conjectures as to
the high qualities of the weaver's art of Tiahuanaco, the ear-
liest cultural period of which we have any knowledge. Thus
the textiles of this period help to support the theory expressed
before : however remote the age of the Tiahuanaco style and
however isolated its position at the beginning of the develop-
ment of Peru, it must have been preceded by an immeasur-
ably long period of cultural development, of which, as yet,
RII
so
LUN
VILUTTAMIST
Fig. 26. Mus. 26748. } n. s.
Fig. 27. Mus. 27596 g.
Fig. 28. Mus. 27596 f.
Fig. 29. Mus. 26863 a.
a
its neck, and a rattle in its hollow base. Mus. 26807; } n. S.
The plastic face wears a peculiarly grave expression, and a
black decoration is painted upon its cheeks like a beard ; the
body of the vessel shows four conventional animal heads upon
a light ground, similar to figs. 15 and 20 above, or upon ves-
sels like pl. 7, fig. 1, belonging to the subsequent period and
to be described below. The open work of the hollow base,
which is filled with balls of clay, shows a textile pattern
resembling those of fig. 7 and fig. 12 of pl. 6.
B. TEXTILES.
we have no knowledge, although there may still be traces of
it in existence.
We probably do not possess specimens of all the different
technical types of the textile art of the Epigone style, but we
can clearly define the following groups from the pieces now
known to us:
1. Tapestries.
2. Plain stuffs.
3. Figured stuffs, the patterns of which are produced by
coloring: a, the figures painted upon the stuff, pl. 4, fig. 1; 6,
the patterns obtained by dyeing in such manner that spots
which were not to be touched by the liquid dye were drawn
tightly together, fig. 31, below.
4. Figured stuffs, the patterns woven in the cloth: a, by
warp threads of special colors, which were according to the
pattern either woven in or left loose, pl. 6, fig. 14.
5. Stuffs with embroidered patterns, pl. 6, figs. 1-4.
6. Woven braids : a, flat braids, pl. 6, fig. 9; 6, tubular
braids, resembling the double stuff and having a round woof,
as in certain narrow head bands.
We possess but a small number of well preserved textiles
of this epoch: some fragments of ponchos, a few forehead
bands, a belt, etc., not enough to furnish a complete picture of
the decorative designs of this period, but of great archæologi-
cal importance. The ceramic art of the period of Tiahuanaco
is sufficiently well known in its various types, but the knowl-
edge of the textile fabrics of the same age was for a long time
limited to a single specimen, the valuable tapestry poncho
found at Ancon by Reiss and Stübel. This piece of very fine
quality testifies to the high degree of perfection reached by
the weaver's art at that early period, which rivals that of
contemporaneous ceramic art.
The peculiarities of the textile art of that period were
not to be learned from this single specimen ; it did not teach
3
7. Various kinds of braided work, such as: a, a finishing
edge for tapestries forming a band of braid, wherein at regular
intervals the loose ends of the woof are caught together, pl.
6, fig. 7, this edge may be seen on several tapestries of this
period; 1, irregularly shaped braids, one of this class repeats
the pattern of the stuff in its doubly edged outlines ; c, tassels,
pl. 6, fig. 9.
1 See pl. 6, figs. 9 and 12.
2 Reiss u. Stübel, pl. 49.
3 The same technique is seen in the string of a wallet, pl. 18, fig. 13, but
the design seems to be of an earlier type.
(29)

Certain types of vessels were described as presenting
important features for the study of the ornament, and there
are tapestries and embroidered stuffs which are valuable for
the same reasons. Both classes of specimens are decorated with
mythological figures, whereby the purpose of these articles is
indicated. The tapestries are of a superior, more artistic type
than the other stuffs, and seem to have been used for robes of
ceremonial character. Embroidery appears upon many stuffs
of a superior class and the designs are mostly of the mytho-
logical order. The dead were occasionally buried with small
pieces of stuff which apparently possessed the character of a
fetich. A small number of these pieces were found, some of
them embroidered with simple mythological figures; the
pieces were laid upon the head of the corpse or fastened to
the shroud; pl. 6, figs. 1-4. These embroideries are square,
or in strips, and seem to have been cut from larger pieces
of stuff before being embroidered. The smallest specimens
measure five and a half by three inches ; the strips are as long
as twelve and a half inches. The patterns consist of human
or animal figures, double or single, or several figures all alike
or different, and cover the pieces closely to the very edges,
which in some examples are neatly trimmed off, so that one
might be misled to take the piece for tapestry. One or more
additional pieces were found elsewhere in the mummy pack-
age, disposed in various ways, in workbaskets, in pouches, or
a
29
bo
with loose threads covering larger surfaces belongs exclusively
to the last pre-Spanish period. The foundation for these
embroideries was a cotton cloth of grayish or natural brown
color, and the embroidery seems to be an imitation of tapes-
try. Two or three warp threads of the foundation are caught
together in each stitch, and covered closely with colored
woolen threads; the closeness of the stitches and the rep-like
appearance of the surface make the resemblance to tapestry
a very good one. The same effect was produced here in a
similar way. While the tapestries remain the same through all
the stages of cultural development, these peculiar embroideries
belong to this period exclusively.
Pl. 6, figs. 1-4, reproduce a number of these embroidered
specimens which were used as fetiches.
Pl. 6, fig. 1: Piece of stuff with a sceptre-bearing figure,
all-over embroidery. Mus. 29647 ; } n. S. The figure in
this piece (fig. 30) stands in the usual frontal position, and
fills the square space evenly; this, in addition to the shape of
the hands and the bearing of two sceptres, are strong points
of resemblance to the figure on the Tiahuanaco Gate; also the
two animal heads at the sides of the belt connect it with the
same. The divinity here seems differently characterized by
the branching, forked headdress, the two-toed, clawlike forma-
tion of the feet, and the ramified ends of the sceptres; the
border around the figure shows birds with upstanding wings
and long beaks.
Pl. 6, fig. 2: Embroidered strip. Mus. 29654; } n. S.
The strip measures twelve and a half by one and five-eighths
inches. It shows four animal figures, which closely cover it;
two condors in flying and running motion, and in opposite
direction two winged figures, position, full front. These figures
have a rhombic head, eyes and mouth, with ribbon pendants,
triangular wings and tail; the claws of one are turned down-
ward. At the first glance this creature resembles an owl,
although it is strange that it should be represented flying, and
in a full front position. Two rows of teeth can be seen in the
beak, which might suggest some flying animal, a bat for
instance. That the animal here is represented with a tail is
no reason to object to this interpretation, as this primitive art
possessed only a few motives, of which the most daring combi-
nations were made, and strange or fabulous animals pro-
duced. The bat is frequently represented in the decorative
art of northern Peru, and its presence in this design is quite
plausible.
Pl. 6, fig. 3: Animal with curled tail, figure upon an
embroidered strip. Mus. 29653; } n. S. The same figure is
repeated twice upon this strip of twelve and a half by three
inches in size; one is in red and green, the other in black and
yellow. The animal, pictured with realistic accuracy of detail,
is seen in profile, in upright, sitting position, with raised fore-
paws, one of which only shows a large claw; the upper jaw is
triangular, enlarged toward the front, and the open mouth
shows the teeth ; the ear on the back of the head is long, and
curves upward; the long tail is laid upon the animal's back,
and its tip has also an upward curl. This figure is repeated
on several of the strips, but it is hard to find an interpreta-
tion for the creature in spite of its very distinct characteris-
tics. The teeth prevent its interpretation as a native rabbit
EUⓇ
Vale
es
1949
tunnin
SL
a
NINA.
man
Fig. 30. Complete design of embroidered fragments of the cloth,
pl. 6, fig. 1. } n. S.
5
as in some instances, used as coverings for the cushion of the
false head of the mummybale. Several pieces were found with
the same design on all, while similar figures also appeared
upon larger cloths that had served as cerements for the dead.
The small embroidered pieces may have been cut from these
in order to be used as fetiches, or the significant symbols upon
the larger cloths may explain their use as shrouds. The
pieces in either case may be regarded as having the signifi-
cance of fetiches. The designs of these embroideries are use-
ful in throwing light upon the mythological ideas of those
Indians who buried their dead with them. The figure of the
design represented the divinity which protected them in life.
These embroidered figures also mark the beginning of the
development of decorative designs and are the key for their
interpretation, as was shown before in the case of the deco-
rated pottery used for ceremonial drinking at religious festi-
vals. Thus we see that the designs of these embroideries
were based upon religious ideas, expressed in the most primi-
tive forms, deviations and changes from which caused their
progressive development. In point of technique these
embroideries form a class of stuffs entirely representative of
this Epigone period, while in the same sense the embroidery
2 Brown cotton is often found in graves; new cotton of a light reddish
color the writer saw at Lambayeque.
3 The figure of the small piece from Ancon, Reiss u. Stübel, pl. 53, fig. 1, is
almost identical with the above, but for the knives in its hands, which lend it a
greater resemblance to the Tiahuanaco figure.
* Teeth are indicated in the face of pl. 6, fig. 7, which was derived from the
above figure.
5 Comp. Stübel u. Uhle, p. 65, note 16, and the fishes in the relief of Tia-
huanaco, l. c., pls. 9, 10.
6 The Maya in Central America frequently made use of bats as ornamental
figures.
1 An embroidered piece from Ancon, Reiss und Stübel, pl. 53, fig. 1, is there
described as tapestry; it resembles the specimens here described and belongs to
the same period.
(30)

a
a
а
(viscascha); the shape of the mouth and the position of the
paw do not resemble those of a cat, nor is its general
appearance that of a fox or a dog.
Pl. 6, fig. 4: Animal with a curled tail; design upon an
embroidered strip. Mus. 29651 ; } n. s. The strip is eleven
and a half inches long, and two and a half inches wide. The
figure in the design is twice repeated in different colors, a
simple textile ornament and repetitions of a conventional ani-
mal head separate the figures; only one of them is shown in
the present view. This animal resembles closely that of fig. 3,
but the face here is seen full front, and one forefoot and one
hindfoot are turned upward. The forehead is divided by a
line and the ears are also indicated. The animal must be of
the same species as that of fig. 3; the curled tail here is less
apparent, as this piece was cut out and sewed into another
piece of the stuff; the wide mouth shows many teeth, whereby
the possibility of its being a rabbit is strictly excluded.
Pl. 6, fig. 5: Sceptre-holding figure in tapestry; all-over
design of a large piece of stuff. Mus. 29492; } n. s. The
piece is eight feet four inches long, and consists of five strips,
each of which is eight inches wide, joined together lengthwise,
repeating the above figure as an all-over pattern; it has sixty-
two woof threads by ten to twelve of the warp to the half
inch. The white woof and warp-threads are cotton, the col-
ored ones wool. The derivation of this design from the types
of the Tiahuanaco style is shown by several features: by its
adaptation to the square space, by the exact repetition of the
figure, which is seen full front, and by the staves which it
holds sideways, instead of sceptres. The pronged crown
resembles ornaments of the figures in the stuff, pl. 4, fig. 1c;
the ornament between the legs is a repetition of one on the
main figure of pl. 4, fig. 1; the body seems to be clothed with a
poncho. This stuff is a characteristic example of the deterio-
ration in art since the period of Tiahuanaco; the color scheme
is not good, the design of the figure lacks style; it is uncer-
tain, changing from square to square and often appearing
awkward in its details. The design does not cover the
given space, as is shown by the shortening of the legs,
the changing of the sceptres into staves, and by the wing-like
additions.
Pl. 6, fig. 6: Square cloth, sixteen and a half inches, in
double stuff with reversed animal figures. Mus. 29701 ; }
The materials which form the design are one of red
wool, the other of white cotton ; the complete design being
in three rows, each with four repetitions of the upright figure
and of the same inverted. The design is finished like an all-
over pattern on both sides, while at the lower edge appear the
upper parts of the animal figures resembling bats.
Fig. 7: Round belt of tapestry, probably a young child’s.?
The little belt is seven inches wide, and four and a half inches
high, closed by a seam. The colors are bright and pretty ;
the ground is red, the head of the principal figure blue. The
blending of colors is soft; black outlines relieve the figures
from the background, as if in imitation of a painted pattern.
A meander in red and blue woven into the stuff forms the
border. Besides the two repetitions of the principal figure, a
bat, there are three rows of zigzag gradations; within the lat-
ter are the head and leg of a small figure in reclining position.
The ground about the same is divided into strips resembling
braid similar to that described above, No. 7 a. Two small
human heads with a feather headdress stand facing the prin-
cipal figure.
Fig. 8: Design of a fragment of tapestry, twelve and a
half by seventeen inches in size. Mus. 29680; } n. s. To
the upper edge of this piece was sewed a band of plain cotton
one and three-eighths inches wide. This fragment seems
to be the ornamental part of a loin-cloth. The stuff is
finished on the side with strips of tapestry showing a cross
pattern in diagonal stripes regularly repeated. The ground
color is purple, while the design is in various colors such as
white, dark purple, green, light red. The figures are outlined
with black or yellow.
Fig. 9: Woolen forehead band with patterns in various
groups; found with the stuff, fig. 5. Mus. 29686 ; } n. S.
The length of the band is five feet eleven inches, the width
one and eleven-sixteenths inches; the tassels are six and a half
inches long. Two inches of the length of the band are divi-
ded at each end, for the attachment of the four tassels with
which they are ornamented. The band is scarlet, except in
seven sections, where colored tapestry patterns, one and one-
eighth to five and five-eighths inches in length, are inserted,
and many cross-stripes, woven in a light pattern, serve as
additional ornamentation. Each of the seven tapestry pieces
is in a different pattern, a-f, and in various colors, such as
green, light red, dark purple, yellow, blue and white.
Fig. 10: Design of a forehead band of tapestry. Mus.
29690; } n. s. The strip is two and a half inches wide and
has a pattern with a figure resembling a star with eight points
set into square sections, in the corners of which are small
crosses, peculiar to this period.
Fig. 11 : Part of a sleeve with narrow tapestry border.
Mus. 29684; } n. s. The border is only one inch wide
and closed in ring shape by a seam, the full length of the
sleeve is fourteen inches. The pattern is formed by square
sections with alternating colors, in each of which is the figure
of a seated animal with curled tail, like the figure of pl. 6,
fig. 3, but in its heavy foreparts it has no resemblance to any
familiar American animal.
Fig. 12: Fragment of a small closed belt, like fig. 7, of
tapestry. Mus. 29682; } n. s. The belt is only two and a
half inches wide and six inches long. The ground colors
are dark red and white, the pattern in red, light brown,
green and black, blending softly with the ground colors.
Fig. 13: Design of a forehead band of tapestry. Mus.
29685 ; } n. s. The band is four feet five inches long and
two and a quarter inches wide. Both ends are divided into
three strands of three and three-quarter inches, and are fin-
ished with tassels five and a quarter inches long. The band
is divided diagonally into twelve broad triangular sections,
each half of which repeats a different design. The point
where both patterns meet is shown in fig. 13. The colors are
bright—white, green, purple, yellow and black.
Fig. 14: Embroidered design upon a plain cotton cloth,
shroud. Mus. 29663; } n. S. The pattern is formed by fig-
ures nearly square, one and three-quarter inches each side, in
chessboard design. The figures of each are the same, but the
colors are different, each color being continued in a diagonal line.
Fig. 15: Short band or sash with wide ends of tapestry.
Mus. 29668 ; } n. S. The connection of the pattern in this
piece with those above described is quite apparent in the spur-
shaped figure in the narrow cross-strip,4 and in the slanting
n. S.
a
a
a
;
a
a
3 These garments were identified as loin-cloths, the same still being worn
by the mestizos in the valley of Lurin, who easily recognized those found at
Pachacamac as similar to their own, which they wear when they go fishing. The
ornamentation of a wide piece at one end of the cloth explains itself from the
way in which the garment is adjusted upon the person of the wearer.
The long
gowns of the side figures in the relief of the Gate of Tiahuanaco were probably
cloaks (yaqalla), like the supa yaqalla worn by the young men of Cuzco at
their initiation into the nobility (Molina, p. 36), and which were fastened around
their necks.
4 Comp. pl. 6, fig. 7.
1 Comp. the figures of the three rows in the Relief of Tiahuanaco; also
description of pl. 4, fig. 2, and Reiss und Stübel, 1. c., pl. 49.
2 In the belt, pl. 7, fig. 4, were found the skull and bones of a young child.
(31)

1
а
crosses which resemble the cross held by the bird in its beak
in pl. 4, fig. 4. The conventional animal figure with hooks
differs somewhat from the preceding, which shows that vari-
ous stylistic tendencies might exist side by side, while so far
we have little knowledge of their origin.
Fig. 16: A woven pouch for spindles, such as is usually
found with all its contents standing beside the mummy.
Mus. 29733; } n. S. The pouch is twenty-six and a half
inches long and three and seven-eighths inches wide, and
consists of a piece of stuff closed along one side and at
the lower end, the upper being open, tied with a cot-
ton string. The cotton stuff of which this pouch is made
has mostly red woolen warp-threads with a few white cotton
ones and a woof of common brown cotton. While the
two ends of the stuff are woven like plain linen, the cen-
tral part shows a design which was produced by skipping the
white threads of the warp in places and leaving them loose
on the reverse side. Thus a complementary design of a diago-
nal meander was made to cover the entire space, in one part
of which the red threads are mixed with the white, in the
other unmixed. This design in the same color is seen in
many textile fabrics of this period, on pouches like the
above or on cloths which cover the false heads of mum-
mies, and also on a poncho. The same design and technique
appear also in different colors, in green and blue, on similar
spindle pouches. The same method is sometimes applied to
produce different designs, such as chessboard patterns, etc.
being joined together with the complementary sections and
forming a new pattern in this combination ; there are violet
and yellow sections with red lines, light and dark blue sec-
tions with white ones. This design was produced by drawing
the stuff tightly in before dyeing, first the red design on violet
ground and then the blue; the red patterns on yellow were
done by hand upon the plain ground.
Although the textiles here described present a great
variety of typical features, the series is by no means a com-
plete one, and many types are found side by side which have
no relation whatever to each other. It is, therefore, not possi-
ble at this moment to furnish a complete survey of the tex-
tile art of this period, neither from the technical nor from
the decorative point of view. However, it can be said, that
the patterns of a majority of these stuffs have been inter-
preted. A number of ornaments, formerly unexplained, are
now understood, and the certainty of their dependence upon
the Epigone period established. The same is assured of
numerous textile patterns found in graves along the coast,
whereby also their descent from the Tiahuanaco style seems
to be proved. The decorative designs of this period, both in
textiles and pottery, are closely related ; they continue orna-
mental features of the style of Tiahuanaco and have a num-
ber of these types in common, while both show the same
tendency of conventionalizing figure patterns into simpler geo-
metric patterns, which is a peculiarity of this period. This
degeneration of the design occurs more frequently in textiles
23
a
Daklee
0 FOEDDE
FE
303OO
Fig. 31. Stuff composed of pieces, woven and dyed
separately. Epigone period. Mus. 29782. {n. s.
Fig. 32. Embroidered human figure. Epigone
period. n. s.
Fig 33. Design of a thick braid-like stuff. Epigone
period. n. s.
than in pottery, since the technique itself would prescribe it.
The products of this ornamental development are taken from
one technical department into the other. Thus certain pat-
terns of importance for pottery are only to be interpreted
after being found in textiles. The discussion has shown that
in textile art four ornamental designs are most frequently met
with :
1. The full front figure of the divinity, holding sceptres
or staves.
However plain the design and technique of these stuffs and
numerous the articles made of them, they still belong exclu-
sively to this oldest period, to judge from the objects with
which they were found.
The stuffs that were dyed after being woven are repre-
sented by a special class of garments, one of which was found
upon a mummy under a poncho of pure Tiahuanaco style.
The rich and harmonious blending of colors, peculiar to these
stuffs, proves that in style and technique they are closely
related to the art of Tiahuanaco, while at the same time they
seem to belong to a more advanced stage of development, to
judge from the details of the pattern which among others
shows rhombic figures, like figs. 25 and 27. This class of
stuffs are composed of a number of small pieces, which were
joined together after being dyed. The outlines of these pieces
occasionally are in hooked shapes, so that two sections form
one whole square. The joining together was done at the end
of the warp by catching the end loops of adjoining pieces with
a woof-thread, so as to give the appearance of having been
woven in one, although the pieces were really sewed together
along the edges. All of these procedures show that the
makers of these stuffs had long passed the early stages of
the weaver's art. The patterns upon these pieces are either
lineal designs or rhombic figures, like fig. 31.
In this instance the pieces had a double design before
2. The bird in running and flying motion.
3. The animal with curled tail.
4. A winged figure resembling a bat.
The two former are found identically the same in pot-
tery. The third explains some of the most important orna-
mental figures. The dog-head of the fourth class occasionally
appears in ornamental derivations upon pottery. The figure of
the first class also occurs without the sceptre, as in fig. 32, of
which a copy exists also in an embroidered stuff. Figures
resembling those in the textiles from Ancon, Reiss u.
Stübel, pls. 50 and 69, fig. 1. Figures with staves in their
hands on some small grave-tablets, copied I. c., pl. 33 a, are
without doubt founded upon the same original design. The
second typical figure can only be identified in designs derived
from it, although it probably has not entirely disappeared
during the subsequent cultural stages. The bird pattern
2
1 Comp. also Reiss u. Stübel, 1. C., pl. 86, fig. 2.
Comp. figures pl. 6, figs. 1 and 5, with that of pl. 5, fig. 2; the bird in
running motion of pl. 6, fig. 2, with that on vase, pl. 4, fig. 4, and fig. 19, above.
(32)

5
a
on the band of pl. 6, fig. 1, fig. 30 above, may be derived
from it.
The animal with curled tail is one of the commonest
ornamental motives during the early periods of Pachacamac.
Only in textiles it appears frequently in the entire figure. It
may be said to resemble the skunk (Mephitis sp.) which was
worshiped at Pachacamac in very early times, 2 but it is not
possible to identify it distinctly as such. The head of this
animal appears in a rather naturalistic treatment upon a small
strip of brown and white double cloth, fig. 33.
The head of the ornamental creature in pl. 6, fig. 6, is
similar. The animal may have been intended as being seated,
as in pl. 6, fig. 3. It is strongly conventionalized, while two
and two of these figures are interlocked in a complementary
pattern. More advanced to a similar completion is the design
in a, pl. 6, fig. 9, where the extended forepaw is still distin-
guishable. The heads with enlarged noses, above, fig. 15 and
fig. 20, belong to the same creature; possibly also the heads
of figures otherwise of human shape, fig. 13. The triangular
enlargement of the nose in the animal, pl. 6, fig. 3, developed
from the former and has a parallel in the decorative figure of
the vessel, pl. 5, fig. 13, but here the back part of the head
is more developed into a symmetrical shape, beginnings of
which were to be observed in fig. 15 and fig. 33 above. In
this form the head continues as an ornamental type during the
later pre-Inca period and appears on specimens of pottery.
In the Tiahuanaco style two animal types are familiar,
both with similar tails, one representing the puma, while the
alized ears. The square head without any side attachments
appears in b on the forehead band, pl. 6, fig. 9, between the
bends of a meandering ribbon pattern. It also forms cross
patterns in many different combinations, as in the stuff from
Ancon, copied by Reiss u. Stübel, pl. 55, fig. 3 ; stuffs with
similar patterns were taken from graves at Pachacamac.
Each of the two designs on the forehead band, pl. 6, fig. 15,
shows one-half of the whole figure of the creature. The
design a shows the head, as in fig. 37 b above ; b shows the
body formed like a stick, with its wings. The design in fig. 8
was composed of the head in the original form, as in 37 a. It
was further developed into a complementary design in f upon
the forehead-band, pl. 6, fig. 9. By the combination of two
heads of fig. 37 a and by interlocking the hooks of the two,
the design a of the same band was formed. A further devel-
opment is shown by g on the same plate and by the design
upon the belt, pl. 6. fig. 12. The form b appears also at the
ends of a meander joined on under c in pl. 6, fig. 9. A later
conventional form of the design is seen in pl. 6, fig. 4, where
the triangular head is used to fill a small square.
C. WOODCARVINGS.
a
The woodcarvings of this period are superior to those of
the following epoch in artistic execution and in a greater
variety of design. The objects upon which the carving
appears are the same as in former periods, such as pointed
staves, weaver's reeds, etc. The material is usually dark-
o
2
Fig. 34.
Decorative design of animal on cup. Tia-
huanaco style. § n. .
Fig. 35.
Fig. 36.
Decorative heads of animal (dog) on vessels. Tia-
huanaco style. n. S.
Fig. 37 a.
Fig. 376.
Typical forms of animal heads in textile patterns.
Epigone period.
are
other has the head of a dog, figs. 34 to 36. Treatment and
pose are different in both, while they must be regarded as
typical ornamental figures of Tiahuanaco, even if we
unable to prove their identity. From the bat figure, as it is
seen in pl. 6, fig. 2, developed a number of different orna-
ments, as may also be seen upon textiles from Ancon.
A very good likeness of the creature, although conven-
tionalized, is to be seen on the belt, pl. 6, fig. 7. The body
here is represented by a straight line, and the mouth does not
show the teeth. Five toes and a claw turned downward seem
to be indicated at the ends of the wings; the tail also has feet
with toes and claws, which strengthens the supposition that
the animal is meant for a bat. The head and other parts of
this figure were much used as ornamental types. The head
developed into forms like the square head of fig. 7, and
into others like figs. 37 a and b. The former differs from
the original square one by its triangular shape and by the
hooked attachments. Although these may be purely orna-
mental, it is possible that they are meant for the hooks
at the ends of the wings in fig. 7. Fig. 37 b shows the
addition of bands at both sides of the mouth, which, how-
ever, occurred before in pl. 6, fig. 2, and may be convention-
brown wood. If the excellent workmanship in all specimens
of this period is an inheritance of the earlier epoch, this fact
must above all be accepted for the woodcarvings, which are
surprisingly realistic in form and pose. The highland prob-
ably was more productive in fine woodcarvings than the low-
land. A pretty specimen is the wooden ear-ornament, pl. 4,
fig. 5. Carved wooden vessels were common in the high-
land, and less so in the plain. Many carved objects were
made in Pachacamac through the influence of the highland
during the period of the Incas. The derivation of the types
of these carvings from the art of the highland, however,
can only in a few cases be asserted. Owing to the perish-
able nature of the material few specimens have come to our
knowledge, those of earlier periods having disappeared.
The ornamental types of these objects reveal less of relig-
ious character than is shown in the pottery and the tex-
tiles. An Indian lying upon his stomach in a lazy atti-
tude, as in pl. 5, fig. 14, and figures of captives with their
hands tied behind their back, can surely not be considered
as mythological subjects. Animal figures occur frequently-
lizards, birds, etc. Often we find small scroll ornaments
joined to their eyes, as in pl. 5, fig. 7; pl. 4, fig. 1, or
as in the figures of the Gate of Tiahuanaco. These orna-
2
1 Pl. 6, figs. 2, 3 and 11.
Garcilaso, B, VI, chap. 31, speaks of the worship of the 'Zorra'; and J. J.
v. Tschudi, Beitr., p. 127, points out that the Zorra was Mephitis sp. and not the
fox. In the myths of Huarochiri the skunk is mentioned, however, among the
animals cursed by the god Iraya (F. de Avila, p. 127).
3 Comp. the figures in Reiss u. Stübel, pl. 69, fig. 3, and the second in fig.
1, 1.c.
4 Comp. also Reiss u. Stübel, pl. 44, fig. 3, 55 a; figs. 2, 65, etc.
5 The type of a meandering band corresponds to that in the ornamental bor-
der of the frieze of the Tiahuanaco Gate, Stübel u. Uhle, pl. 17. In this case two
ornaments were joined to the ribbon and detached faces were set between the
bands.
6 Comp. as the original form of this type the principal figure in pl. 6, fig. 7.
7 Comp. the half-kneeling winged figure, Stübel u. Uhle, pls. 11-15.
(33)

n. s.
ments may possess some religious meaning, while at the same
time they form a certain link of connection with the art of
the classic period.
Instead of numerous examples, only two typical ones may
be mentioned here. Although these carvings are interesting
in themselves they cannot furnish much information concern-
ing the historical position of the period.
Pl. 5, fig. 14: Pointed staff with the figure of a man
lying down. Hard brown wood. Mus. 27784; } n. S. The
carved end of this staff shows the figure of a man, two and
five-eighths inches long, resting upon his stomach with feet
turned upward; the face is looking in a straight line and
rests upon the right hand, while the left lies flatly extended.
The man seems to be stretched upon a mat, the pattern of
which can be seen on the under side. A flat oval object, pos-
sibly a coca pouch, fastened with a band, is indicated upon the
the religious element predominates in them, and three small
objects may serve as examples.
Pl. 5, fig. 16: Animal, flat, carved in bone. Mus.;
The animal is of the well-known type with the
curled tail. The manner of suspension by the head shows
that it is represented as if seated, with its legs extended in
front, as in fig. 3, pl. 6. The tail may have been long, but is
broken off. The ring pattern all over the animal's body
might indicate that it was meant for a cat.
Pl. 5, fig. 17: Animal, winged and in running motion,
carved in shell; n. s. The creature is of a mythological char-
acter, similar to those on pl. 4, fig. 4, and pl. 6, fig. 2, a bird
with head and tail of a quadruped.
Pl. 5, fig. 18: Tooth of a mammal, carved at the root-
end into an animal's head and engraved around the centre.
It was suspended by the hole bored through the midst. Pieces
.
*
a
ODACO
Q
Figs. 38–48. Spinning Whorls.
*
of this class are numerous in this section of the cemetery ; in
some specimens there is a head on each end.
Pl. 5, fig. 19: Small fish, carved in shell. Mus. 27852 c;
a
n. s.
E. SPINNING WHORLS.
head. Ornamental suggestions on the back may be intended
for garments; curves on the thighs and shoulders are meant
for the folds of the loin-cloth and the sleeve ends of the
poncho.
Fig. 15: A head with open-work headdress upon a
staff, similar to the preceding. Mus. 27782; } n. S. The
face shows the well defined features of a man; the headdress
is a circle with two prongs in front and an ornament above
it, three and a half inches high, of a trellis-like open-
work. Similar ornamentation is to be seen upon other carved
pieces.
裝
​The types of these whorls change according to the
periods, and they are numerously represented of all the
epochs; they therefore form a very suitable material for
the classification of graves.
Figs. 38–48 show several types of the spinning-whorls
of this period; figs. 38-40 are of bone; fig. 41 of shell ; figs.
42–44 of stone ; figs. 45-48 of clay. These specimens are in
the shapes of rings or balls or conical, occasionally they
appear in the form of ornamented disks. The stone whorls
are of a yellow or bluish-red stone, resembling soapstone.
The only ornament upon all consists of circles or plain lineal
engravings.
D.
ORNAMENTS OF SHELL, BONE AND TEETH.
A number of small carved figures and kindred objects,
which were worn as talismans upon the person, are typical of
this period. They resemble a large number of such articles
from the classic period of the Chimu. For obvious reasons
.
(34)

CHAPTER XII.
THE GRAVES IN THE NEW SOIL.
I n. s.
IT
T was demonstrated in the preceding chapter that the cul-
ture of the oldest and most untouched section of the
cemetery depended upon that early period to which the stone
monuments of Tiahuanaco owe their existence, while at the
same time was shown its degeneration from the high degree
of perfection of its prototype, both in technique and form.
The graves in the new soil represented the beginnings of
another period, in which was continued the retrograde move-
ment of the preceding epoch. These graves were about one
and a half to five feet below the surface, measured to the top of
the light rush-roofing; each roof as a rule sheltered several
mummies. When more soil was piled up above these graves,
the roofs could not withstand the weight. This accounts for
the finding of so few unbroken pieces of pottery in these
graves.
4
a
Pl. 7, fig. 2: Vessel decorated with cats. Mus. 26815;
Two vessels of this kind were found together in one
grave; the design of one showed four cats, that of the other
three. This motive was probably derived from designs of
cats, such as in pl. 5, figs. 6 and 12.
Pl. 7, fig. 3 : Cup, with crowned heads in its decorations
Mus. 26845 ; n. S.
n. s. The cup, of simple conical shape,
of a very porous clay, is crudely ornamented; the colors are
dull. The crowned heads seem to have a faint resemblance
to heads of the Tiahuanaco and the Epigone period, 4 from
which they may be derived.
Pl. 7, fig. 4: Vase, a seated human figure. Mus. 26906 ;
I n. s. The body of the object is round, with flattened base;
the head, too, is flattened and has a small opening; the legs
are marked in relief, while small handles represent arms. A
wide molded feather ornament surmounts the face and is
tinted in colors.
Pl. 7, fig. 5: Vase with decorations in white, red and
black. Mus. 26832; 1 n. S. The body is bowl-shaped, and
surmounted by a thin tapering neck. Only the upper part is
decorated, as in figs. 1-4. The design upon this specimen
shows diagonal lines encircling the neck, and plain zigzag
lines upon the body.
Pl. 7, fig. 6: Plate decorated with interlocked curved
looks, colored on the inner side. Mus. 26826; 1 n. S. This
plate was among the objects found in fragments, mentioned
above.
A. POTTERY.
B. TEXTILES.
The pottery of this period is inferior in form to that of
the preceding. The vessels are far less artistic, and consist
mainly of a class of large jars with round bodies, flattened at
the bottom, and having short cylindrical necks (see pl. 7, fig.
1). There is, too, a group of cups, flaring, conical, such as
pl. 7, fig. 3; also vessels like fig. 2; plates having a rim
underneath as a foot, fig. 6; jars like fig. 5, and lastly a few
vases with small necks of which fig. 4 is a type. The shapes
of all these vessels, especially of figs. 1 and 4 to 6, are entirely
different from those of the earlier period. The technical char-
acter of the vessels is similar to that of pl. 5, fig. 12; they
are made of porous clay, of crude shapes; the surface has
little polish, and the color looks like whitewash with red and
black figures. If there is any relation in some of the decora-
tive features with those of the earlier period, it certainly is
difficult to find, and the tradition seems to have faded.
Pl. 7, fig. 1 : Vase decorated with a human face. Mus.
26811; n. s. The decoration on the upper part of the ves-
sel and on the neck shows a human face surmounted by a
headdress. The pattern upon the body of the object is in
three sections, formed by vertical bands composed of inter-
locking hooks. The head of an animal appears in each section,
in red or white, similar to those in pl. 5, fig. 14, having a
tufted head ornament. This head resembles that of the ani-
mal with the curled tail of the Epigone period, while the
head ornament was described under No. 5, above, p. 26. The
interlocked hook pattern resembles a design of that earlier
period from which it was derived.?
The textiles of this period are represented only by a
small fragment of tapestry with the design of an animal, the
legs of which are turned into volutes, and by a large brown
cotton cloth of a gauzy thinness, with a red and blue tapes-
try border. This cloth resembles the poncho from Ancon,
(Reiss u. Stübel, pl. 40, fig. 1), both in color and in texture.
The design of the border of pl. 7, fig. 7, shows a repetition of
PA
2
Fig. 49. Design of tapestry border, garment of brown gauze. Later pre-Inca
period ; } n. S.
stiff animal forms. A gauze poncho, similar to the above, was
found in another grave of about the same period; this had a
border with a design like fig. 49. Both these figures are con-
nected by certain stylistic analogies.
1 See above, p. 19, fig. 3.
2 See above, p. 33.
3 See pl. 5, fig. 6, and figs 18 b and 21, above, p. 27. Its original use in the
early period was probably limited to the suggestion of a crown-shaped headdress.
Kultur u. Ind., I, pl. 11, fig. 6; Stübel u. Uhle, pl. 8.
4 Comp. pronged crown in pl. 4, fig. 1 c; pl. 6, fig. 5.
(35)
有
​
n. s.
C. METAL IMPLEMENTS.
The graves of this group yielded the first objects of metal
of any importance.
Pl. 7, fig. 8: Pair of copper tweezers in the shape of a
crowned human figure, seated. Mus. 28777; } n. S. The
figure seems to hold a staff with both hands; the joint of the
two parts of the tweezers is on this staff, while the edges on
the back of the figure served as tweezers; the feet are
damaged.
Pl. 7, fig. 9: Plain copper tweezers. Mus. 28779 ;
The most primitive tweezers in Peru were doubtless
made of bivalve conches; they are still in use for this purpose
with the Indians east of the Andes. Wherever metal tweezers
are not represented among the objects in graves, it is safe to
assume that natural sea shells were used instead. The deriva-
tion of the metal tweezers from bivalve shells may be seen by
the rounded shape of the implement, where it appears in its
earliest stage, while later the shape varied.
а
(36)

CHAPTER XIII.
GRAVES IN A SECOND STRATUM OF THE NEW SOIL IN FRONT OF THE TEMPLE.
cover.
SEVE
YEVERAL graves were disclosed in the course of our
operations, near the base, in front of terrace t (see
view 3, above, p. 19), possessing specific features of interest
in the peculiarities of their construction, and also in the
types of the mummies and of the objects recovered. These
graves were found in a stratum of adobe fragments extend-
ing underneath the front terrace of the temple. The depth
of these graves was about six feet less than that of other
graves, their base being only five feet below the surface. No
special features of construction, such as cells, or of roofing
were observed. The mummies formed bales of irregular
shape, wrapped in large square cloths, five feet by six in size,
and not sewed together along the edges. On a mummy bale
from Ancon (Reiss u. Stübel, pl. 24, fig. 1), otherwise similar
to the above, the edges were all sewed together on the outer
Only two mummies in a good state of preservation
were secured at this spot, and with them a large number of
textile fabrics. Small fragments of similar fabrics were found
mingled with the soil in this locality, showing that originally
there were more graves of this class here, but of different
periods of construction. Some of the crania were of a more
elongated shape than the usual type. The hair upon some of
them was clipped on the top of the head as closely as three-
sixteenths to three-eighths of an inch. The head ornaments
were different from any previously discovered. Several heads
were covered with spiral coils of black woolen cord of the
thickness of a finger. One skull was found with a narrow
white cotton bandage encircling the forehead. The hands,
too, were of an unusual size and of a light yellow color.
Among the bodies were some of women; one, judging from
the shoulder cloth, was that of a little girl; with the belt, pl.
7, fig. 14, were found the bones and skull of a small child.
There were also fragments of girdles, “chumpis,' and a pair
of shoes of an unfamiliar type, pl. 7, fig. 12. In addition to
the above mentioned there must have been three mummies of
A polished shell, bored for suspension ; pl. 7, fig. 17.
Two pairs of ear ornaments ; type of pl. 7, fig. 16.
Some seeds wrapped in wool.
A ball of fibre resembling reeds, used as twine.
A pair of sandals.
A long piece of black woolen braid, pl. 7, fig. 11.
A piece of round plaited braid, used as a binding on a
shoe, like pl. 7, fig. 12.
The ear ornaments are duplicates ; it is probable, there-
fore, that one pair came from another grave.
Grave b. A pair of shoes; pl. 7, fig. 12.
A loin-cloth; pl. 7, fig. 13.
A black poncho of vicuña wool; pl. 7, fig. 19.
A plain black woolen poncho with border of tapestry.
A fine cotton poncho.
A poncho with lineal pattern in openwork weaving.
Two common yellow cotton ponchos.
A brown cotton poncho.
Two large cotton cloths, one embroidered.
A woven belt; pl. 7, fig. 14.
A cotton cloth folded like a bandage.
Two small pouches with articles of food; pl. 7, fig. 18.
A pair of ear ornaments.
A shell (spondylus pictorum).
The contents of this grave may have been mixed with
those of several others; the seven ponchos, however, were
found together, between the arms of one mummy.
Description of objects taken from these graves :
1. The ponchos : most of the ponchos are of cotton, a
few of wool, but all different from any found elsewhere in the
cemetery ; they are long and close-fitting, instead of short and
wide. The measurements of some of the ponchos are as follows:
66
66
66
men.
Length, 323 inches ;
333
32
311
32
26
(imperf.)
31
66
width, 25 inches.
288
272
222
22
272
282
66
66
66
66
66
66
66
66
To convey an idea of the variety and number of articles
taken from these peculiar graves, a list of the contents of two
of them may be given here. The cerements of the bodies are
included in the number.
Grave a. A poncho with border of tapestry; design pl. 7,
fig. 20.
A poncho with woven ornaments; fig. 50, below.
Seven large cotton cloths, one embroidered ; fig. 52,
below.
A quilt.
Two pouches, in fragments.
A plate; pl. 7, fig. 15.
A closed shell (spondylus pictorum).
A half shell of the same species.
The length varies from thirty-one to thirty-four inches,
and the width from twenty-two to twenty-nine inches. The
cotton ponchos are of various kinds, some thin, yellowish, of a
plain cotton stuff, while others are ornamented with openwork
weaving. One has a woven border, Mus. 27542; the design is
in fig. 50. There are two horizontal stripes across the body of
the ponchos, each four inches wide, and embroidered in a
pattern similar to fig. 52, below. The cotton poncho, Mus.
27572, is of a very fine fabric. It consists of two different
materials, one being used as a facing in a strip twelve inches
a
(37)

а
wide under the other. The outside fabric is of orange color,
the other shows vertical gray and brown stripes. The cloth
is as fine as silk ; it has sixty-four threads of warp to twenty-
two of woof to the half inch.
Two ponchos, Mus. 27541 and 27543, found in frag-
ments, are different from the former, being of a thick, brown-
ish-white stuff, resembling felt; one has fifty-two warp-threads
to sixteen woof; the second seventy-four warp to fourteen
woof-threads to the half inch. The patterns upon these,
which resemble that of pl. 7, fig. 19, appear in horizontal
tapestry stripes as shown in pl. 7, figs. 20 and 21 ; one is
black, yellow and red, the other blue, yellow and red.
Only two woolen ponchos were found, Mus. 27569 and
27570; one of very fine texture is shown, pl. 7, fig. 19, in
5 n. s.
Both these ponchos are black, a uniform line of
4. The quilt, Mus. 27533, is sixty-five by fifty-two
inches in size. It has an interlining of cotton batting, the
same as modern quilts.
5. The loin-cloth, pl. 7, fig. 13: Mus. 27573; } n. s. It
is woven of wool, black with red and yellow stripes along the
edges ; the bands of black and yellow are braided.
6. The small belt, pl. 7, fig. 14: Mus. 27574; · n. s. It
is woven of wool and closed in ring-shape. It is ornamented
with five narrow figured stripes, bordered by red ones and
alternating with yellowish-brown bands of the ground material.
The figured stripes are woven in black and white, yellow and
red alternately; the pattern consists of a diagonal line and
two dots in sectional divisions and alternating in color.
7. The shoes, pl. 7, fig. 12: Mus. 27575; n. S. They
are made of buckskin, the hair side turned out, sewed and
Tech
JU
Fig. 50. Design of an embroidered border of a poncho, Inca period,
coast style ; j n. S.
Fig. 51. Design of the border of a poncho, pl. 7, fig. 19, from
Inca grave; } n. s.
embroidery stitching outlines the hem and side edges in
black, red, dark and light yellow; a typical finishing edge for
this period. 27570 is plain, while 27569 is soft and pliable,
apparently of vicuña wool and of much closer weave than
the former; it has seventy-seven warp by twenty woof-
threads to the half inch ; it is ornamented with a fine tapestry
border in black, red, yellow and white. The pattern, enlarged,
is shown in fig. 51. Alternating blocks show two different
designs : the one with three pendent figures (fishes ?), the other
a peculiar design, repeated diagonally in an inverted order.
2. The large cloths : These cloths are of light cotton
stuff, some plain, others figured. The former served as
shrouds for the mummies and are brown, finished with a
round edge of red wool; only one of the number is white,
bound with a narrow black woolen braid. The size is
approximately five feet eight inches by five feet, and nearly
bound with a knit braid. The inner length is eight and
three-quarter inches.
8. The braided cords, pl. 7, fig. 10 a and 10 b : Mus.
26639; # n. s. Figs. 11 a and 6 illustrate the manner in
which this cord was coiled around the head. Beside the one
preserved on this skull, there are two similar braids, twenty-
one and a half and twenty-three and a half feet long. All
three are of very soft black wool, probably vicuña; two are
braided of thirty-two strands, flat, with four edges (see figs.
11 a and 11 b), eleven-thirty-seconds and seven-sixteenths of
an inch wide by three-sixteenths of an inch thick. The ends
are secured by sewing on one, and by tying the strands on
the other. The third of these cords, twenty-three and one-
third feet long, Mus. 27635, braided of sixteen strands, is
almost round, but only one-quarter of an inch thick ; on one
end is a plaited loop three-eighths of an inch long.
经
​Fig. 52. Design of a border in raised embroidery on large cotton cloth,
Inca grave, coast style; } n. S.
Fig. 53. Border like the preceding.
the same in all. Two other white cotton cloths, five and
three-quarter feet by five feet one inch, and four feet five
inches by four feet two inches, are decorated with light
embroideries. One of these, Mus. 27563, is embroidered
with three horizontal stripes in red and yellow, the pattern
like fig. 52. The smaller cloth, 27535, is ornamented with a
wide stripe across the centre, in brown and black cotton, the
pattern being similar to the former; two narrower stripes run
along the edges, the design as in fig. 53.
3. Small shoulder-cloth : Mus. 27544. The size of this
cloth is twenty-five by nineteen and a half inches. The cot-
ton fabric is thicker than that of the cloths described above.
This article shows lengthwise stripes; the central one is black,
bordered by yellow, red and green, all alternating with brown
stripes of the same width as the central.
9. The ear ornaments, pl. 7, fig. 16: Mus. 27548 a ; }
n. s. They are made of strips of leaves, and are one and one-
eighth to one and three-eighths inches wide, rolled into a
spiral two and one-half to two and three-quarter inches thick ;
several sizes exist in pairs.
10. The plate, pl. 7, fig. 15: Mus. 27545; I n. S. It is
shallow of form, and instead of handles has small rim orna-
ments resembling loops ; the color is red and yellow, and the
pattern a familiar one, a diametrical stripe with lozenge-
shaped figures in it, and a circular rim with small zigzags.
11. Small pouch with coca-leaves, pl. 7, fig. 18: This
pouch, of plain linen, is nine inches high by five inches
wide, and filled with coca-leaves; the top is closed by sewing;
an exterior, covering of rushes, held together by a light net-
work, was put over the pouch as a protection against dampness.
;
(38)

pouch, pl. 7, fig. 18. A binding braid, knit of wool, as on
the shoe, pl. 7, fig. 12, was found in the Sun temple, and the
typical stitching around the edges of the stuff, as in pl. 7, fig.
19, was repeated identically in manner and color.?
1
8
This enumeration of objects may serve as an illustra-
tion of the correctness of the method of historical classifica-
tion.
The plate, pl. 7, fig. 15, proves that the style of Cuzco,
also, was represented here, since this type of plates with
identical or similar decorative designs is seen in every collec-
tion brought from the region of Cuzco. The pattern of the
tapestry border, pl. 7, fig. 21, is typical, and is to be found
everywhere, with more or less variation, on the Cuzco ary-
bals ;? the same may be said of the pattern upon the child's
belt, pl. 7, fig. 14. The large plain cloths probably are the
'yaqolla's cloaks, worn generally by the men of the high-
lands.
The forms or designs of numerous objects taken from
these graves were represented among the specimens from the
graves of the Sun temple, which also bear all the character-
istics of the Cuzco civilization. This applies to the shoulder-
cloth with triple stripes (above) ; 4 to the quilt, and the small
Fig. 54. Design of border of a large cloth, similar
to the preceding number.
8
Besides the types of Cuzco, those of the Coastland
are represented among the objects in these graves. To this
class belong, as regards technique and design, the woven
braids of a poncho (fig. 50, above), and the various designs
embroidered in the manner of fig. 52. The style of Cuzco
and that of the last stage of the coast civilization go well
together, as they are contemporaneous and often locally mixed.”
1 Comp. also for form and general design : Medina, Aborijenes de Chile,
No. 162. from Vallenar in Chile; for the design on the rim „see, f. i., the frag-
ments shown in Kultur u. Industrie, I, pl. 12, figs. 12, 13, 25.
2 Comp. Kult. u. Industrie. I, pl. 12, fig. 16, from the Titicaca Island, the
Cuzco arybal ; Congr. des Amér. Bruxelles, Atlas, pl. iii, from the district of
Cuenca in Ecuador; the design on stone pillars of Hatuncolla near Puno, Squier,
pp. 385–356, in the style of Cuzco ; designs on some garments and girdles of the
portraits of Incas in oil colors, originals at Cuzco, copies in Museum at La
Paz, Bolivia, and in private possession in the Hacienda Cusijata, near Copaca-
bana. These portraits were often reproduced, f. i., by Hamy in Galérie améric.,
pl. 51, Nos. 145, 146; by J. Sahuaraura in Recuerdos de la Monarquia Peruana,
1850, and by Rivero and Tschudi on the title-page of the Antiguedades peru-
anas.
Comp. Middendorf, Wörterbuch, p. 102.
4 Comp. below, chap. xxi. Shoulder-cloths, "llijlla,' figured with three
stripes, are to this day a typical garment with the Indian women in the high-
lands. The colors, red, yellow and green, as well as the narrow stripes of this
shoulder-cloth, are frequently seen in the ponchos of the Indians in Bolivia,
where these are the national colors.
5 The two peculiar quilts taken from these graves and from the graves of
the Sun temple are the only ones found at Pachacamac.
6 See bel., chap. xxi.
7 Comp. pl. 19, fig. 7.
8 The lower section of the design in fig. 51 of the vicuña poncho, pl. 7, fig.
19, which shows a conventional animal's face twice repeated, is derived from the
style of Tiahuanaco. The white hooked figure is the same as that found upon
the necks of animal heads (comp. above, p. 27, fig. 20; p. 33, fig. 35, and Stübel
u. Uhle, Introductory remarks with pl. 8, No. 1); the dark square in the corners,
the ring-shaped nose of the animal (see p. 33, figs. 34-36, and pl. 4, figs. 1 a and
c); the small toothed figure, a form of the ear which is not uncommon (comp.
the shape, fig. 54, on the face upon cup, pl. 5, fig. 4). The arrangement of the
minor parts is, however, different, suggesting greater remoteness of time and
loss of tradition. Another explanation might be given of the manner in which
this design was brought into a poncho of the Inca time. It is possible that stylis-
tic traditions of the period of Tiahuanaco survived in the region of Cuzco, or
the poncho itself was made in Bolivia, where traditions of the earliest period
may be supposed to have survived for a long time. It is certain that textiles
(made by the 'mamaconas' of the provinces) were carried to Cuzco from the
entire realm of the Incas.
Comp. the finds of the cemetery of the northern town, below, chap. xvii,
and of some of the graves on the Sun temple. In the poncho, also, where the
patterns are of the style of the coast, see above, p. 38, the proportions corre-
sponded to those of the highlander's poncho, p. 37.
a
MARKS OF THE INCA NOBILITY.
6
n
The interest claimed by these finds does not cease with
their classification according to period and style. They pos-
sess another interesting aspect, for it can be proved from the
objects discovered that the mummies buried in these graves
were members of the nobility of the Inca realm. It is well
known that in the land of the Incas the dwellers of the vari-
ous provinces were distinguished by the headdress, which
also showed the degree of rank of the individual. One class
of these ornaments, the llaut’u, made of braided cords," was
worn in a coil around the head as its only decoration by the
inhabitants of the province of Condesuyu. In the empire
of the Incas the color of the llaut’u was significant of rank,
and therefore very important, certain colors conferring the
highest distinctions and greatest marks of favor. Elsewhere
the color had no special meaning, beyond designating the
province to which the wearer belonged. The llaut’u worn
by the Inca himself was of crimson and blue, wound
2
3
about the head in four or five coils, with a fringe, 'mas-
qua paicha,' in the same colors, covering the forehead
to the temples. The heir-apparent of the Inca, accord-
ing to Garcilaso, wore the same ornament, except that the
masqua paicha was yellow. This forehead fringe was not a
regular part of the llautu, but was reserved for the Inca and
his successor. In confirmation of this statement the fact may
be mentioned that the nearest relatives, probably the mem-
bers of the Qhapaj aillu, wore two red and yellow tassels,
paicha, on the right temple in place of the fringe. The
llautus of the nobles, the orejones, were black. Garcilaso
informs us on this point as follows:10 “The first privilege
granted by the Inca to his followers was permission to wear
the llaut'u like himself, provided it were black.” It is there-
fore unlikely that any of his subjects except the nobility were
permitted to wear black llautus. The fact that the Inca wore
his coiled only four or five times about his head, while the
one represented in pl. 7, fig. 11, appears in thirteen coils, is
no obstacle to its identification as one of the llaut'us of the
2
6.
1 Zarate, I, chap. 6; Garcilaso, I, chap. 22.
2 J. J. v. Tschudi, Wörterbuch, p. 351, and Middendorf, Wörterbuch, p.
514, give the inaccurate translation of “headband” for llaut’u. Congr. des
Amér. Bruxelles, I, p. 608, pl. 2, fig. 6, has the still more incorrect designation of
the same as "an ornament in the shape of a collar.”
6 Montesinos, chap. 17; Garcilaso, I, chap. 22, says: “Of several colors."
The statement of Garcilaso, V, chap. 20, and of Xerez, p. 82, that the fringe was
only crimson, appears to be erroneous, unless the fringe differed in color from the
llaut’u.
7 Xerez, l. c. The translation of “tassel” for masqua paicha, used by v.
Tschudi, l. c., p. 376, appears therefore to be incorrect.
8 I, chap. 23.
9 Garcilaso, IV, chap. 2.
3 Relac. por el Corregidor de Chunbivilcas, Rel. Geogr., II, p. 29 : “las
cabezas descubiertas con algunos llautos al rededor dellas.”
4 Rel. Geogr., l. C., p. 14 (Cotahuasi): “traen .... las cabezas con unas
llautos blancos y negros en ella ;" p. 18 (Alca): “las cabezas con unas sogas de
lana al rededos de la media cabeza ;” p. 22 (Llusco y Quinota): “las cabezas con
y
unas sogas delgadas de lana ;” p. 25 (Capamarca): “en las cabezas traen ....
llautos, que llaman, de lana de colores."
5 The inhabitants of Rucana, by order of the Inca, wore red and white
coiled around their heads. Descripcion de la Tierra de Asunrucana, Rel. Geogr.,
I, pp. 204 and 207. The people of the province of Pacajes in Bolivia were recog-
nized by yellow stripes in their knit caps.
10 Garcilaso, I, chap. 22. Molina, p. 36, confirms the statement that the
llaut'us worn by the young men of Cuzco on the day of their initiation into the
rank of the nobles were black. The Inca Wiraqocha granted the privilege of
wearing the llaut'u to the men of the kindred tribe of the Kechua, as a reward
for their assistance in defeating the Chaucos. Garcilaso, V, chap. 23.
9
(39)

1
2
3
a
these comparisons, there is sufficient proof of the superior
rank of the persons buried in these graves.10 This conclusion
is in harmony with the favored situation of the graves, close
against the wall of the temple; and also with the statement
made by Cieza,ll that only princes, priests and pilgrims were
buried around the temple. These graves would not be con-
sidered as belonging to persons of high rank if they were
judged, as is usually done, by the presence or absence of
precious metals, and if written testimony were disregarded.
This fact proves that it is a great mistake to look merely for
gold and silver objects in graves, in order to establish the
rank of the bodies buried there. It is needless to conjecture
whether the gold in these cases had been retained by the
family, or whether it had been presented to the temple at the
time of the owner's death. At all events it will be necessary
to ascertain the rank of the persons by the objects actually
found in the graves, especially by the textile fabrics in which
the dead were buried. Possibly a statement coming from
Garcilaso might be more considered when he says, in regard
to the classification of the dead :12 “ As tribute they made three
5
Fig. 55. Colored braid, llaut'u. From a grave of Inca period. i n. s.
nobles, the color being the important point. The woolen
braids in figs. 10 and 11 were llautus of the nobility, that of
the Inca being described as of the thickness of one finger,
with four edges. The short hair of these mummies is full of
significance. The Incas had introduced strict regulations for
the hair of their nobles, degrees of rank being marked by its
varying length. “Some men wore their hair to the middle
of the ears, others still shorter, only the Inca had his closely
cropped."
It is possible that haircutting as a mark of distinction
was valued by the Inca nobility more even than the privilege
of wearing the black llaut'u, as it represented a greater
variety of degrees of rank. Judging from the shortness of
their hair, the men buried in these graves must have been of
very high station. Brühl, quoting from a source unknown to
us,' states : “The orejones, the nobles, wore their hair partly
long; those who belonged to the Inca ayllus cut it quite
short, as short as the Qhapaj Inca himself.” The hair of
these mummies 5 is from three-eighths to three-sixteenths of an
inch long, therefore in life the men were certainly nobles
and may have belonged to the ancient ayllus of the Inca at
Cuzco.
Ear ornaments were badges of nobility and seem to have
been granted to the nobles exclusively, according to the regu-
lations of the Inca ; for this reason the Spaniards called them
'orejones.' Their value was according to size and material.
We are not distinctly informed as to the size, but those found
in these graves, evidently belonging to the nobility, may per-
haps be considered as the standard of those of the orejones
of Cuzco. They are about as large as the foot of a goblet,
and none might be worn by subjects larger than half the size
of those of the Inca. Being made only of leaves, their value
might appear insignificant, if it is considered that the chief of
the orejones wore ear ornaments of gold; but we are told
that the Inca occasionally granted the wearing of 'totora' as
a special favor, from the resemblance which they bore to his
own. The shoes, pl. 7, fig. 12, were possibly of some special
significance, and perhaps indicated rank, although they may
be women's shoes. Similar ones were found in the graves of
the Sun temple where only women were buried. The style is
that of the ordinary type of women's shoes, and they may be
compared to the llauki, which were adopted by young girls at
the age of puberty.
A few points of resemblance between the textiles from
the graves and some from graves in the Sun temple may be
reserved for a later discussion. Some peculiarities appear in
both classes alike, such as: a felt-like texture of the two cot-
ton ponchos, a very unusual type of cloth ; also the fine stitch-
ing which binds the edges in exactly the same colors in each
case, white, yellow, red and black. The same colors appear
in the tapestry part of the fine poncho, pl. 7, fig. 19, and in
the two stripes of the little child's belt. Even disregarding
kinds of woolen garments, the first class, awaska,13 being used
only by the common people. The second, cumpi,14 of fine
wool, of several colors and well made, similar to tapestry, was
reserved for the use of the nobles and of the Inca's retinue,
also for captains and chiefs. The third
The third class, compo, was
made of the finest wool existing and worn by all persons of
the blood of the Inca, both in war and in peace, by soldiers,
captains and officers.” The poncho, pl. 7, fig. 19, with its fine
tapestry border, may be of the third class, and the two cotton
ones, with plainer tapestry, of the second. Considering that
the Incas as a rule only renewed old customs observed in
Peru long before their rule began, the stuffs found in these
graves may be classified according to this principle. We must
bear in mind, however, that archeological research of the pre-
Inca period is not advanced enough for us to accept without
reservation the classification of textile fabrics as given by Gar-
cilaso.15
1 Garcilaso, V, chap. 2; I, chap. 22.
2 The llaut'us as head ornaments pictured on the Inca portraits (described
above, p. 39), dating from a later Spanish period, are incorrectly drawn.
Instead of being wound closely about the head, and laid in a spiral decreasing
towards the top of the crown, as described by Garcilaso, the llaut’u here is
arranged in coils becoming wider towards the top and having the shape of a cap.
An object in the shape of a cockade appears as a decoration, and the headdress
looks almost like a helmet. In the face of these erroneous later day representa-
tions it is an important fact to have the only authentic proof of the manner of
wearing the llaut’u pictured upon the skull, pl. 7, fig. 10. As we see it here it
was worn by the Inca nobles and the Qhapaj Inca himself. The masqua paicha
was generally made too narrow and too short in those portraits.
3 Garcilaso, I, chap. 22.
4P. 273.
5 The hair is short, as if cut but lately. Garcilaso, l. C., too, states that the
Inca and his nobles had their hair cut often, so as to have always the same
appearance.
6 The piercing of the ear was a privilege. Garcilaso, I, chap. 23 ; V, chap. 23.
7 The Inca determined the exact size and measurement. Garcilaso, l. c.
8 L, C., chap. 73.
9 Shoes were generally alike for men and women.
10 These men were probably natives of Cuzco, judging from their closely cut
hair and the shape of their crania, which differ from the prevailing type of the
coastland. They were not necessarily members of native chieftain families who,
having adopted Cuzco civilization, had been received into the caste of the nobles ;
but the fact of the two embroidered cloths and one of the embroidered ponchos
having come from the coastland shows some certain connection with that
locality
11 Cronica, I, chap. 73.
12 B. V, chap. 6.
13 Of alva,' to weave, v. Tschudi, Wörterbuch, p. 21, and 'awaska,' that
which is woven.
Qompi,' the costly garb, v. Tschudi, 7. c., p. 166; 'qumpi, fringes, l. C.,
is surely the same word.
15 Near these graves a piece of colored cord was found (M. 32684), seventeen
inches long, greatly resembling the llaut’us of black wool discussed above, fig.
55 ; î n. s. This cord, eleven-thirty-seconds of an inch in thickness, is braided of
thirty-two strands, and almost square in its section. It resembles one of the black
cords and is finished at one end by a loop one and a half inches in length. The
colors are grayish-yellow and citron, crimson and dark blue, so intermingled as
to form zigzag patterns. At the opposite end the strands are unraveled for two
and a half and six inches, and the red and blue strands are tightly bound with a
crimson thread while the yellow are hanging loose. This piece at first claimed a
certain interest by its pretty colors and design, but there is a possibility that it
may possess a still higher one. Red and blue were the colors of the Inca's head-
dress, while his nearest of kin, probably the whole 'gens qhapay aillu,' wore
red and yellow. But as blue, beside red, is sometimes ignored in the description
of the llaut'u of the Inca (Xerez, p. 82; Garcilaso, V, chap. 20), the same omis-
sion may have occurred in stating the colors of the clan of the Inca. We would
thus have a combination of the Inca colors, red and blue, and of that of the
adopted successor, yellow, the very colors shown by this piece of cord, besides a
fourth one. We know furthermore that the chieftain of the town after the Inca
conquest received the rank of an Inca of the “royal blood” (Garcilaso, IV,
chap. 31). The colors yellow, brown, purple and blue are repeated in the
embroidered figures of the cloth, pl. 7, fig. 21, described below.
(40)

CHAPTER XIV.
THE GRAVES IN THE UNCOVERED SECTION OF CEMETERY I.
а
of cloth without false heads, and with or without ornament
on the outside. Mummies with false heads were uncommon,
but in many instances the material of the wrapping was
found to be extended at the upper end like a handle, tied
with pieces of cord and decorated with feathers, etc., as the
real head might have been in life.
While a cultural decline is apparent, even in compari-
son with the preceding Epigone period, there is certain
progress corresponding to the general advancement of the age.
In some cases earthen vessels have been made by the help of
moulds; utensils of metal, such as tweezers, spinning-whorls of
copper, etc., are more frequent; ornaments fashioned of ham-
mered sheet-metal are seen in more developed forms; and
scales appear for the first time in this period. Besides com-
plete scales with nets, about a dozen wooden sticks, former
scale-beams, were found in this cemetery.
A. POTTERY.
THIS
THIS vast burial ground surrounds the Temple upon three
sides, extending as far as the enclosure of the court, and
embraces the much rifled area towards the north and the
mound-like elevation towards the east. The finds of this field,
with few exceptions, belong to the same period as the graves
v of the upper stratum underneath the Temple. All the pot-
tery found here is identical in shape, technique and decora-
tion with that of the former graves, as will be seen on pl. 7,
figs. 1 to 6.1
Not all the earlier types are entirely wanting, however.
A conical grave built of stone was uncovered, wherein the
mummy-bale was wrapped in textiles of the most ancient
class, among which was found a tapestry poncho in the pure
style of Tiahuanaco. Such occurrences may be explained in
this way: this section of the cemetery is of the same date as
the ancient part y, and originally 3 both had the same class of
graves. By being covered with strata of earth and débris,
the old section was rendered impracticable for burial purposes,
while the uncovered ground continued in use. Through
gradual destruction of older graves in making new ones,
the general character of the cemetery was continually mod-
ernized until it came to a standstill in the period of which
the cemetery now bears the impress. The Inca period, as far
as our observation goes, is not represented here.
This explanation throws light on other points of impor-
tance. It does not exclude the possibility of discovering older
graves, which by accident or careful sealing escaped destruc-
tion. It explains the immense crowding of graves, and the
condition of the soil itself, which is a conglomeration of
remains of mummies of every kind—of bones, fragments of
cloth, pieces of rope, human hair, etc. This was brought
about by the long-continued use of the cemetery, extending
through all periods of the history of the ancient Temple, until
the time of the Inca period or soon after.
The graves, as to type, are similar to those described
p. 35. Cane thatching covers grave structures of inferior
durability, from two to seven feet below the surface. In some
places, especially in close proximity to the Temple, there are
found in this soil adobe-walls, which form rooms or narrow
passages, into which the mummies are fitted, and in some
instances closely packed,4 one above another.
The majority of the mummies resembled the types from
Ancon, shown by Reiss u. Stübel, pl. 20, figs. 1; 22; 25,
fig. 1. These are common mummy-bales, wrapped in pieces
7
a
Form and technique of these vessels were discussed
p. 35 ff. Technically they are most nearly related to the
well-known type of Chancay. The decoration of the surface
by coloring without polishing is common along the entire
coast of central Peru, at least as far north as Pativilca.
Similar vessels have been found in the valleys of Rimac and
Chillon, as well as in Ancon. A large number of black ves-
sels were recovered here, all of a plain local type, technically
related to the more familiar Peruvian black pottery.
The ornamental features of this period at Pachacamac
are inferior to the well-known type of Chancay. Figure ves-
sels 8 are largely represented in this class of pottery, without
any individuality of style, however, and of rather inferior,
even crude workmanship. The vessels with white, red and
black decorations show a great variety of patterns : figures of
birds (pelicans), toads, fishes and others, possessing but little
artistic merit. These figures are as simple as is probably
their symbolism. The general character of the decoration
may be seen on the vessels shown, pl. 7, figs. 1, 4, 5; pl. 8,
figs. 2 and 3. Undecorated vessels are more numerous in this
cemetery than in the others; the same may be said of vessels
with primitive engravings, of the type shown, pl. 8, fig. 9,
having very crude faces and decorations. There is a large
series of this class of pottery; also of the figure-vessels and
1 See above, p. 35 ff.
2 These fabrics could not be preserved, as the fine warp-threads had
decayed, owing to their great age.
3 See above, p. 19, fig. 3.
* Graves as described by George Squier, p. 74, with all the details, and
with exact determination of the different members of the family, we have not
found.
5 They may belong to the same period.
6 The small sticks, Reiss u. Stübel, pl. 87, figs. 5, 14 and 15, were such
scale-beams. These beams were made of wood or of bone. The cultural influence
during this period, which came to Pachacamac from the north, is apparent in
everything-pottery, textiles, copper spinning-whorls, which sometimes show
human faces with lip-ornament, etc. From there, probably, was imported the
use of the scales, which could not be traced in earlier graves. Waitz, Anthro-
pologie der Naturvölker, 1872, III, p. 381, mentions several tribes of the northern
coast of South America who used scales for testing gold.
7 Reiss u. Stübel, pl. 96, fig. 3.
8 Comp. pl. 8, figs. 1-3 and 5.
6
(41)

B. TEXTILES.
a
those in red, white and black, none of which give evidence
of much artistic taste.
Among the ornamental types preserved and continued
from the Epigone period, are the animal's head with enlarged
ends and the tufted headdress (comp. both, pl. 7, fig. 1).
Some vessels shown on pl. 8 are intended to illustrate more
clearly the character of the period as generally indicated in
pl. 7, figs. 1-5.
Pl. 8, fig. 1: Vase, formed like a human figure. Mus.
26895; # n. s. Only the face of this figure, with large head
ornament, and the arms, are formed in detail, the legs being
merely indicated in relief. The figure is painted in the ordi-
nary manner. It represents an Indian with a high crown of
feathers in gradations, the head ornament for one of the
numerous festivals which the Indians celebrated in all parts of
Peru. The figure is seated and playing a Pan's flute, such as
is customary at festivals, while one hand is shaking a rattle,
suspended from its neck.
Pl. 8, fig. 2: Vessel in the shape of a cat-like animal.
Mus. 26916; \ n. S. The head, in proportion to the body, is
very large, especially the ears. The face of the animal is
exceedingly unnatural, while the claws are quite realistic and
the tail is curved over the back. The vessel is colored.
Pl. 8, fig. 3: Vessel in the shape of a bird. Mus.
26917 ; I n. S. The bird is seated, so that the vessel rests
on the feet and on the back part of the body. The tail ends
with a broad flat line.
Pl. 8, fig. 4: Clay doll. Mus. 26960; n. s. The
figure is hollow, the head is quite large and out of propor-
tion with the body. The hands are extended and resemble
the type of many clay dolls of the same or later periods on
the northern coast. These figures are usually painted.
Pl. 8, fig. 5: Black double vessel with human figure.
Mus. 26879; # n. S. This vessel consists of two barrel-shaped
parts with separate feet. The small cylindrical neck is
attached to one of these parts, while the other is surmounted
by a seated human figure, about the head of which is coiled
a spiral band.
Pl. 8, fig. 6: Double vessel with bird's head and whistle.
Mus. 26885; n. s. The two bodies of this vessel are
rounded, one of them bearing a large parrot's head, in the
hollow of which the whistle is concealed. On some similar
specimens the wings are indicated on the bowl. These vessels
are usually colored.
Pl. 8, figs. 7–9, reproduce a peculiar kind of vessel. Fig.
10 comes from the region of Lambayeque. Mus. 34760 ; #
n. S. Vessels of this type are also represented among those
from Pachacamac.
Pl. 8, fig. 7: Black vessel in the shape of a woman,
nursing a young animal. Mus. 31796 ; \ n. s. The woman is
seated on a square support and presses the animal with her
right hand against the left breast. A second animal, which
is awaiting its nourishment, lies at her right side. This ves-
sel is well polished and of attractive workmanship. It
belongs to the same class as fig. 10, as may be seen by the
two human figures on the handle, which resemble those on
the side of the vessel, fig. 10. This object is evidently an
imported one. Vessels of the same class occur at Ancon, as
shown by Reiss u. Stübel, pl. 93, figs. 11 and 12; and in the
Santa valley. There is no indication as to the locality from
which were brought pieces of this type, but it may be the
northern part of Peru. None of them are to be regarded as
native to Pachacamac.2
The textile art of this period is represented by a large
number of specimens. There are examples of all the differ-
ent techniques known at that time: tapestries, plain textures,
figured material, in which the figure is produced by warp-
threads in two or three colors (see pl. 8, fig. 14); double
stuffs, open-work materials with interchanged warp-threads,
gauzy fabrics and embroidered ones, besides a great variety of
braids, etc. The same class of objects repeated here is illus-
trated by Reiss u. Stübel in their book on Ancon; but grave
tablets, like those on pl. 32, 1. c., are wanting. In design and
technique these stuffs resemble many of those shown in the
before-mentioned work, with the exception of some of either
an earlier or a more recent date than this period.
The following designs may be specially mentioned as
being characteristic of Ancon, according to Reiss u. Stübel,
and of the open section of this cemetery, at the last pre-Inca
period: rhombic (or rather hectangular) ponchos represented
in pl. 44, fig. 1. The patterns are always the same. Ponchos
identical in texture and design with those in pl. 35, figs.
and 4. Braids such as those on the mummy, pl. 22. Figure
designs in tapestry, as pl. 51, fig. 1; pl. 52, fig. 1; pl. 69,
fig. 1. Separate ornamental figures, mostly indistinct in
detail, but of the general character as those in pl. 67 b, figs.
3 and 4; pl. 53, fig. 2. Bird designs, pl. 56, fig. 1. Lineal
ornaments developed from figures, pl. 63, fig. 3. All-over
designs, pl. 67, figs. 1 and 6; pl. 63, fig. 4; pl. 62 a, fig. 3;
35 a, fig. 1; 59 a, fig. 4. Patterns like that of pl. 67 a, fig. 1,
are often seen in fewer colors and in a different technique, the
pattern threads running through the design. Subdued color-
ing is characteristic of this period. One very important cate-
gory of all-over patterns, found in this cemetery, seems to be
entirely wanting at Ancon; therefore a few typical specimens
are shown, pl. 8, figs. 11-13, 15 and 16. Gauze-ponchos, of
the class shown by Bastian, I. c., pl. 5, are not represented in
the Ancon plates. These ponchos are white, colored, or
with mythological figures painted upon them; the material is
a very fine cotton, with a strip of tapestry down the front,
and a border around the edges and sides to strengthen the
stuff. The ground color of these borders is nearly always
green, rarely blue. The ornaments are typical and designed
from plants. In some cases small pieces of tapestry are
attached to the gauzy material, as may be seen on the poncho
shown by Bastian. A large number of ponchos with similar
borders were found in this section of the cemetery, and it is
possible that the poncho shown by Bastian, the origin of
which is not known, may have come from here.3
Comparing the designs and technique of the weaver's art
of this period with that of the earlier, some great differences
are to be noted.
1. The figures in the tapestry designs show a tendency
to curtailment in details, which is evident in various ways.
a. The figures become more rigid and inartistic, as pl. 7,
fig. 7; above, fig. 49; they are set side by side, or sporadically,
without any connection, pl. 8, fig. 14. b. The figures repeat
the main forms of the design in a characteristic, even natural
To commercial relations point: the beads which were numerous in the graves of
the central Peruvian coast and in Pachacamac, probably coming from the island
of Lobos ; also the fishing-boat, made of a tree from the forests of Equador, still
in use in that region (comp. Reiss u. Stübel, pl. 3), the type of which is a very
ancient one.
3 Wiener shows, p. 514, a strip of border of the same width and design as
the central strip of the poncho mentioned above. He calls it 'the central part of
a cloth' (linceuil); it is shorter in the reproduction, but probably by faulty
reconstruction, and it is evidently the central strip of one of these gauze pon-
chos. As to its origin, Wiener states : “Trouvé dans la ‘huaca’ nord du
Gran-Chimu, non loin du groupe des grands palais”(?). This again seems to be
one of those conscious and systematic falsifications of locality (see Stübel u.
Uhle, l. C., p. 49, note 2), whereby Wiener's book loses all scientific value.
1 The nursing of young animals by women was observed by the writer in
the highland of Bolivia.
2 On ancient trade connections along the coast of Peru, see above, p. 5.
(42)

and animated way, but lack beauty of style and finish in
detail. Examples of this class from Ancon are the figures in
the stuffs by Reiss u. Stübel, pl. 51, fig. 1 ; pl. 52, fig. 1. A
fine and characteristic example from Pachacamac is shown in
fig. 56.1
2. The all-over patterns of the earlier periods followed
the principle of the rectangular outlines, corresponding to the
figures from which these were developed. The edge there-
fore followed the warp and woof.? In the present period the
principle of the diagonal arrangement and outline of the
design become typical. This new principle has an influence
on the figure design, comp. pl. 8, fig. 14.
In the fabrics, pl. 8, figs. 11-13, 16 and 17, it is apparent
that the edges of the designs were adapted diagonally, so as
to follow the direction of the warp. The slits occurring in
tapestry patterns were thus avoided. The specimen, pl. 8,
changed into a diagonal arrangement. There are several
examples of this class, bearing certain marks which may
date
them from a transitional period, when the diagonal outlines
were only being introduced.
3. Tapestries of earlier periods, as Reiss u. Stübel show,
pl. 49; 55 ; 69, fig. 6, here pl. 6, figs. 5 and 9, etc., are of
the basse-lisse’ technique or mainly so. The stuffs of the
present period on the contrary, such as Reiss u. Stübel show,
nt
pl. 51; 52; 69, figs. 1 and 3, are of the 'haute-lisse' order.
That the choice of a special principle is intentional is, per-
haps, proved by the design of the pouch, pl. 8, fig. 14, where
the figures are set side by side, although the effect would have
been better if standing vertically.
4. The ornamental types of the earlier period are devel-
oped more and more into numerous all-over designs, gradually
losing their symbolical meaning. The design, pl. 8, fig. 16,
seems to be a reminiscence of figures like pl. 6, fig. 9a; it
also is in a state of transition into a mere all-over pattern,
similar to that of fig. 11. In the design of the type of fig.
13 the process of transition has come to a close.
DESCRIPTION OF OBJECTS.
CD
Agogogogogog
BE
ANANAND
SE
2
2
TURIS
60
8
Pl. 8, fig. 11: Forehead-band with ends of tapestry.
Mus. 29506; 1 n. s. This band is six feet ten inches in
length and consists of a plain cotton band six feet long by
one and a quarter inches wide, with end pieces of tapestry
five inches long and three and a quarter inches wide. Two-
thirds of the length of these tapestry pieces are divided into
three strips, joined together at the ends and finished with a
long brown fringe. The tapestry shows the colors white, red,
yellow, blue, in a diagonal all-over pattern with many grada-
tions.
Pl. 8, fig. 12: A forehead-band resembling the former.
Mus. 29517; 1 n. s. The length of this band is five feet
eight inches, without the fringe. It consists of a plain cotton
band of five feet two inches by one and seven-eighths inches.
Tapestry ends are woven on to it and terminate in a long
fringe, which has been joined on after weaving. The band
proper is loosely woven, with partly double, partly single
warp threads. It widens out near the tapestry ends, which,
are three inches long by three and a quarter inches wide,
woven in two sections of coarse yellow, red, brown and black
woolen threads. One section at each end has a red and black
ground. The design, however, is different on all four strips,
consisting partly of an all-over pattern with many gradations,
partly of rhombic, triangular, spurred and inverted rhombic
figures. Forehead-bands of this class, all very similar in
their general type, in color and design, were found in large
numbers in this cemetery.
Pl. 8, fig. 13: Strip with tapestry. Mus. 29924; # n. s.
This strip, fourteen and three-eighths inches long, consists of
a loop three inches long by three and five-eighths inches wide,
showing a faint design and a tapestry piece attached to it.
At the joint a piece of cane has been inserted to keep the stuff
stretched flat. The tapestry piece is eleven and a quarter
inches long and four and, a quarter inches wide; the edges are
stitched all around. The design shows in white, green, blue,
red and three shades of yellow, a complicated all-over pattern,
with many gradations, in which birds and cat-like faces are
repeated. Red, white, black and yellow strands ten inches
long form a fringe. This fabric may have served in some
ornamental way, attached to a wooden staff.
Fig. 56. Tapestry with mythological design. From later pre-Inca period.
3 n. s.
fig. 15, however, shows that occasionally the outlines of the
design might follow the direction of the woof, and were thus
1 This interesting piece of stuff is decorated in the centre with a design
showing a staff or tree with crosses and figures about it ; fruit with grain-like
contents streaming from the staff like rain. A number of human figures, stand-
ing, kneeling or running towards the tree, seem to receive gratefully, with out-
stretched hands, the gifts poured down upon them. They seem to be chieftains,
judging from the ornaments on their heads, and each one is marked by several
tablets bearing arms or signs. A similar tablet, a seal of clay with a condor's
head on it, is to be seen among the acquisitions from Tiahuanaco. These signs
.
resemble the figures in Mexican picture-writing. The occurrence in Peru of
these shields or tablets, bearing marks explanatory of the figures, should be more
closely investigated.
This important design is imperfect in its upper part. Two figures, probably
similar to each other, stood on both sides of the pole in the centre. One is still
visible; it stands in a frontal position, different from the lower. Its headdress is
high, blue and ornamented with something like an axe; the almost square face is
red, and the left hand is extended, as if bestowing gifts. It is clear that the two
upper figures were the sources of the gifts which the human beings below
received gratefully. The square face is a reminiscence of the figure on the
Gateway of Tiahuanaco, and of the cloth, pl. 4, fig, 1. It appears that the color
of the Huacas was red (comp. the description of the vessel, pl. 5). The pole in
the centre formed at the same time a continuation of the slit of the garment of
which this piece of stuff was a part, and of the design itself, serving as the
medium, whereby the gifts of heaven were poured down upon humanity. For
the first time, perhaps, we see here a distinct religious significance connected
with the cross in Peru, outside of a vague statement made by Garcilaso, that the
Incas possessed and worshiped a cross of stone. The Chilcha erected a cross of
stone upon the grave of persons killed by snake bites. This reminds us of the
cult of the cross among the Mayas and the Mexicans, where, too, it is connected
with beneficent powers.
The ground of the stuff is white, the figures being mainly yellow with red
outlines; the loin cloths are blue, a few additional colors are grey, blue, brown,
green, all rather faint.
2 Comp. pl. 6, fig. 6, p. 31.
3 The same is the case with the tapestry : Reiss u. Stübel, pl. 66 a, fig. 1,
from Ancon, but in that instance the slits had been dexterously closed by the
tapestry. A very similar piece of stuff was found in a peculiar grave at Pacha-
camac.
(43)

n
It con-
Pl. 8, fig. 14: Pouch with design in colors, part tapestry,
part plain. Mus. 29900; n. S. The pouch is seven and a
quarter inches long by six and a quarter inches wide.
sists of woolen vertical and cotton horizontal threads. The
inside and outside are alike; the sides and upper edges are
stitched. A thick band serves as a strap; it is woven with
a round woof as a double band. The pattern upon it is recip-
rocal ;1 it is one and three-sixteenths inches wide; the colors
are dull white, red and green. The ground color of the
pouch is purple; the pattern consists of two sets of figures in
alternate colors, blue, green, light-red, yellow and white.
The woven stripes are black and white, and show a sort of
animal figure and an all-over design. The texture of the
plain parts looks like fig. 57, which gives the ground a dotted
H
Fig. 57. Three-ply weaving of pouch, pl. 8, fig. 14.
appearance. The ends of the pouch are finished off with
tassels.
Pl. 8, fig. 15: Belt of tapestry. Mus. 29878; 1 n. S.
The belt is ten and a half inches long by five and a quarter
inches wide. It is closed and consists of two thicknesses of
material of the same design, which are joined at both ends by
means of a round knit seam; a similar one secures the edges.
The pattern shows many colors and vivid contrasts—white,
two shades of red, also brown, yellow, blue, green and black.
There are seven stripes of uneven width, the four narrower
ones of which repeat a meander design, and the three wider
ones show an all-over pattern in mostly rectangular grada-
tions. The original figure design of this pattern may be seen
in eyelet ornaments. A red woolen tassel finishes each of the
two lower corners.
Pl. 8, fig. 16: Border of a cotton poncho. Mus. 29934;
4 n. s. The poncho is of a grayish-brown color, of a thin
gauzy tissue ; its dimensions are twenty-five and a quarter
inches in length by forty-one and a quarter inches in width.
The opening for the neck and armholes, three and three-quar-
ters inches long, were bound with colored knit borders. The
lower edge, two and fifteen-sixteenths inches wide, consists
of a wider border and a narrow sort of braid binding, with
a short fringe formed at intervals by the woof. The wider
border shows, in sections, interlaced figures. The figures of
each color cover four of these sections, two of them on both
sides being interlocked with the adjoining figures of contrast-
ing colors.
Pl. 8, figs. 17 and 18, show figures of two central strips
of s
gauze ponchos. Mus. 29576 and 77; } n. S. The two
figures, 17 and 18, appear in several repetitions in both strips.
The design of both is the same. Two figures clothed like
men and facing each other, are holding on to a staff, which is
broad at the upper end and pointed at the lower. In fig. 18
a flat object is noticeable under the point of the staff, which
appears to be resting on a support with two points or feet,
in such a manner that the centre lies hollow. Judging from
the pointed shape of the staff in both reproductions, and the
manner in which it is held, the two figures seem to be engaged
in piercing the object beneath it, and it may be assumed that
the figures are keeping a drill in motion. The drill as seen
in pl. 8, figs. 17 and 18, is being applied to some object, the
nature of which is not clear. It rests upon a foundation hav-
ing two prongs or supports, and is evidently not a fire-drill ;
it may have been used for the boring of pearls. The proce-
dure as here represented is not an ordinary one, as the figures
are pictured with high festive headdresses, similar if not iden-
tical in both plates. The headdresses in fig. 17 are of the
same shape as the support upon which the object for drilling
rests in fig. 18. The faces of the men resemble those of ani-
mals. Square objects at the back of the neck seem character-
istic, as they are reproduced in both plates. This scene evi-
dently pictures some ceremonial act. Such representations
and mythological figures are of frequent occurrence upon pon-
chos, but no interpretation can be attempted until the ancient
customs shall be better known and more specimens brought
together depicting similar scenes.
Pl. 8, fig. 19: Central strip of a gauze poncho like the
preceding. Mus. 29578; } n. s.
n. S. The ground in this strip
is dark blue, and upon it are two figures in a frontal posi-
tion with high headdresses. The staves in their right hands
resemble those in figs. 17 and 18, being wide at the top and
pointed at the lower end. Some object, such as a flag or an
axe, seems to be attached to one of these staves. This strip
is duplicated, and a mythological meaning suggests itself, on
account of the heads which decorate the ground and the
border.
1 Throughout this report we use the term reciprocal for patterns which are
alike on both sides, only in opposite colors, and complementary such designs
which are continued in the same outlines through colored stripes, etc., reflexive
for stripes of a different color which cross designs, only partly changing their
color, but not the outline (as in woven baskets, comp. Kultur u. Industrie, II, pl.
8, figs. 6, 8, 11, 12).
GENERAL REVIEW OF THE PERIOD.
The characteristics of this last period are in marked con-
trast with those of the preceding. Only a few slender
threads connect them, in the form of ornamental types, such
as the head of an animal symmetrically developed (pl. 7,
fig. 1) and the tufted ornament, resembling a staff broadened
towards the top. A few additional resemblances might be
traced in the elements of ornamental design, but differences
and departures from older designs are much more numerous.
The strong contrast between the older forms and technique of
the pottery and those of the present period is very evident,
quite apart from difference in artistic skill. A new decora-
tive type appears, and entirely new principles of technique
arise in textiles, such as the tendency toward diagonal out-
lines of the patterns, together with a complete transformation
of design. It has been suggested that during this period a
distinct transition had been accomplished in the technique of
tapestry from 'basse-lisse’ to ‘haut-lisse. This sudden and
abrupt change? must have been brought about by unknown
circumstances, strong enough to cause a revolution in the cul-
tural forms of Pachacamac. The difference in the head-
bands from those in older gravesmay serve as a proof of
this. In the lands of the Incas each tribe or province was
distinguished by the headband. Doubtless this custom was
an old one, and merely confirmed by law through the Incas.
The uncivilized tribes of eastern South America are now dis-
1 See above, p. 29.
2 Comp. pl. 6, figs. 9 and 15.
(44)

tinguished from one another by the peculiarities of the feather
ornaments in their headdresses. Headbands of plain white
cotton with two colored tapestry ends (pl. 8, figs. 11 and 12)
are entirely wanting in the older section of the cemetery,
while they are so characteristic of the graves in the uncovered
section. Such change in cultural form might suggest a differ-
ence in the predominating population. The nature of the
new elements which came to rule at Pachacamac seems to be
proved by the technical types of pottery and textiles, which
correspond to those extending along the coast of central Peru
as far as Pativilca, or even farther north. Gauze ponchos, so
characteristic of this period, point to the coast, especially its
northern part, as their climatic home and natural place of
origin. It appears, therefore, that previous to the conquest by
the Incas two forms of civilization had met upon the soil of
Pachacamac; the older, that of the earlier graves, depending
upon the culture of the highland, the more recent being intro-
duced from neighbors on the coast and, although inferior,
abruptly supplanting the earlier.
Comp. also p. 37.
(45)

CHAPTER XV.
REMARKS ON THE HISTORY OF THE TEMPLE.
A TOPOGRAPHICAL description was previously given,
recent.
.
p. 19, ff. (with fig. 3), of the site of the Temple and of
the different strata of soil found under and in front of the
Temple terraces. We shall have to draw conclusions concern-
ing the history of the sanctuary from the character of those
different strata in relation to the graves found in them. The
graves of the original cemetery, a (graves No. 1 above),
extended to the wall, 6,1 as has already been shown, and the
characteristic objects found there belonged mainly to the style
of Tiahuanaco or to that immediately following. The wall,
doubtless, formed part of the original building and was in
existence before the foundation of the cemetery. The Temple
structure, therefore, if not older, must be at least contempo-
raneous with the period illustrated by these graves, and must
have been built at a period and by a people holding the same
religious views which found their most artistic embodiment in
the ancient monuments of Tiahuanaco. Some fragments of
pottery found near l in section m, in the rear of the Temple,
were identified as being of the same period. They were dis-
covered at various depths, five to fifteen feet, among adobe
fragments, and were classified with vessels of that period
decorated with textile patterns. After the cemetery at the
foot of the Temple had been in use for a long time, events
took place during the course of which the building was
destroyed. This destruction is shown by the leveling of the
stone front wall, and by the stratum of fragments of ruined
adobe walls above the soil of the original cemetery. The
adobe fragments are from the old Temple, judging from their
position and from the style of their decoration, one which in
this ancient city was used only on buildings devoted to the
cult. The débris of the ruined walls piled up in front of the
structure and formed a slope (d, e). The destruction of the
sanctuary corresponded to a complete change in the dominant
form of civilization in the town, as is shown by the graves v,
found in the débris of the Temple. These graves, made since
the destruction of the original Temple, not only reveal a
different form of civilization, but even a different nationality,
a people by whom probably the ruin was effected. This peo-
ple possessed close cultural relations with the northern part
of the coast, which always had an ascendency over the cul-
tural forms of the highlands. It is a difficult matter to fix
the date of the wall with mural paintings in the manner of
the Chimu, since it was untouched by the destruction and by
the rebuilding of the Temple. If older than the first event,
it seems strange that it escaped unharmed. The character of
the painting would be more in harmony with the later period,
that revealing the supremacy of the tribes which possessed the
general culture of the coastland. On the other hand, the
wall does not belong to the plan of the terraces of the new
Temple, since it was concealed by one of them built in front
of it, and its height, four and a quarter feet, compared to three
to three and one-quarter feet of terraces, p, q, also points to a
different date of erection. No mention is made by any of the
early writers of temple walls frescoed with human figures.
These paintings, therefore, at the time of the discovery,
belonged to a past era, and that they had been in existence
a long time is shown by the fact that the figures have been
renewed, the older ones still being visible under the more
The date might be assigned to three different periods
of the Temple, two of them after its destruction. As, how-
ever, the wall was not destroyed, but only concealed by
another, it may be assumed that the two last periods were
not separated by any great convulsions.
When the Temple was enlarged by the addition of ter-
races certain sections of the ancient cemetery were built over.
The extension included more than one side of the Temple, as
there are graves of the same period underneath the terraces
of the northeast front. The people to whom the destruction
of the Temple must be ascribed evidently made their appear-
ance prior to its reconstruction, as several graves, very charac-
teristic of that period, were found in the ruins under the later
temple terraces (the graves v). Graves bearing marks of the
Cuzco type, however, are situated only outside the Temple, in
no case under its terraces. The graves v4 also seem to reflect
only the first part of the period to which they belong; various
earlier forms peculiar to the same, and which might be
expected in these graves, are wanting. There are only a
small number of graves of this class. They are entirely
wanting in the section of the cemetery which was covered
over with stratum i at the time of the rebuilding of the
Temple. A more complete representation of the character-
istics of the period was prevented by the rebuilding of the
Temple during the period of the graves v, at least in all those
sections upon which the terraces were set, p to t, and under
stratum i.
Numerous types of specimens are found only in the open
section of the cemetery, never duplicated in the part which
was built over, and all of them quite free from any admixture
of Cuzco types. The existence of these objects in the later
graves, which were made after the rebuilding of the Temple,
mark that period as a far earlier one than that of the appear-
ance of the Incas in the valley. It seems, therefore, difficult
to assume that the rebuilding of the Temple should have
taken place as late as the Inca period.
1 See above, p. 19, fig. 3.
2 Comp. above, p. 29.
3 See above, p. 19.
4 See above, p. 19.
5 See above, p. 21.
(46)

THE DIVINITY OF THE TEMPLE-ITS HISTORICAL FIGURE.
There is no doubt that the nature of the tutelar deity of
the ancient city is not yet fully known and interpreted. We
learn from early accounts no more than that Pachakamaj was
the supreme deity of the Incas, and Garcilaso states that the
worship accorded to him was of a spiritual nature. We also
know that the ancient Indians said prayers to the god Pacha-
kamaj on certain spots upon the roads, as do some of their
descendants in the highland to this day, and that his sanc-
tuary was in the lowland near the coast. In the highland the
cult of Pachakamaj was an ancient one, while early rec-
ords show that the Incas in the lowland adopted it shortly
before the arrival of the Spaniards. The name Pachakamaj,
• Creator of the world,' is of Kechua origin, and cannot, there-
fore, be of more recent date than the Inca conquest of the
valley. During earlier times the divinity must have had a
Ι
different name, which would agree with a statement of Fer-
nando de Santillan, that he had formerly been called Irma.
It is difficult to understand why the Incas should have given
him a name in their language, signifying the highest concep-
tion of their own religion, if it were true, as historians will
have it, that they became familiar with the god and his shrine
only after their conquest of the region.
If Dr. Brinton is correct in identifying the name of
Pachakamaj with Wiraqocha, on the strength of the testimony
of Acosta, and of an old song preserved by Garcilaso, it rather
complicates the individuality of the divinity in the ancient
sanctuary on the coast. At once the questions arise, What
relation had the ancient god Irma to the later name of Pacha-
kamaj? Was there any historical reason for the change of
name? If so, what was its nature and what caused the Incas
to make the identification ?
All we really know about the Temple is that the building
is of great antiquity, and that in the early part of its history
it was known by a different name. Cieza makes several refer-
ences to it in his writings;s also Garcilaso,4 when he tells how
the ruler of Pachacamac, in his negotiations with the Inca,
proudly mentions his forefathers from whom he had inherited
customs and laws. Montesinos gives the sanctuary a very
great age, dating it back to an early period of Peruvian his-
tory, 2000 before Christ, but as a rule his chronology is not
trustworthy. Garcilaso states that the Inca had spread the
cult of the supreme creator and preserver of all things,
Pachakamaj, through the entire realm, and that those who
adopted this faith were the ancestors of Cuismancu, who built
a Temple to this deity. His account may be regarded as pure
invention, intended to explain what otherwise seems so incom-
prehensible, the adoption of the divinity of the coastland by
the Incas as if it had been originally their own. The testi-
mony of the old chronicler cannot as yet be used for an
historical examination of this question.
One of the most important points shown by the preceding
discussion is this : that the style of Tiahuanaco, which, accord-
ing to latest discoveries, extended over the entire shore country
of Titicaca Lake, and probably included the southern section
of the Peruvian highlands, spread from the highland to the
coast country, and gained a deep influence over those regions,
also becoming established in the valley of Pachacamac, situa-
ted farther to the north. We have thus acquired a very
different point of view for the appreciation of the importance
of ancient Tiahuanacoan culture on Peruvian soil, from that
with which its historical discussion was introduced in an ear-
lier work on the monuments of Tiahuanaco. It now seems
certain that this ancient form of civilization was widely
extended in oldest Peru. The entire coast of central Peru
has been more or less influenced by it. The splendid poncho
from Ancon is found to be in the pure style of Tiahuanaco,
as well as numerous other specimens from graves in Ancon
and in the Lima valley. Pieces of pottery of the class of
pl. 5, figs. 1 and ff, are occasionally designated at Lima as
'type of Huacho,' which seems to indicate that the period of
that style is represented a little farther to the north in the
Huacho valley. The cultural relations between the ancient
works at Tiahuanaco and the province of Ancacho, estab-
lished by the discovery of the stone relief of Chavin de Huan-
tar, at first so mysterious, now seem almost proved, and doubt-
less before long will be entirely understood. The De Ville
collection at Brussels contains a vessel from Chordeleg, near
Cuenca, in Ecuador, identical with the ceramic types of the
Epigonal period of Pachacamac. A possible doubt as to its
genuineness is removed by the fact that the collections of Ber-
lin, Brussels and Philadelphia contain pieces of pottery from
Ecuador of ancient Peruvian character, and bearing traces of
the Tiahuanaco style. The vessel of Chordeleg must have
been carried from northern Peru to Ecuador at some very
early date. This points to a possible influence which the
archaic forms of the Tiahuanaco civilization may have exer-
cised as far as Ecuador. At the very beginning of the devel-
opment of ancient Peruvian civilization we see the cultural
form, which was only little inferior in the extent of its sphere
of influence to the later one introduced by the Incas.
We are as yet unable to determine its relation to the
Chimu culture, no investigations having been made in this
direction ; but it is possible that the latter either depended,
proceeded from or was strongly influenced by it. If facts
should be brought out to sustain this supposition, the culture
of the Chimu would be the most brilliant outcome of the art
of Tiahuanaco. In spreading this cultural form, its religious
side especially was carried to the various parts of ancient
Peru. The detailed description of the more important finds
has shown that by them the same religious ideas are illus-
trated which caused the erection of the Tiahuanacoan monu-
ments. We see there the same sceptre-bearing divinity, with
winged and unwinged animals of mythological nature appear-
ing as his servants. The figures of these mythological minis-
tering spirits are, as a whole, almost identical with those upon
the monuments and objects of art from Tiahuanaco; and the
ancient forms of religion as revealed by the contents of the
graves at Pachacamac must therefore have been similar in
both places. Several features have been already specially
noted. The custom of placing in the graves pieces of pottery
or textiles, with religious figures of the nature of fetishes upon
them, whereby the dead were recommended to the protection
of the divinity at the foot of whose sanctuary they were
buried, shows that the occupants were adherents accordingly
of the same religious faith. Doubtless the large cloth, bearing
mythological decorations (pl. 4, fig. 1), had been chosen for
this reason as a shroud for the dead. The grave where it
was found (p. 19, fig. 3, u) was situated quite near the Tem-
ple, which might lead to additional conclusions in regard to
4
5
6
1J. v. Tschudi, Beitr., p. 121, discusses at length the meaning of Pachaka-
maj. “Pacha' signifies 'time' or 'space, and world' means as much as earth'
or ‘soil.' The divine activity of Pachakamaj is therefore bestowed upon the
universe in general, and the earth especially 'Pachamama,'' Earthmother,'
seems to be the goddess of Earth. The worship of this goddess is still continued
in the south as far as the valley of the Calchaqui in Argentina. Both she and
Pachakamaj are said to dwell in the 'apachetas,' artificial stone-piles on the way-
side.
2 Hero Myths, p. 197.
3 Crónica, I, chap. 72 ; II, chap. 58.
+ VI, chạp. 30.
Chap 9.
(47)

10
a
monuments
Wiraqocha of the Titicaca Lake country, and with the Peru-
vian Wiraqocha, whose nature was so ably described by Dr.
Brinton. The apparently unimportant feature, alike in the
Relief of the Gateway, in the legends of the Titicaca country
and of the Peruvian Wiraqocha, that a host of mythical mes-
sengers are attending the god and sent out by him,11 serving
as mediums through whom he works, is exceedingly charac-
teristic and definitely connects the three forms. The question
as to the name of the god in whose honor those Tiahuanacoan
were reared, whether Wiraqocha, T'onapa or
Tarapaka,l2 names which all appear for the same divinity,
can only be of interest to us as far as it is connected with
the question of the nationality of the builders of the ancient
works of Tiahuanaco, which cannot yet be solved.
The sanctuary of Pachacamac originally was derived
from the cult of this creator-god of the highland, 13 who
appears in the Tiahuanaco relief, as well as in the legends
and myths, as Wiraqocha, and sometimes as Tonapa or Tara-
paka. By this conclusion various details of the ancient myths
gain a greater importance.14
12
the special significance of the mythological group upon it, in
itself an object of worship. The god pictured upon this cloth
was presumably the same to whom the Temple was dedicated,
and one of supreme rank, not inferior to the divinity of the
Tiahuanacoan Gate-relief. Though he does not bear sceptres,
he is still distinguished by similar attributes of greatness and
power buth to preserve and to destroy, such as the head-orna-
ment surmounting serpents and trophies of human heads.
The pose is similar to that of the figure of Tiahuanaco, and
ministering spirits appear again in mythological figures wait-
ing upon him. It may be assumed that in this picture we
possess one of the oldest representations of the divinity of this
Temple, which evidently was for the coastland the counterpart
of the idol of Tiahuanaco.
In the earlier book on the Ruins of Tiahuanaco, the
question concerning the principal figure of the relief was not
conclusively settled; authors were inclined to regard this deity
as a representation of Wiraqocha, the creator-god, and it did
not seem impossible that he might be a personification of the
sun. All doubts on this subject are now removed ; the
figure does not represent the sun, but a creator-god. The
Chimu had a powerful deity, the ruler of the skies, armed
with club and arrows, and in Peru the god of thunder is gen-
erally shown with these weapons. The attributes resembling
sceptres of the main figure in the Gate-relief are symbols of
the power of the god over thunder and lightning. It may be
noticed that one of the sceptres, the right-hand one, is short
and broad, without ornament, while that in the left hand is
half cloven in two. Both sceptres symbolize weapons; the
short and broad one is the club which causes the thunder, the
cloven one represents the three-armed bola, which, when
swung and thrown at a goal, in flying resembles the form of
a flash of lightning. The bola here takes the place of the
arrows, which symbolize lightning in the religious art of the
lowlands. This fact is another important proof that the art
of the Tiahuanacoan style had its origin in the highland.
The correctness of the interpretation of the two sceptres as
club and bola, or thunder and lightning, is proved by a small
circumstance not noted by others since the publication of
1892. There is a very small yet distinctly visible bird upon
the top of the right-hand sceptre or club, the thunderbird,
waiting for the waving of the sceptre, in order to fly off and
carry the darkness and terrors of the thunder-storm over the
land.? The power over thunder and lightning, attributed to
the god of the Gate-relief, entirely excludes his interpretation
as a sun-god.
The sphere of power of this divinity is boundless. Judg-
ing from the legends, he appears to be almost identical' with
5
9
6
i L. C., p. 65.
2 Relacion anónima, p. 139.
3 Stübel u. Uhle, l. C., pl. 8.
4 According to Ondegardo (v. Tschudi, Beitr., p. 58), the weather-god car-
ries a club in one hand, a sling in the other. The god of the Gate-relief, how-
ever, must at the same time be a creator-god, since the sun is represented in the
frieze in an inferior position, as if depending upon the main figures.
5 Comp. also Brinton, Hero Myths, p. 49.
6 The bola was used as a weapon in the highland of southern Peru and in
Bolivia. It may still be found in use in the Puna of Bolivia for the hunting of
vicuñas. The Uros on the Desaguadero use it for hunting water-fowl.
Myths, p. 178, Ayar Cachi, the brother of Manqo Qhapaj, represents Wiraqocha.
The Indians in northern Bolivia tell various animal and other fables. In one of
these the Inca decides a quarrel as to the relative height of the mountains, Illimani
and Mururata, by taking a sling and knocking off the summit of Mururata,
which becomes the Sajama mountain, over 20,000 feet high, more than 130 miles
distant. The Mururata appears flattened at the top, as if cut off. Similar stories
are told of other mountains.
10 Dr. Brinton, Hero Myths, p. 170 ff. Compare for the nature of Wiraqo-
cha, Acosta, B. VI, chap. 21. 'Aticsi,' conqueror, is not a surname of Wiraqocha,
as Markham states, Narratives, Introd., p. x, and Brinton, p. 170. A verbal
formaticsi,' from 'atiy,' to be able, is impossible in Kechua. The participle
would have to be atij. The surname is caused merely by the misinterpreted con-
nection of the ejaculation 'ah'! with the name of the god : “ Ticsi Wirakocha."
The prayer begins (Molina, p. 33) : “Aticsi (instead of Ah-tiksi) Viracocha," but
in another, l. c., p. 31: "A Viracocha ticsi Viracocha," etc.
11 Stübel u. Uhle, pls. 5, 11 and 16, pp. 55 and 60; Relacion anónima, p.
140 ; Garcilaso, V, chap. 18, and Acosta, 6, chap. 21.
12 Santacruz, p. 236. Brinton takes the name Tarapaka for taripaka, from
taripay, to judge (Hero Myths, pp. 174 and 184). This interpretation does
not recommend itself. It is arbitrary to replace a by i (taripay); the derivation
with ka is unusual ; the conception of Wiraqocha as 'Judge of the universe' may
have been caused by the corresponding Christian idea, but it is not to be found
among the religious conceptions of the ancient Indians of Peru. The Relacion
anónima also avoids (p. 154) reference to this side of Wiraqocha's character.
Cieza has the form Tuapaca (II, chap. 5). Tarapaca is translated as eagle' by
Santacruz. Compare paca-eagle, cocotaapaka and conturaapaka-very big eagle,
similar to a condor (Bertonio, Diccionario Aimará-Cast., p. 241). The condor,
eagle and falcon had many relations with Wiraqocha. Kon Traya Wiraqocha,
in a legend of Huarochiri, blesses the condor and the falcon (Avila, p. 127 ff.).
On the temple of Wiraqocha at Urcos an eagle and a falcon were cut into the
stone (Molina, p. 29). After having won the victory over the Chaucas with
Wiraqocha's help, the Inca had two condors carved in stone at Muyna (Garcilaso,
IV, chap. 23).
As a special emblem of distinction the Inca wore two feathers of two
different birds of the Coraquenque species, which were considered as symbols
of Manco Chapaj and of Mama Ojllo. Garcilaso, VI, chap. 28, describes the
bird as resembling the falcon. According to Bertonio, Dicc. Aim. Cast., p. 241,
and Cast.-Aim., p. 29, Kurikenke or Kurikanke is a small eagle, “speckled black
and white, and with white throat."
The Indians in Bolivia call the condor the Supreme God and a number of
pretty fables are connected with it. They avoid killing a condor. A friend of
ours in Bolivia who had a stuffed specimen of this bird in his country house, for
that reason alone enjoyed the highest respect of the Indians.
13 Very different and more unreliable is the tradition as Montesinos gives it.
According to him, a nation arrived in the plains, one part of whom built the tem-
ple of Pachakamaj, the others moved to the interior and founded Huaitara and
Quinoa. Huaitara, with ancient stone monuments, is situated in the mountains
above Pisco (Paz Soldan, Dicc. Geogr., p. 443). Not far from there, in another
valley, lies Ica, which possesses similar traditions (comp. v. Tschudi, Beitr.,
p. 193).
14 According to Betanzos and Molina, the creator sends messengers from
Tiahuanaco to Condesuyu and Andesuyu, along the coast and the wooded slopes
in the east, all over Peru, as far as Guayaquil (comp. Stübel u. Uhle, p. 55).
The existence of a huge stone image at Jauja is confirmed in Spanish
times (Descripcion de la provincia de Jauja, Rel. Geogr., I, p. 80): "and in Jauja
were monuments of the highest antiquity, which Tupaj Inca went to see.” San-
tacruz, p. 263, speaks of ancient stone figures at Jauja, Pachacamac and Chincha
Yungas, and connects them with T'onapa. According to Molina, the creator-god
turned men into stone at Tiahuanaco, at Pucana (north of Lake Titicaca), at
Pachacamac and at Cajamarca. Santacruz states that the Guancas at Jauja and
the Chinchaisuyus believe that T'onapa had been in their country ; he even iden-
tifies the divinity of the Guancas, Wariwillka, with T'onapa. This shows that
the myths point to a direct dependence of some of the ancient places of worship
of pre-Spanish Peru upon the creator-god of Tiahuanaco. The old stone images
mentioned by Molina and Santacruz as standing at Pachacamac have disap-
peared.
7 A study of the thunder-bird upon pottery of the Chimu was published by
Schurtz, Globus, 1897, No. 4.
8 The power over thunder and lightning as well as over the lives of men,
which is symbolized in the trophy-heads, go well together. According to the
Relacion anónima, p. 139, the thunder-god is the originator of epidemics, of death,
of famine. Similarly the Indians of Bolivia still believe that every flash of
lightning kills a human being.
a
9 The god causes fire to fall from the heavens; he flings flashes of lightning,
burns up entire mountains, drowns whole nations in tidal waves, but also puts
out conflagrations with his staff (Betanzos, chap. 2; Santacruz, p. 237 ff.; Calan-
cha, according to Tschudi, Beitr., p. 182). In the account of Dr. Brinton, Hero
(48)

THE DIVINITY OF THE TEMPLE ACCORDING TO THE MYTHS.
a
The peculiar nature of the god of the coastland is well
illustrated by myths connected with him. Antonio de Calan-
cha (II, chap. 29) writes as follows: “I will here relate the
labors and achievements of the god [ Pachakamaj ] : The
remembrance of Adam and Eve had been lost among these
people, as among the Greeks and Romans; they ascribed the
creation of the world to various agents and gave different
names to the first creatures or first men. It has already been
told from whom the dwellers of the mountains considered
themselves descended, and we will now consider the god
whom the people worshiped—the people of the coast, the
Yungas, the dwellers of the plain and of the sandy desert,
over all the district from Piura to Arica, three hundred
leagues in length and from twelve to fifteen in breadth,
according to the width of the coastland. One particular
legend was firmly believed in by all, up to the time of the
general visitations to exterminate idolatry, which were
extended as far as Huarmey. This legend was found to
be about the same in all villages and districts, and even at
the present day the people believe in it more than in the
articles of the Church. They did not acknowledge that
Adam and Eve were the first parents of mankind, but at
six hearings, in which more than a thousand persons were
examined, all agreed in their beliefs concerning those beings
whom they worshiped.”
P. Luis Turruel, the companion of P. Joseph de Arriago,
gives the following narrative: “ They say that the origin of
the Indians of the plain, and of their wakas' and their food,
is this: In the beginning of the world there was no food for
a man and a woman whom the god Pachakamaj had created.
The man starved to death, but the woman survived. One day
she walked abroad to seek roots among the thornbushes, in
order to sustain life. She lifted up her eyes to the Sun, and
with abundant tears and lamentations and sighs she said :
Beloved creator of all things! Why hast thou brought me
into the light of this world, if I am to die of hunger and
want? Oh, that thou hadst not created me out of nothing,
or that thou hadst suffered me to die immediately upon enter-
ing this world, instead of leaving me alone in it without any
children to succeed me, poor, cast down and sorrowful! Why,
O Sun, having created us, why wilt thou let us perish ?
And if thou art the giver of light, why art thou so niggardly
as to refuse me my nourishment? Thou hast no pity, and
heedest not the sorrow of those whom thou hast created only
to their misery. Cause heaven to slay me with lightning, or
earth to swallow me, or give me food, for thou, almighty one,
hast made me.' These and other prayers and lamentations
she poured forth to the Sun, in such outbreaks of frenzy as
the sting of hunger will cause. The Sun was touched with
pity and descended to her. Pretending ignorance, he inquired
into the cause of her complaints. She told him of her dis-
tressing life, and of having to find her nourishment in so
painful a manner as digging roots. While he listened to her
lamentations he pitied her in her sorrow, and spoke kind
words and bade her give up her fears and hope for comfort,
for she would soon be delivered from the cause of her trouble.
He bade her to continue digging roots, and while she was
doing so he impregnated her with his rays, so that to her
great joy she bore a son after four days. She believed that
the life of anxiety had come to an end for ever, and that
abundance of food would be given her. It was not to be so,
however, for the god Pachakamaj was displeased with her for
having given worship to the Sun, his father, and for having
borne him a son in defiance of himself. He seized the new-
born demi-god, not caring for the cries of the mother, who
implored the Sun to help her child, his son equally with
Pachakamaj. But the latter killed the new-born infant, his
half-brother, and cut him to pieces. The mother cried to the
Sun to revenge this fratricide. Pachakamaj did not intend
the woman to suffer through the lack of foresight of his father,
the Sun, in not producing food for her, and in order that she
should not worship any one but himself, he sowed the dis-
membered parts of the murdered boy. From the teeth grew
corn, which resembles teeth in form ; from the ribs and bones
yuccas, a root which resembles bones in shape and whiteness,
and other roots as well. From the flesh came cucumbers,
pakais and all the other vegetables and fruits. Since that
time men have known no more want, and there has been no
more weeping on account of it. This abundance of food they
owe to Pachakamaj, and the earth has ever since continued to
be fruitful. The mother, however, was not comforted, for
every fruit reminded her of her child. Moved by her sor-
rowing love for the dead son, and by her thirst for revenge,
she again dared to appeal to the Sun to punish the murderer
and to heal her sorrow. The Sun was powerless against
Pachakamaj, still he pitied the woman and descended to her,
asking for the umbilical cord of her murdered child. The
woman showed it to him and the Sun put life into it, thus
giving her another son, and bidding her clothe the weeping
child and to name it Wichama, or Willamo according to
others. She brought the child up until he grew into a fair
and strong youth. Desiring to travel all over the world like
his father, the Sun, he set out on his journey, having obtained
his mother's permission, but had hardly left when Pachaka-
maj, finding her old and feeble, slew her. Cutting her to
pieces, he caused the gallinazos and the condors to devour her.
Only the hair and bones he saved, concealing them near the
shore. Then he created men and women, who were to take
possession of the earth, and set curacas and caziques to rule
over them. Wichama, the demi-god, then returned to his
native valley, Vegueta, about a league from Huaura, and found
it abounding with trees and plants and beautiful with many
flowers. Anxious to see his mother, he looked for her, but
found her no more. He learned from a chieftain of her cruel
death, and his eyes blazed with rage and his heart burned
with sorrow. Calling the inhabitants of the valley together,
he made inquiries for the bones of his mother, and learning
where they were concealed he found them, joined them
together and brought his mother back to life. Then he
planned with her for revenge and decided to murder Pacha-
kamaj, the god. The latter, to avoid killing this second
brother also, and enraged against all mankind, threw himself
into the sea, on the spot where now stands the temple and the
city which bear his name. When Wichama saw that Pacha-
kamaj had escaped from his wrath, with the breath of his
nostrils he set fire to the air and scorched all the fields. He
then turned his wrath against the inhabitants of Vegueta
and accused them of participation in the murder by not pre-
venting it, and even of rejoicing in the deed. In his blind
rage he disregarded their prayers and pleadings and besought
his father to turn them all into stone. After having turned
into stone all the creatures of Pachakamaj, who was invisible
and could not see or help them, the Sun and Wichama
repented of their deed. They could not undo their cruel act,
but to atone for it somewhat they resolved to have the cura-
cas and the caziques accorded divine honor, also the nobles
and braves. They brought them all to the coast, setting up
some there to be worshiped as Guacos ; others they set in the
a
7
(49)

2
sea, where they now stand as rocks and reefs. They were all
to receive divine honor, with annual offerings and sacrifices
of silver-plates, chicha and espinco, thus to be propitiated for
having been changed into stone. The highest place was
accorded to Anat, the chief-a rock surrounded by the waves
of the sea, one league from shore. He was formerly the
greatest among men, and is now worshiped by the Indians
as their supreme deity.” In this legend concerning the origin
of the human race, all the Indians of Huaura, Cupi Bar-
ranca, Ancallama, Huacho, Vegueta and the dwellers of the
coast believed as firmly as we do the articles of the Creed,
as was ascertained by the missionaries, Fernando de Aven-
daño and the PP. Josef de Arriago and Luis Teruel. The
same belief was held by the Indians of Caraguaillo, five
leagues to the north from Lima, and at Pachacamac, five
leagues to the south ; also by the inhabitants of all the vil-
lages on the coast southward as far as Arica, who worship
their rocks and reefs. They believe, however, that Pachaka-
maj created the people who later filled the earth, and who
were to worship the gods and the huacas, and that he accom-
plished it in this manner : he sent four stars down to the
earth, two male and two female, from whom the chieftains,
nobles, the great, as well as the poor common people and
serfs are descended. This is the legend in which they
believed, and still do believe to this day, as firmly as we
believe in the Creed, and that is the reason why they worship
this god and the wakas.
Zárate, I, chap. 10, gives a shorter and slightly different
version of the myth : “They relate that a man came from
the north who had no bones and no joints, and when on a
march he could lengthen or shorten the distance at will, also
decrease or increase the height of mountains. This man cre-
ated the first Indians. When the dwellers of the coast caused
his displeasure, he changed the country into a desert, such as
it now is, and ordered that rain should fall there no more for
He sent them some rivers, however, which flow through
their land, so that they may at least drink and refresh them-
selves. The name of this man was Con, a son of the Sun
and of the Moon. He was looked upon as a god and wor-
shiped as such. It is said that he also provided the people
whom he had created with wild herbs and fruits for their
nourishment. After that another man came from the south,
so it is said, who was stronger than the former. He was
called Pachakamaj, which means “Creator,' and he also
was a son of the Sun and the Moon. On his arrival Con
disappeared and left the people he had created without a pro-
tector. So Pachakamaj turned them into birds, apes, cats,
bears, pumas, parrots and many other creatures which are to
be seen in the land. Then the same Pachakamaj created the
Indians of to-day and gave them for their occupation the cul-
tivation of fields and the growing of fruit. They worship
him, and all the chieftains of the land desire after death to
be buried in the province of Pachacamac, which bears his
name, since he has his shrine there. The valley is situated
at four leagues from Lima. They also say that their Pacha-
kamaj lived during several centuries, until the appearance of
the Christians in Peru, and that since that time he has disap-
peared.”
It
may be seen from these two forms of the myth that
the god personified the creative powers of the earth by giving
to men as their nourishment the fruits of the soil, which
before had not existed. He is the god who gave to man the
fruitfulness of the valleys in the plains, and as this can only
be obtained by the most careful husbandry, besides being a
god of nature, he is also a god of cultural significance. In
this he resembles Wiraqocha Pachayachachij, a world-
instructor' of the lowland, who drives out the savage and
uncultured Con. The myth, as related by Antonio de Calan-
cha, therefore, is a myth of the seasons, describing in mythi-
cal form the phenomena of nature as they are annually
repeated in the clime of the coastland. The description of
the climatical conditions, p. 1 ff, shows, as the most character-
istic feature, the annually repeated struggle of the vegetation
of the valley, which depends entirely on artificial irrigation
against the scorching heat of the sun. The former is personi-
fied in the myth in the god Pachakamaj. The Sun, with
whom Pachakamaj carries on his struggle, represents the solar
year, the first solar son, whom Pachakamaj kills, possibly the
spring-sun, before the beginning of the rising of the rivers in
the highland, when the season of fruitfulness begins. The
scattering of the teeth and bones of the murdered son pro-
duces at once the fertility of the soil. The woman who
bears a son to the Sun-god is the year. From a needy but
toil-free life in the natural wilderness, Pachakamaj leads
her into a life full of care and toil, such as the cultivation of
fields will require. Dissatisfied and still grieving at the death
of her first-born son, she implores the Sun for another son to
be the avenger of the slain one. This is Wichama, who per-
sonifies the autumn and winter sun, with whom Pachakamaj
soon enters upon a struggle. The woman grows old, as does
the year. Pachakamaj kills her, as the course of the year is
ended with the harvest, and there is rejoicing over the death
of the woman as over the harvesting. After the ingathering
of the harvest and the autumnal decrease of the rivers,
Pachakamaj is unable to resume the struggle, and the myth
explains this from the gentleness of his nature, which refuses
to kill the second son.
The flight of Pachakamaj into the ocean to escape the
wrath of Wichama corresponds to the protecting cover of
dense fogs which every winter for a time of six to eight
months overspread the parched fields of the plain. Wichama,
the Sun-hero, seeing himself deprived of the object of his
revenge, wreaks his anger upon these fields, “his eyes ablaze,
his heart kindled into flame and setting the air on fire.” All
the fields within reach of his eyes are those of the fog region,
stretching inland about ten miles from the sea, and which are
even in winter exposed to the arid sun, although they are
still partly watered. For the climatic significance of the
myth may also speak the fact, stated by Calancha, that all
inhabitants of the narrow coastland between Arica and
Huarmey believe in this myth as firmly as in their creed, for
this strip of land is principally distinguished by the climatic
phenomena which served the legend as a foundation. The
climatic form of the myth of the god is related in its origin
to the primitive myth of Wiraqocha, from whom the principle
of the god Pachakamaj has been derived. Even the name,
Wiraqocha, 'foaming sea,' points to a relation with the sea.3
6
1 Cuismanku, the chieftain of Pachacamac, was right in telling the leader of
the Inca army that his people did not need the Sun for a god, as he only caused
them trouble. In Santillan's report, p. 32, Pachakamaj tells Tupaj Inka 'that
he gave being to all creatures here below and to the Sun above.' This is ambigu-
ous and may refer as much to the difference between the creative powers of the
earth and those of the heavenly bodies as to the different forms of cult of both in
the highland and the lowland.
2 J. v. Tschudi, Beitr., p. 126, thinks it surprising that the Sun should have
so unwisely neglected to supply mankind with food, and concludes that the myth
was invented by a people of a different mode of thought, based upon a racial
difference. This difference in the conception of the Sun, however, was caused by
the difference in its climatic importance in the lowland.
3 This interpretation, which we owe to Pacheco Zegarra, seems to have
much to recommend it. Brinton, Hero Myths, p. 19 ff, mentions several facts in
support of the same. Others of similar tenor are as follows: Ondegardo, Doc.
ined., XVII, p. 80, says : “The name of the creator is Pachayachachij, which is a
general name, and when they speak of the sea they say: "Tiksi Wiraqocha,
that is the foaming sea, the origin of the world. The vast market place of
Cuzco was covered with sand from the coast to the height of two spans and a
half, in honor of Wiraqocha” (1. c.), and “the ashes of the sacrifices which were
offered up to Wiraqocha were scattered into the two rivers running through
Cuzco), probably in order to let them flow into the sea ” (comp. also Molina,
(50)

According to the oldest myths the god has risen out of the
sea, whither he returns at the end of his labors. It would
not have been easy to find elsewhere in Peru climatic condi-
tions which could more readily have been referred to the
continually repeated rising out of the sea of Wiraqocha and
return into it as on that coast of central Peru, and it may be
possible that this added to the high degree of reverence in
which the sanctuary was held all over the country.?
We may go still further. The god Irma of Pachacamac
is the same as the god Iraya of the myths of the province
of Huarochiri, in the upland of the river of Pachacamac.
Francisco de Avila gives the following account :3 “In ancient
times Kon Iraya Wiraqocha came to Huarochiri in very
ragged attire. By his word he created terraces and fields
along the steep slopes of the ravines. By wielding a simple
wand of cane he opened channels for irrigation. In the
shape of a bird he descended from a tree upon the fair
Kawillaka (“Sweetmouth'), who bore him a son as a vir-
gin. Ignorant of the paternity of her son, she called the
gods together and identified Iraya as the father when the boy
ran to him laughingly. With shame and humiliation at his
ragged appearance she fled into the sea. Iraya, however, who
was the highest of the gods in spite of his poor attire, desired
to win the favor of the goddess. He, therefore, clothed himself
The god Iraya evidently signifies productiveness of the
soil as the result of cultivation. The staff with which he
opens channels suggests the sceptre wielded by the thunderer
Wiraqocha of the highland. In the mountainous upland,
with its narrow valleys of inferior fertility from climatic
causes, the otherwise powerful god appears in beggar's
attire. But while following the fair Kawillaka downstream,
he dons splendid robes and “by his splendor lighted up the
whole country," just as the golden cornfields in the plain are
shining brightly between barren strips of desert. Thus he
transforms himself again into the resplendent and powerful
god Irma of Pachacamac. To him it is not difficult to
pacify the serpent who is guarding the daughters of Pacha-
kamaj. His wrath at the absence of Urpi Wachaj is natural,
since she is his wife. By filling the sea with fish he ren-
ders a cultural service to mankind, one that would become
Pachakamaj himself. As the personification of the fertility
of the soil, Pachakamaj marries women who represent rivers.
The fish-pond of Urpi Wachaj, Pachakamaj's wife, seems a
sufficient proof of this relation. The 'mother of the doves,'
Urpi Wachaj herself, may denote the river of this valley.
The season of the year when the god shows himself in the
valley in his resplendent robes is the time when the rivers
are swollen by the rains in the highland. Then Urpi
-دوسی
متر
Coca
ca
--
se
THE
aza
-تحص
1 /
ma
23
Fig. 58. Ruins on the Mama river, above Chosica.
in splendid golden robes, and by his splendor lighted up the
whole country' and followed her, begging her to turn back
and'look upon him, but she refused. When he reached the
shore, she had turned to stone in the ocean. The wife of
Pachakamaj, Urpi Wachaj, 'mother of doves,' was visiting
the goddess Kawillaka in the sea, and he found only her two
daughters, guarded by a large serpent. While he was trying
to pacify the latter one of the daughters made her escape in
the shape of a dove. Angered by the absence of Urpi
Wachaj, he emptied her fish-pond into the sea, whereby
the sea became stocked with fish, while originally it had
contained none.
Wachaj is absent in the sea; the waters of the river flow
in a current of a different color far out into the sea toward
the Pachacamac Islands, the rocks into which Kavillaca and
her child were changed. The ponds of the legend are always
numerous in winter. They are formed by the power of the
surf, which is stronger than that of the river, and so the
water is backed up in pools along the coast, diked in by the
sand washed into the mouth of the river. Upon the appear-
ance of the rising waters these dikes are broken and the
ponds are empiied out into the sea, which in the myth is
ascribed to the wrath of the god. The water-course of the
Lima valley is also regarded as Pachakamaj's wife. At San
Pedro de Mama, in the neighborhood of which is the conflu-
ence of the headwaters of the Rimac of the Lima valley,
there stood a famous temple dedicated to the wife of Pacha-
kamaj. These two rivers were her breasts. The fertility of
p. 49). The sea, therefore, seems to have been considered as the origin of the
world.
1 Wiraqocha came out of the sea, according to different legends (Betanzos,
chap. 1; Garcilaso, I, chap. 18 ; Oliva in v. Tschudi, Beitr., p. 130), and returned
into it (Cieza, II, chap. 5 ; Santacruz, pp. 239, 240; Betanzos, chap. 2). The dis-
appearance of Pachakamaj over the sea at Pachacamac is told by P. Maestro
(Lima, 1602). According to S. Jimenez de la Espada, Congr. des Amér., Brux-
elles, I, p. 593, in the same manner as that of Wiraqocha in the myths on Lake
Titicaca.
2 From all these sources it is clear that Pachakamaj should not be regarded
as originally “a simple tribal deity of the Yunka,” as Bruehl, p. 465, calls him.
3 Markham, Narratives, p. 124 ff.
* Compare the piece of fabric, pl. 4, fig. 1, and figs. 10 to 12, in the text,
6 In the central figure of the design, fig. 1, pl. 4, interpreted as the divinity
of Pachacamac, the sceptres seem to be replaced by the knife and the severed
human head, that of the murdered solar son. Thunder-storms being entirely
unknown in the coastland, the omission of the staves symbolizing them in the
highland seems to explain itself.
Comp. Santillan, p. 32.
8 Descripcion y relacion de la provincia de Yauyos; Rel. Geogr., I, p. 75.
Mama is synonymous for breast' and 'mother.'
p. 24.
5 Doves, which live on grain and are only to be found in richer countries,
go well with Pachakamaj as the god of fertility of the soil.
(51)

a
a
the valley of Lima was apparently the child of this divine
pair, nourished by the breasts of the goddess. Starting from
Lima in the company of a friend, the writer visited a ruin on
the left bank of the Mama river, the left arm of the head-
waters of the Rimac, close above its confluence with the right
arm. The ruin is an extensive though plain building of
broken stone, with doors and windows, and having several
terraces in front which slope down toward the river, Fig. 58.
As this building is distinguished above all others in that
neighborhood by its size and importance, it may be the ruin
of the temple dedicated to the wife of the divinity, which is
the subject of the preceding chapter.
THE GOD AS THE CENTRE OF AN ORGANIZED NATURE CULT.
Occasionally, however, images of the creator, 11 of the
sun,
12 and the idol of the sanctuary on the mountain of
Guanacaure near Cuzco are called wakas. Still this use of
the word may be an erroneous one, as we know that the
Incas were zealous persecutors of the wakas, while they wor-
shiped the ancient idols, and before they acknowledged the
sanctuary of Pachacamac.13 The word waka
The word waka may be supposed
to designate the powers of nature, especially the productive
force of the earth, in contrast to other and more independent
powers. The wakas, generally speaking, were divinities of
local character, dwelling in springs, mountains and caves, also
mummies of the dead, who by their return to the creator had
become mysterious powers. These powers of nature were
grouped about the figure of the creator as emanations of the
We do not know if all the wakas of the realm since
the earliest centuries were thus connected with the creator-
god of Pachacamac. Certainly something of this kind must
have been the case in Pizarro's time, when Atawallpa told
him that, although each village had its own ‘mosque,' yet the
common divinity of all the people together was at Pacha-
same.
camac.
One of the most important clues to the interpretation of
the ancient god Pachakamaj is the account of the Inca realm
by Fernando de Santillan. From this alone may be gained
sufficient information concerning his significance as the creator-
god throughout this coastland. According to this account
Pachakamaj formed the centre of the waka' cult, which
possessed a fundamental importance among the Peruvian
religions. With the acceptance of the cult of Pachakamaj,
that of the 'waka' came to be generally adopted in the land
of the Incas. The original meaning of 'waka' is not known.
It corresponds in a general sense to the Latin “sacer.'2 This
word designates all idols, temples, all extraordinary products
of nature, either attractive or repellent, especially monstrosi-
ties, graves, sacrificial offerings; all such localities or objects
in nature which seem to reveal the creative power or a mys-
terious presence of the divinity, such as high mountains,
springs, caves, etc.
In many of these places the god had an oracle. This
idea seems to have been attached especially to stone images.
Santillan, therefore, interprets the word waka as “ certain
stones through which the devil spoke.” The real meaning,
however, is not so simple. As we are surprised to learn in
the Relacion anónima that the images of gods were called
'willka' and not 'waka,' we are furthermore obliged from
other statements to distinguish between images which were
waka’ and those which were not. Thus Santillan distin-
guishes the cult of the sun, the moon and the earth from
that of the wakas. He repeatedly speaks of “the wakas and
the Sun,” “ neither the Sun nor the wakas.” Molina makes
a distinction between the sacrifices offered to the creator and
those to the wakas. 6 The Relacion de la Prov. de Vilcas
Guaman, l. C., says also : “In ancient times they worshiped
the wakas, and after the subjection under the Incas also the
sun, the moon and the special idols of the Incas.” From the
foregoing quotations it appears that the images of the sun,
the moon, the earth and the creator were not generally taken
for 'waka,' being specially designated as 'villka.?? Thus Ber-
tonio in his dictionary gives for 'villca' the meaning of
adoratorio dedicado al sol o atros idolos,” and “villca el sol
como antiguamente decian.”8 Some other idols of great
importance or of special antiquity are termed 'villka'-
so an idol mentioned by Santacruz, “Guarivilca,' the chief
idol of the province of Jauja,' and 'Aisavilca,' 'a great
devil of the neighborhood of Lima.' 10
966
A people of the simple intelligence of the Peruvians
could not have preserved this religious system in an abstract
form. It was, therefore, by the priests adapted to the wants
of the people and clothed in the garb of sensual relationship.
Besides Pachakamaj, there were four principal shrines in the
realm-- Mala, Chincha, 14 Andahuailas and, since the conquest
of the valley by the Inca, Cuzco—which were called the sons
of Pachakamaj. “By these sons," says Santillan, “ the wakas
increased more and more, for the devil who spoke through
them made the people believe that they could bear children,
and caused them to build ever new temples and chapels for
those who, according to their belief, were the descendants of
the wakas, and as such considered to be their gods. Some
they worshiped as men, others as women, and they laid their
wishes before each of them according to his special nature.
One they would implore to cause the rain to fall, another to
make corn to grow and fruit to be abundant, others again
to cause women to bear children and so forth.15 In this man-
ner the wakas were enormously multiplied, that at present
there is a waka for nearly everything, and by their means
the devil keeps the people so blindfolded that the greatest
difficulty in that country is met in trying to teach the natives
our holy Faith.” Some of the wakas may have won their
לי
3
9
>
1 Santillan, p. 32 ff.
2 The view of J. v. Tschudi, Beiträge, p. 152, that it meant 'cave,' lacks all
foundation.
3 Comp. J. v. Tschudi, Beitr., p. 150 ff.
4 P. 34. Similarly the Descripcion de la Provincia de Vilcas Guaman, Rel.
G., I, 149. “Wakas, which were images of stone, and which stood in certain
places set aside for them.” Molina, p. 7, speaks of wakas which had been turned
into stone. Zarate, I, chap. 11: “Of certain stones which the Indians worship
and call wakas."
5 v. Tschudi, Beiträge, p. 171, doubts this statement (not in his Wörter-
buch, p. 99).
6 Pp. 30, 34, 35.
7 Willka are also the highest classes of the priesthood (v. Tschudi, Beitr.,
11 Betanzos, chap. 5.
12 Zárate, I, chap. 11 : “Certain stones which represent the sun and are
called waka.”
13 The mythical Manqo Qhapaj was looked upon as hostile to the waka
(Santacruz, p. 244). The same is said of other Incas, such as Maita Qhapaj
(p. 255), Pachakutij (p. 280), Tupaj Inca (p. 283), etc. V. Tschudi's statement,
Beiträge, p. 158, that the ancient Peruvians had never distinguished between the
waka and their anthropomorphized divinities, and had given them the same
names, from their chief gods down to any kind of monstrosity or abnormal crea-
tion, does not seem to be quite accurate.
14 Also Garcilaso, VI, chap. 18, terms the god Chinchaikamaj an early imi-
tation of the god of Pachacamac. The chieftains of Chincha and Mala, among
others, appeared at Pachacamac with presents during Hernando Pizarro's sojourn
at that place (Estete, p. 133).
15 “ Provinces had special wakas as well as villages, owing to the general
ramification of the sanctuaries” (J. v. Tschudi, Beiträge, p. 150).
p. 171).
8 Diccionario Aimará-Cast., p. 386.
'P. 262, v. Tschudi, Wörterbuch ; p. 315, Huari, god of strength.
10 Santacruz, p. 275.
(52)

6
7
1
ner.
10
fetish-like significance in that they derived their attributes
indirectly from the great Creator and were appealed to inde-
pendently for certain matters only, according to the division
of power. The fear of the all-embracing creator-god was
hereby enhanced, as in him was seen the comprehensible
source of all those powers by which man is influenced every
day of his life and which for that reason were objects of wor-
ship. 1 Only after all this is known is it possible to under-
stand the extreme degree of awe with which the sanctuary of
Irma was invested for believers. Hernando Pizarro” says in
reference to this statement: “ The temple is held in such fear
by all the Indians that they believe, if any of the priests
belonging to it should ask them for anything, it would mean
instant death to them to refuse.” This seems natural, as
they were surrounded by the power of the god in innumer-
able sub-agents, each of which was specially feared and rever-
enced, even if the entire being of the creator-god remained
incomprehensible to the people. In view of this, it is not so
difficult to understand that pilgrims flocked to this sanctuary
from great distances, even as far as eight hundred miles, since
every one of their own local sanctuaries was but a branch of
the shrine of the supreme deity at Pachacamac, and that
envoys were willing to fast a whole year before being per-
mitted to invoke the deity for a good harvest for their chief
and his country. This also explains the large number of
wooden images standing, according to Estete, in all the streets
of Pachacamac, at the principal gates and all around the tem-
ple, which the people worshiped “as effigies of their devil.”
All the different powers emanating from the creator-god were
embodied in these images, thousands of which surrounded the
people in this city and formed the objects of their worship.
It is, therefore, easy to understand that it would have been
very difficult for the Incas to abolish the sanctuary, even had
they desired to do so, on account of its powerful organization,
as Cieza assumes. Even before the Incas had reached Pacha-
camac in their conquest they had the hardest fights with the
faithful adherents of the sanctuary. There are
indications that the struggles of the Incas with the Chancas,
and which nearly destroyed the flourishing empire of the
Incas, were not fought for political reasons only, but also on
account of important religious differences between the adher-
ents of Irma and those of Wiraqocha of Cuzco. The Chancas
occupied the region of Andahuailas, whence they had at some
early period driven the Kechuas.? At Andahuailas stood a
minor shrine of Irma, according to Santillan. The priests
had instigated the last decisive struggle and the last rebellion
of the Chancas, previous to their final defeat, and the Chancas
of Andahuailas were allied against the Incas with the Chin-
chas, who possessed another of the small sanctuaries of Irma.
Montesinos ° points out the religious motive for these struggles,
saying that the Chancas wished to “subjugate the son of the
Sun.” The god Wiraqocha was also said to have taken side
with the adversaries, the Incas, in the most wonderful man-
He had given them warning of the outbreak of the
rebellion long before it occurred. In the greatest stress of
the battle the Inca implored Wiraqocha for help, 11 which he
granted by personally entering the fight with auxiliary
troops. 12 All this indicates that the Incas in these struggles
were actuated by religious motives. The outcome was an
increase in the fervor of the worship accorded to the god, as
well as various monuments erected in his honor in fulfillment
of vows,13 and the entire religious and political subjugation of
the enemy. The cult of Wiraqocha was forced upon them
by the victors,14 a number of the priests of the idols were
made prisoners, brought to Cuzco and deprived of their
dignity forever, while a new sacerdotal order was intro-
duced.15 Henceforth the priests should always be poor and
of lowly extraction, and in future rebellions they should be
subject to the common law and to the death penalty. For
this reason the Inca was surnamed Pachakutij, he who
changes the world, perhaps meaning as much as the great.'
With minor sanctuaries the Incas could deal in this manner,
but political and religious principles must necessarily be influ-
enced and purified as soon as they came into closer relations
with the chief sanctuary of Irma.
3
numerous
1 Apparently we have to deal here with a transformation of the highland
creator-cult to correspond with the religious views of the coast dwellers. In the
highland we notice the worship of the powerful creator, together with a leaning
toward the development of a waka-cult. Molina, p. 4, says: "The creator
formed models of various nations and caused them to spring from the earth.
Some came forth out of caves, others rose from springs, again others descended
from trees. For that reason and according to the places from which they came
they made wakas and places of worship of them.” The Relacion anónima,
p. 137, states: “The devil told them that the creator had invested various crea-
tures with some of his power and divinity, and that these were attendants and
ministering spirits of the supreme deity and had their dwelling in the heavens,
as the sun, the moon and the planets." We are hereby reminded of the winged
figures in the Relief of the Gateway Compare with them also the interceders
with Wiraqocha, mentioned in Coleccion de documentos ineditos, XXI, p. 142 ff.
The changes which religions undergo with different races of men, and under
different climatic conditions, are as a rule very great, as is shown by the differ-
ence between the Buddhism of Thibet and the original Hindostanic form of the
same religion. A similar process may have taken place in Peru. It was a pecu-
liarity with the Yunka peoples to have an exceedingly large number of religious
observances, as may be seen by the numerous symbolical figures on the Chimu
pottery. Under the influence of such a highly gifted people the cult of the crea-
tor-god of the highland might well have assumed the peculiar form of worship
accorded to the same divinity under the name of Irma.
2 Letter, p. 124.
3 Pizarro, Letter, p. 124.
4 P. 132.
5 Crónica, I, chap. 72.
6 Garcilaso, IV, chap. 15 ; V, chap. 17; Bétanzos, chaps. vi-x; Cieza, II,
chaps. 44, 45.
7 Garcilaso, IV, chap. 15; Cieza, I, chap. 91.
8 See above, p. 52. Stübel u. Uhle, p. 579 and 60a, note 1. The writer
erroneously concluded from the fact that Wiraqocha was worshiped at Cuntisuyu,
that he would also be adored at Andahuailas. Cuntisuyu is not the region about
Andahuailas to the west from Cuzco, but to the south on the other side of the
Cordillera, near Cotahuasi and Chumbivilcas.
9 Relacion anónima, p. 172. Garcilaso, who pays less attention to the reli-
gious element, and whose historical interest in the Incas was more especially
directed toward the Sun-cult than toward that of Wiraqocha, speaks nevertheless
of an alliance of the Chancas with Chinchaisuyu in these struggles (Book IV,
chap. 23). It must be mentioned that according to Balboa, a chieftain, Cus-
manco of Conchucos, guarded the rear of the Chancas during their retreat, and
later on was himself defeated and killed by the Incas. In the name of this
chieftain we have a resemblance to that of the chieftain of Pachacamac.
Cuismanco, who was defeated by the Incas.
10 Garcilaso, IV, chap. 21.
11 Betanzos, chap. 8.
12 Garcilaso, V, chap. 18; Betanzos, l. C.; Acosta, VI, chap. 21.
13 Garcilaso, Betanzos, Acosta, Ondegardo in Col. de doc. ined., XVII, p. 15;
also compare Stübel u. Uhle, p. 60.
14 Montesinos, p. 175.
15 The religious reforms are confirmed by Blas Valera, Garcilaso, IV,
chap. 36,
CONQUEST OF THE SANCTUARY BY THE INCAS.
The chronology in Peru, up to the last decades before the
conquest by the Spaniards, is at best very unreliable, in con-
sequence of the entire lack of any kind of script and of the
inadequate makeshift of knotted strings (kipus) as the only
assistance to memory. The date of the incorporation of the
valley of Pachacamac into the Inca empire comes under this
same chronological difficulty, although that occurrence was not
a very remote one from the arrival of the Spaniards in Peru.
According to Garcilaso the last ruling Incas had the fol-
lowing dates :
(53)

7. Wiraqocha
6. Pachakutij
5. Yupanki..
4. Tupaj ......
3. Waina Qhapaj.
2. Waskar.....
1. Atawallpa...
... 1289–1340
..... 1340-1400
..... 1400–1439
...... 1439-1475
.... 1475–1526
... 1526–1531
.... 1531–1532
1
10
11
The statements of the various authors concerning the
name of the conqueror vary between the four Incas :
7. Wiraqocha, according to Montesinos.
6. Pachakutij, according to Garcilaso. 1
5. Yupanki, according to Balboa and Santacruz. 3
4. Tupaj, according to Cieza 4 and Santillan. 5
Following the order of the names in the lists of the
Qhapaj Incas, as given by the different authors, we find as
the sixth Pachakutij, according to Garcilaso ; Wiraqocha
and Yupanki, according to Montesinos, Balboa and Santa-
cruz, the fifth from the last. Tupaj was the fourth from the
last according to all the lists.
The conquest of the valley would therefore fall between
the years 1340 and 1400, according to Garcilaso, or, since it
was one of the later achievements of the Inca (Pachakutij),
probably not earlier than 1360, about one hundred and
seventy years before the Spanish invasion. According to
Balboa, Santacruz and Montesinos, the event took place
between the years 1400 and 1440, ninety to one hundred
and thirty years prior to the arrival of the Spaniards, and
according to Cieza and Santillan it was even as late as 1440
to 1450, about eighty years before the Spanish conquest.
Whatever date between the years 1360 and 1450 may finally
be accepted for the conquest of the valley, the fact remains
that the Inca period of the valley and the sanctuary, in pro-
portion to the age of the latter, must be considered a very
short one, while the transformations which the city underwent
in its appearance were relatively considerable.?
Garcilaso gives a detailed account of the political sub-
jection of the sanctuary under the Inca rule. The Inca’s
army came from the south.8 The Inca first sought to force
the prince of Pachacamac to an unconditional surrender and
renunciation of the original cult. The discussions held on
this subject, however, convinced the Inca that the divinity
worshiped under the name of Irma was really identical with
his god Pachakamaj. He, therefore, felt induced to moderate
his conditions, leaving the prince in possession of his lands
and cult, and receiving him with full honors as an ally and
near relative of his own family, incorporating his possessions
into the Inca realm. However the prince was obliged to con-
cede changes in his cult and to accept that of the Sun in his
lands in addition. J. v. Tschudi thinks this account untrust-
worthy and partial to the Inca, since it represents him as
reaping advantages without any apparent struggle. The
details he considers incorrectly stated, since other chroniclers
tell of a voluntary submission by the Inca to the foreign
cult.' Garcilaso's account must, nevertheless, be considered
the most complete and correct for several reasons. A volun-
tary submission by the Inca to the foreign cult did not take
place, since the Sun-temple was erected on the finest and most
prominent site in the ancient city of Pachacamac,1° the Inca
thus reaping advantages for his own religion. The details
which Garcilaso is said to have incorrectly stated are under
dispute, and even if they should be proved to be false
would not include the principal terms of peace as stated by
Garcilaso. In his account of the recognition by the Inca of
his own supreme deity in the divinity of Irma, J. v. Tschudi
sees a palliation of the fact that the Inca had submitted to a
strange god, and in order to conceal this and for the purpose
of introducing the new cult among his own subjects he had
given it a Kechua name. Tschudi, however, did not realize
that such a superficial interpretation of the historical facts
only renders much more inexplicable the renaming of the
deity as Pachakamaj, 'world-ruler,' by the Inca. Garcilaso’s
account, therefore, corresponds in even a higher degree with
the real facts, and is also in itself more harmonious and prob-
able. It tallies, besides, with the account given in the preced-
ing chapters of the derivation of the sanctuary at Irma from
the creator-god Wiraqocha (Tonapa, Tarapaka) worshiped
in the highland. The course of events may have been as
follows: The cult of the creator-god Irma was derived from
that of the creator-god of the southern highland, who in
different localities had different names, Wiraqocha, T'onapa,
Tarapaka. This original cult of the highland was of pre-Inca
date. The Inca realm grew up in it and retained it. In the
course of time the two kindred cults of the highland and
lowland became estranged, to be later reunited after the con-
quest of the valley of Irma by the Inca. Mindful of the
religious kinship, the Incas reintroduced this cult, while it
served their political purposes to adopt the sanctuary again as
one of their own. In the kindred deity of Irma, the giver of
all the fruits of the fields and the preserver of the powers that
6
12
1VI, chap. 30,
2 Balboa relates in chap. 8 that Tupak visited the temple of Pachacamac
and found it managed in accordance with the statutes of his father, Titu Yupanki
Pachakutij. This would seem to indicate that the predecessor of Tupaj subju-
gated the valley of Irma.
3P. 275.
4 II, chap. 58.
5 P. 32, p. 172.
6 A mean date of the conquest of the valley by the Inca is the most prob-
able one. Garcilaso has two Incas by the name of Yupanki, the older of whom,
immediately preceding the younger, was distinguished by the surname of Pacha-
kutij, under which he is known in history; the other chroniclers only speak
of one Yupanki having the same surname. Thereby the existence of another
Inca Titu Yupanki Pachakutij, as distinguished from the former, becomes doubt-
ful, being insufficiently supported. On the other hand the accounts of Cieza and
Santillan of Tupaj's visit to Pachacamac admit that Pachacamac may have been
conquered under the preceding ruler. The conquests in the highland were
pushed as far as Jauja and Tarma under the preceding Inca Yupanki (Cieza, II,
chap. 50).
Concerning the great age of the sanctuary, compare above, p. 46. It is not
possible at present to estimate correctly the age of the temple or to state the
exact date of its foundation, nor even the century. The ancient monuments of
Tiahuanaco were the outcome of a civilization which flourished in southern Peru
prior to the beginnings of the development of the Inca realm, therefore at least
as early as 1100 A.D. (see Stübel u. Uhle, pp. 52-54), and since the sanctuary of
Irma (Pachacamac) may have been founded at the same time its approximate
age would be the same. But we do not know as yet how long before the founda-
tion of the Inca empire the works of Tiahuanaco were constructed. The high-
land of southern Peru having been but little explored, we are not in a position
to know anything about the genetic relation of the Cuzco or Inca style to
that of Tiahuanaco, from which it differs strongly. Nor can we tell if the deriva-
tion from the same was a direct one or transmitted by other intermediate
types. The final settling of the question as to the relatively earlier or later date
of the works of Tiahuanaco, before 1100 A.D. approximately, will depend upon
the determination of these doubtful matters.
8 Garcilaso, VI, chap. 20. According to Montesinos, Cieza and Santacruz,
the Inca who conquered Irma came from the north. The first visits of the Incas
seem to have been from an eastern direction, from accounts of Santillan and
Balboa.
9J. v, Tschudi, Beiträge, p. 128.
10 The erection of a temple of the Sun is mentioned as one of the condi-
tions of the conclusion of peace in Garcilaso's book. Below, chap. xix, will be
seen how the architecture of the Sun-temple on the one hand and that of the
women's convent on the other was in keeping with the style of the people who,
according to Garcilaso, were obliged to construct these two buildings in fulfill-
ment of peace conditions.
11 Garcilaso mentions idols which had been removed from the temple of the
creator by order of the Inca. He made a mistake if he thought that this had
included all the figures there, for Pizarro and his companions found an “ugly
image” of the divinity in the building (Pizarro, Letter, p. 124; Estete, p. 131).
However, no record is found in Spanish reports of the conquest of the idols of
fish and fish-like character (“zorra ') “which the Yunkas had set up in the tem-
ple previous to the times of the Incas.” To connect with these the image of the
god which Pizarro found and to call it fish-like (“the wooden fish-idol”), as
Markham does (Reports of the Discovery, Introd., p. xvii, ff), is absolutely
unsupported by facts; though there were fishes, etc., painted upon the walls
of the terraces and temples (see above, p. 21).
12 The statement made by Garcilaso that the cult of Pachakamaj had been
spread to the coast in earlier times by the Incas rests evidently upon an excel-
lent tradition, which claims that the original cult of Irma was derived from the
highland, and it proves the value of the sources from which Garcilaso drew his
material when he wrote his history. He, of course, ascribed the spread of the
cult to the agency of the Incas themselves, since he, together with many early
chroniclers, ignored the existence of a pre-Inca period.
(54)

3
1
slumber in the earth, they saw embodied principally those sides
of their creator-god which correspond to his surname 'world-
ruler.' Thereby the sanctuary came to be called by the name
of Pachakamaj instead of that of Irma. Only the recognition
of this kinship of their religion to that of the worshipers of
Irma can explain the unusual amount of honors with which,
according to Garcilaso, the Inca received the prince of Pacha-
camac into his realm, and the high veneration which the
Incas bestowed henceforth upon the sanctuary throughout the
days of their prosperity.
Each of the succeeding Incas either visited the sanctuary
in person, consulted the oracle of the god and celebrated bril-
liant festivals with many sacrifices, or sent to consult the
oracle on all the most important affairs of his empire. Since
the conquest, the god spoke through the oracle to princes and
caziques only. 3 Tribute was partly paid at this shrine instead
of being sent to Cuzco, 4 and some of the Incas bestowed great
riches upon it. It was only with the overthrow of the empire
and his own captivity at Cajamarca that Atawallpa became
shaken in his faith in the ancient religion, and in the trust-
worthiness of the oracles of the god. As one of the last acts
of authority before he himself was put in fetters he ordered
the chief priest of the sanctuary to be thrown into prison,
when he visited him at Cajamarca, “as the god had counciled
him to wage war on the Christians, and had made the false
promise that he would slay all of them.” 5
1 Markham, l. c., disputes the assumption that the divinity of this sanctuary
had been a creator-god, and that he was himself called Pachakamaj. If the fish-
form is not in keeping with the idea of a creator-god, the fact that the god was a
creator-god and was called Pachakamaj forms a refutation of the fish-shape
(comp. Cieza, Crónica, I, chap. 73 ; II, chap. 58 ; Santillan, p. 32 ; Estete, p. 132 ;
further the proofs furnished by the objects found there-the historical derivation
of the deity, the myths, the importance of the god as the centre of the waka cult).
The Kechua name Pachakamaj being given to the god, also designated his sanc-
tuary, and this in turn caused the city to be distinguished by the same appellation.
2 Montesinos, chap. 26: Inca Wiraqocha sacrifices many llamas after the
conquest. Garcilaso, VI, chap. 32: Pachakutij, the conqueror of Irma, visits the
shrine and worships there before setting out on the campaign against the Chimu.
Santillan, p. 32: Tupaj visits the temple, consults the oracle of the divinity and
makes great sacrificial burnt offerings of llamas and garments. Similar, Balboa,
chap. 8. Also, Cieza, II, chap. 58: Yet upon this occasion the god demands
human sacrifices also. Santacruz, p. 302: When Waina Qhapaj visited his shrine,
the god asked for “more buildings” and for riches. Cieza, II, chap. 65 : Waina
Qhapaj celebrates brilliant festivals at Pachacamac in honor of the god with
dances and feasting, and makes rich presents to the temple; in return for which
the god sends him favorable oracles. Similar, Montesinos, p. 229. Comp. also
Xerez, p. 108. Santacruz, pp. 314, 315 : During the war between Waskar and
Atawallpa the leaders of both armies send offerings and cause the oracle to be
consulted, whereupon victory is promised to both of them.
3 Garcilaso, VI, chap. 31.
4 Hernando Pizarro, Letter, p. 123; Estete, p. 132.
5 Xerez, p. 115.
(55)

CHAPTER XVI.
THE CITY WITHIN THE INNER WALLS.
C'ROSS
a
a
YROSS streets divide the ancient city into quarters or sec-
tions; one-fourth of its entire area being occupied by
the temple district, there remain sections B, C and D, as will
be seen on the map of the city. As the eastern street does not
extend quite so far as the eastern limit, the dividing line
between B and D is not distinct and the division of the city
into sections has not been completely carried out. The
eastern district, section E, appears to be different from the
rest of the town by reason of its situation, being separated
by hill Z and built upon slightly rising ground; the build-
ings, too, show some peculiarities. In comparison with sec-
tion B, the sections C, D, E appear to be of more recent
date, especially the parts situated in the plain. The division
of the city's area by walls is upon a large scale, and the
ground is not so closely cumbered with buildings. The
graves in sections C, D, E are limited to some particular
spots, such as the courts of the dwelling houses, while the
ground of section B is filled with graves, which are, as a
whole, of much greater age than those in the other three sec-
tions. The ground is full of ancient masonry, discovered
unexpectedly during explorations; here and there traces of
ruined buildings were found, covered by a thick stratum
of apparently natural soil. Depressions in the ground seem
to be ancient courts, a, a; mounds are found to be débris of
buildings, 6.1 From all these indications there is no doubt
that section B, situated between the hills X, Y, Z, is the
oldest part of the city, situated nearest to the sanctuary of
the god. There are no closely built-up rows of houses in
Pachacamac, nor do we find such labyrinthine crowding of
rooms and buildings as may be seen in the ruins of Caja-
marquilla, about eleven miles above Lima, 2 and in the remains
of the village in the valley above Pachacamac. The city for
the most part consists of large detached buildings of palatial
proportions, some situated with their courts toward the street,
others lying within the sections. A considerable area has
been cut into court-like sections by walls, devoid of any
buildings; which are also wanting in another part of the city,
B, the oldest section. The general ground plan of construc-
tion in parts B, C, D corresponds to that of the temple and
follows the coastline, while that of the secluded section E is
different, as is also the shape of its structures. On the
whole, the ground plan of the city resembles those of the
ruins in the neighborhood of Lima, as, for instance, that of
the ancient town of Huatica, and in a modern sense has more
resemblance to a suburb than to a city. A group of closely
crowded buildings is seen in the extreme eastern part, in one
locality only, in E; here were also found several indications
of the architectural influence of the highland, and to the same
influence probably may be ascribed a divergence in the
manner of building.
The city wall enclosed the town completely upon the
most exposed northwestern side; an addition to it, two hun-
dred and sixty feet long, is joined to it on the northern end,
forming an imperfect enclosure of the northeastern side, leav-
ing between the wall and a hill a strip of ground for a road,
which may have served since early times for intercourse
with the southern parts of the coast. The wall of the north-
western front and the northeastern section are eleven to thir-
teen feet high and about eight feet broad, while, judging from
its foundations, the enclosure around C on the west side must
have been about sixteen feet broad. This section at present
is the most ruined, which accounts for the numerous gaps in
it. The city wall is in a very dilapidated condition, and
according to Estete 4 it must have been so for a long time.
The northern and eastern streets have a width of sixteen and
thirteen feet, and side walls similar to the city walls in
breadth and height, so that the width of the streets was the
same as the thickness of the two walls flanking them. A
şort
of
open square appears on the point where the four
streets cross each other, and various piles of small stones are
to be seen there. Another regular square is at the end of
the southern street, and closely adjoins the temple enclosure.
Both the city walls and the streets were of rather unusual
appearance; the wall was more exposed to and the houses
more secluded from the streets than is usually the case.
Estete speaks of city gates, wherein idols were set up. It
is doubtful, however, whether there were any means of clos-
ing the entrances to the city besides the guard-houses and
the idol gates. Explorations in the extreme ends of the
northern and western streets brought to light no traces of
them. The city was almost open on the most important side,
the northeast, where much traffic passed along the road.
That the streets 6 were used by crowds of people is proved
by the condition of the remaining well-preserved sections of
the adobe walls along the eastern street, which are worn
quite smooth by much use. The number of entrances from
the streets into the palace courts are in no proportion to the
size of the areas which they enclose, for it is still clearly to
be seen, in spite of the ruined condition of the walls, that
but few entrances existed and that some of the palaces could
5
6
3
1 In certain places buildings have been superposed upon others, thus one
wall near c extends over an ancient cemetery. The terraced house near d stands
above older sunken rooms.
2 The plan of a part of Cajamarquilla, Squier, p. 93, gives an idea of the
crowding of buildings in that place.
3 They are not rectangular, as they appear to be and as stated by Wilkes,
4 See above, p. 11.
5 A similar arrangement may be noticed in the ancient city of Surco, near
Chorillos.
6 One of these is shown in pl. 10, fig. 1.
p. 288.
(56)

only be entered from the interior of the courts. It appears,
therefore, that the different sections were much more difficult
of access than the city itself, as may be seen in the bar-
rios' or wards of the ruins of Trujillo, which, however,
resemble these in no other respects. The streets must have
been used mostly by strangers, pilgrims, etc., and have served
also to connect the different buildings.
1 Squier, p. 159 ff.
TYPES OF BUILDINGS.
Among the foremost in number and prominence are the
terraced houses which are scattered over various parts of the
city. Some of these houses may be seen in pl. 9, fig. 2 and
pl. 10, fig. 3. The characteristic feature of these buildings
is that they are set upon massive terraces from ten to seven-
teen feet high, which they entirely cover. An inclined path
leads up to them, seven to sixteen feet in width, furnished in
some instances with side-rails. 1 Above the first terrace, and in
the rear, rise similarly constructed grades of a lesser height,
of from three to five feet each, approached by winding paths.
The terraces reach to the top of the building, sometimes
twenty-seven feet and more in height. The rooms within the
houses are arranged in a semicircle along the three closed
sides and enclosing the entrance-terraces. Deviations from
this plan were noticed only in a few cases. The view, fig. 59,
shows a section of one of these houses, and gives an idea of
their construction. It must be observed, however, that the
partitions of the rooms are not preserved, except in a few
instances. The ruined condition of the buildings is caused
not only by centuries of dilapidation, but also by the van-
dalism of treasure-hunters. In front of the houses, on the
side of the entrance, there is invariably a more or less spaci-
ous court (comp. pl. 9, fig. 2). Buildings of this class, on
terraces, and portals thatched with mats.” These houses in a
measure resembled the medieval castle, with cheerful activity
inside and a sharp watch kept outside. Cieza, l. C., says in
regard to this subject : “ Around the house lay a wide area
where dances were held, 5 and whenever the chieftain dined, a
crowd of people would assemble and drink of his chicha of
maize or peanuts. There were gate-keepers in these palaces,
who carefully watched all who came and went.” The courts
in front of the houses must, according to this statement, have
served for the execution of popular festive dances, witnessed
by the chieftains from the terraces of their houses. Construc-
tions for the safety of the palaces are still to be seen in some
places. There were gates, g, watch-towers, t, and a peculiar
kind of entrance easily watched, l. Only in a few cases
could any real gates be proved. There were two in the large
palace, pl. 9, fig. 2. In front of the entrance of the court, on
the side fronting the street, were two walls through which
gates opened. One can be seen from the left in pl. 10, fig. 2,
and from the right in pl. 10, fig. 1;6 the other from the
opposite side can be seen from the right in pl. 10, fig. 2,
from the left in pl. 10, fig. 1. These two gates must have
existed as, though there have been walks over both walls,
the entrances may still be distinguished as doors, pl. 10,
103'4"
364
as I
27'
28'9"
>
1
< 13'2">
>
16.5"
K 27'
27'
11'3"
55'9*
<
v
<67 x 11'6" > < 14*
°
Fig. 59. Sectional view of a terraced house at Pachacamac.
条
​来
​account of the terrace construction, might, though errone-
ously, be taken for temples. The terrace is on one side of
the house only, and, as a whole, differs considerably from
that which surrounds a temple on all sides. Houses also
lack the painting on the outside, so characteristic of temples,
and they are all partitioned off into rooms, which can only
have served for living purposes. The large number of graves
found in them also mark the buildings as dwelling houses,
as it was a very general custom to bury the dead in their
own dwellings. The earliest writers specially speak of this
form of houses as the dwellings of chieftains and nobles.
Hernando Pizarro writes: “ The houses of the common
Indians are of reeds, those of the chieftains of adobe.” ?
Estete, p. 133, speaking of Pachacamac, says: “ There are
houses with terraces, as in Spain,” whereby he surely referred
to private houses and not to temples. Finally the description
of the houses of chiefs on the coast of Peru as given by
Cieza exactly fits these buildings : “Each chieftain had a
large palace in his valley, with many brick pillars 4 and large
figs. 1 and 2. From the small terrace, i (see the plan on the
right), it was easy to keep a watch over all those who passed
in and out of the first gate. The entire building in pl. 10,
fig. 2, in the foreground, may have served for the purpose
of watching the palace, pl. 9, fig. 2 ; it is marked t, t, in the
plan. Another watch-tower may be distinguished in pl. 11,
fig. 2; to the right in pl. 11, fig. 1; to the left on the hill.
This watch-tower, marked t on the plan, near entrance 1 of
the house, overlooks the plain to quite a great distance. The
long terraces into which the hill in pl. 11, fig. 2, is graded,
doubtless served also as defenses. Other towers of similar
character, t, are to be seen to the northeast from the house in
section E, and on the side of a house near the city wall in D.
The manner of watching the houses probably was the same in
the platform buildings near s, although here a watch-tower
cannot be traced so distinctly. To secure the approaches to the
palaces there were also narrow winding lanes which led into
angles, so that it was impossible for the enemy to cover them
with their arrows, while the defenders were concealed. Of this
nature are the approaches l near some of the palaces," the first
four of which at least could be overseen from the watch-towers
2
3
7
5
5 Cieza, l. C., also tells that the chiefs of the coast land were always sur-
rounded by troupes of dancers who executed dances, while others made music and
sang for them.
1 An inclined path where both side-rails are still distinctly seen does not
exist at Pachacamac, but may be observed in other ruins, such as those of the
ancient Surco Chorillos.
2 Letter, p. 122.
3 Crónica, I, chap. 61.
4 Similar ones may possibly have existed at Pachacamac, but have been
destroyed.
6 At this place only one jamb of a door is still preserved.
7 Comp. pl. 10, fig. 2.
8
(57)

*
4
1
is quite evident. They are generally connected either on the
sides or in the rear, 4 and the occupants of both buildings
evidently had intercourse with each other. In one instance
these closed rooms are joined to the house on the court-side
and connected with the main building by a wall. In another
case a high walk, covered on the south side by a parapet, u,
conceals the connection with the closed rooms belonging to
this palace. It is easy to see that there was a great difference,
however, between these rooms and the living apartments of
the palace proper. The closed rooms are situated with their
base as low as sixteen feet, sometimes twenty-six feet, even
forty-two feet, below the level of the palace rooms. They
were much more secluded than the main building; although
grouped together in honey-combed fashion, they formed
dwellings for large numbers of occupants. They were per-
haps used by the retainers and servants of the chief, by his
bands of dancers, by armed men for his protection, always
ready at his call, and possibly may have formed the rear
defence of the palace.
The peculiar formation of these dwellings entered only
from above has its parallel in other parts of America ; for
instance, with the Pueblo Indians in New Mexico, also in
Argentina in the extensive Ruins of Quilmes, the town of
the Calchaquis. In the latter are dwellings built against
each other in honey-comb fashion, only separated by stone
walls ten feet thick, which at the same time form connecting
walks, from which inclined planes lead downward into the
dwellings. These three examples of this mode of building
also possess in common the peculiarity of the crowding
by which they were flanked. If it is remembered that several
of these strictly guarded palace entrances were at the same
time the only entrances to the interior of the city sections it
will be seen how secluded the latter must have been on cer-
tain sides. Many of the terraced houses used as palaces were
connected with artificial plateaus, ten feet or more in height
and sometimes very extensive, on which in certain instances
there have been rooms. The construction of these plateaus
seems to be the same as of those found near Lima by Bande-
lier and in the neighborhood of Trujillo, of which he has
given a good description. In making cross cuts through ter-
races adobe walls were found, forming the skeleton of the
terrace, which were worn so smooth that it might be sup-
posed that they had served as rooms before being filled in
and covered to form the terrace. However, many of these
rooms or sections are far too small for such use ; in some cases
they are only seven feet in length by ten to twelve feet in
height. Several were found together at Pachacamac in one
house in an unfinished condition, in various stages of con-
struction, so that an intermediate use as rooms seems entirely
out of the question. In Pachacamac they were mainly filled
with driftsand, together with refuse from the houses. The
covering on the top of the terraces consists mostly of sand or
of large flags of adobe. Fragments of big earthen vessels
were found upon several of these terraces which could have
held as much as fifty gallons and may have served for domes-
tic purposes. There is another type of dwellings which is in
close relation to these castle houses. These consist of rooms,
square or rectangular in shape, formed of plain walls from
eight to thirteen feet high, without any mode of entrance
whatever except from above. These rooms are mostly
. 3
arranged in groups of two or four, or simply in rows of six
or in double rows of eight, ten, twelve or twenty, or they are
in three rows in a network arrangement. The different
rooms are thirteen to twenty feet square on the average, but
larger as well as smaller ones are found. Near one palace a
room of this class is seventy-two by sixty-five feet in size.
The rooms or groups of rooms are either accessible directly
from the walls above, or they are enclosed in other walls five
and one-half to ten feet higher, and in each case very thick,
from which descent has to be made to the top of the room
walls, and from there still further down into the room itself.
In this case the lower walls are built against the upper ones
as retaining walls, of two to five feet in thickness, forming a
walk around the rooms, which may be as low as sixteen feet
or more below its upper edge, and which has quite the appear-
ance of a dungeon. Niches in the walls of these sunken
rooms are not uncommon. We have found but one instance,
however, where two of these rooms were connected by a door
opening through the lower dividing wall. The means of
descent into these rooms were most difficult, consisting of
small grooves in the wall down one corner of the room into
which the point of the foot would fit, as may be seen in
pl. 11, fig. 4. These steps may be seen in nearly all the
rooms of this class. Other means of descent, such as beams,
notched like ladders, may possibly have existed, but only
exceptionally. The descent from the higher enclosing walls
down to the lower walk around the top of the rooms was
usually made in the same way, although in a few instances use
was made of an inclined plane, such as is indicated in fig. 60
or in pl. 11, fig. 4.
Fig. 60. Mode of entrance to rooms from above.
together of the groups. There is significance in the fact,
as this and many other examples show, that close relations
must have existed at one time among the peoples along the
west coast of America and the Andes.
Large sections of districts B, C and D are cut into
smaller courts by walls without any buildings. These courts
were, like the above-described rooms, entirely closed, as may
still be seen by the walls, although they are in a very dilapi-
dated condition. It is not likely that these walls formed
mere enclosures for fields, as suggested by Bandelier in the
Ruins of Trujillo, since within the entire ruined city not a
trace of ancient water-channels was found. The courts are
quite empty, except when in rare cases they contain an
unimportant grave. Many graves, numerous enough for a
cemetery, were discovered in a few fore-courts of buildings
of palace-like dimensions. When, in one of these otherwise
empty courts a pile of household rubbish is found, it is safe
to trace it back to some adjoining house. These court-like
enclosures, devoid of any buildings, are to a certain degree
still unsolved riddles. Their walls, however, served for roads.
It is evident that intercourse within the sections of the city as
far as possible was carried on by means of the walls. It has
a
The close relation of these structures to the castle-houses
Comp. pl. 11, fig. 3.
5 In pl. 11, fig. 1, in the foreground.
6 Comp. also the good description given them by Sen. Ambrosetti, Bol. del.
Inst. Geogr. Arg., 1897, XVIII, pp. 40, 41.
7 Comp. Hodge, l. c., p. 305.
8 For the signification of the massing of the graves in these courts in its
bearings upon the ancient city communities, compare Cunow, Die sociale Verfas-
sung des Inkareichs, p 45.
1 The plateau adjoining the palace in the western side is about one hundred
yards in length by twenty-five in width.
2 Hodge, Bandelier's Researches, Amer. Anthropologist, 1897, 306.
3 Comp. pl. 11, figs. 3 and 4, and fig. 1 in the foreground.
(58)

1
rooms
been previously shown how walks were laid above the walls
of the gates, and how it was necessary to pass along these
walls to reach the enclosed rooms and to descend into them.
On the top of many walls were rooms and paths built of
bricks, as their remains still show, m, also inclined planes for
the descent, o, etc. Some of these walls are worn down in
the centre, by the traffic that must have passed over them, for
the closed courts made roads over the walls a necessity. This
manner of intercourse was evidently intentional and not due
to accident. In the interior of section D of the city, eastward
from the house near which the network of walls begins, and
where the entrance is open from the east side across the plain,
there is a peculiar construction for the guidance of travelers
upon the walls, as may be seen in pl. 11, fig. 1, in a broad
inclined plane near p. At first having a width of seventeen
feet it gradually grows narrower, corresponding to the ramifi-
cations of the roadways over the walls, and in a width of nine
or ten feet it runs around the net-work of the closed ro
which are located there, and to the terrace of a house, from
which point more walls radiate as roads into other directions.
The roadways over the walls were a very practical mode of
connection, since the plane is deeply covered with driftsand,
making communication across it very difficult. But as this
system of roads forms a part of the peculiar architecture of the
city, there must have been stronger reasons than mere local
ones for its adoption. Even now the low tapia walls which
enclose the fields are used as walks in preference to the dusty
paths running alongside.
In addition to the castle-houses there is another class 2 of
buildings to be found at Pachacamac, though only a few speci-
mens exist, which to all appearance was derived from the
former' type, but lacking the character of the stronghold. The
plan of apartments in this type of houses is similar to the
former, but the court is generally smaller and the defences
insignificant, especially wanting the high terraced understruc-
ture. The terrace consists of one grade only, three to five feet
above ground. Some of the buildings of the type under con-
sideration are still in a good state of preservation. This class
of houses is not found in the district of the castle-houses and
occurs more frequently in the northeastern part of the town,
where there is less organization. This fact in itself speaks
for the more recent date of the houses. Besides these two
types of buildings, there is another class of a somewhat differ-
ent order. A specimen of this, ninety yards long by seventy-
six yards wide, and resembling a court, lies on the northern
extremity of the town, in an isolated position, near the termi-
nal point of the northeastern city wall. This court was cut
into six parallel sections by walls of masonry, slightly raised
one above another. The structure does not seem to have
served for domestic purposes; judging from its location it
might be taken for a tambo (Kechua : tampu, inn), and there
are additional grounds for such a supposition. One building,
with another adjoining, resembling a hall, is found in the
eastern part of the town, leaning against the hill Y. This
house shows the peculiarity of a simple mural painting in a
checkerboard pattern, upon a wall in the main room, near the
western entrance. The elongated hall structure is rounded
off at its eastern end like an apsis. The inner side of the wall
is pentagonal. The height of this room is only three feet
three inches on the hillside by five feet nine inches on the
other, and a view of its interior may be obtained from the hill,
it must, therefore, have been used for purposes which would
allow such conditions. Two buildings in the eastern town, in
section B, and in the northern, E, have a very similar cross-
shaped ground-plane, formed by two small side-terraces flank-
ing a wider central one, simply graduated. The purpose of
neither structure can be conjectured. One building in the
northern part of the town, now a complete ruin, is worth
noting, because it shows traces of a former complete circuit
and a transept. This building is remarkable because it con-
sists only of a court, graded in terraces, two and a half and
five feet toward the west, and a line of six closed rooms, all
enclosed together in broad and high walls. The few parti-
tions of rooms in the court are not sufficient for dwelling
purposes, and must therefore have been used in some other
capacity, at present unknown. The palatial edifices con-
structed upon terraces at the northwest foot of hill X are
also unusual, but not strikingly so. Near q, q, q, in the east-
ern town B, are the remains of plain chambers, with poly-
chrome mural paintings, which may have served as shrines
for the cult of some idol. The three last-named buildings are
situated in the neighborhood of the enclosure of the Pachaca-
maj temple, one in the centre of the square at the end of the
southern street. They may have been shrines for the idols
of which Estete speaks as existing in large numbers in the
vicinity of the temple.
a
1
Comp. pl. 9, fig, 2, and pl. 10, fig. 1.
2 This type is found in the neighborhood of Lima in a few cases, as is
proved by the dwelling house at Huatica, shown by Middendorf, Peru, II, p. 84.
This ruin is at present so fallen into decay that the living-rooms proper have all
disappeared.
3 Visible from the distance in pl. 9, fig. 1.
CONSTRUCTIVE DETAILS.
It was
The type of the buildings as well as that of the details
of construction are those of the native style of the land. The
roofs were flat throughout on account of the almost entire
absence of rain. Although they are nowhere preserved, in
many cases traces of such construction can be seen.
easy to build a flat roof over the even walls of the rooms.
Walls apparently designed to support others were often
utilized to carry rafters on their upper edges, and a hori-
zontal line of grooves is frequently met with (comp. pl. 10,
fig. 2). These grooves are always on a level with the oppo-
site wall, and served to hold the spars for the roof. This
arrangement is found mostly in rooms of lesser width ; in
wider rooms the construction of the roof may have been
different; possibly posts were used to support the rafters along
the centre. There is no doubt as to the material employed
for the roof, since Hernando Pizarro 1 states that it consisted
of branches of trees; and to this day the roof of the natives
is formed of poles, with cane, leaves and sand spread over
them. The cell-like tombs for the dead, apparently an
imitation of the huts of the living, show the same manner
of roofing
Doors.—These are preserved only in a small number,
partly for the reason that the rooms in the principal buildings
have been entirely destroyed, and that many, as shown above,
were without doors. That a few are still remaining in perfect
condition is owing to the good climate. The apertures for
-
a
Letter, p. 122.
(59)

doors may be generally termed rectangular, for a decrease of
width toward the top of one half inch, rarely an inch, can-
not be considered as of material importance. 1 Height and
width vary. There are doors of only three feet in height, and
others of but one foot seven inches in width. Those of five
feet three inches to five feet eight inches in height, and two
feet four inches to two feet six inches in width, may be
regarded as of average size. The upper sill, in accordance
with material furnished by the coast land, is usually made of
Each sill was formed of three layers of cane sticks,
the upper and lower closely laid, while the middle one, cross-
ing the other two, consisted of three or more canes, laid at
wide intervals. The three layers were closely interwoven
with ropes, so as to form a stiff board (fig. 61), before being
placed over the door as the upper lintel, and afterwards plas-
tered with wet loam, overlaid by adobe bricks, as in ordinary
walls. The excellent state of preservation may be owing to
the mortar, which was packed into the cane-sticks and
thoroughly dried.?
cane.
rooms are erected in tiers, one above another, as frequently
seen in Inca buildings, for instance the houses on Titicaca
Island and Coati in Lake Titicaca. From a pathway on the
western side, several doors opened into upper rooms; other
walks, situated at a lower level, led into rooms lying below
these, with another entrance thirteen feet on the east side;
unfortunately the rooms were filled with sand, so that no
details could be ascertained. One of the two upper chambers
shows two narrow apertures for light, seventeen and a half
inches high by seven inches wide, on the south side, such as
are frequently seen in Inca buildings, but not otherwise found ,
at Pachacamac. There are no windows at all at Pachacamac,
as the climate there renders them unnecessary. The wall of
one house, figs. 64 and 65, near the storied building, forms
the only exception. It has in its lower part five niches,
arranged in two rows, a cornice running above them, and over
that a row of windows. Cornices were used to support floor-
rafters for an upper story, while the wall above retreated and
tapered toward the top. This is a special feature with Inca
a
Fig. 61. Lintel of a door, made of cane.
Fig. 62. Same lintel, from below.
Fig. 63. Typical form of niches in the buildings of the town.
buildings on the Islands of Titicaca and Coati. The architec-
tural influence of the highlands is very distinctly marked in
these cornices. The wall is remarkable as being the only
example known in Pachacamac in which two rows of niches
were employed as ornament. The ornamental cornice, sery-
ing as base to a row of windows set immediately upon it,
shows great similarity to the constructive principle of the
western façade of the large Gateway of Tiahuanaco.5 The
analogy is, however, limited to this one particular, but the
Niches.— Although niches are numerous at Pachacamac,
the proportion of their number is small compared with that
usually found in connection with the Incas. There are three
types, all rather primitive : 1, more or less cubic in form ; 2,
shallow niches in walls; and 3, small, high and narrow
(fig. 63 a, b, c). Class 1 is usually between sixteen and
twenty-two inches in depth by the same in height, rarely
varying more than two to four inches. Class 2 is insignifi-
cant, height, width and depth in proportion to 2: 3: 4: for
instance, 20 + 30 + 44. The width of Class 3 is usually
two-thirds of the height, the depth about fifteen inches. The
sides of the niches in Classes 1 and 3 are sometimes a little
inclined, and in that case bear a stronger resemblance to the
Inca type. Wide niches are roofed with cane, the same as
doors, narrow ones with bricks. The ornamental use is not
developed, they are only found in rows in certain cases exclu-
sively upon houses of more recent date. Judging from
various indications, their use may perhaps be traced to
Cuzco influences. The same influences are seen in a few
instances upon the buildings of the town, aside from the
Sun temple and the Convent, but there they are very pro-
nounced, though found only in the northern part. It is a
characteristic feature of the architecture of the coast land to
build a massive sub-structure under the living-rooms, but in
section E, near k, k, k, one example is to be found, in which
M.
10
л
4101
12"
K 12"
8'
2'
410
2 1/4"
1
Fig. 64. Wall decorated with
niches.
Fig. 65. Cross-section
of the same wall.
single, ornamental characteristic, carried out in both build-
ings, may have been for centuries the distinctive feature of
the highlands in the construction of windows.
1 A variation in the width of even two inches, which was observed in a
few instances, is insignificant.
2 Wiener, p. 63, shows the upper sill of a door alleged to come from Pacha-
camac. Aside from the fact that in Mamacona, where he says this was found,
there are no doors preserved intact, there is in all Pachacamac no construction of
the kind, and it is altogether impossible for ancient Peru.
3 Between the inner rooms of one house may be seen a door with the lower
part closed like a window.
4 In the ruins of the ancient convent on the Island of Coati, a wall is found
with one window above and two niches below ; compare also the ornamen-
tal portals (pictured by Wiener, p. 489, giving erroneously the Island of Titicaca
as the locality).
5 Stübel u. Uhle, pl. 5, fig. 1, and pl. 6.
(60)

CONTENTS OF GRAVES.
Although no untouched graves were discovered in the
houses of the town nor in the courtyards near them, some ves-
sels were found intact, a great variety in fragments; very many
pieces of fabric, some large, some small; various utensils, such
as plates, cups and other implements of wood; a clay mould,
soine articles of copper, skulls, etc. These discoveries were
sufficient to identify their period of chronology with the last
era of the town, that of the Incas, during which objects of
native manufacture were mingled with others, of Cuzco char-
acter. They were utterly different from those unearthed in
the cemetery at the foot of the temple, as well as from those
obtained in graveyards II and III, though both sets of graves
belong to the same period. There is great resemblance, how-
ever, between them and the objects taken from cemetery VI,
some examples of which are shown in pl. 13. In quality,
articles found in graves in houses may have been superior to
those of Cuzco characteristics in cemetery VI ; arybals,
plates, etc., are more numerous, but the fundamental attributes
in both cases are those of the general culture of the coast land
at that period.
>
CONCLUSIONS.
66
2
3
The description of the inner town has shown that at the
time of discovery its characteristics were not purely those of
the coast, but in the architecture of certain localities, as well
as in the culture of its inhabitants, it had been influenced by
the Inca conquest. In general, it appears that the inner town
was a dwelling place for a number of chieftains, each of
whom was surrounded by a large host of retainers. Against
the assumption that the whole city was one palace, 1 speaks
the fact of its division into sections and the existence of one
or more important homes in each district. It is hardly prob-
able that these houses were hostelries, ? or family or county
temples, as indications point to their having been constantly
inhabited, and their appearance is more that of castles or
fortresses. Arrangements for the security of some of them
are very noticeable, and many of the courts contain a large
number of graves. The division of the older part of the town
into B, C and D was probably according to the existing social
order. The different sections or wards were occupied by
people of various social grades. Intercourse between the mem-
bers of each grade was much more active than that between
different classes. The sections or wards corresponded to the
Barrios, and the different palaces with their courts bore
some affinity to those which were found 4 in the Ruins of Chau-
can, Trujillo; resembling also the ‘Bandas' or 'Kanchas'
(courts) which at Cuzco formed the subdivision of the 'Par-
cialidates' (aillus ?). The division of Pachacamac into four
connecting wards was similar to the division of Cuzco by
means of two intersecting streets into four quarters, corre-
sponding to the four quarters of the empire, Cuntisuyu, etc.;
also to the division into aillus, which still prevails in principle
in the villages of the highlands, for instance, in Bolivia.
There are, as a rule, from three to eight aillus, generally four,
all sharing in the plaza. In the case of four aillus, each repre-
sents one corner, together with half of each adjoining side, and
the part of the village extending behind it. 6 Calancha states :
:
“At the time of the Incas many nations settled near that
sanctuary. Priests (ministros), together with their assistants,
were sent there and supported by every town.” In connection
with this statement is probably his designation of the terraced
houses as family, village, or provincial temples. Calancha is
exceedingly unreliable in his interpretation of the ancient
buildings, since he follows a posthumous and, as a rule,
erroneous tradition. There may, however, be some truth in
this particular, and that the terraced houses were, in a cer-
tain measure, occupied by religious communities, supported
here by different parts of the coast. In that case the plan of
the ancient town of Pachacamac represented in a microcosmic
manner a religious macrocosm, corresponding to the political
one at Cuzco.? We are not, however, in a position to further
discuss such a possibility, since the objects gathered from
houses in various sections of the town showed no cultural
differences. If we are justified in supposing the largest
palace of the ancient town to have been that of the ruler of
the valleys of Lima and of Pachacamac, it can hardly be any
other than that shown in pl. 9, fig. 2. Pizarro, when he came
to Pachacamac, took quarters with his followers in a very
large palace, situated on one side of the town.8 This may
have been the palace on the summit of hill Z,' which at the
same time afforded the greatest degree of safety to the invaders.
a
7
5
9
1 Ulloa, p. 301.
2 Rivero y Tschudi, p. 288.
3 Calancha, l. c.
Squier, p. 159; Cunow, p. 47.
5 Cunow, p, 48.
6 The divisions known in upper and lower Cuzco as Hanan and Hurin have
been preserved everywhere in the villages of the Bolivian highlands, under the fol-
lowing names, in Kechua : Hanansuyu and Hurinsuyu, or in Aimará : Alayajas
and Manqhajas. Each of these divisions usually embraces several aillus, and the
imaginary line of separation between Hanan and Hurin, or Alayajas and Manq-
hajas, always cuts through the centre of the plaza. At Cuzco two sections
belonged to each of these, Hanan or Hurin (comp. Squier, p. 428).
7 See above, p. 11, note 11 ; also Squier, l. c.
8 Estete, p. 130.
9 In the panorama, pl. 1, to the right on the hill.
4
(61)

CHAPTER XVII.
TIIE OUTSKIRTS OF TIIE TOWN.
1
3
4
а
AN
N apparently boundless desert stretches outside the city
limits to the west, north, and northeast, in which may
be noticed at a distance of half a mile from the city a dark
line of masonry. On closer inspection this wall? is found
to measure twenty feet in breadth, and to extend for four
hundred and seventy-five yards, slightly curving over a
depression in the ground. The extreme ends seem to dis-
appear in the driftsand, 4 which is abundant everywhere,
and sometimes as high as the level of the wall, in places
entirely submerging it. Close investigation reveals that
the masonry is not continued beyond its visible length. At
the ridge, which almost touches the northeastern part of the
inner wall, it is crossed by another, situated at about two-
thirds to one mile distant from the inner town beyond the hills,
and fronting the hacienda Las Palmas, still upon desert soil.
This wall is almost fifteen feet broad and several hundred
yards long, running from east to west. Again, when crossing
the fields which separate the inner town from the sea, the
ruins of a massive dam may be seen at about one hundred yards
from the shore. Only small portions of it remain, but at one
time it must have been an imposing structure. Its extreme
end lies far to the northwest, near the hills which abut closely
on the sea (comp. the panorama, pl. 1). Beginning with a
sharp angle, the dam extended along the shore-line and ended
somewhere in front of hill W, on the summit of which the
Sun temple stands. It has been broken by the sea, which
occasionally washes over it. The present dam may be the
remains of an ancient wall. The construction of this work
may, without doubt, be ascribed to the same people who
erected the northwest and the northeast outer walls, and it
must have formed a part of their general plan. On closer
examination of the territory between the inner city wall and
those at the northeast and northwest, numerous lines of an-
cient buildings may be traced over the desert soil, only slightly
noticeable upon the surface. Fragments of pottery are to be
found everywhere. Cemeteries are numerous in various direc-
tions and at a considerable distance from the town, and
although covered by the sands of the desert, have not escaped
the ravages of treasure-hunters. The large area outside the
inner city wall was occupied by extensive quarters of huts,
which have now disappeared, being merely cane shacks, and
only their foundations of adobe bricks remain. The entire
territory formed an outer zone lying around the town. All
objects recovered from this zone, from graves, as well as from
other sources explored in the desert on the site of ancient
dwellings, belong to the Inca period of the city. In one
instance a necklace was found inside a clay vessel, consisting
of beads of shell and of sea-green glass, the workmanship of
which showed that the occupants of this district of the town
belonged to the Spanish period. Buildings were in process
of erection here at the beginning of Spanish rule. On the
northern extremity of the northern limit may be seen the
beginnings of two streets, one in a northwestern direction,
continuing older lines, and one running due north. Of the
former one section only of its western enclosing wall had been
finished; of the latter a section was formed by the construc-
tion of a building, the southeastern end of which, together
with the inner city wall, enclosed a part of the street, the
width of which was twenty-three feet. This leads to conclu-
sions as to the scope of the builders' plans. Probably the
new street was planned long after the old ones had been
finished. To accommodate increased traffic, the new street
was designed to be almost twice as wide as the older north
street, and once and a half as wide as the east street of the
town proper. One building at this point was in process of
erection, as may be seen from the terrace adjoining it on its
southern side. The retaining walls for the understructure
had not all been erected, others were not quite finished, and
interstices between others had not been completely filled in.
The broad northwestern outer wall must also have been left
in an unfinished state, and was probably intended to be con-
nected with the northeastern one. In some places it has a
height of twenty feet, while in several others near its west-
ern extremity it is level with the ground, or has only an
elevation of one and a half feet. Although the ground rises
gently, the average level of the wall is the same everywhere,
and such variations cannot be supposed to be part of the origi-
nal plan. Its surface is exceedingly uneven in some places,
in others projections occur of uneven length, in height vary-
ing from two to ten feet,' or terraced, rising above the general
level of the wall. Since there is no other explanation to be
found for the condition of this wall, it may be assumed that
the city, at the time of the conquest, was just being extended.
5
6
7
Compare the panorama ; also pl. 20, fig. 1, in the background on the left.
2 Comp. pl. 20, fig. 3.
36.10 m. The breadth of the wall was stated correctly by Bandelier
(Hodge, p. 305), while Rivero and Tschudi's estimate of five' varas' (thirteen
and three-fourths feet) is erroneous.
4 According to Rivero and Tschudi and to Middendorf, 1. c.
5 On the south edge of the valley, about three miles from the ancient city,
the remains of another ancient dam are to be seen, which protected the district
between the coast and the hills farther inland.
6 Rivero and Tschudi estimate it ten feet below its actual height, which is
varying and, owing to the drifting sand, can only be measured in a few places.
7 See pl. 20, fig. 3.
(62)

OBJECTS RECOVERED FROM THE OUTER ZONE OF THE CITY.
Although the objects obtained in the outer zone of the
town all belong to one cultural period, the specimens to be
discussed here were found in one cemetery, situated about four
hundred yards from the town proper, and from the outer
northwestern wall just about two-fifths of the distance between
the two. The cemetery was quite untouched. Numerous
older graves, but there is a marked similarity to those from
the castle-houses of the town, the occupants of which must be
considered as natives. They also made use of numerous
objects of Cuzco type, such as vessels of pottery, implements
of wood and certain textile fabrics, which fact may be
explained by the system used by the Incas in order to
assimilate newly conquered provinces with their realm. Gar-
cilaso 1 relates that the Inca carried the first cazique of the
newly acquired province, together with all his children, to
Cuzco, so that by intercourse with the people there they
might acquire a knowledge of the laws, language and cus-
toms, as well as the religious cult of the realm ; this accom-
plished, they were reinstated in their former dignities. San-
tillan has it that the Incas caused the sons of all the Kuraka
families to go as yanakonas: to Cuzco, and devote themselves
to their service. After having become assimilated with the
Inca people, they were sent back to their own provinces, in
some cases invested with official dignities. The common peo-
ple went through a similar process of fusion ; the Inca,
following the well-known system of the mitimaes, would send
inhabitants of newly conquered provinces to older assimilated
districts, 5 " and citizens of Cuzco and other loyal towns” to
the provinces as garrisons, as well as “for the purpose of
teaching the natives the laws, customs, ceremonies and lan-
guage of the realm.”
This was the reason that in some of
the villages on the coast most of the inhabitants spoke
3
Fig. 66. Large vessel from grave in an ancient house, outer
zone of city. Inca period.
5 66
excavations in the neighborhood, in ancient house foundations,
resulted in the recovery of various large vessels, two of which
were added to the collection (fig. 66), also several arybals
of the style of Cuzco and of large size, but imperfect, and
therefore not preserved, and some ornamented cooking vessels,
specimens of all of which abounded in the cemetery. The
vessels shown in fig. 66 have sides five-eighths of an inch in
AM
a
Fig. 67.
Fig. 70.
Fig. 68.
Fig. 69.
Types of black pottery from graves of the outer city Inca period of coast land.
Kechua perfectly, although originally three or four differ-
ent languages had been employed in that district. The
form in which the Cuzco tongue and that of the coast land
became fused is, doubtless, due to the cultural change of the
coast land, in consequence of the intermingling of peoples
which the Incas effected after their conquest of new terri-
tory. & Certain types, such as pl. 12, fig. 6, show quite a
local stamp, both in technique and ornamental details. It
8
thickness. Their decoration consists of white lines marking
off sections, like a melon, a style typical of the period under
discussion. In the cemetery perishable objects have mostly
vanished; perhaps because the sand here possessed less pre-
serving qualities than clay, or sand mixed with clay, as is
found in cemeteries I, II, and elsewhere. On this account no
mummies were found, only crania and bones, pottery, a few
textiles in a somewhat better state of preservation, owing to
some favoring circumstance, various implements and utensils
of wood and metal. The discoveries were sufficiently numerous
to establish the character of the period in every particular.
They do not represent any distinct style; there are specimens
of Cuzco type, as well as of the coast land, but between these
two extremes are a number of objects which belong exclu-
sively neither to the one nor to the other, but may be said
to form a combination of both, and to create new types, as
may be seen in the ceramics of the period. The specimens,
although dating from the same period, do not correspond
with those taken from graves in the Sun temple, which reflect
Cuzco influence in such a striking manner; nor do they
resemble those found in the ancient cemeteries of the inner
town, I-III, which belong to another period of the style of
the coast, and are entirely devoid of Cuzco objects, as well
as of products of a mixture of both types. The contrast
between the discoveries in cemetery VI and those of the older
burial-places is in some respects very marked. In ottery,
for instance, not one vessel resembles any of those from the
8
(6
I VI, chap. 12.
2 P. 39.
3.Yanakona' is really the plural of 'yana,' servant (see v. Tschudi, Wör-
terbuch, p. 82), as 'mamakona’ of mama, but is used for the singular, in Peru,
perhaps as much as 'vassal.' The usage must have taken root early, as Santil-
lan employs it, making from it a further plural, 'yanakonas.'
The chiefs are said to have lived always near the Inca. Even through the
women the realm was kechuanized, as every year many good-looking ones were
transported to Cuzco, some of them to be given in marriage to the vassals as a
special favor (Santillan, p. 38).
5 Santillan, p. 40, and many others.
6 Garcilaso, 1. c.
7 Cronica, I, chap. 62.
Brinton, American Races, p. 203, says: “It is an error to suppose that the
extension of the Kechua was the result of the victories of the Incas. These
occurred but a few centuries before the arrival of the Spaniards, and their influ-
ence was not great on the native languages.” This position is hardly tenable in
view of the complete transformation of the culture on the coast, brought about
by the Inca conquest, and of the effectual introduction of the Kechua. The
rapid change in culture which the Incas worked wherever they extended their
sphere must certainly be called remarkable. They introduced their speech with
their culture. It is the same in the Argentine. The Calchaqui, according to the
prevailing older names of places, were surely not Kechua. Objects of Cuzco
character, nevertheless, are found everywhere, not only at Cuzco, but also
throughout the provinces of Catamarca and Rioja (comp. collection of the Berlin
Museum). Many are the 'Rios del Inca,' 'Fuertes del Inca,' Tambos del Inca,'
although the conquest occurred at a very late date, and at the same time as the
introduction of Kechua.
a
(63)

seems, therefore, that objects found in the graves of the subur-
ban city must be considered as products of the coöperation
of the Cuzco style with that of the native of the coast, the
latter preserving its local peculiarities. The possibility that a
certain proportion of the inhabitants of Pachacamac, during
the Inca period, were colonists, transplanted there from other
parts, cannot be disputed. The theory might lead to the
expectation of finding the fact expressed in a great diver-
sity of types of cranial deformation ; but this is not the case.
Outside the prevailing type, with broadened and flattened
colors. Among them only a small number show as highly
developed decorative designs as those of purely Cuzco origin.
A few examples of black vases of this class are decorated with
human faces on the neck, as shown in pl. 18, fig. 2. Some
small variations in shape or decoration may be noticed, as an
evidence of outside influences, such as might be exercised by
the style of the coast. A black arybal of this order is shown
in pl. 13, fig. 1 ; it has two arm or ball-like attachments, from
which protrude three thorns, and is represented by two speci-
mens, while in Cuzco ? it is unknown. Another specimen is
Fig. 71.
Fig. 73.
Fig. 72.
Types of black pottery from graves of the outer city. Inca period of coast land.
occiput, the common one of Pachacamac and other cemeteries
on the coast, no additional types of deformation were noticed
in cemetery VI and in the graves of dwelling-houses.
A. POTTERY.
In regard to the technique of the pottery of this ceme-
tery, it must be observed that black vessels, or those of inferior
make, yellow or red in color, with ordinary primitive decora-
tions, form the principal contingent. The black specimens
always show a good polish, at times even excellent, and occa-
sionally with quite a metallic lustre. The shape in many
cases is more artistic than even the finest arybals of Cuzco
type. The sides are sometimes thin as porcelain, such as is
seen in no object of purely Cuzco style. The yellow and red
vessels, with primitive decorations, are of a coarser clay and
in rare cases well polished. The colors are black and a
sort of dull white. A careful treatment in both decoration
and polish is seen only on articles of unmistakably Cuzco
technique. The preponderance of black color in a large pro-
portion of the pottery, and the fine finish of some of it, far
surpassing anything of Cuzco workmanship, prove the influ-
ence of the very highly developed ceramic art of some parts
ornamented with a lizard in relief, the third has a flat base
and only one horizontal handle, etc. 3
2. Vessels, such as are represented in pl. 12, fig. 3, and
small jugs, Cuzco style (pl. 18, fig. 7), have been found in
several instances. The decoration is in good Cuzco manner.
It is surprising that no plates were obtained nor any vessel
like pl. 18, fig. 5.
3. Low flasks or vases with human faces, mostly black,
rarely colored (pl. 13, figs. 3 and 4), are more numerous.
Both black and colored are similar in the moulding of the
face, and in the manner in which the arms are suggested on
the bowl of the vessel, either painted (fig. 4) or engraved, or
even moulded in relief. The partial influence of Cuzco is
unmistakable in the contour of the neck (comp. the arybals),
in the bowl-shape of the body of the vessel, which in general
is the predominant fashion of Cuzco pottery (comp. pl. 18,
figs. 1 and 7), and the type of the face (comp. pl. 18, fig. 2).
The flask (pl. 13, fig. 4) also shows the coloring of the face,
so often seen on vessels of Cuzco type (comp. pl. 18, fig. 2),
while the arms are painted in the dull white color of the coast
Purely Cuzco ornaments decorate the handle.
Fig. 74.
Fig. 75.
Fig. 76.
Types of black pottery from graves of the outer city. Inca period of coast land.
1
of the coast land. The red and yellow ware, on the other
hand, represents in its technique a wide local branch of the
art. The use of a dull white color is characteristic of some
kinds of pottery from Pachacamac, and the valley of Lima,
Ancon,' Huarmey, up to the northern parts of the country.
The use of moulds can be seen in several specimens of this
type, though they are not so extensively employed here as
they were farther north (Trujillo, Chimbote, etc.).
2 Philippi (Verh. der Berl. Ges. f. Anthr., 1885, XVII, p. 268) assumes that
all vessels of Cuzco type, and all those found in remote parts of the Inca realm,
Fig. 77.
1. The arybals are mainly black, a few of different
were made at Cuzco and introduced from there. This may be the case to a great
extent with vessels of purely Cuzco workmanship, but the theory would be
difficult to maintain in regard to vessels of the same size as the largest in the
Centeno collection at Berlin. Many fragments of these were found on the site
of an ancient city near Chilecito, in the province of Rioja, in the Argentine, and
are now in the Museum of Berlin. There were local productions of vessels of
Cuzco style, numerous specimens of which have been previously known, and
which may be distinguished by peculiarities of technique or decoration ; such as
the corner-ornament in the inner design on the arybal from Huarmey (Kultur u.
Industrie, I, pl. 10, fig. 9), which distinctly shows influences of the ornamenta-
tion of the coast. The spiral-winged decoration, however, does not belong to
this class, as erroneously stated by the writer (l. c., p. 326 ). The arybal (1. C.,
pl. 10, fig. 10) from Pueblo Nuevo, on the Peruvian coast, is remarkable for its
white coating; the arybal (l. c., pl. 11, fig. 11) from Corocoro, Bolivia, for its
exceptional metallic red glaze; the plates from Freirina and Valdivia Medina
(A borij. de Chile, Nos. 165 and 173) for their shapes and decorations.
3 Comp. also the vessel (Reiss u. Stübel, pl. 93, fig. 3) from Ancon.
1 Reiss u. Stübel, pl. 97.
(64)

4. Simple flasks are also found of the greatest variety in
the shape of bowl, neck or handle, many being perfect in
form and technique. The majority are black, while some
show the natural color of the clay, but none are painted.
The connection between neck and bowl in many specimens
is formed by a rim, mostly engraved in a simple pattern of
the class often seen in Cuzco textiles, and the handle in some
cases is engraved in a Cuzco pottery design. The shape of
the vessel frequently represents a conical fruit, with an
upright ape or a cat in the act of springing, in plastic mould-
collection, which comes from Quito ;5 another, from Cuenca, is
at Brussels. Vessels of the second type, fig. 9, show a differ-
ence in the proportion of the body, in the ornaments of the
figures, and in the dots in relief, which are peculiar to the
coast, as is also the face with indented brow-an indication of
the deformation of the skull. A similars piece of pottery
may be seen in the Æquatorian Collection at Brussels. 9
Vessels with four bodies were also used at Pachacamac.
They consisted of four conical fruits, surmounted by a bow
and neck. A specimen from Cuenca can be seen at Brussels.10
a
MESSA
Fig. 78.
Typical flask of black pottery. From graves in the
outer city. Inca period of coast land.
Fig. 79.
Fig. 80.
Flasks with flattened sides. From graves in the outer city. Inca period of
the coast land.
1
a
ing, perched on the neck.' Pl. 13 shows in figs. 13 and 14
two types of this vessel, the first of which is very beautiful in
outline. Similar specimens are often found in later ceme-
teries near Lima. Several vessels from Ancon belong to this
class; also one discovered at Calpi, near Riobamba, in
Ecuador, 2 which proves them to have been used over a large
part of the Inca realm. These vessels resemble the Cuzco
style in many particulars, while representing a more developed
type. Points of similarity are to be seen in the bowl-shaped
body, the tapering base, the characteristics of the handle, and
in the engraved ornaments upon it.
5. Flasks with curved neck, nearly all black, like fig. 78
in shape. The only red and white specimen is pl. 13, fig. 12.
The type of these flasks belongs to the coast, and they occupy
the same technical position as the former class.
6. Double-flasks; those in the collection are all black,
and mostly silvadores,' that is, provided with pipes. Two
7. Oblated flasks; the majority of them are colored, a
few black. Some specimens of this class are decorated with
face and coloring in the Cuzco style; others are ornamented
in the style of the coast, with raised figures and dots. A col-
lection of pottery from Cuzco, described by Ewbank, con-
tained a few oblated flasks. 11
8. Jugs with small necks, representing the entire human
figure, head and arms plastic, the lower parts indicated in
relief only. They are black or colored, and represent a
variety of types. On the specimen, pl. 12, fig. 2, a poncho
is indicated, ornamented in the manner of tapestry-ponchos
of the district of Cuzco. The pattern is indicated by engrav-
ing: 12 To this class belong some of the vessels from Ancon. 13
Another jug, resembling the preceding, colored, is shown by
Bastian.14
9. Various plain flasks, or flasks representing animals or
fruits with a bow. To this group belong pl. 12, figs. 11 and
doje
w
um
Fig. 81.
Fig. 83.
Fig. 82.
Types of vases. From graves of the outer city. Inca period of the coast land,
14. They are red or black, and show a large variety of
main types of these exist, which are shown in pl. 13, figs. 8
and 9. The first belongs to the class of vessels described
under No. 4, both as to form and workmanship. These
double-flasks are generally very light in weight, and possess a
metallic lustre. A bird or similar object decorates the closed
neck.4 To this type also belongs a specimen in the Leipzig
9
5 Kultur u. Industrie, I, pl. 7, fig. 4.
6 Congrès des Améric. Bruxelles, Compt. Rend., II, p. 127, Atlas, pl. 19,
fig. 1.
7 Comp. Virchow, Crania ethnica, pl. 14, and fig. VI, p. 12.
8 Comp. Rivero y Tschudi, 1. c., pl. 17, fig. 1; pl. 18, fig. 2 ; pl. 20, fig. 1.
y
Congrès des Améric., I. c., p. 126; Atlas, pl. XVII.
107. c. with pl. XVIII. Another Rivero y Tschudi, l. c., pl. 20, fig. 2.
y
11 Thom. Ewbank, A description of Indian antiquities, U. S. Naval Astron.
Exped., App. E. p. 130, d and j.
12 Even the seams of the poncho are indicated.
13 Reiss u. Stübel, pl. 97, fig. 11; pl. 93, fig. 4. Comp. also Rivero y
Tschudi, l. c., pl. 14, fig. 2.
14 Die Culturländer Amerikas, III, 2, p. 68 and pl. 1, fig. 5.
1 Comp Reiss und Stübel, pl. 93, fig. 6; pl. 96, figs. 11, 13; also the ape,
pl. 100, fig. 4.
2 Kultur u. Ind., II, pl. 7, fig. 5.
3 Comp. the vessel from Calpi, in Ecuador.
4 The vessel, Rivero y Tschudi, Atlas, pl. 12, fig. 1, is of the same class.
9
(65)

be confounded with the older specimens discovered near Chim-
bote ; they are also met with in the vicinity of Lambayeque,
in burial-grounds of more recent date. These vessels, so plain
Fig. 87. Clay figure. From a grave of the outer city. Inca
period of the coast land.
objects, such as stags, cats, fishes, crabs, birds, fruits or vege-
tables, all in plastic reproduction, rarely in relief. They
doubtless belong to the style of the coast, but from the manner
in which they were found they are proved to date from the
last pre-Spanish period.
10. Red and black vessels with faces upon the neck, the
hands moulded in relief, sometimes raised to the head or
painted upon the bowl; the lower parts never indicated. The
base is conical or flatly rounded. The black vessels have
thick sides, the colored are of coarser clay, and mostly red, or
decorated with black and white. Pl. 13, figs. 5–7, and fig.
79 in the text reproduce some of the types. The technique
of this pottery is that of the coast, the conical base resembling
the Cuzco arybals. Several similar objects with a pro-
nounced local character come from Ancon, 2 others with the
same characteristics from Ecuador, and certain modern ones
are produced in various places in Bolivia. They all, whether
from Ancon or Pachacamac, have the hands in the same
position, supporting the head.
11. Vessels with heads of birds on the neck or bowl, as
in fig. 10. Objects of this class are found also in burial-
places of later times in the vicinity of Lambayeque. The
bird's head causes resemblance to objects like that in pl. 18,
fig. 10, and similar bowls of wood, of the type of pl. 18,
fig. 15.4
12. Cooking utensils, decorated, such as are shown in fig.
in appearance, do not therefore represent a limited local type,
although they belong exclusively to this one era. They form
a proof of the cultural uniformity of the coast land through-
out a vast region during the Inca period.
13. Various flasks with two handles, like fig. 86. Some
of these are of plain red, with stripes of dull white, in melon-
shape; others are well polished and decorated, after being
baked, with a bird or other figure in the style of the coast
Many of these decorations have faded owing to the peculiar
technique.
14. Among other objects a clay doll was found, made in
a clay mould (fig. 87). The Berlin Museum possesses clay
moulds from Peru for the making of similar figures. 8
Fig. 84.
Fig. 85,
Types of cooking utensils. From graves of the outer city. Inca period of the
coast land.
Fig. 86.
Typical flask of red clay. From graves of the outer
city. Inca period of the coast land.
B. TEXTILES.
84. The vessels are found in great numbers in the outskirts
of the town. One class shows ornamental bands in various
designs in relief, below the rim; another group is decorated
with undulating lines in relief upon the neck, or bowl, or both.
Many specimens have a sort of embossed ornamentation,
which was made by pressing the sides of the vessel into the
desired shapes from the inside, before it was dried (see
fig. 85).
Both types belong to the same period. Pottery resem-
bling the second variety has been unearthed at Ancon. The
undulating lines in relief may be intended for serpents and
can be compared to those on stone vessels of Cuzco type.
Such objects almost certainly originated at Cuzco. Vessels of
the first-named description, with small local variations, are
also found at Ancon,' in later burial-places, which must not
5
The textiles of this era do not show any marked differ-
ence from those of the last pre-Inca period. A practised eye
may, however, in many cases easily distinguish whether a
fabric belongs to the pre-Inca or Inca time, the difference
consisting chiefly in a more advanced technical development.
It is impossible to give a minute description of the textiles
of this period in so small a space; only a general outline can
be presented. Students are referred to the exhaustive work
of Reiss and Stübel on their discoveries at Ancon, which goes
minutely into the subject.
In many instances textiles of more recent date are distin-
guished by the brightness of their coloring, while earlier ones
favored designs in dull tints. The colors are also more har-
moniously blended and softer, not showing such crude con-
Any fabric showing delicate color-blending will
always, as a matter of course, belong to this period.10 Red
and yellow begin to appear together, a combination hardly
ever before used ; 11 while yellow as a ground color is much
trasts.
8 Uhle, Ausgew. Stücke des Mus. f. Völkerk. Berl., Veröff. aus dem
Museum, 1889, I, Taf. 6, fig. 1. Frogs of the kind reproduced, l. c., fig. 3, in
the model, are often used as handles of vessels during the same period at Pacha-
camac.
9 Comp. Reiss u. Stübel, pl. 66, fig. 3 ; 67 a, figs. 1 and 4; 57, fig. 1 ; 67,
fig. 4; 68, figs. 3-6, etc.
1 Of this class are also Rivero y Tschudi, 1. c., lam. 10, fig. 2 ; lam. 11, fig. 1;
13, fig. 2; 16, fig. 2; 19, figs. 2 and 3. All these might readily have been found
at Pachacamac, but whether they were so or not is unknown. Comp. Reiss u.
Stübel, pl. 93, fig. 14, as belonging to this class and period.
2 Reiss u. Stübel, pl. 97, figs. 1, 8 and 12.
3 Congrès des Améric., l. c., II, pp. 117 and 122 ; Atlas, pl. 7, figs. 6 and 7;
13, fig. 2.
4 Good examples are given in Kult. u. Indust., II, pl. 1, figs. 1, 3 and 8.
5 Reiss u. Stübel, pl. 96, fig. 15.
6 Comp. Rivero y Tschudi, l. c., pl. 31, fig. 1. A clay vessel of this kind
was the first object found by the writer in the suburbs of the city, and was by
him classified as of Cuzco type, on account of undulating lines, resembling the
ornamentation on stone vessels. For this reason he assigned it to a recent date.
The contents, among them a necklace with glass beads, confirmed him in this
conjecture and helped to support the assumption that undulating lines indicate
Cuzco influence.
Reiss u. Stübel, 1. c., fig. 13.
10 Comp. I. c., pl. 67, fig. 4 ; 68, figs. 5, 6.
11 L. C., pl. 67 a, fig. 1; 59, figs. 1, 2 ; 67 b, fig. 1 ; 57, fig. 1 ; 66 a, fig. 4 ; 66,
fig. 3 ; 68, fig. 6; 67, fig. 4.
9
(66)

favored, 1 having been rarely seen in earlier textiles. Many
tapestries of this period are distinguishable from the earlier
ones by the large size of the ornamental figures.? Designs on a
large scale, similar to those shown by Reiss and Stübel, pl. 55,
fig. 1, belong to this time. In figure designs the tendency is
more and more toward large patterns. The technique is cor-
respondingly treated. Texture, even in rich designs, is often
W
belong to this period; also compare the braids of Ancon (I. C.,
pl. 68, fig. 3; 66, fig. 3).
All technical processes of textile art already known were
continued, such as open-work weaving by interchanging warp
threads, the weaving of small pieces of material to form pat-
terns in combination, as well as those produced by dyeing or
painting or other similar treatment. In textiles, as well as
in pottery, certain types appear to belong specially to the
coast region and others to the highland, which is caused by
the influences at work during the rule of the Incas. Com-
pare, for example, a poncho with horizontal border in a
pattern such as pl. 8, fig. 16; with fragments of ponchos
resembling those found in the graves of the Sun temple, and
the pieces of a woman's girdle (chumpi) (pl. 20, figs. 1-4,
etc.). Certain attributes which modified the style of the coast
may be referred indirectly to other agencies. As regards
colors, gray appears at this epoch, sometimes in woolen mate-
rials, but generally in cotton. It may have been produced by
some newly-discovered process. The same color is often seen
in Cuzco textiles, taken from graves in the Sun temple. The
frequent occurrence of textile designs in white and brown
only, rarely found previous to this date, or in white and blue,
white, brown and blue, 8 brown and blue,' must be attributed
to highland influences. In addition to the new color, gray,
a number of combinations appear at this time 10 in striped
Fig. 88. Cross-shaped ornament of a necklace. From a grave
of Inca period. Outer city.
less close than in earlier textiles ; long slits along the outlines
of figures or colors are seen again, similar to the tapestries
of the earliest periods, when they were retained on purpose
to mark the pattern. The tendency to replace oblique lines
by graduated ones is often quite apparent, and the same gen-
eral effect is thus produced. The variety of ornamental
designs seems infinitely greater at this period in the evolu-
tion of new figures, as well as in the continuation and devel-
opment of earlier types. A pronounced tendency toward
7
8
9
wa Kasur
GOOOOOO
| Diwand
100
OOOOOOOO
ODO.
CATASUTUS
mm
ооооооо
OOOOOOOO
JOG
Fig. 89. Decorative design (trees) in a tapestry. From a grave of
the outer city. Inca period of the coast land.
Fig. 90. Section of embroidered decorative design. From a grave
of outer city. Inca period of the coast land.
naturalistic reproductions may also be noticed. Variations
and transformations of designs, such as may be seen in some
of the discoveries of Reiss and Stübel, belong to this time.
See, for instance, l. c., pl. 67, fig. 4 (design in the squares);
68, fig. 1; 66 a, fig. 1; 67 b, fig. 1; 59, fig. 2 (border). Cross-
shaped figures, like l. c., pl. 68, figs. 5, 6, are frequently seen
on textile fabrics, and are characteristic of this period. A
small cross from a necklace, found in a house at Pachaca-
mac, shows the same outlines as the figures above mentioned
(fig. 88.) All kinds of animals are represented for the first
time in the tapestries belonging to this era : spiders, scorpions,
fishes, birds with a peculiar tuft on the head like that of a
rooster, etc. Plants had already made their appearance in the
preceding period, as tree-shaped figures, with many varia-
tions, upon a certain class of fabrics. All these figures
developed into richer designs (figs. 89 and 90) or were
embellished by the addition of a leaf-pattern (fig. 91), which
is often seen on borders or braids. More naturalistic repro-
ductions of common ornaments of earlier times, such as birds,
or figured materials, as well as in tapestries, clearly showing
in every case the effect of highland industries.
Certain kinds of stripes are peculiar to this period, and
may be traced to the same influences, for instance narrow
stripes in two to four of the above colors ; 11 also patterns in
which wide stripes, frequently brown, are bordered by narrow
ones in the same color, often repeated over the surface of the
material. This pattern is very often found on the ponchos
from graves in the Sun temple. 12
Tapestries of the earlier era were either of pure woolen
warp or part wool, part cotton; cotton being sometimes used
for white in the pattern. Later, occasionally cotton is used
for both white and brown, and wool for the other colors.
Tapestries of pure cotton only exist from the last pre-Spanish
period, and these are usually in the highland color combina-
tions of brown, blue and white, though they occasionally vary.
The three-ply weave (as in fig. 92) is, in a certain degree,
5
1 L. C., 59, fig. 2; 66, fig. 3; 68, figs. 5, 6 ; 67, fig. 4.
2 Comp. l. c., pl. 55, fig. 1.
3 Comp. 1. c., pl., 55, fig. 1 ; 40, fig. 2.
4 For instance like l. c., pl. 66, fig. 3.
5 Bastian l. C., pl. 4 (on the central strip of the gauze poncho).
6 Comp. l. C., pl. 15.
Compare several modern cloths from the highland in the Stübel collection
at Leipzig ; also one in Kultur u. Ind., II., pl. 12, fig. 6 (from Chachapoyas).
8 Comp. Reiss u. Stübel, pl. 24, fig. 1; 38, pl. 12, fig. 1.
9 Comp. I. c., pl. 24, fig. 2.
10 Compare the stuff, 1. c., pl. 57, fig. 1 ; 24, fig. 3.
11 L. C., pls. 24, 38 and 57.
12 Also pl. 19, fig. 11.
(67)

6
7
typical of America. Almost, if not entirely, unknown in
Europe, it is found generally in Peruvian textiles, also
among the Pueblo Indians. The weave of fig. 92 produces
a cloth with reciprocal colors on both sides. This class of
textiles belonged to the earlier period, and is often represented
at Pachacamac. Older fabrics show dots over the surface, the
ply being doubled (as in fig. 57, above, p. 44). This weave
allows only for two colors, therefore figures of a third color
are embroidered into the cloth. Textures with a dotted sur-
face are made to this day in the same weave in the highland
of Bolivia. Many designs in brown, gray and white are
1
quite similar to those of the lowland during the last period,
except that all three colors are evenly woven in. The texture
is of this nature: On one side of the stuff, thread No. 1 covers
thread No. 3, while No. 2 remains visible as a dot; on the
reverse side, thread No. 3 covers thread No. 1, while No. 2
again appears as a dot. This weave is only exemplified in
3. Openwork patterns in textiles. The openwork weave
produced by interchanging and drawing together the warp
threads, appeared in earlier years, but the application of these
methods of weaving to more complicated patterns in the cloth 6
seems to be of earlier date. This applies especially to the
delicate embroideries with openwork ground. Of figures,"
birds, puma-heads, etc., Pachacamac gives us a number of
examples from this last period, and from this only. In shape
the textiles generally correspond to those of the earlier periods,
with one interesting exception.
In graves of the earliest times we find certain long shirts
(fig. 94), with openings for the neck and arms. They have
as an ornament a square piece of plaited white cotton sewed
on in front, on the breast. These garments are, perhaps,
referred to by Cieza, I, chap. 62, when he describes the cos-
tume worn by the women of the coastland: “All were clothed
in their ponchos and long cloaks; the women also, only that
000
V
Fig. 91. Leaf design. Tapestry border.
From a grave of outer city. Inca
period of the coast land.
Fig. 92. Three-ply weaving of ancient Peru-
vian textiles. From graves of outer city.
Inca period of the coast land.
Fig. 93. More developed three-ply Peruvian
textures. From graves of outer city. Inca
period of the coast land.
their garment was ample and long, like a Spanish closed cape,
open at the sides to allow the arms to pass through.”9 Gar-
ments as represented (fig. 94), or fragments of plaited fabric
which may have served as ornaments upon similar articles of
clothing, are of frequent occurrence in graves of women
dating from this period; their pre-Spanish origin is thereby
attested. This is, curiously enough, the first and only
,
VID)
..
=
பாவம்
0.
cloth showing the color-scheme of brown, gray and white
borrowed from the highland.
Patterns embroidered into woven cloths also belong to
this period; they are of the same kind as the fabric from
Ancon (pl. 59 a, fig. 3, Reiss u. Stübel ). There are, in addi-
tion, quite a number of different patterns, colorings and stitches
Besides the above, there are still three more technical
methods apparently belonging exclusively to the Inca period
of the coast, and which show no relation to imported cultural
elements :
1. Woven borders or braids of the class shown by Reiss u.
Stübel (pl. 67 a, fig. 1). In these braids threads of one color
only run as woof through the whole piece, in the example
cited, red or yellow, while other colors are worked in by
embroidery. The texture has only one right side ; on the
wrong side the threads appear loose wherever noi required in
the pattern. This method was followed to a great extent in
later times, especially for poncho borders, of which numerous
examples are to be seen. A poncho taken from one of the
Inca graves, and ornamented in the style of the coast, has
such a border, the pattern of which was reproduced above
(p. 38, fig. 51).
2. Surface embroidery upon cotton cloths with loose
woolen threads. The embroidery threads cover the material
quite loosely in large sections, being intended for wide spaces,
and are laid side by side. 3 Of this class is the pattern of
some modern scarfs (fajas) from Ecuador in the Stübel collec-
tion at Leipzig. 4 At Pachacamac, surface embroidery belongs
as exclusively to latest Inca times as the method of embroid-
ering upon cloth, giving a rep or tapestry-like appearance,
belongs strictly to the earliest period; it is extensively used
at this time. The pattern of the garment from Pachacamac,
pictured by Hamy in Gal. améric. du Trocadero, pl. 48, as
well as many similar ones, have been produced in this manner.
Fig. 94. Woman's garment with knife-plaited trimming. From graves of the
outer city. Inca period of the coast land.
instance in all America during pre-Spanish times of the use
of knife-plaiting as trimming for garments. The origin of
the 'kuchma' is now also clear; this is a garment resem-
bling a poncho, knife-plaited, worn at weddings by the
Indians of Otavaló, in Ecuador, 10 and which could not be
satisfactorily traced back to an imported Spanish custom.
5
Comp. W. H. Holmes, 6 Ann. Rep. of the Bureau of Ethnology, 1888,
pp. 211, 309.
Comp. I. c., p. 212, fig. 311.
8 In some graves pieces of gauze are numerous, folded in a triangular way,
or lengthwise into a band, and used as a sort of shawl. They were tightly
sewed together lengthwise, with a border. It is not certain if these articles
belonged exclusively to the earliest period.
9 The two examples in the collection have such small openings that it is
doubtful if they have ever been really used. Perhaps these garments had been
purposely prepared to serve as covers for the mummy bales.
10 Kultur u. Ind., II, pl. 13, fig. 2.
1 Comp. Kult. u. Indust., II, pl. 14, figs. 1-4.
2 Textiles of the same class were repeatedly found at Pachacamac.
3 Comp. Reiss u. Stübel, pl. 66 a, fig. 3.
4 Kultur u. Industrie, II, pl. 11, figs. 4, 5; 12, fig. 5.
Comp. pl. 6, figs. 1-4.
5
(68)

C. IMPLEMENTS OF METAL.
attributed to Cuzco influences. The general shape of the
plate in pl. 13, fig. 16, is the same as that of the clay plates
of Cuzco style, with a round handle. Birds as ornaments
for plates have not yet been traced, but the bird ornament
a
A few objects made of copper from cemetery VI are
shown in pl. 13, figs. 21–26. They represent an advanced
stage in the use of metal, in regard to size, technique and pur-
pose, when compared with the insignificant specimens found
in earlier cemeteries.
Fig. 21. A knife, ornamented with a human face, made
with a handle, showing that this tool was used by hand.
Fig. 22. An implement resembling a chisel, having a
well-preserved handle of dark wood. A wide groove sur-
rounds the handle at one end, which is flattened on one side
to support the blade. A small fragment of thin cord, with
which the latter was fastened, is still preserved.
Fig. 23. An ordinary flat chisel. To judge by another
specimen, the clřisel must have been fastened into a handle of
Fig. 95 a.
Fig. 95 6.
Bowl carved of wood. Shape of a fish. From the graves of the outer city.
stag horn.
upon this one may have been an imitation. Its general
appearance is that of the Cuzco type, and it may have been
carved by some denizen of the highland. Fragments of
other wooden plates, good imitations of Cuzco plates, without
handles, were frequently found in the houses of the inner
Wooden bowls in the shape of birds or fishes of the
type of pl. 18, fig. 15, were also occasionally discovered. In
the bowl, figs. 95 a and b, the fish is formed differently and
resembles the fishes upon a wall of the temple of Pachaca-
mac. This type is undoubtedly of Cuzco character.
Fig. 24. A necklet of a narrow strip of copper, the ends
of which are bent so as to form hooks.
Fig. 25. Earpick, the upper end of which forms a well-
executed figure of a bird with long beak.
Fig. 26. Round object of two pieces of hammered copper,
joined together on the rims. It is ring-shaped, with a hole in
the centre. A similar object, both as to size and shape, made
town.
а
Shah
Fig. 96. End pieces of a weaver's beam. From
the graves of the outer city. Inca period of
coast land.
Fig. 97. Small weaver's knife. Carved wood.
From graves in the outer city. Inca period
of coast land.
Fig. 98. Netting needle with carved ends (fish).
From graves of outer city. Inca period of
coast land.
a
of clay, serves as a rattle, and differs from the preceding only
by being open in the centre and not perforated on the surface.
It was probably filled with small pellets of clay, or with small
stones, and fastened upon a stick. This same construction is
to be seen in the magic rattles (“maraka') of eastern South
America, which are made of vegetable shells. Several silver
finger-rings were found in the cemetery.
The tweezers in this burial-ground are of the usual well-
known shape. One pair, possibly of later date, was found near
a house in the northern part of the city. The shape is a
developed form of the original bivalve sea-shell.
Wooden spoons, with a fish, a bird or a llama's claw upon
the handle, were found occasionally. As spoons are very rare
in earlier burial-places, it may be safe to infer that the use of
this utensil was extended through the Inca conquest.
A small idol of wood, inlaid with pieces of shell, is shown
(pl. 13, fig. 18); although of crude style it is carefully made.
Wooden dolls of this period may easily be recognized by the
triangular shape of the head. 4
Weavers' beams, weavers' slays and netting needles, orna-
mented like figs. 96, 97 and 98, were found in cemetery VI,
but not in older burial-places. The same applies to the well-
known variegated spinning-whorls made of clay. In this
burial-ground were also found two types of ear ornaments (pl.
13, figs. 19 a and b and fig. 20). One is a plain ring of dark
wood, one and three-sixteenths to two and one-fourth inches in
diameter, one-half to five-eighths of an inch thick-in a single
instance only fifteen-sixteenths of an inch—and grooved along
the outer rim, which is smooth. On one side are carved birds,
repeated again and again, exactly alike on all specimens,
embossed by a few incisions. In one case two rings of the
D. UTENSILS OF WOOD.
These show highland influences, either in shape or orna-
mental detail. Three wooden cups were found, not in the
cemetery, but in analogous spots in the town. Two, entirely
alike, were taken from the grave of a young girl at the south-
ern foot of the hill of the Sun temple; the third was picked
up during excavations in the terrace of a house, where it had
probably found its way during a libation while the house was
being built. All three cups, from their decorations, must be
3 See above, p. 21.
4 Reiss u Stübel, pl. 89, fig. 1.
Comp. Reiss u. Stübel, pl. 86, figs. 3-17.
5
6 Spindles and weavers’ beams sometimes contain small clay pellets, neatly
introduced into small cavities as rattles.
1 Reiss u. Stübel, pl. 81, figs. 22, 23.
2 Comp. I. c., figs. 2, 4, 6, 8.
(69)

4
vessels, when found on the coast, must be regarded as dating
from the last pre-Spanish period. A gourd from the graves
of the Sun temple shows ornaments burnt into it, although of
a simpler Cuzco type.
Flasks, represented in fig. 99, consist of three parts : the
bowl, made of a gourd; a cylindrical piece in the centre, made
of wood, and a small stick of wood or bone, extending through
the cylindrical part into the bowl and finished at the top with
same kind are tightly fitted into one another. The second
type, fig. 19 a and b, is represented by three specimens. It
consists of a hollow wooden plug one and one-fourth inches
high, the front side with a diameter of nine-sixteenths of an
inch, and inlaid with shell fragments forming a man's head,
surrounded by simple ornaments. Both types are wanting in
the earlier burial-grounds. Cylindrical ear-plugs from the
region of Lambayeque, belonging to the second type, had a
diameter of one and one-sixteenth by one inch in height; they
showed on the front side an inlaid figure of a fish, formed
of bits of shell. Since all these ornaments were obtained
from graves of earliest date, they were probably made in
accordance with regulations imposed by the Incas, as to ma-
terial, decoration and size, and were the ear-plugs worn as
marks of distinction, of a special kind for each province. I
To this class may belong the second type, being closely similar
in all three specimens, although each is of a different pair.
In the first type the general character is uniform, while the
size varies. 2 The occurrence of two distinct types in one
cemetery still requires future elucidation.
H
ER
Fig 99. Lime-flask for coca-chewing. From graves of the outer
city. Inca period of coast land.
E. GOURDS.
a knob, prettily ornamented in some specimens. Flasks of
this class are found in this burial-place exclusively. They
still contain lime, which was used with the chewing of coca.
Similar flasks for lime are still in use in the highland, and it
must be noted that their occurrence on the coast is limited to
the Inca period.
Vessels of this class are always quite plain in the earlier
cemeteries, while those found in the outer zone of the city are
decorated with burnt-in figure and scroll designs. These
3
3 All vessels made of gourds, Reiss u. Stübel, pl. 84, figs. 5–12, must be
ascribed to the last pre-Inca period.
1 Garcilaso, I, chap. 23.
2 The width is generally uniform, except in the one instance. According to
Garcilaso, the width was regulated by the Inca.
4 Compare also the modern gourd with burnt decorations from Chacha -
poyas in the Stübel Collection at Leipzig, Kult. u. Ind., II, pl. 4, fig. 11.
(70)

CHAPTER XVIII.
DISCOVERIES IN OTHER PARTS OF THE VALLEY.
SUCCESSFUL operations in the ancient city naturally
aroused an interest in the cultural conditions of other
parts of the same valley. Lack of time forbidding extensive
investigations, only two burial-places in the vicinity were
examined, the one being situated to the east from the Centi-
nela rocks in the hacienda Huarangal, on the southern side of
the valley, about two miles from the ancient city ;1 the other,
at a similar distance, in the vicinity of the hacienda Ventu-
rosa, 2 towards the northern limit of the valley.
1 v. Tschudi, Peru, II, p. 301.
2 See above, p. 5.
THE CEMETERY OF LA CENTINELA.
2
a
Specimens obtained in this burial-ground entirely corre-
spond with those of cemetery VI of the outer city. They are
of the last period of the coast style, as well as of the Cuzco
type, together with blendings of both, and must be classed as
belonging exclusively to the last era of the valley, that of
the Incas. It was surprising to find here several graves con-
taining silver vessels, and in one of these a pair of golden
tweezers. There were three graves with such contents. The
mummies were bale-shaped, bound with thick cotton cords,
interwoven with straw-rope. The use of cotton cords, instead
of ordinary straw-rope, is a distinct proof of wealth. From
the shape of the bale, it is evident that persons buried here
were natives of the valley, who practiced ancient methods of
burial, side by side with the newly introduced customs of the
highland—carrying on the forms of their own culture, besides
adopting those imported from Cuzco. The false head on
the mummy bale was ornamented with a face of hammered
silver. The use of silver vessels, like that of ear ornaments,
the length of the hair, and the style of headdress, was subject
to regulations by the Inca. Only the Inca himself, the Sun
maidens at Cuzco and at the temples had golden and silver
vessels for ordinary use, while the kurakas and chiefs were
only permitted such costly objects at grand ceremonies, their
number being strictly regulated. Common people might use
such vessels only as a special privilege, granted by the Inca.
Gold and silver articles in graves are, therefore, a proof of
the high rank of the persons buried there. 1
Two objects obtained from these graves are shown on
pl. 7. Figs. 22, 23 a and 23 b illustraté a trephined skull,
found with a large silver cup, ornamented with a face, and
remains of a feathered stuff. The cranium shows a slightly
flattened occiput, and is that of an adult in the prime of life.
Besides the trephining on the left side it shows the scar of a
former operation on the right, a little further towards the
frontal region, the inner bony part of which is one inch long
and seven-eighths of an inch wide, covered by a circular
cicatrice, forming a deep groove one half-inch in width. The
second operation of trephining was made on the left side of the
crown, removing an oval piece of bone (figs. 23 a and 6) by
a diagonal cut of one and one-fourth to one and one-sixteenth
of an inch in diameter. The cutting surfaces on the skull and
the piece taken out are surprisingly even, still a loss of bone
of three-sixteenths of an inch on the outside was caused by the
diagonal incision, while on the inner side it was not one-
thirty-second of an inch nor even in some spots one-sixty-
fourth of an inch. The bone was not cut through, and was of
the thickness of paper in one part, five-sixteenths of an inch
in length in the line of the cut; here in removing the
detached piece the bone had scaled off. The thickness of the
bony structure at the spot of operation is one-fourth of an
inch. The wound must have been of a very severe nature,
judging from existing indications, and the preservation of such
an instance of trephining is very unusual (figs. 23 a and 23 b).
A flat object, probably an arrowhead, had entered the skull
diagonally seven-sixteenths of an inch deep, with a point of
the width of a pocket-knife, not piercing but splintering the
bone radially on the inner side (fig. 23 b). The bone on this
side seems to have become so loosened during the operation,
that in one point it could not be cut any further and had to be
broken through. The operation was begun probably by the
removal of the foreign object which caused the wound, for
which purpose all manner of introductory incisions were made
to loosen it (fig. 23 a). Around the oval cut, the outside of
the cranium in a strip of half an inch to an inch is darkly
discolored by blood. It seems, therefore, that the scalp was
loosened from the skull to this extent prior to the operation, so
that the blood could color the exposed skull. This strengthens
the supposition that the patient died during or very soon after
the operation, as there is no cicatrization along the edge of the
wound, nor was the blood cleaned away. The opening in the
skull was closed by merely putting the detached piece of bone
back into place for burial, and they were, therefore, found
together.
a
а
2 In fig. 22 the detached piece of bone is shown in its original place.
3 The Bureau of Ethnology at Washington owns an excellent and rare col-
lection of trephined skulls from Peru, among which is one from Pachacamac,
illustrated in the important publication on these skulls by Muñiz and McGee (Six-
teenth Ann. Rep. of the Bur. of Ethnol., 1897, No. 11, p. 36, and pl. XXII). It
has an oval incision lfke that above.
1 Garcilaso, II, chap. 28; IV, chaps. 5 and 11 ; V, chaps. 11 and 12.
(71)

Another unusual cloth taken from the same grave is
represented in fig. 100. The size is ten feet by five and
three-quarters. It is made of cotton, of a thin brown texture,
which has sixteen by sixteen threads to the half inch. Into
this cloth are inserted at each end triangular pieces, woven
separately, each of which is three feet deep and has nine
gradations on both sides. A narrow dark stripe woven into
the cloth extends from these white insertions to the ends of
the piece. Simple as this cloth may appear, the manufacture
Fig. 21 shows an embroidered cotton cloth, taken from
another grave containing silver vessels. It is forty-six by
forty-three inches in size, well preserved and composed of two
pieces. It is ornamented with all-over embroidery, described
on p. 43, No. 2. Fig. 21 reproduces one-fourth of a section
of the cloth. The figures embroidered upon it with woolen
threads resemble birds, and are repeated all over the surface;
the colors are purple, blue, yellow, and brown, alternating.
The ends show only one-half of a narrow embroidered border
in blue and purple, on account of having been composed of
two sections. The cloth belonged to a chieftain, being found
together with silver vessels. Owing to its colors, blue and
purple, the colors of the Inca, and those of his nearest of
kin, the Qhapay aillu, yellow and red, it gains a certain
deeper significance. The brown color, which seems out of
place, might have been intended to mark the province, as it
does in the kipus.? It is possible that the lower grade took
its own color, and added it to that of the higher, thereby
making the color combination more significant. This would
explain why the fewer colors of the Inca were so generally
recognized and respected, that one thread of the frontal fringe
of the Inca was sufficient to secure for its wearer the deepest
submission to his orders. It is possible that the colors of the
Inca may owe their origin to the synthesis. According to
Oliva, 2 the god Wiraqocha emerged from the ocean in a violet
tunic and a crimson cloak. Red and blue, accordingly, might
be the colors of the universe. Crimson in the knotted strings
signifies the Inca, and blue might be the color of the origin of
the world, tiksi.” Blue is also among the colors which signify
religion.4 Blue, yellow and white might be the combination
of the Creator, blue, the sun, white, and yellow instead of
brown (? ) for the wakas, representing the three forms of cult.
1
a
Fig. 100. Large cloth from the grave of a chieftain. Graves of
La Centinela. Inca period of the coastland.
a
of a fabric ten feet long, and in which large triangular
pieces had to be shaped in the weaving, requires a certain
amount of skill. The size of this cloth precludes its use as a
garment. As this grave, from all indications, is that of a
chieftain, it is possible that the cloth is a mark of his rank,
the brown color representing the province. Considering the
character of the ornamentation, it must have been intended
for distance effect, and the decoration of both ends gives
ground for the supposition that it was used as a pall.
5
1 Uhle, A modern Kipu, Bulletin of the Mus. of Science and Art, Phil., 1898.
2 In v. Tschudi, Beiträge, p. 130.
3 Elsewhere it means 'nothing,' refusal or subtraction (see Uhle, l. c.).
4 Uhle, 1. c.
5 Brown as the color of a province was at the same time that of the pro-
vincial waka cult. White signifies day, probably also sun. The sun-maidens
and priests were clothed in white (see below, chap. xxi).
6 Comp. the ornaments on the fabrics shown by Bastian (l. c., pl. 5).
THE GRAVES OF VENTUROSA.
2
Two vessels only were secured from these graves, one of
them being decorated. The design upon this object (pl. 7,
fig. 10) consists of three birds in frontal view, their type being
the same as those upon the arybals of the Inca time, and
clearly an imitation. This shows that during the period of
the Incas some settlement must have been situated in the
vicinity of the present hacienda Venturosa. The discovery of
graves dating from the Inca period wherever explorations were
conducted in the valley outside the city, refutes the theory?
that Pachacamac was abandoned during that time and left to
fall into ruins. Whatever the Inca rule may have been in
other parts of the coast, it was beneficial in the Pachacamac
valley. Settlements were then extended outside the city,
probably for the main reason that personal safety was guaran-
teed to the settlers in the valley.
1 Rivero and Tschudi (pl. 36). The same vessel also Bastian (1. c., pl. 3,
fig. 2).
2 See above, p. 6.
Inca period of
Fig. 101. Decorated vessel from graves of La Venturosa.
coast land; À n. S.
(72)

CHAPTER XIX.
THE SUN TEMPLE.
3
THE
HE Sun temple, situated upon the hill, W, and the con-
vent at the western end of the city are the only promi-
nent buildings of the old town of Pachacamac which date
from the Inca period. The Sun temple has been more fre-
quently described than any building in the ancient town, still,
so far only one plan exists of it, and four or five views by
George Squier, which, however, give no adequate idea of the
edifice. The writer here offers a more reliable plan (pl. 161),
elevations of the four fronts of the Temple (pl. 17) and a
number of photographic views in reproductions, pl. 2, fig. 1,
the Temple from the north ; pls. 12, 14, 15, figs. 1, 2, which
will assist in the study of its details. 2
On the whole, the dimensions of the Temple have been
so carefully established by the plan, the notes, and the meas-
urements of elevations, that at any time it will be easy to
make any number of cross-sections, twelve of which have
been prepared and are in the collection. 3
The bill W may be two hundred and fifty to three hun-
dred feet high, and is of a rocky nature. Its base is broad,
and slopes down in many curves ;º at one point only, on the
southeastern side, near the Temple, it is somewhat steeper. A
narrow ridge, lower than the hill,' continues it in a north-
eastern direction, and at the foot of its northwestern slopes
was the site of the ancient temple of the Creator as above
described. Excluding the Temple, the circumference of the
hill at this point may be one mile and a quarter.
1 The plan of the Temple is especially difficult to obtain on account of its
intricate construction and the ruined state of several terraces, therefore some
glaring errors in Squier's plan (p. 69) must not be taken too seriously :
1 The confounding of north and south, the same as on p. 275 in the plan,
and the description of the Ruins of Tiahuanaco by the same author (see Stübel u.
Uhle, p. 15 b, note 5).
2. The rectangular shape of the structure.
3. The complete misconstruction of the plateau, absence of structures on
the southern part of the plateau of the entire fourth terrace of the southeast front,
the misplacement of the centre of the western front too far south, whereby the
rooms adjoining it on the north become detached from it, and those on the south
side too much crowded.
4. Misdrawing and misinterpretation of the third northern terrace.
5. Misinterpretation of the displacement of the terraces between northeast
and southeast on the eastern angle.
6. Removal of the ascent of the second terrace from the inside to the front.
2 Among earlier views those of Hutchinson (p. 159) and of Wiener (p. 484)
are intended for reproductions of the west front of the Temple, shown in pl. 15,
fig. 2. Those by Wiener (p. 62), Rivero and Tschudi (Atlas, pl. 54), show the
Temple from the northwest. The view given by Squier (p. 68) and one by
Wilkes are intended to show the north angle of the Temple, but are not exact;
better views of the same point may be seen in the collection.
It might be suggested that a perfect model in clay could be made of the
Temple in its present condition (1:100 or 1 : 200) from photographs and exact
measurements. In some respects it would be more instructive than the original,
as its small dimensions would facilitate the study of the structure, while at the
same time it would preserve the memory of the famous monument which seems
doomed to destruction at no distant period.
4 Wilkes and Hutchinson (p. 158) estimate the height as above; Markham
and Middendorf, 400 feet; Rivero and Tschudi (p. 288), with remarkable correct-
ness, 458 feet ; Cobo and Squier (p. 68), 500 feet; v. Tschudi, Peru, I, p. 291,
558 feet ; Bancroft, Native Races of the Pacific States, V, p. 796, 600 feet. In
some of these estimates the difference of the original hill from the summit of the
building is too high, so that of Wilkes.
5 Hutchinson, I, pp. 160, 172 and 176, regarded the hill, with a remarkable
lack of judgment, as being artificial and piled up above a large cemetery. San-
tillan is doubtful on this subject (p. 32). Wiener addresses an unscientific public;
as in nearly all his work, so in his erroneous explanation (p. 711), when he
describes this hill as being of breast-shape, representing in a figurative sense the
* creating Earth’ under the Pachakamaj temple (elsewhere he calls it the Sun
temple).
6 See plan of the city, and pl. 2, fig. 1.
7 See plan of the city, pl. 1.
8 Compare pl. 2, fig. 1 ; 9, fig. 2 ; 13, fig. 1.
SITUATION AND GENERAL REMARKS.
2
The distance from the base of the hill to the sea may be
estimated at about one-third of a mile. As the Temple lies
from six hundred to one thousand feet away from the south-
western foot of the hill, the present distance of the Temple
plateau from the sea would be about one-half mile, as estima-
ted by Markham. From the prominence of its site, the
Temple gives the impression, as Squier says, “that it reaches
close to the shore, so that the ocean may be said to break at
its feet.” On the side facing the ocean the most delightful
sea-breeze is felt all day long, making the crest of this hill a
pleasant place, even on the hottest day. The Temple with its
broad base has been built around the top of the hill and into
a ravine, as Squier expresses it, facing toward southwest, as
may be seen both on Squier's plans and in the views here
shown (pl. 14, fig. 2). The edifice really is fitted into and
around the hill, following its form closely. Although the
section of the hill-crest with the Temple as given by Squier
is quite imaginary, it still serves to give a general idea of the
manner in which this structure was built around the hill, and
which accounts for the varying height of the masonry on the
different sides. It is seventy-five feet in the ravine in the
centre of the southwestern front, and only forty-nine feet over
the western base of the hill, 4 while at the extreme western
end, within the terraces, the hill was left in its natural state
near 69, 26 and 29, and rises there at a distance of seventy
feet from the plateau to a height of only sixteen feet below it.
4
2 The natural hill may be seen on the entire lower terrace on the west side
(comp. pl. 14, fig. 2; not so in Squier's plan). It is not known how deep the hill
lies beneath the masonry, outside this one point. The summit, as given in Squier's
cross-section, is also imaginary.
3 See pl. 17, cross-section 1.
4 See pls. 16 and 17, fig. 2.
1P. 68.
10
(73)

ance.
6
1
2
may be seen
The Temple is an imposing terraced structure, built of
adobe. It has four fronts. The shape is trapezoid, conform-
ing to the outlines of the hill, the longest front looking sea-
ward in a southwestern direction. The northeastern and
northwestern fronts are nearly parallel, the two others con-
verge in a short angle toward northeast. The sides measur-
ing 715 feet (southwest), 515 feet (north west), 565 feet (north-
east) and 347 feet (southeast)," it is safe to estimate its area at
about twelve acres ; 2 and as the height varies on the three last
mentioned sides from forty-nine to sixty feet, 3 it
that it is in a very small proportion to the size of the Temple
about one-tenth to the length of the sides. The Sun tem-
ple of Pachacamac is one of the most prominent among the
Peruvian terraced buildings, while it cannot compare in size
with the largest Mexican structures. The pyramids of Teoti-
huacan surpass the Sun temple of Pachacamac in the length
of the base on three sides (about 680 feet) as well as in height
(about 180 feet); the pyramid of Cholula is twice as long in
its base (about 1300 feet), and has more than three times the
height of the Sun temple. The proportion of the length of
base to the height is quite different in the Mexican pyramids ;
the proportion would be of 1:13 to 1: 6.
The Sun temple, nevertheless, is an imposing and mag-
nificent structure, and must have been much more so when
seen by Cobo early in the seventeenth century, from whose
description we are now able to reconstruct some parts no
longer existing in its present state of ruin. The entire build-
ing was painted red. An imposing main front, decorated
with niches, faces southwest, and presents a beautiful appear-
The opposite front, facing northeast," is noteworthy on
account of the passageway which winds up to the terraces and
the plateau, and for the temple-cells on both extreme ends of
the same, rising above it symmetrically for more than twenty
feet.? A vast plaza extended over the summit of the struc-
ture, facing northeast and southwest, and bordered by temple-
cells on the other two sides. Here the great festivals in honor
of the divinity were held. The view from this point is superb,
commanding a panorama embracing the town, the desert back
of it, the convent with the adjoining fields, the ocean and the
islands near the shore, 8 the valley in its perennial verdure
against the background of the desert, and to the left the
towering peaks of the Andes.
the Andes. Even the terraces of the
northwest, northeast and southeast sides were by no means
monotonous. Their grouping was peculiarly varied; they
had different details and formations, while various superstruc-
tures broke the straight lines, and the pious and initiated
native may have attached special ideas to each part of the
structure, always in his full view from the town.
1 Former measurements, Wilkes : length of the Temple, 500 feet, width,
400 feet.
2 Comp. pl. 17, fig. 4.
Squier gave 54 feet, Cobo 74 feet (Span.) as the height of the Temple, a
difference which may be explained from the different sides where the measure-
ments were taken.
* Pl. 14, figs. 1 and 2. From the existence of a main front of this order,
around which three terraced sides of less importance were grouped, Squier (p. 68)
and v. Tschudi (Peru, I, p. 291) derived the rather unfortunate description of the
structure as “crescent shaped” or as “resembling an amphitheatre.”'
5 Pl. 13, fig. 1.
6 Pl. 15, fig. 1.
7 See the reconstruction as marked in pl. 17, figs. 2 and 3.
8 Comp. pl. 13, fig. 1; pl. 14, fig. 3.
THE NORTHWEST, NORTHEAST AND SOUTHEAST FRONTS OF THE TEMPLE.
3
1
It is fortunate that the main lines of the front were so
simple and regular that an adequate idea may be formed of
many details of the original construction without making
many excavations, simply by studying the existing parts of
the building. So far the descriptions of the terraces were
misleading in regard to their number and form, as well as
their height, width and relative position, which accordingly
were assumed to be far too uniform. Instead of measuring
each terrace separately, travelers were satisfied to take meas-
urements at random, at any spot, often not even typical ones.
The results were the most varying descriptions, frequently
giving no remote idea of the size and individual character of
buildings.
The system of terracing on the north west, northeast and
southeast sides, compared to that on the fourth, the southwest
side, is, as a whole, quite uniform. The foot of the Temple is
correspondingly regulated in a symmetrical plan on these
three sides 2 and different from the southwest sides, where it
is formed by the natural slope of the hill. Its height and
width both vary about ten feet. In this it differs from the
terraces. It is also graded and distinguishes itself from the
surroundings by artificial slopes, so that P. Bernabé Cobo
described it as the lowest terrace. Kitchen refuse, containing
chips of pottery, fragments of fabric, parts of shells (mytilus)
cast out from the Temple ( near 75), were thrown down the
northeast and southeast slopes.
The system of terracing turns at the southern angle, near
41, toward the southwest front of the Temple, in such a man-
ner that the second terrace, southeast, lies on the same level
with the southern part, 23, of the second southwest terrace.
But here was a dividing wall, and this piece, 23, of the second
terrace, southwest, extended without connection along the
third and fourth terraces, near 40 and 39. Near the corner,
the terraces of the southwest front and that of the northwest
are in no wise connected, on account of the hill which sepa-
rates them. The walls divided near 26, 28 and 29, only three
areas of various sizes, like courts. The second and third
terraces of the northwest front terminate near 65 and 64 with
walls. 4 The formation of the hill did not prevent the con-
tinuation of the lowest first terrace farther toward the west ;
this latter therefore terminated only below 69, near a former
wall, 68, in the line (33) of the northern continuation of the
second terrace of the southwest front. But since the hill rose
above the first terrace and, near 67, above the foot of the Tem-
1 Very mistaken views concerning the period of the destruction of the Tem-
ple were given by v. Tschudi and Hutchinson, which originated from their failure
to see the Sun temple in this building. The Temple was not in such a state of
decay at the time of the Incas (v. Tschudi, Beitr., p. 127), or at the time of H.
Pizarro's visit, as it was when Stevenson saw it in this century (Hutchinson, I,
p. 167), for Estete describes it as being beautifully ornamented with niches ( bien
labrada'), and at that date, evidently, the building was as good as new. It
doubtless suffered much during the conquest, but probably less than the temple
of Pachakamaj, the destruction of which alone by Hern. Pizarro is reported
(Estete). Cieza does not specially mention it as being in an advanced state of
ruin. When Cobo saw it in the seventeenth century it was indeed 'desportil-
lado,' but the temple of Pachacamac must have been much more so. He could
still see in an almost complete state of preservation the winding passageways up
to the terraces, the height of the walls of the cells on the plateau-then twice
their present height-and the roof of the uppermost terrace of the south west
front. It is probable that Roman, before the close of the sixteenth century,
saw carved ceilings in the Temple cells of the plateau, as may be assumed from
his words. Under these circumstances there is no need to resort to earthquakes
as causes of the early decay, as is done for instance by v. Tschudi. Adobe walls
crumble to pieces during the course of a few centuries without any additional
destructive agent. In what degree earthquakes may have contributed to the ruin
cannot be conjectured.
2 See the diagrams 2-4 in pl. 17.
3 Comp. plan, pl. 16.
4 See pl. 17, fig. 2.
(74)

1
ple, probably from æsthetic reasons, the hillside west of the
second terrace end, near 63, was built over with another second
terrace, and west from the end of the first terrace, near 66,
with still another first and second terrace, which gradually
merged into the hillside. As a whole, four terraces may be
distinguished, including that of the plateau, on the northeast
and northwest sides, while there are five on the southeastern
side. At the eastern angle, 52, of the first terrace, the system
of the two former sides is interrupted. The southeast side is
divided into terraces of different degree, branching out into a
fifth terrace. During the transition into this altered arrange-
ment of the front, the first, second and third terraces of the
northeast side are raised in a sort of half-story, three and a
half to nine feet, and then form the second, third and fourth
terraces, while a new first terrace is constructed outside of
these; the latter is only about four feet lower than the first of
the northeastern side, but is separated from the former by the
masonry of the second southeast terrace. But, as if asserting
its claim to be a part of the Temple, its eastern angle (44)
lies in the direction of the first northeastern terrace front,
while the terrace itself, with its superstructures of rooms ( 45,
46), projects farther toward northeast.
Although the terraces vary in width, they are generally
laid in parallel lines and symmetrical with the outlines of the
plateau. This regular arrangement has been reversed in the
terraces of the north western front, and in the lowest of the
northeast front; the frontal lines of the second and third ter-
races northwest converging with the edge of the plateau above
them on the north, while that of the lowest terrace again
pursues the contrary direction of the former. In a symmet-
rical way the lowest terrace, northeast, joins closely at the
eastern end (52) the second terrace, but leaves it again to
follow a northern direction, thereby becoming united with the
first terrace, northwest, at almost a right angle (60), resem-
bling that of the northern angle of the plateau just above it,
but facing differently. This creates an open plaza, near 60,
more free and open than any spot on all the other terraces,
and having the peculiarity that the terrace near 66 has the
edge of the plateau above it in a perpendicular line, which
occurs at no other place.
The terraces of the northwest front are the best preserved
of all the three sides, but even here the second is almost
entirely ruined. The same is the case with nearly the whole
of the second terrace and the fourth (the plateau ) on the
northeast side;s also the terraces on the east corner and along
the entire southeast side. Nevertheless it was possible to
ascertain from these remains the entire plan, and to prepare
diagrams of the fronts in minute detail, as shown by pls. 16
and 17, figs. 2–4. The height of the terraces on these sides
varies between twenty and ten feet; it is mostly about thir-
teen feet, and generally decreases in the upper stories. The
width varies from five and a half feet in the third terrace of
the northwest front, in some parts, to fifty-nine feet in the first
terrace near the northern angle ( 60 ). The general width is
between thirty-three and forty-six feet. The grading of the
terraces seems to have been perfect, although at present in
their state of ruin this is no longer the case. The explora-
tions by treasure hunters had a great share in the devas-
tation, especially upon the first southeast terrace on the
southern end. In many instances the terraces were sub-
divided into lesser grades, and occupied by a number of
smaller buildings.
The most varied gradations were found in the southeast
side, which now unfortunately is the most ruined. Thus the
second terrace to the east sloped in one grade, the third in two,
the fourth in one; on the west side the second was in one
grade, the third in two, the fourth in three. While the ter-
races of the northeast front were all on the same level, those
of the northwest front have in the third terrace a remarkable
and quite well preserved subdivision. Its eastern part
formed a ledge between the third terrace northeast and its
western part (62). On the western end of the ledge, how-
ever, another piece of the terrace (63) projected above the
original level like a half-story, and that continued westward
in its divided form. One grade (48) extended as a half-ter-
race on the inside of the third terrace northeast across the
front, as far as it may still be traced in its present ruined
condition. Indications of similar small terraces are to be seen
near 53 and 55, on the northeast foot of the structure. A
short projection adjoined the half-terrace (48) at the back
near 49, on the third terrace northeast. A retaining wall
(72), which probably supported a roof, was built against
the wall of the fourth terrace near 61. On the third ter-
race' there is a slight elevation like a bank along the rear
wall of the second northwest terrace, near 73.10 Rooms
were built upon the different terraces, the remains of which
are to be traced in the plan (pl. 16 ) at the following points :
45, 67, 43, 42, 47, 50, 54, 56, upon the first terrace north-
west, the first terrace northeast, and at the foot of the Temple
northeast and northwest, on the first second and third
terraces southeast.
The superstructures were of varied nature, as shown
by the plan, with regard to extent, division and the man-
ner in which they were set upon the terraces. Rooms
(67) were built against the front edge of the terrace, and
the passageway lay behind them; those at 50 were set
against the rear of the terraces, leaving an open walk in
front four feet wide. Rooms (43) apparently occupied
the entire depth of the terrace,11 one wall being set
upon the front, while the others were built against the
2
3
rear.
Very curious are rooms 45, 46 and 47 ; the first two are
on the first terrace southeast, and 47 is built up against them
at the foot of the Temple. 12 In the adjoining building (56 ) a
connecting passageway (58) may be traced, apparently con-
tinuing over (57) to the first terrace southeast, and forming
another side entrance to the temple terraces. Numerous
evidences of room partitions are also noticeable on the
first terrace northeast near 51, and on the third terrace
southeast, west from 42. Remains of small square rooms,
partly fallen down the slope, may be seen near 70 on the
1 It is somewhat difficult to count the terraces, as may be seen from the fact
that Squier in his plan has partly overlooked one on the southeastern front (the
third), and another (also the third) on the northwestern front. But when the
number of the terraces is given as three (Ulloa, Markham, Reports of the Discov.
Intro., p. xix) and elsewhere as six, the difference is caused by the manner of
counting, either by taking on the northeast and northwest sides the terraces only,
without the plateau, or by counting all half-terraces, and also the foot of the
Temple (Padre Cobo). Miquel Estete numbered five terraces, which may easily
be done, if his manner of counting is followed.
2 See pl. 2, fig. 1.
Comp. pl. 13, fig. 1.
4 Comp. pl. 13, fig. 2.
• Comp. pl. 17, figs. 2-4. The decrease in the height of the terraces in the
upper grades was noticed by Middendorf, II, p. 114.
6 Earlier measurements reversed the proportions between height and width
and gave general extreme measurements of all the terraces, which are so varied
in their relative size. The general height was given by Rivero and Tschudi as
thirty-two feet; the same by Markham following Cieza ; Bancroft (probably on
the authority of Rivero and Tschudi) gave it as thirty feet; Markham, Cuzco,
twenty feet; Hutchinson, six to eight feet; Middendorf, lower terraces, seven to
eight and one-third m. = 25–30 feet.
25-30 feet. The width was stated to be seven to eight
feet by Rivero and Tschudi ; five to eight feet by Bancroft.
Comp. pl. 17, fig. 4, and pl. 13, fig. 2.
8 Comp. pls. 16 and 17, fig. 2.
Comp. pls. 16 and 17, figs. 2 and 3.
10 Comp. pls. 16 and 17, fig. 2.
11 Similar near 42 on third terrace, southeast.
12 Comp. pl. 17, figs. 3 and 4.
Comp. pl. 13, fig. 1.
9
13
(75)

1
northwest foot of the Temple. Plain as these room struc-
tures upon the terraces are, they still show by the variety of
niches constructed upon them that they dated from various
periods. The niches in chambers 43 and 47 are of the type
described above (p. 60, No. 1). In rooms 50 and 54 are niches
of the type No. 3, 1. C., and an attempt was made to claim for
this type a later date. The impression is strengthened here
where it is apparent that the rooms form a later addition,
since near 54 the red paint of the original terrace wall was
continued on the parts covered by the more recent buildings.
1 Niches of a similar shape are also to be seen in the rooms of the convent,
built by the Incas.
THE TEMPLE ENTRANCES.
The plan of the city shows that the Temple with its
north angle, the whole of the northeast front, and half of the
northwest front lie within the enclosure surrounding the
sanctuary of the Creator god. As the entrances of the Sun
temple, which face northeast, are situated within the wall, it
is clear that the Temple as a whole was enclosed in it. This
serves as a proof of the amicable relations between the two
sanctuaries, which may be compared with the circumstance
related by Betánzos, that an image of the god Wiraqocha at
Cuzco stood in a temple dedicated to the Sun. Similar
relations, however, cannot be assumed to have existed at
Pachacamac, as Estete and others make a local separation of
the Sun temple and of that of Pachakamaj. The women of
the convent were, according to Garcilaso, consecrated to the
Sun, although H. Pizarro calls them the wives of the devil,'
thus referring to the god Pachakamaj. It is possible that
they served both divinities. Certainly it is an additional
proof of the good relations between the two sanctuaries that
the women had to pass through the outer and the inner courts
of the temple of Pachakamaj in order to reach the Sun tem-
ple. The wall of the enclosure must be of a much earlier
date than the Sun temple, judging from its advanced state of
decay, and the Sun temple was evidently built into its area.
The irregular way in which the enclosure runs into a point on
the northwest front of the Temple favors this theory. Appa-
rently the Temple had two entrances, both opening within the
enclosure, and both situated on its northeast side. The first
(35) of these was in the centre of the northeast front, and
continued as a winding passageway, cutting through the
various terraces, up to the plateau of the Temple. The
second entrance seems to have been near 56, as stated above,
and after passing through the passageway it led into minor
rooms, primarily to the first terrace of the southeast front.
The former was the main passageway up to the Temple. The
plan shows that the passageway leading to the three lower
terraces are still to be traced in some of their principal lines,
while that of the fourth terrace is completely obliterated or
choked with débris. It probably turned an angle, similar to
that of the lower passageways, over 38, about half way up
to the front of the plateau. Pl. 15, fig. 1, and pl. 14, fig. 4,
show to what degree the passageways of the three lower ter-
races are yet discernible. The lower passageways ( 35, 36, 37)
were inside the terraces, and to make the ascent easy they
were not constructed radially up to the top, but were cut into
the terraces in a slowly ascending grade, with one angle
each. It must also be mentioned that none of these pas-
sages run parallel with the section of the terrace in which
they are located; that in the lowest terrace is turned too far
to the outside, those in the two following run within the
general line of the terrace, but with an inward incline. This
may have been the case in a more marked degree with the
fourth terrace, and it seems not merely by accident that the
passage inclined more and more toward the plateau the nearer
the shrine of the divinity upon it was approached. The
width of the passages varies between five feet ten inches and
six feet six inches. In several places, as for instance half way
up to the second terrace and in the centre of the third, the
width was reduced by pillar-shaped projections on one or both
sides to only four feet. To these points P. Cobo probably
referred when he spoke of gates. What he calls ‘resting
places' may have been only landings on a level with the
terraces, or possibly other places specially prepared. A pecu-
liarity of the passageways was that the lowest one in the
centre of the first terrace (35) was cut into the terrace, while
the upper one apparently projected from the middle of the
topmost terrace (near 9); that the different openings were
not placed one above another, and that after reaching the first
terrace the passage made a backward turn to the entrance
into the second passageway (36). With the rest ( 37, 38)
the progress was always in a forward ascending direction, as
far as the second backward turn (to 9). This peculiarity
might suggest that access to the first terrace was a special
privilege, and ascension to terraces two, three and four possi-
bly was denied to many who might have obtained the right
of entrance to the first terrace of the Temple. By the angles
in the passageways through the terraces the line of the actual
ascent was increased to 150 and 200 feet. With an alti-
tude of fifty-four feet to climb, the average grade would have
been as low as 1:3, 6. This figure agrees with the state-
ment of P. Cobo, who saw one part of the original passage-
way. He describes it as being built of quarried stone, without
mortar. “The grades were so low that one could ascend
without fatigue, although it was a long way up.” In one part
of the stairway he counted twenty steps, estimating from these
the entire number to the top as 150, or more, originally.
Therefore with fifty-four feet of altitude, and 150 to 200 feet
of length, each step must have been on an average four inches
high and one foot to one foot four inches wide, a proportion
justifying P. Cobo's remark as to the ease of ascent. Midden-
dorf1 supposes the passageway to have been only a supple-
mentary one, on account of its narrowness, and believes the
main ascent to have been a spiral, leading over the terraces
up to the Temple, as is often seen in Mexico. However, con-
sidering the usual proportions of Peruvian architecture, the
stairway is really unusually wide and luxurious, and the plan
of the Temple, together with the sections of the fronts, show
that a spiral ascent would have been impossible. It is not
likely that under original conditions it would have been easy
to reach the southeast terraces from the northeast.2 The
graded ascent on the northwest front, looking seaward, which
Middendorf claims to have seen from the town, does not in
reality exist. Superstructures barred a convenient passage on
the terraces, and wherever graded steps are actually pre-
served, as between 62, 61 and 63 on the third terrace north-
west, the gradations are three and one-fourth to four feet high,
and so disconnected that they could not possibly have been
used for processions. There is absolutely no trace to be
2
1 The stairs mentioned by Middendorf, II, p. 114, are no longer in existence.
2 Comp. pl. 17, fig. 3.
(76)

discovered of an additional entrance to the Temple, the former
existence of which Middendorf supposed. The architectural
construction forms the strongest argument against the theory
that a spiral ascent, as in the Mexican temples, could have
existed here. However, in consideration of the graduated
slope of the terraces on the southeast front, it is not impossi-
ble that over these, or at least over some of them, a side
passageway led to the top of the Temple, a point which it
would be hard to determine in the present state of the ruins.
1P Roman, in his description of the newer Sun temples of the Inca realm,
probably had in mind the Temple at Pachacamac, mentioned by him together
with that of Cuzco. All the details of his description apply to this structure,
but in addition he mentions two large temple-gates, perhaps rooms on the plateau,
reached by two stairways of thirty steps each. These stairs are not to be found
at all, and the number of steps, each of which would have been two feet high,
excludes the possibility of a double ascent. Roman either took this detail from
another temple, or meant a double stairway in the upper part of the ascent only,
the existence of which might be proved by exploration.
2 Comp. pl. 17, fig. 4.
THE SOUTHWEST FRONT OF THE TEMPLE.
1
The southwest front is apparently the main façade of
the Temple, its elevation exceeding that of all the others by
seventeen feet. The effect of height is increased by the hori-
zontal and vertical divisions of the façade and by ornamental
niches forming a frieze. Another notable feature is its close
connection with the plateau, the rooms of which are situated
toward the southwest side. Judging from the termination of
the terrace on the southeast front near 39, 40 and 41, the
rooms 14, 16, 19, 22 and 24 may be considered as belonging
to the plateau, although situated on the southwest slope of
the structure. For this reason, and on account of their share
in the general effectiveness of the building, they were included
in this discussion, hence the irregular outline of the rear. It
embraces on the south end only one outer grade of the terrace,
situated in front of those of the southeast front (23), and
including farther north 14, 22, 19, 16, 24, 26, 28 and 29.
While all other fronts are divided by horizontal lines only,
the southwest façade was divided by radial walls into four
principal sections, marked 30, 31, 32, but now partly
destroyed. They seem to be closely connected with the
rooms upon the plateau, being simply continuations west-
ward of lines of masonry dividing it in the same direction.
The two northern of the four sections, with courts 26, 28 and
29, are situated mainly upon the natural soil of the hill. A
small room, now ruined ( 27), was built into 26. This with
room 24 belonged to the second section, and in its ruined
state still shows the peculiarity of a transverse division by
four parallel dams from north to south. They stand near
together, but vary in height, so that a cross-cut through all
would show a waved line, decreasing toward the south west.
The two outside dams, 11 and 25, were the walls of the room.
It may be supposed that here three narrow and short passage-
ways ran parallel to each other at different heights, and that
the dividing walls between have crumbled into their present
state, presenting the appearance of dams. The two southern
sections are subdivided into terraces. The third, the most
beautiful of all, is situated above the southwestern ravine or
depression of the hill, exactly in the centre of the front. The
entire elevation of the Temple at this point is about seventy-
seven feet from the base. As the foundations are sunk deeply
between the spurs of the hill, the lowest terrace grade fitted
into the ravine (34), lying outside the limiting line ( 33), and
curving inward along the entire southwest façade, which is
more than six hundred and fifty feet in length. The trans-
verse wall (32) extended over 34, or was intended to do so;
its line may still be traced upon the ground. A graded eleva-
tion of the hill, requiring only the addition of a frontal wall
to form a terrace, seems to have been utilized in constructing
the lowest plateau (34). In pl. 14, fig. 2, the natural soil
of the hill is still visible in the terrace. The third, near 21,
and fourth sections, near 14, of the southwest front are simi-
larly divided by vertical lines. Apart from the almost
identical measurements of height and width of the terraces,
they differ from the other fronts in the very small variation in
the height of the topmost grade ( 16 and 14) and that of the
plateau, and the rapidly increasing distances downward. The
variations in the height of the terraces from the plateau here
measure eight, eighteen and thirty-three feet. In other
respects the third and fourth sections are different from the
first and second. On account of the southern limiting wall, 13
to 22, the lower terrace (23) was continued farther to the
south to 41. There were probably no niches along the front of
grade (14), now entirely ruined, but some may be seen along
the front line of 16, with its continuation northward. P. Cobo,
who saw many of the more important parts of the structure
while still in a fairly good condition, does not mention any
niches at this point. It would have marred the general effect
of the façade if the decoration of the central section had been
continued too far to the south. The superstructures of the
highest terrace (16) are wanting near 14. Decorative niches
were used sparingly, judging by their absence from the
southern side of the wall ( 32). In the third section of the
southwest façade, 16 to 34, the principle of increasing both
height and width toward the top was carried out to a greater
degree by the construction of the high and broad grade ( 34)
in front of it. The measurements of the grades, taken from
the top downward, are as follows: Eight, eighteen, thirty-
three and fifty-eight feet, the terrace grades increasing in
height approximately in the proportion of 1 : 2. The widths
of the grades, taken from the top, were approximately fourteen,
sixteen, fifty-nine and seventy-three feet. The second terrace
grade of the centre ( 21 ) now almost entirely merges into the
third (19; comp. pl. 14, fig. 1), which appears to have been
much disturbed by treasure hunters. It is impossible now
to see how the connection between 19 and 21 was constructed.
It appears that passage 19 must have been faced with stone
coping along the front edge, extending as far as wall 31, pass-
ing it northward in an angle and enclosing the area of 21.2
In three wide grades of three feet each (20), the passage fell
away from the northwest toward 21. The rear wall of 19
is decorated with large niches, visible from the sea at a great
distance. This wall is a little over ten feet high. The niches,
now open on top, stand with their base about one foot above
the passage. They are built of adobe of parallelopipedic
shape and are nine feet high by two feet seven and one-half
.
3
2 Middendorf, Peru, II, p. 115, gives the height of the passage under the
plateau as fifteen feet, which is approximately correct (comp. pl. 17, fig. 1).
3 Markham, Engl. edition of Cieza, chap. 72, note 2, and Middendorf, l. C.,
misunderstood the ornamental purpose of the niches, taking them for buttresses.
The presence of a cornice as base of the niches (comp. pl. 17, fig. 1) proves that
they were not intended for supports. They were certainly roofed over. Ulloa,
who took the Temple for a fort, interpreted (p. 301) the large niches as chambers
for military guards.
1 Comp. pl. 14, fig. 2, and pl. 17, fig. 1.
(77)

inches wide and one foot nine inches deep. The distance
between them equals their width. Architectural ornaments
are lacking. Middendorf counted about fourteen of these
niches; but it may be clearly seen from pl. 14, fig. 1, that
they extended along the entire front from 19; their number
may have been thirty-six. At the northern end, the pillars
dividing the niches are either broken off or the niches are
filled up. Originally they were painted red like the rest of
the Temple, now they appear blackened by fire, similar to the
rear walls of some of the superstructures. The discoloration
must have occurred after the plaster and frescoes had in places
scaled off, as both the injured and uninjured spots are scorched
alike. From this fact it may be assumed that the blackening
was not caused by the smoke of burnt offerings, but simply
by the burning of a wooden roof of some kind, occurring after
the Temple had ceased to be used for the cult.
The topmost terrace grade (16) in its original form was
not less remarkable than the one here described (see pi. 14,
figs. 1 and 3). In the first view it may be seen near 20,
above the large niches of grade 19; in the second, from the
Temple plateau near 17, from the rear facing south. The
plateau has a gentle slope down to the level of the terrace 16,
in the rear near 18 (comp. pl. 14, fig. 3). A thick wall
(17), standing like a slanting earth work, and set with niches
upon the inner side, divides a narrow passageway (18) from
the outer side of terrace 16. It is lower than the plateau and
one foot six inches lower than the piece of wall adjoining it on
the south side (15). A number of mounds of refuse (comp.
fig. 1), similar to those in the third forecourt of the Pacha-
kamaj temple, extend along the full length of the terrace on
the outer side. Digging near wall 17 resulted in the discov-
ery that the cane roofs of the niches (pl. 14, fig. 3) observed
in the rear extended through to a length of ten feet eight
inches, and that this dam (17) originally had been a wall of
that thickness, through which the niches were cut.
The
mounds, judging from similar ones elsewhere, must be the
remains of pillars for the support of a roof. A notice of
P. Cobo may be mentioned in reference to this subject : “In
the rear there was a lobby (18 in the plan ), like a narrow
corridor, running through the entire width of the building, its
walls garnished with niches. The windows were toward the
top. In front of this corridor was a gallery twelve feet wide,
its walls likewise decorated with a row of niches (huecos de
puertas), resembling closets in the wall (alacenas), with a
row of pillars ( danza) in front of them, rudely made (labra-
dos ) of adobe. The roof of this passage and gallery was like
a white floor, and the construction could not be seen when
walking across the court or plaza of the plateau, until, step-
ping upon its roof, it was found to be hollow and different
from the rest of the plateau, the ground of which was solid.”
From this description it appears that P. Cobo saw a roof, cov-
ering gallery (16) and the lobby (18) with its supporting
pillars, in a good state of preservation. The difference in
height of one foot six inches of dam 17 and the piece of
masonry 15 from that of the plateau, may have corresponded
to the thickness of the roof resting upon it, and having the
same level with the plateau. The lobby (18) at that time
also had niches on the eastern side. An entrance to this
lobby, nowhere to be seen now, must have been in the centre
of 17, where the row of niches seems to be interrupted, or at
the northern end, now covered up. There is no entrance at
the southern end, near 15, but only a sort of bay-extension
toward the south west. The front gallery, with its pillars,
must have been a beautiful sight from the distance, while it
afforded a cool place for those under the shelter of its roof.
The rear gallery (18) being covered, received but little light
through the niches, as the space in front of them was itself
obscured by the roof and the pillars. It is, therefore, a point
of interest that Cobo mentions upper windows. Gallery 16
had an entrance from the south in wall 15. This aperture
is partly preserved in the masonry, and it is at the same time
the only door to be seen on the plateau.
1 Comp. Hutchinson, I, p. 160. Concerning the use of burnt offerings in
Peru, comp. Garcilaso, I, Book VI, chap. 22; Betanzos, chap. 11.
THE TEMPLE PLATEAU.
融
​As shown in the preceding discussions, the area of the
plateau was occupied by chambers 18 and 16, as far as the
pillars, and probably also by 24, including the roof. The
space 14 seems to have been uncovered, and might be con-
sidered part of the southwest front of the Temple, being
situated about six and one-half feet below the level of the pla-
teau, and opening in terraces facing southwest. Its floor,
however, is on a level with the adjoining closed chambers (7),
and it may, therefore, be included with the plateau, as was
done by P. Cobo, who saw it in a much better state of preser-
vation. The plateau is shaped like a large irregular trapeze,
the southeast and northwest fronts being only approximately
true, and the two other sides converging in similar angles.
The roof over 18 and 16 has disappeared, probably also that
of 24, which extended the plateau toward the southwest; the
enclosing walls of the other three sides have fallen into ruin,
and may be traced at a few spots only. The approximate
measurements of the plateau were about two hundred and
forty feet on the northwestern side, about two hundred and
thirteen feet on the southeastern, while on the northeast
and southwest the sides measured three hundred and fifty-
four to three hundred and fifty-seven feet. Thus the trapeze
on its smallest side faces southeast. The outlines of the Tem-
ple and of the plateau stand in the closest relation to one
another. The terraces of the southwest and southeast fronts,
similar to the upper ones of the northeast front, run parallel
with the outlines of the plateau above them. The reason for
selecting the southwest front for the most ornamental façade
of the Temple was given, on the one hand, by its situation in
the ravine of the hillside, and on the other by the location of
the open plaza of the plateau, the southwest extension of which
it forms. The position of the decorated part of the southwest
front in the centre of the façade was determined after fixing
the south angle by the terrace of the southeast front, and the
building thus extended westward as far as 66. From that
point the northwest front was not built parallel with the line
of the plateau above it, but sloped toward the east, in order
to follow the shape of the hill as much as possible. While
the plateau originally measured eighty thousand square feet,
or one and four-fifths acres, it has but sixty-eight thousand
square feet at present, or about one and one-half acres," since
the roofs extending it toward the southwest have disappeared.
The entire area was divided into three sections. Bordering
the northwest and southeast sides were buildings (2 to 7)
enclosing the plaza (1), which opened southwest and northeast.
The width of this plaza was about one hundred and sixty-five
1 Hutchinson, I, p. 158, estimates the size of the plateau at ten acres, which
would agree more exactly with the measurement of the base.
(78)

walk may yet be seen. Near the north angle of the plateau
the distance of the buildings from the edge was about
1
A
110%
3'7"
A
< 1972
1773
5'15"
3'5"
7'5"
Fig. 103. Cross-section of same wall.
feet, by a length of two hundred and forty feet. Even
now, after the disappearance of the extending roof, and with
the width the same as formerly, the length is still about two
hundred and twelve feet. The original area was six-sevenths
of an acre; it is one-half of an acre at present. On the north-
eastern edge, near 9, was the entrance of the passageway, now
ruined; there is another depression near 4, which may be
the remains of a second entrance to the plaza. At the point
where presumably the main entrance was laid may be seen
the remains of a chamber-like cavity, five feet deep (18), the
inner walls of which still bear traces of frescoing. This is
the only instance of mural painting in the entire building; it
shows a pattern in yellow, ochre and green upon a red
ground. The room may from this fact be assumed to have
been one of the more important upon the plateau. A deep
cavity (10) in the soil of the plateau suggests an attempted
exploration at this spot.
The buildings upon the plateau are described in detail
by Cobo. According to him, they extended on both sides of
the plaza to a length of one hundred and sixty feet, while
not touching the edge of the plateau ; there was a level circuit
about sixteen feet wide around them on the outside. The dis-
tance between these buildings from the northeast edge of the
plateau was thirty feet, from the southwest edge fifty-six feet.
2
twenty-one and one-third feet toward the northeast, and
twelve and one-third feet toward the northwest. The various
chambers are almost completely ruined, as they were to a
great extent at Cobo's time. He could still see some sec-
tions of the walls in their original height of twenty-four feet.
At present only in one spot, near 2, the wall is still as high
as ten feet. The measurements given by the conscientious
Jesuit father may be considered approximately correct, hence
they were accepted in the reconstruction of the original height
of the buildings in the elevation shown (pl. 17, fig. 3). The
room marked 2 is in the worst state of ruin from unceasing
1
37"
1
1
<
16"
>
1
1772
172
الناس
V
< 15%
16/22 17"
دور
>
12
172
k1²12
3'5"
85
7'
<
13'
2'7"
1722
25 v v
<1'10>
1
k 19"
7'10"
10'2'
121
V
Fig. 102. Elevation of north wall adjoining the two upper terraces of the southwest front of the Sun temple.
In this estimate of the buildings must be included chambers
2 to 7, traces of which are still to be seen. Cobo remarks
that the buildings on both sides of the plaza were of even
length, extending toward the southwest ; this applies to 2 to 3
and 4 to 7. Both groups measure one hundred and sixty-
three feet, or one hundred and sixty feet according to Cobo’s
measurement. The chambers 14 and 24 were not included
by him, being parts of the superstructures, and were meas-
ured with the section from the southwest end of the plateau.
It may also be noticed that the length of both groups of build-
ings represents about the double of their width, about eighty-
one feet (Cobo, seventy-five feet). A few traces of the circuit
explorations, and only a few fragments of walls and cornices
were found during present operations. The entrance to this
chamber could not be determined. No. 3 is in a similarly bad
condition. It has an opening toward the north west, which
may have been an entrance, or only an aperture made by
explorers. No. 4 is in the same state. In No. 5 a small
room, painted red, was found, which still had niches in the
wall; it had been disclosed during some recent exploration.
In room 6 were two small elevations or platforms built of
brick and facing each other, but now decayed into a mere
dam. The cross-section from northeast to southwest and the
frontal elevation are shown in figs. 102 and 103. The small
platforms were on a level with the plaza, or possibly one and
one-half to two and one-half feet below it. The chamber No.
7 has a level floor, but it is enclosed by walls of débris seven
feet high, the remains of the ancient masonry.
Middendorf measured the plaza by paces, and gave the result as being one
hundred and seventy paces long by eighty-five wide (l. C., II, p. 114).
2 A fragment of brick, with a piece of the frescoing upon it, from this room
is now in the collection here.
(79)

DETAILS OF THE EXTERIOR OF THE STRUCTURE.
.
It is surprising that no doorways are now to be seen,
except the one mentioned in the description of the plateau
(p. 79), the entrances to the passages, and a few remains in
chambers or superstructures of the terraces. Rooms, such as
2, 6 and 7 upon the plateau, now appear as having been
closed all around. There is no doubt; however, that once
there were numerous doors. Passageway 18 bas now no visi-
ble entrance, although there must have been one on the side
where the roof covered it from above. We have, moreover,
Cobo's testimony that he saw many doorways along the ter-
races and outer walls of the Temple, while there was but one
gateway leading into it. It seems, therefore, that the learned
observer was referring to the superstructures on the plateau
when he spoke of outer walls. In that case the rooms would
have opened by means of doors on their outer sides toward
the circuit walk along the edge of the plateau ; of these doors
there are now no traces. It is still more difficult to determine
the manner of communication between the various terraces on
the southwest and southeast sides of the Temple. Further
explorations may give a partial solution of this question, but it
is doubtful if it will ever be fully known. The shapes of the
roofs can be determined with some degree of certainty, as they
possibly resembled the one which extended the area of the
plateau over 18 and 16. Gable-shaped elevations of the walls
for pointed roofs are nowhere to be seen, and the walls of
chambers wherever preserved are of even height. The roofs
must, therefore, have been flat in this city, and not pointed as
at Cuzco.
Besides the trapezoid type of niches, only seen on more
recent additions and discussed above (p. 60), there are only
two other types, the square and the vertically elongated. 1
The former appear frequently in the city, the latter is found
nowhere but in this immediate location. It may be seen in
the Temple only on the fine main terrace of the west front,
while the other type is generally used all over the building.
Niches in rooms built up against the terraces, such as 43 and
47 on the southeast front, may be considered in this group;
it is an important architectural characteristic of the Temple,
used almost exclusively for wall decoration upon the plateau,
and upon the third and fourth terraces of the southwest front.
The niches are nearly all about the same size, being one foot
eight inches to ten inches in height by one foot ten inches in
width, and one foot seven inches to eight inches in depth ; on
wall 15 alone are some with a width of one foot four inches
to six inches, the other measurements being the same as the
preceding. This type of niches appears upon the west front
on both sides of wall 15, 2 and on the northeast side of wall
17 ;3 there are also traces on the eastern corner of chambers
2 and 6 of the plateau, proving that the fronts of the cham-
bers 1
upon the platform were decorated in this manner. Two
visitors who saw the Temple in the sixteenth and seventeenth
centuries bear out this observation, P. Cobo stating that “both
groups of chambers on the Temple plateau were decorated
with a row of niches all around them” (“un orden en torno ');
while P. Roman says that “the walls of the four temple cellas
had many niches, in the manner of ancient Spanish churches,
upon their outside walls, such as may be seen
|
be seen on Sancta
Leocadia at Toledo.” However simple and uniform the type
of the niche appears in its proportions, it will be seen that its
application always followed a certain regular scheme. This
is shown by the variety of the spaces between the niches, and
in the various distances measured from below. P. Cobo states
that “the niches surrounding the rows of chambers on the
plateau stood only so far from one another as they measured
in width.” A space of nine feet ten inches, however, may be
assumed for the inner side of room 2 facing the plateau. The
intervening space on the exterior façade must have been six
times less than that upon the interior, where the niches could
not be seen from the outside. The space varies in the same
proportions elsewhere; in the ruined wall 17 it is four feet
two inches to four feet six inches; on wall 15, near 14 and
16, it is two feet five inches, but near passage 18 it is only one
foot nine inches. On the chamber walls upon the plateau
the height of the niches from the floor measured over eight
feet, as may still be verified. This height was in proportion
with that of the walls, and must have greatly added to the
impressiveness of the building, judging by the reconstruc-
tion shown in pl. 17, fig. 3. On wall 15 the space from the
floor in the direction of 16 and 18 was only three feet four
inches, the roof above lowering the general height, while
toward 14, where the wall was as much higher as the thick-
ness of the roof, about one and two-thirds feet, it measured
five feet six inches. The niches continued along terrace 19,
and with it followed a downward line. 4 Here again their
height above the ground is about eight feet. In this way
were obtained the peculiar cross-sections of terraces 16 and
19 (shown in fig. 102). Similar niches probably decorated
wall 31, terminating the terrace on the opposite side, but these
have disappeared together with the wall.
The Temple was painted a uniform deep red, traces of
which color may still be seen on all the terraces, on the inte-
rior of passageways, on the exterior of the chamber walls
upon the plateau, on the inside of room 5, on wall 15 near
the terrace of the large niches 19, etc. The paint had been
applied not only once but several times, as in many spots a
coat of the same color is seen underneath a thin layer of plas-
ter, wherever the outer paint had scaled off. In some spots
ochre yellow is found beneath the red paint, as in the corners
of many of the niches, which may have been painted yellow
at one time. Room 8 on the plateau with its mural painting
proves that occasionally the chambers of the Temple were
decorated in this manner. P. Cobo speaks of “ many rooms,
terrace walls and doors, painted in many colors with animals,"
etc., which could be seen from the outside. The doors at
every turn of the passageway had been much decorated, not
the faintest trace of which is now visible, and it would be
unsafe to credit the statement had it not been made by so
reliable an observer.
From Garcilaso's description there is no doubt that the
Temple at one time was richly ornamented with gold. “Each
chieftain since then undertook to decorate with gold and silver
the Sun temple of his own particular province, in proportion
to his wealth. These temples hence were covered with gold
1
5
9
6
3
4
Comp. pl. 15, fig. 2.
5 See above, p. 79.
6 Calancha makes a similar statement.
7 Of all the earlier statements concerning mural paintings upon the Temple,
perhaps only those of P. Cobo, were founded upon careful observation. He
describes them as “ pintura de varios colores, con mucha laboras curiosas á su
modo, si bien al nuestro toscas, y diversas figuras de animales mal formadas,
como todo lo que estos indios pintaron.” It is not likely that he should have
confused this Temple with the painted Pachakamaj temple. J. J. Tschudi,
Peru, I, p. 291, speaks of "several halls and niches," upon the walls of which
might be observed half-faded, faint “frescoing”in brick-red and yellow ; Rivero
and Tschudi, p. 290, mention "fragments of animals on the walls upon the white-
washed plaster.” Their statements, however, regarding the details of the Tem-
ple are not always reliable.
1 Comp. pl. 15, figs. 1-3.
2 Pl. 15, figs. 1-3, and pl. 14, fig. 2.
3 Pl. 15, fig. 3.
(80)

incrustations, like the principal temple at Cuzco."1 In the
latter the interior of niches and the doorways were covered
with gold and silver, and animal figures made of gold; sacri-
ficial offerings were suspended along the walls, possibly also
on the roof. Cieza writes that Waina Qhapaj, the Inca, had
bestowed 4 costly gifts in gold, silver and precious stones upon
the Sun temple at Pachacamac, of which at present every
trace has vanished, even the fastenings having disappeared.
2
1 Book III, chap. 24.
2 L. C., chaps. 20 and 22.
3 Roman, l. c.
4 Crónica, II, chap. 65.
2
THE ARCHITECTURE OF THE TEMPLE AND ITS BUILDERS.
8
In
2
now.
a
It has long been observed that the Temple is lacking in
all the distinctive marks of Inca architecture, even if it may
be considered at all as one of the Inca buildings. In its
ruined state it presents few characteristic features, but to any
one familiar with the ruins of the ancient convent the differ-
ence becomes at once apparent. There the niches belong to
types peculiar to Inca style, while here those of the third
terrace of the southwest front, high and rectangular, corre-
spond with no architecture of that period so far as it is known
The plain, square niches ? represent a more primitive
type, probably local and native, differing from that of the
convent, while quite common in buildings of the ancient
town. The small trapezoid niches may have originated with
the Incas, but are only to be seen in more recent additions, 3
and cannot be considered when judging of the original style
of architecture.
In the convent at Mamacona the material and quality of
the adobe bricks, their shape and mode of laying, the plaster-
ing of the walls with a colorless clay, all seem to be of superior
order, distinguishing this one building of Cuzco style from
all others in the town. The Temple, in all these features, has
more in common with the town than with the convent, where
the architectural details of ornamentation were quite profuse,
and determined the entire effect of the structure. In the Sun
temple, the expression of color surpassed that of architecture,
while it undeniably shared the general marks of Inca influ-
ences with all the other buildings of the coast. No one, for
instance, will fail to see that the order of ornamental niches of
the third southwest terrace is out of keeping with the stylistic
tendencies of the coast, while entirely agreeing with that of
the Incas, verified on numerous edifices. They favored ele-
vated points for their structures, so that the lofty terraced and
niched buildings commanded the surrounding plains and
valleys. A fine example is the large niched terrace struc-
ture at Ollantaytambo (Squier, Peru, pp. 496 and 497); of a
similar character are the niched terraces of Colcampata at
Cuzco, 4 and the ruins of ancient Inca monuments near Cañete.”
In the plains, the Incas erected handsome buildings, gar-
nished with airy niches, such as those now in ruins at Chin-
chero (pictured by Squier, p. 484), and the rows of beautiful
niches upon several buildings situated close to the shore in
the hacienda Yuumani on Titicaca Island. The type of the
niches at Pachacamac, while not an exact repetition of that
of Cuzco, represents in proportion of height and width that
generally used by the Incas, which is entirely absent from
6
all buildings of the coast. The decoration of walls and enclo-
sures on the plateau with long rows of small niches is only to
be ascribed to Inca style, while elsewhere in the town, wher-
ever timid attempts at rows of niches are found, they may
safely be attributed to the influence of Cuzco culture. 8
no other building was the principle so perfectly carried out as
in the Sun temple, described above (p. 80, and pl. 14, figs. 1
to 3).
Two entirely different styles of architecture were
clearly combined in it, that of the coast and that imported
from Cuzco, the latter however being mainly limited to the
superstructures upon the plateau, and possibly including the
peculiar winding passageways leading up to the terraces, which
were specifically destined for the cult. Even here, foreign
influence could only determine the general character of the
buildings and not the minor details. Therefore the local type
of small niches was used decoratively, while for the high ones
a new style was introduced, forming a transition between the
technique of the coast and that of Cuzco.
If the question be raised why a Temple erected for the
supreme deity of the Inca realm should have been built in a
style different from that of the reigning cult, the answer can
only be, that the Temple had more than one purpose in the
minds of the ruling people. The Incas imposed Sun-wor-
ship upon newly-acquired provinces. Garcilaso quotes one
of the conditions of peace between the Inca and Cuismanku,
to the effect 10 that the Yunkas should worship the Sun and
erect a Temple for him, while apparently the Inca charged
himself with the building of the convent, which to this day
shows pure Cuzco style.
Cuzco style. We know besides that ministrations
in the Sun temples of the provinces were performed by repre-
sentatives of both nationalities, ruling and subjugated. Garci-
laso tells repeatedly that the priests of the Sun temples in the
provinces were members of the reigning chieftain's family of
that locality (of the “Aillu' of the prince), while the chief
priest in every case had to be a relative of the Inca, the
Qhapaj aillu. 11 He (l. c.) states also that all the Sun temples of
the provinces were built after the model of that at Cuzco, and
therefore he did not give special descriptions of them. Judg-
ing, however, from the ruins of some of them, they varied
largely, according to the provinces. The following of the
Cuzco model was probably limited to the arrangement of
essential parts, on account of the similarity of the cult, while
transition to the new worship was made easier for the people
by having the Temple built in the style of the individual
province, forming at: the same time a monument of victory for
the conquerors.
We possess information concerning the workmen by
whom the Temple was built, given by Santacruz, p. 273 :
“At Vilcas Huaman Pachakutij Inca Yupanki met seven
wakas, Aisawillka, Pariaqaqa, Chinchaqocha, Wallallo,
5
6
1 See pl. 21, fig. 2.
2 See pl. 15, figs. 1-3.
3 See above, p. 60.
* Comp. Squier, p. 477, and plan; also view of the terrace, Wiener, p. 312.
5 Middendorf, Peru, II, p. 128.
6 Comp. Squier, pp. 477 and 484.
7 The great preference of the Incas for the ornamentation of walls with long
rows of niches is illustrated by all the Inca monuments of the Islands of Coati
and Titicaca in the Lake of Titicaca ; also views, Squier, pp. 499, 504, 515
[Ollantaytambo] and 527 [Pisac].
8 Passages cut into the terrace are to be found only in one house at Pachaca-
mac, situated in section D, and which distinguishes itself from other buildings by
additional peculiarities. For Ollantaytambo compare the view, Wiener, p. 512.
Compare, besides, many references in Garcilaso's works; also to Relaciones
geograf., I, p. 149.
10 B. IV, chap. 31.
11 B. II, chap. 9; III, chap. 24.
6
(81)

Chukiwakra and two of the Cañares. He made them pris-
oners, and as a punishment for having tried to kill him he
sent them to Cuzco, to build the fort Saksawaman. When
this was finished they constructed another building, of the
character of a summer palace, on the coast of Pachacamac or
of Chincha, which rose directly from the sea.” The waka
Aisawillka was near Lima, 1 Wallallo and Pariaqaqa (of the
same name as the mountain ) was in the province of Yauyos, 2
Chinchaqocha was farther to the north.3 The exact location
of Chukiwakra is not known;4 the two wakas of the Cañares
were probably in a more distant part of the country. Four of
the seven wakas were only twenty to one hundred miles dis-
tant from Pachacamac, and it is possible that the names of
these wakas stand for the provinces where they were wor-
shiped, and which had to share in the construction of the Sun
temple.
1
2
1 Santacruz, p. 275.
2 Descripcion de la provincia de Yauyos, Rel. Geogr., I, pp. 71 to 72.
3 The lake between Cerro de Pasco and Tarma.
4 Mentioned again by Santacruz, p. 255.
INTERPRETATION OF THE TEMPLE.
1
4
It is only possible, at this date, to make conjectures as to
a satisfactory interpretation of the structure. Conclusions
drawn from the building itself may serve as a basis for
speculations, to which may be added information on the gen-
eral construction of Sun temples in the Inca realm, supplied
by P. Roman and Garcilaso in their descriptions of the one
at Cuzco,' the supposed model for all those of the provinces.
A comparison of the interior arrangement of the Sun temple
at Cuzco and that at Pachacamac is unfortunately rendered
impossible by the fact that the ruins at Cuzco 2 are in even
worse condition than those at Pachacamac. It is almost
impossible to verify Garcilaso's description from the scanty
débris. All that can be ascertained is that he is not con-
tradicted by anything in the actual condition of the ruins. It
is still apparent that an open plaza lay in the centre, and was
surrounded by temple cellas as he stated. The Cuzco Sun
temple also had terraced sides. P. Roman asserts that in all
the Inca Sun temples there were four halls, or cloister cells,'
in a quadrangle. This probably refers to the rooms on the
plateau described by P. Cobo. From Garcilaso we learn that
in the Cuzco temple, on the side near the shrine of the sun
image, there was a sort of cloister or court with five pavilions,
dedicated to the moon, the morning star and other planets ;
also to the thunder and to the rainbow, besides one which the
chief priest himself occupied, all these being comprised in the
' House of the Sun.' The Temple was accordingly a centre
for the worship of astral bodies and other phenomena of the
firmament, the sun cult being merely the chief. A compari-
son may be made between the pavilions mentioned by Gar-
cilaso and the cella rooms surrounding the plaza of the
plateau, to which a purpose would be difficult to assign.
Garcilaso states that the altar of the sun faced east; room
No. 2 on the north corner of the plateau facing, both north-
east and northwest, may have been the shrine. This room is
the most devastated of all upon the plateau, an indication in
itself that it was from the first held to be of the greatest
importance. Squier assigns to it the altar of the sun. The
additional remarks by Father Roman, interesting in them-
selves, cannot now be connected with any visible parts of the
Temple. He says: “The Temple had at one angle a room for
worship, in the east, where the sun rises, having a high wall
and a terrace six feet wide above it. In the wall there was
an opening, into which the sun image of gold was set; at
sunrise it was placed upon the altar so as to face the east, and
from morning to noon the real sun and the sun image faced
each other. Then the image was placed upon another altar or
stand (encaje) so that it faced the sun for the rest of the day
the same as in the morning.”
Cobo may be correct in surmising that the open plaza of
the plateau served for sacrifices, feasts and dances. Calancha
also is probably right in assuming that the grades of the large
ornamental terrace of the southwest front were used for sacri-
fices. The platform upon the terrace near 203 may easily
suggest a former use of that nature. We know that at Cuzco
the sites for sacrifices varied according to the festivals. 4
Calancha supposes, and probably correctly, that the human
sacrifices took place on the southwest side in front of the
façade of the large niches, as from there an excellent view
was to be had of the sun setting over the ocean.
There is,
however, no proof for these suppositions. Chamber No. 8,
decorated with mural paintings, may have had some relation
to the sacrifices. A reference to the sacrificial methods prac-
ticed at Huamachuco may serve as a key to the significance
of that room. In the Documentos Ineditos, III, p. 14, is
stated: “For the worship of this false trinity they had large
courts, with a very high wall on one side, having apertures in
which poles were erected for the festivals. A pole wrapped
with straw stood in the centre. The sacrificing priest, clad in
white robes, ascended this pole and killed a guinea pig, offer-
ing the blood to the divinity and eating the flesh. Other
priests sacrificed llamas and sprinkled the poles with the
blood, also eating the flesh, so that nothing was left. At the
foot of the poles the chief priest sacrificed a large quantity of
chicha and zaco, a concoction of cornmeal. Many niches
along the walls were used for the disposal of the remains of
sacrificed llamas.” The practical use here ascribed to the
niches is quite interesting. Roman furnishes an exceedingly
clear description of the purpose served by the lowest terraces
of the Temple, which may have a deeper foundation.
“ Within the two lower terraces,” he says, “were the apart-
ments of all the servants, priests and priestesses connected
with the Temple. There were rooms for the manufacture of
the Temple decorations, cellars and storerooms for the liquors
(vinos), and living and slaughtered animals and birds were
kept here for the sacrifices. There were sacristies where
might be seen hangings of wool and cotton of the finest colors
and texture, and so the entire edifice was excellently arranged
( estava muy claro todo ) and everything easily to be found.”
This account, if founded upon truth, would largely explain
the disposition of the lower terraces of the Temple.
It was pointed out above that a large open space (60),
situated at the northeast corner of the first terrace, seemed to
a
1 B. III, chap. 20.
* Compare the excellent description of the ruins with plan by Squier, Peru,
pp. 439-442.
3 It resembles one of the western part of the plateau of the Pachakamaj
temple (pl. 3).
4 Garcilaso, III, chap. 23.
(82)

1
have been planned for a specific purpose. It was near the
Sun image, which stood just above it, three terraces higher.
It has also been shown that the only turn in the passageway
to the plateau was upon the first terrace, forming a sort of
stopping-place, and that this winding path was laid in a
slightly outward inclination up to the first terrace, while the
upper passages all leaned slightly toward the centre. These
facts might suggest that the first terrace was a sort of vesti-
bule, admission to which was granted to persons limited to
this place alone, and not allowed to ascend higher. This
assumption is supported by the fact that the plaza (near 60)
seems to have served for assemblies and feasts. The wall of
the second terrace, above the plaza, about seven feet from the
ground, is the only instance in the entire structure in which
stone takes the place of adobe. The reason for this may be
that it was to serve as a protection for the wall from being
worn away by the crowds of people passing here continually.
Moreover, the ground of this plaza is strewn with chips of
pottery, among which are numerous those of well-decorated
arybals, used for chicha, and which were broken during feasts
or similar occasions, when large crowds would congregate
here.
The Temple was also used for a certain class of burials.
Hutchinson observed three graves upon the plateau." The
occurrence of single graves at this spot is possible, but from
the description of Hutchinson these were not of an unusual
nature. Middendorf apparently thinks that all the terraces
on the southeast front were used for burial purposes. There
were no signs of such a use upon the third and fourth terraces.
A number of chips of pottery were found in the western part
of the third terrace during an attempt at excavation, but
no human remains. There may have been graves in this
locality, but they must have been destroyed long ago. The
lowest terrace only could be proved to have been a former
cemetery, and it will be described in the following pages.
II, p. 161.
2 Peru, II, p. 14.
(83)

CHAPTER XX.
THE CEMETERY OF THE FIRST SOUTHEAST TERRACE OF THE SUN TEMPLE.
GENERAL REMARKS.
THE
THE first terrace of the southeast front of the temple has
a length of about three hundred and ten feet and a
width of forty-seven to fifty feet. Rising abruptly above
the base of the temple on its southwestern end at two sides,
it was bordered at the eastern end by chambers 45 and 46,
and by side-passageway 57, which from there passes 56 and
so out of the temple. This terrace, which was filled in with
earth, was an ancient burial-ground. In chamber 45 were
found the fragments of a large decorated vessel. The ceme-
tery closely adjoined the chamber on the outside, but was not
continued underneath it. The western half of the burial-
ground had been disturbed in the past. The devastation
caused at that time is still noticeable, even the front wall of
the terrace being completely destroyed, and only a few scat-
tered crania found in the ground, with no other objects. The
eastern half of the terrace, however, was still intact. This
section was now explored thoroughly and forty-six skulls col-
lected from it. It is supposed that about twice that number
of burials had originally been made here. Where the ceme-
tery was undisturbed, it touched in its full width the thick
foundation wall of the front which extends through the ter-
race. In parts of the rear, as well as at the base, it touched the
rocky soil of the hillside around which the temple-structure
is built. A few cross-walls supporting the front wall of the
terrace divided it into sections. The burials had been made
at various periods, and earlier graves had been occasionally
disturbed or destroyed by the construction of later ones. As
far as soil and climate alone were responsible for the preser-
vation of objects, they were found to be in perfect condition.
Numerous fabrics looked like new; the skin of the mummies
had hardened to a leathery condition. Grains and vegetable
substances were found in an excellent state of preservation,
while even the ancient dried-meat preparation, 'charki,'
seemed fresh enough to serve its original purpose.
1
1
Comp. pl. 13, fig. 2; pl. 15, fig. 2.
CUZCO CHARACTER OF THE CEMETERY.
2
The objects found in this cemetery possess an unusual
degree of importance as regards the cultural type represented
by them. The cemeteries along the Peruvian coast as a rule
reflect in different degrees the civilization of the coast, accord-
ing to the period to which each belongs. It is due to the
favorable climatic conditions of this region that its type of
civilization is so much better known in its details than that
of the highland. There the annual rainy season, lasting six
months, had naturally destroyed the contents of the ancient
cemeteries. Only a number of types of pottery, which are
so characteristic of Cuzco culture, resisted this destruction, as
clay withstands dampness more easily than textiles. Until
now a complete and adequately exhaustive picture of Cuzco
culture from a single cemetery was lacking. By slow degrees
enough fragments might have accumulated from later burial-
grounds along the coast to enable us to reconstruct a picture
of Cuzco culture as a whole, but such a collection could never
have had as great a value as the discovery of the entire cul-
ture united in one burial-place. The treasures unearthed
upon the first terrace of the Sun temple offer the advantage
of showing us the pure Cuzco forms of culture in the most
favorably situated cemetery on the coast. The packing of the
mummies is of the simple and yet peculiar class described in
connection with those of the Incas found at the foot of the
Pachakamaj temple. The pottery of this cemetery compares
with the best specimens found near Cuzco in perfection of
style and workmanship. The fabrics represent types so far
unknown in our collections, both in regard to technique and
ornament. We are able to recognize in these pieces the cos-
tume worn by the highland population of the sixteenth cen-
tury, some of which may still be seen there in certain localities
of Bolivia. Their connection with Cuzco style is proved
3
by a number of well-known patterns, and by the close stylistic
resemblance of decorations upon Cuzco pottery. Simplicity
of design, severity of form and a general absence of figure-
patterns are the characteristic features of textiles and pottery
of this period.
It was noticed that the common varieties of food of the
coastland, such as yucca ( Manihot sp.), camote (Convolvulus
Batatas), lucuma (Lucuma obovata), pakai (Inga Feuillei),
so frequently found in the graves of the town, were entirely
absent from those of the Sun temple, while all kinds peculiar
a
1 See above, p. 22 ff., p. 26 ff., p. 35 ff.
2 Rüss, in Verh. d. Berl. Ges. f. Anthrop., 1879, p. 291, declares that the
finds in Ancon do not represent the race of the Incas.
3 As by the Indians in Tora palca, in the upper valley of Pilcomayo between
Potosi and Tupiza, with the Uros on the river of Desajuadero, on various islands
in Lake Titicaca, in some parts of the province of Carangas, etc.
(84)

1
2
a
3
5
9
to the highland were represented, such as chuño (black),
t’unta? (a preparation of white potatoes), uma- or wipi-
kaya? (prepared of Oca, Oxalis tuberosa), apparently also
ulluku“ (so-called yellow potatoes ), quinoa (Chenopodium
4
Quinoa), and the beans of aji (Capsicum sp. or cayenne pep-
per) strung on twine ; 6 of coca and its various components
(Llijta), prepared of vegetable ashes, many pieces were found;
also a quantity of lime.? Beans only occur in the small
round, black variety, not the black and yellow kind fre-
quently seen elsewhere of a large, flat shape ;8 maize was found
only in short ears, as it is grown in the highland.'
Under these circumstances there is no need to hesitate
in regard to the bulldog-like Inca-dog' (Canis Ingæ molos-
soides Nehring), 10 of which two perfect specimens were dis-
covered in this cemetery and nowhere else in the town below;
they should be accepted as true Inca-dogs, brought here by
the Incas from the highland.
It would be a mistake to assume that objects of a differ-
ent type, belonging to the style of the coast, were entirely
absent. The city with its distinctly different culture was too
near for that, but such specimens amount to about two per
cent. less than might be expected in a temple built jointly by
the Incas and the people of the coast, and which, at the same
time, is so well preserved. The purity in which the style of
the Incas is represented here must be explained from specific
It would be impossible to find a more wonderful
confirmation of the interpretation of the building as a Sun
temple, founded upon the combined testimony of all the early
chroniclers, such as Estete, Cieza and P. Roman, than is here
represented in this group of Cuzco culture, found in a burial-
ground of the Sun temple. The classification as Cuzco of
certain types of pottery, such as are shown in pl. 18, figs.
1 to 7, up to now disputed by some explorers; the interpreta-
tion of the temple as an Inca Sun temple, the discovery of a
cemetery in its precincts revealing the same type of culture
represented by the pottery, together with the entire absence
causes.
of objects of earlier types—all these proofs support each
other, each standing for itself as well. A solid basis has thus
been gained for the elucidation of the early history of Peru
according to strictly archæological principles, which this
present work is intended to further. A closer examination of
the objects recovered here will show the following facts :
1. All persons interred in this cemetery were women.
The costume of women is represented in complete sets, not a
single garment being found which might be considered as
belonging to a man-no poncho,11 no cloak (yaqolla), no loin
cloth ; no tweezers used by men for pulling out the hair, but
instead numerous topos, employed by women for fastening
their clothes. In order to establish the details of men's cloth-
ing one would have to look elsewhere, a limited supply being
furnished, for instance, by the Inca graves at the foot of the
Temple of Pachakamaj.12
2. The persons buried here did not die a natural death,
but were victims of strangulation. Proofs of this statement are
to be seen in several specimens of the collection. In some may
be noticed the unnatural thinness of the neck, which seems
drawn into two inches in diameter. It may also be observed
that the scalp in its dried state covers the lower opening of
the skull, which is only possible when the neck has been dis-
located before death. Again, in other cases, several vertebræ
were found in a distorted position beneath the back of the
lower jaw-bone, and the skin below was tightly drawn in.
A summary illustration of these dislocations, as it occurs in
many instances, is shown in the cranium of a woman, pictured
in pl. 18, fig. 13. A plain white cotton cloth encircles the
neck, which is drawn into one and five-eighths inches of thick-
ness and two and three-fourths inches of width. The thick
folded cloth fits the neck closely, so that drying is evidently
not the cause of its reduction in thickness. The cloth is not
tied in front, but closed in the back by a thick hard knot, the
ends about one foot long, hanging down. If this cloth were
a neckerchief, which it could not be on account of its tight-
ness, it would have been closed in front. Neckerchiefs are
never tied in the back, while it is quite plausible that the vic-
tims were strangled from behind. A most convincing proof
that the cloth was used for the purpose of strangulation and
not for neckerchief is, that in the front, immediately in the
centre, is tied another hard knot, which would have closed the
larynx mechanically, and thus increased and quickened the
deadly effect. By the forceful pressure the lower jaw-bone is
dislocated, and the skin has dried hard over it, so that it
cannot be reduced. The manner of strangulation which we
can here trace with the assurance of a coroner was the cause
of death of all the persons buried in this cemetery. The
above-mentioned analogous observations upon numerous other
skulls are a proof of this statement. All were women, in
the same strange town, buried in the same manner, together
with objects of the same type. They were all buried after
strangulation. They may have been maidens, but no children
were found among them ; all were adults; one of the skulls
shows gray hair, nearly white. The hair is mostly parted in
front, and hangs down loosely or is plaited in two braids.
The eyes apparently dried up when closed. Faint traces of
vermilion paint may still be seen on some faces, sometimes
also traces of covering with cotton.
1
1 Compare Hartmann and the writer in Verhandlungen der Berl. Ges. für
Anthrop., 1890, p. 300 ff; v. Tschudi, Beiträge, p. 113. The manner of prepara-
tion is well known and varies but slightly. Those found by the writer at Pacha-
camac are possibly the first specimens from ancient graves. They are not repre-
sented at Ancon. Wittmack (Compt. Rend. du Congr. internat. des Amér., Berl.,
1888, p. 340) correctly concludes from this circumstance that no highlanders were
buried at Ancon.
2 Their name in Aimara, and known elsewhere as chuño blanco, also called
moray by v. Tschudi. The preparation varies slightly, probably according to
the climate only. Besides the ways of preparation told by Hartmann and v.
Tschudi, the writer learned of several more. The Indians not only prepare
chuño blanco, as told by v. Tschudi, but also chuño negro (the real Chuñu) of
lluq'i (bitter potatoes). This preparation is unpalatable to European taste; the
Indians eat it thinking it strength giving. T’unt'a, white chuños, have not yet
been found in graves.
3 Of Ocas three kinds of food preparations are made : qawi, by drying;
umakaya, by the simple process which produces the t'unt'a from potatoes, and
wipikaya (ripe kaya) by the process of the chuño negro. The writer learned
these three ways of preparation on Titicaca Island, finding them quite palatable,
contrary to v. Tschudi, l. c., p. 114, who describes kaya as a black paste cooked
with tschuñied Ocas and very unpleasant to the taste. Tschudi also maintains
that the Ocas preparation will not keep a year, but to our great satisfaction we
found it in ancient graves.
4 A yellow vegetable of a potato character, of glassy appearance, found in
the graves, might be ulluku (Ullucus tuberosus), from its resemblance to the
fresh variety. It seems to have been classified before from graves by Roche-
brune (Wittmack, l. c.). It is much used as food in the highland.
5 According to Wittmack, l. c., p. 333, the leaves of Quinoa were found at
Ancon as packing material. Quinoa seeds were discovered in larger quantities in
the graves of the Sun temple, some of them quite well preserved in their boxes of
cane.
6 Found in quantities. Rochebrune confirms their presence in graves (Witt-
mack, p. 346).
7 Small chuños and coca were found also in the graves of the earliest period.
As proofs of the use of coca on the coast during the Inca period may serve the
above-mentioned gourd-flasks, used for lime.
8 Wittmack discusses beans, l. c., p. 334 ff.
9 On the coast maize grows in short ears only when imperfectly cultivated ;
since all the specimens of food are of highland character, the small size of the
ears may be explained as being of that class.
10 Nehring in Comptes Rendus du Congr. des Amér., Berlin, 1888, p. 312,
with pl. III, fig. 3. One of these specimens is of a blackish color, the other
spotted yellow and brown. A perfect specimen of the shepherd-dog species
(Canis Ingæ pecuarius, l. C., fig. 1) was found in the graves of the Sun temple.
It was light yellow and reddish brown in color.
ll Hamy, in Galérie Américaine du Musée d'Ethnographie du Trocadero, No.
157, pictures a poncho, described as the garment of a vestal found in the Temple
of Pachakamaj, meaning the Sun temple. However, the garment is not that of
a woman (comp. Cieza, chap. 61), and never belonged to a vestal virgin, because
the Sun maidens wore the Cuzco garb described below, and not the costume of
the coastland. At all events the garment never could have been found in the
Sun temple, since it does not correspond to any of the styles represented in the
graves there. It was brought to Paris in 1786, at such an early date when not
much attention was paid to the locality where objects were discovered.
* 12 See above, p. 37.
(85)

1
2
3
What could have been the reason for strangling all these
women ? First, the idea of punishment might suggest itself.
Garcilaso reports that women of the convents in the Inca
realm were buried alive when they had sinned against the
law of chastity. With burying alive, from analogies to be
discussed later, might be understood burial after strangula-
tion In another place? Garcilaso speaks of a governor of
the Inca empire who was hanged and strangled as a punish-
ment for dishonesty in his administration. But the pun-
ishment of strangulation would not necessarily have been
followed by burial in consecrated ground, as was considered
every part of the temple precincts. It is more than probable,
therefore, that the victims may have been strangled as sacri-
fices to the divinity. This conjecture necessitates a closer
review of human sacrifices as practiced among the Incas.
We find that they are explained in literature in a manner
entirely satisfactory, and their prevalence at Pachacamac
explicitly stated. Many peculiarities in the corpses, and
in the objects buried with them, would be inexplicable with-
out the assumption that they were human sacrifices offered by
order of the Inca. The custom of offering human sacrifices
by the Inca was obdurately disputed by Garcilaso, although
he was refuted by v. Tschudi with the enumeration of many
proofs. We gain from his words merely the impression that
the Incas did not abolish human sacrifices, but restricted
them by legislation. This view, however, does not bear
close investigation. The difference between the customs of the
various peoples before their subjugation by the Incas and after
that event was only this : While formerly each of the tribes
or peoples could kill their own human sacrifices, the right
to do this afterwards passed over to the Inca exclusively, and
he himself offered them, or caused them to be offered to the
gods. Garcilaso is, therefore, correct, in a certain sense, in
saying that the Incas forbade human sacrifices, but he leaves
unstated the fact that the right to offer them was thereafter
vested in the Incas exclusively, and that they availed them-
selves of the privilege. The Descripcion de la Provincia de
los Collahuas 6 says distinctly : “Whenever the Inca wished
to make a high sacrifice in order to propitiate an offended
waka, upon advice of the sorcerers, he would order a few
human beings to be sacrificed to those wakas or mountains,
and without his orders it was not permitted to sacrifice human
beings.” The Inca personally officiated at many of these
sacrifices, sometimes offering several persons in succession to
different gods." The human sacrifices for all the places of
worship in the empire were organized and directed from Cuzco
on a large scale. The different modes of sacrifice in the
empire were as follows: 8
1. Strangulation and burial of the dead bodies. This
mode of sacrifice is confirmed by Cieza, Crónica, II, chap. 28;
Molina, p. 54; Ondegardo, P. Ramos,10 etc. The victims
were buried in hallowed ground, in the precincts of the tem-
ple.11 Betanzos, chap. 17, and Santacruz, p. 259, speak of
burying alive. But since they are apparently discussing the
same class of sacrifices as Molina and Cieza, the term might
merely signify that the victims were killed in some way with-
out shedding blood.
2. Strangulation followed by cremation of the bodies ;
confirmed by Molina, p. 58.
3. Tearing the heart out of the body while living, using
the heart and blood for sacrifice, followed by burial of the
bodies ; confirmed by Molina, p. 57.
4. Tearing out the heart, and afterwards burning the
body; mentioned by Montesinos, p. 129.
5. Cutting the throat and using the blood for the sacri-
fice, according to the Relacion de la provincia de Vilcas
Guaman, Relac. Geogr., I, p. 167, Xerez 12 and P. Ramos
chap. xvii, p. 45.
The strangled bodies buried on the first terrace of the
southeast front must have been sacrificed according to the
first of the above enumerated methods. Children of both
sexes were sacrificed, babies in arms 13 as well as older ones,
especially those from seven to ten years of age.14 Child-
sacrifices were the most numerous, as many as two hundred
being occasionally killed at one time, 15 but this was only in
honor of chief wakas of the provinces, 16 or of Cuzco.17 Adult
were sacrificed after campaigns, 18 both men and women ; in the
provinces also at unusual events.19 Outside of that our knowl-
edge in regard to the sacrifice of adults is quite limited. In
Cuzco maidens were set apart for sacrificial purposes ; 20 also
youths and men.21 On the mountain of Guanacaure, one of
the most important sanctuaries of the Inca, were sacrificed
both men and women.22 Besides these, there are other proofs
of human sacrifices at Pachacamac.
Since the sacrifice of adults was not a common occur-
rence in the sanctuaries of the provinces, the fact that it took
place at Pachacamac forms another proof of the great distinc-
tion of this temple among others of the empire. Garcilaso 23
states that up to the time of the conquest the Incas had sacri-
ficed men, women and children at Pachacamac. Calancha
also mentions the frequency of human sacrifices, and Cieza
and Santillan suggest it in two passages.24 Santillan observes :
“Inca Tupaj asked Pachakamaj, the god, what kind of sacri-
fice would be acceptable to him.” Cieza relates : “ The Inca
(Tupaj) prayed to the god, asking in what manner he could
best serve him; whereupon the god sent him the reply that
he should sacrifice to him a great quantity of blood of human
beings and of llamas." 25 These quotations clearly show that
human sacrifices were common at Pachacamac during the Inca
period as well as before. The question where they took place
and in which temple remains unanswered. The statement that
men and children were sacrificed was not verified by our dis-
coveries in the graves. Some light is thrown upon the subject
by a notice of Santillan : “ Occasionally at Cuzco, and at
Pachacamac, they sacrificed adult young women ( doncellas )
and buried them alive, but this happened a few times only.”
This statement corresponds with the sex of the victims in
а.
6
>
5
1 Book IV, chap. 3.
2 Book V, chap. 2.
3 Crónica, II, chap. 28.
4 Beiträge, p. 46 ff.; also Stübel und Uhle, p. 60 a.
5 B. II, chap. 8; III, chap. 24; comp. also I, chap. 11, and II, chap. 10.
6 Rel. Geogr., II, p. 45.
Descripcion de la Provincia de Vilcas Guaman, Rel. Geogr., I, p. 167.
8 V. Tschudi, Beiträge, p. 49, assumed that killing for sacrifice had consisted
merely in strangulation and shedding of the blood. This seems a mixture of sev-
eral modes, ignoring the great variety of sacrificial customs. Molina, p. 55, and
Santacruz, p. 259, discuss separately the methods of human sacrifices, with blood-
shedding and without. Nowhere are both mentioned together. Strangulation,
previous to the shedding of blood, would have represented a motive of humanity,
of which the Incas were certainly not capable. V. Tschudi assumes that the
human sacrifice in Peru was only carried out by bloodshedding, but sacrifices by
strangulation in pursuance of other purposes exclude this one-sided view.
9 Tschudi, l. c., p. 49.
10 Ramos, chap. xvii, p. 45.
11 Cieza, II, chap. 28: “á la redonda del oráculo.”
12 Barcia, Hist. primit., I, p. 190.
13 Molina, p. 58.
14 Santacruz, p. 249 ; Ondegardo in Tschudi, Beitr., p. 49.
15 Tschudi, Beitr., p. 49.
16 Molina, p. 58.
17 Colecc. de Docum. inéd. del Archivo de Ind., XXI, pp. 141, 154, 183, 196,
208; Cieza, II, chap. 27.
18 Molina, p. 59; Descripc. de la prov. de Vilcas Guaman ; Rel. Geogr., I,
p. 149; Montesinos, p. 129 ; Cieza, II, chap. 34.
19 Rel. Geogr., II., p. 45.
20 Ondegardo, p. 166.
21 Betanzos, chap. 17; Cieza, chap. 27.
22 Cieza, II, chap. 28.
23 Book IV, chap. 30.
24 P. 32.
25 Crónica, II, chap. 58.
16
(86)

1
a
our graves. That they were maidens may also be assumed,
since in a number of instances the hair was plaited after the
manner of young girls. The fact that the two places, Cuzco
and Pachacamac, are mentioned together points to a ceremony
in the cult of the Incas, and consequently to the Sun temple ;
the “burying alive” would correspond with the mode of
death of the bodies found in the graves. Santillan, doubt-
less, referred to the class of sacrifices, which was verified by
the exploration, and thus the remains found in the cemetery
of the first terrace of the Temple confirm in a remarkable
way the statement made by him so long previously.
The crania in this burial-place show a remarkable differ-
ence from the usual type found in the graves of the ancient
city. There the enlarged and flattened occiput predominates,
while highland types of deformation, long and pointed heads,
are exceptions. The crania in the burial-place of the Sun
temple present a variety of deformations, of a greater uni-
formity in the eastern section of the terrace, where highland
types are chiefly in evidence. 4 A more careful study by
anthropologists, which is to be desired, will surely serve to
further elaborate the result here communicated. The differ-
ence in type, together with a few characteristics of objects
buried with the victims, are to be explained from the manner
in which human sacrifices were largely regulated in the
empire.
In common with the requirements of other cults, inhabi-
tants of the Inca realm sacrificed every year to their divinities
a sort of tithes. Human sacrifices were included in this
tribute, each tribe furnishing one boy and one girl annually
for the purpose. At a fixed period of each year, the victims
from all the tribes of the realm were assembled together at
Cuzco. There a division was made of them according to the
four principal parts of the empire, and the chief wakas of the
provinces. They were then sent to the points at which they
were to be sacrificed. The management of this part of the
service was admirable. One minister of the cult ( willka
kamayoj) for each of the four principal parts of the empire
at Cuzco, kept a strict account in kipus of the number due to
each waka in his special part of the country, while the chief
priests of the sanctuaries correspondingly kept their own
accounts of the number of victims received by them and duly
sacrificed. 6 The mingling of natives of different provinces
was not limited to children only. Ondegardo states that
every year young girls from all provinces were taken to
Cuzco. These were distributed among the different cults,
that of the Sun, or of Pachamama, or of others, to become
sun-maidens, etc., while a number were set aside for the sacri-
fices, which occurred frequently in each year.”? In this
manner young women could be sent as sacrificial victims to
the prominent sanctuaries of the provinces, such as Pachaca-
mac, and the result must have been that in the different sanc-
tuaries of the country human beings were sacrificed whose
cranial types did not generally correspond with those peculiar
to that particular part of the realm.
The great uniformity in the Cuzco character exhibited by
objects in these graves is remarkable, when we consider the
very different cultural type of the town, the great diversity
of cranial formations and the gathering of individuals from
such a variety of provinces. This uniformity in the first
point is explained beyond a doubt by the passage of the vic-
tims through Cuzco, where they received their uniform reli-
gious training and garb, and a good supply of necessaries for
the long journey to the places of sacrifice.
The vessels found in these graves show a surprising
purity of style, and appear to have been brought directly
from Cuzco, to be used only for this ceremonial purpose, while
many of the other articles were of an inferior and ordinary
class. It is difficult to conjecture what purpose there was in
burying at the same time with ceremonial objects, dogs,
parrots, pieces of textiles in course of weaving, etc. Numer-
ous specimens show signs of prolonged use, even of wear; the
cooking vessels in particular are of this class. Besides girdles,
apparently quite new, there are others which must have been
worn for some time, and are even partly worn out; work-
baskets showing long use, combs with hairs in them, a vessel
and plate of pottery, and a wooden cup, bearing a mark such
a
ts
ㅜ
​a
c
Fig. 104. Engraved marks of ownership on vessels of Cuzco type. From graves
on first terrace of Sun temple; n. s.
as was put upon articles of daily use by the owner (figs. 104,
a, b, c). These things were not of a ceremonial character, but
simply those in every-day use. Accordingly the objects in
the graves in general had no religious significance; the per-
sons had been interred “as they went and stood,” and with
them the articles which served them in their daily life. The
vessels of pure Cuzco style were, no doubt, brought here from
Cuzco by the victims themselves. If the objects in these
graves do not bear a ceremonial character, the highland types
of the articles of food seem so much more strange, in the
midst of a valley abounding in more excellent and refined
edibles. Their appearance here can only be explained by
assuming that the victims came from a distance, moreover
from regions in the mountains, and that they were seized and
sacrificed immediately upon their arrival. Such dispatch in
the sacrifice of the chosen victims is admitted by Molina.8
The presence of specific highland articles of food proves the
exclusively highland extraction of the victims, while their
coming by way of Cuzco is convincingly proved by the char-
acter of the pottery. The purpose of these sacrificial rites
must serve as explanation for the large quantity of pottery
found in some of the graves, which would be still greater if
the older graves had not been constantly disturbed by later
burials. Persons who were strangled and buried in this man-
ner, according to the belief of the Peruvians, went to serve
1 Above, p. 86.
Comp. 0. C. Blake, On the Cranial Characters of the Peruvian Races,
Transact. of the Ethnol. Soc., II, N. S., p. 227, on crania from Pachacamac
in European museums, quoted by Markham, English edition of Cieza, Crónica, I,
chap. 72, note 1.
Virchow observed a constant mingling of various types of cranial defor-
mations in one locality, in a number of crania collected by the writer in North
Argentine, in separate territorial groups (Compt. Rendu du X Congrès des Amér.
Stockholm, 1894, p. 38 ff.). The places of discovery referred to by Virchow were
situated in the mountains, where various groups of people (aillus) lived more
closely together, and buried their dead in common in certain places. This obser-
vation is at variance with a statement by v. Tschudi, Beiträge, p. 139, which must
have been borrowed from an older source, that when a certain cranial shape had
once been adopted by an aillu, no other might be used within that tribe. His
accordance with Virchow in what he calls a well-known custom of the North
American tribes, speaks for the accuracy of his observation in general. But
however correct and important the conclusions may be which Virchow has
drawn from these Argentinian crania, it would hardly be safe to apply them to
the whole of ancient Peru. The populations of the Bolivian Puna and of the
Peruvian coast seem to have shown a much greater uniformity of types of
cranial deformation, which must have extended over a larger territory.
4 In Peru four cranial types were known: Rumpu (round), p’allta (broad),
wanka (long), and saitu uma (pointed head); V. Tschudi, Beitr., p. 139.
5 The qhapaj qocha (rich lake), or qhapaj hucha qocha (rich lake of dues),
of Cieza, II, chap. 29; of Santacruz, p. 259 ; of Molina, p. 54.
6 Molina, 1. c., p. 419.
Ondegardo, p. 166.
3
7
8
8 Molina, p. 58.
(
(87)

the deity in the next world, and for this reason
abundantly provided with all household articles. Although
the outfit of the victims was a very rich and costly one in the
neighborhood of Cuzco, and even seems to have comprised
objects of gold and silver, the absence of such articles in
graves here cannot upset our faith in the general character
of the sacrifices. It must only be considered as a feature of
secondary importance, just as the similar absence in those
graves at the foot of the lower temple, identified as Inca
graves. The course of the preceding investigation served to
destroy the theory that sepulchres of distinguished persons
may be recognized by the silver and gold articles found in
them. There is no such rule, and future investigations will
have to decide the question whether these valuables represent
more than a very inferior means for the identification of tombs
of illustrious dead. It might be inferred that the victims were
sacrificed to the deity as a tithe, that is, in the very solemn
rite of the qhapaj qocha. Although Molina speaks only of
children of about ten years being used for these rites, it is
probable that in the most prominent sanctuaries adults were
sacrificed as well. The sacrifice by strangulation and imme-
diate burial was one of the most solemn modes of the qhapaj
qocha. Santacruz seems even to identify that method of
sacrificing with the offering up of the tithes. But from
Molina’s 4 account it appears that the regular annual sacrifice
of the tithes was always accompanied by shedding of blood,
while on certain other occasions, such as the wedding of the
Inca Wiraqocha, the victims were offered by strangulation
and burial. The idea of paying a tribute with persons who
went to serve the deity rather than with others who continued
to live, has probably always been the underlying one in these
sacrifices. An example of this is shown in the offering, at the
occasion of the marriage of an Inca, of a youth and maiden,
who were buried together after having been strangled. This
pair went to serve the deity as substitutes for the Inca, who
hoped thereby to acquire the right to be spared for a long
life, together with the 'Qoya,' his wife.
I Betanzos, chap. 17; Cieza, II, chap. 28; Santacruz, p. 259.
2 “ Those who had to escort the victims for the Qhapaj qocha’ to the
provinces avoided the regular roads of the empire ; they sought with their
charges a path in a direct line across mountains and valleys, running with
unceasing crying and screaming, in order that the Creator might grant the Inca
victory, health and peace. Natives meeting them upon their pilgrimage would
silently prostrate themselves, not daring to look up while the procession passed,
or they would hide in their houses in deepest awe” (Molina, p. 59).
3 P. 259.
4 Molina, pp. 54, 57, 58.
5 Betanzos, chap. 17.
(88)

CHAPTER XXI.
OBJECTS FOUND IN THE CEMETERY OF THE SACRIFICED WOMEN.
CLOTHING
9910 and
>
ance.
GARMENTS taken from these
YARMENTS taken from these graves are of different
shapes from any others obtained in the city, or else-
where on the coast. Their interpretation is facilitated by the
fact that some articles of dress of the same shapes are still
worn in the highland, and a number of accounts in the six-
teenth century furnish minute descriptions of this garb. It
was worn by all women in the highland and differed from the
costume of the plain, as Zarate states. Accidentally most
descriptions refer to the garments of the province of Ayacu-
cho.2 Cieza, however, observed the same among the women
of the highland of Ecuador, from Quito southward, and terms
it “the costume of Cuzco, adopted with elegance by some of
the women of Ecuador.” 3
Since this same dress is still being
worn by the women of some remote districts of Bolivia, there
is no doubt that it had been spread all over these regions by
the Incas, as well as among the Araucans in Chile, apparently
also among the Chibchas in Cundinamarca (Colombia), and in
every other place in which traces of it are found.
This ancient costume for a woman consisted of the fol-
lowing garments : One large square cloth, ajsu,' worn for a
'
skirt; two girdles,' mamachumpi' and 'chumpi,' by means of
which the cloth was held in place; one shoulder-cloth of the
nature of a cloak, “llijlla'; one forehead-band, wincha’;
sandals, “ usuta’; occasionally also a cloth, ‘ñañaka, folded
and worn upon the head. All these articles are represented
in large numbers among our discoveries, except the ñañaka. 5
The ajsu was worn next to the skin ; 6 beginning at the right
side, it was wrapped about the whole body, covering it from
the shoulders down, and from the hips to the feet, leaving
the arms free. It was held fast on both shoulders by clasps,
and the right breast could easily be bared by unfastening one
of them. It must not be supposed that this garment was
necessarily unattractive. Xerez states : “They trailed their
garments on the ground like the ladies of Castile,
Zarate : “ They girded them tightly in order to make their
waist appear long.” The same may still be seen among the
Indian women of that region. The peculiar girdle consisted
of two parts. Both were made of wool. One girdle, the mama-
chumpi, mother-belt,' short and wide and thickly woven,
was worn next to the ajsu.'1 It was intended to confine the
waist. The other, chumpi, more resembled a band. It was
narrow, fine, of many colors, attractively woven and quite
long. 12 Its length is given as five to six brazas. 13 It was
wrapped many times around the waist, above the mama-
chumpi, and may have served mainly to give a trim appear-
The cloth, llijlla, square in shape, was worn around
the ajsu above the shoulders and fastened over the breast with
a clasp of smaller size than those on the ajsu. It reached
from the head of the rather short Indian woman down to her
knees ; 14 although not of large size in itself, sometimes if hang-
ing straight down it almost touched her feet.15 The pins, topo,
or thopu, were usually sharpened at their broad ends, so as to
be used also as knives. 16 The forehead-band, wincha, was
intended to confine the hair, and was made of wool, about two
inches wide, and of gay colors.17 Men wore winchas sometimes,18
but of a different kind. In the sandals, usuta, there was, as a
rule, no difference between those of men and women.19 The
mall cloth ñañaka (Aimará : iñaka) for the head was worn
by women of a certain rank only. Thus it is mentioned in
the Descripcion de la Provincia de Rucana. According to
Bertonio, “iñaka’ in Aimará means 'palla,' married woman
of Inca blood or another woman of high rank, and ' iñaka-
chasiña'had the meaning of 'acting the great lady.' The
Indian women in Bolivia of to-day, when their husbands
prepare a costly feast, spread their iñaka over the floor during
the dances, and offer coca upon them to the dancers. The
ñañaka or iñaka, therefore, was a characteristic garment for
6
5
7
8
10 Xerez in Barcia, Historiadores primit, I., p. 190.
11 Rel. Geogr., I, pp. 149, 189, 208; Tour du Monde, 1883, p. 182.
12 Rel. Geogr., I, pp. 173, 189, 208; Cieza, l. c.; Xerez, I., p. 80.
13 Aimará : llimphi. Bertonio, Vocabulario Aim.-Cast., p. 204. There are
also belts for men. Modern ones of this kind are pictured in Kultur u. Ind., II,
pl. 11.
I, chap. 8.
2 From Rucana, Sora and Vilcas Huaman, Rel. Geogr., I, pp. 149, 173, 189,
208.
3 Crónica, I, chap. 41.
4 Descripcion de la provincia de Vilcas Guaman, Rel. Geogr., I, p. 149.
5 The types of an Indian woman from Sylvia and one from Pitayó (Colom-
bia) in A. Stübel und W. Reiss, Indianertypen aus Ecuador und Colombia, 1888,
pls. 23 and 24, give a general idea of the ancient costume of the Peruvian high-
land. There are a few important differences to be noted, as in the short and scant
ajsu, which corresponds to the more primitive cultural condition of those Indians.
Descripc. de la Tierra de los Rucanas Antamarcas, Rel. G., I, p. 208 ;
Desc. de Vilcas Guaman, l. c.; Descr. de la Tierra de Atunsora, Rel. G., I, p. 173.
J. J. v. Tschudi, Wörterbuch, p. 9, erroneously terms the ajsu an "upper gar-
ment."
7 Cieza, I, chap. 41.
8 Rel. Geogr., I, pp. 149 and 189; Desc. de la Tierra de San Francisco de
Atunrucana y Laramati, Rel. G., I, p. 189. The ajsu is not a “chemise without
sleeves and collar," as Middendorf describes it, Wörterb., p. 20. The statement
that this garment was worn by men also is erroneous.
9 The hip-cloth of the Araucan women is fastened over the hips by a girdle,
then drawn underneath the latter up to the left shoulder, leaving the right arm
and breast uncovered (Ochsenius, Chile, Land und Leute, 1884, p. 116). This
garment suggests the ajsu of the ancient Peruvian women. But more suggestive
still is the parallel to the shoulder-cloth or kerchief.
8
14 Rel. Geogr., I, pp. 149, 189 ; Xerez, p. 80.
15 Cieza, 1.c. A resemblance to this shoulder-cloth may be seen in the icula,
worn by the Araucanian women, wrapped around neck and shoulders, and
secured at the throat by heavy silver pins; even the name is evidently derived
from llijlla. The women of the Chibchas (Colombia) also wore a shoulder-cloth
fastened with a topo. Restrepo, Los Chibchas antes de la conquista española,
1895, p. 135. The ancient author Piedrahita even gives its name 'liquida.'
6 Zarate, 1. c.
17 Rel. Geogr., I, pp. 189, 208; Cieza, l. c.
18 Montesinos, p. 135.
19 Rel. Geogr., I, pp. 173, 189, 208.
20 Bertonio, Vocabulario Aim.-Cast., p. 175.
9
16
12
(89)

the mistress of the house; though it is not represented among
the graves its absence does not have much weight, as it would
probably not have been appropriate as the dress of women
who went to serve a divinity.
A. TEXTILES.
The material is sometimes wool, sometimes cotton ; the
latter is usually of medium thickness. Thin cotton fabrics,
so numerous in the graves of the town, are absent here. Dark
colors, such as brown and black, predominate, although all
the principal ones are represented. Green is only found in
a single specimen. Compared with the older textiles, gray
appears as a new color-in wool as the natural shade; in cot-
ton as an artificial dye. On page 67 it was stated that gray,
which only appeared during the Inca period in textiles of
the culture of the coast, seemed to have been introduced by
contact with the Cuzco civilization.
In design as well as in technique these fabrics show great
simplicity. Tapestries, so numerous in the graves of the town,
and also appearing on men's ponchos of the Inca graves at
the foot of the temple of Pachakamaj, are entirely absent.
Larger cloths, with narrow woven stripes, show the plain linen
texture. The pattern is generally produced by strips of one
color, introduced among the multi-colored warp threads. In
contrast to the great variety of textiles found in the ordinary
burial-places of the coastland, but few technical processes are
LILOMMODOREDA
ders two of the edges. This stripe is either plain white or
black, or as in some specimens composed of several narrow
stripes of one or more colors—blue, white and some shades of
brown.
2. The wide girdle, 'mamachumpi’: There are twenty-
nine specimens of these; four are pictured (pl. 19, figs. 1 to 4),
representing the principal types. Among them the second
type predominates, more than twenty belonging to this class,
while there are four of the third, and only one each of the
first and fourth. Characteristic peculiarities of these girdles
are : their great width, which varies between three and three-
fourths and six inches, and their thickness (pl. 19, fig. 4 is
an exception), while the length is irregular, generally between
four feet four inches and seven feet three inches. The short
girdles of the third class (pl. 19, fig. 3) are exceptions, being
exceedingly stiff ; they cannot be tied, and have, therefore,
thick cords at the ends, attached by several thinner ones, by
means of which the belt is evenly drawn together.
The girdle (pl. 19, fig. 1) is very thick. Its length is
four feet nine inches without the heavy cords ending with
tassels. It is divided into three sections, two end-pieces
twenty inches long and two and three-fourths inches wide,
and a shorter piece for the centre, three and three-fourths
inches wide and shield-shaped, covering the abdomen. While
the end-pieces generally show the same technical character as
fig. 2, that in the centre was made by weaving the running
warp threads into a linen-like texture, and by the thickness
of the woof, formed by heavy strands, the surface obtained
a rep-like appearance. The warp, as it approached the
centre, was distributed into thirteen strands, while in the
end-pieces it formed only seven. The colors are red and
yellow. On the two end-pieces the figures of llamas are red
on yellow ground, and on the reverse yellow on red.
Pl. 19, fig. 2: A girdle typical of a group of more than
twenty of the same kind. The girdles in this class vary in
length from four feet four inches to six feet two inches, and in
width from three and three-fourths inches to six inches.
They are thick and stiff and of a peculiar technique. The
woof runs double, in two layers, so that the material is woven
double throughout. The two woofs change their position at
the ends, running alternately on the surface, or on the reverse,
and for each woof there are two threads used alternately. In
this manner a thread of the woof of a first course is only
repeated in the reverse in the third, that of the second below
is above in the fourth ; the thread of the first below is above
in the third, and so forth. This makes the edges much
stronger, while by the alternating of the woof the material
has a firm substance of its own. The warp is woven in such
a way that each thread crosses on the surface two threads of
the woof, in an irregular manner, with no resemblance to the
texture either of linen or of twill. The color of these girdles
is always brown or black, and the design generally the same:
narrow stripes with parallel lines in black and white, or in
red, yellow and brown, etc. In some instances the design
forms a zigzag pattern over the entire width.
Pl. 19, fig. 3: Girdle with bird design. This girdle rep-
resents the third type: a short and broad belt, the ends of
which are stiffened by short pieces of cane woven into the
material. The four specimens of this group are twenty-seven
to thirty-three inches long and three and one-fourth to four
and three-fourths inches wide. A zigzag pattern is on two
of them; one is all black with red edges, closely woven with
a constant alternating of the warp threads, giving the appear-
ance of braid. The fourth girdle, here shown, is decorated
with a bird design in the style of the coast; it is thirty inches
long by four and three-fourths inches wide, the reverse show-
Fig. 105. Woman's garment, square ajsu, folded. From
graves of Sun temple : 1o n. s.
a
4
represented. These are : striped patterns, produced by warp
threads of different colors, interwoven with each other ;
double-faced material, woven by a continued interchange of
the warp threads throughout the piece; and cloth with
designs embroidered upon it. The manufacture of textiles
with a double face, and the weaving of entire pieces with
continued interchange of warp threads are so peculiar to the
thick mamachumpi, that they cannot be mistaken for the
fabrics of a similarly produced class of textiles of the
coastland.
1. The large cloths, 'ajsu,' of the costume : Fig. 105
shows one-fourth of one specimen. There are fourteen of
these cloths, six of wool and eight of cotton. They are
nearly square in shape, except one which will be described
later, a white one, with an interwoven striped pattern ; the
rest are all of plain linen texture, generally numbering from
fourteen warp threads to seven woof to the square inch, up to
twenty-four warp to nine woof. The length is four feet three
inches to six feet two inches, width three feet eight inches to
five feet seven inches generally. The length of nine of them
varies only between five feet and five feet ten inches; the
width between four feet ten inches and five feet seven inches.
There is no difference in size between the cotton and the
woolen cloths. The color, except in two instances, is brown
of various shades. A stripe, seven to ten inches wide, bor-
a
(90)

nate.
ing the pattern of the right side in reversed coloring. It is
one of the few examples of Cuzco objects showing influences
of the style of the coast in the treatment of ornament. All
four of these girdles are uniform in color, red and black.
Pl. 19, fig. 4: Thin band-like girdle; a section only is
given here. It is seven feet one inch long by four and one-
eighth inches wide, the woof is single, the design formed of
white zigzag lines and conventionalized birds upon a back-
ground of black and brown stripes. On the reverse the color-
ing is of course in inverted order.
3. The narrow girdles, 'chumpi.' Five of these are
complete; in one specimen the warp is finished, while the
weaving is only partly done ; there are a number of fragments
cloths are gray, one red, one white. The five specimens will
be fully described later. One of these cloths, folded in eight
lengthwise folds, is presented in pl. 19, fig. 7.
5. Forehead-bands, wincha' (pl. 19, fig. 9). They are
of wool, in bright colors, in which red and yellow predomi-
The length is seventeen to nineteen inches, in one
specimen twenty-five and one-half by thirteen-sixteenths to
one and three-eighths inches in width, finished off with plaited
or corded ends, to be used for tying. The usual patterns are
zigzag lines, small diamond patterns and small squares, such
as the above (figs. 106 a, b). One band in red, yellow and
white shows a design in stripes like pl. 7, fig. 14.
6. Pins, “thopu.' The pins mostly resemble the speci-
,
a
NOM N
BUDOVUN
des
JODAS
DES
AN
818183183
8181831res
GCG
2 12
28.6%
மாரு
When
WWW
XXX
XXX
Hon mini NUNTIITINUM UNUTRIMENTATIO TURINT
Barcello
WWUmmunmun DUNNIN
GEO26
0 0 0
muik.
a
b
с
d
e
f
Fig. 106 a-f. Various designs of women's girdles, chumpi, of the Cuzco costume. Graves of Sun temple; n. s.
of others besides. All these specimens are of wool, fine and
like ribbons. They show a variety of rich designs. Figs.
106, a to f, reproduce some of the typical patterns; the colors
are mostly dark, white, brown and black; red or yellow are
rarely seen in them. The measurements of the present five
girdles (pl. 19, figs. 1-5) are as follows: Thirteen feet four
inches, fifteen feet eight inches, eighteen feet seven inches and
nineteen feet eleven inches.
4. Shoulder-cloths, “llijlla.' There are fourteen cotton
and five woolen ones in our collection. They are square,
except one which is thirty-seven to forty-eight inches long by
twenty-eight to thirty-three inches wide; the edges are secured
by fancy stitching in technique similar to figs. 6, 7 in pl. 19.
mens of pl. 19, fig. 10. Some have a small semi-lunar upper
part which is thick and solid. The majority are made of
silver, with a legation of copper, so that they are oxidized.
Some are of copper only.
7. Shoes and sandals. Among the discoveries are sandals
of leather, and others of fibres of Agave sp., both of the
shape pictured (pl. 19, fig. 12); one sandal resembles pl. 19,
fig. 13, and one braid, the binding of a shoe, plaited in a
rhombic pattern in black, red, yellow and white, is shown
(pl. 7, fig. 10). The sandals of the first type (pl. 19, fig. 12)
have a leather sole, or one of braidwork, with four leather
thongs, or with four loops made of Pita (Agave sp.). If of
braid-work, through the loops runs a woolen or cotton string
Fig. 107. Woman's shoulder-cloth, llijlla. Graves of Sun
temple; to n. s.
Fig. 108. Typical stripes on woven ponchos and other fabrics. Cuzco
style. Graves of Sun temple; } n. S.
The texture is plain, with the exception of a figured stripe
in one of them, which will be discussed later. All these
cloths are striped crosswise except two, where the lengthwise
stripes are so arranged that two similar stripes enclose another
of the same width, of different color. As a rule the two out-
side stripes are of one color, while the middle stripe is of
another plain color or of several narrow lines of different
colors (pl. 19, fig. 7); the opposite order is seen occasionally
in cloths which, in various other points, possess peculiarities
of their own. The two outer bands are in most specimens of
a plain brown; white is also found sometimes. The central
stripe is either white, brown or black; white, blue and brown ;
or red, green, yellow and brown. Three of these shoulder-
crossing the heel horizontally. The sandal is easily slipped
on and is fastened by means of the string at the point of the
foot, where it is drawn together in several loops across the toes.
The leather soles consist of one layer of stronger leather,
probably llama, or are made of three or more thicknesses of
the skin of some other animal (* capybara '). The hair upon
the latter is not yet worn off in some specimens; the differ-
ent layers of leather are sewed together with thin leather
strings to form the sole (pl. 19, fig. 12). The sandal (pl. 19,
fig. 13 ) has only two thongs near the heel, while the sole at
the point of the foot extends in two points with slits for eye-
lets. The strings on this sandal are fourfold, and crossing
over the foot they are fastened into the two thongs at the
(91)

a
back and the two eyelets at the point. The sandals of the
former class are six and seven-eighths to eight and one-eighth
inches long, that of pl. 19, fig. 13, measures on the inside
eight and one-half inches, the average sizes for women's feet.
In addition to the above there are :
8. Pouches (pl. 19, fig. 11) in a large number, which
should perhaps be regarded as part of the costume; they were
apparently used mostly to hold coca and the ingredients for
mixing it (^llijta, etc.). Small bags with coca, like pl. 7,
fig. 18, and sticks of the llijta were found in many of the
graves, as before mentioned. The pouches are woven of wool
or cotton, or of cotton with woolen stripes, generally narrow
and vertical. Bird patterns and serge designs are also seen.
The woolen pouches are as a rule of a dark color, the cotton
ones light, many of them striped in red and yellow. The
stripes shown here are the typical kind. Some varieties of
stripe patterns are pictured in fig. 109.
Among the articles of clothing described above in a
general way, there are three requiring special discussion.
1. A white cotton garment, ajsu : The length and width
of this cloth are even (sixty-six to sixty-seven inches), and it
differs from the other ajsus by its pure white tint and its pecu-
liar and superior technique and ornamentation. The cloth
was woven with an unusually fine warp in spite of its large
size. The woof is thicker than the warp, as in many woolen
fabrics ; the threads are exceedingly close, thirty-six to forty-
six of warp by eleven to thirteen of woof to one-half square
a
forty-three inches in length. Compared with the others
described above, it may be said to claim a place of its own,
on account of its peculiar color and fine silky material (alpaca
wool ?), as well as its beautiful technique and ornamentation.
The material is fine and closely woven, twenty-seven to
twenty-nine warp threads by eight to nine woof to the half
inch. It has a stripe like the white llijlla, No. 2; two
bands, sixteen and one-half inches wide, of crimson border
both sides of a white one, nine and one-half inches wide. A
threefold band, two and one-fourth inches wide, repeats in
three symmetrical lines a pattern similar to that of the white
ajsu (fig. 8). The two outer lines show the pattern in yellow
and red, while in the line in the centre it appears in black and
red. The edges are worked in the same colors as in No. 2.
In a former report was pointed out the significance
attached to certain colors in ancient Peru, each having a
special meaning—for instance, in the knotted string, kipu,
mentioned above, p. 72. These significations were not chosen
arbitrarily, nor could their meaning have been limited only
to the knotted strings. On a former occasion, when discuss-
ing a colored cord, which doubtless had been worn as a mark
of distinction (p. 40), and a colored cloth from the grave of
a chief (p. 72), an interpretation of the probable meaning
of the colors was given. The principle of the significance of
colors also extends into the province of woman's garb. The
people, in a certain sense, were divided into classes by colors,
and these colors corresponded with those of the kipu. White
signifies silver, day, doubtless also sun; black, time, in the
sense of night; scarlet or crimson means Inca; brown stands
for the chief of the provinces. These color distinctions were
followed by the Inca in the classification of the women of the
realm. de Santillan,' in reference to this subject, says: “The
Inca commanded that in each valley they should build houses
for the women of the Sun, others for those of the wakas (the
wakas are the provincial deities); also for those which he
reserved for himself.” From the rest, the common class, he
selected the prettiest, and had a home built for them, where
they would work in his service. This would represent a
division of women according to four distinctive ideas : Sun,
Inca, Province, Chieftain, common people. The same author
(p. 18) says: “ The Inca's governor of each province had
to make the selection of women, to be sent by the province as
a tribute to the Inca, both for himself and for the Sun.
Others he distributed among the chiefs, and the rest he gave
to the ‘atunlunas' (' hatunrunas,' adult men), to the plebe-
ians.” This division elsewhere is carried out according to
color—white, red, brown and black women-ʻyuraj ajlla,
wairuru ajlla,'' p'aqo ajlla' and 'yana ajlla. 'P'aqo’some-
times is translated as blonde, yellowish-red, or as red, red-
brown. In the Aimará and Kechua of Bolivia 'pako' means
alpaka ;4 in the Uro it means dog. Probably the above p'aqo
is the same word, and designates the reddish color of these ani-
mals. "Wairuru' is not clear in the dictionaries as a name for
a color, although evidently used for one. Bertónio translates
'huayruru' as 'cosa muy hermosa.'5
hermosa.'5 Middendorf takes it for
a shrub, with red clusters of blossoms and red seeds 6 (Papilio-
nacea). Accordingly the meaning of 'wairuru' may be the
original or a derivative one of red. Santacruz, p. 284, says:
“Inca Tupaj brought with him from his campaigns many
young women, principally the chosen maidens (ajlla) for Tiksi
Qhapaj Wiraqocha Pachayachachij, called Yuraj ajlla,” etc.
;
Fig. 109. Various designs on pouches. Cuzco style. Graves
of Sun temple ; } n. S.
6
inch. The material, therefore, has the appearance of rep,
while it is as close, heavy and stiff as many woolen fabrics, and
yet much finer. The threads are hardly visible on the sur-
face, and it seems almost felt-like to the touch, similar to one
of the ponchos with tapestry borders from the Inca graves at
the foot of the temple of Pachakamaj, although much finer
and of a purer white. On two sides, a red and yellow woolen
stripe is woven in (pl. 19, fig. 8). The pattern in this stripe,
a familiar one of the Inca style, resembles that of the stripes
in pl. 7, fig. 14; it consists of a zigzag line, with thick jointed
ends. In gradations it repeats alternately contrasting colors,
red upon yellow and yellow upon red. The reverse also
shows the pattern in reversed colors. The edges are secured
by the familiar row of stitching (comp. pl. 7, fig. 20).
2. A white cotton shoulder-cloth, llijlla : The length of
this cloth, which is not complete, now measures about fifty-two
by forty-seven inches. It resembles the preceding specimen
in its fine texture; it is of the same pure white and as closely
woven (forty-four to forty-eight warp by sixteen to eighteen
woof to the one-half square inch). A narrow blue band, with
a black one above it, is near one edge. The edges are worked
in the same manner as pl. 19, fig. 7-around the corners in
white, red, yellow and black; on the shorter side simply in
red; on the two long sides in red and yellow.
3. A red woolen shoulder-cloth in fragments: This cloth,
the design of which in half its length is shown in pl. 19,
fig. 6, is thirty-three inches wide and originally measured
1 P. 37.
2 Middendorf, Wörterbuch, p. 692,
3 v. Tschudi, Wörterb., p. 417.
4 Hon. Mossi, Diccion. Quich-Cast., p. 197.
5 Diccion. Aim.-Cast., p. 157.
6 L. c., p. 406.
(92)

7
1
P. 304 : “ The Inca was unfortunate in the war, as he trusted
too much in the promises of the Waka Pachakamaj, and was
not satisfied with so many women wairur (u) ajllas.” P. 253 :
“He had houses built for the four classes, Yuraj ajlla, Wai-
ruru ajila, P’aqo ajlla and Yana ajlla. The first of these he
destined for the Creator, Pachayachachij, the Wairuru ajlla
for himself, the P’aqo ajlla for the chieftains (the apukurakas)
and the Yana ajllas for the common Indians.” With this
classification agrees that which, according to Balboa, the Inca
Tupaj made in the convents, into Mama ajlla, the older ones,
probably the Sun maidens, Guayor ajlla, that is, wairuru
ajlla, about fifteen to twenty years old, who were being edu-
cated for future Inca wives, and the Sayapaya, “who stand
still,” the novices, twelve to fifteen years of age. The
Descripcion de la provincia de Vilcas Guaman? must, how-
ever, be mistaken in mentioning for Vilcas Guaman a con-
vent of Sun maidens, guayran calla, which might stand for
wairuru ajlla, and another next to it for Inca women.
This classification of the women of the empire according
to colors corresponds entirely to the significance of the colors
in the kipus, with the modification that black, besides night,
also means the common class of people. It was carried out also
in their clothing. Women dedicated to the Sun, or as Santa-
cruz says, to the Creator, wore white garments. The accounts
of Anonimus, So valuable in some respects, furnish this
notice : 3 When one of the maidens, after the three years of
her probation, at the age of fifteen, decided to remain in the
convent as ajlla, Sun maiden, she was clothed in white, a gold
circlet, “qoriwincha,' was placed upon her head, costly shoes
upon her feet, and a white shoulder-cloth, “pampakuna,'
wrapped about her.” The officiating priests also wore white
garments. The Willkaumu wore a sleeveless unbelted tunic, .
and over this a cloak, falling to his knees, of white wool, with
red tassels and borders. From this it may be assumed that
the Sun maidens wore white clothing only, and the color being
so characteristic of them, they were named after it as a class.
It may be safe to say that no woman in the realm of the Inca
had the right to wear white unless she belonged to this rank.
Hereby we are led to the conclusion that the fine white ajsu
and the white llijlla, described above, were worn by women of
the order of Sun maidens. Doubtless also other classes of
women wore the colors of their own station. Wherever in
Bolivia the ancient ajsu costume is still worn it is uniformly
black. The women corresponding to the black color, women
of the people, probably wore black in ancient times just as
they do now. Clothes of any other color therefore must theo-
retically designate a different class in the costume of the high-
land. For this reason the brown garments, ajsus and llijllas,
cannot be regarded as having belonged to women of the lower
class, but as having been worn by women of the rank of the
chieftains. The fine red llijlla, described above, must have
been that of a woman belonging to the station which is
marked by this color, which would be a wairuru ajlla, one of
the women selected for the Inca's special service. All these
deductions show forcibly that the women here sacrificed to the
glory of the divinity were not of the lower classes, which
might otherwise be inferred from the simplicity of their garb.
P. Ramos 6 confirms distinctly the statement that wairuru,
yuraj and paqo ajllas were sacrificed to the Sun. This fact
is of so much consequence that small importance need be
attached to the incident of his being acquainted with no other
means of sacrificing than that of throat cutting. For we
know that the manner of sacrificing was more subject to
changes than the sacrifices themselves. In the white ajsu,
the colors white, red and yellow appear; in the red llijlla,
red, yellow and black. It is also worth noting that white,
red, yellow and black are combined in the row of stitching
with which the edges of the red llijlla are secured, and also
those of the white one." The same stitching is repeated on
the striped llijlla, the ground of which is gray (pl. 19, fig. 7).
This might suggest that llijllas of such patterns and colors
had a special significance; a note of Anonimus, p. 181, may
serve here as an explanation : “ The probationers in the con-
vents were clothed in brown or gray llijllas (velo), and their
garments were gray, very quiet.” All these observations help
to strengthen the supposition that the sacrificed women
found in this peculiar manner were not of the lower classes,
and Ondegardo's word may be accepted that “the victims
were taken from among the maidens, either chosen to serve
the god here in temples, as Sun maidens, or to whom fell the
lot, not less coveted in some cases, of having to serve him
in the other world by going to him as sacrificial victims.”
Among the other textiles discovered in these graves are:
Two cotton cloths, sixty and sixty-three inches long and
fifty-seven and one-half and fifty-nine inches wide, striped in
white, blue and brown, similar to the ponchos (Reiss u.
Stübel, pl. 38, fig. 1).
A cotton cloth, fifteen feet eight inches long by eight feet
eight inches wide. This cloth shows wide brown and white
stripes, and was pieced together across the width of four
sections.
Woolen pouches with brown stripes and many small
white cotton bags, similar to pl. 7, fig. 18; some filled with
coca leaves, others with peanuts.
3 66
B. POTTERY.
Pl. 18 shows a typical group of vessels taken from these
graves. The types are mostly familiar ones, often. seen in
collections. Some of them were frequently reproduced, while
others were never shown before. This may be said especially
of the pottery of the style of Cuzco, no complete group of
)
which had previously been in existence. The specimens of
pottery pictured in pl. 18 therefore may serve to emphasize
the statements concerning the Cuzco character of the burial-
ground, and to establish the types specifically characteristic
of Cuzco for the benefit of future investigation. The objects
(figs. 11 and 12 of pl. 18) are exceptions to this class. They
belong to the style of the coast, and were here represented for
chronological reasons, since they were discovered together
with the other specimens in the same graves. The pottery of
this burial-ground shows far more varieties of forms and
ornament than can be indicated here. A summary report on
explorations at Pachacamac, such as the present one, is not a
suitable vehicle for the treatment of all the details of Cuzco
pottery.
Fig. 1. Vessels of the shape of aryballos : ' This type is
1 Rel. Geogr., I, p. 168.
2 It may be erroneous for 'ajlla huasi,'
3 Tres. Rela., p. 184.
4 Gomara, Historia de las Ind. in Barcia, Histor. primit., II, p. 112 ; Zarate
I, chap. 11; Docum, inédit., III, p. 14.
5 Anonimus, Tres. Relac., I, p. 159,
6 Ramos, chap. ix, p. 23.
7 White, red, yellow and black, when combined, possibly signified the
'people'; when variegated, as in the kipus, the color for government' (Uhle,
p. 11). This combination of colors may have been the original one.
8 The illustrations given by Ewbank (U. S. Naval Astron. Exp., App. E, in
A Description of Indian Antiquities, p. 130), showing a number of pieces of Cuzco
pottery, are inadequate; there are even objects belonging to the Spanish colonial
period. Hamy made the same error in presenting a vessel of Spanish times, of no
style in particular, decorated with a cat's head (jaguar, African lion ?), as "a
very archaic” piece from the Peruvian highlands (in Galérie Américaine du Mus.
d'Ethn. du Trocadero, pl. 35).
9 A similar vessel, only larger, from San Sebastian, near Cuzco, is shown
by Hamy (Gal. Amér, No. 110).
(93)

6
:
shown in fig. 1. There are about twenty specimens, from five
and one-eighth to four and one-eighth inches in size. Some
are black. The decoration is in the well-known Cuzco man-
ner-rhombic figures, series of small triangles, a leaf-pattern
as in fig. 7, hooked meanders, various kinds of insects, etc.
All are of strictly Cuzco type. One of the vases is provided
with a tube, to allow of its being carried upon a stick. Some
of the smaller pieces are double-bottles.
Fig. 2. Jug with face: Its general shape is like that of
fig. 1, but the handle is different and the base appears more
flattened ; the face may be compared to those upon arybals
presented before.
Fig. 3. Decorated vessel : Vessels of this very common
Cuzco type were possibly never before reproduced. There
are fifteen pieces of this class in the collection, two to six
inches high, with an upper diameter up to six and one-half
inches. With most of these specimens the height is in a
better proportion to the diameter than in these shown here.
Several of the vessels are blackened by fire.
Figs. 4 and 5. Cup-shaped objects, fig. 4 with cover :
As remarked by Hamy,' a large number of vessels of this
general type are to be seen in the principal museums; but he
is mistaken in asserting that the type is characteristic of the
pottery of the shores of Lake Titicaca; the type is Cuzco.
From the burial-ground under discussion ten of these cups
were recovered, from three and one-half to eleven inches high.
The partial type shown by Hamy is also represented in large
numbers. About one-half of the vessels were used near the
fire, but so that the foot did not come in contact with it. In
one of them were crabs, in the foot of another was a rattle.
Covers of vessels, as of fig. 4, were found here, but in no
other place in the town. They seem, therefore, to be more
typical of the highland than the coast. The shape of covers
shown in the Galérie Américaine is also represented in several
specimens, besides some with bird heads for handles. A better
shape is fig. 9.
5. Plates: One plate decorated on the inside in a pecu-
liar manner is pictured in fig. 6, instead of the entire group,
which presents many varieties, both in shape and decoration.
There are forty plates, twenty of which are in pairs. Several
pairs were spoiled by the breaking of one of them. These
plates are either with handles or without. Some of the
handles terminate in a bird's head, or in a sort of knob. 3
Among the bird's-head handles all the different types of
plates are represented. 4 This shows that no special shape of
plate, with or without handles, is typical of any particular
locality or period. The plates are black or colored. Many
decorations appear upon them, fishes, toads, birds, llamas,
insects, fruits, such as pepper-pods, etc. Geometrical patterns
predominate, similar in design to one of the two plates shown
by Rivero and Tschudi, pl. 36.5 The plate, fig. 6, is decorated
upon the inner side with frogs, cross-shaped figures and orna-
ments resembling arrow-heads, in white, yellow, red and
black. Cross and toads combined also appear in the orna-
mental design in relief on a stone pillar of Inca character in
Hatuncolla, south of Peru, 6 and the repeated association of
the two in one design may have some significance, as the cross
has in itself.7
Fig. 6. Small jugs: The small jug stands for a larger
number of vessels, which though of kindred type, still show
many variations in the details, in size, in the position of the
handle, in the proportions of height and width, and in the
opening of the neck.
Pl. 18, fig. 8. Vessel with cylindrical neck and large
bowl: This type is represented in one specimen only. Simi-
lar types are objects shown in figs. 74 and 86 above (pp.
64, 66 ); also the ring forming the connection between neck
and bowl is here repeated, while the white design in pure
Cuzco manner disposes of the doubt as to whether the shape
of this vessel may be regarded as typically Cuzco. The orna-
mental squares resemble doors, reminding one of pottery
decorated with pictured houses.
Pl. 18, fig. 9. Cover of vessel, with an animal resembling
a cat upon it: The figure decorating this cover proves that
plastic ornaments of a superior class are not entirely lacking
in Cuzco style. A similar animal of the cat species was pic-
tured by Ewbank (p. 138, R); the outstretched tongue is
repeated in l. c., p. 136, K. The cover is grooved around the
edge, and appears to have belonged to a vessel similar to
fig. 4, since it has the same band-like decoration, interrupted
on the rim, which is to be seen on many of these vessels.
Fig. 10. Bowl, representing a bird with spreading
wings : Wings, neck and head are indicated on the rim of
the bowl, which has some resemblance to the bird-shaped
wooden bowls in pl. 18, fig. 15, also recovered at Pachacamac.
A round bowl of pottery was found, the exact size and shape
of the brain-pan of a human skull.
Fig. 11. Flask with small neck and gavel-shaped bow,
representing a fish, called meru, coast style. The fish is
black.
Fig. 12. A fine black double vessel, having the figure of
a monkey upon the neck, coast style. The bodies of both
bowls resemble fruits. The vertical bow connecting the two
parts is broken off.
a
C. VESSELS AND HOUSEHOLD UTENSILS OF Woon.
3
It was shown previously (p. 69) that the richer develop-
ment of wooden utensils, discovered in graves of the Inca
Fig. 110.
Fig. 111.
Figs. 110 and 111. Spade-like utensils of wood. From the graves of
the Sun temple ; } n. S.
Hamy, l. C., No. 113, shows one of this class from Copacabana in Bolivia,
where a strong Inca colony was once established. Both in shape and design,
this object seems identical with one (m) shown by Ewbank. Another reproduc-
tion of this type was given by D'Orbigny, Voyage dans l'Amérique méridion,
Atlas of the historical part, Ser. 4, Antiquités, No. 10, fig. 6.
2 A plate shown by Rivero y Tschudi, Atlas, lam. 36, is of that class.
3 L. C., pl. 28.
4 Only one type of this class is represented here in two plates, l. c., pls. 29
and 36.
5 The transformation of a bird-design may be seen in this pattern. The
painted central stripe of the plate may be the body running up to the handle, the
head, and the sides with zigzag edge the wings. The as yet uninterpreted one-
sided decoration on the arybals may be of similar origin. Birds appear elsewhere
as decorative designs upon arybals (comp. Rivero y Tschudi, l. c.).
period, must be attributed to the influence of Cuzco, which
prevailed on the coast at that time. An abundant supply of
Squier, p. 385.
7 See above, p. 43.
(94)

2
all kinds from the graves of the Sun temple supports this fact.
These are mostly cups and bowls.
Pl. 18, fig. 14. Small cup with engraved ornaments of
blackish wood, which were originally filled in with white
paint. They show distinct outlines of the human form,
besides other designs. On the bottom of the cup appears the
mark of the owner, one of those shown in figs. 104, a-c.
Wooden plates copying the shape of the pottery plates of
Cuzco type were also found. Two are without handles, one
with a handle, the end of which is carved in the shape of a
llama's claw.
A square wooden bowl with straight sides is seven and
one-eighth inches long by five inches wide, and one and one-
half inches high.
A small utensil of wood, composed of four compartments
(pl. 18, fig. 16), resembles in its division a stone vessel from
the neighborhood of Cuzco, pictured by Ewbank (p. 138, K).
Two small bowls, fish-shaped, similar to that of pl. 18,
fig. 15, were also recovered. In one a small sediment of a
red pigment was still visible.
Pl. 18, fig. 17, is a small square ornament, resembling a
bell, with an eyelet on the upper part, and slightly engraved
outside.
of a blue stone, lapis lazuli or sodalith, which has not yet
been tested, but may be the same as that of a bead from a
grave at Ancon, which was proved to be sodalith. A simi-
lar bluish stone is found occasionally in workshop chips (to
one and one-half inches in length) on the site of the ancient
city, which might suggest the possibility that it was intro-
duced by Cuzco civilization. It must, however, be observed
that beads of this kind have been found in graves of the ear-
liest Epigone period of Pachacamac, which might imply rela-
tions with the highland, as is inferred by the general style of
the time, and by the presence of chuños, coca, etc. The
writer collected many pounds of workshop chips on the site
of the ruins of Tiahuanaco, some pieces being of the size of a
hand, a proof of its extensive use in the locality. In the
Museum of Science and Art, Philadelphia, a number of blue
beads of the same stone are to be seen among the objects from
Chulumani, in the Yungas of Bolivia, east from La Paz, but
from whence it originally came has not yet been decided. A
quarry of a blue stone, resembling lapis lazuli, seems to be
near Cuzco ;3 another in the province of Ayopaya, in the
interior of Bolivia. Specimens from both these sources are
still untested.
Among the objects recovered in this cemetery are two
а
a
HIT
CROSSW.
320
Cumpama
3333333
PLACESSES
US
Assistir MMITTEE
CIBULELELLEEHICSERE
bine
נונה (תמונות נעלי
BAIN DOBILI OD
LONDINES
VESTILLIN
Fig. 112. Pendant ornament of shell beads. Cuzco. From
graves of Sun temple ; } n. S.
Tone DOUP
Fig. 113. Necklace. Cuzco. From graves of Sun
temple; n. s.
copper bells (figs. 114 and 115). Both types are numerous
in graves of the highland, for instance in northern Bolivia,
which is a proof of their Cuzco origin.
a
Spoons were unearthed in large numbers. Their length
varies between four and one-half and fifteen and one-fourth
inches. Many of them have carved ornaments upon the
handles, such as llama claws, human figures, etc. One is
decorated with the figure of a pelican, which suggests the
coast (fig. 19). A few utensils of similar shape were also
recovered, such as one spade-like object, thirteen inches long
(fig. 110); another with spatula-shaped lower end, nine and
one-half inches long (fig. 111).
Pl. 18, fig. 18. A bowl made of a calabash: The decora-
tions in Cuzco manner are burnt into the shell.
;
D. ORNAMENTS.
Fig. 114.
Fig. 115.
Copper bells. From graves of Sun temple.
E. COMBS.
The persons buried here being women, a large quantity
of personal ornaments were obtained, consisting largely of
plain necklaces, bracelets and small pendants of shell
(fig. 112). A certain typical significance belongs to one
necklace (fig. 113), made of several rows of shell beads
(wampum ?), alternating with small squares of wood. Some
are bored for the string, carved in animal form and inlaid
with bits of mother-of-pearl. A frog carved in white shell,
with one green and one red eye, is suspended from the chain.
Certain trinkets are embellished with beads or small pendants
A large number of combs rewarded the search of the
explorer. Three types are represented (figs. 116 to 118), two
of which have a handle. They are made of fine sticks of
palm-wood, bound together and held in place by pieces of
cane or, as in some specimens, by a fine plaiting of cotton, the
patterns of which present a great variety. The combs of the
2 Stübel und Uhle, p. 43 b, note 4.
3 To judge from a large block of that stone coming from there, and
exhibited in Chicago in 1893.
1 Similar to one shown by Rivero y Tschudi, pl. 36.
(95)

1
2
Karaya in eastern Brazil correspond in their general form
to the third type. The combs of the earliest period of
Pachacamac are different.
A sling: Of apparently ceremonial character was a sling
taken from one of the graves. It is almost four feet long,
not including the thick tassels of two feet two inches in
length, one of which is brown, the other white. Each of the
strings is three feet and a half long, the centre eight and one-
half inches. The colors are white, red, yellow and black ;
period. The whorls are of a triangular form in the half sec-
tion (figs. 120 and 121), and are made of wood, bone or clay;
the type is that of the clay implements shown by Reiss and
Stübel (pl. 86). They are ornamented by means of stamps, a
,
technical process peculiar to this period. Several small black
vessels recovered in the Sun temple were decorated in a
similar manner.
The sewing needles are made of copper or of a thorn (of
Agave sp.).
a
Fig. 116.
Fig. 117.
Figs. 116-118. Typical combs. Cuzco. Graves of Sun temple ; j n. s.
Fig. 119. Weaving staff. Cuzco type. Graves of the Sun temple ; ſ n. s.
the symbolical and ceremonial significance of which was
pointed out in the foregoing pages.
F. IMPLEMENTS FOR SPINNING, WEAVING, ETC.
One weaving implement, which in its peculiar type was
not found in older graves, is a small staff with knob-like
swellings near the pointed ends (fig. 119). It is probable that
articles of this class may be regarded as belonging to the last
ARCH
AmmainamamanMindhong
Fig. 118.
Work baskets taken from these graves resemble those
from older cemeteries.
Fig. 120.
Fig. 121.
Figs. 120 and 121. Spinning whorls of wood. Cuzco type. Graves
of Sun temple ; fn. s.
1 Ehrenreich, Beitr, zur Voelkerk. Brasiliens, p. 24, fig. 9.
2 Comp. also l. C., figs. 3 to 7 and 17.
(96)

CHAPTER XXII.
THE CONVENT.
a
8 ܕܪ
a
THE
HE ruins of a building situated to the west from the
western angle of the inner city wall, and away from
the city proper, were (p. 11) identified as those of the con-
vent. This structure (pls. 20 and 21 ) is marked as of special
historical significance by its site, and by its distinct architec-
ture, while the ancient name, Mamakona, still clinging to it,
points to its former purpose. No other building at Pachaca-
mac has a similar situation, standing, as it does, with its front
to the edge of the green fields. The architecture cannot be
classed with that of any other building in the town; it is
neither a temple nor a dwelling-house. Small and attractive
in design, it is built in three sections, each of which shows a
distinct character, around three sides of a quadrangle. Niches,
frame-like, and of a trapezoid form, decorate in long rows
nearly all the walls visible from the outside. No other build-
ing in Pachacamac has niches like these, which, however,
conform entirely to the familiar type of Inca architecture.
The building, from these facts, was evidently erected by
the Incas, and as such has been generally acknowledged by
persons in a position to know. It is the only one at
Pachacamac still bearing its ancient name, Mamakona, which,
however, does not belong to the edifice alone, but also to the
adjoining fields belonging to it. “Mamakona' means mothers,
and was especially applied to the older nuns, both at Cuzco
and in the provinces. In this sense it is used by Cieza4 and
3
by Cobo for the women definitely appointed to the service in
the temples at Pachacamac; Cobo even mentions “a house
which the Mamakona possessed,"5 and Pizarro, distinctly
specifies the locality as the site of the ancient convent. It
seems, therefore, impossible to be uncertain on this point. 6
Garcilaso particularizes one of the conditions of the treaty of
peace concluded between the Inca Pachakutij and the prince
Cuismanku of Pachacamac, 7 in fulfillment of which this con-
vent was built : “ As a much finer monument of his greatness,
:
(probably that of Pachakamaj the god], they [the Incas]
would build a house in the valley of Pachacamac for the
chosen maidens, as the greatest distinction that could be be-
stowed upon this district, for in all other provinces these houses
and the temples of the Sun were more admired than every-
thing else, as they were really the finest buildings that existed
at Cuzco.”
Squier’s carefully prepared plan of the convent is too
much reconstructed in some parts and wrong in others. The
main building, c in the plan, was photographed by Midden-
dorf; his representation resembles ours (pl. 3, fig. 2).10
Some other views of this interesting building have been
already published, but they are generally unsatisfactory in the
details; such as one corresponding to pl. 21, fig. 2, by Hutch-
inson and Wiener; one of the south west front by Hutchinson ;
another of a peculiar arch and the niche of a room by Squier,
and one of the north west of the convent by Wiener. 11
In the present work the following views were given as
indispensable for the interpretation of the building: One of
the southwest main wing, taken from the south (pl. 3, fig. 2);
one of the large niches of the northeast front, facing the court
(pl. 21, fig. 2), and one of the central and eastern parts
(pl. 21, fig. 1). These views will show that the convent in
its present condition is only a shadow of its former grandeur.
Some parts could merely be traced along their foundation
lines, such as the rear wall in nearly its entire length ; others
are covered several feet deep with débris, probably due to
some elementary cause, such as inundations, etc.; others again
are completely buried in driftsand. A careful exploration
would, no doubt, uncover many important details of the origi-
nal structure, which might be helpful in the study of the
arrangement of women's convents under Inca rule. The con-
vent is situated at a distance of seventy-six yards from the
southwest side of the inner city wall, and about seven hun-
dred and fifty yards from the Sun temple, 13 near the edge of
the southwest slope of the desert. The fields, skirting it for
about one mile in length and four hundred yards deep, stretch
between the city and the beach, the hills in the northwest
and the river in the southeast, and bear the same name as the
95
1 Similar niches may be seen in the palace of the first Inca at Cuzco (Squier,
p. 451); on the terraces of Colcampata, at Cuzco (1. c., p. 477); at Ollantaitambo
(Wiener, pp. 335 and 491); Vilcas Huaman (l. C., pp. 266 and 267); in all the
buildings of Coati and of the Titicaca Island ; in a so-called Inca palace near
Hervey, in the valley, of Cañete (Middendorf, Peru, II, p. 128; 1. c., p. 139, etc.).
2 P. Cobo, Squier, p. 69; Middendorf, l. c., p. 113.
3 Kechua mama=mother, mamakona=mothers. Compare Garcilaso, B. IV,
chaps. 1, 2, 5 ; also I, chap. 26; Relac. anónima, p. 181, for Cuzco; Garcilaso,
IV, chap. 4; Ondegardo, English edition by Markham, p. 165 ; Santillan, p. 38,
and Relac. anónima, p. 180, for the provinces; also Balboa, chap. 9. Garcilaso
and Balboa write the plural Mamakuna, most of the others Mamakona. Both
forms are grammatically correct (comp. v. Tschudi, Organism. der Khetšua
Sprache, p. 140). The frequent interchange of the vowels o and u is peculiar to
the Kechua and Aimará. Compare also Yanakona.
4 Crónica, II, chap. 58.
5 “Las mamakonas tenian casa aparte junto a el" (el templo). Calancha,
however, erroneously considers Mamacona as a later name of the temple, "el
templo,” that is, the Sun temple : “se llamava del Dios Pachacamac, que despues
fue llamado Mamacona, nombre de las virgines dedicadas al Sol.”
6 Middendorf, Peru, II, pl. 112, is uncertain whether the building could be
accepted as the convent, even with the name Mamakona and the apparent Inca
architecture.
7 Book VI, chap. 3.
8 Hutchinson considered the building as some place of worship ; Rivero
and Tschudi (p. 154) equally erroneously saw in it a Sun temple, subsequently
changed into a convent.
9 Squier, p. 70. The plan of ruins given by Wiener, p. 69, judging from the
niches, must be intended for Mamacona. This is the only ground for identifica-
tion however. In the city plan by Rivero and Tschudi, lam. 55 of Pacliacamac,
the building was entirely omitted.
10 Peru, II, pl. with p. 112.
11 Hutchinson, I, pp. 161 and 163; Wiener, p. 71 ; Squier, p. 71.
12 Pl. 21, fig. 1, shows also a glimpse of the desert stretching back of the
convent, and to the right in the background a part of the ruined western section
of the inner town wall, while to the left, in the distance, parts of the northwest
outer city wall may be distinguished as a dark line against the desert.
13 See plan of city, pl. 2.
13
(97)

1
3
a,
buildings, Mamakona. A pond, thickly grown with rushes,
the abode of many wild ducks, lies near. The green plain
adjoins the city, reaching out to it with one arm six hundred
feet wide by five hundred feet long. Here stands the con-
vent, facing the northwest corner of the city, and at a greater
distance also the Sun temple. The site for the building was
prepared by leveling the slope of the desert, originally
twenty-three feet high at this eastern curve. The plot was
cut into a rectangle, and graded to about eight feet above the
general level of the plain. The convent was built with the
rear against the slope of the desert, while the front overlooked
the plain. Toward the east and west at the foot of the
building, the desert was graded in a straight line toward the
plain, with angles and corners, to a distance of seven hundred
feet. Here and there traces of walls are yet to be seen,
which formerly separated the plain from the desert. All
these changes prove that the fields still bearing the name
Mamakona were connected with the convent, and therefore
represent the Inti chajra, a complex of arable land, set apart
by the Incas for the cult of the Sun in newly conquered
provinces, 2 to provide the women in the convents with all
their sustenance. A grove of date palms near the southern
end of the building serves as a picturesque background to the
ruins. The grove is of a later date, although it was men-
tioned as early as the oldest deeds of the hacienda, which
proves that from early times the Spaniards began to change
the character of the surroundings of the edifice.
The eastern wing, a, of the structure has the appearance of
a terrace, open to southeast and southwest, adjoining in the
north and northwest the more elevated plain. The terrace
lies thirteen feet above the level of the surrounding ground,
and is divided into three small cell-like rooms by partition
walls, two of which still show along the rear walls the traces
of decorative niches, of frame-like shape, set upon the floor,
and flanked by smaller ones; the back wall of the third
room was not decorated. The front walls, facing southeast
and southwest, are so decayed that it is impossible to tell if
they ever were garnished with niches or not. The building in
the rear is adjoined by a wall with an opening like a gate-
way, near d; this gate served as an entrance from the north-
east; its sides are much worn away, a proof of its extensive
The gallery, b, must bave been reached from d. It
forms the rear terrace of the court, the back wall of which is
now partly ruined. The gallery is eleven and one-half feet
high by fourteen feet wide, and two hundred and eight feet
long, while at the back it was originally over eight feet high,
as may still be seen in one place, where the entire former
height is preserved. Niches were used as decoration both in
the front and back. The eastern and western sections were
divided by a wall, near k, which must have had a door
through it; more walls may be seen to the west from this
point, separating small cells, each containing a large niche,
while to the east no such traces can be seen in the present
ruined state of the building. The plan of Squier, showing the
terrace as being cut into cells, is a reconstruction founded upon
suppositions only.5 The niches in the lower front of the ter-
race stood about ten feet eight inches apart and alternated
with smaller ones. Whether the same order prevailed on the
rear wall would depend upon the former presence of dividing
walls between the larger niches, as assumed by Squier, but
in that case there would have been no room for smaller niches
between the large ones.
The western wing, c, was the main part of the building,
C,
one hundred and twenty feet deep and one hundred and five
feet wide, and arranged in a peculiar manner. It consisted of
two platforms facing each other diagonally, e, f, one being
connected with gallery b. An inner court, i, is formed by
both platforms, together with the western part of the gallery,
and a wall, g, which terminates in the northwest end of the
,
east front of f. The platform, e, must have been accessible
from the court, i, and perhaps from the western part of the
gallery, b. It lies on the same level with b, about eighteen
feet above the plain. Its floor space is divided up into rooms,
,
the walls of which are still preserved in their full height, in
one place of twelve and one-half feet. These rooms are
planned so peculiarly that three oblong passage-like rooms,
n, surround four smaller ones, o, in the centre on three sides,
in such a manner that they touch the front wall of the terrace
only with their fourth side, which faces the court, i. The
terrace, f, is about three feet lower than e. The partition
,
walls of rooms upon it have almost entirely disappeared, and
even some of the lines as given by Squier in the plan can
only be suppositional. Both terraces were separated by a pas-
sage, six and one-half feet wide, leading to the southwest end
of the building, and apparently touching the wall g in the
east. room, p, was built into court i. The different
walls of the court and the interior of the room were decorated
with niches, and a gateway may have been cut through the
wall g, separating the court i from the passage m. Several
small rooms were added as a wing at the foot of the terrace, f,
on the southeast front near r, and to the southwest from that
point. The back walls of these formed a side of the passage-
way m. The condition of all these rooms now is so dilapi-
dated that their original appearance cannot be guessed. All
these buildings were decorated with a number of niches, large
and small. Besides terrace a and gallery b, the building con
its northeast, southeast and southwest sides was entirely sur-
rounded with niches. More large ones appear on the wall,
separating platform e from the plain at the back of it; several
additional niches were in the smaller rooms within platform e.
The large niches are numbered from 1 to 28 in the plan. It
will also be seen that a row of about ten handsome niches
faces court h. Of these, those shown in pl. 21, fig. 2, are the
only ones complete and fairly preserved, 6 to 8. The divid-
ing pillars are worn smooth on their lower parts by the pass-
ing of people, similar to the gate-entrance d. Five of the ten
niches are on the wall in the rear, g, which continued the
terrace front to northwest and separated courts h and i. A
door seems to have been between the fourth and fifth niches,
but the wall has disappeared and only a few traces of it are
left.? Two niches, 11 and 12, ornamented the southeast wall
of terrace f, but the walls of the rooms r concealed them from
view; three more 13 to 15, may be distinguished on the south-
west front of f; the traces of two, 16 to 17, on the southwest
wall of the room s, built up against terrace f; five others,
18 to 22, on the south west front of terrace e, while similar
ones, 23 to 25, decorated the inner rooms of the terrace, and
three more, 26 to 28, its posterior wall. Smaller ones were
distributed over the rooms upon the platforms of the terrace.
All these niches, including those of gallery b and of terrace a,
a
е
use.
a
a
1 The distance is given a mile and a half too high by Squier, p. 69.
a
2 Comp. Santillan, p. 37, and others.
3 Garcilaso, IV, chapter 1: “They were provided with all the necessaries,
which they derived from the farm of the Sun, in whose service they stood.”
4 The plan of Squier shows niches also in this third eastern section of the
terrace, which is erroneous.
5 The number of cells as given by Squier must be incorrect, as he does not
take any notice of wall k, and a bay-like curve in the hindmost wall, in pl. 22
near 1, whereby the row of niches is interrupted.
6 Wiener considers these pillars as being breast-shape and as the remains of
caryatides.
7 The front of the niches as given in the plan by Squier is based upon a
reconstruction. The steps indicated by Squier at the eastern end of the building
are in reality not to be distinguished with any degree of certainty ; those indi-
cated on the side of the front with the large niches are not existing.
(98)

с
were of nearly corresponding measurements. The inner depth
was one foot eleven inches to two feet, and the upper width
one foot eleven inches to two feet two inches (two feet nine
inches in No. 12, two feet three inches in No. 23), and a
lower width of two feet two inches to two feet four inches.
The inner height varies ; it is five feet in some, while in others
it may be six feet seven inches to six feet eleven inches, as
in those of the east front of c (pl. 21, fig. 2). An outer frame
projected from this, one foot six inches to one foot nine inches
deep, with an upper width of four feet seven inches to four
feet eleven inches, and a lower of four feet eleven inches to
five feet three inches. The difference of the height of the
interior part of the niche was about one and two-thirds inches.
The smaller niches, generally two feet four inches high and
about one foot four inches deep, tapered toward the top from
one foot to two and one-half inches.
C
The roofs were flat. They covered not only the rooms
upon the terraces and those on the ground-floor, which may
still be traced in their outlines, but other parts of the building
as well, where indications of rooms are wanting. The grooves
in the walls and over the large niches of the northeast front of
c3 and of the eastern wall of the court,4 must have served for
the support of rafters to carry the roofs. In this way the
worn condition of the lower part of the pillars between the
large niches of the northeast front of c is explained, as the
space in front was in constant use, being under a roof.
Similar roofs were probably constructed along the northwest
rear wall of the court, the front wall of gallery b, and along
the southwest front of c. The large number of roofs permits
us to estimate the number of occupants of the buildings as
much higher than could have been done otherwise.
In the surroundings of the convent are found a few
remains of structures which must have had some relation to
the buildings. In the plain, some low mound-like piles, of
round or square shape, may be noticed; these are remains of
former apartments. Another line, resembling a mound or
dam, may be traced, extending like a road from the convent to
the opposite end of the plain, in the direction of the Sun
temple.
The area between the western end of the city wall and
the plain lies at a level of forty feet above the latter ; in this
place ruined foundation walls and a number of small mounds
may be seen, here as elsewhere representing the débris of a
row of pillars, formerly supporting roofs. They must have
had some relation to the convent, although situated outside of
its precincts.
a
b
Fig. 122. Cuzco types of large niches in the convent; bo n. S.
The roofing of all the niches was made of wooden boards,
bound together with ropes, a mode which differed strongly
from that of the niches in the ancient city. The boards are
now nearly all torn out of the walls, and thus the general
effect of the niches has been destroyed. Their former shape
was like fig. 122.
2
a
be seen that the quality of the more recent masonry and plastering are of an
inferior order, compared with the ancient. These modern alterations are not the
only ones in the building. Near 30 a deep recess in a wall has been closed by a
shallow arch ; the niches 23 have been altered by removing the pillars in the
inner recess; similar changes took place in niches 26 to 28. The alterations
prove that the buildings were in use in Spanish times and probably occupied by
Spaniards. This may be connected with the former independent existence of
Mamacona as a hacienda and the early appearance of the palm grove near it.
1 Hutchinson, I, p. 164, supposes the openings to have been used for fas-
tening draperies to cover the niches.
2 Squier pays too much attention to an arch to be seen in these ruins, near
29. It can be proved to date from the Spanish period, as was recognized by Mid-
dendorf (Peru, II, p. 113). It has been put into the place of a former niche,
cutting it out, fitting the recess and closing it over the top with an arch. It may
3 Comp. pl. 21, fig. 2.
4 Comp. fig. 2.
(99)
*

CHAPTER XXIII.
INTERPRETATION OF THE CONVENT.
10
THE
HE description of the ruins is only one part of our task.
The ruins themselves can only give a partial explanation
of the peculiar purpose for which the building was intended,
and as they are among the best preserved of the more promi-
nent among the old convents of the Inca realm, it seems well
worth the attempt to interpret them, as far as may be done
with the help of the records at hand.
The institution of convents for women living in celibacy
and consecrating themselves to the service of the various
deities was quite common in the land of the Incas. They
were to be found in all the larger provinces, and generally
attached to a Sun temples erected by the Incas. From all
appearances, however, there were different classes of convents
in the provinces, and it is necessary to distinguish between
them from the beginning. H. Pizarro distinctly states that he
found in all principal cities two classes of nuns—one conse-
crated to the cult of the Sun, or to that of the dead Inca; the
members of the other destined to be presented to the Inca
when he should pass through the town, besides having the
office of brewing chicha, the Peruvian beer, made of maize,
intended for any warriors who might enter the neighborhood.
With this agrees the note in the description of the province of
Vilcas Huaman, which mentions two large convents in connec-
tion with that locality, one for the Sun maidens and one for
the women set apart for the Inca's service. Santillan even dis-
tinguishes between four different convents; but it is doubtful
whether all four were always to be found side by side in the
same place. Probably three classes might correspond to the
first class as H. Pizarro assigns it, and the fourth would be
the same as the second described by him. The convent of
Pachacamac was of the first class, and the women occupying it
were, like those of the convent Ajllawasi,“ house of the
chosen ones," at Cuzco,4 set apart for the service of the Sun, as
may be seen in the oldest records. The pure Inca style of
the building is also a proof that the women were destined for a
cult of purely Inca origin. Therefore this convent very prob-
ably was an Ajllawasi, as Ondegardo calls the convents of the
provinces, and the name Mamakona merely survived the
former designation, as it may have been the popular name for
the building:
The peculiar design clearly explains itself by the special
purpose for which it was intended. This purpose was of the
same nature in the convent of Cuzco and in the provincial
institutions, except the slight difference that the occupants of
the latter came from provincial families, and possibly also were
presented to the Inca. It is quite plausible that, as Garcilaso
repeatedly tells, the convents in the provinces were all built
after the model of the one at Cuzco. It should therefore be
possible to find certain points of resemblance between the
different convents of the realm and the description given by
him of the Ajllawasi of Cuzco.?
Garcilaso emphasizes the fact that the Sun maidens lived
secluded lives, separated from the world, and so far from the
temple that they could not go there, nor could any man enter
the convent. The distance from Cuzco was only 200 yards.'
However, it is of much more importance that Garcilaso lays
more stress on the distance from the Sun temple than on
its proximity. In the province of Pacajes, in Bolivia, the
Sun maidens also lived outside the general settlements. Jo
On Coati Island, Lake Titicaca, are the ruins of an
important structure, a very fair plan of which is given by
Squier (Peru, pl., p. 360 ) remarking that this building was
known as the palace of the Sun maidens, but that the better
name would be Temple of the Moon. Here again the local
tradition in connection with a monument proves of more value
than later conjectures. The building shows striking points
of resemblance to the convent of Pachacamac in its former
state, and both resemble the convent of Cuzco in many of
their details. “The Sun maidens saw neither men nor
women; they lived in entire seclusion, and were never seen by
anybody,” according to Garcilaso. A passage extended
through the large building, wide enough for two persons to
walk abreast; on both sides were rooms, and at the end lay the
apartments of the Sun maidens, never entered by any but
themselves. The two convents of Coati and of Pachacamac
show the same features, inasmuch as they are both secluded
from the outer world upon three sides, while the front faces the
fields or gardens, with a view over the vast plain or the sea.
A definite principle seems to have been followed by the Incas
in establishing the convents in this particular way, although
in the populous city of Cuzco the feature could not be so
pronounced. Both convents form a rectangle, similar to that
of Cuzco, four times as long as it was wide, 11 according to
Squier. In the ground-plan of the two buildings of Coat-
1 To the service of the Sun, Pachamama (Ondegardo, p. 166), to the
service of the dead Incas (H. Pizarro, Letter, p. 121), to the wakas of the
provinces (Santillan, p. 38).
2 Garcilaso, Book IV, chaps. 1 and 4.
3 Cieza, Crónica, I, chap. 37, Caranqui, Rel. geogr., I, p. 168, Vilcas
Huaman, Cieza, I, chap. 73, for Pachacamac. Comp. also Cuzco (Cieza, V, chap.
27), Coati in Lake Titicaca, etc.
4 Garcilaso, Book IV, chap. 1. The translation in Garcilaso's text as 'house
of stars 'rests on some inexplicable error.
5 Cieza, I, chap. 73 ; II, chap. 58; P. Cobo, H. Pizarro refers to them as the
'women of the devil.' He must therefore have connected them with the temple
of Pachakamaj, the 'mosque.'
6 Garcilaso, Book IV, chap. 4.
7 Book IV, chap. 1 to 2.
8 Garcilaso questions whether the Sun maidens had any duties in the
temple, as is maintained by others (f. i. Relacion anónima, p. 184), perhaps not
without cause.
Compare the plan of the place (Squier, p. 428).
10 Relacion de la provincia de los Pacajes, Rel. geogr., II, p. 59.
11 Peru, p. 445.
9
(100)

and of Pachacamac there may be seen besides long
extended court, open at one side, a reminiscence of the long
and narrow passage extending through the convent building
in Cuzco. The subdivision into cells, in which the women
worked, seems to have been copied at Coati, and must have
been similar at Pachacamac; for this reason the reconstruction
of the gallery of the cells, as seen in the plan of that building
( Squier, Peru), may be considered a very good one. Finally,
may be noticed in both buildings that one of the wings of the
crescent-shaped construction is much larger and wider than
the other. This was probably the most important apartment
given up to the residence of the Sun maidens. The points of
resemblance to the convent of Coati cannot be carried any
further than this. Still the building of Pachacamac becomes
more intelligible with the help of the description of the
convent at Cuzco.
The terrace b, in its gallery-like appearance, was repre-
sented at Cuzco, by the long passage which Garcilaso calls a
gallery. As in Cuzco, the apartments, “where no one might
enter," were reached by the narrow passage-way, so also at
Pachacamac gallery b and the line not to be crossed by any
outsider lay near the gate in the wall k. The rooms a proba-
bly were the most secluded ones, in which the Sun maidens
lived, while in the long front of the building, and near the
door in wall g, was the main entrance, entered only by women
of the highest rank, and through which new Sun maidens
passed when about to be received into the circle of the chosen
virgins. The rooms in the gallery were occupied by women
working in the service of the Sun maidens, while the court
and gallery served for the numerous novices who usually
lived there. The Relacion Anónima states, in regard to the
convent of Cuzco: “ The novices lived apart from the real
Sun maidens, and although there were no keys or doors, all
doorways being closed with hangings, their obedience and
discipline were so great that no one ever walked over to the
novices from the Sun maidens' apartments, nor from those
of the novices to their superiors’, without the permission of
the abbess or the chief directress.”
There was a well regulated gate-service in these convents
attended to by women. Porters guarded the entrance at the
door near d and kept up communication with the outer world.
They allowed the visitor to go as far as the second door,
probably near k, or they received the articles brought there
and carried them so far, but they were forbidden under
penalty of death to cross that threshold. It is also likely
that the convent was guarded from the outside, and it is
possible that the hall, of which the traces are to be seen on the
higher eastern edge of the plain near t, had some such relation
to the convent. To judge from the small size of the building,
the number of its occupants might be underestimated. But it
may be assumed that the women lived very much crowded
together in the apartments. The deeply worn sides of the
gateway d, and of the pillars between the niches 5 to 9, may
be considered as proofs of this, as well as the roofing over
parts of the courts. Garcilaso may have exaggerated when he
gave the number of the Sun maidens in the convent at Cuzco
as 1500, with 500 women servants. It is probable that all
over the country the convents were crowded with women.
We always hear of high estimates given for the numbers in
each of them. Two hundred is mentioned for the convent in
Caranqui, Ecuador; five hundred each for the two convents in
Vilcas Huaman, and Ondegardo states that in each Ajllawasi
of a province little girls from eight to nine years of age were
chosen to be trained for novices and lived together with one
hundred Mamakonas.?
The convent in Cuzco was 800 feet long by 200 feet
wide, and occupied about four acres of ground. The number
of inmates being estimated at 2000 persons, and not allowing
any space for walls, etc., this would give eighty square feet to
each person. If it is assumed that the convent at Pachacamac,
which occupied only about one acre of ground, housed 200
persons, the above figure does not seem too high.
3 Garcilaso, Book IV, chap. 1.
L. c. chap. 2.
5 In connection with the convents the early chroniclers always mention the
large number of guards and doorkeepers; comp. Cieza, II, chap. 27; Ondegardo,
p. 166 : Santillan, p. 38; Rel. geogr. I, p. 168 ; H Pizarro, Letter, p. 121 ; Doc.
inéditos, III, p. 41; Relacion Anónima, p. 180 ; Gomara, Hist. de las Ind in
Barcia, Hist. prim., II, p. 113. The three latter speak of eunuchs, whose exist-
ence in Peru is disputed by J. J. v. Tschudi, Beitr., p. 94.
6 See the plan of the city, pl. 2.
7 The Spanish text, which is full of errors, gives here ciertas Mamakonas,'
instead of ciento. The text of the English translation by Markham (p. 165) gives
more correctly hundred '.
8 Squier, p. 445.
1 Ondegardo, p. 165; Garcilaso, Book IV, chap. 4; Relacion Anónima,
p. 180, for the provinces.
2 P. 182.
(101)

CHAPTER XXIV.
TECHNICAL POINTS OF BUILDINGS.
THREE different kinds of material have been used in the
buildings of Pachacamac: sun-dried adobe bricks, stone
and plain adobe clay used as a sort of plaster upon walls and
simply dried in place, 'tapia.' The stone was quarried in
the neighborhood of the ancient city. The yellow clay for
the adobe and tapia is found near the river, which washes
it down from the mountains. In some terraces of the temple
of Pachakamaj a blackish kind of adobe has been used for
more important purposes, a kind not found elsewhere in
Pachacamac, and seemingly baked of the mud from the bottoms
of canals or ponds. Residents of the neighborhood who dis-
tinguish the adobe bricks by their quality assured us that
those used for the buildings of the town are of the best
quality. Difference in quality does not consist in the mate-
rial used but in the method of preparation—whether the adobe
bricks have been baked immediately after taking the clay
from its bed, or whether it has been allowed to "rot" in water
for one week or longer. 1
The buildings of Pachacamac are nearly all constructed
of brick. Nevertheless Bandelier's statement is correct, to the
effect that there is hardly one large pile of ruins to be found
along the coast entirely without stone. The foundations of
buildings especially are built of stone. As above stated, an
ancient wall in the interior of the temple of Pachakamaj, 2
disclosed during the excavations, and certain terraces and
structures along the hill were constructed entirely of stone ;3
this is also true of the high substructures of the original
enclosure of the Pachakamaj temple. In the Sun temple
stone has been used in the construction of the foundations,
for sections of the interior masonry, previous to a facing with
adobe, in the construction of the northern angle of the second
terrace, also for the terrace in front of the large niches of
the southwest façade, 4 and in some other places. The founda-
tions of houses and also the side walls of many grave cells of
the earlier times (in Cemetery I) were built of stone. Never-
theless, the use of stone as building material is so secondary
in Pachacamac to that of adobe that in a general survey of
all the buildings in the town it is not at all noticed.
Tapia has been still more sparingly used in buildings.
We only noticed it in five buildings of the town, where it
occurred in a secondary way, such as in the platform f in
Mamakona, in some of the houses, in the ruins to the north
from the inner city wall and the road skirting it. In these
cases the tapia had been used in some partition walls and in
some of the terrace walls.
The almost exclusive use of adobe in buildings at Pacha-
mac is peculiar, for the reason that in all other ruins of the
neighborhood tapia constructions are in the majority or
exclude all other material : thus in the ruins of the Lima
valley in Cajamarquilla ; also the other ruins in the valley
of Pachacamac,“ in the Rinconada, are built entirely of
tapia ; likewise those on the rock and at the foot of La
Centinela, as well as numerous field enclosures in the lower
valley.? Adobe constructions, however, are not exclusive to
Pachacamac. In the valley of Lima, near Maranga, numer-
ous mounds are to be seen, of the same nature as that situated
above the temple of Pachakamaj, m, in the plan. In the
ruins of the ancient Surco, near Chorillos, tapia is more exten-
sively used, while adobe is also represented. Judging from a
view, the buildings in Paramonga also seem to have consisted
of adobe. Farther up the coast to the north, adobe construc-
tions are to be found, for instance in the valley of Santa.
Middendorf raises the question as to whether all of the
city of Pachacamac does not date from the time of the Incas,
since it is built of adobe ;8 but it is much more likely that
adobe represents the earlier form of building. Very ancient
portions of the temple of Pachakamaj, dating before the Inca
period, are built of it. The mound at Pachacamac and those
of a similar nature in the valley of Lima impress one with
their great antiquity. The occasional use of tapia at Pacha-
camac is found to be in buildings of the Inca times, also in
Mamacona and the walls in some of the more recent buildings,
whereby an increased use of tapia is proved during the period
of the Incas, while in all buildings of apparently greater age
tapia is not to be found. It would seem correct, as seen from
former discussions, to classify the other ruins in the lower
valley, which are built of tapia, as belonging to the last Inca
period. If the adobe construction was the older mode of
building in the lower valley of Pachacamac, it would by no
means prove that all those buildings must have a higher
antiquity than those built of tapia in the valleys of Pacha-
camac and Lima. An originally older form of construction
may be retained longer in one place than in another. Several
kinds of building material rarely exclude each other in the
same periods, as several may be used at the same time, and
this use mostly depends on local conditions. Therefore it is
possible that at Pachacamac adobe was still used for building
long after other places had changed to the use of tapia.
Although the adobe bricks at Pachacamac have about the
4
a
5
1 Mixing grass into clay (“tchu ') is quite general in the highland ; we
never found grass mixed with the clay of the adobes at Pachacamac (Bandelier,
Amer. Anthrop., l. c., p. 305).
2 Above, p. 20, fig. 4.
3 Above, p. 51.
4 Comp. pl. 15, fig. 1.
5 Prescott, I, p. 291, erroneously speaks of the temple as built entirely of
stone; Hutchinson, I, p. 173, states that the main portion of the material in it
is stone.
6 However, the ancient walls of the hacienda Las Palmas, 1-13 miles from
Pachacamac towards northeast, consist also of adobes.
7 Wiener, p. 494.
8 Peru, II, p. 121.
(102)

a
same proportions as baked ones, they are generally larger in
size and their measurements are of endless variety. The
smallest bricks are those of the mound, about eight inches
long, seven inches wide and four inches thick. The largest
seen by us at Pachacamac are two feet long, one foot two
inches wide and about six inches thick.1 The size varies
according to the purpose, if used as fillings of terraces or plat-
forms (mound, terrace in the rear of the Pachakamaj temple,
Sun temple), or for ordinary masonry and front walls of ter-
races, or for the floors of terraces and plateaus. The former
class are the smaller everywhere, the latter the larger. In the
Sun temple, for the filling of terraces bricks of the smallest
size were used, little larger than those of the mound; a larger
size served for the fronts of walls, but the largest of all for
the flooring. It can not, therefore, be said that buildings of a
certain given period at Pachacamac, such as the Sun temple,
consisted entirely of one even size of bricks, nor may the bricks
of one building, even in their peculiar use, be compared to
their use in another; indeed the size of bricks varies within
the same building for the same purposes. The bricks in the
flooring of one terrace were found to measure one foot eight
inches and one foot two inches in length. Five different sizes
were observed among the bricks in one of the younger ter-
races of the temple of Pachakamaj, the length of which varied
from one foot one inch to one foot six inches, the width
from eight inches to one foot three inches, the thickness from
three to six inches. From this may be seen the difficulty in
fixing the age of a building by the size of the bricks. No
safe distinction may be drawn between the sizes of adobes of
apparently pre-Inca buildings and of those dating from Inca
times. It may be said, however, in accordance with the view
expressed by Middendorf, that the bricks of the most recent
period seem as a whole somewhat larger than those of earlier
times. Future explorers will have to decide how far it
is possible under these obvious difficulties to draw conclusions
concerning the relative ages of the various buildings. The
bricks were laid in mortar composed of the same kind of
clay as the bricks themselves. A glance at the plates will
show that the walls of buildings incline inward, which gives
increased strength. In a wall twenty feet high the inclina-
tion
may be between four and eight inches. Partition walls
did not taper towards the top, while this is frequent with
tapia walls, such as may be seen in the ruins of the Lima
valley and in the walls built of stone in the ruins of the upper
part of Pachacamac valley. Terrace walls more than twenty
feet high are, as a rule, strengthened by a lower retaining
wall. This class of walls is found up to twenty-three feet
high. The highest point of elevation of terraces is to be seen
in the houses of the town, forty-three feet above the city level.
Partition walls occur in all degrees of thickness, ranging from
one foot six inches to twenty feet (the northwestern outer
city wall), as the city plan will generally show. The walls
were formerly coated with a smooth layer of plaster, still
adhering in some spots on the walls of the convent. 3 In the
town this coat of plaster is preserved upon walls inside of
ni
terraces, which were filled in with sand and overlaid with
flagstone, while from exposed walls nearly all traces of plas-
tering have disappeared. Mural painting was uncommon and
is only found upon sacred edifices. The peculiar mode of
decoration upon walls with large grooved ornaments, seen
on the buildings of the Lima valley, cannot be proved to
have been in use at Pachacamac, while it is not impossible
that it was once known here, but may have disappeared in
the course of the centuries,
1 Pl. 16, fig. 1. They were set on their long and narrow sides, correspond-
ing to the different purposes for which they were used.
2 Middendorf, Ollanta, 1890, p. 21.
2
3 Pl. 21, fig. 2
(103)

ERRATA.
On p. 63, line 2 from below; on p. 64, line 13, and on p.
65, lines 20 and 26 from above, read pl. 13 instead of pl. 12.

Zima
OH
IND
AMA
W
20
=
of the
WA
TOMI
MAP OF THE RUINS OF
Marthern Street
Unfinished Terraces
Old City Wall
Wakiket witte
ya
HVATA
CARITHI
ity
TH
ちなみに
​Houd to the
HEZ
درارارار
allele
alget
wa
Min
رانا
411F/HA
19TO
wit
Convent Mamacona
لرد اند برای رانا اور
Old City Wall
anunu
PACHACAMAC
الحال
ilayah
We won
.
litantulliya
VM
CRANE
AR
MUS
PASAP
C
F
Why
AVA
MA
MA
IN MIDDLE PERU,
Wale umilla
امراه
www.d. WWW.EU
امتداد دارد ولی این دوري
Wazlik
el
w
MT
کو سیاه در خراسانی و فورم
BY
CHUDI
mer
Sunken Chambers
LE
#111
WA
WA
w
Nr.
M. UHLE.
:اسان ادرااور
m
V
20:00.74
0.tw
Pics
VALA
immons
son DIY
WN
Infolent 123
.
che st. Walia
ವಸತಿ
es
in
13
NIS
Sepet
Wewe
62
.۱۱/
Hében!
11.11
lin
RY
PRE
Colliers
Sveltolle
.
)
3/14
Juliwa
NUAL
ALL
AN)
WHEN
716
DIE
அவர்
11
WITH
ALAM
LAS
e.
$. tot Waniti
Tambo
Wi..
Sun Fields
Sunken
nal
I will
S**
os
linna
STA
MAM
out
31.5 was
morruumis
Chambers
)) //
22
Sen
VAK
ws
venementen
While the
mirem
kahwi
VII
og
materworth
would
Woes
talen
Itthianza
AB
/ را
h
SAH ******
C
C
white
Autor
olish
W/ml ball)...
NA
...di
Ingen
را اور
111.11
2.
2
billi
۱۱ / ۱ //
190 HP
D
wa!!
AR
INIZ
*****977
Wwwwww!
اند
سال اول و سیما
PH
TI
La
MINNET
by fungolla
:۱۱ زرار
***** .
Ultid Mondell
re
lors
www
pH.
15
Modern Tambos
HI
PAN
ww
.
18.
Ville
...
propin
Mende
Units
www.well, de onder
Fullwidth
Winnie mousehdet vore WAV WAV
Wanawa
Notes
Tv
/////ello... making sile, ali
here
awy
VAVAAN
w skiller ...!
Whil
8
.
WO
SWELL
///
l} full
Son
Bahan
un
5
MORE
fyrir
VA
WA
&
ANO
wie
راديويات
Wann
***NERA
E
US
0
www.in
MELVIN
und
Lisa kita mim
Mond
k?
EVEN
1. Umine
eng
16 We
There
4 r.
map w
114
TV W
oner!
MAY
wi,
ಸsKSticiaWS1313
VUA ON
Luna
11
whe
within
WAKAWA in Williams
w
19
11!
با ایفا ۲۷ / ۱۰ / ۷۰
hasa
Minthe
ollands
Quarters of_Pizarrón
ЈАмери
2018
W
Pne vs
Job
11 "/", VN + V + a + +
NEW Brennade
wwwwwwwwwwww
shah
Airten
Abg
www.hall
*** W
will
.
.: ине "уруутг
WHO
WWW
1
171
W
why
th
OALS
县
​Whendeltem ilm, mely wllle, wolke
Sep
.
SA
illi,
WA
0.01mina
WWW.
a
V11
Nasar
Alue,
Peny
1/47
که آن را ار
pl
......... en
Wate
wome
MU
R14 PAS
Web
wwwURE
از ۱ از
With
emili
...3
22
Wewtwo
PM MINUS
Namba
Myy
MAIS
tu.
Pro
Vus
Wh
D
۱۱۲۰ ۱۱۱ ۱۱۰ ۱۱۰ ماده ۱۱۰۰ مانده بود و
VIDZA
andere
w
الاولاها الورد
WAM
11x
WW
LENJE
UL
பகயை
..lu
ws
iesletin.
We would
***********
اوراد والبوابادوامه
de Www.ch. Naw
VA
AUSW. . . . Note the whole Mariste...
4/11/11
.
W PRICE
w
HA
Western Streetupy
who
ANAWheckoma
www
و کار او بود، در ادامه اداری و ابزار الاتهام بانی
*
more
wie.
Se
IN
اباحي
enland
S
إليه لقا بالقوة و
AWG
willy Wo sulle all well by. we wille Will Webes esthuth
w Wowowanie do
Re
pation
want. Wh, ti wurde
mihi
Willial/
WAN
MIN
VE
WWW.
m
...
zo
어
​m'n
HALOZI
Rott
VUE
1.1., Fore' Court
'
還
​Foi plan
Sihim
Luli /
un
M.
シルバー
​خانه دوانده، از
she wear
sühkers
SAMA
Wilhelm
Whitekten
With
Ww
TE
herry
VIH
bull, no
III UN
WAN..Whereveile
Why Men
wir
Wirbel
www.gute MAMAQWW.
VA
VW VW
Modern Bridge
WOJVU
.77
V11
Me
11 و 11
ama
HII
W4
Chambers
...WWW.
Wilmaiselle
112
weitling
Alle
ble to win the
koht
DIWANI
ورود راه قرار
mai
m
74
S
AFYA
w
SMS
W
WWW
TAS
121
We will
.
Wel
BE
www.
will solelhetlanda
The WM111.
Son
WANG
an
WIN!
MAALAIlii
that Allik
ill
Video
Wiel.
Eastern Street
/
3/11
111
Wit
194
will,
Say when
.
Awal
WM
Min
འ''*#***
- و -۱ ,۱۱۰ ...
7
11.
Z
F.W..."
27
...
2. Fore Court
MM
.
Im
ANN
Wor
why they
When we
"11
Twins
Wedig
"A905!!!
kunta choo
hots
Trishhh kiri
! / اب
humanity
stille
WW2
altrike
AWA
HENNES
IN
rawie
int
4
が​かり
​Wmu!
Ah!
ONCERT
merenn
Izowy
Powe
WWW.
wymy
full
44601
11!
Х
SNES
!! (
دان
ch
WAN
W
wird
bu
.
ayam
Hill
Showball
*****
rehr
lill
X
WI
WW
Win
WiN
feed
by Wine
pastnil
स
Water
, Mwiat
ko
Muller
HERE
.
17/07/2
W
Nie
Autis an
.
#role, West
411717WTON
11!
Alinin
wavunma
LAMA
NEWS
him
WWE
நாங்காக ப
NIM
RE
اااااااا أ
7/10
ith Wale
Wild
Wewe
ma
WA
WAWMW...,
...ch
B
TA
re
VIVA
.
கா
We w... WAVIN
R?
SIR
2
25
Empre
Pilgrints' Houses
Snow
A.
2.
www
4910
il
IV
111
אני וייוין קליין /
(!
NA
llow
.و ۱۱۰۱ به او دادن
VAATA
are
ww
ws.
ent
EN
AP
Surs
ats
w
MES
FLORIN
www
az
if
worldwide
live
Hulle
WU
s
hli
2. W
3. Fore Court
وع
2
A
dalali
www.
Pet
...
11 11
hum,
Pellente
11
FI
WMA,
.
.nl
SPA
Compare
法定
​19
Cötönnå de
HE
*
*
**
*
*
Si
**
*
亮
​9
5
MOR
..
* **
WANCAIS
山
​卷
​as
to sa
"11/1,
*
*
in
*
*
TK
25
ik
Square
ming
AN
AL
Will
VZZ Z
(/N/1/72. tal
۱۱۱
dith
AVAIL.
bulan
antara
W
11/12
1
WAM
AC
Ali
(\ //
Wri
P98
Will
* *****
Monte
will
1)
WL
பூக்கள்
22
le
* * * * *
Colonnade
JIVI
www.
IN
Zine
www
ARU
.
MOV
சாப்பயாம்
WAMY WW, We
Then, 1 page your fingers
"
* * * * * * te
111
MIIN
Uwavu.
hik
w
au
تال
WW
NE
#
دولا ولا بماند تا 15 بها
WWA
ee
Meme
::
یه به ما 1 یا 4
llimi
Hewletinguts
AM
SAMA
& River of Lurin
HAMU
Nawilidhani nimir-se al suo all she was able vir die water.
in the US
wawi w
1/15
ががが
​TAS
118 را مازه ل
be
FI
WA
wi...
h
mi
and would be listed
NAVENA
Women
wai
116.
ipsius
VANY
mimi
WHERBST
We
WILL
IN 2012
NIVI
komthing
"/
while
.
10**"
While
11!
.
Waway
WWW.
en un moment en een hot
lill
VLON
WIN
..
...
\W
.
WWW.41
19
.
RA
Auto
101
.),
I!!! with
softer will be remo
ll!
B!
She W/
his
WWW
ان رو، برادر /\ر,۱۱
wy
المتقدمة هذه البلاد. و
W
wer
MA
lor More
PRAWIAM
gim
.
53
WWW>>
Whe
Band
SI
VAN
Tom
Templo Enclosure
Temple Enclosare
Külli
M
nog Vichar e vetar
11
والدها من الطلاب المسالك والسلوان السادا
با
12
ورواه اااه
MIMI
WW
sesi
MANUS
str. 1
""wli with,
HD
ولارداد ادا
WE
ژیانی
llll
1,16
SIN
Altre
14.
M.
14
***/*/*,
Ruined Building
yull
qilinin
Watki,
a
76
illalba
Vikiba
Wiatra
ithil), la thuil, ...,
1136
SEMINE
Air
Hall
Sur
hu
21?
Mel... toththu
Fields of Hacienda Mamacona
VI
ME
MIT !!!!
We Are
יווווו אילולא
/ \\ // ۱۱ / ما را /
ان //ادرار
, ,{ا 1 / ۱۱ / ۱۱ / ۱۱ / ۱۱
אמא אמא
71215
\
1 / 1 kg
./11
Terhi Mo LAMA
பாடம்
enkele
WA
My
Want
Vanda
4914.4.2
*N
FRIA
sowie do
7100
Hi
*|༌།
11
ol
11
Sabina
View
పురం
wpd
292
... lende
.
M'Atocongo
cum
مر/111
SUOTA
EUR
ya!
7!
WS
pill!
Menu
۸۱)
Wall
..
Il
Agns
HEIMA
hvern
ly's
Venturosd
W!!!
:))
'1111!!!
1197
11"
ار
***
. \
chuin.
ni
YN
fe
...V.......
Phyperlite
الوان / ۱۱/۱/i
القهر والامان الدهانه : ( اقلیمی
1:12,
-
۱, ۱۱/
14., 16.wrl,
Ví.
Live
Malt!
2
AWAY
Montagnettiin
7.AM
لااله امي وانا امان الله
Homeworytitle I., la,
la,
n.
\\
inweit war
//
...AL
tha...IN..
II
WWA
بار بار در ما
Palasport
E
Will
Villan
glesias
ابرار
tavalla
//
"A
ht
WW
9!!!
Parlaces
11
991
VIL
mwille,
win the sales while
Mwana
/
4.74
tel.de
masin/
7,111
MO
ll
Watud
Am
totesi! Here They
11/\/\/
،،،، اما الان
ارزو امان، ادا
the ident
نوروز ۱۱ را 19, ولوالد، (۵۸۸
mana
!
sca
ton
Pachacange
agare
lihintais
Las Palmas
/ رام ولار! / او را
Wall
WAL
alle
WA
17,
4.11
-1/2
I HAVE
mille
Mound
wil hul
WALIWA
d
BILAN
MINE
L27124
Apetent
ul
Part. 4*
wit
(61)
VA
20
Mens
llon
Ruins
Tablada de Lurin
"T
WW
La Rinconada
L
"11317
1.
hihollow
11
varig
Wielkie
Woluwarden
1111,
NI
Jual
AYAWAV
three por
: .
MA
Italie, will
May
Pachacamas
Urrinya
Buenavista
spore >
V.
Small
Wiki
VA
$
70 alesef
Werstand that
dk.si
2
74
hly
141
:)
...
TA
Goglas
Mamacona
/۱۱ را « ها
.
(1271M:
will
Urban
1/
San Pedro
1141
La Centinelo
We
W
Mer...vitenlle uthu, vtedy
INI
ใน
วัน
ill
Lurin Valley
all/
Warangal.
DEL
Ene
*****
Wit
Malawi,
Willi Westli
th
e per un sentit a Benue on her Amyrass
1119
wy
Curin
will be with the hiciell,
HE
us
9
ampte Enelosure
DI
fy
ze
1/10) ||
pripo
Hall
:تار
*****
14.11
27
11. "
.
T!
ahliane
E
W
WARN
WA
MILAN
MIA
3
Milouhi
howlittle
·S
Scale 1:2000
M
Lurin Valley
0
100
200
300
400
500
600
700
800
900
ma
Pachacamac I:
1000 Feet
IE1
וווווווווווו
ENTITLEHT 11111111111
1119) IUNIITWINTI1*11
110UNTUTISTUMPURIIL
MA
MVH
IM
0
50
100
150
200
250 Meter
VALLEY OF LURIN
111111111
11011
||| 411 TF أ و قرر
IDATI
18AUTY
117770
UNTUT
Scale 1:60000
RIVER
WAAHA
1 Sea Mile
Kilometer
70459 W.
TO

НОВИЯ АНТЕНО ЧАМ
ПАНДА ОАН ОАО
ULO
TROHE
ли
ООО ѕоа


Uhle, Pachacamac
PANORAMA OF PACHACAMAC
SEEN FROM THE TEMPLE OF PACHAKAMAJ
ONIL
WICH
OF
1



1
THE SUN TEMPLE AND THE TEMPLE OF PACHAKAMAJ
FROM THE NORTH
2
Uhle, Pachacamac
THE CONVENT, MAMACONA
SEEN FROM THE SUNFIELDS
UNAL
OF
HICH
2


3
PLAN OF THE TEMPLE OF PACHAKAMAJ
Meters
که
0
250
Uhle, Pachacamac
0
10
139
20
330
Feet
23
007
73"
مه
0
230
: و20
Scale
description of the Temple in the text.
The letters correspond with those used in the
x
+
*
X
*
*
۶
Y
\۱۱۱,
\\\"
*
777777
X
7}}}}
*
+
*
او را
کم
กร
(17 /1) //
4
1
-
مام
+
X
۱۴( (
X
۱۹۱ در
//
لا
(
1((
ار
لا
را. ||\
در روز ۱۰ رارا ان را
۱۱۱/
/۱۱
۱۱
،،
XC
لا
(1/{\/
۱۱}}
11
(
(((((
/ ا / ۱۱ / ۱۱ / ۱۱ / ۱.۱۱/۱۱
)))))
۱۱ // \\ // \\ // \\ //
////\ /\ /\ / ۱۱ /\ /\ /\ /
۱۱ / ۱۱ / ۱۱ // \ / 11 / 17 ) 1 (
))))
%
۱ ( 11 / \ //
/۱۱/۱
//)
//
ر ر ر ر ر ر ر ر ا // //
) ۶۱۱۱
را وارد بر او آن را
{ /////
1) /
راه وارا
. در ۱۱
{\\
، زار زار را در با او
(
ز
۱۱ // // ۱۱۱۱۱۱۱۱
با ... // // ا / کار را
// {
، ا . )
ر(
///////
17/۱۱) | \\ //
• ۱۱۱۱۱۱/۱۱ // ..
וניירנוזייו
)۱/۱}}/11
} }
}/)
پی
رز
۱)
N
17/
(((((((((((((((((( رب(11در1/2 در رود در((((( ((((رد. (((( 12 (۱۱ /اراده ۱۹
ور
(1)
\/
مالی را ر/ نه
راز
(((
۱۰ ۱۱ ۱۱۰ ۱۱-۱۱/۲۱۰۹(11 111 رای ۱۱» را
)
M
.را ,(
(
(
(
را
/// ۱۱ ۱۰ ۱۱ را \\ //\
۱/۱۱ :
۱۶۱۲
//
دورا )1121 برار
E
۱۱۰۰*۱۰۰۰۰۰
j/
/////////
// ( از 1 // // // // // //۰
((
۰۱ / ۱۱ / ۱۱ / ۱۱ / ۱ / ۱۱ / ۱۱/۱۱ارد را داد ( را ارزان (((((
S
(....:
C
/از
//)
12_۱۱ رای // / ۱۱ارد
,ال
۱۲/۱۱/۱۱/۱۱۷۱۹/۱۰)۱۱۹۰٫ /۱/۱۰/۱۱/۱۰ ۱۱۱ / ۱۱ / ۱۰۰
۱۷/۱۰ // // ۱۱ / ۱۱۰۰
l :
۱۰۰ \ /-
۔
ر11 /16
،
OH.


lc
la
1b
CE
ne
10
E
DO DEE
S
데이
​14
CON
1
E
10
PTS
없
​1
.
미
​ld
le
2
4
3
릴
​- 0
이과
​-
터
​5b
ба
힘
​0
0
1:
:
1
Uhle, Pachacamac
CEMETERY UNDER THE TEMPLE OF PACHAKAMAJ
OF
UNIL
OBJECTS IN THE STYLE OF THE MONUMENTS OF TIAHUANACO
FIGS. 1a-e: io N. S. ; FIG. 2: AB. %; FIGS. 3, 4:; FIGS. 5a-b : N. S.
4
1er


1
4
3
2
水
​AT
ol
3
o
6а
66
5
7
se
CO
三​)
jye
哥
​22
/22
in
\
%
三
​10
8
11
9
13
15
{
g
18a
12b
16
17
18
14
19
Uhle, Pachacamac
CEMETERY UNDER THE TEMPLE OF PACHAKAMAJ
OBJECTS IN THE STYLE OF THE EPIGONE PERIOD
FIGS.1-11, 13:%N.S.; FIG 12:4; FIGS. 14,15:%; FIGS.16-19: N.S.
ONIC
OF
,
5
少​ies


3
2
1
5
4
E TO
8
6
이
​를
​10
9
15
BE
b
11
12
10
14
13
16
Uhle, Pachacamac
CEMETERY UNDER THE TEMPLE OF PACHAKAMAJ
TEXTILES OF EPIGONE PERIOD
ALL OBJECTS: 1% N. S.
G MED
Os
6
ICH


1-14
4-14
3-14
2-14
In
5-14
6-24
YY%
SE
8-)
9-12
2
SE
13-16
105-12
11a-14
11b-14
12-24
10a-92
16-12
14-14
15-14
18-14
ஐR 882888888 28.3ERS
988888888888230aea9
adagasca&83 1982
swala sa Sara BH
US
storan & BUMB
17-14
20a-ab. 14
21-10
19-1.
合
​B
A
20b-ab.14
atau salah
TE
A
22-14
23b-
23a--
Uhle, Pachacamac
FIGS. 1-9-GRAVES IN DÉBRIS OF BUILDINGS UNDER THE TEMPLE OF PACHAKAMAJ, ABOVE THE ORIGINAL CEMETERY. OBJECTS OF THE LATER
PRE-INCA PERIOD
FIGS. 10-21-OBJECTS FROM INCA GRAVES IN FRONT OF THE TEMPLE OF PACHAKAMAJ
FIGS. 22-23b--OBJECTS FROM THE CEMETERY OF LA CENTINELA NEAR LURIN
GNIL
OF
7
MICH
/


1
5
2
3
4
EN
00
9
10
ng
6
14
12
11
13
2
15
18
19
17
Ela! ER
G
16
ce
G
ALA
La
D
M
hy
Uhle, Pachacamac
CEMETERY SURROUNDING THE TEMPLE OF PACHAKAMAJ
OBJECTS OF THE LATER PRE-INCA PERIOD OF PACHACAMAC
Figs. 1-16: 14 N. S.; 17-19: % N. S.
UN
8
MICH


1
VIEW OF THE OLDEST PART OF THE TOWN
FROM THE NORTH
2
Uhle, Pachacamac
VIEW OF THE SOUTHWESTERN PART OF THE TOWN
FROM THE NORTH
GNIE
9
ICH


1
2
THE EASTERN STREET OF PACHACAMAC
ENTRANCE TO THE PRINCIPAL PALACE OF THE TOWN
FROM THE EAST
3
4
UN
Uhle, Pachacamac
OF
GNIE
RUINS OF A TYPICAL PALACE OF THE TOWN
RUINS OF A PALACE WITHOUT TERRACED FOUNDATIONS
10
CH


2
1
ایده الهه ها
VIEW OF THE NORTHERN PART OF THE TOWN
TERRACED SLOPE OF THE HILL WITH WATCH TOWER IN THE
NORTHWESTERN PART OF THE TOWN
FROM THE WEST
3
4
Uhle, Pachacamac
PALACE WITH SUNKEN CHAMBERS IN THE EASTERN STREET OF PACHACAMAC
SUNKEN CHAMBERS IN THE SOUTHWESTERN PART OF THE TOWN
11
UNIV
OS


1
VIEW OF THE SUN TEMPLE
FROM NORTHEAST
2
Uhle, Pachacamac
30
ONE
VIEW OF THE SUN TEMPLE
POHI
FROM SOUTHEAST
12


1
2
4
3
6
5
7
a
M
au
C
antivi
o
9
8
11
10
AAANA
WE
12
17
15
14
13
25
23
22
26
18
16
24
I
21
20
19b
19a
w
Uhle, Pachacamac
CEMETERY OF THE OUTER CITY. TOWARDS THE NORTH
OBJECTS IN THE STYLE OF THE INCA PERIOD OF THE TOWN
Figs. 1-8, 10-15: 74 N. S.; Figs. 9, 16-18, 21-26: %2; Figs. 19, 20: % N. S.
UNIL
OF
MICH
13


1
2
CENA
TERRACE OF THE LARGE NICHES
THE SOUTHWEST FRONT
FROM THE WEST
FROM THE SOUTH
3
4
Uhle, Pachacamac
NICHED WALLS OF THE SOUTHWEST FRONT
ENTRANCE TO FIRST TERRACE
SEEN FROM THE PLATEAU
SEEN FROM ABOVE
HO
SO
VIEWS OF THE SUN TEMPLE
14


1
2
NORTHEAST FRONT WITH THE RUINS OF THE TERRACE PASSAGEWAYS
SEEN FROM THE THIRD TERRACE
2
3
Uhle, Pachacamac
TERRACES OF THE SOUTHEAST FRONT WITH THE CEMETERY OF THE
SACRIFICED WOMEN
MOUND BUILT OF ADOBES.
TO THE NORTH FROM THE TEMPLE
UNIL
OH
VIEWS OF THE SUN TEMPLE
15


76
77
16.11.11
"1111
(333
F
I!!!
69
Mancany
li
|/\
68
Wikile
I
1117:/
ANAWA
MINIAI
34
Suui
הירון
This
WUM
will be
!!lolla
nelly
NI
TISCIVO
در ایران
را
wu,
29
.
19
27
PUUTTUU
hallowing
70
26
au ga
like
\\\\\\\\/\// ۱۱ / ۱۱ / را. || || ({اورا/ \/ \\\\\\1\{1
18
33
/۱/
76
Wallet
mum
th
יהיה להוזילי
B2
“সে যা
25
65
70
کیا اور بازار را دارد و او را (۱) سلام 71
اراده اند. 8/1/18
دارد و در
FOOT OF THE TEMPLE
We
.
OHAU DAN
Waltesaada al TATTO,
70
867
WHIT
I la
اسبانيا -
افراد اور اولاد 1 1. ، رانا مراد اور اس
With
الدور او را با
New
?יוזיוןי ]
Male
11!
baten Wahab Wale wanak
Wakati
Wille
A Wall / WWW.WISAVALLA Wall ///
/*
33 zudeler BLAUWVAINK: 3166 lb.luralisasyon Valle,
il trattati,
WOO
*33 mm
VLAND. Without
II 21
w Wawa 2020 20
23
www.All H../../
1111W WWW We will allow
WALANG
140
WII,
WWW WA
39
lie
12 2,132
IT
· 19
nale Monday malkasta. Walimwi
SAKAL
WEMA WAWATAN
1.1) 11/23 (1)
ch
崇​崇​蒙​器​深紫​禁​禁​禁​禁​禁​禁​禁​禁​禁​茶​熟
​BRAVO
30
28
GORMER
Eiwum
14
MBALAAAAAA
64
091918.1186/12/blotil)
, sales Jah WROOFED / lille grip
Wom
ey pampangan
V
allo, w 24
little
To wille wil gelina
pir
JAWA
رادرانادا الملکرد
لالالالالالالالالالا
VAATA
//
و الأد,{11/أ ((ارا لیا اور
12/14
VA
VA
Jaunatv/
an
se bo
3
N.
TEMPLE
GB
172
IMM
7
шп
IT
Wasthu.
PLATEAU
6362
113
all mill USUADALUNG
WWW JUALAN
1)
名
​THE
st
10
5
wwwwwww
ikaw Nalle meci care surse
MAAVAT
ia
وداد و الدرام(
2
ا ا ا ا ا ا والد ہ لالالالا
na USA
59
When
LP
/ 9 / 11 / 1 /1 /11/18 144, ای ۱
, دار و در
61
1holhouj News یا 15. V/1U34
Strips
MW MILLES
ENGOSURE
STAR
V
، اگر را القرار الان به نام داد
SANCTUARY
6
WAALALALA
N
காக
A
BÖBIA
111
AVAHENY, Atill
"A
ma
/IIVITIM
Wom
WAVE
WWWXXU
DAVID
7
بود را ۱ د ما.بها
GA
20
2017 yeni ma
/ والنار
1
יויוריון:YN
H42
remy MICO
WALYA
FE8
it! petering
jy 48/74
66
wi
mm Wein
3a
49
GROUN 43
74
Sil
li
-37
TI
NI
TIN
IN
fu
VITALY
HY
AVC
Gwade
HA
MI
564
50
50
I
类
​4
60
RY
اوايلد. ماد لما بدا اما
المسألك
2017
OPEN PLACE
frT7 [vt(s
my
I
YA
54 54
Woulutus
371
I/100
الطد علادالطال
25
AMAL
54 54
U
53
Miran
MITY
54.54 1541 //IV. - 14414{{{E41-7г ичhtter
FOOT OF THE TEMPLE
rivillilitch
iyorum,
VI
TADO
min
Where
mu
Valhal
m WAT 2
apygli
OF
ru, que
Aluno
11901
WWW.11.
ENCLOSURA
171717190
IN
Manan
12 2014/1271) ITV
WYMO ////////
M.
MIT...
Uhle, Pachacamac
The red Arabic figures correspond with those used
in the description of the Temple; the red Roman
numerals mark the height of stories.
PLAN OF THE SUN TEMPLE
Scale
1: 1000
4 30 20 10
30
100
130
200
250
Feet
50
40
30
20
2
50
100 Meters .
ONIC
ICH
16


I
1
77
12
V
31
713
17
41
63
15
bla
JS
16
I
14
I
os
6.2
33
(
IT
32
S3
22
33
69
20
40
32
77
33
33
11
7
South-western Front
7
2
2
U
V
Ꮧ
63
J
49
48
ol
64
62
政
​2.
-
-
North-western Front
TILTE
59
-
I
I
09
T
I
7
67
படிபடி
I
69
3
1
7
2
5
IV
IY
(I).
38
хи
1
-9-
8
ತಿ.
North-eastern
Front
eti
पक्ष
7
1
-
(17)
179
49
TI
II
37
(IT
37
---
I
II
1
36
74
36
60
99
57
52
51
45
I
(1967
35
57
55
35
In
571
56
54
54
34
47
U
54
53
4
Leon
-
1
-
7
TY
17
South-eastern
Front
MI
(7
40
1
13
и
!
1
쁘
​45
.
47
44
Uhle, Pachacamac
ELEVATIONS OF THE SUN TEMPLE AT PACHACAMAC
GNIL
17
MICH
'


1
2
3
UNULUTUSU
Unununununun
4
40
5
8
6
9
10
7
11
12
13
15
16
19
14
17
18
了
​Uhle, Pachacamac
CEMETERY OF THE SACRIFICED WOMEN. SUN TEMPLE
OBJECTS OF POTTERY,
POTTERY, WOOD, ETC.
Figs. 1-8, 10-12, 15, 16, 18, 19: % N. S.; Figs. 9, 13:14; Fig. 14:%2; Fig. 17: N. S.
U NHL
HIGH
Os
18


1
2
3
5
6
9
10
00
8
SC00322100DD32DDID MER
4
Oro ff's
11
acute1.ticum
12
13
rece
Uhle, Pachacamac
CEMETERY OF THE SACRIFICED WOMEN
ARTICLES OF DRESS
Figs. 1-4, 6, 7: N. S.; Figs. 5, 9, 11-13: 74; FIG. 10:72; Fig. 8: AB. N. S.
NE
OF
19
ICT


1
VIEW OF THE NORTHERN AND EASTERN PARTS OF THE RUINS OF THE CONVENT
2
3
Uhle, Pachacamac
FRONT WITH LARGE NICHES. CONVENT
WALL TWENTY FEET WIDE IN THE NORTH OF THE EXTENDED TOWN
SEEN FROM THE COURT
SNE
20
CH


N
W
E
OINA
29
EN
A4-
22
To
1
PILMI
!!!!!!
21
12
20
!
P
1112
19
3
Via
h
18
! 4 .
71
U
m
4.,
HICS
((((( ( ( ( ((
11%
5
17
161
115
un
6
17
7
אוליי יוון/יייי יוון '
lllull, Ill/hiller
Prane
11171113
8
9
INH14141417 44571 개
​4910
11!
li!!!
Uhle, Pachacamac
Scale
1:500
The letters correspond with those used in the
description of the Convent, in the text.
23"
0
50
73-
100
123"
239
1
بابليليا
Feet
23"
20
13
10
0
27
30
1 Meters
PLAN OF THE RUINS OF THE CONVENT.
.
PACHACAMAC
blow
UN
21

1



BOUND
H23 1931
OF MICH
UDRARY