Highland: Excavations 'Huaylas..And Chavin DE Huantar

TH11 HIGHLAND
EXCAVATIONS IN THE CALLEJON.

.DE ~'HUAYLAS..AND AT: CHAVIN
DE HUANTAR
WENDELL C. BENNETT.
VOLUME 39 ,:PART 1
ANTHROPOLOGICAL PAPERS OF
THE AMERICAN MUSEUM OF NATURAL HISTORY
NEW YORK: 1944
El ...
THE NORTH HIGHLANDS OF PERU
THE NORTH HIGHLANDS OF PERU
EXCAVATIONS IN THE CALLEJON DE HUAYLAS

AND AT CHAVIN DE HUANTAR
WENDELL C. BENNETT
VOLUME 39 : PART 1
ANTHROPOLOGICAL PAPERS OF
THE AMERICAN MUSEUM OF NATURAL HISTORY
NEW YORK : 1944
PREFACE
THIS EXPLORATION in the North Highlands of Twenty-six days at the famous site of Chavin
Peru represents the fifth expedition in the An- de Huantar allowed the excavation of 16 pits
dean program of the American Museum of Nat- in these ruins, a chance to examine much of the
ural History. Introduced by the survey of Dr. little known stone carving, and a general review
Ronald L. Olson in 1930, the program was con- of the site. The excavations uncovered pottery of
tinued by the author in Bolivia in 1932 and Chavin style and several post-Chavin styles.
1934, and on the North Coast of Peru in 1936. These are treated in detail in this report.
Three reports have already been published.' Following this work in the North Highlands,
The work was financed by the Frederick G. we spent ten days at Chiclin Hacienda in the
Voss Anthropological and Archaeological Fund. Chicama Valley with the Rafael Larco Hoyle
My wife and I sailed from New York on family. A further study of their famous collec-
January 1, 1938, arriving in Lima on the fourth tion was made in the light of our recent dis-
of February. We had the pleasure of spending coveries. A final month was spent in Lima,
ten days en route on the Island of Barro Colo- studying the collections in the various museums
rado in Gatun Lake, Canal Zone, through the and arranging for the division of our collection,
courtesy of Mr. James Zetek and Dr. Frank M. part of which is now in the National Museum
Chapman. After twelve days in Lima, securing of Peru, and part in the American Museum of
a permit and equipment, we proceeded to the Natural History. We returned to New York on
town of Huaraz, via plane to Chimbote, train July 25.
to Huallanca, and motor to Huaraz. Today, the To thank individually all the numerous per-
whole Callej6n de Huaylas is accessible by auto- sons who assisted us in our work in Peru is
mobile via the new road up the La Fortaleza obviously impossible. To the Honorable Gen-
Valley. eral Oscar Benavides, President of the Republic
A total of two months was spent in the gen- of Peru, and to Dr. Ernesto Montagne, Minister
eral region of Huaraz. The local museum, under of Public Instruction, we wish to express our
the direction of Dr. Augusto Soriano Infante, gratitude for their continued interest in the
contained a representative collection of ceram- archaeological investigations of the American
ics and over 125 stone statues- from the area. Museum. Dr. Luis E. Valcircel, Director of the
Exploration in the region led to intensive exca- National Museum of Peru, and Dr. Julio C.
vation around the ruins of Wilkawain and Tello of the University of San Marcos con-
Ichik-Wilkawain, and minor testing at a num- tinued, as formerly, to give every assistance and
ber of other sites. These ruins and excavations attention. Dr. Alberto Giesecke and Sr. Jorge
are described in detail in the body of this report. Muelle were among the many who aided us in
Ten days were devoted to the region of Lima. Both Dr. Augusto Soriano Infante and
Carhuaz, particularly the ruins of Copa Chica, Sr. J. Eugenio Garro accompanied us as official
where minor excavations were effected. The col- government representatives in different periods
lection of the La Rosa Sanchez family in Car- of our work. To both we express our pleasure
huaz is said to have come from their Copa Chica for their good company, useful assistance, and
Hacienda, and a reasonablylargecollection in the collaboration. Dr. Emiliano Ciceres of Lima
Museo de la Universidad de San Marcos was ob- kindly presented us with many useful letters of
tained by Dr. Julio C. Tello from the Copa area. introduction. His brother, Sr. Ludovico Caceres,
Eight days at Katak, near Recuay, permitted was our good host in Katak. Sr. Tomas La Rosa
examination of a dozen sites in the region. The Sanchez invited us to use his hacienda at Copa
famous Macedo Collection in Berlin came from Chica. Dr. Rafael Larco Herrera, and his sons
this section, and the name Recuay has thus Srs. Rafael, Constante, and Javier Larco Hoyle
been applied to one style of ceramics. Dr. Julio offered their kind hospitality at Chiclin Ha-
C. Tello has excavated sites in this region, and cienda. To these and to many others we wish to
we tested several others. offer our sincerest thanks.
1 This Series, vol. 34, pt. 3, 1934; vol. 35, pt. 4, 1936; vol. WENDELL C. BENNETT
37, pt. 1, 1939. October 14, 1942
5
CONTENTS
PREFACE . 5
PART 1. EXCAVATIONS IN THE CALLEJ6N DE HUAYLAS . 11
The Geographic Setting . 11
Huaraz Sites 12
The Wilkawain Section. 14
A. Temples .................... 14
Wilkawain 14
Ichik-Wilkawain. 17
B. Above-Ground House Sites. 17
Excavation of House 1 (Site 6H-2). 18
Materials from the Grave at House 1 (Site 6H-2A) . 18
Refuse Material from House 1 (Site 6H-2) 18
C. Deep Stone-lined Tombs. 21
Materials from Deep Stone-lined Tombs. 22
Observations on the Deep-Tomb Ceramics 30
D. Stone Box Graves . 34
E. Unlined Graves. 35
Observations on Unlined Grave Materials . 36
F. Subterranean House Sites. 38
Materials from the Second Subterranean House Site. 38
Observations on Subterranean House Collection 41
G. Subterranean Galleries ..... . . ....... 42
Materials from Subterranean Galleries.. ... . 44
Observations on Subterranean Galleries and Collections 48
H. House-Gallery Stratification. 48
The House Collection . 48
The Gallery Collection. so
Observations on the Wilkawain Region. so
Site 7H-15 . . . . . . . . . . . . . . . . . . . 50
Middle Period Collections . 51
Deep-Tomb Collection. 52
House Site 6H-2 . 52
House Site 7H-5A. 52
Subterranean House Site 9H-2 . . . 52
Small Grave Collections . 53
Resum6 . 53
Other Sites near Huaraz 54
Shankaiyan, Site 1H. 54
Site 1H-A .................... 54
Site 1H-B .................... s7
Site 1H-C, 1H-D ................. 58
Observations on the Shankaiyan Gallery Collection . 59
San Jer6nimo, Site 2H. s9
Ayapampa, Site SH 60
The Carhuaz Region. 62
Copa Chica Ruins. 62
House Type A . 62
House Type B . 63
Copa Grande Ruins 63
The Katak-Recuay Region. 64
Some Construction Types in the Katak-Recuay Region . 64
Miscellaneous Constructions ............ 69
7
8 ANTHROPOLOGICAL PAPERS AMERICAN MUSEUM OF NATURAL HISTORY Vol.39
Sites in the Katak-Recuay Region . . . . . . . . . . . . . . . . . . . . . . . 69
Excavations at Site 2K-A . . . . . . . . . . . . . . . . . . . . . . . . . . 69
Observations on the Katak-Recuay Region. . . . . . . . . . . . . . . . . . . . 70
PART 2. EXCAVATIONS AT CHAVfN DE HUXNTAR. . . . . . . . . . . . . . . . . . . . 71
The Ruins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
Unit A, the Castillo . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
Unit B, the Idol Section . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
Unit C, the Chapel Mound.. . . . . . . . . . . . . . . . . . . . . . . . . . 73
Unit D, the Castillo Terrace . . . . . . . . . . . . . . . . . . . . . . . . . . 73
Unit E, the Sunken Plaza . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
Unit F, the South Plaza Platform . . . . . . . . . . . . . . . . . . . . . . . . 74
Unit G, the North Plaza Platform . . . . . . . . . . . . . . . . . . . . . . . 74
Unit H . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
Unit I, the Inca Amphitheater . . . . . . . . . . . . . . . . . . . . . . . . . 74
The Adjacent Regions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
Excavations ................................. . 75
The Chavln Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
Chavin Ceramics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
Chavin Architecture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
Stone Carving . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
Post-Chavin Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
89
Post-Chavin Ceramic Styles ..... . . . . . . . . . . . . . . . . . . . . . 90
Inca Style . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
San Jer6nimo Style . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
White-on-Red Style . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
Minor Styles . . . . . . . .
.. . . . . . . . . . . . . . . . . . . . . . . . . 92
Other Post-Chavin Artifacts . . . . . . . . . . . . . . . . . . . . . . . . . . 92
PART 3. THE NORTH HIGHLANDS . . . . . . . . . . . . . . . . . . . . . . . . . . 93
Chavin as a Whole . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
Architecture ................................. . 94
Stone Carving . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
Ceramics ....... . . . .. . . . .. . . .. . . . .. . . . .. . . . . 95
Chavin Comparisons and Distribution . . . . . . . . . . . . . . . . . . . . . . . 95
Chavin in the North Highlands . . . . . . . . . . . . . . . . . . . . . . . . . 95
Chavin and Tiahuanaco . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
Chavin and Coast Styles.. . . . . . . . . . . . . . . . . . . . . . . . . . . 97
Chavin and Early Ancon . . . . . . . . . . . . . . . . . . . . . . . . . . 97
Chavin and Puerto de Supe . . . . . . . . . . . . . . . . . . . . . . . . . 97
Chavin and Cupisnique . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
Chavin and Coast Styles . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
White-on-Red Style . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
Recuay ....... . . .. . . .. . . .. . . .. . . .. . . .. . . .. . 99
Recuay Ceramics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
Recuay Sculpture. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104
Recuay Architecture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104
Chronological Position of Recuay . . . . . . . . . . . . . . . . . . . . . . . 104
The Middle Periods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106
The Wilkawain Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . 106
Other Tiahuanaco-influenced Materials . . . . . . . . . . . . . . . . . . . . . 106
Callejon House Types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
Late Period and Inca . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108
Tentative Chronology for the North Highlands. . . . . . . . . . . . . . . . . . . 109
LITERATURE CITED ........... ..................... . 109
EXPLANATION OF ILLUSTRATIONS . . . . . . . . . . . . . . . . . . . . . . . . . . 111
ILLUSTRATI ONS
PLATES
(AT END OF TEXT)
1. View of three-story temple and deep stone-lined tombs at Wilkawain
2. Wilkawain-Recuay ceramic types
3. Subterranean galleries, Wilkawain
4. Construction types in the Callej6n de Huaylas
5. Subterranean houses at Katak
6. Chavin architecture
7. Views of Chavin
8. Stone carving of Chavin and Aija
TEXT FIGURES
1. Map of the Department of Ancash, Peru. . . . . . . . . . . . . . . . . . . . . 10
2. Groundplan of Wilkawain three-storied temple . . . . . . . . . . . . . . . . . . 15
3. Cross-section of Wilkawain temple . . . . . . . . . . . . . . . . . . . . . . . 16
4. Ceramic styles from above-ground house site, Wilkawain. . . . . . . . . . . . . . 19
5. Groundplan and cross-section of deep stone-lined tombs, Wilkawain . . . . . . . . . 22
6. Blackware style of deep stone-lined tombs, Wilkawain . . . . . . . . . . . . . . . 24
7. Redware style of deep stone-lined tombs, Wilkawain. . . . . . . . . . . . . . . . 25
8. Negativeware plate from deep stone-lined tombs, Wilkawain . . . . . . . . . . . . 26
9. Tiahuanaco style of deep stone-lined tombs, Wilkawain . . . . . . . . . . . . . . 28
10. Tiahuanaco style of deep stone-lined tombs, Wilkawain . . . . . . . . . . . . . . 29
11. Miscellaneous artifacts from deep stone-lined tombs, Wilkawain. . . . . . . . . . . 31
12. White-on-redware from unlined graves, Wilkawain . . . . . . . . . . . . . . . . 37
13. Ceramic styles from subterranean house site . . . . . . . . . . . . . . . . . . . 39
14. Groundplan and cross-section of Recuay style gallery, Wilkawain . . . . . . . . . . 43
15. Recuay style ceramics from galleries, Wilkawain . . . . . . . . . . . . . . . . . 46
16. Groundplan and cross-section of house-gallery stratification, Wilkawain . . . . . . . 49
17. Recuay style artifacts from gallery at Shankaiyan. . . . . . . . . . . . . . . . . 55
18. Recuay style artifacts from gallery at Shankaiyan. . . . . . . . . . . . . . . . . 56
19. Groundplan and cross-section of house at Ayapampa . . . . . . . . . . . . . . . 61
20. Groundplan and cross-section of house at Copa Chica, Carhuaz . . . . . . . . . . 62
21. Groundplan of house at Katak, Recuay . . . . . . . . . . . . . . . . . . . . . 65
24. Groundplan and cross-section of house at Katak, Recuay, Site 4K-D . . . . . . . . 68
25. Groundplan of ruins at Chavin . . . . . . . . . . . . . . . . . . . . . . . . . 72
26. Groundplan and cross-section of Pit 3, Chavin . . . . . . . . . . . . . . . . . . 76
27. Groundplan of excavations of Pit 9 and Pits 11-14, Chavin. . . . . . . . . . . . . 79
28. Chavin style rim types and post-Chavin vessels from Chavin. . . . . . . . . . . . 82
29. Chavin style decorated sherds from Chavin. . . . . . . . . . . . . . . . . . . . 85
30. Chavin style decorated sherds from Chavin. . . . . . . . . . . . . . . . . . . . 86
31. Post-Chavin ceramic styles from Chavin and San Jeronimo. . . . . . . . . . . . . 91
32. Recuay ceramic shapes.. . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
33. Stone carving styles in the Callej6n de Huaylas . . . . . . . . . . . . . . . . . 105
9
10 ANTHROPOLOGICAL PAPERS AMERICAN MUSEUM OF NATURAL HISTORY Vol. 39
Pomabamba
0
Caraz
1-
+
4t Jt
4
4c
if
Jr
4
IC
4
4c
PacLific Ocea-n y-
IK
y-
f
)t
)t
x
44 .
FIG. 1. Map of the Department of Ancash, Peru.

PART 1. EXCAVATIONS IN THE CALLEJON DE HUAYLAS
THE GEOGRAPHIC SETTING
THE CALLEJ6N DE HUAYLAS (Fig. 1) is the basin passes are over 4200 meters. Until very recently
of the Huaraz River, the principal mountain communications from the coast to the Callej6n
branch of the Santa. The Huaraz originates in were limited to the railroad from Chimbote to
Lake Conococha at 3944 meters altitude and Huallanca and thence by truck through the
flows north northwest between the two moun- Callej6n, and to combination truck and mule-
tain ranges called the Cordillera Blanca and the back trails up the coastal valleys such as La
Cordillera Negra. At the northern end of the Fortaleza, Huarmey, Casma, and Nepeina. To-
Callej6n, the Huaraz River cuts through the day an automobile road has been opened through
Cordillera Negra forming the fantastic Cano6n the La Fortaleza Valley. In the past consider-
de Pato, after which it is joined by the Chuqui- able mutual influence existed between the coast
cara River to form the coastal Santa. The and the Callej6n, although not to the extent of
intermontane Huaraz and the coastal Santa are wiping out the local character of each area.
both frequently designated by the term Santa. The mountain ranges which flank the Calle-
As such, the Santa forms the largest river on the j6n de Huaylas are rich in minerals, and numer-
west coast of Peru, rivaled only by the Chira in ous mines are operated in the search for gold,
the north. Kroeber1 quotes Adams' statistics silver, copper, mercury, iron, lead, sulphur,
indicating a basin for the Santa of 10,500 square coal, and salt. The principal occupation, how-
kilometers, of which only 800 square kilometers ever, is, and long has been, farming. The
refer to the coastal section. In spite of the large numerous tributaries to the Huaraz River make
run off, estimated as averaging 5093 million water abundant in all parts. The area is within
cubic meters per year, the river is not navigable. the direct rain belt and besides has considerable
Likewise its coastal and intermontane sections underground water seepage which makes irri-
are so sharply divided that two distinct regions gation unnecessary. The intermontane charac-
are formed. Archaeologically, the coastal Santa ter of the valley produces rich soils. In general
must be treated with other north coast valleys, the region is so broken up by foothills and
while the Callej6n de Huaylas forms a distinct streams that the average farms are small. To-
unit. day small haciendas and some comunidades are
The Callej6n is effectively shut off from the the rule, and the archaeological evidence indi-
rest of Peru by its two flanking mountain cates a similar concentration in small village
ranges. The Cordillera Blanca along the eastern units in the past. Population centers are found
side is composed of a long series of snow- at regular intervals in the Callej6n. The princi-
covered mountain peaks, many of which rise pal central towns from north to south are
well above 6000 meters in altitude. Some of the Huaylas, Caras, Carhuaz, Yungay, Huaraz
outstanding peaks are Ranropalca, Copa, (the capital of the Department), and Recuay.
Huascaran, Chopicalqui, Contrahierbas, Hual- Farther south the swampy wind-blown Pampa
c6.n, Huandoy, Aguja Nevada, Artesonraju, de Lampas supports some grazing, but little
Santa Cruz, and Chacraraju. Borchers2 gives an farming. According to Raimondi (1873) the
excellent description of these, and also includes ruins of rather good-sized temples were found
the best map of the Callej6n de Huaylas. The in each of the principal modern towns, so that
Cordillera Blanca is crossed by mule trails one gathers that the population centers were
above snow line. At all times it has formed an about the same in the past.
effective barrier, although more noticeable in The work of this expedition was limited to
terms of modern transportation than in the past. the Provinces of Huaraz and Huari in the De-
The Cordillera Negra along the west side of partment of Ancash. The principal work was
the Callej6n de Huaylas is less majestic than around the town of Huaraz, with more limited
the Cordillera Blanca, although still a formid- exploration near Carhuaz and Recuay. The work
able range. Peaks rise to 4853 meters and most in Huari was limited to the ruins of Chavin
I Kroeber, 1930, 76. de Huaintar, reached by crossing the Cordillera
' Borchers, 1935. Blanca by mule from the town of Olleros.
11
HUARAZ SITES
In the following pages are listed the sites 1 Black, brown, yellow, white-on-red
which we examined in the region of Huaraz. 1 Black, yellow-on-red with squared face and
The list is by no means a complete survey of trident headdress
sites in this area. Reports on other sites both 4 Pieces of round-bottom cups
here and in the whole Callej6n de Huaylas may 1 Black, white, gray-on-red geometric design
3 Black, white, gray-on-red with circle plus four
be found by consulting Tello,' Raimondi,2 the attached hooks
chapter by Hans Kinzl in Borchers, and Mid- 1 Black-on-red piece of an angular-bodied open-
dendorf.4 On the basis of our investigation some bowl
of the sites merit no more than this listing.
Others, at which we did more extensive excava- The designs and shapes of these fragments cor-
tion and survey, are included here for complete. respond closely to the style found by Uhle at
ness, although the detailed descriptions are re- Pachacamac and labeled "Epigone" by him.
served for the sections following. Field numbers Complete pieces of this same style may be seen
of some of the sites are included. in collections from the Callej6n de Huaylas, and
1. Pomakayan. In the northern part of one assumes that isolation of the type should be
Huaraz itself are the remains of what Tello' possible.
has described as a large, terraced, truncated 3. Akobichai. On a hill slightly north of
pyramid with interior stone-lined galleries. Huaraz is a stone-lined and covered under-
Raimondis reported dressed stones as part of ground gallery about 8 meters long.
the facing and the excavation of a cut-out stone 4. Wakrajirka. On a ridge near No. 3 are a
bathtub. He also mentions that statues were number of rough stone walls which may once
not found here, in spite of a popular belief to the have outlined some structure.
contrary. The pyramid is now badly destroyed, 5. Wancha. In the same general region of the
and we attempted no more than a cursory ex- above sites are two large upright stones and
amination. some miscellaneous walls.
2. Patay Katak (Site 3H). In a section at the 6. Kepijirka. Another meandering series of
northern edge of Huaraz itself some 27 mounds walls with much recent displacement is found in
have been reported, although most of them are the region of the above sites.
now hard to distinguish. One such mound con- 7. Shankaiyan (Site 1H). On a hill to the
tains a stone-lined and once covered pit about 3 north of Huaraz on the property of Sr. David
meters long, 1.5 meters wide, and 2 meters deep. Alvarado a number of subterranean stone-lined
The walls are of rough stones set in rows. An- and covered galleries have been excavated.
other mound contains stones which have been Three days' excavation allowed us to complete
roughly dressed, and a doorway with a stone the uncovering of one gallery which had already
lintel. These mounds have long since been been opened, discover and excavate a new one,
opened, and further work at the site did not and find two stone-lined box graves. The collec-
seem practical. A collection of sherds in the tion of six complete vessels, 620 sherds, and
Huaraz museum is said to have come from this eight copper fragments is described later as a
site. Several styles are represented, among them sample of Recuay style.
a fairly good Coast Tiahuanaco including: 8. Kekamarka (Site 4H). East of Huaraz on
a ridge between the Rio Hauqui and Rio Quill.
9 Pieces of straight-sided "kero" goblets cay is a rocky outcrop with several natural flat
5 Black, white-on-red with trident and other de- terraces augmented by artificial stone terraces.
signs The ruins are extensive and consist of terraces,
2 Black geometric design with white outline on a
red base walls, and some delimited rooms, rectangular in
shape and about 7 by 2 meters or smaller. The
1 Tello, 1929, 1930. walls are made of double facings of stones filled
2 Raimondi, 1873.
3 Borchers, 1935.
and chinked with smaller stones. Uprights are
4 Middendorf, 1893, vol. 3. not common in the wall construction. A large
6 TeLlo, 1930, 273. granite block, 2.5 meters long, 1 meter wide,
6 Raimondi, 1873, 40-41. and 50 centimeters thick, has a relief figure cut
1944 BENNETT: THE NORTH HIGHLANDS OF PERU 13
on its convex face. The figure represents a male type of remains. For convenience, a r6sume is
with raised hands of five fingers each, and sepa- included here of the actual excavation and ex-
rated legs with five-toed feet. Features are in ploration in each of the five adjoining sections.
relief, the mouth being rectangular with incised Wilkawain is the name which Tello3 has given
teeth. Roosevelt' mentions these ruins and to the three-storied stone temple which is the
illustrates the carved figure. Two other small outstanding construction of the region. Meas-
statues are found in the ruins, but without urements and plans of this temple were made.
definite association. Sherds of plain red and Slightly to the east of the temple are many
orange occur on the surface. The general ap- houses of one and two stories. Six of these were
pearance, except for the statues, is that of Inca measured and one excavated (Site 6H-2), re-
construction. sulting in the discovery of a grave with three
9. Ayapampa (Site SH). About 5 miles east bowls, some 441 sherds, seven spoon fragments,
of Huaraz, on the same ridge as Kekamarka a stone amulet, and two copper pieces. Many
(Site 4H) are numerous stone houses, uprights, natural boulders are found on the slopes of the
rows of stones, and terraces. Twenty-five of the hills, and under some of these boulders are
houses were examined, ten measured in detail, graves of stone box type. Four test cuts under
and one excavated. A burial with one vase and boulders uncovered three graves (Sites 6H-3
five toy vessels was found. The houses and the and 6H-5) containing a total of nine vessels.
excavation are described later in the text. Two small test pits in the region (Sites 6H-1
10. San Jer6nimo (Site 2H). Just over the and 6H-4) furnished a few unimportant sherds.
bridge which crosses the Santa River in Huaraz 13. Ichik-Wilkawain (Site 7H). Slightly
is a rocky promontory with remains of crude southeast of the previous region and adjacent
stone walls. The hill is covered with sherds, to it is another area with numerous archaeo-
many of which have washed down to the road logical features. A three-storied temple differing
below. A collection of 123 surface sherds from from Wilkawain temple in size and arrange-
this site are described in the text. All appear to ment of interior rooms is the dominant struc-
be Inca style. ture in the section. A plan of this was made.
11. Orojirka. In the hills to the west of the On a terrace below the temple four deep stone-
Santa River and slightly north of Huaraz is the lined graves were found (Sites 7H-1, 2, 3 and 4).
hacienda of Sr. Pando. It lies just below the The pottery found on the floors of these graves
hacienda of Pongor from which came many of had evidently been broken even before the
the statues now in the Huaraz museum.2 Long, graves had been filled and the slab covers put
stone-lined, and covered subterranean galleries in place. The result is a large collection of some
have been discovered here. One is over 4 meters 2208 fragments of clay vessels, 27 complete
long, 1.2 meters wide, and about 90 centimeters bowls, 25 toy vessels, and four clay spoons. The
high. We photographed four stone statues and sherds may eventually result in some 72 more
measured a previously opened circular tomb 1.5 restorable vessels. In addition to pottery, seven
meters in diameter. We excavated test pits in flint and obsidian lance points, four copper
several parts of this hill, but discovered only pieces, and six miscellaneous stone artifacts
one stone box tomb, 35 by 30 centimeters on the were found.
sides and 70 centimeters deep. It was lined with One house site excavated (Site 7H-5A) con-
four upright slabs and covered by another. tained one olla, a figurine, seven toy vessels, and
Adult bones were the only content. 148 sherds, some of which are partially restora-
12. Wilkawain Region (Site 6H). About 13 ble vessels.
kilometers north by 16 degrees east of Huaraz Seven stone-lined and covered subterranean
lies a section with numerous archaeological re- galleries were seen (Sites 7H-5B, 8, 10, 12, 13,
mains which we excavated extensively. Actually 14, 16). While some of these had apparently
this region and the four following (Nos. 13-16) been opened before, two (Sites 7H-12, 13) were
all form a continuous area which will be treated definitely intact, and all yielded at least some
as one in the text, subdivided according to the sherd material. The material from all the gal-
leries is not great in quantity and is largely
1 Roosevelt, 1935, 33, 36; Figs. 17, 18.
2 Raimondi, 1873, 40-41. 3Tello, 1929, 44; 1930, 272.
fragmentary. In total we found three complete vessels, ten sherds, a clay, and a stone whorl.
vessels as well as 136 sherds, two copper pieces, 15. Region below Ichik-Wilkawain (Site 9H).
two shells, and a stone bead. All belong to the On the hill below the temple of Ichik-Wil-
Recuay style. kawain are two large house sites under im-
Again stone box tombs were sought under mense flat surface stones. The first (Site 9H-1)
natural boulders. Ten were found (Sites 7H-6, contained only 39 sherds. The other (Site 9H-2)
7, 9, 11), all but two without artifacts. Of the was filled with refuse material of a mixed type,
two, one (Site 7H-7) contained a complete including four complete vessels, 829 sherds, one
bowl, and the other (Site 7H-6) contained complete spoon, six stone beads, and four cop-
twelve beads, three silver fragments, and a per fragments.
piece of copper. Finally, under a large boulder, 16. Irwa (Site 10H). Across a stream to the
we found a tomb (Site 7H-15), not of the box east of Ichik-Wilkawain two galleries were ex-
type, which contained thirteen vessels of a cavated. These contained Recuay-style ma-
white-on-red style. terial, including three complete vessels, 131
14. Region above Ichik-Wilkawain (Site 8H). sherds, a stone bead, and a copper pin head.
On the hill above Ichik-Wilkawain temple In the excavation descriptions which follow,
two tombs and some slight refuse material Sites 12-16 will be treated as a single area of the
were found. The collection contains two Wilkawain region, and Sites 7, 9, and 10 will be
complete vessels, two partially complete described in more detail.
THE WILKAWAIN SECTION

The numerous archaeological remains in the A. TEMPLES
general vicinity of Wilkawain are not all con- WILKAWAIN
temporaneous. None the less, since the terri-
tory is not large, the whole section must have The most outstanding building in the region
formed a unit at any given period. For field is this three-storied stone structure which rises
convenience in excavating, the area was divided to a total height of 9.25 meters and which, for
into five sections (Sites 12-16), but for the pur- descriptive purposes, may well be called a tem-
poses of description a grouping based on type ple (PI. la). It stands on a flat platform in part
remains seems practical. Later a r6sum6 in natural and in part artificial, some 54 by 35
terms of the time periods suggested can be meters in extent. Remains of walls surround
added. The materials will be described then in three sides of this area, although many changes
the following order: have been made in recent years, and the fourth
side is a stone-faced terrace. Today this plat-
a. Temples: 2 form or court is filled with eucalyptus trees
Wilkawain (Site 611) which cut off the view, but previously the tem-
Ichik-Wilkawain (Site 7H) ple site must have commanded a good section of
b. Above-ground house sites: 16 plus the Huaraz region. To the west of the temple
Excavation of one house (Site 6H-2) section the slope is covered with old terraces
c. Deep stone-lined tombs: 4 and house foundations, the same being true of
(Sites 7H-1, 2, 3, 4) the ridge which continues to the east. To the
d. Stone box graves: 13 north and south the slope is sharper, although
(Sites 6H-3, 5; 711-6, 7, 9, 11) even today it can be planted with potatoes.
e. Unlined graves: 3
(Sites 7H-15; 8H-1, 2) The temple consists of a three-storied cen-
f. Subterranean house sites: 2 tral building, 10.7 by 15.6 meters in groundplan
(Sites 9H-1, 2) and 9.25 meters high (Fig. 2). This central core
g. Subterranean galleries: 9 is surrounded by a terrace, 2 meters high and
(Sites 7H-5B, 8, 10, 12, 13, 14, 16; 10H1-, 2) from 2.1 to 2.3 meters wide, which is well
h. House-gallery stratification preserved on the north and west sides, but badly
(Sites 7H-5A, 5B) fallen on the east and south. The terrace was
0, I X.
FIG. 2. Groundplan of Wilkawain three-storied temple.

probably built after the central building was meters. To the south are two rectangular
completed, since the ground floor rooms not gallery-like rooms, each 6.8 by 1.15 meters,
only do not extend into it, but actually have separated by a 90-centimeter wall. At each of
thick side walls independent of it. Along the the four corners of the building is a rectangular
well-preserved northern side the terrace has room about 3.1 by 1.8 meters. Due to the ex-
three T-shaped niches, more or less evenly tent of the stone fill, we were unable to enter
spaced. The entrance to each niche is a regular two of these corner rooms on the first floor, but
doorway with side blocks and lintel. One door- the groundplan indicates that they exist.
>.22 5
z~~~~
:NICIIE
FIG. 3. Cross-sectioz n off Wilkawain temple.

way measures 1.15 meters wide, 92 centi- Along the northern edge of the building, from
meters high, and 78 centimeters thick. The the first to the third floor, is a narrow ventilat-
niche proper is 75 centimeters deep and 2.11 ing shaft about 43 centimeters wide. Vents to
meters wide. The other two niches vary slightly this shaft were seen in the large north-central
from these measurements, but are essentially room on the first and third floors and in the
the same. Although not visible today, it is quite northeast corner room of the third floor. Prob-
possible that the south terrace also had three ably at one time all the rooms were served by
similar niches. this shaft. Even today fresh air circulates
Each floor of the central building has seven throughout the building.
rooms. Their arrangement can best be appre- Entrance to the ground floor is at the west
ciated from the plan. Measurements by candle- end, through the surrounding terrace. The sec-
light in rooms partially filled with stones are ond floor is entered from the south (P1. lb), and
none too accurate. Nevertheless the sym- the top floor from the east. No communication
metrical plan of the building is obvious. The from floor to floor was seen on the inside of the
arrangement of rooms on each floor is identical, building, and the ventilating shaft certainly
the only variation being the communicating could not have served this purpose. Likewise
doorways. Thus a general description is possi- windows between rooms were found in only two
ble. A central wall about 1.1 meters thick runs places on the top floor. They measured 50 by 55
the full length and height of the building centimeters.
(Fig. 3). To the north of this wall is a long The first and second floors are about 2 meters
rectangular room measuring about 6.8 by 2.25 high. Great slabs form the roof of each room
and likewise serve as the floor of the room of a central building of three floors, 16 by 7.2
above. The roof of the third floor is different. meters surrounded by a terrace about 3 meters
Here the central wall rises to a height of 2.95 wide and 1.5 meters high. The long axis of the
meters, while the outer wall of the rooms is only building runs roughly east to west, although not
2.08 meters high. Great slabs slant from the oriented, and in general faces southward. Down
central to the side walls, forming a gabled roof. the slope in front are terraces, graves, and house
On the outside dirt and stones are heaped up sites to be described later. Behind the temple
above the peaked roof, producing a dome-like are some ten small ruins, probably house sites,
appearance. the largest of which measures 9.15 by 6.5 meters
The inner and outer walls of the temple are and has four interior rooms.
built up of horizontal rows of large stones, al- The ground floor of the temple has an en-
ternating with rows of small stones. Likewise trance through the terrace on the south side,
small stones are used to fill all the chinks be- but it was so blocked with stone as to prevent
tween the larger ones. In some places remains entry. The second floor is composed of three
of clay between stones can be seen. The ma- sets of two connected rooms, 2.2 by 3.5 meters,
sonry cannot be called coursed, but there is a with doorways on both the north and south
definite effort to maintain roughly horizontal sides of the building. All doors have stone lintels
rows. Although some of the stones may have and side blocks. Inside walls have stones which
been split or cut for fitting, they are not dressed. project some 20 centimeters and inset niches
The edge of the roof is formed by a series of 36 by 38 centimeters, and 26 centimeters deep.
projecting flat stones weighted down at the in- The eastern set of rooms has a total of five
ner end. Underneath the eaves thus formed an niches and fourteen projecting stones. The top
inset niche, 45 centimeters deep and 55 centi- floor has six rooms, today barely di-stinguishable
meters wide, encircles the building. Below this by their remaining low walls.
decorative niche-band are regularly spaced holes In general the stone masonry is of the same
from which carved stone puma heads are said type construction as that of Wilkawain, with
to have been extracted. None remains in posi- horizontal rows of large and small stones. The
tion today, but at his house near the temple inner walls are finished in the same fashion with
Sr. Miguel Peinaranda has one such carved head smaller stones. Good evidence of clay cement is
which he says came from this site. This head has found on the inside. The roof of the second floor
a projecting tenon at the back for insertion in is composed of large stone slabs.
the wall. Furthermore, Dr. Soriano Infante has
a photograph of a somewhat similar ruin called B. ABOVE-GROUND HoUSE SITES
Catihama, near Caras, with the cat heads still Mention has already been made of house sites
in position. on the terraces around Wilkawain temple and
Excavation within the temple seemed im. behind Ichik-Wilkawain temple. These and
practical. Some of the rooms are literally filled other houses in the area are variations of but
with stones, a condition suggestive of inten- one style, differing in such details as floors,
tional obstruction, since no such accumulation niches, and the number of rooms. The construc-
would be possible with the walls and ceilings tion is in general similar to that of the temples,
still intact. but less carefully done. The use of large stones
ICHIK-WILKAWAIN
with smaller ones filling chinks and roofing of
large stone slabs is the technique used through.
About one-half kilometer to the southeast of out. The stones in the house sites seem even less
Wilkawain is another temple of somewhat simi- worked than those in the temples. They are
lar construction, but on the whole smaller and somewhat selected, but certainly neither split
less complex. Its present appearance is affected nor dressed. Also less attention is paid to main-
by the fact that the third floor is largely de- taining the horizontal rows.
stroyed, so that only the base foundation of the Six house sites on the slope to the east of
rooms may be seen. The surrounding stones, Wilkawain temple were examined in detail and
however, do not suggest the type of peaked roof one was excavated. Another house site (Site
found at the first temple. 7H-5A) was excavated, but the description is
This temple, like the previous one, consists reserved for special discussion later. In all cases
dwellings consist of single-unit buildings sepa- the refuse material from this site can be treated
rated from each other by some 7 to 30 meters as a unit and described in terms of wares.
and scattered between walls and terraces.
House 1 (excavated). MATERIALS FROM THE GRAVE AT HOUSE
House 2. A two-story building has an upper (SITE 6H-2A)
floor, 4.6 by 7 meters with a lintel-covered door- The three clay vessels which accompanied the
way to the south, 90 centimeters wide. The burial consist of a pair of small redware double-
walls are 90 centimeters thick and still stand spout vessels with rounded bridges (Fig. 4a) and
over 1.5 meters high. The large roofing slabs a double whistling jar. The containers of the
have caved in. This second floor rests on a ter- double-spout vessels measure 6 centimeters high
race covering an area of 8 by 9.6 meters and by 5 centimeters in diameter, and the spouts
1.5 meters high. A doorway on the east enters are slightly tapering. The double jar, half of
the ground floor through the terrace. which was badly broken, is decorated in a two-
House 3. This one-room building measures color (black-on-orange) negative design of rows
2.1 by 2.65 meters. of lines and dots (Fig. 4b). One half is a plain
House 4. A badly destroyed building meas- black container with a tapering spout. This is
ures 3.1 by 5.4 meters. connected by a large flat bridge, and a smaller
House 5. A one-room house measures 2.3 by round one, to a modeled seated figure.
3.45 meters with walls still 2.5 meters high. A
lintel-covered door to the south is 65 centi- REFUSE MATERIAL FROM HOUSE 1 (SITE 6H-2)
meters wide. The fallen roof seems to have been Plainware
a corbeled dome of stones covered with dirt.
House 6. The foundations of this house meas- Orangeware, 212 fragments. All plainware
ure 3 by 3 meters. fragments are orange, and include the following
varieties:
EXCAVATION OF HOUSE 1 (SITE 6H-2) 27 Thick orange fragments from ollas with flare-
House 1 measures 4 by 3 meters on the out- rims and globular bodies. No handles are found.
112 Thin orange fragments of which 95 are body
side, with walls from 60 to 80 centimeters thick, pieces, nine are from flat bases of rounded-sided ves-
leaving an inside measurement of 2.52 by 1.78 sels, seven are from straight-rimmed vessels suggest-
meters. The walls are made of large and small ing goblets, and one is from a constricted bulge-collar
stones, plus clay cement. The inside was com- vessel
pletely filled with dirt and stones, and the large 35 Pieces of shallow plates with annular bases
roof-slabs, while partially fallen, still topped the (Fig. 4f)
pile. When cleared, the height of the room on 14 Pieces of crude orange toy vessels
the north side was 2.9 meters, but no doorway 22 Pieces of shallow orange plates pierced with
was found. Likewise, the fill of the inside of the sieve-like holes. One is complete and measures 5.5
centimeters in diameter.
house contained no artifacts of any kind. Con- 1 Piece appears to be the base of a conical handle
siderable dirt had accumulated outside of the 1 Thick orange piece has a relief face
house, and excavation here produced many
refuse sherds. On the south and west sides a Polished Monochrome Wares
low enclosure wall about 30 centimeters wide Blackware, 32 fragments. This is well-fired,
leaves a passage some 78 to 82 centimeters wide polished blackware including several fragments
along the side of the house. An entrance through from the same vessels. The pressed-relief design
this wall is on the center of the west side. This is interesting as a possible diagnostic.
passage between the enclosure and house wall
was filled with dirt and sherds, and next to the 5 Pieces of one double-spout vessel with a zoo-
southwest corner of the house was an intrusive morphic face engraved with broad grooves (Fig. 4i)
burial. This was a direct burial in an unprepared 5 Pieces of a plain double-spout vessel
12 Pieces of other double-spout vessels, including
grave (Site 6H-2A), and while doubtless some- three spouts, a round bridge, and a ball whistle
what later than the fill, the material is not very 8 Pieces with pressed design, including a bird and
distinctive. Three vessels accompanied the bur- a figure holding a spear-thrower and darts (Fig.
ial. For comparison with other sites in the area 4h, k)
FIG. 4. Ceramic styles from above-ground house site, Wilkawain.

2 Modeled pieces, one of two hands holding a Miscellaneous Paintedwares
shell; the other a human head with seven pierced Forty-one pieces are grouped together and
holes (Fig. 4g, j)
can best be described in tabular form:
Redware, a fragment of a taper spout. Other 7 Fragments of black-on-orange flat-handled
pieces of polished redware are parts of painted spoons
vessels. Polished redware is common in other 4 Pieces of one small open-bowl with flare-rim
sites of the region and its absence here is sur- and two body nubbins, painted in purple-on-orange
prising. 1 Fragment of small open-bowl with flare-rim,
black-on-brown
Negativewares 2 Fragments of black-on-orange taper spouts
No negative-painted piece, either two- or with flat bridges
1 Fragment of white-on-orange taper spout with
three-color, is found in the sherd collection. The flat bridge
piece in the intrusive grave has already been 4 Pieces of one black, white-on-red constricted-
mentioned. collar vase with flare-rim and triangle and hook de-
sign
Tiahuanacoid Wares 11 Pieces of one black, white-on-orange constricted
Thick black, white-on-redware, 118 frag- bulge-collar vessel with band and circle design
ments. All of these pieces come from open 11 Miscellaneous fragments of spouts and bridges
straight-sided bowls with plain straight rims,
and flat bases (Fig. 4c). These thick U-shaped Copper Fragments
bowls have been found associated with Coast
Tiahuanaco style at a number of sites, notably The only metal fragments were the pierced
Pacheco. Only 27 of these fragments have flat head of a pin and the stem of another.
painted design and none is definitely Tiahuana-
coid. Four are simple step designs in black out- Stone Artifacts
line filled with white; two have black circles
filled with white; and 21 are band or linear de- A rectangular small flat stone, pierced as if
signs. The division is as follows: for a pendant, was the. only piece found.
A tabular resume of the refuse material fol-
23 Straight-rim pieces; 11 plain and 12 painted lows, first for suggestions of shapes of vessels
73 Body fragments; 58 plain and 15 painted and then for type painting.
22 Plain heavy flat base pieces
Polychrome tripods, 24 fragments. These No. Per Cent
pieces probably come from about five vessels. Olla. 28 6.25
All are shallow plates with simple short conical Double-spout. 22 4.91
tripod legs (Fig. 4d, e). The design on the in- Double-bowls . 19 4.24
side consists of alternating perpendicular and Spouted vessels . . 15 3.35
wavy lines in black, yellow, gray-on-red, or Constricted flare-collar . 4 .89
black, gray-on-red. The identification as Tia- Constricted bulge-collar 12 2.68
huanacoid is tentative. U-shaped thick bowl . 118 26.34
Modeledwares, 20 fragments. Modeled pumas Pedestal plates . . 35 7.81
with painted design are fairly characteristic of Bowls with flare-rims. 5 1.12
some Tiahuanacoid collections and thus are in- Sieves. 22 4.91
cluded here. The 20 pieces can be described as Toys . 14 3.12
follows: Modeled vessels . . . 3 .67
Goblet shapes (?) . 7 1.56
1 Piece of a modeled puma head with black, Tripod plates . 24 5.36
white-on-red design Spoons . 7 1.56
19 Pieces of one double-bowl, one half red, the Conical handle (?) . . 1 .23
other painted in black, white-on-orange and modeled Shape undetermined . 112 25.00
as a jaguar or puma. Pieces of the spotted body,
feet, ears, and eyes are included. Totals.
. . 448 100.00
No. Per Cent Totals Per Cent

Plain orangeware . . . . . . . . . . . . . . . 212 47.32
Orange vessels . . . ... . . . . . . . . . . 176 39.29
Sieves . . . . . . . . . . . . . . . . . . 22 4.91
Toys . . . . . . . . . . . . . . . . . . . 14 3.12
Monochrome polished . . . . . . . . . . . . . 33 7.37
Blackware . . . . . . . . . . . . . . . . . 32 7.14
Redware . . . . . . . . . . . . . . . . . 1 0.23
Paintedwares . . . . . . . . . . . . . . . . 203 45.31
One-color . . . . . . . . . . . . . . . . 5.80
. 26
(Black-on-orange, 9) . . . . . . . . . . (2.01)
.
(Purple-on-orange, 4) . . . . . . . . . . (0.89)
.
(Black-on-brown, 1) . .(0.22)
(Miscellaneous, 12) ..(2.68)
Two-color . . . . . . . . . . . . . . . . . 167 37.28
(Thick black, white-on-red, 118) . . . . . . (26.34)
(Black, white-on-red, 5) . . . . . . . . . . (1.12)
(Black, white-on-orange, 30) . . . . . . . . (6.70)
(Black, gray-on-red, 14) . . . . . . . . . . (3.12)
Three-color . . . . . . . . . . . . . . . . 10 (2.23)
(Black, yellow, gray-on-red, 10) . . . . . . (2.23)
Totals . . . . . . . . . . . . . . . . . . 448 448 100.00
Further analysis of this site is reserved for a large stone slabs. The floor is hard-packed clay.
later section in order to make use of compara- Most of the artifacts were found within 20
tive materials. The slight suggestion of stratig- centimeters of the floor in spite of the fact that
raphy in the intrusive grave cannot be stressed a fill of dirt and stones extended almost to the
because of its small content. It is interesting to covering slabs. Since the stone-lined walls and
note that the grave contained both redware and the covering slabs were in undisturbed condi-
two-color negative, neither of which appears in tion, it is obvious that this dirt and large stone
the refuse collection. fill was intentional, not accidental. No bones
were found, probably because of the heavy
C. DEEP STONE-LINED TOMBS water seepage in the tombs. Traces of ash were
Some 6 meters in front, or roughly south, of found on the tomb floor.
the temple of Ichik-Wilkawain is a terrace wall. A few complete vessels were encountered.
Immediately in front of this terrace, cover- Most, however, were badly broken, and the
slabs of four stone-lined tombs (Sites 7H-1, 2, 3, pieces of a single vessel were scattered over the
4) were exposed on the surface. In construction full length of the floor, rather than in one place.
the four tombs form but two units, as the fol- This suggests that the pottery was perhaps
lowing description of the paired graves (7H-1 thrown in from the top and thus broken, or
and 7H-3) will show (Fig. 5; PI. ic, d). shattered by the stones of the fill as they were
The two tombs in the set are rectangular and thrown in. While many vessels have been re-
parallel to each other, separated by a common stored, it is none the less clear that fragments
central wall, some 75 centimeters thick, which of previously broken vessels also found their
also serves to support the covering stones. way into the grave. This is particularly notice-
Both tombs are 1.12 centimeters wide, although able in the polished blackware pieces which are
7H-1 is 3.55 meters long and 7H-3 only 3.32. easily isolated and is also true of modeled puma
Due to a slight rise in the ground surface one is vessels, none of which has proved restorable.
about 2.5 meters deep and the other 2.6. The Thus the ceramic collection presents difficulties
walls are lined with large and small stones set in of description and analysis, since it can be
clay cement. Each tomb is covered with four treated neither as pure grave material nor as
22 ANTHROPOLOGICAL PAPERS AMERICAN MUSEUM OF NATURAL HISTORY VOl. 39
7
#3
A/
FIG. 5. Groundplan and cross-section of deep stone-lined tombs, Wilkawain.
refuse. A grouping by outstanding wares is One thousand twenty-nine fragments of ollas.

again followed, by which means both complete The commonest shape is a globular olla with
vessels and sherd material can be described at flare-rim and two horizontal flat-loop handles.
once. Forty-six handle pieces were found. Other olla
The first pair of graves, 7H-1 and 3, was by shapes are represented, including a plain verti-
far the richest in material, and forms a unit on cal rim; a flare-rim with bulge below; and a
typological grounds. The materials from the thicker rim of an open olla with slightly con-
other pair, while presenting somewhat similar stricted mouth. Pieces of large vessels with con-
styles, were relatively poor and far more frag- stricted straight high collars are included here.
mentary. There is no sound basis, however, for Most fragments are of medium thickness, but
distinguishing one set of graves from the other, 29 are extra thick.
and consequently the collection will be de- 5 Pieces of shallow plates with annular bases
scribed as a whole. Whether the graves should 5 Pieces of straight-sided goblets, and one com-
be associated with the temples is unanswerable plete goblet with a raised band (Fig. 7e)
in definite terms. The location in front of the 2 Fragments of flare-collars with faces, one in re-
temple of Ichik-Wilkawain and the general na- lief, the other incised
ture of the stonework suggest a relationship, but 2 Incised fragments
this is by no means certain. Search was made for 1 Fragment of a spout-and-bridge vessel
other deep graves of this type, but without suc- 9 Pieces of sieves, and one complete sieve, measur-
cess. ing 8.3 centimeters in diameter and 1 centimeter
deep (Fig. hlf)
MATERIALS FROM DEEP STONE-LINED TOMBS
Brownware, 94 fragments. The olla fragments
Plainware included here might well be placed with the
Orangeware, 2 complete, 1053 fragments. orangewares. Fragments from other vessels,
Plain orange pieces were common in all four however, show a better finish and a distinct
tombs. Most of these are from ollas, although brown color.
reconstruction of vessels is difficult. An esti- 13 Pieces of one flare-rim, globular-bodied olla
mate of the possible number of vessels repre- 48 Pieces of large vessels with constricted high
sented is impossible at this time. The orangeware collar
is somewhat crude and crumbly. Some of the 14 Pieces of two shallow plates
fragments have been burnt black by use. 8 Pieces of a flask-shaped vessel of thick ware
7 Pieces of one vessel with relief legs 3 Fragments of one vessel with globular body and
2 Pieces from goblets with raised bands slightly flaring collar (Fig. 6a)
2 Pieces of modeled faces 3 Complete double-spout bowls of small size (Fig.
6i). The bodies are round, and the largest is 9 centi-
Whiteware, 6 fragments. meters in diameter and 6.5 centimeters high. The
short spouts (4 to 5 centimeters long) have a raised
ridge around their bases and are connected by a
Polished Monochrome Wares round bridge.
Blackware, 14 complete, 134 fragments. Pol- 8 Pieces of one similar vessel with a simple relief
ished blackware occurs in all four tombs, al- design on the upper part of the body, and two spouts
with rounded bridge
though most of it was found in Sites 7H-1 and 12 Fragments of similar double-spout bowls with
7H-3, and is usually well polished, thin, and round bridges
well fired. While some of the shapes are sugges- 1 Complete double-spout vessel representing a
tive of the Late Chimu style, a number of curved animal with two heads (P1. 2c). The tapering
features distinguish it. Most of the double- spouts are connected by a flat bridge. The animal
spouts have round bridges; no pressed-relief base is about 5 centimeters wide and thick and is
design is represented; no stipple design is found. modeled in a curve so that the total length is 17
Likewise, except for two simple incised crosses, centimeters.
incision is not used. The shapes and designs of 1 Complete double-bowl with two simple globu-
blackware distinguish it completely from the lar containers. One has a taper spout connected by a
Chavin style. The following descriptive list in- flat bridge to the modeled bird whistle on the top of
the other (Fig. 6g).
dicates the variety. 9 Pieces represent modeled vessels. Two are mod-
5 Complete shallow plates. Three are from 1.5 to eled animal legs, 3 relief legs, 3 animal heads, and
2 centimeters deep and from 6.5 to 7.5 centimeters in one a face lug.
diameter and without decoration (Fig. 6h). Two are 59 Miscellaneous body and base fragments
slightly deeper plates, about 2.5 centimeters deep,
and 12 centimeters in diameter. Redware, 4 complete, 460 fragments. Mono.
12 Fragments of one open-bowl with rounded base chrome redware pieces occur in sufficient quan-
and an incised cross on the side tity to constitute a definite type in the grave
5 Pieces of an open-cup with small vertical loop collections. Most of the ware is covered com-
handle pletely with a red slip or paint. While many of
1 Complete goblet with straight, slightly diverg-
ing sides and a raised band below the plain rim. It is the painted pieces have a solid red base, here
9 centimeters high, 9 centimeters in diameter at the the red coloring is the only decoration. In re-
rim, and 7 centimeters at the base. view, only two shapes are represented with any
5 Fragments of similar goblets (Fig. 6c) frequency: a constricted high collar jar and a
1 Complete goblet without a raised band and shallow plate. Incised or appliqu6 decoration is
with slightly bulging sides is 9 centimeters high, 8 absent, and modeling is present in only one
centimeters at the rim, and 5.5 at the base (Fig. 6d; piece. A r6sum6 list of the redware follows:
P1. 2b).
1 Complete angular-bodied bowl with slightly 1 Complete modeled monkey standing on four
constricted mouth and no rim. It measures 8.5 centi- legs with head twisted to one side and tail curled up
meters high, 9 centimeters wide, and 4 centimeters at (Fig. 7a). A short, cylindrical-collar spout projects
the mouth. from the back of the animal.
2 Fragments of similar vessels (Fig. 6b) 2 Complete constricted-collar jars with flat
1 Complete angular-bodied vase with constricted bases and globular bodies (Fig. 7d). One is 12 centi-
tapering collar with a vertical loop handle at the meters in diameter and 9 centimeters high for the
base of the collar (Fig. 6e). It measures 9.5 centime- body, with a straight collar 5 centimeters high and 7
ters high, 9.5 centimeters diameter at the body angle, centimeters in diameter. The other is slightly larger,
has a flat base and a taper collar 4 centimeters high. with a slight flare to the collar.
8 Fragments of a similar vase 382 Fragments of similar constricted-collar jars.
10 Fragments of one globular-bodied vase with Straight, slightly flared and bulge-collars are repre-
constricted bulge-collar and a loop handle at the col- sented. Globular bodies and flat bases are the rule,
lar base although one slightly angular body fragment was
1 Fragment of a vase with constricted flare-collar found. A few flat-loop vertical body handles suggest
I
i
I
I
I
PI
B C
A
D F
E
G
FIG. 6. Blackware style of deep stone-lined tombs, Wilkawain.
1944 BENNEfT: THE NORTH HIGHLANDS OF PERU 25
C
B
'<~'
D
FIG. 7. Redware style of deep stone-lined tombs, Wilkawain.

this as a variant feature. Also one has a lug at the Tiahuanacoid Wares
base of the collar. While most fragments seem to rep. A number of pieces can be tentatively identi-
resent vessels about the size of those described fied by shape, modeling, and design as related
above, some 47 pieces come from similar, but larger Middle period Tiahuanacoid styles of the
jars. to the
25 Pieces of four shallow plates, one of which has Coast of Peru. Included here are five complete
a groove around the rim edge vessels and 113 fragments. Tripods are found
1 Complete vertical-sided, flat-based open-bowl
with a plain rim (Fig. 7c). The rim diameter is 12.5
centimeters and the height 4 centimeters. Two small
rim lugs project on either side.
1 Piece of a deep open-bowl with slightly convex
sides
3 Pieces possibly from goblets
22 Globular flare-rim olla fragments with one
horizontal flat-loop handle
6 Base fragments of a bowl with convex body
and perhaps a wide flare-rim (Fig. 7b)
21 Miscellaneous fragments
Negativeware
Two-color negativeware, 2 complete, 207
fragments. Only two-color negativeware is
found and largely in one tomb, 7H-1, although
a few fragments occur in each of the others.
The negative design is commonly superimposed
on a red base, and rarely on an orange base.
With the exception of the two complete plates,
designs are simple lines, wavy lines, and dots.
Neither in shapes nor designs, nor in the addi-
tion of a third color, does this negativeware re-
semble the classic Recuay. In its simplicity this
ware is somewhat like the Gallinazo style, but
the shapes are distinct. Actually only two vessel
shapes have negative design: a flat plate and a
constricted-collar jar.
2 Complete plates identical in size and design
(Fig. 8). They are 26.5 centimeters in diameter and
6 centimeters deep, with a curved bottom. The base
is red and the design on the inside only. The design
in negative consists of a double circle in the center
containing a cross. Between the central circle and the FIG. 8. Negativeware plate from deep
rim are three repeated design units composed of an stone-lined tombs, Wilkawain.
angular scroll flanked on both sides by a step and
scroll. A series of tabs mark the rim design. in the tombs, but none is decorated in a
207 Fragments of constricted high collar jars (Fig. Tiahuanacoid style. It is difficult to identify
lie). Three such jars can probably be reconstructed this material as any specific style of Tiahuana-
from the 163 fragments ofTomb 7H-1; the remaining coid ware now known. In total the comparisons
fragments from other tombs seem to be from the must be with the Peruvian Coast styles rather
same type vessels. All are globular-bodied vessels than the Bolivian Tiahuanaco. In the closing
with straight or slightly flaring, constricted high col- r6sum6 of the tomb contents, an attempt will
lars. One has a horizontal handle on the body and an
oval lug. The design is limited to the collar, the body be made to analyze this style in more detail.
of the vessel being black. The collar designs are Thick black, white-on-redware, 32 fragments.
simple series of alternating vertical lines, wavy lines All of these pieces may represent one vessel with
and dots. a globular body, two horizontal loop body
1944 BENNEIT: THE NORTH HIGHLANDS OF PERU 27
handles, and a constricted face-collar with flare- 1 Complete flask-shaped vessel with a constricted
rim. Seven pieces of the collar show a face with and slightly flaring collar (Fig. 1Oe). The flask is 16
relief features and a band beneath each eye, all centimeters in diameter, 3.4 thick at the flat edge,
painted in black, white-on-red, and of thick and 8.5 centimeters thick in the center. The collar is
clay. The other pieces are also thick and painted 5.5 centimeters high and 3.7 thick at the rim. It is
painted in black, white-on-red in a simple design
in the same colors. The design consists of verti- which quarters the face of the flask.
cal and wavy lines plus bands. While thick 1 Complete angular-bodied vessel with con-
ware, it is not from the same type of vessel as stricted neck and flare-rim. A design in black, white,
that described for the house site, 6H-2. yellow-on-red consists of a double circle with seven
Modeled wares, 31 fragments. All the mod- hook appendages ending in circles (Fig. lOb).
eled pieces are from puma vessels painted in 9 Fragments of a similar vessel have a horizontal
black, white-on-red. None of these is complete, S-design painted in black, gray, yellow-on-red
but the general shape is probably a four-footed 5 Fragments of another angular-bodied vessel
puma vessel with constricted collar, curled tail, have a cross-hatch design in black, white-on-red
1 Restorable double-spout vessel has a muffin-
and modeled head. Fifteen modeled feet (Figs. shaped container and two short straight spouts con-
9c, 10c) were found and one pair of joined feet nected by a round bridge (Fig. lOa). The body is 8.5
in relief. Sections of the curled tails, constricted centimeters in diameter and 7 centimeters high, and
collars, and modeled heads were also found. the spouts are 4.5 centimeters long. A faint cursive
Paintedwares, 5 complete, 50 fragments. With design in black-on-red is still visible.
one exception, all these pieces are painted in 3 Fragments of an open straight-sided bowl of
two or three colors on a red or orange base. common Tiahuanacoid shape have a triangular de-
The colors are common in Tiahuanacoid styles, sign in black, white, brown-on-gray
including black, white, yellow, gray, and brown, 19 Pieces from one vessel are painted in black,
plus the orange and red base color. The designs white, gray-on-red in a faded design like a stylized
puma
are composed of elements suggestive of the Miscellaneous Paintedwares
Tiahuanacoid styles, such as chevrons, puma
heads, S-curves, pumas, and the like. The The remaining 140 painted fragments cannot
shapes are less typical, but not exceptional for at the moment be grouped with any of the pre-
the style. viously mentioned wares. These are listed below
1 Complete face-collar jar with elongated globu- for completeness.
lar body, a flat base, and one vertical loop body 2 Fragments with painted orange design
handle (Fig. 9d). The face is modeled on the collar 49 Black-on-red fragments. Five are from one high
with some skill, has vertical lines on each cheek, and constricted-collar vessel with chevrons around the
a band above the eyebrows. The rim is missing, but rim edge. One is from a similar vessel with broken
probably had a flare. The painted design in black, wavy lines around the rim edge. The rest have linear
gray-on-red is repeated on the front and back of the designs.
upper half of the jar. It consists of a horizontal 14 Black-on-orange fragments. Five are from shal-
S-band with a stylized puma head at each end. The low plates with conical tripod legs. These tripods are
vessel is 27 centimeters in diameter and 22 centime- all from Tomb 7H-2, the other tombs having none.
ters high, including the 7-centimeter collar. 13 Red-on-orange fragments. Twelve are from
1 Fragment of a face-collar vessel with a flare shallow plates, and one is a relief face.
collar and a face with modeled nose and incised 3 Orange-on-gray fragments
features. It is painted in black, gray-on-orange, with 3 White-on-red fragments. One represents a
vertical lines beneath the eyes, and an irregular de- round handle, and two are fragments with chevron
sign above the eyebrows (Fig. 9b). A flat handle is at designs. These are not typical of the so-called white-
the back of the collar. on-red style.
1 Restorable constricted straight-collar jar with 4 Brown-on-white fragments from shallow plates
a globular body, 15 centimeters in diameter, and a 3 Black, white-on-red fragments
round body lug (Fig. lOd). It is painted in black, 10 Black, red-on-orange fragments. One is a piece
red-on-orange on the front and back of the upper of a modeled animal, the others seem to be from one
half of the body. The design consists of a horizontal, round-bodied vessel with a wide flare-rim.
outlined S-curve with two red-filled dots at each 38 Black, red-on-whiteware fragments from shal-
point of the S. low plates
13 Fragments of a similar straight-collar jar, de- 1 Black, white, red-on-orange fragment of a mod-
signed in black, white-on-orange eled collar
Dt 2/-
FIG. 9. Tiahuanaco style of deep stone-lined tombs, Wilkawain.
1944' BENNETT: THE NORTH HIGHLANDS OF PERU 29
16
4wI
FIG. 10. Tiahuanaco style of deep stone-lined tombs, Wilkawain.

Clay Spoons Stone Artifacts
Four complete clay spoons and one handle Twelve stone artifacts can be described
were found in Tomb 7H-1. All are slightly dif- briefly.
ferent in shape and in painted designs. They are 3 Chipped flint spearpoints (Fig. lib), the com-
described here for comparative purposes: plete one measures 5 centimeters in length and is 2.6
1 Spoon with an oval bowl and flat handle is or- centimeters wide. The others are broken.
ange on the outside and painted with black and white 4 Chipped obsidian spearpoints (Fig. lid). One is
longitudinal designs on the inside of the bowl 5 by 2.6 centimeters; another 6 by 2.7 centimeters;
1 Spoon has a round bowl and straight flat handle. the others are broken.
It is of whiteware with red, wavy lines on the inside 2 Plain hammerstones
of the bowl (Fig. Ilc). 1 Ring stone
1 Spoon has an elongated bowl with a flat loop 1 Waterworn stone
handle. It is orange outside and has black and white 1 Small red stone bead
lines across the bowl.
1 Spoon has a round bowl and a long straight han- OBSERVATIONS ON THE DEEP-TOMB CERAMICS
dle. It is of whiteware, and the inside of the bowl is As previously mentioned, although minor
painted in light brown and a reddish color in a cur-
sive style. It is a style distinct from anything else differences do exist, the four tombs are treated
found in the tomb. here as a unit. In any case, in either Tomb 7H-1
1 Spoon handle is flat with a nubbin at the end and or 7H-3 the association of several styles is
has black and gray cross stripes on the orange clay. shown. The position of the pottery on the floor
of the tomb, covered by about 2.5 meters of fill
Toy Vessels capped by large cover-slabs, makes the associa-
tion of these styles unquestionable. This is im-
Twenty-five small clay vessels were found in portant since the combination is one as yet un-
the tombs. In spite of the variety of shapes reported from any other Peruvian site. A brief
these are described as a group under the label resum6 indicates the styles and types which
of toys. must be considered contemporaneous in refer-
ence to this site.
4 Toy plates, one orange and three red-on-orange 1. Polished blackware of good quality is well
2 Toy tripod plates, both orange represented by fourteen complete vessels and
1 Toy bowl with annular base and orange color 134 fragments which indicate at least 18 more
2 Handleless toy ollas, both red-on-orange
2 Toy ollas with one handle, one orange, one red- vessels. The shapes are principally shallow
on-orange plates, containers with double-spouts and round
11 Toy ollas with two handles, eight orange, one bridges, goblets with raised bands, constricted-
red, one black, and one red-on-orange collar jars with taper or slightly bulge-collars,
1 Toy orangeware bowl with lip on one side and fragments of modeled animals. To these can
1 Toy blackware bowl with animal-face collar be added a double-bowl with whistling bird and
1 Toy spoon of red color a curved double-headed animal with double-
spout and flat bridge. While the shapes are sug-
Metal Artifacts gestive of coastal blackware affiliations, the
absence of pressed relief or stipple design is
Four pieces of copper were found in Tomb distinctive.
7H-3. Two long copper pins with cut-out ani- 2. Polished redware forms a definite style in
mal heads are 21.7 centimeters in length (Fig. the tombs and is represented by four complete
11a). The decorated animal head is 4 by 2.5 vessels and 460 fragments indicating at least
centimeters. A flat-headed copper pin, 5 centi- 17 more vessels. The most typical shape is a
meters long, and a pin with a flat nail-top head, constricted-collar jar, with either a straight or
3 centimeters long, are the other two pieces. a slightly flaring collar. Shallow plates, ollas,
A narrow wooden point, 5 centimeters long, a vertical-sided open-bowl, and a modeled
overlaid with thin gold leaf, was found in the monkey are other shapes. With the excep-
first tomb. tion of modeling in one piece, these vessels
B
C D
I
I
I
I
i
t
t
t
k
k
E F
FIG. 11. Miscellaneous artifacts from deep stone-lined tombs, Wilkawain.
32 ANTHROPOLOGICAL PAPERS AMERICAN MUSEUM OF NATURAL HISTORY Vol.39
R£SUM£ OF DEEP-TOMB CERAMICS IN TERMS OF COLORS
Frag- Per Com-
ments Cent plete Spoons
-~~ ~ ~ ~ ~ ~ ~
Plainwares 1153 55.22 2
Orangeware, 1053, 2 complete . . .. .. (47.69)
Brownware, 94 ........ . . (4.26)
Whiteware, 6 ............. (.27)
Polished monochrome wares. 594 26.90 18
Blackware, 134, 14 complete . (6.07)
Redware, 460, 4 complete. (20.83)
Negative paintedwares . . 207 9.38 2
Positive paintedwares . 254 11.50 5 4
Orange painted, 2 (.09)
One-color, 86, 1 complete, 1 spoon . (3.89)
(Black-on-red, 49, 1 complete).
(Black-on-orange, 14) .........
(Red-on-orange, 13) ..........
(Orange-on-gray, 3) . .. . . . . ...
(White-on-red, 3) .
(Brown-on-white, 4) .........
(Red-on-white, 1 spoon) ........
Two-color, 134, 3 complete, 2 spoons (6.07)
(Black, white-on-red, 71, 1 complete)
(Black, gray-on-red, 1 complete) .
(Black, gray-on-orange, 2) .
(Black, red-on-orange, 10, 1 complete)
(Black, white-on-orange, 13, 2 spoons).
(Black, red-on-whiteware, 38) . . ..
(Brown, reddish-on-whiteware, 1 spoon)
Three-color, 32, 1 complete.. . . . ... (1.45)
(Black, white, yellow-on-red, 1 complete)
(Black, gray, yellow-on-red, 9).
(Black, white, brown-on-gray, 3) .
(Black, white, gray-on-red, 19) .
(Black, white, red-on-orange, 1) .
Toy vessels. 25
Totals .. | 2208 | 100.00 52 4
are not decorated in any way. Otherwise, the huanacoid affiliations include five complete ves-
redware suggests that found by Strong' in sels and 113 fragments, possibly representing
Middle Ancon I graves. The Ancon redware some 11 more. Again the typical shape is a con-
however, is characterized by incised design. stricted-collar jar, including three complete
3. Two-color negativeware includes two large and at least three fragmentary vessels. Of these,
shallow plates with geometric design and 207 three have modeled faces on the collars, two
fragments representing probably four con- have straight collars, and one a flask-shaped
stricted-collar globular jars with simple nega- body. Six modeled puma bowls can be counted,
tive patterns on the collars. Only the jar if one judges by the fragments of feet, but none
and plate are represented. Neither in shape, is complete. Finally, a double-spout vessel with
design, nor number of colors does this nega- rounded-bridge, and one complete and two
tiveware suggest clear Recuay affiliations. fragmentary angular-bodied bowls with flare-
4. Painted and modeled wares with Tia- rims can be added to the list of shapes. Actually,
Strong, 1925, see PI. 47. then, the identification of these pieces as of
RiSUMi OF DEEP-TOMB CERAMICS IN TERMS OF SHAPES
Shape Fragments Complete aEstimate

Ollas . . . . . . . . . . . . . . . . ... . 1064 5 (32)
Small shallow plates . . . . . . . . . . . . . 93 (12)
Large shallow plates . . . . . . . . . . . . 2
Open round-bottom bowls . . . . . . . . . . 13 (2)
Flaring-sided cups . . . . . . . . . . . . . 3 (1)
Shallow plates with annular base . . . . . . . 5 (2)
Angular-body open-bowls . . . . . . . . . . 2 1 (2)
Vertical-sided open-bowl . . . . . . . . . . . 1
One-handle cup . . . . . . . . . . . . . . 5 (1)
Goblets . . . . . . . ... . . . . . . . . . 15 3 (4)
Constricted-collar jars . . . . . . . . . . . . 637 (7)
With straight collars . . . . . . . . . . . 16 2 (5)
With flare-collars . . . . . . . . . . . . . 7 1 (5)
With bulge-collars . . . . . . . . . . . . 10 (2)
With face-collars . . . . . . . . . . . . . 35 1 (4)
With taper-collars . . . . . . . . . . . . 8 1 (1)
Flask shape . . . . .8 1 (1)
Angular-body bowl with flare-rim.. . .14 1 (2)
Bowl with wide flare-rikn (?) . . . . . . . . . 15 (3)
Double-spout . . . . . . . . . . . . . . . 21 5 (4)
Double-bowl. ....... . . . 1
Modeled vessels . . . . . . . . . . . . . . 52 1 (14)
Tripod plates . . . . . . . . . . . . . . . 5 (2)
Sieves . . ... . . . . . . . . . . . . . . . 9 1 (2)
Toys . . . . . . . . . . . . . . . . . . . 25
Spoons . . . . . . . . . . . . . . . . . . 1 (4) (1)
Miscellaneous unidentifiable . . . . . . . . . 170 (1)
Totals ....... . . . .. . . . .. . 2208 52 plus (109) plus
4 spoons s(1poon)
1 Much restoration has been done, but the task is by no means complete. The estimate here refers to the fragments, ex.
clusive of the complete vessels. It does not follow that 110 vessels will eventually be restored, but that at least 110 vessels
are represented by the sherd collection. Thus a single fragment of a double-spout in one tomb represents one vessel in this
estimate. In reference to the constricted-collar jars, any indication of the type of collar was utilized.
Tiahuanacoid affiliation is based on the designs Coast Tiahuanaco sites as Middle Ancon I is
and colors, rather than on the shapes. Except not based entirely on the ware described as
for the typical puma bowls, the other shapes do Tiahuanacoid, but also on other details, such
occur in Coast Tiahuanaco sites, but are rarely as the following:
typical. On the other hand, no piece in this col- Painted designs in colors of typical Tiahuanacoid
lection has a design as closely resembling the styles, including black, white, yellow-on-red and
Classic Tiahuanaco as some in the Uhle col- other two- and three-design colors on red or orange
lections from Pachacamac, Ancon, Supe, and base.
Moche. The designs, in fact, are somewhat like Some designs, such as the horizontal S-band end-
those of the Nazca Y style,2 particularly those ing in two stylized puma heads.
from Nieveria classed as Nazca Y influenced.' Modeled puma vessels painted in black, white-on-
The suggestion that the tomb collection as a red.
whole bears a certain resemblance to such mixed Goblet-shaped vessels in black, orange, and per-
haps redware. It is interesting to note that no deco-
2Gayton and Kroeber, 1927. rated goblets are found.
'See Gayton, 1927, PI. 92c, d; P1. 96j. Double-spout bowls with painted design, or of
blackware, and especially the curved-body, double- many of them. Certainly such graves were
spout, double-headed, black vessel. made by similar excavation under these rocks,
The presence of polished redware, including shapes as it is inconceivable that the boulders were
found in Middle Ancon I, such as straight-collar jars, actually moved. Some of the boxes were so well
flask shapes, straight-sided bowls, and ollas with hori- placed that they were opened with considerable
zontal handles. difficulty. We excavated a total of thirteen of
Face-collar jars with suggestive Tiahuanacoid de-
signs. these box graves, and certainly more could have
The absence of flat-collar handles on the con- been found had the contents of those excavated
stricted-collar jars. justified the effort. As it was, only three con-
The absence of pressed-relief design. tained pottery (and of these only one had more
Hand, rather than mould-made vessels. than one piece), one other contained non-
In spite of the case that can be made for the ceramic artifacts, and the remaining nine were
resemblance to Middle Ancon I, however, there empty. Presumably all were intended as graves,
are so many weaknesses that it seems preferable although only four contained any trace of
at this time to leave the question open. bones. Some are built immediately under the
5. Clay spoons with straight and loop handles surface boulder so that the latter serves as a
are associated. One in particular has a design cover. Some are well under the boulders and
of light brown and a reddish color in fine-line covered by a stone slab. Since most of these box
cursive style on a white clay. Fragments in graves are more or less alike and since so few
this style are frequently found in the region of had contents of any importance, the description
Cajamarca,' but without information on their of each is presented in tabular form.
chronological position. 1. Site 7H-9A: A 35-centimeter square slab-
6. Tripod plates are sparsely represented by lined and covered box 50 centimeters deep, lo-
five fragments, probably indicating two vessels. cated 50 centimeters under a surface boulder.
7. Orange and brown plainwares are fre- No contents.
quent. Although ollas are commonest, shallow 2. Site 7H-9B: A 50 by 80 centimeter rec-
plates with annular bases, simple plates, and tangular slab-lined and rough stone-covered box,
pierced plates, like sieves, are also included. 70 centimeters deep, located 1 meter under a
8. Miscellaneous painted pieces should be surface boulder. No contents.
mentioned, although they present no startling 3. Site 7H-9C: A 70-centimeter square slab-
variations. Noteworthy is the absence of the lined and covered box, 50 centimeters deep,
white-on-red style which is found at other sites located 1.2 meters under a surface boulder. No
in this same region. The three small white-on- contents.
red fragments which were found are not defi- 4. Site 7H-9D: A 60-centimeter square slab-
nitely of this style. lined and covered box, 50 centimeters deep,
9. Toy vessels do not seem to have great sig- located 1 meter under a surface boulder. No
nificance as diagnostics, but 25 are included in contents.
the tomb collection. 5. Site 7H-9E: Box tomb like above.
10. Finally, the presence of copper pieces, 6. Site 7H-9F: Box tomb like above.
flint and obsidian points, and a few miscellane- 7. Site 7H-1 A: A 40-centimeter square slab-
ous stone objects completes the mixture. lined and covered box, 50 centimeters deep,
More evidence will be available for the com- located just under a surface boulder. Contained
parative discussion of this grave collection when traces of bones.
the description of the remaining sites in the 8. Site 7H-llB: A 30 by 20 centimeter rec-
Wilkawain region is presented. tangular slab-lined and covered box, 30 centi-
meters deep, located just under a surface boul-
D. STONE Box GRAVES der. No contents.
The slopes of the hills around Wilkawain and 9. Site 6H-5: A rectangular slab-lined box,
Ichik-Wilkawain have numerous natural boul- located just under a surface boulder which
ders, some of immense size. By pit excavation serves as the cover. Contained traces of adult
we discovered stone-lined box graves under bones.
10. Site 7H-6: A rectangular slab-lined box,
1 See Muelle and Blas, 1938, PI. 72. located just under a surface boulder which
serves as the cover. This box grave contained A globular olla with flare-rim and no handles, with
three fragments of silver, one piece of a copper a simple red band on the orange clay base
pin, an irregular bead of lapis-lazuli, a short Two shallow plates, 9.5 centimeters in diameter
tubular bead, and ten disc beads. and 3 centimeters deep, decorated with red curves
11. Site 7H-7: A 70 by 80 centimeter rec- and dots on the inside. The base is orangeware.
A large shallow plate, 20 centimeters in diameter
tangular slab-lined and covered box, 55 centi- and 8 centimeters deep, decorated on the inside with
meters deep, located 30 centimeters under a red bands on the light gray base clay
surface boulder. It contained traces of adult Two shallow redware plates, 17.5 centimeters in
bones and a single vessel of a unique type. This diameter and 7 centimeters deep
is an angular-bodied bowl with a flare-rim (PI. The contents of the stone box graves are too
2a). It is 16.5 centimeters in body diameter,
13 centimeters high, 11 centimeters in rim di- limited for serious analysis. The unusual incised
ameter, and 7 centimeters in base diameter. vessel of Site 7H-7 is interesting, but unplace-
The bowl is of brownware, decorated on the able in relation to the other ceramics, with the
upper and lower halves of the body with wide possible exception of the fact that it was in a
incised lines and punch marks. On each side is stone tomb similar to the others. The contents
a double nubbin with a deep punched hole be- of Site 6H-3B conform in general to the house
tween. The wide incised lines show traces of a site material of Site 6H-2 and are not too in-
fill of red and yellow paint. A similar vessel consistent with the deep stone grave contents
is owned by Dr. Augusto Soriano Infante but, (Sites 7H-1 to 4).
like this one, without any information on possi- E. UNLINED GRAVES
ble associations. No fragment of such incised Other graves found cannot be grouped with
brownware was found in any other site in this
region. Of the total Wilkawain collection, this is the stone-lined boxes, since apparently they
the one piece which suggests Chavin ceramics, represent direct burials, without any particular
and it is indeed unfortunate that no data for preparation of the grave. Three of these are de-
placing it in relation to other ceramics of the scribed, although one is dubiously a grave.
Wilkawain region were obtainable. Site 8H-1: A grave in the southeast end of a
12. Site 6H-3A: A 40-centimeter square slab- low mound was about 1 meter in diameter and
lined box without covering, 50 centimeters the same deep, without lining or covering of any
deep, located 1.58 meters under a surface boul- kind. It contained two vessels and one piece of
der. It contained adult bones and an orange a third which is probably not part of the origi-
olla. The bowl is globular with a flare-rim and nal contents.
two horizontal flat-loop body handles. It meas- A complete modeled puma vessel is painted in
ures 22 centimeters in diameter, 19 centimeters black, white-on-red (PI. 2f). The elongated body is
high, and 14 centimeters in rim diameter. The supported by four short cylindrical legs. The two
bottom is round. forelegs have two projections representing toes. The
13. Site 6H-3B: Under the same boulder head is modeled in the round with upright ears and
an open mouth showing rows of teeth and two sets of
which covered the previous box grave was an- crossed fangs. The tail is conical and projects straight
other, above and behind it. This second one back. The bowl has a constricted straight collar. The
was an 80-centimeter square slab-lined box, lo- design consists of circle bands outlined in black and
cated immediately under the surface boulder filled with black dots on white paint. One band en-
which served as the only cover. No traces of circles the neck, and four circles are on the body. The
bones were found, but eight vessels were in- collar and legs are simply decorated. This vessel
cluded: closely resembles the type indicated by the frag.
ments in the deep-tomb collection. The total length
An orangeware constricted straight-collar jar with from the tip of the tail to the nose is 27 centimeters.
flat base. The body is 14 centimeters in diameter and The height to the rim of the collar is 19.5 centimeters.
14 centimeters high. The collar is 4.5 centimeters The constricted collar is 4.5 high and 7 centimeters in
high and 6 centimeters in diameter. diameter.
A redware globular jar with constricted straight A complete orangeware vessel in the same grave
collar. The body is 13 centimeters in diameter and 14 has a slightly angular body and a constricted straight
centimeters high. the collar is 3 centimeters high and collar (PI. 2e). A flat wide loop handle, 3 centimeters
5 in diameter. wide, extends from the center of the collar to the
body. The vessel is 22 centimeters high and 6 in di- sists of white horizontal and oblique lines on a red
ameter. A band around the upper half is decorated base.
in the pressed-relief technique. The design is com- A similar vessel with a flat base is 13 centimeters in
posed of two warriors facing each other. The bodies diameter and 7.5 centimeters deep (Fig. 12f). The
are front view and the heads in profile. Two puma design is of white lines and triangular areas filled with
heads are attached by ribbon bands to the waist of small dots on a red base.
each figure. Two sets of similar ribbons, ending in A flaring-sided open-bowl with flat base is 12 cen-
puma heads, project from the head ofeach figure, the timeters in diameter and 6 centimeters deep (Fig.
front pairs crossing between them. Each man holds 12h). The white-on-red design consists of pairs of
an ax in one hand and grasps the ribbon band of his vertical lines, alternating with sets of four oblique
opponent with the other. In both, long tongues pro- wavy lines.
ject from the mouth. Neither pressed relief nor such A similar vessel, 11 centimeters in diameter and 5
elaborate design units were found in the deep-tomb centimeters deep, has sets of vertical zigzag white
ceramics. lines on a red base (Fig. 12b)
The upper section of a face-collar blackware ves- A vessel similar in shape and size has pairs of
sel with incised body design was also found with the white oblique lines, the space between filled with
above pieces. Other fragments of this same vessel dots on a red base
were found in other parts of the mound, so that the A vessel similar in shape and size has a simple
association is probably fortuitous. white-on-red linear design
A small brownware globular jar has a tall cylindri-
Site 8H-2: A partially hollow space, 1.2 meters cal neck, 3 centimeters in diameter, ending in a
in diameter and 1.15 meters deep, may possibly slightly flaring rim edge, and a flat neck-to-body
have been a grave. It contained broken pieces handle (Fig. 12d). The body is 9 centimeters in di-
of a blackware constricted-collar jar with a ameter and 7.5 centimeters high, and the neck is 4.5
slightly flaring collar and a pierced nubbin centimeters high.
handle at the collar base. Also included were A white-on-red double whistling jar has one con-
pieces of an orange tripod plate, an orange olla tainer about the size and shape of the last vessel
with flare-rim and horizontal handle, a brown- described connected by a tube and a flat bridge to a
ware plate with annular base, and a redware container capped with a human bust (Fig. 12c). Two
short arms are in relief. The head with its trisected
constricted-collar j ar. crest is poorly modeled, and has a relief nose, incised
Site 7H-15: An unlined tomb partially be- eyes, and an open mouth containing a whistle. The
tween and partially under two large surface flat handle is decorated with incised lines and dots.
rocks contained a total of thirteen vessels: two Traces of white design on the red base can be seen.
redware, two brownware, and nine white-on- OBSERVATIONS ON UNLINED
redware. The pieces are described in detail GRAVE MATERIALS
since the white-on-red style is rare in the ma-
terials found elsewhere in the Wilkawain ex- The two grave lots of Sites 8H-1 and 7H-15
cavations, but appears in the post-Chavin are quite distinct in style and were described
collection. together for convenience only. Treating the
A convex-sided round-bottom open-bowl of red-
first pair of vessels then (8H-1), it is obvious
ware, 10 centimeters in diameter and 2.5 centimeters that both are consistent with the Coast Tia-
deep (Fig. 12e) huanaco periods. The modeled puma vessel is
A similar brownware vessel, 8 centimeters in di- quite similar to one found by Uhle (1903) at
ameter, and 4 centimeters deep Pachacamac in the Epigone level (which is his
A similar brownware vessel, 14 centimeters in di- earliest clearly isolated period at that site).
ameter, and 6 centimeters deep The accompanying orangeware vessel, with its
A redware open-bowl with flaring sides and slightly complex pressed-relief design, is remarkably
rounded base, 14.5 centimeters in diameter and 6 similar to a redware piece illustrated by
centimeters deep Kroeber' from the Supe Middle Period. The
A similar vessel, 15.5 centimeters in diameter and two struggling warriors on the Supe pressed-
7 centimeters deep, is decorated with a white angular relief piece are the same in detail as those on
scroll design on a red base (Fig. 12a) the Wilkawain vessel. Kroeber is inclined to
An open-bowl with straight diverging sides and a
slightly rounded base is 14.5 centimeters in diameter place these Supe pressed-relief vessels in a
and 6 centimeters deep (Fig. 12g). The design con- 1 Kroeber, 1925b, PI. 71d.
1944 BENNEfT: THE NORTH HIGHLANDS OF PERU 37
B
A
................ ..
H
FiG. 12. White-on-redware from unlined grave, Wilkawain.
slightly later phase than the purer Tiahuana- 4 by 3 meters and over 1 meter thick, served
coid styles at the same site. Even in the ab- as a roof for a second subterranean dwelling.
sence of other finds this one grave would estab- A stone-lined doorway, 50 by 50 centimeters,
lish a relationship with the Coast Middle was at the south side. The interior was filled
Periods. with dirt and house refuse practically to the
The white-on-red style from Site 7H-15 is roofing stone, probably due to washed-in fill.
more puzzling. Some of the vessels have the Partial excavation of this fill produced a mis-
shape of the flaring-sided cup so typical of cellaneous collection, including four complete
Highland Tiahuanaco. Even the simple oblique bowls, a spoon and various fragments, and 829
zigzag, wavy lines, the angular scroll, and the sherds. A number of styles are represented, but
dot-filled triangles are found in the Decadent unfortunately the nature of the fill and the
Tiahuanaco style, although usually with black cramped excavation required if a large scale
color as well as white. In the Wilkawain region operation was to be avoided, make it impossi-
only the mixed refuse of Site 9H-2 contained ble to state that all the materials are contem-
more of this white-on-red style. On the other poraneous. The mixture is described again in
hand, vessels which must belong to the same terms of wares, plus a few special categories.
style were found in intrusive graves at Chavin.
A comparison with the white-on-red Chancay MATERIALS FROM THE SECOND SUBTERRANEAN
style as isolated by Kroeberl is suggested, in HOUSE SITE
spite of the fact that the differences are many. Plainware
On the basis of Grave 7H-15 it is clear that a
white-on-red style can be isolated, but the time Orangeware, 1 complete and 588 fragments.
relationship with other Wilkawain material is By far the largest proportion of the pieces
not clarified. in the total collection is plainware, basically
orange, but frequently with a brownish tinge
F. SUBTERRANEAN HouSE SITES and a burnt black due to use. The pieces are
In the area of Site 9H immense boulders and listed in tabular form.
large flat rocks are exposed on the surface. 572 Fragments of ollas. The typical olla form rep-
Several of these had served as coverings for resented has a globular body with rounded base and
subterranean houses. Four were seen in total, a flaring rim. A few fragments represent a high flar-
and two examined in more detail. ing rim, more like a somewhat constricted collar.
House 1, Site 9H-1. Two large flat surface Only two flat loop-handle pieces are included, and
two plain body lugs.
rocks had served as a covering for a room ex- 9 Thick orange fragments from an open vessel
cavated beneath. In general its shape follows with a projecting ridge just below the rim
that of the surface rocks, with a central section 1 Complete shallow open-bowl with a plain rim
5.6 by 2.25 meters and a niche, 1.5 meters wide and rounded base, 13 centimeters in diameter and
and 1.2 meters deep, at one corner. The original 4.5 centimeters deep
entrance at the south was 1.57 meters wide, 3 Pieces of solid round handles
but was mostly filled with washed dirt. The 1 Piece of flat handle or bridge
floor of the room was 2.5 meters below the roof 2 Conical handles suggesting Recuay types
stone, and the walls were made of rough stone (Fig. 13c)
mixed with some good-sized boulders. The 1 Fragment of an open-bowl with an inflaring
walls are poorly made, and one section of the rim, flat on top and decorated with a relief hook
room had partially caved in, although the rest Polished Monochrome Wares
was still open. Clearing the floor produced some
30 fragments of straight, orange clay trumpets, Blackware, 7 fragments. Blackware is de-
including four flaring-mouthed pieces, and nine cidedly rare, and the fragments, while well
plain, crude orange sherds. The trumpets re- polished, are without pressed-relief or stipple
semble those found in the Recuay style gal- decoration. Four fragments are from straight-
leries to be described later. sided goblets, one representing a goblet with
House 2, Site 9H-2. A large flat rock, about raised band below the rim. The remaining three
pieces are from a globular-bodied vessel.
1 Kroeber, 1926, PI. 87c, d, e. Redware, 84 fragments. These pieces are all
1944
j<3z.7*i;!UX.r> t@'i8rs,W;Avtg.-*7g
''.',.'.
;
t
BENNEfT: THE NORTH HIGHLANDS OF PERU
-.- 3; 7
+ ; ;.W..,n
\|---'-''|'----"'''.'o'
=X
A
;e ;
;|-
8t;\;<
/
\~~~~~~~~~~
B /
39
.~-t..
,:
E t't5,-^ :* :t '1:
H
G I
.~~~~~~~~~:
K
FIG. 13. Ceramic styles from subterranean house site.
from red-slipped or painted monochrome ves- Nine pieces are from flat-based flaring-sided
sels, and most of them are polished. cups (Fig. 13h,j). A semi-complete open plain-
77 Fragments of constricted-collar jars, including rim dish with an annular base has a crude white
13 pieces of flare-collar. Other collar types are not in- wavy design inside (Fig. 13k). One is a fragment
dicated. of a small bowl.
4 Fragments of shallow plates
2 Fragments of open convex-sided bowls
1 Fragment of a cylindrical neck or spout and Modeled and Relief Pieces
rounded handle Twenty pieces represent modeled or relief-
Negativeware decorated vessels. While not a distinctive group
Negativeware is represented by a fragment in themselves, they do not properly belong with
of a large high-collar vessel with a step design. the other wares described.
This is two-color negative, in a black and 1 Redware fragment with relief design represents
orange combination. There are also two small two arms of a human figure. The right hand is a disc,
fragments which appear to be three-color nega- but the left is represented as holding a club. Part of a
tiveware. Twenty-three pieces of a large vessel necklace is portrayed. The general impression is that
have a wide orange line on a burnt clay base of Early Chimu style, but the fragment is too small
which may possibly be negative design. for identification.
1 Redware fragment with traces of white paint is
the head of an animal which seems to form one end of
Painted Grayware a handle (Fig. 13g). This might possibly be part of
Three complete vessels and 20 fragments of the white-on-red style.
a special painted grayware are part of the col- 1 Redware fragment of relief animal feet
lection. No trace of a burial was found, but the 1 Orange animal head lug with crossed fangs
4 Redware pieces with relief designs, one part of
pieces from these vessels were found in the a human face, two bird heads, and one band
refuse. The base color of the clay is gray. The 3 Pieces of a vertical-sided open-bowl with annular
design on the outside is painted in a reddish- base of a thick orangeware, with an animal head lug
brown color on a whitened base. An orange on either side, and a snake-like relief band on the
band forms an irregular part of the pattern. side (Fig. 13i)
Complete open-bowl with convex sides and a 9 Slightly modeled pieces
flare-rim measures 15.5 centimeters in diameter and
7 centimeters high (Fig. 13a). The base is annular Trumpets
and 5 centimeters in diameter. Three rough cuts on
the base resemble a potter's mark. A step-and-hook Twenty-three fragments of straight, orange-
design painted in reddish-brown on a whitened area ware trumpets with flaring mouthpieces are
forms a band around the side. Two orange bands like those found in the first house (9H-1) and
encircle the vessel, one near the rim edge, and one in the Recuay galleries.
through the middle of the above design.
Complete open-bowl with plain rim and annu-
lar base is roughly the same size as the above. A Miscellaneous Painted Pieces
simple interlocking linear design is painted in the
same color combinations over a gray base clay. 6 White-on-orange fragments of a large olla with
1 Restorable convex-sided plain-rim vase with flat handle
annular base has a relief face on one side (Fig. 13b). 3 Red-on-orange pieces of a large olla with high
Two relief arms are along the rim edge with the face collar
immediately below them. On the reverse side is a 2 Orange-on-white pieces, one with cross-hatch
simple linear design in the typical colors. design
19 Pieces of a grayware vase, but without painted 2 Yellow-on-red fragments
design 12 Red-on-buff pieces from a shallow plate with
Red-on-gray painted piece an interior design of a line and concentric circles
5 Black-on-red fragments of a constricted, bulge-
White-on-Redware collar jar with linear design
4 Red-on-white clay fragments, slightly modeled,
Eleven fragments are of the white-on-redware which resemble Recuay style
style described previously from Site 7H-15. 7 Miscellaneous small painted fragments
1944
Plainware.
. _~ ~ ~ ~ ~ ~ ~
Grayware, 19 .......
BENNETT: THE NORTH HIGHLANDS OF PERU
R£SUM1 or
Orangeware, 624, 1 complete .

COLORS OF SUBTERRANEAN HousE SITE COLLECTION
Blackware, 7 .....................
Redware, 90 ... .. ... .............
Negativeware.
.
.
.
.
.
.
.
Polished monochrome . . . . . . . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Fragments
643
97
26
Per
Cent
77.56
(75.27)
(2.29)
11.70
(.84)
(10.86)
3.13
Complete
Pieces
1 complete
41
Two-color, 1..................... (.12)

Three-color, 2. ................... (.24)
Doubtful, 23 .. . . . . . . . . . . . . (2.77)
Positive paintedware. 63 7.61 3 complete
One-color, 55, 1 spoon ................. (6.64) 1 spoon
(Red-on-gray, 1) ..................
(Black-on-orange, 1 spoon, 1) ............
(White-on-red, 12) .................
(White-on-orange, 6) ................
(Red-on-orange, 3) . ..... .... . . . ...
(Orange-on-white, 2) . . . . . . . . . . . . . . . .
(Yellow-on-red, 2) .. . . . ... . . .. ... ..
(Red-on-buff, 12) ..................
(Black-on-red, 5) ..................
(Red-on-white, 4) . . . . . . . . . . . . . .
(Miscellaneous, 7) .................
Two-color, 8, 3 complete .. . . . . . .. . . .
(Reddish-brown, orange-on-gray, 3 complete)
Light brown, dark brown-on-white, 8 spoon fragments
Total . . . . . . . . . . . . . . . . . . . *complete
spoon
Clay Spoons long and 7 millimeters wide, and the other a

The one complete clay spoon and nine frag- fragment of a copper plate.
ments represent two styles. The complete spoon Stone Artifacts
(Fig. 130 and part of another are of one type,
namely, an oval bowl (4 by 6 centimeters) with Six stone beads of small size can be described
a flat-loop handle, painted in vertical black as a flat disc, a tubular bead, three grooved
lines on an orange clay. The remaining eight turquoise (?) beads, and a small club-shaped
fragments are of round-bowled spoons with black stone.
straight handles (Fig. 13d, e). The design is
cursive and is formed of fine lines in both light OBSERVATIONS ON SUBTERRANEAN HOUSE
and dark brown on a white clay. It is again the COLLECTION
style of Cajamarca, and of one spoon in the It was stated previously that the house col-
deep-tomb collection. lection need not be contemporaneous, in con-
trast to the material from the deep tombs. On
Metal Artifacts the other hand, there is no excavation evidence
Two pieces of copper were the only metal for the separation of the various styles repre-
artifacts found, one a thin bar, 6.5 centimeters sented. The concept of a subterranean house is
TABLE OF SHAPES SUGGESTED BY SUBTERRANEAN HOUSE COLLECTION
No. Per Cent
Ollas . . . . . . . . . . . . . . . . . . . . . . 578 69.72
Large, high-collar jars . . . . . . . . . . . . . . 4 .48
Shallow plates . . . . . . . . . . . . . . . . . . 16 1.93
Open convex-sided bowls . . . . . . . . . . . . . 2 .24 1 complete
Open-bowl with annular base . . . . . . . . . . . 1 .12 2 complete
Open-cup with annular base . . . . . . . . . . . 1 complete
Open-bowl with ridge below rim . . . . . . . . . . 9 1.08
Open-bowl with inflare rim . . . . . . . . . . . . 1 .12
Open-bowl with relief heads . . . . . . . . . . . . 3 .36
Flaring-sided cup . . . . . . . . . . . . . . . . 9 1.09
Goblet . . . . . . . . . . . . . . . . . . . . . 4 .48
Constricted flare-collar jar . . . . . . . . . . . . 77 9.29
Constricted bulge-collar jar . . . . . . . . . . . . 5 .60
Flare-collar vase . . . . . . . . . . . . . . . . . 19 2.29
Cylindrical neck and handle . . . . . . . . . . . 1 .12
Modeled . . . . . . . . . . . . . . . . . . . . 21 2.56
Solid round handles . . . . . . . . . . . . . . . 3 .36
Flat bridge . . . . . . . . . . . . . . . . . . . 1 .12
Conical handle. ....... 2 .24
Trumpets . . . . . . . . . . . . . . . . . . . . 23 2.77
Spoons .. 9 1.09 1 complete
Undetermined . . . . . . . . . . . . . . . . . . 41 4.95
Total . . . . . . . . . . . . . . . . . . . . 829 100.00 4 complete
1 spoon
suggestive of Recuay, although it is not a cer- namely, the simple blackware, the redware
tain identification. In any case the contrast constricted-collar jars, the loop-handled spoons,
with the above-ground houses in the same re- and the orangeware ollas. On the other hand,
gion is marked. Certain features of the collec- the differences are enormous. The absence of
tion are also reminiscent of Recuay, namely, the Tiahuanacoid features, so obviously a part
the straight trumpets, the conical handles, the of the deep-tomb collection, is startling. There
two possible three-color negative pieces, and the are no modeled pumas, no designs suggestive of
red-on-white slightly modeled fragments. How- Tiahuanaco Epigonal, no typical Tiahuanacoid
ever, the lack of negativeware or any other out- color combinations, and no two or more colored
standingly characteristic Recuay style piece wares (the two groups of two-color wares are
should be emphasized. covered by the special grayware and the cursive
Represented by complete pieces in an other- spoons). Even the lack of two-color negative is
wise definitely refuse site one new style is en- surprising.
countered in the collection-grayware annular- Treated as a unit the collection is almost im-
base vessels with simple designs in reddish- possible to place in relation to other finds in the
brown plus orange on a whitened area. region. The breakdown into styles is conven-
The white-on-red style, isolated in the grave ient, but not secure evidence for time placing.
of Site 7H-15, and the cursive-designed spoons
of Cajamarca style are both represented, but G. SUBTERRANEAN GALLERIES
little is added to the general problem of placing Seven subterranean galleries were located on
these styles. the slopes below Ichik-Wilkawain, and two
In spite of the geographic proximity to the more were found across the ravine at the site
deep tombs, previously described, only a few at Irwa. All of these are more or less alike and
parallels in the two collections can be traced, can be described together. Most of the galleries
were at least partially opened, or apparently to the surface, where it was covered with top
had been, but two were found with the en- soil. After the stone fill and the central slab
trances still intact and sealed. All were exca- were removed, one had access to the hollow
vated down to the hard-packed clay floors. The part of the shaft, and in turn to the entrance
variations in measurements and details are of the gallery some 62 centimeters wide and
given in a tabulated list below, and a descrip- 60 centimeters high (definitely smaller than the
Oa 7.3Z -.P-
FIG. 14. Groundplan and cross-section of Recuay style gallery, Wilkawain.
tion of one intact gallery should serve for the gallery itself). The actual gallery is 7.32 meters
type (Fig. 14). long, 95 centimeters wide, and 90 centimeters
The entrance to the gallery of Site 7H-12 is high. The floor is of hard-packed clay, and ex-
an irregular shaft, lined with a stone wall, meas- actly 2 meters below the ground surface. The
uring some 65 centimeters wide and 1.18 meters sides are faced with eight slabs each, and an-
long (PI. 3b). The shaft extends to a depth of other slab faces its back end. Small stones are
2 meters. The lower half of the shaft is hollow carefully inserted to fill in the irregularities be-
and, when discovered, was covered by a stone tween the slabs. Eight large slabs, again care-
slab which extended across it, resting on a spe- fully chinked, form the roof. The whole gallery
cially constructed ledge. The upper half, above had been so well constructed and sealed at the
the slab, had been filled with small stones up entrance that only a thin layer of silt covered
the hard clay floor. In spite of the fact that no about 85 centimeters by 63 centimeters. Three
one had opened this site before, nothing but the slabs form the roof. A niche in the eastern wall
powdered remains of a few bones was found in is 30 by 30 centimeters square and 52 centime-
it. ters deep. This is the only gallery that had any
Other galleries are essentially the same as variation on the plain walls. Sherds of Recuay
this one. Some are longer; some are much closer type were found.
to the surface; some have rectangular rather Site 10H-2: A gallery 5 meters long, 1.3 me-
than irregular oval entrances. Others contained ters wide, and 90 centimeters high, with the
some pottery. Orientation is not consistent in covering slabs just below the ground surface.
spite of a general north-south arrangement. A It had previously been opened, but a few sherds
brief tabular description of all the galleries fol- were still to be found.
lows:
Site 7H-SB: A gallery over 20 meters long, MATERIALS FROM SUBTERRANEAN GALLERIES
1.5 meters wide, and 1.2 meters high with the
roofing stones 1 meter below the ground sur- The nine galleries furnished relatively little
face. This site is described in more detail later. material. Actually, six complete vessels were
It contained a few sherds of Recuay style pot- found and 267 sherds representing at least 21
tery. more vessels, although neither complete nor re-
Site 7H-8: A gallery 5 meters long, 1 meter storable. No gallery contained enough material
wide, and 90 centimeters high with the roofing to justify separate treatment and so the collec-
slabs just below the ground surface. The en- tion is described here as a unit. Actually the
trance was at the southeast end. It contained materials from the various galleries are quite
one complete vessel. consistent in style. Three-color negativeware,
Site 7H-10: A gallery 4 meters long, 1.1 fragments of clay trumpets, and indicated
meters wide, and 70 centimeters high with the shapes all conform to the Recuay style and are
roofing stones just below the surface. The en- found in nearly all the galleries. No style or
trance had caved in. Two restorable vessels shape inconsistent with the known Recuay was
were found. found. Consequently, it is even more logical to
Site 7H-12: The type gallery described previ- treat the collection from the nine galleries as a
ously. unit. The materials are described in terms of
Site 7H-13: A gallery 10.1 meters long, 1.15 type of painted decoration.
meters wide, and 95 centimeters high, with the
roofing stones 80 centimeters below the ground Plainwares
surface. Like that at Site 7H-12, the entrance Brownware, 28 fragments.
to this gallery was intact, with a stone-filled
shaft 1 meter by 66 centimeters and 1.75 meters 4 Fragments of straight trumpet (Recuay 02)1
deep. The actual gateway to the gallery from 8 Pieces of a plain open round-bottom bowl
this vertical shaft is 77 centimeters wide. It con- (Recuay A-1)
14 Pieces of a thick brown globular vessel with a
tained fragments of Recuay style pottery. relief design of a hand holding a disc (possibly
Site 7H-14: A gallery 6 meters long, 1.15 me- Recuay E)
ters wide, and 90 centimeters high with the 2 Miscellaneous fragments
roofing stones about 25 centimeters below the
ground surface. The entrance had been opened Orangeware, 91 fragments. Eighty-four of
previously. It contained a few sherds. these fragments are from straight clay trum-
Site 7H-16: A gallery 10 meters long, 1.2 me- pets. A typical trumpet has a flaring mouth-
ters wide, and 1 meter high, with the roofing piece about 3.2 centimeters in diameter and a
stones 1.5 meters below the ground surface. It straight tube which diverges gradually to a
had previously been opened, but still contained plain end about 5.3 centimeters in diameter.
a few sherds. The remaining seven pieces are of crude orange.
Site 10H-1: A gallery 5.27 meters long, 1.32 Whiteware, 1 fragment.
meters wide, and about 90 centimeters high,
with the roofing stones just below the ground 1 Reference to the Recuay pottery classification, see pp.
surface (PI. 3a). The entrance was rectangular, 101-102 and Fig. 32.
1944 BENNETT: THE NORTH HIGHLANDS OF PERUI 45
Negativewares White-on-red, 2 complete. These two pieces
Black-white-red negativeware, 78 fragments. represent Recuay styles with no obvious rela-
Negative-painted vessels with sharp black- tionship with the white-on-red style of Site
white design with the addition of red coloring, 7H-15.
either in negative technique or positive paint- 1 Complete globular vessel with a constricted
ing, are typical of Recuay style (Fig. 18b). No flare-collar and short bridge to a modeled human
complete vessel was found, but a number are head (Recuay H-i), painted simply in white on a red
partially restorable, and indicate shapes and base (Fig. 15a). The head is reasonably well modeled
designs equally typical of Recuay. with a wide headband and a simple headdress above.
Two disc earplugs are decorated with small circles.
16 Fragments of one large vessel. The upper part On each side of the head is a modeled animal.
is missing, but the general appearance is that of a 1 Complete open-bowl (Fig. 15e) with conical
collar-jar with a short bridge to a modeled head handle (Recuay B-1). The bowl is 14.5 centimeters
(Recuay H-1). On the body is a relief arm with a in outer rim diameter and 7 centimeters high. The
square shield. A stylized cat head represents the sides of the bowl have a very slight bulge, the base
hand. Sets of vertical and horizontal straight and is flat, and the rim projects at a right angle, being 1.6
wavy lines are on the back of the vessel. centimeters wide on top. Around the rim is a simple
20 Pieces of one large vessel, with all the upper design, in white paint on a red base, a square with a
part missing. Again, this is possibly a shape like the dot inside. The conical handle projects from near the
previous one described (Recuay H-1). The base clay base and is 9 centimeters long and 3.5 centimeters
is buff, on which the black-white-red negative de- at the base.
sign is superimposed. Around the bottom part of the
vessel is a band of rectangles filled with rows of dots. Red-on-orange, 1 complete, 14 fragments.
Above this is part of a complex design probably rep- The fragments represent two vessels. All three
resenting a Recuay cat. in this style are the same shape.
12 Pieces of a vessel similar to the above, with part
of the collar and bridge. On one side is a good section 1 Complete open-bowl (Fig. 15c) with a plain rim
of a seated Recuay cat design, flanked by an inter- and an annular base (Recuay A-2). The bowl is 14
locking pattern resembling a ray fish. centimeters in diameter and 8 centimeters high, with
30 Miscellaneous pieces include part of a base with an annular base 6.5 centimeters in diameter. Red
two feet in relief, a section with a wing-like body lug, bands form a simple design on the orange clay.
a piece of disc rim, and a constricted collar. All show 9 Fragments of a vessel similar in shape and design
traces of complex designs. 5 Fragments of a similar vessel, with wide red
bands plus a cross-hatched area on the orange clay
Two-color negativeware, 1 complete, 9 frag. (Fig. 15d)
ments. None of these pieces is of the negative
type found in the deep tombs. Red-on-gray, 1 complete (Fig. 15b). An elon-
1 Essentially complete dipper with open bowl, 9.3 gated globular vessel with a flat base, con-
centimeters in diameter and 6 centimeters deep, has stricted neck, and flare-rim is 19 centimeters
a hollow conical handle, 3.8 centimeters at the base high, 14.5 centimeters in body diameter, and
and 10.5 centimeters long (Fig. 15f). The tip is miss- 9 centimeters in rim diameter (Recuay F). The
ing, so that it is impossible to state whether or not it clay is gray and crude, and the design consists
ended in a modeled human head. The general shape of elongated red daubs around the neck. While
is common in Recuay collections (Recuay B-4). On unique in the collection, it is not the special
the handle is a simple geometric design in black- grayware of Site 9H-2.
orange negative. Red-on-buff, 1 complete. An exceptional ves-
4 Pieces of one open-cup (Recuay A-1) with black- sel represents a deep open-bowl, with plain rim
white negative design and annular base. The rim diameter is 12.5
4 Miscellaneous pieces with black-white negative centimeters, the height 11 centimeters. A clay
design. It is possible that these are small fragments
from black-white-red negative vessels. trumpet starts at the rim and encircles the bowl
1 Fragment of black-red negativeware three times. The mouthpiece projects above the
rim, and the horn extends outward from the
Positive Paintedwares lower part of the bowl. The trumpet is painted
The remainder of the ceramic collection con- in red lines on the buff clay. The shape is a spe-
sists of positive painted pieces, five complete cial category of Recuay A type.
and 60 fragments. Red-on-white slip, 38 fragments. All but one
FIG. 15. Recuay style ceramics from galleries, Wilkawain.

RAsUMi OFTYPE OF PAINTING INGALLERY COLLECTION
Pieces Per Cent Complete
Plainware . . . . . . . . . . . . . . . . . . . . 120 44.95
Brownware, 28 .(10.49)
Orangeware, 91 .(34.08)
Whiteware, 1 .(.37)
Negativeware . . . . . . . . . . . . . . . . . . 87 32.58 1
Black-white-red negative, 78 .. ......... (29.21)
Black-white negative, 8 . . . . . . . . . . . . . (3.37)
Black-red negative, 1 ...... . . . . . . . .
Black-orange negative, 1 complete.
Positive paintedware . . . . . . . . . . . . . . . 60 22.47 5
White-on-red, 2 complete . . . . . . . . . . . . (5.24)
Red.on-orange, 14, 1 complete . . . . . . . . . .
Red-on-gray, 1 complete.
Red-on-buff, 1 complete.
Red-on-white slip, 38 . . . . . . . . . . . . . . (14.23)
Red-and-white, 4 . (1.50)
Black outside, red inside, 2 (.75)
Miscellaneous,2 ..... . . . . . . . . . (.75)
Totals ... . . 267 100.00 6
of these fragments are possibly from a single ears and two short arms. Behind the figure is
vessel. On top of a globular vessel is the bust of some kind of wall with a cut-out step design.
a modeled figure covered by a roof supported The vessel is far from complete. Three con-
by two pillars. The figure has appliqu6 eyes and stricted collars are included, but possibly only
R£SUMm OF SHAPES REPRESENTED BY GALLERY COLLECTION
Recuay Descriptive Shape

Recuay
Shape
Frag
Frag-
ments
Per
Cent
| Coi-
Com-
plete
~~~~~~~~~~Estimate
Of Frag-
ments
(Recuay A-1) Plain open-bowl . . . . . . . . . 12 4.49 (2)

(Recuay A-2) Annular-base open-bowl . .14 5.24 1 (2)
(Recuay A-3) Trumpet around bowl . . . . . . 1
(Recuay B-1) Conical-handle bowl . . . . . . . 1
(Recuay B4) Conical-handle dipper . . . . . . 1
(Recuay E) Globular body with relief . . . . . 14 5.24 (1)
(Recuay F) Flare-neck vase . . . . . . . . . 1
(Recuay E) Flare-rim deep bowl . . . . . . . 3 1.12 (2)
(Recuay G) Disc-rim jar . . . . . . . . . . 1 .37 (1)
(Recuay H) Collar and bridge to head . . . . . 21 7.87 (2)
(Recua'y H-1) Collar gnd bridge to head . . . . . 28 10.49 1 (2)
(Recuay H4) Head under roof . . . . . . . . . 37 13.86 (1)
(Recuay I) Connected fruit . . . . . . . . . 4 1.50 (1)
(Recuay 0-2) Straight trumpets . . . . ... . . 88 32.96 (7)
Unidentified sherds 45 16.85
Totals . . . . . . . . . . . . 267 100.00 6 (21)
one belongs to this vessel. Also, a short cylindri- style is not represented, and that Tello's An-
cal body-spout seems to be from the same dean Archaic style is likewise missing. On the
vessel. The other piece, from a different gallery, other hand, this negative evidence is not suffi-
is a constricted-collar with red lines on the white cient to validate either Kroeber's or Tello's
slip. styles. An attempt to glean a few more crumbs
Red-on-white, 4 fragments. All four are from of interpretation from these gallery collections
one vessel which represents four connected is made later in the general discussion of Recuay
oval fruits, painted alternately red and white style.
(Recuay I).
Black outside, red inside, 2 fragments. H. HOUSE-GALLERY STRATIFICATION
Site 7H-5A-B has been reserved for discus-
Metal Artifacts sion here, as it presents the one good example
Three galleries contained fragments of cop- of construction stratification (Fig. 16). Located
per. One piece is a pin with a wide disc head, on the slope below Ichik-Wilkawain was a four-
like a nail. room two-story stone house, partially covered
Stone Artifacts by wash from above, so that it had the appear-
ance of a mound. The lower floor was composed
Two beads, one of disc type and one with of two long narrow rooms, 6.13 meters long,
groove, were the only stone objects found. 1.3 meters high, and slightly different in width,
one being 1.1 meters and the other 1.23 meters.
OBSERVATIONS ON SUBTERRANEAN GALLERIES
AND COLLECTIONS
The ceilings of both rooms were formed of slabs
resting on stones which projected from the side
Both in architecture and collections recov- and center walls. The two rooms were con-
ered the galleries form a consistent unit. The nected by a doorway through the middle of the
general construction of the subterranean gallery 50-centimeter thick central wall. Opposite this
and the particular type of slab-wall are distinct doorway, at the north, was the original en-
from all other building units in the vicinity. trance to the building. The second floor which
Likewise, these subterranean rooms have the was caved in had two similar rooms roofed with
only architecture which seems clearly identified large slabs which had once been capped with
with Recuay style. The function of the galleries dirt and stone.
presents a problem. They do not seem practical Excavation in the fill on the first floor re-
as dwellings, because although well made, they sulted in a collection quite similar to the gen-
are damp and without proper ventilation. Fur- eral material from the deep tombs, as will be
thermore, the ceramics are essentially of grave seen in the analysis which follows. At one
type. The absence of house-type refuse or utili- point, the floor had caved in slightly, revealing
tarian wares is marked, in spite of the fragmen- the cover-slabs of a subterranean gallery be-
tary nature of the collection. Some were neath the house. Considerable effort was re-
certainly used as graves as witnessed by the quired to break through the roof, but once
presence of powdered bones. The fragmentary entered it proved to be not essentially differ-
nature of the ceramics is difficult to explain. ent from those previously described. It is in-
While most of the galleries had been previously cluded in the list of galleries as Site 7H-5B.
opened, the two which had the entrances intact It extended some 20 meters in length, 1.48 me-
also produced fragmentary pottery or nothing. ters in width, and 1.2 meters in height. It was
The ceramics are clearly of the Recuay style. actually under the north wall of the house,
None of the wares found in other sites in the rather than directly under the north room. The
vicinity is represented. In fact the unique type entrance had once been on the far side of the
which occurs in other sites is the straight, house, but had long since caved in. The rela-
orange trumpet. Otherwise no Tiahuanacoid tionship of the house to the gallery is purely
influence, none of the typical black-orange neg- fortuitous.
ative, no blackware, redware, or other style is
represented. The collection is unfortunately too THE HOUSE COLLECTION
small to confirm any subdivision of Recuay Some burial material seems to be mixed with
style. It is obvious that Kroeber's Recuay B the house collection, although everything ex-
cept the toy vessels was broken. Most of the 10 Pieces of polished redware, including three
material appears more like house than burial pieces of an open straight-sided bowl
refuse. As such, it follows the general type of 4 Pieces of a constricted-collar flask-shaped ves-
the deep tombs. sel, decorated in black-orange negative design. These
four pieces and the toy plate mentioned above are
7 Toy vessels complete. Except for a small plate the only negative painted pieces.
which has a negative, linear design inside, all are 1 Large fragment (18 by 13 centimeters) of
crude orange or brown, without decoration. Four toy 7-millimeter thick brownware represents a section of
ollas with two horizontal handles, a plate, a shallow a face with details in incision and relief (Fig. 10f).
tripod vessel and a vase with two off-center vertical The eye, a raised dome, surmounts a stepped tear-
I Ia L.
0 'I
2. 3
METERS
FIG. 16. Groundplan and cross-section of house-gallery stratification, Wilkawain.
body-handles and a body-lug comprise the toy ves- band. The nose is in high relief. Incised step designs
sels. cover the nose and the rest of the face. Above the
1 Complete orangeware olla with globular body, eye are three incised, stylized puma heads. The
flare-rim, and two horizontal flat-loop handles Tiahuanacoid influence is obvious.
100 Orangeware fragments from high-collar ollas 4 Pieces of black, white-on-orangeware repre-
2 Extra-thick orangeware fragments sent a constricted-base vessel with animal heads
1 Orangeware conical handle or tripod leg projecting from each side. A fragment of one of these
7 Plain brownware fragments heads has the crossed fangs. The vessel is of a type
16 Polished blackware fragments, including nine found in the Coast Tiahuanaco periods.
from one goblet with a thick rim edge and three relief 3 Miscellaneous black, white-on-orange frag-
faces (Fig. 6f0, measuring 8.5 centimeters in diameter ments, one representing a flat handle, one a bird
and 8.5 centimeters high. Six other pieces are from head, and one with grooves
the base of a blackware vase; one is a blackware 1 Partially complete orangeware figurine, 8 cen-
spout. timeters high, 2.6 centimeters wide at the head, and
about 1.5 centimeters thick. It represents a solid A comparison of grave and refuse material
female figure, with separated legs, and two hands on presents a difficult problem, further compli-
the chest. cated by the fact that much of the grave ma-
THE GALLERY COLLECTION terial from the subterranean galleries and the
Only ten fragments were found in the gal- deep stone-lined tombs is incomplete. Conse-
lery, but all of these are consistent with the quently, the calculation of a straight percentage
Recuay materials found in the other galleries. of occurrences for comparative purposes be-
comes difficult in that sherd units versus com-
2 Fragments of plain brownware plete or partially complete vessel are not
3 Fragments of straight, orangeware trumpets equivalents. It seems advisable, therefore, to
5 Fragments of black-white-red negativeware, all
with segments of complex designs of Recuay type base the comparison on the presence or absence
of certain selected diagnostics. This method has
On the basis of this site, it is clear that the its weaknesses, as will be apparent, but is cer-
mixed Tiahuanacoid style of the deep tombs is tainly better than forcing categories for statisti-
more recent in time than the Recuay style. cal purposes.
The collection from the subterranean galleries
OBSERVATIONS ON THE WILKAWAIN REGION is consistent, representing the Classic Recuay
In the preceding pages the collections from style. It is probably the earliest material found
the Wilkawain region excavations have been in the excavations. In Sites 7H-5A and 5B the
described. An attempt to organize the miscel- house site material was found superimposed on
laneous materials follows, although with com- a subterranean gallery. Furthermore, the Re-
plete awareness that in many cases the final cuay style, as represented by the galleries, shows
evidence is lacking. virtually no connection with the other collec-
In terms of the ceramic collections, there are tions. Possible exceptions are the straight clay
five major and four minor sections to be con- trumpets and the rare occurrence of conical
sidered. handles. Both of these are found in the mixed
1. Sites 7H-1, 2, 3, 4. The materials from the collection from the subterranean house site,
four deep stone-lined tombs are considered as a 9H-2, but without positive proof that they
unit, representing a single time period. No posi- should be considered contemporaneous with the
tive association of ceramics with building units, rest of the refuse. The plan and type of con-
outside of the tombs themselves, is possible, struction of the galleries, with the slab walls
although the masonry suggests that of the and covers, are distinct from other architectural
temples. features in the region. Actually Recuay style
2. Site 6H-2. The house refuse collection was not found in any other association.
from one of the above-ground stone house sites. SITE 7H-15
Included with this is the material from one
grave, which, while intrusive, is not too distinct The thirteen vessels from the unlined grave
from the rest of the materials. serve to isolate the white-on-red style. This is
3. Site 7H-5A. A collection from a four-room the only design style represented in the collec-
above-ground stone house site, interesting for tion, appearing on eight of the vessels, while the
its stratigraphic position above a subterranean remainder are plain brown or redware. The style
gallery. is nicely isolated in this grave, but little evi-
4. Site 9H-2. A mixed collection from a sub- dence is presented for its relative position.While
terranean house site. not identical in every detail, the shapes, de-
5. Subterranean Galleries. A collection from signs, general carelessness of workmanship, and
nine galleries of Recuay style, treated here as a colors seem related to the white-on-red style at
unit. Chancay and at Chavin. At Chancay the white-
a. Site 7H-15. A collection from an unlined on-red style is definitely pre-Coast Tiahuanaco,
grave representing a white-on-red style. as shown by the work of Uhle and Kroeber4 and
b. Site 6H-3A, 3B. Materials from two box as confirmed by the 1941 excavations ofGordon
graves. Willey. At Chavin the white-on-red style ap-
c. Site 8H-1. Bowls from an unlined grave. pears in intrusive graves of definitely post-
d. Site 7H-7. A bowl from a box grave. 1 Kroeber, 1926, 291.
Chavin period. The recent work of Larco (1941) is not too satisfactory. Consequently, for the
in the Chicama Valley suggests that a related time being, the white-on-red style is placed be-
white-on-red style may be pre-Early Chimu at tween Recuay and the definitely Middle periods.
Salinar, although more evidence is necessary to
prove this. The total effect of this evidence is MIDDLE PERIOD COLLECTIONS
to place the white-on-red style of Grave 7H-15 The remaining four large collections are not
in a relatively early position, at least pre- as easily disposed of as that from the galleries.
Tiahuanacoid Middle period. What would be None except 9H-2 shows any positive connec-
its relationship to the Recuay material is not tion with the Recuay style, nor any trace of
answerable at this time. The only possible evi- Chavin influence. Furthermore, no definite
dence comes from the mixed Site 9H-2, and this Incaic influence can be noted. On the other
WILKAWAIN MIDDLE PERIOD SITES
Sites 7H-1-4 Site 6H-2 Site 7H-5A Site 9H-2
Category Com- Frag- Com- Frag- Com- Frag- Com- Frag-
plete mentary plete mentary plete mentary plete mentary
I II I_
1 Orangeware ....... 2 1053 198 1 103 1 624
2 Brownware ....... 94 7
3 Blackware ....... 14 134 32 16 7
4 Redware. 4 460 2 1 10 90
5 Two-color negative . 2 207 1 4 1
6 One-color positive 1 86 26 56
7 Two-color positive . 3 134 167 7 3 8
(all sspecials)
8 Three-color positive 1 32 10
9 Toy vessels. 25 14 7
10 Shallow plates . . . . . 5 93 16
11 Large plates . . . . 2
12 Annular-base plates 5 35
13 Goblets ......... 3 15 7 4
14 Flask shape . 1 8 4
15 Face-collar vessel. 1 35
16 Constricted-collar .... 4 678 16 82
17 Double-spout ...... 5 21 2 37 1
18 Double-bowl ...... 1 1 19
19 Tripod plate 5 24 1 1? 1
20 Sieves . . . . ... 1 9 22
21 Modeled puma bowl 31 1
22 Straight spoon . 3 7
23 Thick black, white-on-red . 32 118
24 Tiahuanacoid designed 5 50 43 8
25 Open round-base bowl 13 2
26 Flaring-sided cup . 3 3? 9
27 Cursive-design spoon . 1 8
28 Loop-handle spoon. . 1 1 1
29 Pressed-relief design . 8 9
30 White-on-redware . . 3 11
31 Conical handle ..... 1 1? 2
32 Annular-base bowls . 3 1
33 Straight clay trumpet 23
34 Brown-red, orange-on-gray-
ware. 3 1
35 Grayware . . . . .. . I
19
hand, three of the collections are definitely suggesting any major change is added in the
comparable, in spite of certain local differences, house site collection, but rather, the U-shaped
with the Middle periods as represented on the thick black, white-on-red bowls and the pressed
coast of Peru. In the accompanying table a blackware design technique are further con-
comparison of the four major collections (Sites firmation of the Coast Middle period relation-
7H-1-4, 6H-2, 7H-5A, 9H-2) is made in terms ship. In final analysis, it seems impossible to
of 35 diagnostics. A discussion of this table fol- differentiate the two collections in any but in-
lows, starting with the deep grave collection. significant detail. Even if the three pieces in the
intrusive grave were given exaggerated impor-
DEEP-TOMB COLLECTION tance, the only significance achieved would be
Since there is no question as to the contem- to place the refuse collection as slightly earlier
poraneity of the styles represented, the collec- than that from the deep-tombs, but still well
tion from the four deep-tombs forms a basic within the generalized Middle period.
unit. In the table, the first 24 diagnostics are HOUSE SITE 7H-5A
typical. The position of the collection in the
general Peruvian framework of Middle periods The material from the four-room house site
is quite evident, as mentioned previously. The is difficult to use for comparison because of its
combination includes blackware; redware; paint- relatively small quantity. Even so, it presents
ing in one, two, and three colors on a red or no startling differences from either of the other
orange slip base; the particular color combina- two collections under discussion. Eleven of the
tions emphasizing black, white, and red; the deep-tomb characteristics are found and like-
Tiahuanacoid designs plus other influence, as wise ten of the significant features of house site
depicted by shapes and colors. All is consistent 6H-2. The one large fragment of incised brown-
with Coast Middle periods, particularly as rep- ware has a design more clearly Tiahuanacoid in
resented by Middle Ancon I and related styles. style than any piece from the other two collec-
Obviously the correspondence is not absolute. tions.
Not only are some Coast characteristics lacking, SUBTERRANEAN HOUSE SITE 9H-2
but two-color negativeware, tripod plates,
sieves, toys, painted spoons, and other de- It has already been emphasized that this col-
tails are added. In total, the collection pre- lection may not represent a contemporaneous
sents a sharp contrast to the Recuay style as unit. None the less, it seems to fall roughly into
found in the subterranean galleries. Considering the Middle period category by virtue of the
the Recuay style as "early" for the Wilkawain absence of any typical Early material, as repre-
excavations and the deep-tomb style as "mid- sented by Recuay style, or of any positive Late
dle," one is impressed by the gap which sepa- material, as represented by Inca style. How-
rates the two and emphasizes the coastal paral- ever, the comparative table emphasizes the
lels. differences between the main body of the sub-
HOUSE SITE 6H-2
terranean house collection and those from the
other three Middle sites. Thus, while orange-
The house site refuse and the three vessels ware, blackware, and redware are still present,
from the intrusive grave are treated as a single a number of significant characteristics are lack-
unit in the comparative table. While this ma- ing. These include negative painting; positive
terial is essentially refuse, the close relationship painting in more than one color on a base slip
to the deep-tomb collection is none the less (there are three complete and eight fragments
apparent. Of the 24 diagnostics which distin- of two-color, but all of these are special styles
guish the deep-tomb style, nineteen are likewise which do not correspond to the other sites);
characteristic of the house site collection. It is tripods; toys; sieves; annular base plates; and,
true that the redware and two-color negative- most important of all, the majority of the char-
ware are represented by the intrusive grave acteristics suggesting Tiahuanacoid influences.
pieces only, but otherwise the characteristics of Actually this collection contains neither the
the deep-tomb style which were comparable colors, shapes, nor the design elements which
with coastal Middle periods are all found in the were used in the other three collections as indi-
house site refuse. Furthermore, no new style cators of Tiahuanacoid affiliations. Of the 24
characteristics of the deep-tomb collection, only Site 6H-3A, 3B. The nine vessels from these
seven are found in the subterranean house col- two stone-lined box graves are consistent with
lection, and these are the least distinctive ones. the collection from the deep-tombs and from the
The subterranean house collections add cer- above-ground house sites in shape, colors, and
tain new styles not found in the other collec- designs.
tions. Most distinctive is a grayware pottery Site 7H-7. The single brownware incised bowl
with design in brownish-red on a light orange from a stone box grave is left in its unique posi-
band over the gray base. This style is associ- tion. Not only is it a distinct style from that of
ated, furthermore, with two new shapes, namely, any other piece found, but absolutely no evi-
an annular-base open-bowl and an annular-base dence is available for even the suggestion of a
cup. The cursive-style spoons are more frequent relative time position.
in this collection, although one was found in the
deep-tomb collection. White-on-red style frag- R£SUMi
ments of the type isolated in the grave of Site 1. Recuay Style (Sites 7H-SB, 7H-8, 7H-10,
7H-15 are represented here, while only three 7H-12, 7H-13, 7H-14, 7H-16, 1OH- 1, 1OH-2). It
dubious fragments appear in the other collec- is associated with subterranean galleries and is
tions. demonstrably earlier than the Wilkawain Tia-
While it is easy to establish the differences huanacoid style.
between the subterranean house site material 2. White-on-red Style (Sites 7H-15, 9H-2).
and the other three sites, the relative chrono- Although isolated in one grave and also repre-
logical position of the two groups is more sented in the mixed collection from the sub-
difficult. In spite of the lack of typical Recuay terranean house, the tentative chronological
material, it has already been pointed out the position is, by analogy, with the style in other
straight trumpets, conical handles, three-color Highland and Coast sites.
negative, and slightly modeled red-on-white- 3. Subterranean House Mixture (Site 9H-2).
ware are all found in Recuay collections, and
are lacking in the other Wilkawain Middle While the materials are probably not contem-
period sites. The presence of white-on-redware poraneous, a position between Recuay and
would, by the previous argument, also suggest Wilkawain Tiahuanaco is indicated by some
a slightly pre-Middle position. Without insist- slight Recuay resemblances, the presence of the
ing that all styles in this collection are contem- white-on-red style, and the absence of Tia-
poraneous, or that further evidence will not huanacoid influences and of those of still later
affect the analysis, the materials from Site 9H-2 periods. The new style of reddish-brown,
are here considered as slightly earlier than the orange-on-grayware is distinct in color, de-
Tiahuanacoid-influenced collections from other sign, and shapes of vessels, and quite possi-
Wilkawain sites. bly represents a later intrusion.
4. Wilkawain Tiahuanacoid Style (Sites
SMALL GRAVE COLLECTIONS 6H-2, 6H-3A, 6H-3B, 7H-1, 7H-2, 7H-3, 7H-4,
The material from the four graves can be 7H-5A, 8H-1). It is well isolated by its associa-
treated briefly, since obviously it is insufficient tion in the deep stone-lined tombs with slab
in quantity to affect deeply the interpretations covers, in stone box graves, in unlined graves,
based on the large collections. and in the above-ground house sites of one to
Site 8H-1. The modeled puma bowl and the four rooms. The architecture of the house sites
orange olla with pressed-relief design, found in and graves also suggests the association of the
an unlined grave, quite obviously fit into the three-storied temples. If this is true still another
general group represented by the deep-tomb characteristic can be added, namely, carved
collection. If anything, this grave serves further stone puma heads with tenons for wall insertion.
to unite the deep-tomb collection with that The closest affiliations of the style are with
from the above-ground house sites, since it com- Coast Tiahuanaco periods such as the Epigone
bines pressed-relief (absent in the deep-tomb) at Pachacamac, Middle Ancon I, and the Supe
with the painted and modeled puma (charac- Epigonal.
teristic of the deep-tomb). Furthermore, both 5. Late period materials were not found in
pieces are typical of Coast Tiahuanaco sites. the Wilkawain region excavation.
OTHER SITES NEAR HUARAZ
Sites in the Huaraz region have already been 2 Rims of somewhat constricted neck ollas
listed. Aside from the Wilkawain section, three 4 Short vertical rims of ollas with horizontal
other sites near the pueblo of Huaraz deserve flat-loop handles
more detailed attention because of the excava-
3 Fragments of shallow plates
tions carried out.
3 Fragments of convex-sided open-bowls (Re-
cuay A variant)
4 Pieces of round-loop handles
SHANKAIYAN, SITE IH 4 Pieces of conical handles of Recuay B vessels
In the fields on this hill near Huaraz three 1 Complete orange open-bowl, 7.5 centimeters
subterranean galleries were visible, and more in rim diameter and 3 centimeters deep, with two
were reported as previously opened. One meas-
bulges on the body (Fig. 17h)
ured 4 meters long, 82 centimeters wide, and 2 Miscellaneous fragments, one representing
crossed arms, and one decorated with simple incision
1 meter deep with the cover-stones just under
the surface. The side walls were formed of three Thin orangeware, 43 fragments. An orange-
large slabs each, with the intervening cracks ware like the above, but much thinner and pos-
filled with small stones. At each end was a large sibly once decorated with painted design,
upright slab, and four more formed the roof. although no traces are visible on these pieces.
No evidence of an entrance was seen.
24 Miscellaneous body pieces, including one flat
IH-A
SITE base
8 Rim pieces of an open-bowl with straight di-
A subterranean gallery 5.6 meters long, 1.05 verging sides
meters wide, and 1 meter deep to the packed 5 Pieces of a vessel with constricted cylindrical
clay floor had three covering slabs at the north neck
half starting 10 centimeters below the ground 6 Fragments of straight clay trumpets with flar-
surface. This north half had apparently been ing mouthpieces (Fig. 18d, i)
excavated previously. The south half was filled Plainware, 7 fragments.
with dirt and house refuse debris, but without Redware, 4 fragments. A thin polished red-
any trace of covering slabs. Excavation in this ware including a rounded base fragment, two
south half produced one small complete vessel straight rim pieces, and a body piece.
and 570 sherds. No bones or other indication of Whiteware, 44 fragments. All pieces are of
burial were found, nor is the material of the white clay and quite probably were once part
grave ceramic type. On the other hand, the of negative painted vessels. As it is, the pieces
open stone-lined gallery does not seem impres- included here show no remains of painted de-
sive as a dwelling. sign.
Materials from Gallery 1H-A 36 Miscellaneous body pieces
1 Piece of flat base with vertical sides
The materials from this gallery can be de- 4 Fragments of large flat rims of Recuay G type
scribed briefly in terms of wares, with indica- 1 Body fragment with elongated wing lug
tions of the shapes represented. 2 Pieces of flare-collars
Orangeware, 1 complete, 265 fragments. All Two-color negative, 1 fragment. One black-
pieces are rather thick, crude orangeware with- orange negative piece was from a shallow open-
out slip or painted decoration.
bowl (Fig. 18c).
249 Fragments of olla types Three-color negative, 27 fragments. All pieces
215 Plain body fragments are black-white negative plus the addition of
10 Base pieces showing five rounded and five red. The base clay is in all cases whiteware. All
flat bases are typical of Recuay style.
18 Rim pieces indicating flare-rims in 13,
and flare plus a high collar in five. Five 8 Miscellaneous body fragments
of these have shoulder nubbins either 11 Pieces of one modeled figure with a complex
plain or oval with groove (Fig. 18e). geometric design on both sides (Fig. 18a). The figure,
1944 BENNETT: TWE NORTH HIGHLANDS OF PERU 55
-- - -
---
A B
C.
D E
........
H
FIG. 17. Recuay style artifacts from gallery at Shankaiyan.
De
I . - I
I
It
t i.I
. i.
t,
I
.I '...S 1-
I U
FIG. 18. Recuay style artifacts from gallery at Shankaiyan.
while incomplete, seems to be the central one of a 10 Body pieces probably from shallow open-bowls.
group on a flat-topped vessel of Recuay J style. Three are painted inside with parallel bands; two are
2 Pieces of a flat-topped vessel with a cut-out painted outside with band and wavy line (Fig. 17c).
opening behind, and two modeled birds in front The other five have linear designs.
(Fig. 18f) 1 Fragment of an annular base open-bowl of
1 Body fragment with a short projecting spout Recuay A-2 style with linear design (Fig. 17a)
3 Pieces of a long wing lug with simple triangular 3 Fragments of an open-bowl (Recuay A-1) with
designs crossing red bands. Five fragments of a similar bowl
1 Piece of an open-bowl with line and dot design with zigzag bands (Fig. 17b). Three fragments of a
1 Head disc of a modeled figure similar bowl with fret-like rim lug, and decoration of
horizontal lines and wavy lines.
1 Piece of a somewhat irregular open-bowl (Fig.
White slip on orange clay, 87 fragments. 17d)
Orangeware fragments are covered with a thin 1 Fragment of a rounded handle
white slip. Apparently in some vessels the slip
does not cover the entire body, but rather out- Black-white-red, 3 fragments. Three small
lines a simple design.
modeled figures are painted in black-white-red
72 Body fragments of olla-like vessels, including (Fig. 18g). All are of a type found on flat top ves-
one flare-rim and five rounded base pieces sels of Recuay J style.
2 Fragments of angular-bodied vessels Spoons, 1 fragment. One bowl of an elongated
5 Pieces of open, convex-sided bowls clay spoon with a flat handle is painted in black
6 Fragments of decorative bulges on body pieces
1 Constricted cylindrical neck piece and red on an orange clay (Fig. 18h).
1 Piece of headdress of modeled figure Metal artifacts, six pieces. Six copper pieces
were found, including a needle 8.7 centimeters
Black-on-orange, 9 fragments. Seven pieces long; a pin with a nail-head top (Fig. 18j); a
are from one round-bottom open-bowl with fragment of a disc; a 5-centimeter long rounded
three parallel black lines, and two are body bar hammered flat at each end; a flat arm,
fragments. 5 centimeters long, ending in a hand with the
Red-on-orange, 37 fragments. Most of these middle finger curled over; and a flat bell-shaped
are miscellaneous body pieces (30) with red knife, 3.8 centimeters wide at the blade and
lines or wide bands on an orange clay. Besides 5.7 centimeters long.
the body fragments the following can be noted: The collection as a whole shows obvious
affiliations with Recuay style, in spite of some
1 Piece of horizontal loop handle, rounded variants. The discussion is reserved to follow
1 Flare constricted-collar the description of the second subterranean gal-
1 Piece of flare-rim lery excavated.
4 Rims of open convex-sided bowls, three of which
were painted inside SITE 1H-B
Red-on-white, 18 fragments. An orange clay About 3.4 meters west of the above-de-
is covered with a white slip on which a design of scribed gallery was found another which had
red lines or bands is imposed. not been previously disturbed. It was 3 meters
11 Pieces of flare-rim bowls, including two rim long in total, from 75 to 85 centimeters wide,
fragments and 75 centimeters deep to the clay floor. The
6 Pieces of two open-bowls of Recuay A-1 style. cover-stones were about 35 centimeters below
One is decorated with a wide line forming an angular the ground surface, and the spaces between the
scroll, and the other with bands crossing each other. slabs were filled with smaller stones. The side
1 Body piece with a short tubular spout and sim- walls were built of small and large stones, not
ple red line design as well matched as in other galleries. The
groundplan is in the form of two rectangular
Red-on-white, 24 fragments. A base white sections which form an angle. Fifty fragments
clay is decorated with wide red lines or bands. and five complete or restorable bowls were
Most of the pieces are from open-bowls. found but no bones nor other indication ofburial.
Materials from Gallery 1H-B and design of groups of parallel lines. Two pieces of a
similar vessel with more convex sides, and with a
Plainwares, 31 fragments. These include design consisting of straight and wavy horizontal
seven crude orange from olla bodies; seven lines. Two pieces of a similar vessel with linear deco-
thin orange, six from globular vessels and one a ration both inside and out.
flat base with two feet lugs; nine orange clay 1 Flat top of a vessel with the bases for two mod-
with red slip from the flat base of one globular eled figures and traces of red paint
vessel; two redware pieces, one a flat handle; Metal artifacts, two complete. Two copper
and six whiteware, including a flare-rim, a pieces were found, one a needle 7.5 centimeters
cylindrical neck, three body pieces, and a sec- long, and the other a long pin with a flat disc,
tion of a modeled vessel. nail-top head. The pin is 19.2 centimeters long
Three-color negative, 4 fragments. One piece and the head is 3.5 centimeters in diameter.
is from the wide flat rim of a Recuay G style
vessel; one is a modeled body spout; and two SITES 1H-C, 1H-D
are body fragments.
White slip-on-orange, 1 complete (PI. 2d). A Two stone box graves were found, one 55 by
conical-handled vessel of Recuay B-2 style with 45 centimeters by 1.1 meters deep, covered by
slightly flaring rim and short tubular spout. The two large stones. The other was 50 by 50 centi-
body of the vessel is 10 centimeters in diameter, meters and 58 centimeters deep, without cover-
8.3 centimeters at the rim, and 7.5 centimeters stone. Both contained fragments of infant
high. The handle is solid, 5.5 centimeters long burials.
and 3 centimeters at the base. The spout is 1.5 SHAPES REPRESENTED IN SHANKAIYAN GALLERIES
centimeters in diameter and projects for 1.7
centimeters. Traces of white slip are the only Site 1I-A Site 1H-B
signs of decoration on the orange clay.
Red-on-orange, 2 complete, 7 fragments: Shape Com- Frag- Com- Frag-
Complete shallow open-bowl, 15 centimeters in
1
plete ment plete ment
rim diameter and 4.5 centimeters deep, with sets of
short parallel red lines around the outside of the rim (Recuay A-i) . . 12 3
on the orange clay
(Recuay A-2) . . . 1
1 Complete bowl of a dipper (Recuay B-4 style)
(Recuay A, variants). 1 35 1 7
with the handle missing (Fig. 17e). The design is on (Recuay B-2) . . . . 2
the inside of the bowl and consists of a simple cross, (Recuay B-3) . . . . 1
plus two stylized jaguars (Fig. 17f). (Recuay B-4) . . . . 1
Conical handles . . . 4
3 Pieces of one open-bowl of Recuay A-1 style (Recuay G). 4 1
with parallel red lines around the outside near the (Recuay J) . . . . . 14 1
rim (Recuay 0-2) . . . . 6
2 Pieces of a globular bowl with flare-rim Shallow plate . . . . 3
2 Miscellaneous body pieces Cylindrical neck. . 7 1
Body spout . . . 2 1
Wing lug on body . 1
Red-on-white, 2 complete, 8 fragments: Base fragments . . 17 9
1 Complete conical-handled bowl of Recuay B-2
Body fragments . . . 397 21
Angular body . . . 4
style (Fig. 17g). The bowl is 9 centimeters in diame- Wide flare-rim of olla . 20 3
ter and 8.5 centimeters deep, with a flare-rim, a Constricted flare collar 6
curled-up conical handle, and a short tubular spout. Straight olla rim. . . 4
The design is in cross-hatch lines of reddish-brown on Modeled . . . . . . 8 2
the white clay. Round handle. . . . 6
1 Restorable dipper-bowl (Recuay B-3 style) with
Spoon . . . . . . . 1
red bands on the white clay.-The handle of this vessel Miscellaneous . . . . 18 1
was found just above the cover-stones and the body
pieces inside. Totals . . . . . . 1 570 5 50
3 Pieces of an open-bowl with fret-like rim lugs
RECUAY STYLES FROM GALLERIES
Shankaiyan Wilkawain
Sita 1H-A Site 1H-B All Galleries
Ware Fragments Fragments Fragments
Complete Complete Complete
No. Per Cent No. Per Cent No. Per Cent
Crude orange . . . . 1 265 46.49 7 14 91 34.08
Thin orange . . . . 43 7.54 7 14
Redware . . . . . . 4 .70 11 22
Brownware. . . . . 28 10.49
Plain . . . . . . . 7 1.23
White . . . . . . . 44 7.72 6 12 1 .37
Two-color negative . 1 .18 1 9 3.37
Three-color negative. 27 4.74 4 8 78 29.22
White slip-on-orange. 87 15.26 1
Black-on-orange. . . 9 1.58
Red-on-orange . . . 37 6.49 2 7 14 1 14 5.24
Red-on-white slip . . 18 3.16 38 14.23
Red-on-white clay. 24 4.21 2 8 16 1
Red-on-gray . . . 1
White-on-red 2
Red-and-white . . . 4 1.5s
Black outside, red in-
side . . . . . . . 2 .75
Black, red-on-orange, 1 .17
Black-white-red posi-
tive . . . . . . . 3 .53 2 .75
Totals . . . . . 1 570 (100) 5 50 (100) 6 267 (100)
OBSERVATIONS ON THE SHANKAIYAN GALLERY represented in the Wilkawain collection can be

COLLECTION matched in the various Recuay collections. The
The affiliations of the pottery from both the shapes from Shankaiyan, on the other hand,
Shankaiyan galleries are undoubtedly with Re. present variants and styles not ordinarily as-
cuay style, by virtue of colors, negative paint-
sociated with Recuay. None of these variants,
ing, shapes, and some designs. At the same time however, suggest closer affiliation with any of
Shankaiyan presents a great quantity of utili- the middle styles of the Wilkawain region. Still
tarian material not ordinarily found in Recuay no Tiahuanacoid influences are seen, nor any
collections. This contrast holds not only for the other direct tie-up with Middle periods. Thus,
selected display material of most collections, for lack of further evidence, the Shankaiyan col-
but also for the material from the nine galleries lections are considered as a sample of refuse site
of Wilkawain. Plainwares are rare in the material of general Recuay style.
Wilkawain galleries. The total number of pieces, SAN JER6NIMO, SITE 2H
119, actually includes 88 fragments of orange
and brown straight trumpets which, if elimi- On the hill just across the Santa River from
nated, would make the contrast with Shan- Huaraz, a surface collection of sherds was made,
kaiyan more striking. Most of the shapes although no excavation was carried out. A cata-
logue list of these sherds is included here. The 1 Red, orange-on-white outside, red inside with
variety of colors, their combination, the geo- band design
metric design, the habit of using different color 13 Black and red-on-white outside, white inside;
schemes on the inside and outside of vessels, the five rim, one base, seven body; four horizontal
parallel lines, two vertical parallel lines, one disc,
hard firing, and the few suggestions of shapes, two line and circle, one hook, three complex geo-
all seem to indicate an essentially Incaic affilia- metric designs
tion for most of these pieces. (See Fig. 3la-p for 1 Black, red-on-white outside, orange inside; a
style.) rim piece with horizontal parallel lines
2 Black, red-on-orange outside, white inside; a
7 Orange slip, inside and outside, including four rim and a body piece with linear design
body, one lug, one flat handle, one incised piece 2 Black, red-on-orange outside, orange inside; a
3 Blackware pieces rim and a body piece with linear design
1 Redware, unpolished flare-rim 1 Reddish-brown, white, black, yellow on out-
7 Red polished slip on orange clay, including two side, white inside; body piece with horizontal and
rims of open-bowls, two thick rims with angular vertical stripe design
ledge, two base and one body fragments 1 Red-on-orange spoon fragment with round
1 Yellow clay rim of large vessel handle
3 Brownware, one with grooved nubbin, one in-
cised, and one with stipple AYAPAMPA, SITE SH
1 Grayware section with two raised ridges On this ridge, some 5 miles east of Huaraz,
4 White line on red slip, including a rim and two a number of above-ground houses, commonly
body pieces with linear design and a collar base with
band design called chullpas, were examined. All correspond
2 Black-on-red slip, one linear and one disc de- to one type of house and vary only in details.
sign A description of one house is sufficient for the
14 White-cream clay, inside and outside, with group. One chullpa (Fig. 19), then, was rect-
seven rims of open-bowls, two flare-rims, and five angular, measuring 1.85 by 1.9 meters on the
body pieces outside and standing some 1.35 meters high. It
14 White outside, orange inside, with two open- contained only one room, without subdivisions,
bowl rims, one flare-rim, and eleven body pieces niches, projecting stones, or other interior modi-
7 Orange outside, white inside, body pieces fication. A doorway to the east measured 50 by
3 Orange outside, red inside 50 centimeters and was topped with a single
6 Red-on-white outside, white inside; with four stone lintel. The roof was formed of one long
rims from open-bowls and two body fragments. stone laid across the north and south walls at
Three have parallel horizontal lines, one a line and the center. Shorter stones were laid with one
triangle, one a band, one angular lines. end on this center ridge and the other on the
3 Orange-on-white outside, white inside, all linear east and west walls. The slightly gabled effect
with two open-bowl rims, one body piece thus produced was obscured by a heap of dirt
4 Black-on-orange outside, orange inside, all and small stones which covered the roofing
linear with one open-bowl rim and three body
pieces beams and gave a dome-shaped appearance to
1 Black-on-orange outside, red inside, with the whole roof. The walls of the chullpa were
parallel lines on open-bowl rim composed of rough stone, both large and small
3 Black-on-orange outside, white inside with one stones being utilized but with little plan. Mud
diamond and two linear designs; two open-bowl rims, or clay was used as a cement.
one body piece Other houses vary in size. Some are placed on
5 Black-on-white outside, reddish-orange inside; large natural boulders. Some have low enclosure
three rim and two body pieces; two horizontal paral- walls surrounding them, or low platforms along
lel lines and one vertical, and two semi-curvilinear an outside wall. These differences can be briefly
designs tabulated.
6 Black-on-white outside, white inside; four rim House 1. (Described above as type.)
and two body; four horizontal parallel lines, one House 2. A one-room, rectangular house,
square and dot, and one band design 1.7 by 2.25 meters and 1.80 meters high. The
door is at the south and 55 by 75 centimeters. wide flat handle from rim to body. It is 7 centimeters
Along the west side is a platform 1.45 meters in body diameter, 4 in rim diameter, and 9.5 centi-
wide and 50 centimeters high. meters high.
House 3. A house 1.85 by 2.7 meters and 1.3 1 Redware shallow plate, 8 centimeters in diame-
meters high, with a small doorway at the east. ter by 3 deep
House 4. A house 1.9 by 1.8 meters and 1.9 1 Redware open-bowl, 6 centimeters in diameter
and 2 centimeters deep
meters high. The doorway at the east is 80 by 2 Shallow tripod redware plates, about 9 centi-
60 centimeters. A platform 1.1 meters wide is on meters in diameter and 4 centimeters high
each side of the house. 1 Small orangeware sieve, 8 centimeters in di-
House 5. A house 3.4 by 2.1 meters and 2.2 ameter by 3 deep
meters high. The doorway at the south is 50 by 1 Toy orange olla with two loop handles
75 centimeters. About 3 meters away from the
house on all sides is an enclosure wall made of
a single row of stones set on edge.
House 6. A house 2.6 by 2.4 meters and 2.2
meters high, with a south doorway, 50 by 50
centimeters.
House 7. A house 2.6 by 2.9 meters with the
door at the east and a side platform.
House 8. A fallen house 2.2 by 2.5 meters,
with door at the east.
House 9. A house 2.4 by 2.75 meters and 2.3
meters high. The door, 55 by 60 centimeters, is
at the east. A platform 50 centimeters wide and
25 centimeters high is on all sides.
House 10. A fallen house 4 by 3.5 meters.
Other houses in the region were seen, but not
measured. Six were noted as badly destroyed,
and about nineteen others were noted briefly.
Miscellaneous terraces, platforms and uprights
were also seen in the region of Ayapampa. One 0
ti,
upright stone stood in the center of a stone
platform, 4.1 by 2.2 meters. It measured 80 by
50 centimeters and 1.8 meters high. Two other
uprights were placed 1.4 meters apart on a plat-
form 6.5 meters long. Still another 60 by 60
centimeters and 1.6 meters high stood on a
platform 1.3 by 2.1 meters. The purpose of the FiG. 19. Groundplan and cross-section of
uprights was not clear. house at Ayapampa.
One house, No. 11, was partially cleared out.
It rested on a large natural boulder and meas- While the data on this burial are not ade-
ured 1.65 by 1.27 meters and 1.3 meters high. quate enough for any sound conclusions, the
The roof had fallen in and the doorways were material certainly resembles that described for
completely covered. Excavation along the floor
uncovered two flexed skeletons and eight small the Tiahuanacoid Middle period at Wilkawain,
vessels: as represented by the deep-tomb collection. In
spite of this one burial, it does not seem likely
1 Orangeware jar with a straight-collar with sim- that the houses were constructed for that pur-
ple relief face, and two flat-loop body handles. It is pose. All the others examined disclosed no
7.5 centimeters in diameter and 12 centimeters
high. burial remains, nor any sign of having been
1 Blackware pitcher with constricted-collar and deliberately sealed up.
THE CARHUAZ REGION
Our work in the Carhuaz region was limited COPA CHICA RUINS
to an examination of the ruins at Copa Chica The ruins occupy a long hill, above and
and the neighboring hacienda of Copa Grande. northeast of the hacienda house. The glacial
Much excavation has been done on these two peaks of Yan-Raju and Huascarin form a back-
haciendas which, although nothing has been ground, and the fertile fields of the hacienda
published, is represented by the private collec- spread out below. The long ridge of the hill runs
roughly north to south and is covered with
ruins of houses, walls, terraces, and tombs. In
the houses two definite construction types can
be noted, and within these the structures are
remarkably uniform. In general the construc-
tions of Type B are concentrated on the knoll
. -. at the northern and highest end of the ridge,
while those of Type A are distributed down the
ridge and along the eastern slope (PI. 4c).
HOUSE TYPE A
Thirty-five house units were counted. Sixteen
of these average 4 by 4.5 meters in groundplan
measurements, with a range of 2.6 to 5.9 meters
in width, and 3.3 to 6 meters in length. No one-
room house was seen; two had four rooms and
the rest had two rooms each. Some of these
house units are actually combined to form a
long house with two separate dwelling units, un-
connected internally. The total of 35 house
units treats such combinations as two distinct
units. Other houses may form clusters, within
a few meters of each other, but houses well iso-
lated from each other are, if anything, more
typical. In spite of some difference in measure-
ments and the number of interior rooms, all the
houses are constructed in more or less the same
fashion. Thus the following description of one
house unit may serve for all in Type A.
FIG. 20. Groundplan and cross-section of A typical Type A house measures 3.6 by 5.1
house at Copa Chica, Carhuaz. meters (Fig. 20). A doorway 50 by 50 centi-
meters, covered by a stone lintel, is at the center
tion of Sr. Tom"s La Rosa Sanchez and the of the south side, although in general the orien-
collection in the Universidad de San Marcos tation of the house or the door is not consistent.
assembled by Dr. Tello. The La Rosa Sanchez Within the house a lengthwise central partition
Collection contains numerous pieces of Recuay divides the interior into two rooms. A door,
style pottery, as well as Tiahuanacoid pieces of similar to the outer one, connects the two rooms
the Santa type, and a few decadent Early in the center. The resulting rooms are about
Chimu types. All are said to have come from 1.2 meters wide, 4.1 meters long, and 1.3 meters
the ruins on Copa Chica hacienda. The Recuay high. The rooms are covered by four to five flat
style pieces were classified and will be discussed slabs each, the ends of which rest on the central
in a later section. The following descriptions wall and are built into the outer walls. Above
refer to house structures at Copa Chica essen- these slabs is piled as much as 2 meters of dirt
tially and include a few notes on Copa Grande. and stones, held in place by building up the
The few days' excavation produced no further outer walls. No evidence of a second floor is
information on the ruins. seen, and this accumulation of dirt was evi-
dently intended as protection against the heavy HOUSE TYPE B
rainfalls. Viewed from the end, the roof is The knoll at the north end of the ridge is
slightly peaked. The outer walls are from 40 to covered by houses of an open type. Instead of
50 centimeters thick. The inner side utilizes being spread out as are Type A houses, these
large slabs with the chinks filled with small are close together, utilizing the same walls in
stones and mud. Small niches and projecting many cases. A village plan is laid out in so far
stones are frequently seen inside. The outside as the contour of the knoll permits. Rooms of
of the walls may employ the same slab tech- these houses are of large size, one measuring
nique, but more often alternating rows of large 4.25 by 5.6 meters. The doors are ample; one
and small stones are seen. None of the stones is measures 1.4 meters wide and 1.55 meters high
dressed, although selection and perhaps splitting and is covered with a lintel. The walls are over
are noted. The fact that many of these houses a meter thick and contain niches, but no win-
are intact today testifies to their excellent con- dows. None of the rooms has any remains of a
struction. Although bones were found in some, roof, implying that the roofs were once covered
the houses were dwellings rather than burial by perishable materials. Many houses and ter-
places. Measurements on other houses of this races were seen, and the whole knoll is sur-
type can be tabulated briefly: rounded by a wall still standing in parts, over 1
House 1-2. Two adjoining room units, one meter thick and from 2 to 3 meters high. Two
4.1 by 5.5 meters, the other 2.7 by 4.5 meters, stone gateways remain in this wall (P1. 4b), one
both with doors at the east. 1.35 meters wide and 1.95 meters high; the
House 3. Two-room house, 3.6 by 4.7 meters, other 1.75 meters wide by 1.65 meters high.
with door at the east. An oval wall surrounds Both are covered by stone slabs. The house
the house. walls are built by placing stones of about the
House 4. Four-room house, 4.5 by 4.6 me- same size in irregular rows, while slabs are used
ters, with door at east. only for door facings.
House 5. Two-room house, 5.8 by 6.3 meters, In spite of the contrast in the two types of
with door at the north. houses, there is no good evidence to indicate
House 6. Described as type above. relative antiquity. It is true that the general
House 7. Two-room house, 3.3 by 3.3 meters, style and lay-out of the Type B houses suggest
with a south door. Inca building, but no ceramics were found
House 8-9. Two adjoining two-room units, which would confirm this.
one 5.9 by 6 meters, with a door at the north Tombs have been excavated in many parts
and at the south, the other 5 by 5.1 meters, of this ridge, and a large part of the Rosa
likewise with two doors. A step around the Sanchez Collection is supposed to have been
building is 40 centimeters wide, and a cornice found here. Our excavation encountered five
projects around the roof. stone box graves, but all without contents (P1.
House 10. A two-room house, 4.3 by 4.4 me- 4a). One measured 30 by 50 centimeters and 30
ters with the door at the north. The peaked end centimeters deep, with the cover slab 80 centi-
wall stands 4 meters high. About 3 meters away meters below the surface. Otherwise even sherds
from the house is an enclosure wall. were not encountered.
House 11-12. Two adjoining two-room units,
each 3.7 by 4.3 meters. COPA GRANDE RUINS
House 13-14. Two adjoining two-room units,
each 3.2 by 3.3 meters. Although some ten sites were reported for
House 15. Two-room house, 4 by 4.5 meters, Copa Grande by the owners, Eliseo and Tomis
with south door. Dextre, only a few were visited by us. One,
House 16. Two-room house, 2.6 by 4 meters, called Jatun Allankay, was located along the
with south door. top of a long narrow ridge. Terraces and open
House 17. Four-room house, 4.3 by 4.4 me- platforms were seen, as well as series of rooms
ters, with door at the east and an addition to the without roofs. These were arranged along
west 1.5 by 1.5 meters. streets or passageways, had wide doorways,
House 18. Two-room house, 3 by 4 meters, niches in the walls, and a construction of rough
with south door. stone not arranged in files or rows. In total the
Houses 19-35 seen, but not measured. style is similar to Type B at Copa Chica.
THE KATAK-RECUAY REGION
Katak, the hacienda of Sr. Ludovico Caceres, since the excavation was limited, the presenta-
lies a few kilometers south of Ticapampa and tion here will describe some of the house types
Recuay in the upper Santa Valley. To the south and the excavated material, and discuss briefly
extends the great Pampa de Lampas on which the various sites examined.
plains the Santa has its origin. Inca roads are
reported for this swampy area. Our work here SOME CONSTRUCTION TYPES IN THE
consisted of the examination of a dozen sites in KATAK-RECUAY REGION
the region of Katak and Recuay, and some
minor excavations. While this region is famous Several Type A: Central Subterranean Chamber with
as the center of Recuay style our work con- subterranean Connected Side Chambers. This type of
tributed little to clarify the problems relating to monest in the construction
region of
is one of the com-
Katak.
that period. We examined numerous subterra- always the same, although the The concept is
nean constructions, possibly dwellings, and ex- vary in number from two to six, withchambers side
an aver-
cavated one of these which furnished a small age of between three and four. Unit 1K-A
collection. (Fig.
Most of the constructions in this area are sub. 21) has a central room, 2.78 by 2.2 meters,
terranean. Tello has done considerable work in oriented from the
north to south, and 1.5 meters high
clay floor to the slab cover-stones. The
the region and has published a brief note.' Here roof of this room is covered by two long parallel
he says that he counted 148 subterranean cham- slabs
plus cross slabs.
bers in the Katak region alone, and notes that room is a shaft and lintel-covered The entrance to the main
Don Agustin Icaza made the Recuay style col- doorway at
lection which now forms part of the Macedo athedoor southeast corner. At the northeast corner
Collection in the Museum fur V6lkerkunde in meters.leads To
to a connected chamber 2.5 by 1.23
the
Berlin. Tello, however, was puzzled like our- 1.15 by 3.77 meters. north is a connecting chamber
selves by the object of these chambers and by is at"the west, and Aatslightly its
smaller chamber
south end is a still
the actual association of materials. To quote: smaller connected chamber. The fourth con-
The object of these buildings has not yet been nected chamber, 1.04 by 2.28, is at the south.
satisfactorily ascertained. The natives affirm that The inner walls are made of upright slabs and
bodies and ceramics have been extracted from them. horizontally placed thin small stones (P1. 5). All
Nevertheless, I have not found in Katak nor in the spaces between the slabs are carefully filled
subterranean chambers of Gecosh, Castillopampa, or with small stones and clay. Traces of red paint
Wari-raxa, any of these archaeological elements.2 are seen. Actually a low mound is formed by
dirt
Thus far in this report several types of house roofing heaped up above this house, so that the
sites have been mentioned. (Wilkawain: above, slabs are actually 1.63 meters deepj al-
though
ground houses of one and two floors; three-floor surface. only 1 meter below the general ground
temples; subterranean one-room houses under Other sites show no trace ofmounds.
Close
large natural rocks; subterranean galleries. pler, unit (IK-B) to the above is a similar, although sim-
Ayapampa: one-room above-ground houses, central room it has of the same type. Besides the
with slightly gabled stone roofs. Copa Chica two or three connecting
two- to four-room above-ground houses with room to chambers, depending on whether the now open
heaped up dirt roofs the east,
over flat slabs [A]; roofless ered a chamber or 1.08 by 2.38 meters, is consid-
houses laid out in village pattern [B].) Virtually the entrance. A wall niche,
none of these types appears in the Katak-Re- 25 by 23 centimeters and 20 centimeters deep,
is a new feature.
cuay region; the same is true of the various the doorway leading from the Most interesting in this unit is
types found in the latter. Since our exploration main room to the
did not include every unit in the region, and west chamber. It is flanked by two upright
rectangular slabs and covered with a rectangu-
1 Tello, 1930, 270-271. lar lintel. Both side slabs and lintel are covered
' Tello, 1930, 271. with a white paint on which a red design is still
0 so
DOOR
DOOR
DOOR
C%)
DOOPR
D°OOR
/. sO
Fic. 21. Groundplan of house at Katak, Recuay.

WALL
9.30
METERS
O S0 1.0
N I
-J 0
3 Go
-I
FIG. 22. Groundplan of house at Katak, Recuay.

visible. The main lintel has a two-headed figure deep is at the northwest corner. The roof is
with single simple body, flanked on each side by formed of seven great slabs, the four on the
an animal. The side panel on the north has a north forming the floor of the room above. The
stick-like human figure. The figure of the main upper room is again entered from the east, pre-
lintel is similar to some of the carved slabs in sumably from a hallway which gave access tQ
the Huaraz museum. both floors. The main room of the upper floor
Unit 2K-E has a central room, 1.7 by 5.1 is roughly 2.5 by 1.8 meters, and only 1 meter
meters, with an entrance on the east, and six high. A niche-like room is connected on the
hook-like connected chambers, three on the north with two others on the west. Actually the
west and one each on the north, south, and east
sides. The height of the main room is 1.3
meters; a low mound is formed above the roof
which supports a surface wall, 3 by 3 meters
square. Traces ofsurface construction are found
above several of the subterrranean houses of
this type.
Unit 4K-E has a central room and four con-
nected chambers. The new feature is found in
the east chamber, 1.4 meters square, which is at
a higher level than the main room and is en-
tered by a window-like niche. 4K-B, 4K-C, and
4K-D (Fig. 23) all illustrate clearly the type of
entrance suggested by 1K-B. Through a side
door one enters a good-sized hallway (1.1 by
2.12 meters in 4K-D), slab-covered, but at a
higher level than the main room and its con-
nected chambers. The main room is entered
from the hall compartment via a window-door.
This illustrates the main features of the
Type A construction. Excavation of Unit 2K-A
and the collection will be described later. The
principal features of the type can be sum-
marized briefly:
1. Central subterranean room plus two to six
connected chambers FiG. 23. Groundplan of house at
2. Direct shaft and door, or hallway entrance
3. Slab roofing Katak, Recuay.
4. Slab walls with chinks filled with small hori-
zontally laid stones, held by clay cement second floor forms a sort of balcony room above
5. Red and white paint on some walls the first floor, since the height of the first floor
6. Clay floor is 1.6 meters at the south end and only 1.3
7. Doorways to main room and to side chambers, meters under the second floor. Above this sub-
flanked by slabs and covered by lintels terranean unit is a surface mound, so that the
8. Niches in walls cover slabs of the first floor are 1.6 meters below
9. Some mound-covered, plus traces of surface
constructions the surface. Surrounding the mound is a rec-
tangular enclosure wall, 9.3 by 19.8 meters.
Type A, Special: A special variant of Type A This unit is listed as a distinct variant since
has a second floor at one end (Fig. 22). The it is unique among those examined by us. Actu-
ground floor is a subterranean room 6.6 meters ally it shows numerous variations from the
long, 2.8 meters wide at the south end, and regular A type.
tapering to 1.8 meters wide at the north end. Type B: Simple Subterranean Gallery. Site
The entrance is on the east side, and a niche or 6K-B is a simple one-room gallery, 1.66 meters
chamber 1.6 meters wide and 90 centimeters wide and 3.29 meters long, oriented north to
south. The entrance, at the north end, is a shaft the lower floor, but a stone in the floor of the
1 by 1.15 meters, with a window-door into the upper gallery had been removed permitting
gallery. A single wall niche was seen in the main entry. The lower floor is about the same size,
gallery. This type is the simplest in the region but is partitioned lengthwise into two com-
and closely resembles the subterranean galleries partments by thin slabs. On the ground above,
found in the Wilkawain region. Site 6K-E was traces of a foundation of a surface building were
the only other gallery of this type seen. visible. Two other units of this same general
Type C: Two-Floor Subterranean Gallery. type were seen.
Closely resembling Type B are galleries which, Type D: Two-Floor Above-Ground House.
in addition to the main room, have a 9imilar At Site 6K-D an above-ground stone house had
floor below it. Site 6K-F has a covered hallway, at least two floors (P1. 4d) and in many ways
.35 .5ss l
I$,
II
I 1 REPIOV(L
IIgI
II
1.30 g,
_I
v ,,
II
I.20 ,I I
I,I
I
I
I, I
_ _ _ _ I l_ _
- _ -
1 .65 .10 .65 I

I I
FIG. 24. Groundplan and cross-section of house at Katak, Recuay, Site 4K-D.
1.44 meters long and 68 centimeters wide, en- resembled the above-ground house sites of
tered through a doorway on the east end, 44 by Wilkawain. The groundplan is a square of 7.5
58 centimeters. The hallway has two window- meters. On the east an entrance to the ground
doors, one above the other, on the north side. floor was seen, but it was not possible to enter.
Through the upper one, 65 by 79 centimeters, The second floor is set back on all sides, leaving
one enters the upper gallery, 1.65 meters wide, a terrace from 1 to 1.4 meters wide. Outside
3.01 meters long, and 1.56 meters high. Three measurements of the second story were 4.7 by
small niches and a niche-like chamber are in the 5.5 meters, and the floor contained two rooms
walls of the upper gallery. Through the lower about 1.4 meters high inside. The whole unit
door from the hallway one enters the lower gal- was about 3.5 meters high.
lery, exactly the same size as the one above, but Type E: Chullpas. At site 5K a row of twelve
only 1 meter high. Site 4K-F (Fig. 24) is a simi- chullpas, all badly destroyed, was seen. These
lar two-floor gallery, but with the hallway de- are small above-ground stone structures, with
stroyed. A window-door gives entrance to the two or more small rooms. All are made of rough
upper gallery, 1.25 meters wide and 2.2 meters stone. Certainly they were not intended as dwell-
long. No doorway of similar type was found to ings.Tello is said to have excavated two of these.
MISCELLANEOUS CONSTRUCTIONS Site 7K, Yana-yako. A site east of 3K con-
Three types of graves were seen, but our own tained a number of galleries of Type B.
excavations uncovered none. Grave Type A is Site 8K, Garbanzo-oko. A site near 7K with
round, stone-lined, and slab-covered. Two seen Type B galleries.
at Site 1K were 85 and 53 centimeters, respec-
Site 9K, Yaullish. Another nearby site with
tively, in diameter. Grave Type B is a stone- Type B galleries and graves of Type B.
lined box under a surface boulder, similar to Site 1OK, Inca-tana. A site near and similar
those seen in the Wilkawain region. Some are as to Site 9K.
much as 2 meters below the boulder, with the Site IlK. Castillo de Katak. On the flood
intervening space hollow at present, although plains below Katak hacienda house is an irregu-
not stone-walled, but this may be the result of
lar hill with remains of walls. Some claim that it
recent excavation. Grave Type C consists of a
is artificial, but little trace of construction can
rectangular box tomb, 1.84 meters by 84 centi- be seen today.
meters, stone-lined and covered, with a special
Site 12K, Puruway. A site near 2K with some
48-centimeter square opening in the top. Those house ruins.
seen were up to a meter below the surface rock
EXCAVATION AT SITE 2K-A
with the space between grave and boulder hol-
low, although unwalled. We excavated in one of the low house mounds
and were fortunate in discovering the entry
SITES IN THE KATAK-RECUAY REGION way. This consisted of a fallen-in hallway,
Site 1K, Yunapampa. A site south of the something over 1.36 meters square, which en-
Katak hacienda house, on the edge of a que- tered the main room of the house via a window-
brada. Two Type A house units and two Type A door, 60 centimeters wide, 65 thick, and 56
graves were examined. high. The house floor was at a lower level than
Site 2K, Ichik-Kayanan. East of the Katak the hallway. The main room of the house was
hacienda house, about a half hour's walk, are 2.42 by 1.28meters, oriented north to south, and
some sixteen mounds representing subterranean
1.42 meters high. Five connected chambers off
house sites, and four stone circles about 1 this main room were located as follows: one to
meter in diameter, some with double rows of
the southeast, three along the west side, and one
stones for walls. One house was excavated, and
off the north side. The floor was of packed clay,
two others were measured. Two are of Type A
and the walls were of the large slab and selected
and one is a special two-floor variant of this small stone technique so typical of all Type A
type.
house structures. Clay cement was used in the
Site 3K, Ichik-pata. One of a series of three walls, but no traces of paint were found. The
ruins on a ridge to the north of the hacienda roof was composed of three large slabs and
house of Katak. Graves of Type B and C are smaller cross stones. Actually, the roofing
numerous,as well as galleries andTypeA houses.
slabs were 1.62 meters below the top of the
Site 4K, Raku-ama. South of Katak hacienda mound. On the surface above the house was the
house, across a quebrada, is a large site said to foundation of a 2 by 3 meter enclosure wall, al-
be one source of the Macedo Collection. though this was not precisely located over the
Graves of Type A and B were seen, as well as main room of the subterranean house. An en-
numerous house sites. Of those measured, five
closure wall, 8 by 10 meters, surrounded the
were of Type A and two of Type C. Two chull-
mound. Excavation along the floor furnished
pas, house Type E, were seen.
some 336 sherds.
Site SK, Parill6n. A site along the north bank
of the quebrada and east of Site 1K presents a Materials from House Site 2K-A
dozen Type E chullpas. Plainware
Site 6K, Gekosh. A section on top of a high Thick orangeware, 78 fragments. Includes
ridge east of Ticapampa pueblo. Forty-eight or eight flare-rims of ollas, one straight rim of an
more units are represented. Of the six measured, olla, and four flare-collar rims of large vessels,
one was Type A, two were Type B, two were five flat-loop handles and two collar lugs.
Type C, and one Type D. Thin orangeware, 96 fragments.
83 Olla fragments, including pieces of fourteen Red-on-buff, 8 fragments, all from open-
flare-rims bowls. One has a fine-line design of line, circle,
13 Orange tubes, including two with closed ends, and simple face, and two others have relief de-
although the others resemble the Wilkawain clay signs of a modeled head.
trumpets White-and-red, 7 pieces, probably all from
Brownware, 6 fragments. All from a bowl one vessel with relief design, including a
with sieve base, and flat-top angular rim. modeled foot lug, a head comb, and parts of a
Crude reddish olla pieces, 41 fragments. relief arm. From the small fragments this vessel
might represent either Recuay or Early Chimu
Polished Monochrome style.
Black, white-on-red, 1 fragment, representing
Blackware, 23 pieces. the base of a vase, with a simple linear design.
8 Goblet fragments, representing plain and raised
band goblets OBSERVATIONS ON THE KATAK-RECUAY
2 Pieces of a slightly angular-bodied bowl with a REGION
plain rim
12 Pieces of globular vessels The collection from Site 2K-A is obviously
Piece of an angular-bodied double-spout bowl too limited for detailed comparisons, and over-
with a relief face design weighted with plainwares. From its general
Redware, 33 fragments. appearance it looks like still another sample of
the Wilkawain Tiahuanacoid material, al-
29 Pieces of globular vessels, including a flat-loop though lacking many of the characteristics. On
handle, and a flare-rim the other hand, toy vessels, sieves, tripod plates,
2 Goblet rims large shallow plates, blackware goblets, red-
2 Straight rims of cups
ware, and the general type paintedwares are all
Paintedwares typical. The question as to whether this mate-
rial was left by the builders of the subterranean
Red-on-gray, 14 fragments, including two house or whether it represents a later intrusion
pieces of toy ollas, an annular base fragment, cannot be answered from the evidence at hand.
and a straight rim piece. Ten pieces come from Our limited work in the area presents certain
one large shallow plate with wide bands forming details but little basis for interpretation of the
a simple locked design. house types, or for a solution of the Recuay
Red-on-orange, 9 fragments, including two problems. The absence of stone statues in the
pieces of a shallow tripod plate, four rims of region is noteworthy, since one style is found in
shallow plates, two simple bowls with flare-rims abundance at Aija and another in the Huaraz
and flat horizontal handles, and one body frag- region. On the other hand, none was seen by
ment. In all cases the designs are composed of us, nor have any been reported from the Katak-
simple lines. Recuay area.
PART 2. EXCAVATIONS AT CHAVIN DE HUANTAR
THE RUINS OF CHAVfN DE HUANTAR lie just certainly outstanding, even though affiliations
south of the pueblo of Chavin, east of the with other Peruvian sculpture may ultimately
Cordillera Blanca, on small tributaries of the be demonstrated. Recent discoveries on many
Mara-non River. Some general descriptions have parts of the coast of Peru have augmented the
already been published,1 and the discussions of importance of Chavin by demonstrating affili-
Chavin sculptural style are numerous. Some ex- ated styles in ceramics and wall decorations. In
cavations have been made by Julio C. Tello and other words, Chavin, as a period or a complex,
Santiago Antu'nez de Mayolo, but no accounts is slowly taking its place as the third Pan.
have as yet appeared. Peruvian style, along with Tiahuanaco and
Chavin as a site is important in itself. The Inca. Consequently, the results of our survey
massive platform buildings with their well- of the ruins and the excavation of 16 pits
faced walls and maze of interior galleries are therein are presented in some detail, even
unique in Peruvian archaeology. The associ- though we are quite aware that the solution of
ated stone sculpture presents a style which is the total Chavin problem is still to be found.
THE RUINS
For descriptive purposes, the building units tions on top. Actually then, as viewed from the
in the Chavin ruins are divided into nine sec- east, the Castillo achieves a total height of over
tions, lettered A to I (see sketch plan of general 13 meters above the surrounding terrace, and
layout and location of excavated pits, Fig. 25). the buildings on top would add even greater
The principal ruins cover an area of about 250 height. It is doubtful whether the building was
meters square, although the extension of the of this height on all four sides, since it is set
artificial archaeological area is much greater. on the terrace sloping down to the Mosna
In general, the orientation of all ruins is on an River at the east. In fact, only about 4 meters
east to west and north to south basis, although are exposed along the west side, although the
some deviations are found. For descriptive pur- modern road and the accumulation of dirt may
poses the orientation is considered relatively hide part of the construction.
true, and in the plan the ideal layout is sketched. A stairway (PI. 6a), first uncovered by Tello,2
cuts through the east face of the temple, 17 me-
UNIT A, THE CASTILLO ters from the southeast corner. It is not centered
One of the best-preserved and certainly the on the east side and it is possible that a match-
most frequently illustrated unit is this princi- ing stairway once existed farther to the north,
pal structure forming the southwest corner of although, due to bad destruction, no trace of
the ruins. Actually the southern half of the this is now visible.
Castillo is beautifully preserved, while the At the southwest corner are preserved de-
northern half is badly destroyed. Thus the tails which can no longer be seen in other parts.
southern end measures 72.5 meters in width, Sections of a cornice are still present in the form
and the north to south length is estimated as of projecting flat slabs, well dressed and carved
about 75 meters. Measurements at the well- on the edge and bottom surface, at about the
preserved southeast corner of the Castillo level of the setback terrace ledge. One stone
show that the inclined, well-faced wall rises and a fragment of another cornice are still in
to a height of 9.5 meters, where it is inter- position; several other flat decorated slabs in
rupted by a flat ledge 2.5 meters wide, followed the area can be considered as additional ex-
by another rise of 2.5 meters. Behind this ter- amples. Below the projecting cornice, heads
raced ledge the earth is built up for at least carved in the round, with projecting tenons at
another meter to the level of the house founda- the back, were inserted in the wall. Two of these
heads are still in position at the south end of
1 Middendorf, 1895; Raimondi, 1873; Roosevelt, 1935;
Tello, 1929,1930; Wiener, 1880. 'Tello, 1930, 277 and Fig. 11.
71
I--~~~~~~~~~~~~~~
ICHt-6Ia~~~~~~~~~~~~~~~~
I
I
I
'13 I
A I I C
I I I 3F -e
I
I CASTILLO I I To
C"-16
I UzSPE I I CHAPEL MOUND
I a
I Il
I
I
I
I I CHAPEL.
I
I
I
I
D
TERRACE
I
I I
,,I~ I
I
IOUD
H c5
100'"
CARVED I I
R4EAD o- I MOUND TO-I
S
°STONE
I SCALE
I IN METERS
m- I I0 10 25 50
E G I * Tl
'H PLA7A'
SUNKEN PLAZA ATFORM I
CH- I
0
stXps
L____
CH-9 O
SIKETCH PLAN OF CHAVIN

SHOWING PRINCIPAL BUILDINGS
and
LOCATION OF EXCAVATED PITS
FIG. 25. Groundplan of ruins at Chavfn.
the east wall, and the base of another is seen either from the presumably contemporaneous
on the south wall. Many heads with projecting houses on top, or from post-Chavin periods.
tenons are found in the local museum and While some late material was found, the exca-
private collections, all presumably taken from vations were not too important.
the ruins. The terrace which is seen on the south and
On the flat top of the Castillo, foundations east side of the Castillo is described as Unit D.
of rectangular houses made of well-dressed
stone may still be seen. Tello1 has illustrated UNIT B, THE IDOL SECTION
some of these. The north side of the unit is
badly destroyed, although traces of two wide Between the Castillo and the large mound to
descending terraces are seen. The west wall, be- the north (Unit C) is a setback section which
side which runs the road to Chavin pueblo, ap- Tellos describes as a narrow terreplein connect-
parently continues and serves to unite the ing the two. The west wall shows no distinct
Castillo with Units B and C to the north. break from that of the Castillo. On the east
The interior of the Castillo is honeycombed side are two sets of steps and two entrances to
with stone-lined slab-covered galleries and rect- the galleries. The south entrance has been de-
angular rooms, forming at least three floors and scribed, with a plan, by Roosevelt.4 The north
perhaps more, many connected by stairways entrance leads to the now famous room contain-
and inclines. Likewise, the interior galleries are ing the monolithic statue (the lanzon) still in
ventilated by a complex system of vent shafts, situ.
not only connecting galleries on the same floor, UNIT C, THE CHAPEL MOUND
but running vertically as well. The interior gal- North of the main Castillo is a badly weath-
leries have been frequently described.2 They ered mound or platform on which is located a
vary from 1 to 1.5 meters in width and from small modern chapel which gives it a local desig-
1.5 to over 3 meters in height. The walls are nation. The west wall of this platform or mound
made of selected split stones neither dressed nor seems to be the unbroken continuation of the
carved, placed in horizontal rows of large west wall of Uniis A and B. It is true that this
stones, alternating with very small ones. Niches, western edge of the ruin group is badly dis-
as well as the openings to the ventilating shafts, turbed in spots, but in the total length of 150
are found in the wall. All galleries are roofed meters no definite division units can be dis.
with selected slabs which either extend entirely tinguished. The chapel mound is about 40 me-
across the gallery width, or rest on slightly pro- ters wide (north to south) and some 84 meters
jecting stones at the top of the walls. The from east to west. Along the north edge is a ter-
rooms are about 2 meters wide and from 4 to race, about 10 meters wide, corresponding per-
5 meters long. All doorways are covered with haps to a similar terrace at the south edge of the
lintels. Castillo. The chapel mound is also honeycombed
The outer walls of the Castillo are faced with with interior galleries. One pit was excavated
well-dressed granite slabs (PI. 6b), again set in at the base of the mound at the eastern edge
horizontal rows, with wide and narrow slabs (Ch-2).
alternating. Sections like the corner of the
Tello stairway present beautiful stone masonry. UNIT D, THE CASTILLO TERRACE
Behind the facing is a rubble fill of dirt and
mixed stones. Quite obviously, the Castillo was Along the south side of the Castillo is a ter-
built up floor by floor with galleries, vents, fac- race about 9.5 meters wide. This apparently
ing, and fill carefully planned. continues along the east side, where it broadens
Three test pits (Ch-4-5-6) were excavated on into a wide terrace of 43 meters with a facing
the top of the Castillo, not with any hope of wall along its eastern edge still standing 2.5 me-
uncovering architectural features or achieving ters high in some places. At a point directly
great depth, but to discover dwelling site refuse, east of the Tello stairway in the Castillo, an
angle in the terrace facing wall suggests the
Tello, 1930, Fig. 10.
I
Raimondi, 1873, 213; Tello, 1929, 50-53; 1930, 277-
2 Tello, 1930, 277.
278; Roosevelt, 1935, 3742. ' Roosevelt, 1935, Fig. 23.
possibility of a stairway there also. One pit was The details uncovered will be described with the
excavated on this terrace (Ch-7) without re- excavation notes.
sults.
UNIT G, THE NORTH PLAZA PLATFORM
UNIT E, THE SUNKEN PLAZA
A platform similar to Unit F flanks the north
Directly east of the Castillo and the terrace side of the plaza. It is well preserved along the
is a square area, 48 meters on a side, which ap- south side and eastern end, but the northwest
pears to be a sunken court. At the time of ex- delineations are vague. Tellol states that this
amination the whole section was planted with platform runs from northeast to southwest,
oats which obscured some details. This plaza, although we frankly failed to notice this dis-
if such it be, is fairly well centered in reference crepancy. The platform is some 28 meters wide
to the Castillo, and about 80 meters from it. and about 90 meters long with a projection to
Just outside the southwest corner of the plaza the south at the western end. A terrace runs
is a large flat boulder with seven grind holes along most of the south side. One pit (Ch-1)
and a rectangular niche. The stone is in rough was made on this platform, revealing vents
line with the Tello stairway and the break in which indicate interior galleries.
the terrace wall of Unit D. Some local people
claim that the Raimondi monolith once stood UNIT H
in this niche in the flat boulder, but no positive
evidence of this is available. The plaza is out- Just northeast of the chapel mound (Unit C)
lined on three sides by terraces, one along the is another mound which is not oriented with the
river front, and two at the bases of the side rest of the ruins. A saddle gap separates it from
platforms. Both to north and south of the plaza Unit C on the west, and a rough facing wall
are platforms described here as separate units. along the east is 36.5 meters long. The east-to-
west measurement is roughly 51 meters. Non-
UNIT F, THE SOUTH PLAZA PLATFORM Chavin ceramics and building features were
found in the excavations along the west edge
Flanking the south side of the plaza is a plat- (Ch-3).
form, oriented east to west, and about 30 me- UNIT I, THE INCA AMPHITHEATER
ters wide. A rectangular projection is seen at
the west end. The length can only be esti- In the northeast corner of the total ruins are
mated, since the Mosna River has cut deeply a few vague, somewhat curved walls which have
into the platform. By comparison with the the impressive name of Inca Amphitheater.
north platform the length should be about 90 Some terrace walls of massive stone can be seen
meters. The river cut shows the rubble con- (PI. 6d). Our investigations neglected this sec-
struction of the inside, as well as some poorly tion, so no accurate description can be given.
constructed interior galleries (PI. 7c). The base
of the platform is close to the water level, and THE ADJACENT REGIONS
great stones are utilized in the foundation. The Aside from the main construction units the
top of the platform is about 10 meters above region around Chavin has many features of
the river at the cut. Its sides show no trace archaeological interest. The fields to the north
of the dressed-stone facing of the Castillo. of the ruins contain many sherds, some of
A terrace bounds the north side, next to the Chavin style (Ch-15, 16). Tunnel mouths were
plaza. seen along the river cut to the south of the
Near the northeast corner of this platform is ruins. Others are reported under Chavin pueblo
a dressed rectangular stone with a longitudinal and across the Mosna River. Obviously, much
groove cut from one side like the canal stones of more work is needed in this whole area. For one
Tiahuanaco. Nearby is a set of monolithic thing, in spite of the amount of casual investiga-
steps, although not in position. One pit was tion, no Chavin cemetery site has yet been dis-
excavated on top of this platform (Ch-8). covered.
Along the river side of the plaza a number of
pits were excavated (Ch-9, 10, 11, 12, 13, 14). 'Tello, 1930, 277.
EXCAVATIONS
In the following pages a brief description of galleries. A small canal ran from the mouth of
the excavation of the sixteen pits is given. For one vent for about 4 meters. Six sherds were
the most part, the analysis of artifacts en- encountered, all plain, but one with the thick
countered is reserved for special discussion, rim of Chavln style.
since such a description for each pit would in- Pit Ch-2. At the eastern edge of Unit C, in
volve considerable repetition without great sig- the saddle which separates Unit C from Unit H,
nificance. Search within the galleries now open a pit 3 by 1.25 meters was excavated. This was
proved fruitless, either because sherd material near a long trench made by Dr. Antunez de
was not deposited in them, or because the years Mayolo. A rough wall cut across the north end
of casual investigation have made their impres- of the trench from the surface to 1.1 meters
sion. Surface sherds were not too abundant. depth. Another rough wall from 1.25 to 2 me-
Doctor Soriano encouraged the school children ters deep was found at the south end of the
to bring him "Chavin" style fragments of pot- trench. At 2 meters depth we encountered the
tery, but even with a monetary incentive rela- cover-stones of a gallery which had long since
tively few were discovered. A cut like that made been almost entirely caved in, but a clear sec-
through the south plaza platform, Unit F, by tion was found about 2 meters long. The gallery
the river has revealed virtually no sherd ma- was small, only 1.3 meters deep and of about the
terial, and the occasional pieces found in the same width, with walls of small rough stone and
river wash might well have been carried down good-sized covering slabs. Although the gallery
from the fields above. obviously continued to north and south, it was
Judging by the river banks, much of the re- impractical and dangerous to attempt to follow
gion around the ruins is artificially built up as, it.
obviously, are the constructions themselves. Sherd material was abundant. A total of
Such artificial building and leveling of terraces 1950 pottery fragments includes nine which
limit the reward of the excavator, since sherds might be considered Chavin types. A bone and
would be mixed in only by chance, or as part a stone artifact were also included. In excavat-
of the fill dirt itself. In fact the potentiality of ing, the materials were separated into six lots
pre-Chavin stratification in the ruins them- on the basis of arbitrary half-meter levels, but
selves is limited by the difficulty of cutting pits the only suggested stratification was between
below the artificial level. Other parts of the Incaic and Tiahuanacoid types.
Chavin region might well present other cir- Pit Ch-3. The third pit (Fig. 26) was located
cumstances, but our digging was limited to the just northeast of the second along the western
ruins themselves. Consequently, about the only edge of the mound of Unit H. The mound rises
hope we had was for good post-Chavin styles. steeply at this point making surface depth
The quantity of post-Chavin material was measurements somewhat meaningless. Late
enormous, but for the most part disappoint- stone-facing walls were discovered, and the
ing in distinctiveness. pit was extended to uncover them. The facing
The location of all pits is indicated on the walls run roughly north to south along the west
sketch plan and consequently will not be de- edge of the mound. Starting at the north is a
scribed in detail again. straight wall of rough stone somewhat poorly
Pit Ch-1. Near the eastern end of Unit G, laid, 5.5 meters long and about 2 meters high.
two stone-lined vents were discovered. One was At the level of its base, and about 30 centi-
40 by 60 centimeters at the surface and was meters in front of it, was a 2-meter section of
surrounded by a stone slab wall. The vent de- a stone-lined and slab-covered canal. It was
scended in steps at about a 45-degree angle and 40 centimeters wide and shallow. Between the
then turned sharply downward. It was largely canal and the wall a pair of complete thick red-
filled with stone, so that entry became impossi- ware goblets (Ch-3A) was found without grave
ble. The other vent was narrower, but entered association (Fig. 28a). Both are good Inca
at about the same angle. Obviously such shafts types.
are too small to have served as entrances to At the south end of the facing wall is another
wall, possibly either another facing wall or a ered with slabs. The cover-slabs were at the
house wall. The north end presents a sharp level of the base of the wall, and the walled
curve, then a north to south stretch 2.42 me- section extended to a depth of 30 centimeters.
ters long, a sharp angle to the west, of 64 centi- We excavated to a depth of 1 meter, finding
TOP OP 1OUIwo
CDCD METERS
0 9.0 x,0
--- N -
HOUSE OF CLAY
WITH PLASTER
ROUGH FACING WALL

tAJ9uo.oo
cooooc oo
CANIAL SECTION
.aO ,c&o
/ g ~~~HOUSE
WALL
=I.........
pOZO 0
I
FIe. 26. Groundplan and cross-section of Pit 3, Chavin.

meters, an angle to the south of 1 meter, an nothing but sherds. Near the pit, in the north
oblique turn to the southeast of 2.4 meters, an- to south section of the wall, was a carefully
other turn to the south of 1.5 meters to a final made window, lined on all four sides with slabs,
angle to the west. and measuring 41 centimeters square and 33
Several features mark this second main sec- centimeters thick. The window leads through
tion. West of the wall, at its north end, was a the wall into the mound, and its function in
1-meter circular pit, lined with stone and cov- this position is dubious. Excavation on the back
side uncovered a burial (Ch-3B) associated with this was a hard yellow clay, undoubtedly the
a crude black spoon and a small olla. Another original covering of the temple, and without
spoon and olla found nearby may well belong sherds. Of the 233 sherds encountered not one
to the same burial. Since this burial appears to seemed Chavin in style.
be intrusive from the present surface of the At the north side of the pit, at 1 meter depth,
mound, it is quite possibly post wall in time. the entrance to one of the Castillo vents was
In the corner formed by the sharp right angle found. This starts with a 45-centimeter square
of the wall to the south of the window was a shaft, descends for 1 meter, then slopes slightly
stone-lined box, 30 by 60 centimeters, covered to the north for 2 meters, drops again for 1.5
by a thin slab. It contained the skeleton of a meters, followed by a horizontal stretch for 1
young female (?) with an olla (Fig. 28c) and meter and a final vertical drop which we could
a small jar (Ch-3C). Continuing along this ang- not examine. Undoubtedly this vent leads to
ular wall another window was found, 55 centi- some of the interior galleries, but the peculiar
meters wide and 37 high. West of the window, feature is the fact that the mouth was not only
on either side, were two burials without arti- 1 meter below the surface but was also covered
facts. Ch-3E contained the deformed skull of a with a stone slab. Intrusive burials had broken
child and pieces of an infant skull. Ch_3F con- through the roof of the first sloping part of the
tained one flexed burial in seated position and shaft, and bones and about eight broken ollas
miscellaneous bones. Both Ch-3E and Ch-3F were found.
were burials in very crude stone boxes. Pit Ch-5. Another pit on the same terrace as
Behind the house or facing wall and higher Ch-4 was made just north of the Tello stairway.
on the mound was a house construction of a Sherds of Late styles were found between the
distinct type with burnt clay plaster walls fallen stones, as well as a shallow burial accom-
(P1. 7a, b). The house was probably about 2.86 panied by a crude vase and an -annular-base
by 2 meters, although only the north wall is open-bowl (Fig. 31q). Further excavation
well preserved. This is 75 centimeters thick and seemed impractical at this point.
contains a window-door 1.1 meters wide and Pit Ch-6. A test pit on the highest part of the
50 centimeters above the clay house floor. The Castillo in the southwest section did not even
original height of the wall is unknown, but it produce sherds.
still extends 1.13 meters to the ground surface. Pit Ch-7. Just east of the southeast corner of
Along the eastern wall was a clay bench 50 by the Castillo a pit uncovered a number of stones
50 centimeters. Other walls were completely de- in rough rows, but even down to 1.5 meters
stroyed. The composition of the walls is stone, depth contained no sherds.
and clay cement, all covered, inside and out, Pit Ch-8. A pit 2.5 meters square was exca-
with a 1-centimeter thick clay plaster, tempered vated in the northwest sector of the south Plaza
with straw, and fired to a very hard consistency. platform (Unit F). Here the yellow clay starts
No ceramics could be identified with the house, at the surface, and contains but few sherds. A
and an extended burial at the western edge stone slab was found at 90 centimeters depth
(Ch-3D) contained no artifacts and is probably and some 20 centimeters under this a stone-
later than the house. lined and slab-covered box grave, 37 centi-
The first sherd material was divided into meters square and 65 deep. This box contained
three arbitrary levels, but the later work showed some badly disintegrated bones and two vessels,
that sherd stratigraphy was meaningless on this one a plain orange olla with two flat loop
steep slope. Six complete bowls and two spoons handles (Fig. 28b), and the other a small bowl
were found, as well as 996 sherds, of which only decorated with a red-on-orange band. Just out-
five seem to be Chavin style. Other artifacts side of the box two plates were found, one
include two of stone, five of bone, a shell and a orange and one red-on-orange. The pit was con-
clay whorl. tinued to a depth of 1.5 meters. Outside of the
Pit Ch-4. On the southeast corner of the main intrusive burial only eight sherds were found.
Castillo (Unit A), we made a pit 3 by 2.25 me- One of these, near the bottom of the pit, is
ters. This is not the highest part of the Castillo black polished with a groove design of the
and some wash has resulted. Forty centimeters Chavin type. It suggests the association of
of top black soil contained Late sherds. Below Chavin sherds with the temple construction.
Pit Ch-9. Reports of the finding of incised Ch- lA. A stone box grave, 50 by 44 centi-
sherds along the river bank led us to cut a pit meters and 64 centimeters deep, was based on
on the flood plain just about east of the center one stone and utilized the end of a large fallen
line of the plaza (Unit E). A section, 2 by 1 me- slab for a cover. It contained the skeleton of a
ters, was excavated to a depth of 1.3 meters and youth with deformed skull, in flexed position on
then cut back into the bank for a maximum size the back with the knees up. No artifacts found.
of 2 by 4 meters. Most of the soil was disturbed Ch-IIB. A grave of oblong shape, 60 by 40
by the river into a mixture of black soil and centimeters, and 30 centimeters below the sur-
waterworn stones. The 138 sherds found seemed face. The same fallen slab which partly covered
consistently of Chavin style. Fragments of shell Tomb 1 A formed a side wall of this grave. The
and coal were also found. other three sides were stone-lined, but without
Pit Ch-10. Since the material from Pit Ch-9 a cover-stone. It contained a flexed burial and
was disturbed by the river floods, the next pit mixed fragments of pottery.
was made to the west, on the top of the bank. Ch-11C. A stone-lined and covered box grave,
This was started as a cut 2.5 by 4 meters and 52 by 80 centimeters and 44 centimeters deep,
later enlarged to 3 by 6 meters. Chavin type contained a flexed burial, but no artifacts.
sherds were found throughout. The top meter Ch-llD. A rough stone tomb, 58 by 65 centi-
of the pit contained a fill of many small rough meters and 65 centimeters deep, contained
stones and an occasional large one. The next bones in bad condition and some sherds.
1.1 meters was composed mainly of yellow clay Ch-IIE. A skeleton was found at 40 centi-
with some sherds and stones. At 2.1 meters meters depth in a poorly delineated grave.
depth a layer of charcoal crossed the whole pit. Ch-1lF. A second grave under 11C had three
Mixed in it were sherds of good Chavin style. flexed skeletons.
Along the north side of the pit a rough stone Ch-IIG. Another burial was 50 centimeters
wall extended from 1.7 meters depth to this under lID.
charcoal line. Apparently this charcoal layer Ch-IIH. A burial was just north of 11A.
represents an old surface since it reappears at Ch-III. A burial was near liE.
about the same level in Pit Ch-12. No complete vessels were found with these
At about the center of the pit and just above graves, but sherds were plentiful in the section.
the charcoal level was an angular canal, lined Of a total of 486 sherds, eleven are of the
with a double row of stones and covered with Chavin style. The rest represent the white-on-
small slabs. The inside measurements are 17 red style and others of post-Chavin period. Two
centimeters wide and deep, and the cover slabs large cover-slabs and many other stones in
are 1.86 meters below the surface. Actually this mixed position indicate considerable disturb-
canal section forms an angle 82 centimeters on ance. The two slabs possibly were the roofing
each side, and the ends are carefully closed with stones of another cell like those in the series of
small slabs. Similar canals are seen in the bank Sites Ch-12 to Ch-14. The position of the graves
along the river cut, but this one with the ends in relation to these slabs would leave no doubt
walled seems meaningless. Within the canal as to their intrusive and late position.
were Chavin sherds and a piece of stone. Pits Ch-12, Ch-13, and Ch-14. The next
The pit was excavated to a depth of 2.5 me- three pits (Fig. 27) are grouped together here
ters with the yellow clay continuing, but with- because all uncovered cell-like houses formed a
out sherds. The river cut indicates over 4 me- row (P1. 6c). Actually six cells were seen and
ters of artificial construction. The material was the row may once have been longer, although
divided into four arbitrary levels in excavating, it is difficult to prove this since the river has
but no significant stratification appears. A total cut its bank close to Cell 6. The two cover-
of 305 sherds of Chavin style was found, as well slabs described in Pit Ch-11 are suggestive.
as a piece of a stone bowl. It was mentioned previously that the re-
Pit Ch-11. Another pit (Fig. 27) along the mains of a terrace along the river bluff, east of
edge of the river bank, northeast of Ch-10 and the sunken plaza, can be seen. Almost exactly
east of the plaza, failed to repeat the Ch-10 east of the center of the plaza at the inner edge
set-up because of disturbance and intrusive of this terrace is an upright stone. Directly east
burials.The intrusive burials can be listed briefly: of this stone, although not actually meeting it,
11-!
H1-1,4. i-
HOUSE Z
HOUSE 3 o I 2- 3
ol
..
MHETR$
* h.i
CH-13 HOUSE Lt
HouSE 5
CH-II
RIVERMOSNA '/ I , /
FiG. 27. Groundplan of excavations of Pit 9 and Pits 11-14, Chavin.

is a wall which extends some 20.75 meters to Pit Ch-15. To the north of the principal
the river bluff. Six small cells open off the ruins of Chavin are flat fields, some planted,
south side of this wall. All are of about the same and some with groves of eucalyptus trees.
size and construction, so that a detailed de. About 100 meters north of the mound of
scription of one (Ch-12, Cell 6) will serve for Unit H, in a field belonging to Sr. Ramirez,
the rest. a pit 4 by 1.5 meters was excavated to a depth
Ch-12, Cell 6, is rectangular, measuring 1.4 of over 2 meters. The field is level and has had
meters wide and 1.82 meters deep. Rough stone little washing. From the surface to 1.5 meters
walls outline three sides and the fourth, on the depth Chavin style sherds were found in abun-
north, is open. The walls, about 35 centimeters dance. Below 1.5 meters some sherds were
thick, are not as well made as those in the found, but as the greater depths were reached
Castillo galleries, but are about the same work- they ran out completely. Tests in various cor-
manship as the galleries in the south plaza plat- ners below 2 meters showed no continuation or
form. Two large slabs serve as covers, one the sherd material. The cross-section shows no
placed flush with the front opening. A clay obvious layers. Animal bones, charcoal, and
floor is 1.42 meters below the roof, with two sections of small and large stones are seen. In
stones set in the floor at the door opening. The general, the material seems to be house refuse.
side walls extend only a few centimeters below Some 1465 Chavin style sherds were found,
the floor. Below the floor, at about 2.4 meters separated into four levels. Eleven Late sherds
depth, the charcoal layer mentioned in Pit were found in the top layer only.
Ch-10 was again encountered. The charcoal Pit Ch-16. In the field immediately to the
layer represents, then, a pre-cell level, although north of the chapel mound (Unit C) a pit 4 by
in both cases Chavin sherds are associated. In 1.5 meters was excavated on the property of
this particular cell only Chavin style sherds Sr. Poza. A rough stone wall crossed the pit
were found. from east to west, 50 centimeters below the
The other five houses or cells are essentially surface, meeting a north-to-south wall. It is
the same, each separated by about 1.45 meters, possible that these walls are house foundations.
but connected with the north wall. There is no North of these walls, at 1 meter depth, was a
indication that these cells were originally part stone box grave 30 by 60 centimeters. Another
of a subterranean gallery, nor that any other stone box grave, 88 centimeters below the sur-
construction unit was definitely associated with face, was 60 by 60 centimeters and 80 centime-
them. ters deep. Both of these graves had disintegrated
The cells were numbered from west to east. bones, but no artifacts. An open burial in front
Of the six, three were excavated by us, Nos. 1 of the second box contained a complete bowl. In
(Ch-14), 4 (Ch-13), and 6 (Ch-12). Doctor the southeast corner of the pit at 1.35 meters
Soriano Infante cleared two more, Nos. 2 and 3, depth, was a stone-lined and slab-covered oval
and Cell 5 was left untouched. area, 1.52 meters long and 1 meter deep, which
Four cells (Nos. 1-4) contained intrusive extended for 95 centimeters in depth. It con-
burials and broken pottery vessels of Late tained a few sherds. Most of the top meter
types. No. 4 contained five burials associated contained sherds of Late types. From 1 meter
with broken and restorable vessels of the white- to slightly below 2 meters Chavin type sherds
on-red style (Fig. 31r). Three complete vessels were found, although still somewhat mixed with
and 256 sherds of similar Late vessels were later types due to intrusions. In total, a bowl,
found in these cells plus 692 sherds of Chavin a spoon, and 357 sherds of Late type were as-
style and three stone artifacts. It seems clear sembled, plus eight stone, nine bone, and two
that the cells were associated with Chavin miscellaneous artifacts. To this can be added
ceramics. some 334 sherds of Chavin style.
THE CHAVIN MATERIALS
The survey of the Chavin ruins and the ex- analysis is thus limited to Chavin style itself,
cavations show a Chavin building, sculptural plus several post-Chavin styles. Due to the in-
and ceramic style as the earliest material. The creasing importance of Chavin style in the total
history of Andean archaeology, the materials are The predominance of monochrome black,
here treated separately in terms of ceramics, brown, and redwares is obvious. Most of these
architecture, and stone carving, and the dis- are well polished, and some achieve a high
cussion of the post-Chavin styles is reserved for polish. Sand and mica temper seem to be the
a later section. rule, and the fragments show good firing, al-
CHAVfN CERAMICS though not to the almost metallic hardness of
Cupisnique. Orangeware appears in greatest
Chavin ceramics are represented by two com- quantity in the top half meter of Pit Ch-15
plete vessels (purchased by Dr. Soriano) and (184 pieces) and is only sparsely represented in
2990 sherds from the surface and the excava- the deeper levels. The rim shapes and general
tions. These pieces can be distinguished from finish of the orangeware do, however, suggest
the other sherd material, and the validity of the Chavin affiliation.
style is established by its isolation in Pits Ch-9, The nine positive painted two-color pieces all
10, 12, and 15. That the ceramics pertain to the have black and red areas separated by incised
ruins of Chavin is indicated in a number of lines. Most of the fragments are small, but none
ways. First, no stratigraphic earlier material the less the band pattern which typifies the
was found. Second, the pieces were found in the two-color pieces of Cupisnique is not repre-
temple top clay fill (Ch-8), in the artificial ter- sented. The only negative painted body frag-
race along the river (Ch-10, 12, 9), in an old ment in the collection has a linear design in red
charcoal layer apparently antedating some of and orange negative. It was found in the deep-
the stone construction work (Ch-10, 12), in the est level of Pit Ch-10, together with typical
stone-lined and covered cells along the river black and brownware. The two tan pieces are
bank (Ch-12, 13, 14), and in the fields to the quite distinct from the general brownware and
north of the ruins (Ch-15, 16). Third, the style both have a high polish. Four remaining pieces
is obviously related to Coast Chavin ceramics have a slightly whitish finish. Two pieces, not
which in turn have designs comparable to included in the positive painted group, are
Chavin stone carving. Fourth, styles repre- blackware with incised grooves filled with red
sented by the other excavated sherd material paint.
can all be shown to be in definitely late positions RIMS
in reference to the stone construction work of Of the total number of sherds 517 or 17 per
the ruins. In three pits (Ch-10, 12, 15) the cent are from rims. The major rim types are
Chavin style sherds were segregated into three shown in Fig. 28, and their frequency and per-
to five excavation levels, but the analysis shows centage are given in the following table. Char-
no significant differences in levels, nor does the acteristic of Chav:'n rims is the thickness at the
collection as a whole indicate startling variation edge, frequently augmented by an inward or
in style. Thus, for descriptive purposes, the col- outward bulge, from which they taper rapidly
lection must be treated as a whole. to a relatively thin body. Most of the rims are
COLORS from open-bowls, a fact which will be elaborated
in the discussion of vessel shapes. Black, brown,
The following table presents a resum6 of the red, and orangewares are all proportionately
Chavin wares: represented by rims. Some of the rim pieces
Ware Pieces Cent show incised design (72 pieces), but the ma-.
jority are from plain vessels.
Monochrome blackware . . . . 1374 45.95
Monochrome brownware . . . 901 30.13 SHAPES
Monochrome redware . . . . . 427 14.28 Shapes of vessels are indicated by two com-
Monochrome orangeware . . . 272 9.10
Black and red painted . . . . 9 .30 plete bowls purchased by Dr. Soriano, the rim
Miscellaneous . . .7 .24 fragments, base fragments, and some of the
Two-color negative 1 body fragments. Since most of the fragments
Tan 2 are relatively small, it is not possible to give
Whitish 4 a detailed analysis of these shapes, but some
Totals . . . . . 2990 100.00
. . . . . mention can be made of those suggested.
AI A2 A3 A4 A5 BI B2 B3 B4
B5 B6 CI C3 C4 D-l
D4
-~~~~ElI E2 E3
D2 D3
A
FIG. 28. Chavin style rim types and post-Chavin vessels from Chavin.
FREQUENCY AND PERCENTAGE OF measuring 2.5 centimeters high, another 2 centi-
CHAViN RIM TYPES meters.
Fre- Per One complete constricted-collar jar from the
Type quency Cent surface collections had a globular body 10.5
A Straight rims . (186) (35.98) centimeters in diameter and 9 centimeters high,
1 Thick ..... 18 3.48 with a flat base 6 centimeters in diameter. The
2 Thin. 70 13.54 collar is straight and measures 4.5 centimeters
3 Wedge . 38 7.35 high and 5.5 in rim diameter. The rim has a
4 In-curve 36 6.96 beveled edge of Type A5. Other fragments sug-
5 Bevel edge . . . 8 1.54 gest seven more such vessels. A tall collar, with
Miscellaneous 16 3.09 slight out-flare rim, is represented by four
B Bulge-in rims (190) (36.75) pieces, and four more show perfectly straight
1..... 90 17.41
2 . .. .. .. . 6 1.16 collars and plain rims. Twenty-five fragments
3 . .. .. .. . 9 1.74 suggest constricted-collar jars with flare collars.
4..... 20 3.87 Small globular bowls with angular-flare rims
5..... 29 5.61 are represented by nine fragments, and two
6. 29 5.61 more pieces indicate straight-sided bowls with
Miscellaneous 7 1.35 a rim at right angles.
C Bulge-out rims (40) (7.74) Sherds definitely belonging to stirrup-spout
1..... 8 1.54 vessels were not found, although 28 pieces
2. 7 1.35 seemed to be from spouts of some kind, and
4. .. ..
3. .. . 7 1.35 five more fragments might possibly represent
34.....
... .. .. .. 6 1.16
the arcs of stirrup-spouts. None of the frag-
6 1.16
Miscellaneous 6 1.16 ments showed the heavy spouts so characteris-
D Flare-rims .. .. . (63) (12.19) tic of Coast Chavin, or Cupisnique, stirrup-
1 Angular flare-out 8 1.54 spouts.
2 Curve flare-out . 21 4.06 Doctor Soriano purchased a plain blackware
3 Angular flare-in . 18 3.48 vessel of goblet shape with annular base and
4 Curve flare-in 7 1.35 diverging sides with plain edge rim.
Miscellaneous 9 1.74 Bowls with in-curving rims and convex sides,
E Spout or collar-rims (36) (6.96) sometimes called yarn bowls, were represented
1..... 16 3.09 by 44 fragments, and a variation on these with
2. S .97 angular in-flaring rims adds eighteen more frag-
3 . .. .. .. . 7 1.35 ments.
Miscellaneous 8 1.54
Most body fragments are either from straight-
F Flare-lip rim (2) (.38) sided bowls or globular vessels, but three indi-
Totals . 517 100.00 cated angular bodies, and one a square body
with projecting edge. Variation on symmetrical
Open vessels form the bulk of the collection, shape was suggested by two rim fragments with
at least 423 pieces belonging to this type. Most definite flaring lips. A single piece of a redware
of the rims of categories A, B, and C are from modeled foot suggested a puma vessel like those
open flat-bottom bowls with vertical, diverging, at Wilkawain. This was found in Ch-12 with
flare or convex sides and considerable variation other Chavin sherds, but a chance for intrusion
at the rim edges. Only one fragment had an is possible. Otherwise modeled vessels were not
annular base, all others were flat. Base frag- represented. Two fragments of flat handles are
ments represented three vessels with perpendic- probably intrusive, since handles are definitely
ular sides; twenty had straight diverging sides not a characteristic.
meeting the base at an angle; eleven had convex As in the variation of wares and rims, shapes
sides meeting the base in a curve; six had flare also seem limited to a few basic types with little
sides; and one had a convex side with a pro- divergence. This material is not from graves,
jected base. Four surface pieces from one vessel which might make some difference, but in any
indicated a bowl 11.5 centimeters high with case the uniformity throughout is noteworthy.
flaring sides. Others are shorter, one section In r6sum6:
Open vessels with rim variation and Nineteen pieces have fine incised linear de-
flat bases . . . . . . . . . . 423 signs. Of these seventeen have designs com-
Straight constricted-collar jars . . 16 posed of single lines, parallel lines, or extremely
Flare constricted-collar jars . . . . 25 simple combinations. One piece has an angular
Goblet shape . . . . . . . . . . 1 (complete) hook and bar, and one is decorated with two
Bowl with flare-rim . . . . . . . 11 fine intersecting lines.
Spout and possible stirrup-spout ves-
sels .33 Four pieces have fine incised lines which form
In-curve rim yarn bowls . . . . . 44 complex designs. On one large piece a rectangu-
In-flare rim yarn bowls . . . . . 18 lar area contains a double-line figure-eight de-
Angular body vessels . . . . . . 3 sign in the center with two triangles on either
Flare lips on rims . . . . . . . . 2 side of it. Within the triangles are small incised
Modeled foot . . . . . . . . . . 1 circles. A second piece has a zigzag line running
Handles (?) . . . . . . . . . . 2 across it, below which are a series of connected
DESIGN
diamonds, each containing a rectangle and dot.
A third piece from a rim section has an incised
Only 213 fragments or 7.12 per cent of the design of an inverted stylized Chavin feline
total collection showed any kind of design or head (Fig. 291). The last piece has a geometric
decoration. Again it is noted that the variation design with the lines filled with red paint.
in technique of decoration and the range of de- Ninety-six pieces are decorated with grooves,
sign are decidedly limited. Starting with design in contrast to the fine line incision. Eighty-six
techniques, a brief r6sum6 can be made, based of these have straight or curvilinear grooves
on the frequency of occurrence. Since two tech- made with a dull pointed instrument. Of the
niques may appear on the same fragment, the remaining ten pieces, one has a zigzag groove
totals are obviously greater than the actual between two parallel lines (Fig. 30h); one has a
number of decorated pieces. double diamond design; one has two interlock-
DECORATION TECHNIQUES IN ing angular lines; one has a horizontal line be-
CHAViN CERAMICS low which are vertical grooves (Fig. 29q); three
have exceptionally wide grooves; and one has a
Technique Fre-
quency sharp cut groove.

1 Fine line incision. . . . . . . . . . . 48 Five pieces are decorated with complex
(Linear, 30; circle, 15; dash, 3) groove designs, consisting of curvilinear fig-
2 Groove or wide line incision . . . . . 144
. ures, although none of these can be specifically
(Dull wide line, 112; circle, 26; sharp identified (Fig. 30i, j). Eighteen pieces are
wide line, 3; extra wide groove, 3) decorated with combinations of grooves and
3 Positive painting . . . . . . . . .
. 6 other designs as follows:
(Black and red areas separated by in-
cised line) 3 Grooves and fine point punch design
4 Negative painting . . . . . . . . .
. 1 1 Groove and dull gouge marks (Fig. 29m)
(Orange and red) 2 Grooves and semicircular punch units
5 Incised line with red paint fill. 2 2 Circles between parallel grooves (Fig. 29e, j)
6 Raised band . . . . . . . . . . .
. 1 2 Combination of parallel grooves and small cir-
7 Pierced holes . . . . . . . . . . .
. 1 cles (Fig. 29i, k)
8 Champlev6 . . . . . . . . . . . .
. 3 5 Combination of curvilinear groove lines and
9 Punch and gouge . . . . . . . . .
. 21 small circles (Fig. 30e)
(Fine point punch, 8; dull gouge, 3; 1 Grooved circle and rectangular punch marks
ratchet, 2; stamp punch, 8) (Fig. 29b)
10 Relief . . . . . . . . . . . . . . . 7 1 Grooves and gouge marks, possibly representing
(Simple relief, 6; modeled face, 1) a simple human figure
11 Thumb-nail impression .... . . . 1 1 Grooves plus a triangular area
12 Rough scratch surface . . . . .
. . . 4
13 Raised tabs with incised cuts . . . . . 1 Thirty pieces are decorated with circular de-
Totals of Technique Occurrences 240 signs. Two have incised circles arranged in rows
(Fig. 29c). Four have incised circles, each with
. .
The decorated pieces themselves can be de- a dot in the center (Fig. 29a). Eight have
scribed in terms of their designs: grooved circles, each with a dot in the center
~~?' q'~~~~i".
I..)%
G .'l:
-.:,~~~~~~~K
K
0
/
::"'!' 'o
Li6 .
R "I'llS'
FIG. 29. Chavin style decorated sherds from Chavfn.
'S
4 .-.
.L.77$f74{t
.ri rf.
(74: i.>k' ;vfri P. 0
(.4" K ;'
VII B
A
N
(% Ut.c< crw--- -
>.gC
C
Ct -
.3 fl.
C. 'C*'
¼''
CI - F - I-i
cc
E 6
r
-A-
..ZtKCCk. .*fj.t I;A
-6-v.-.'
),J--.
I I' K
c23 N
f$.9 (1+ 0
R
Fia. 30. Chavin style decorated sherds from Chavin.
(including one in which the grooves are filled Raised ridge. 4
with red paint [Fig. 30q]). Two pieces have a Diamond . 2
single line of small incised circles. Four have a Rough scratch lines 4
simple groove circle. Ten have designs com- Circles . 41
posed of two concentric groove circles. Painted areas . 6
Punch dot .. 8
Eighteen pieces are decorated with punch, Punch square . 1
gouge, or stamp marks. Of these, three have Open circle . 4
fine point punch marks; one has a square punch Cross . . . . . . . . 2
design; one has simple gouge marks; one has Angular U. 1
fine punch dots between incised parallel lines; Zigzag. 2
two have a rectangular ratchet impression, in Gouge oval . 3
one case combined with fine point punch marks; Figure eight. 1
one has a design made with an angular U-shaped Triangle. 2
stamp; three have designs made with an S. Dash or broken line 3
shaped stamp (Fig. 29d); one has punched Horizontal S .. . . . 3
holes which perforate the piece (Fig. 30o); one Rectangle. 2
Short arc . 1
is decorated with a thumb-nail impression (Fig. Loop. 1
29h); three have rows of short dashes (Fig. 30d); Face . 1
and the final piece has a row of long dashes. Chevron. 1
Nine pieces have relief design. All but one of Angular hook . 1
these designs are comparatively simple: two Band. 1
have simple ridges (Fig. 30a); one has a raised
band (Fig. 30n); two have relief ridges forming CHAVIN ARCHITECTURE
curvilinear designs (Fig. 29r); one has relief The ruins have already been generallyde-
crescent shapes; one has relief rows incised with scribed. The following r6sum6 is intended to
chevrons, separated by incised circles; and one stress certain architectural features which char-
has rows of raised tabs, each of which has three acterize the Chavin style.
vertical cuts on top (Fig. 30b). The one complex
piece has a modeled human face with relief nose SYMMETRY IN PLAN
and eyes and an incised mouth with double The general appearance of the Chavin ruins
crossing fangs (Fig. 30p). It is a well-finished as a whole is one of attempted symmetry. That
fragment and one which closely approaches the this is not totally achieved is obvious from the
Chavin stone carving style of design. sketch plan, particularly in reference to the two
Four pieces are decorated with rough scratch outlying northern mounds, Units C and H.
(Fig. 30c). Considering the remaining units, however, it is
Seven pieces are painted, one possibly in seen that the square plaza (Unit E) is flanked
negative technique (Fig. 301). The other six by two roughly equivalent platforms, Units F
have positive painted red and black areas sepa- and G, and backed first by the terrace (D) and
rated by incised lines or grooves (Fig. 30m). then by the main, virtually square Castillo,
Three pieces have geometric designs in Unit A. Likewise, the interior arrangement of
champlev6 technique (Fig. 29n). rooms and galleries within the Castillo follows a
Two hundred and thirteen decorated pieces symmetrical plan.
in total.
The frequency of the design elements as such THE MASS EFFECT
is summarized in the following table. The platform type units which make up the
OCCURRENCE OF DESIGN ELEMENTS IN
Chavin ruins give a definite effect of mass.
CHAVfN CERMICS First, the platforms in groundplan and height
Fre- are large in gross size. The high solid east wall
Design Element quency of the Castillo is somewhat inclined and broken
Linear incised or groove . . . . . . . . 90 near the top by two step-back terraces, but the
Curvilinear ............. . 48 appearance is that of a great block, not of a
Dash ........ .. . .. . . . 3 stepped pyramid. No windows or other open-
Complex figure . . . . . . . . . . . . 5 ings break this effect.
The two platforms, Units F and G, have the three or more floors of galleries, the wall
sometimes been called forts, but none of the niches, the connecting ramps and stairways
details normally found in fortifications are should be noted.
present. In spite of the great mass of building,
the effect of fortification is not conveyed. MISCELLANEOUS DETAILS
Several details which need no elaboration
CONSTRUCTION NOTES here but which are useful for comparative pur-
The general technique of construction is one poses should be mentioned: first, the projecting
of planned interior galleries and vents, lined and decorated cornice still to be seen at the south-
covered with stones, and an exterior facing wall west corner of the Castillo; second, the carved
of stone, with the remaining interior filled with heads set into the walls with tenons; third,
rubble and capped with clay. The large mass the presence of a carved stone within one of
platforms thus built up serve, in the Castillo, the galleries; fourth, the house foundations
as a base for house construction. All platforms with their dressed stone walls on top of the
are flanked by low terraces. The height of some Castillo. These are the only free standing walls
of the units is attained by inclining the facing found around the ruins. Other house sites were
walls and setting back terraces near the top. not seen, unless the small cells between the
Corners are well finished, in both the dressed plaza and the river are so considered. Fifth,
and the rough stone walls. Slabs covering wide the steps leading up the east face of the Castillo,
rooms or galleries are reinforced by stones which and a set of monolithic steps near the river bank
project inward from the top of the side walls, should be mentioned, as well as a dressed canal
although the corbeled arch itself is not em- stone near the river bank. Small canals in vari-
ployed. ous sections of the ruins might well have served
STONEWORK for drainage.
Carefully dressed stone slabs form the facing STONE CARVING
of much of the Castillo, Unit A, and part of
Unit B, but are not seen in other parts of the Our excavations at Chavin did not uncover
ruins, and never in the interiors. The stone is any carved stonework, nor did they add ma-
skilfully dressed, but such Tiahuanaco features terially to the existing knowledge of this impor-
as niche joints, tenon joints, and copper cramps tant aspect of Chavin culture. Illustrations of
are not employed. 72 pieces ofstone carving, all probably from the
The standard facing technique is to set stones ruins, were assembled. Some of these were seen
in horizontal rows, alternating wide and narrow and photographed by us, some have been pub-
courses. This applies to the split stone walls of lished by Julio C. Tello, some were photo-
the interior gallery as well as the dressed stone graphed by Donald Collier, and many others
facings, although the rougher slabs required were traced by Humberto Hidalgo who sold
small stones to fill the intervening chinks. copies of the drawings to us. The 72 pieces can
While all walls are not equally well faced, the be grouped as follows:
technique shows no major variation. The Tia- 25 Heads carved in the round (Pi. 8b, c), including
huanaco facing style of upright slabs set at in- twelve representing human heads with tenons for
tervals with small stones between is not found. wall attachment, and 13 representing animal heads,
Selected and split stone is used in the interior all but one (PI. 8a) of which have tenons
and on some outer facing walls. Large stone 33 Flat slabs with incised design, of which 17 may
slabs are used as gallery covers. be cornices, like the one still in position in the Chavin
Castillo
8 Flat slabs which appear to be parts of stelae
GALLERIES like the Raimondi Stone or the Tello obelisk
The internal galleries, small rooms and the 1 Lanzon, with carving in the round, still in
connecting ventilation shafts are the outstand- position in the Chavin gallery
ing characteristics of Chavin construction. 5 Pieces so different in style as to be classed as
While these have been sufficiently described non-Chavin in spite of being found in the region
in this report and elsewhere, such details as Twenty-six of the above pieces had been illus-
trated in the literature before our expedition' concept dominates all of Chavin design. Only in
and in a recent publication' of 30 of Humberto the one group of carved heads is the human be-
Hidalgo's drawings, 24 new ones are included. ing depicted. Otherwise the feline figure is uni-
This total of 50 illustrated pieces furnishes an versal, either as the main design in profile or in
excellent basis for a study of Chavin stone carv- zoomorphized front view, or as the basic figure
ing style. Julio C. Tello has published3 a de- with appendages attached, such as bill and
tailed analysis of Chavin stone carving style, wings for a condor, undulating body for a snake,
particularly in reference to four outstanding tail and fins for a fish. In decorative design de-
pieces. As a preface to the Hidalgo drawings4 tail the feline appears most frequently and with
an analysis of stone carving design elements is considerable variation, but can be recognized in
presented. Other discussions of Chavin style profile, front and top view. Even more limited
can be found in Means,5 Kroeber,6 and else- design elements seem to be derived from feline
where. Consequently, only a brief resum6 is eyes, ears, whiskers, tongue, and teeth. In no
included in this report. other Peruvian art style is a single design con-
Chavin carving is divisible into two cate- cept so dominant.
gories: carving in the round and flat carving. The brevity of the present discussion of
Exceptions to this rule are few and not of major Chavin stone carving is not to be taken as in-
importance. The carving in the round is largely dicative of its importance. For the most part,
limited to human and animal heads with tenons Chavin stone carving is known from the site
for wall insertion. Both types have been ade- itself, although Tello7 reports the style from the
quately described in the literature cited above. provinces of Huari, Pomabamba, and Pallasca,
It is interesting to note that none of the Chavin and future exploration may widen the distribu-
carvings can properly be called a statue. This tion still more. On the other hand, the stone
is an outstanding contrast to the Callej6n de carving design is the chief diagnostic of the
Huaylas with its numerous stone statues. Chavin period. On the coast of Peru materials
Flat surface carving is definitely related to with Chavin stone carving design are varied
that in the round. The techniques are low relief, and numerous. Thus the design appears on in-
champlev6, and incision. The styles range from cised clay walls at Nepefna, on ceramics at many
the completely curvilinear, through combina- sites on the North Coast, on goldwork at
tions of angular and curvilinear, to a rare angu- Chongoyape, and on small objects of shell,
lar design. Other generalized characteristics are bone, and stone from many sites. In fact, the
symmetry and a tendency to fill the surface numerous recent discoveries indicate that Cha-
space by elaboration of design detail. vin stone carving style is one of outstanding
As has been frequently pointed out, the feline importance in Andean archaeology.
POST-CHAVIN MATERIALS
Definitely post-Chavin materials were found facing walls with window niches in Pit Ch-3; a
in eleven of the sixteen sites as well as in surface baked clay plastered house in Pit Ch-3; walled
collections (Sites Ch-1, 2, 3, 4, 5, 8, 11, 13, 14, pits with slab covers in Pits Ch-3 and 16; and
15, 16). Post-Chavin materials are represented a small canal in Pit Ch-3.
in several categories: Graves. At least 31 graves are represented,
Constructions. Rough stone division walls including ten clearly of the stone box type.
were found in Pits Ch-2 and 16; rough stone These were found in Pits Ch-13, 14, and 16.
1
Borchers, 1935, 352; Larco Hoyle, 1938, Fig. 25; Tello, Eighteen of the graves were without definitely
1923, Figs. 59,70, and 77, and Pls. I, II, III, IV; Tello, 1929, associated contents, and the remaining 13 con-
Figs. 22 to 38 inclusive; Roosevelt, 1935, Fig. 29. tained a total of 25 vessels and three spoons.
2 Bennett, 1942, Figs. 1 to 30 inclusive, of which Figs.
1, 2, 3, 6, 7, 8 have been previously illustrated. Sherds. A total of 4306 sherds from the 11
' Tello, 1923, especially 256 to 320. sites with post-Chavin materials are available
4 Bennett, 1942. for analysis.
6 Means, 1931, 139-144.
6 Kroeber, 1926, 36-39. ' Tello, 1923, 256.
Miscellaneous artifacts of post-Chavin style POST-CHAVfN CERAMIC STYLES
were found, including 14 stone objects, 15 bone The post-Chavin materials lack two of the
objects, three clay whorls, and a shell. important Huaraz styles, namely, Recuay and
Stone sculpture of post-Chavin style was not Wilkawain-Tiahuanacoid. No three-color negai
encountered. Some pieces of stone carving from tive fragments, modeled pieces, trumpets, shortE
Chavin can be called non-Chav'n style, and spouts, modeled figure pieces, or any other posi-
possibly these represent a later period. The ab- tive suggestion of Recuay style occur in the
sence of positive evidence, however, makes it excavated collection. The Recuay style is repre-
preferable to omit these from consideration here. sented in collections from the pueblo of San
No materials suggesting a pre-Chavin hori- Marcos, so its absence from the Chavin collec-
zon, nor even styles contemporaneous with tions is not indicative of any limited distribu-
Chavin were found. The materials here dis- tion of Recuay. Likewise the Tiahuanacoid
cussed are all definitely post-Chavin by virtue style, found in such abundance near Wilkawain,
of stratigraphic position in some pits, grave in- is not clearly represented at Chavin. Tiahuana-
trusion into the Chavin construction in others. coid designs are not found. While some simple
The styles represented present no discrepancies one-color on slip pieces, such as black-on-red,
with this interpretation. red-on-orange, might conceivably be Tiahuana.
All of the materials, the construction work as coid, only five pieces of two-color ware were
well as the ceramics, are not only post-Chavin found and of these only one was the typical
in temporal position, but definitely non-Chavin black, white-on-red. Three-color wares were not
in style. None presents the slightest indication represented. (These statements about two- and
of Chavin influence. three-color ware do not include those fragments
Treated as a whole, for the moment, the of the San Jer6nimo type.) Furthermore, such
post-Chavin ceramics are predominantly plain- associated characteristics of Wilkawain as two-
wares. The 25 complete vessels include 13 plain color negativeware, double-spout black and red-
orange or crude brown, two polished red, and ware, modeled pumas and the like, are not rep-
ten with one-color design on a slip background. resented in the collection. Both a Recuay and a
The 4306 sherds can be grouped as follows: Tiahuanacoid style should be found in the
3766 Plain brown, orange, gray region, and thus it is obvious that our Chavin
285 Polished monochrome red, black, brown, excavations give a very incomplete picture.
orange
215 One-color on slip background INCA STYLE
33 Pieces of San Jer6nimo painted style in The pair of redware goblets (Fig. 28a) found
which interior colors and combinations dif- in Ch-3A are good Inca types, although they
fer from exterior lack painted design. No other pieces, either
5 Pieces with two colors on slip background complete or fragmentary, represent the Cuzco-
2 Negative painted pieces (two-color)
4306 Total Inca typical design or shapes.
Direct stratigraphy in reference to the post- SAN JER6NIMO STTLE
Chavin materials proved of little significance. A surface collection from San Jer6nimo, near
The materials from Pit Ch-2 were segregated Huaraz, has already been described. The style
into six arbitrary levels, and Pit Ch-3 was di- appeared to be Inca-influenced, characterized
vided into three levels. These arbitrary divi- by change in color combinations on the inside
sions have no significance in terms of the and outside of fragments. Similar pieces were
materials. Considering the pits as units, certain found in the sherd collections from the Chavin
differences can be pointed out, but their ulti- Pits Ch-2, 3, 4, and 16 (see Fig. 31a-p for
mate temporal significance cannot be finally style). At San Jer6nimo these painted pieces
determined. Most of the materials do not seem had a surface association with brown, yellow,
of sufficient importance to warrant detailed gray, red, and black fragments, as well as white-
analysis. Consequently, the recognizable styles on-red, black-on-red, red-on-orange and black,
are discussed below and the majority of the white-on-orange. All of these associated types
plainware sherds are omitted from major con- are also found with the painted style at Chavin.
sideration. Although we cannot insist that all fragments
*. *.. *-'
F I
- v...:
P.i
Q R
FIG. 31. Post-Chavin ceramic styles from Chavin and San Jer6nimo.
associated with the San Jer6nimo style at tance in checking Coast sequences in which a
Chavin are contemporaneous, it does appear Chavin style is also involved.
that the bulk of the collection from Pits Ch-2 MINOR STYLES
and 3 (2941 fragments in total) belongs to a
Late style, probably of about Incaic time. The remaining post-Chavin material is not
This is confirmed in part by the pirca facing too distinctive. The following minor styles are
walls of Pit Ch-3. Likewise, the grave pieces mentioned, although the final significance is
from Ch-3 do not disturb this analysis, since, unknown.
aside from the two Inca goblets of Ch-3A, only RED-ON-ORANGE
crude ollas (Fig. 28b, c) and spoons are in- The two graves of Ch-8 contained two shal-
cluded. low plate-bowls, one orange and one with a
WHITE-ON-RED STYLE wavy red line on an orange base; a toy-like olla
with a red band on an orange base and a two-
A grave, Site 7H-15, at Wilkawain contained handled orange olla. Red-on-orange fragments
a series of vessels which isolate a white-on-red occur also in Ch-2 (13 pieces), Ch-3 (two
style. This same style occurs at Chavin in in- pieces), Ch-4 (four pieces), Ch-5 (two pieces),
trusive graves in Ch-13 and 14 (seven complete and Ch-16 (five pieces). The general position is
[Fig. 31r] or partially complete open vessels that of a Late style.
plus some fragments), and in Ch-16 (one com-
plete and some 20 fragments from open ves- RED-ON-WHITE
sels). Perhaps the fragments of seven other An open, plain-rim bowl with annular base
vessels from Ch-11 should be included, and was found in Ch-5 (Fig. 31q), together with a
likewise the ten small fragments from Ch-3. crude brown jar. The open-bowl was decorated
The shapes of these vessels are all variants on with red bands on a dull whitish base. Frag-
open-bowls. These include a plain rim bowl ments ofvessels of similar shape and design were
with straight diverging sides and a rounded found in Ch-2 (14 pieces), Ch-3 (five pieces),
base; a similar bowl with slightly converging and Ch-4 (two pieces). Again the associations
sides; a plain rim bowl with convex sides; a suggest this as a Late style, in spite of super-
flare-rim bowl with convex sides; and a some- ficial resemblance to one type of Recuay vessel.
what plate-like vessel. Of the fifteen complete
or partially complete bowls which might be OTHER POST-CHAVfN ARTIFACTS
grouped into this style, three have white-on- Eleven miscellaneous artifacts found in Ch-2
red design, eight are plain orange, two are red, and Ch-3 apparently belong to a Late period.
one black, one red-on-white, and one red-on- These include half of a circular stone with a
orange. The twenty fragments from Ch-16 are depression on each side, a ring stone, and a
all white-on-red designed. The decorated ves- grooved stone bead; a cut knuckle bone, two
sels, then, show combinations of straight hori- deerhorn points, and three polished bone points;
zontal, vertical, and diagonal lines; wavy lines; a flat disc clay whorl; and a seashell. In Ch-14
triangular areas filled with dots; and circles. a leaf-shaped stone knife and part of another,
The white-on-red is definitely post-Chavin. plus a pierced flat stone, might possibly be
It is isolated in the graves of Ch-13 and 14. part of the white-on-red period.
Both white-on-red and San Jer6nimo style The 19 miscellaneous objects of Ch-16 are
were found at Ch-16 in the upper meter, but probably of a Late period, since most came from
definite stratigraphy was not possible. the upper meter, although this is not certain.
Recent excavations by Gordon Willey at The stone objects are a flat ring stone and part
Chancay are giving new importance to the of another, a pierced ball club head, a polished
Chancay white-on-red style. The white-on-red T-shaped ax, a rough chipped ax, a piece of a
of Chavin and Huaraz show certain similarities bowl, a long thin knife, and a fragment of obsid-
to the Coast style. If new evidence enhances the ian. Bone pieces are a spoon, three points, a
significance of these similarities the position of needle, a knife, and three polished pieces. Two
white-on-red as post-Chavin will be of impor- flat disc clay whorls were also found.
PART 3. THE NORTH HIGHLANDS
SINCE 1938 WHEN THESE EXCAVATIONS were The material resulting from our excavations
effected, Peru has become the scene of intensive falls largely within those broad Peruvian time
archaeological activity. Julio C. Tello has car- categories called "Early" and "Middle" peri-
ried out considerable exploration in many parts ods, plus a certain amount of material which
of Peru, including the Callej6n de Huaylas and might be classed as "Late" period and Inca.
Chavin. Rafael Larco Hoyle has directed ex- Three distinct styles fall into the Early period
tensive excavations in Chicama, Viru, Santa, category. These are Chavin, Recuay, and the
and other coastal valleys. He has uncovered white-on-red styles. The excavations at Chavin
numerous new graves of the Chavin type and demonstrated that the white-on-red style was
has isolated a new style of material called "El post-Chavin, but we encountered no evidence
Salinar." Other Peruvian archaeologists such for the relative positions of Chavin and Recuay
as Luis E. Valcircel and Augusto Soriano In- or Recuay and the white-on-red. At Wilkawain,
fante have continued their archaeological work. Recuay materials were found stratigraphically
The 1941 program of the Institute of Andean under the Epigonal or Tiahuanaco-influenced
Research brought many new archaeologists into style. Since the latter generally characterizes
the field and allowed several others to return the Middle period everywhere in Peru, the plac-
again. Theodore McCown excavated in the ing of Recuay in the Early period is justified.
North Highlands, principally around Huama- The white-on-red style is classed as an Early
chuco. Duncan Strong, Gordon Willey, and period because of its position on the coast of
John Corbett worked at many points along the Peru in pre-Tiahuanaco levels. A Chavin style
Central Coast including Chancay, Supe, Ancon, has now been found on the coast of Peru in
and Pachacamac. A. Kidder, II, and John Rowe positions apparently older than the Mochica
continued their work at Cuzco and at Pucara. or Early Chimu period. There is some slight
S. K. Lothrop and Marshall Newman ran a evidence that the white-on-red style might
program which included surveys and a careful also be pre-Mochica. Judging by this compara-
study of skeletal material. More recently A. L. tive evidence, the sequence of the Early period
Kroeber returned to-Peru for special studies in styles in the Highlands of Peru would be
the field of archaeology. The work of all of Chavin, white-on-red, and Recuay, although
these scholars bears directly on the problems this order is not definitely confirmed by direct
encountered in our excavations in the Callej6n evidence from our excavations.
de Huaylas and at Chavin. While some of the The great bulk of the material we found
work has been published, much of it is still in around the site of Wilkawain belongs to the
the process of preparation. As a consequence Middle period because of the obvious Tiahua-
detailed conclusions and elaborate distribution naco influence. Still other Tiahuanaco-influ-
studies of the materials which we found in the enced styles are found in many of the local
Callej6n de Huaylas and at Chavin are imprac- collections in the Callej6n de Huaylas. The
tical at this time. A careful review of the total Middle period, then, can be divided into a
materials involved is obviously needed, but this Wilkawain-Tiahuanaco and an "Epigonal," al-
must wait until these important new discover- though evidence for the order in time of these
ies have been published. In the light of these two is missing. Late periods are poorly repre-
facts, the final section of this paper is nothing sented in our collecitons. While a great mass of
more than a resume of our own work done in material was encountered in a Late position at
1938, plus some general observations on the Chavin none of it represents any clearly defined
Callej6n de Huaylas and Chavin and few com- style. Finally, a certain amount of material
parative notes. The essential value of this re- seems to correspond in a generalized way to the
port lies in the descriptive sections already pre- Inca period. The resume which follows is ar-
sented, and we are aware that any but the ranged in this general chronological order of
most generalized conclusions would be out- Early, Middle, and Late periods.
dated even before they appeared.
93
CHAVfN AS A WHOLE
Our excavations at the site of Chavin, plus antiquity among the materials excavated by us,
study of other materials, reveal a Chavin period and by certain associations with architectural
followed by a series of post-Chavin styles, the features. In many ways the ceramic design is a
most important of which seem to be the white. decided contrast to the stone carving design.
on-red and an Inca-influenced style. That further excavation might reveal ceramic
The Chavin period at its home site is repre- designs as similar to the Chavin stone carving
sented by an architectural style, a stone carving style as the Cupisnique pieces in the Larco
style, and a ceramic style. With the exception Hoyle Collection at Chiclin Museum is, of
of a fragment of an open stone bowl, other type course, possible. On the other hand, both the
materials pertaining to this period were not Early Ancon and Puerto de Supe ceramic styles
found in our excavations, although it is logical tend to confirm the validity of a Chavin ceramic
to assume that new excavation will produce style as distinct from the stone carving.
more evidence, including gold objects. The While sufficient description of the major
stone carving is definitely associated as parts of features of the Chavin site have been given in
some of the buildings. The ceramics are in- that section, a brief resume follows, for con-
cluded by virtue of their exclusive claim for venience of comparison with other sites.
ARCHITECTURE
1. Platform construction of massive type, in 7. Symmetry of general plan.
which the effect of mass is gained by window- 8. Orientation following roughly on east to
less walls and unbroken terraces. west and north to south alignment.
2. Interior rooms and galleries arranged on 9. Complexity of total site including ar-
three or more floor levels. rangement of buildings, courts, terraces, and
3. Stone-lined ventilating shafts. the like.
4. Planned construction based on building 10. Free standing walls only as foundations
the stone-lined and slab-covered rooms, galler- of houses.
ies, and vents and exterior facing walls and I1. Grooved canal stones.
filling the space in between with rubble. 12. Stairways and steps cut out of single
5. Dressed stone facing walls and corners. blocks.
6. Building technique of horizontal rows of 13. Projecting carved cornices and carved
slabs, alternating thick and thin. Chinks are heads inserted in walls.
filled with small stones. The horizontal row 14. Sunken court or plaza.
technique is employed both in the well-dressed 15. Small cell-like rooms apart from the main
stones and in the split stone. buildings.
STONE CARVING
1. Carving in the round and on flat surfaces. and snake figures. Design detail emphasizes
2. Human and feline heads carved in the the feline features in decorative units represent-
round with tenons for wall insertion (PI. 8b, c). ing profile, front view and top view heads, plus
3. One head carved in the round, but with- other feline-derived units.
out tenon (P1. 8a). 6. Design style ranging from curvilinear to
4. No true statues. No depiction of human angular.
being except for carved heads. 7. Flat carving technique of incision and
5. Dominance of feline motive, particularly champlev6, but always low relief.
in the flat carving. The feline is represented as 8. Design highly stylized and flamboyant.
the main figure in profile head, profile body, and 9. General tendency for symmetry in design.
anthropomorphized front view. The feline fig- 10. Emphasis on detail and filling of total
ure forms the basis for depiction of condor, fish, available space.
CERAMICS
1. Well-polished monochrome wares, includ- such as lines, circles, triangles, angles, S-shapes,
ing black, brown, red, and orange. and the like. Some more complicated curvilinear
2. Some black and redware with color areas designs, but designs paralleling those on the
separated by incisions. stone carving are not common.
3. Great rim variation, emphasizing thick These brief summaries may not give an ade-
rim edges, beveled rim edges, and bulging rim quate picture of the Chavin materials. It can
edges. be said, however, that architecture, stone carv-
4. Shapes dominated by variations on shal- ing, and ceramics are all easily distinguishable
low open-bowls with flat or slightly rounded in Peruvian materials. In spite of this fact the
bases, and convex or flaring sides. Other forms combination of the three at the site of Chavin
include a constricted-collar jar, a globular bowl is, to my knowledge, unique in the Andean
with flare-rim, and some spouted vessels pos- region. One cannot be certain that Chavin de
sibly representing stirrup-spouts. Annular bases, Huintar represents the center of distribution of
tripod bases, and handles are not found. the Chavin culture and style. In fact Larco
5. Design dominantly fine-line and groove Hoyle (1938) is inclined to believe that it is
incision, but also includes punch, gouge, scratch, a Highland manifestation of an essentially
champlev6, raised strips, and tabs. Positive Coastal period. Whatever the final answer may
color painting limited to two colors. Highly be, Chavin is, so far, the most elaborate mani-
questionable negative painting. Modeling oc- festation of Chavin style as a whole and stands
curs, but is rare. isolated.
6. Designs are largely geometric in pattern,
CHAVfN COMPARISONS AND DISTRIBUTION

In a former paper1 a brief discussion of the Since the majority of definite Chavin-style
Chavin distribution problem was given. In a finds have been made on the coast of Peru one
more recent publication2 the problem was re- would naturally expect that the Callej6n de
viewed in the light of numerous new discov- Huaylas, being geographically between Chavin
eries. It is not necessary to repeat these de Huintar and the coast, would be a key region
discussions here, but certain new points may in the distribution problem. Doubtless it will
be considered in reference to the excavations at be, but up to the present little evidence of
Chavin itself. The comparisons which follow Chavin in the Callej6n has been presented. Of
refer to the architectural style, the stone carv- the known ceramic styles, stone carvings, and
ing style, and the ceramic style. architectural features of the Callej6n, only the
CHAVflN IN THE NORTH HIGHLANDS last category has much bearing on the problem.
The ceramics seen by this expedition were
The area of the North Highlands is obviously definitely non-Chavin in style. While the Cal-
the most logical place to seek further evidence lej6n contains much stone carving, only the
of Chavin. Unfortunately little work has been carved puma heads with tenons for wall inser-
done and still less evidence published. Julio C. tion might be said to resemble Chavin heads,
Tello reports Chavin-style finds in many sec- and even in these, the style is distinct. In the
tions, but his published evidence has not yet three-storied temples, such as those described
appeared. Augusto Soriano Infante collected at Wilkawain and Ichik-Wilkawain, one finds
Chavin ceramic fragments near the town of the closest parallels. The Wilkawain temple
San Marcos, not far from Chavin itself. Other- had three floors of galleries and rooms with a
wise, this important aspect of the Chavin dis- built-in vent system. Cornices and projecting
tribution problem must await further discovery carved heads decorated the outer wall. The
and clarification. walls of split stone were laid in the horizontal
1 Bennett, 1939, 127-133. technique of Chavin. One feels that some rela-
2 Bennett, 1943. tionships exists, but in the absence of other
supporting material it is impossible to define style of Chavin. Outstanding in Tiahuanaco
it with accuracy. architecture are the elaborately jointed and
grooved stones employing copper cramps. These
CHAVfN AND TIAHUANACO are not found at Chavin. From the point of
Most Andean archaeologists have been im- view of architecture, then, the resemblances of
pressed by the fundamental resemblance of Chavin and Tiahuanaco are generalized rather
Chavin and Tiahuanaco. There is, however, than specific.
considerable disagreement in the interpretation In ceramics even a greater contrast is found.
of this resemblance. Thus, Joyce, Posnansky, Here one can go so far as to say that there is
Markham, Gonzailez de la Rosa, and Means virtually no resemblance between the poly-
argue that the Chavin style is probably later chrome Tiahuanaco pottery and the mono-
than that of Tiahuanaco. In fact, all of these chrome incised Chavin pottery. We are thus
men feel that Chavin is a highly stylized phase forced to return to stone carving as the only
of the Tiahuanaco period. Means' sums up the ground for the resemblance of the two sites.
argument in his comparison of the Raimondi Like Chavin, Tiahuanaco has sculptured
Stone and the Viracocha figure on the Gateway stone heads (although these are usually on
of the Sun. He concludes that the two are un- blocks rather than on tenons for wall inser-
questionably derived from the same funda- tions) and carvings in low relief on gateways
mental concept. At the same time, he suggests and on flat stone. The chief difference is the
that the differences are the result of mixing of occurrence of numerous statues at Tiahuanaco.
Tiahuanaco style with Nazca and that the While the feline figure, usually identified as a
combination explains the Chavin style. Tello puma, is an important design in Tiahuanaco
and Uhle take the opposite view, namely, that carving, the carving is not dominated by the
the Chavin style antedates that of Tiahuanaco. feline concept as at Chavfn; instead, human,
Uhle2 states definitely that the Raimondi Stone bird, and snake figures play an important part.
is not a copy of the Gateway of the Sun but Without going into a detailed comparison, it
rather a stylistic predecessor. Muelle (1937) still appears that the resemblance between
has summed up the argument on this problem. Tiahuanaco and Chavin carving, from a sty-
His own conclusion stresses the resemblance of listic angle, is generalized rather than specific.
Chavin carving, particularly the Raimondi The only concrete evidence for the chrono-
Stone, and Nazca B style. In all these discus- logical relationship of Tiahuanaco and Chavin
sions the arguments are based on a comparison occurs on the coast of Peru. Here there is no
of stone carving, with little or no mention of question that the Chavin style antedates, by
architectural style or ceramics. considerable time, the distribution of Tiahua-
Both Chavin and Tiahuanaco represent ag- naco style. While this in turn would suggest
gregations of architectural features, buildings, that Chavin as a whole was older than Tiahua-
platforms, and the like, arranged according to naco as a whole, it does not offer conclusive
a large scale plan. Both employ well-dressed proof. In general formation one feels that both
stone, stairways, grooved canal stones, and Tiahuanaco and Chavin are derived from the
other similar features. Both have sunken courts same fundamental concepts, but that these con-
and somewhat oriented units. Both have stone cepts are generalized Andean rather than spe-
carving associated as an architectural feature, cifically either Tiahuanaco or Chavin. At this
both in the form of low incised relief and time there is no good evidence to suggest that
sculptured heads. At the same time the differ- Tiahuanaco is another manifestation of Chavin,
ences are numerous. The built-up platforms or the reverse.
with the interior rooms, galleries, and vents are The south highland Peruvian site of Pucara,
characteristic of Chavin, but are not found at which has been extensively excavated by Alfred
Tiahuanaco. While both use dressed stones in a Kidder, II, is thought by some archaeologists
facing technique, the Tiahuanaco style con- to bear a reasonably close resemblance to
sists of upright pillars with small stones fitted Chavin. At the same time the Tiahuanaco re-
in between, in contrast to the horizontal row semblances cannot be denied. Until Kidder's
1 Means, 1931, 139. work is published no sound conclusions can be
2 Uhle, 1920,
155. drawn in reference to Pucara.
CHAVfN AND COAST STYLES objects besides pottery. The absence of stone
In recent years a great mass of new evidence carving or building styles leaves only the
on Chavin on the coast of Peru has been ac. ceramics for comparison with Chavin, since
cumulated by Larco Hoyle, Tello, Willey, at the latter site such perishable materials as
Strong, and others. In the light of this mass of baskets, nets, and textiles are not preserved.
evidence, much of it still unpublished, it is in- The ceramic collection is unfortunately small
advisable to draw conclusions at this time. (eight whole pots, three fragments of one, and
Consequently, we will pass over rather briefly a few sherds), but even so it is amazingly similar
the consideration of this important aspect of to Chavi'n ceramics in thickness of ware, colors,
the Chavin period. shapes, incised decoration, and the like. Kroebers
states that the Puerto de Supe material is simi-
CHAViN AND EARLY ANCON lar to but not identical with the Early Ancon.
Early Ancon1 materials come from a settle- The differences are not obvious from the de-
ment refuse. No architecture or stone carving scriptions and illustrated materials. The un-
is available, and thus the comparison is between decorated vessels of Supe are another point of
the small sherd collection and the ceramics of similarity with Chavin where only a small per-
Chavin. Such a comparison shows an amazing centage of the fragments are decorated.
similarity between Chavin and Early Ancon Just why the closest parallels to Chavin
ceramics. Some points of similarity are listed: ceramics are represented at Early Ancon and
1. Thick polished wares ranging from black
Supe fishing sites cannot be answered. Pre-
through brown to red sumably other sites which will help explain this
2. Decoration by incision, in which groove in- paradox will eventually be found. Again the
cision is typical 1941 excavations of Willey must be awaited
3. Design detail similar, such as parallel lines, for Puerto de Supe conclusions.
circles with dots, zigzag, punch areas, curvilinear
grooves CHAVfN AND CUPISNIQUE
4. Wide incised lines filled with paint (red, at Cupisnique Quebrada in Chicama Valley con-
Chavin; white, pink, and yellow at Ancon) tains surface sherds,4 many of which resemble
5. Two-color areas separated by incisions
6. Suggested shapes of Ancon similar to Chavin. Chavin. Likewise, complete vessels resembling
Most pieces from open-bowls plus rim variation. this style are found in the Chiclin Museum of
7. No stirrup-spout vessels suggested at Ancon Sr. Rafael Larco Hoyle and in other Peruvian
(some are suggested in the Chavin collection, but museum collections. No significant construction
certainly they are rare) nor stone carving has been found, so the com-
8. Dr. W. D. Strong reports rare pieces of nega- parison is essentially one of ceramics. A com-
tiveware at Ancon, and one rare piece found at parison of Chavin sherds with one collection
Chavin from Cupisnique Quebrada shows:
9. Stone mortar at Ancon like stone bowl of
Chavin
10. No metals associated Chavln Cupisnique
Early Ancon has a figurine type and a stone Per Per
club head not encountered in our Chavin col- No.N. Cent No. Cent
lection. Even though the Early Ancon collec- Blackware . . . . . 1374 45.95 412 54.07
tion is small and further excavation is neces- Brownware . . . . 901 30.13 232 30.45
sary, it seems clear that the relationship of Redware . . . . . 427 14.28 79 10.37
Chavin and Early Ancon ceramics is remark- Orangeware . . . . 272 9.10
ably close. The recent work of Willey at Ancon Grayware . . . . . 17 2.23
should clarify this situation. Two-colorware . 9 .30 22 2.87
Miscellaneous . 7 .24
CHAVfN AND PUERTO DE SUPE
The finds of Dr. Max Uhle at two sites near Totals . . . . . 2990 100.00 762 100.00
Puerto de Supe2 include burials and numerous 3 Kroeber, 1925a, 255.
1 See
Strong, 1925. 2See Kroeber, 1925a. 4See Bennett, 1939, 90-93.
A general comparison of Chavlin ceramics the relative ages of the two styles, but the argu-
with Cupisnique, both of the sherds from the ment that the Chavin ceramic style is older
Quebrada and the complete vessels from collec- seems to be a simpler one than the reverse.
tions, shows a convincing series of similarities
and also a fair number of differences. Some of CHAVfN AND COAST STYLES
the outstanding similarities are: With the exception of Ancon and Supe ma-
1. Black, brown, and redwares in about the same terials the identification of Coast styles with
proportion Chavin characteristics is essentially based on
2. Dominance of incised design as against painted the Chavin stone carving style. The Early
3. Similarity of design techniques, such as punch, Chimu pieces in the Uhle Moche collection
scratch, raised tabs, relief, stamp, and the like with Chavin designs, the Lambayeque gold col-
4. Similarity of design elements, such as circles lection from Chongoyape,1 the relief decorated
and dots, circle combinations, linear and curvilinear, walls of Nepenia all correspond to the Chavin
and broken lines stone carving style. Furthermore, it must be
5. Some similar shapes, such as a few open-bowls, noted that relatively no variation is presented
and a constricted-collar jar
6. Painted areas separated by incisions by these manifestations of Chavin design in
such widely separated regions and in such di-
Such a list of similarities would be sufficient vergent media. The recent work of Larco (1941)
to establish a relationship between the two presents a far sounder basis for discussing
groups, but the differences must also be con- Chavin coastal periods than previously existed,
sidered: but has not completely clarified the Coast-
1. At Cuspisnique the stirrup-spout is the most Highland relationships.
characteristic shape. Chavin may have stirrup- Although, as previously stated, it is prema..
spouts, but they are certainly rare. ture to give detailed conclusions, the following
2. Open-bowls are rare in Cupisnique, character- general statements about Chavin seem valid.
istic of Chavin 1. Chavin presents a distinct architectural,
3. Cupisnique incision is fine line in most pieces; ceramic, and sculptural style.
Chavin is groove incision 2. Materials classified as Chavin style have
4. Cupisnique designs, especially on complete
vessels, are directly comparable to Chavin stone been found over a wide area in the Andean re-
carving designs. Chavin ceramic design does not gion. On the coast of Peru evidence of Chavin
contain many of these elements. influence or style has been produced from the
5. Cupisnique lacks many Chavin ceramic design Piura Valley on the far north coast to Paracas
techniques such as fill of incised line with paint, on the south coast, though with greatest con-
champlev6, negative painting, ratchet, and stamp centration on the central and north coasts. In
punching, thumb-nail impression the highlands the site of Chavin is itself out-
6. Cupisnique uses design techniques not found in standing and there is certain evidence of Chavin
Chavin, such as fine scratch, cross-hatch, appliqu6 influence in the south highlands.
pellets and strips, modeling
3. Subdivisions of Chavin will be made even-
A comparison of Chavin and Cupisnique tually. There are at least two ceramic styles and
leaves no doubt of the basic relationship of the two architectural styles.
two styles. The Cupisnique style differs essen- 4. Chavin materials are definitely early on
tially in the abundance of stirrup-spout vessels the north coast, preceding the Mochica or Early
as against open-bowls and the use of Chavin Chimu period. In general, it appears that Cha.
stone-carving designs in its incised and relief vin everywhere is definitely early in terms of
decoration. No new evidence is available for known Andean periods.
WHITE-ON-RED STYLE
The white-on-red style was found in isolated local evidence for the relative dating of this
graves at Wilkawain (7H-15) in a mixed house style. The following shapes are represented in
site (9H-2), and in intrusive graves at Chavin. the collection:
The position of the white-on-red as post-Chavin
is well established; otherwise there is no good 1 See Lothrop, 1941.
1. Open-bowl. This is by far the commonest A simple incised design is found on the flat
shape and consists of a simple open-bowl with con- handle of the double bowl. Designs consist of
vex or flaring sides and with either a flat or rounded triangles filled with dots, parallel and wavy
base. lines, angular scrolls, and simple daubs. Burials
2. Double-bowl. One double-bowl has a modeled were direct in unprepared tombs; no architec-
human head on one side and traces of white-on-red tural features could be associated with the
design. It is associated with the isolated white-on-red
graves at Wilkawain. style; and no other artifacts besides ceramics
3. Constricted-neck bowl with flat-loop handle could be definitely associated. While not iden-
4. A shallow plate with an annular base tical with the Coast style found at Chancay, the
5. Shallow plate relationship seems more or less certain. How-
6. Small jar with horizontal handle ever, further interpretation of the antiquity
7. A modeled animal and position of the white-on-red style must
White-on-red is the only painted design, al- await publication of the important new evi-
though some of the vessels are plain red, orange, dence from the Peruvian coast.
and brown.
RECUAY
Recuay style materials were found in our own give an accurate picture of Recuay ceramics.
excavations in stone-lined, subterranean gal- The refuse material furnished some new data,
leries near Wilkawain, in gallery refuse sites as will be mentioned later. The grave material
near Shankaiyan, and in a mixed house site conforms in general to the Recuay style as rep-
refuse. The quantity of the materials is not resented in collections elsewhere. Consequently,
great and the information from excavations a brief classification of Recuay ceramics is
leaves much to be desired in the understanding based on the following material: N nf
of the complexities of distribution and position Pieces
of this important style. The Recuay style has Huaraz Museum Collection ....... 60
been reported in many parts of the Callejon Rafael Larco Hoyle Collection . . 53
de Huaylas, and local collections contain many Augusto Soriano Infante Collection . . . 52 .
examples. Although we examined many sites Tom6s La Rosa Sanchez Collection . . 51

around the town of Recuay we uncovered no American Museum of Natural History Col-
new material. Recuay style has been found east lection ...... . . . . . . . . . 7
of the Cordillera Blanca and west of the Cordil- Seler (1893) illustrations .. . . . . 51. .
lera Negra, particularly around the site of Tello (1923, 1929, 1930) illustrations . . . 56
Aija. Recuay influence has long been noticed in Wilkawain excavations . . . . . . . . . 27
Coast ceramics on the north coast of Peru and Total . .357
recently Rafael Larco Hoyle has reported the Most of these collections contain a greater
discovery of pure Recuay style graves in the number of pieces than those classified in our
lower Santa Valley. hurried survey, and the large Tello Collection
Recuay is represented by a definite ceramic in the Universidad de San Marcos is a major
style, an architectural style, and a stone carv- omission. None the less, the material is suffi-
ing style. Although textiles are not preserved, cient to give a general view of Recuay ceramics
the indirect evidence of stone carving design in spite of the fact that most of the pieces in
and ceramic design indicates that they were the collections have come from graves, and a
made. Our own excavations at Shankaiyan add great amount of selection has taken place.
the association of copper pins, needles, and The frequency of the types and subtypes as
pendants and a few minor shell objects to the well as the occurrence of various styles of deco-
Recuay period. ration is given in the table which follows the
RECUAY CERAMICS descriptive classification. Fig. 32 shows the
major shapes in simple outline. Illustrations
The materials encountered in our excavations can be found in Tello (1923, 1929, 1930) and in
in the Callejkn de Huaylas are not sufficient to Seler (1893).
A-I ='2 B-i

I3'
cI
C-1
D='}
D-2 F
H-I H2
I-1
FIG. 32. Recuay ceramic shapes.

1944 BENNEIT: THE NORTH HIGHLANDS OF PERU1101
Recuay A. Open-Bowls. Simple open-bowls C-1. Annular-Base Cup. This vessel has
are common in collections and are probably slightly convex sides, a plain rim, and an an-
more important than is indicated by the tabular nular base. Some are decorated with a relief
frequency because the collectors are not in- figure on the side, the head extending above
terested in the plainwares. The refuse material the rim edge.
at Shankaiyan contained fragments of many C-2. Head Goblets. Deep goblets with some-
plain bowls and of others with simple red-on- what vertical sides and annular bases have a
orange design, but most of those in the collec- front view face design which occupies the major
tions are painted in two- or three-color negative part of one side, giving the whole goblet the
designs, or in red-on-white patterns. appearance of a head. The face has some relief
A-1. Plain Bowls. Simple open-bowls with features and some painted design. The rim is
plain edge rims, convex sides, and rounded bases. varied in accordance with the head modeling.
A-2. Annular-Base Bowls. Simple open-bowls C-3. Crude Cups. Tello's illustrations' of his
with plain edge rims, convex sides, and annular Andean Archaic style include a number of
bases. varieties of crude cups. Actually his material
A-3. Miscellaneous. A few variations on the forms a distinct group, and a style rarely found
simple open-bowl type occur. One variant is a in private collections.
bowl with a flat base, somewhat vertical sides, Recuay D. Tripod Vessels. Tripod vessels are
and a rim which flares at right angles. Another not too common in the collections examined.
variant represents an annular-base bowl with All were monochrome.
a clay trumpet coiled around the outside. The D-1. Open-Bowl with Tripod. Plain open-
mouthpiece of the trumpet is at the rim edge bowls have three thick, somewhat cylindrical
and the horn is near the base. legs.
Recuay B. Conical Handle Bowls. A hollow D-2. Flare-Rim Bowl with Tripod. A small
or solid conical handle projecting from the side olla-shaped container with flare-rim has elon-
of a bowl is one of the typical features of Recuay gated, conical tripod legs.
ceramics. Several subtypes are distinguishable. Recuay E. Ollas. Common olla types are not
Many of these vessels are monochrome and frequent in the collections, but are found in the
form one category of Tello's Andean Archaic. refuse sites. Most of them are standard
B-1. Plain Bowl with Conical Handle. A sim- globular-bodied ollas with flaring rims. Some
ple vessel with slightly flaring sides and plain are decorated with painted and relief figures
rim edge has a long conical handle projecting on one side.
from one side near the base. Recuay F. Vessels with High Flaring Collars.
B-2. Flare-Rim Bowl with Spout and Conical These vessels have elongated globular bodies
Handle. A small olla-like bowl has a rounded and high flaring collars (PI. 2j). A variant has
base, convex sides, and a flaring rim. A conical two flat loop vertical body handles. Many are
handle projects from one side, and on the op- decorated with negative painted designs.
posite side is a short tubular spout. Recuay G. Flat Disc-Rim Vessels. Excep-
B-3. Measuring Bowls with Conical Handles. tionally wide flat disc rims are a characteristic
This shape vessel, which is typical of both of Recuay. The body of the vessels is frequently
Mochica and Recuay ceramics, has been called globular, with a short collar flaring into the
by many names, such as conical-handled dipper, wide disc. A modeled human, bird, or animal
medida, and the like. It is a squat vessel with head may be under the disc rim, on the shoulder
constricted orifice and no neck or rim. The body of the vessel, and some have a flat handle from
is usually angular, with a rounded base. A the edge of the rim to the shoulder.
conical handle projects from one side. Recuay H. Collar and Short Bridge. Short
B-4. Conical-handled Ladle. A cup-shaped flat bridges, virtually functionless, are typical
bowl, ladle, or dipper has a long, somewhat of Recuay ceramics. Kroeber suggests that
conical handle, usually decorated with a these are probably the residue of a deeply in-
modeled head at the end. grained stylistic trend. Considerable variation
Recuay C. Cups or Goblets. A number of in the containers and the details occurs in this
deep open vessels are classified as cups or
goblets. I Cf. Tello, 1929, Figs. 49, 89.
category, although the commonest container vessels represent modeled buildings, with steps,
shape is globular, with a truncated cone base. shelters, and small modeled figures.
Many of these vessels are decorated with three- Recuay N. Modeled Llamas. Modeled llama
color negative design. Four major subdivisions vessels, with positive or negative decoration,
occur, although all are closely allied. can be subdivided into two categories.
H-1. Collar and Bridge to Human Head. A N-1. Llama Alone. A llama vessel which in
straight cylindrical collar is connected to a some cases has two heads.
modeled human head by a short bridge (P1. 2g). N-2. Llama and Man. A modeled llama with
H-2. Collar and Bridge to Bird Head. Similar a man standing beside it (P1. 2h).
to the above, but with a bird head in place of Recuay 0. Trumpets. Long clay trumpets
the human head. are fairly typical of Recuay.
H-3. Collar and Bridge to Animal. The collar 0-1. Coiled Trumpet. An elongated trumpet
is connected by a short bridge to an animal with a complete coil between mouth and horn.
head, or to a complete modeled animal. Some have negative painted designs.
H4. Collar and Bridge to Roof. A modeled 0-2. Straight Trumpet. Plain orangeware
human head or bust is covered by a roof with trumpets with a mouthpiece and tube which
pillars. The roof is connected to the spout by a gradually diverges to the horn is common in the
short bridge. Wilkawain collections (Fig. 18d, i).
Recuay I. Double Jars. The commonest type Recuay P. Modeled Jars. A number ofvessels
of double jar has two oval containers, one with represent modeled figures (P1. 2i), arracacha,
a cylindrical collar spout, the other with a heads, and the like. None falls into the previous
modeled bird head. The head and spout are categories and thus all are grouped roughly
connected by a very short bridge. Human and together.
animal heads are variants. Some unusual double Recuay Q. Miscellaneous. A number of ves-
jars represent two struggling warriors, con- sels were represented by a single example in the
nected modeled fruits, and copulating figures. collections examined. A flaring-sided vessel
Recuay J. Flat Top Jars with Groups of with wavy rim edge looks like a typical Mochica
Small Figures. Hemispherical shaped vessels piece, but has a Recuay design. A squat con-
with somewhat flattened bases and flat tops tainer with convex sides has a high diverging
have slightly flaring collar spouts and groups collar rim giving the appearance of a spittoon.
of small modeled figures on the flattened top. A ring-shaped container has a collar-spout on
Frequently, one figure is larger than the others one side and two modeled bird heads on the
and the rest are grouped around it. A short other. A tall vessel has a series of bulges from
tubular spout may project from one side of the rim to base. Other miscellaneous shapes also
vessel. Practically all of these jars are decorated occur.
with three-color negative designs. The small Although the Recuay ceramics in the collec-
figure modeling and the arrangement of group tions examined by us fall into the shape cate-
scenes are Recuay characteristics. Obviously gories described above, great variety is still
great variation is achieved. possible. Recuay pottery is hand made and the
Recuay K. Stirrup-Spouts. Stirrup-spouts minor variations in shapes, the combinations
are not too common, but occur with sufficient of small modeled figures, and the range in de-
frequency to be considered a characteristic of sign are enormous. Many of the pieces are
Recuay. These are not mere copies of the monochrome, including redware, blackware.
Mochica type. Instead, the two tubes meet in brownware, grayware, and some whiteware,
an open bowl-like mouth. Containers are gen- Monochrome is even more common in the refuse
erally squat, and may be decorated with a relief collections at Shankaiyan. Here plain orange
modeled bird head. Containers in the form of a pieces are dominant. Likewise cooking ollas,
ring, a bent leg, a coiled snake, and other vari- shallow plates, and open-bowls with simple
ants are found. designs are the most frequent in the refuse
Recuay L. Trifid-Spouts. Three tubes meet- collection. In spite of the differences between
ing in an open bowl-like mouth occur on ring- the refuse collections and the grave material no
shape containers and on other styles. basis for a chronological division of Recuay is
Recuay M. Modeled Castles. A number of available.
FREQUENCY OF SHAPE AND DECORATION IN RECUAY CERAMICS
lMono- Two- Three- R Three- Two-

Shape Totals Mono-|color color Redon color | color Incised Unknown
chrome Negative Negative white Positive Positive
Recuay A. 37
A-i. 16 2 1 4 9
A-2. 18 5 3 4 6
A... 3 3
Recuay B . 53
B-i. 17 8 2 1 6
B-2. it 5 1 1 4
B-3. 19 4 4 8 1 2
B.4. 6 2 1 3
Recuay C . 23
C-1. 7 2 3 2
C-2. 6 3 2 1
C-3. 10 10
Recuay D . 8
D-i 3 3
D-2 5 3 2
Recuay E . 4 1 3
Recuay F . 28 4 2 i3 1 2 6
Recuay G . '9 2 1 15 1
Recuay H . 79
H-i 54 7 34 9 1 3
H-2 12 5 2 4 1
H-3 8 1 1 6
HA.. 5 5
Recuay I 22 6 2 10 2 2
Recuay J '7 16 1
Recuay K. I0 2 3 2 3
Recuay L . 7 4 2 1
Recuay M. 8 1 4 1 2
Recuay N . x6
N-i 8 3 5
N-2 8 1 1 5 1
Recuay O. IO
0-1. 3 2 1
0-2. 7 7
Recuay P . 9 5 4
Recuay Q . 7 1 4 2
Both positive and negative painting are com- isolated as a distinct period style. Three-color
mon in the grave ceramics of Recuay. Red-on- negative of black, white, and red is the most
white is the commonest two-color painting, al- typical.
though some other combinations are noted, and As Tellol and Kroeber2 have pointed out, Re-
the three-color positive is black, white, and red. cuay painted design hinges around figures of
In the refuse material a number of other two- jaguars with combs, condors, and serpents. All
color combinations are all significant, including of these are highly conventionalized and form
white-on-orange, black-on-orange, red-on-or- a consistent design style. Added to this are
ange, and red-on-white. The white-on-red as a
style, however, does not occur. Two-color 1 Tello, 1929, 97.
negative painting is frequent, but cannot be 2Kroeber, 1930, 103-104.
numerous geometric elements including paral- period. This style of statue has been found only
lel, straight and wavy lines, cross hatching, in the Aija region, although the Callej6n de
scrolls, steps, triangles, circles, diamonds, zig- Huaylas has some which are variants of the
zag lines, rows of dots, crosses, and the like. same general theme.
The refuse material in general tends to em- Type C stone carving includes slabs with
phasize the simpler geometric designs. relief design on one surface (Fig. 33a, b). A
Our excavations furnished no new informa- common design is a profile feline figure with
tion on the subject of subdividing the Recuay profile or front view head and it is frequently
period. Nothing was discovered which would combined with a small, front view human
confirm or deny Kroeber's distinction of a figure. Type D is a carved stone head on
Recuay A and Recuay B, nor Tello's distinction squared tenon for wall insertion (Fig. 33c;
of Andean Archaic and Classical Recuay, in P1. 8d). One group represents human heads
spite of the distinctiveness of the Archaic style. with decorated headbands, like the Type A
RECUAY SCULPTURE statues; another includes feline heads, like some
of those at Tiahuanaco or, more vaguely, like
In the general region of the Callej6n de those at Chavin. Of the four major types of
Huaylas are four major types of stone carving stone carving, only the Type B statues can
with a number of minor variants. Type A (Fig. be definitely associated with the Recuay period.
33e) is a seated, cross-legged statue found in the The relief slabs of Type C and the heads of
Callej6n in the vicinity of Huaraz. Statues of Type D are found both in the Aija region and
this group all have the legs drawn up and in the Callej6n. Both types might be associated
crossed with the hands resting on the knees. with Recuay, but an equally good argument
All are male figures. The heads are enlarged, can be advanced that they belong to the Tia-
with straight or slightly flaring noses; round huanaco-influenced Middle period.
button eyes; simple, slit mouths; and half-
moon ears. The headband has simple decora- RECUAY ARCHITECTURE
tions, frequently in the form of hands. Other- In our own explorations Recuay style mate-
wise, these figures are undecorated. These rials were associated only with a simple type of
statues are found in isolated spots around the subterranean gallery. Near Wilkawain these
town of Huaraz and in the local museum, but galleries were from 7 to 20 meters long and at
are not definitely associated with any ruins. least 1 meter below the surface of the ground.
Type B is another kind of seated statue They were entered by stone-lined shafts at one
(P1. 8e, f, g) found in the region of Aija and end. The galleries are lined and covered with
amply illustrated by Tello (1929). Here the large flat slabs, with the intervening chinks
knees are drawn up and the toes turn in, but filled with small stones. The technique is quite
the legs are not crossed. Hands are represented distinct from the characteristic one of Chavin.
on the chest, the right hand frequently holding Near Shankaiyan, similar galleries were evi-
a club and the left hand a shield. A trophy head dently used as habitation sites, judging by the
may be depicted as hanging around the neck. refuse material found therein. These were much
The faces are more elaborately decorated with shallower than the burial galleries near Wilka-
teeth indicated in the mouth, and the eyes are wain. Obviously, Recuay should be associated
formed by a double ring. The headband has with some of the other types of buildings found
elaborate decorations, and quite frequently the in the Callej6n, although we encountered no
whole figure, front, sides, and back, is deco- direct evidence to demonstrate this. Some pos-
rated with incised design (Fig. 33d). These sibilities will be discussed in the brief section on
statues have been found in association with Callej6n architecture.
ruins around Aia, principally that known as
Illawain. The statues definitely belong to the CHRONOLOGICAL POSITION OF RECUAY
Recuay period. Many of the modeled ceramic At Wilkawain we encountered a gallery with
vessels in Recuay have the same warrior figure typical Recuay ceramics stratigraphically un-
with club and shield. Likewise, the detailed, der an above-ground house site containing
incised design on the statue is in many cases Middle period material. Thus, it is clear that the
identical with ceramic designs of the Recuay Recuay period precedes the Middle Tiahua-
E
D
Fio. 33. Stone carving styles in the Callej6n de
Huaylas (Huaraz Museum, Huaraz).
naco-influenced period, and the contrast in to the Gallinazo style of Viru Valley. We had
styles makes the separation more marked. hoped to isolate the Gallinazo style in the
Otherwise, we encountered no evidence for the Callej6n de Huaylas, but were unable to do so.
chronological position of Recuay which would The position of Recuay between an earlier
change the arguments presented.' It seems Chavin and a later Tiahuanaco-influenced pe-
obvious that Recuay and Mochica styles are riod seems probable, although its time relation-
related. Consequently, one is tempted to align ship to the white-on-red style is still a mystery.
them as more or less contemporaneous in a Much more work is needed on this important
chronological sequence. Recuay is also related style before a final statement can be made.
THE MIDDLE PERIODS
Remains of the Tiahuanaco-influenced Mid- modeled monkey, and other shapes. None of this
dle period are prominent in the Callej6n de redware is incised.
Huaylas. Available for discussion in this r6sum6 2. Blackware is fairly common, either plain or
are two categories of material: one, the exten- with incised decoration. It differs from the Late
sive collections excavated by us in the region Chimu materials of the coast. The shapes include
tapering collar jars, bulging collar jars, shallow
of Wilkawain, which form a definite unit and a plates, cups, goblets, double-bowls, and bowls with
distinct style in spite of the Tiahuanaco in- double spouts and rounded bridges.
fluence; two, Tiahuanaco-influenced material, 3. Two-color negativeware, black on an orange
not of the Wilkawain style, found in collections clay base, is represented by two shapes: a high collar
throughout the Callej6n. This second group jar and a wide, shallow plate
may or may not represent a period unit, since 4. Orangeware is common in ollas with two hori-
several styles are included. In both categories, zontal flat handles, goblets, plates, and shallow
however, the relationship is to the Coast Tia- sieves
huanaco periods and not to the site of Tia- 5. A painted ware has Tiahuanaco influence in the
huanaco in the highlands of Bolivia. There is no design. Color combinations include black, white-on-
red; black, white-on-orange; black, white, yellow-
evidence that Tiahuanaco influences reached on-red; black, gray, yellow-on-red; black, white,
the north highlands directly and not by way of yellow-on-orange. Shapes include collar jars, jars
the coast. with face collars, flare-rimmed bowls, straight-sided
In this r6sum6 we will discuss first the Wilka- cups, double-spout jars, some rare tripod vessels, and
wain excavated materials and then the mis- modeled puma bowls. The designs resemble the Coast
cellaneous Tiahuanaco-influenced materials in Epigonal styles such as the Nazca Y and Nieveria.
collections. We do not have good evidence for Clay spoons with a cursive design resembling the
chronological subdivisions of the Middle period. Cajamarca style are also associated. Non-ceramic
artifacts include copper pins, stone beads, flint and
THE WILKAWAIN MATERIALS obsidian points, wooden points with gold overlay,
Materials belonging to essentially one style and rare pieces of shell.
were found in our excavations around Wilka- Except for the lack of clear-cut Tiahuanaco
wain in deep, stone-lined and stone-covered design, this material compares most closely
tombs; in small, stone-lined, box-like graves, with Strong's Middle Ancon I. The differences
under natural boulders; in the refuse material are numerous, however, and the addition of
from one-room, above-ground houses; inside two-color negativeware at Wilkawain is a com-
four-room, above-ground houses; and in unpre- pletely new feature. Associations with various
pared graves. The materials themselves have types of architecture have been mentioned
been described in detail in the body of this briefly and will be reiterated shortly.
report and only a brief r6sum6 is needed. A
number of ceramic styles are associated in one OTHER TIAHUANACO-INFLUENCED MATERIALS
time period: A small collection in the Huaraz Museum is
1. Redware, usually polished, is represented by said to have come from the site ofPatay Katak.
constricted-collar jars, open-bowls, and plates, a Here large stone-lined tombs were found built
1 Bennett, 1939. up above ground and covered with dirt to form
low mounds. The ceramic shapes represented small dots. The shapes include double jars with
are variants of cups and goblets of good Coast tapered spouts and flat bridges; face-collar jars;
Tiahuanaco pattern, and the designs are equally globular vessels; face-collar flasks with two
typical, including such color combinations as small handles at the collar; double-spouted
black, white-on-red; black, yellow-on-red; black, vessels; goblets; a combination of a goblet and
white, gray-on-red; and black, yellow-on-red. a sea lion; and some open-bowls. We have no
Unfortunately, no further evidence on the asso- evidence as to the origin of this material, nor
ciations of this material is available. of any associations with architectural style. It is
Another type of Coast Tiahuanaco material quite possible that the Wilkawain style is the
is seen in local collections. This is of the style earliest in the Middle period, and that the oth.
commonly found in the lower Santa Valley. The ers follow, although we have no direct evidence
vessels are painted in a black, white-on-red de- for this conclusion.
sign with many checker patterns, Isein, and
CALLEJON HOUSE TYPES

A variety of house types and other construc- Type 7. A subterranean house with a large
tions were observed in our exploration in the central room, plus a series of side niches. The
Callej6n de Huaylas. Without going into elabo- chamber is entered by a special shaft and is
rate detail, a brief resum6 of these house types covered with great stone slabs. As many as
follows. seven niche-like rooms may open off this cen-
Type 1. A one-room stone house above tral chamber.
ground is built up of rough stone, with a roof Type 8. Two-floor subterranean house. Two
composed of slabs covered with dirt. A low door types were seen by us.
with a stone lintel is on one side. 8a. Two gallery-like rooms, one below the
Type 2. A two-room above-ground house. other and usually entered by the same shaft.
Two parallel, gallery-like rooms are roofed with 8b. Two floors, each with a central chamber
slabs on which is piled a high mound of dirt. and small niche-like rooms opening off them.
The two rooms are connected by an inside door- The general distribution of these house types in
way and one of them has an exit to the outside. the five major sites examined by us is given in
Type 3. A four-room, above-ground house is the following table. These eight major divisions
similar to Type 2, except that it has four rooms do not represent all of the architectural styles
instead of two. in the Callejon, but only those which we ex-
Type 4. Above-ground house with two floors. amined.
Each floor may have one or more rooms. In Period associations cannot be given for all of
most of those examined by us, two rooms were these types of construction. The possibility is
found on each floor. ever present that a house, particularly of the
Type 5. A three-floor above-ground house. In durable, subterranean type, was built in one
the two measured by us, each floor had a num- period and re-used in a subsequent period. The
ber of interior rooms. Because of the size of following notes referring to the various types
these constructions they have been described as are suggestive, although not completely con-
temples. clusive. A Type 1 house excavated by us at
Type 6. A one-room subterranean house. Wilkawain contained refuse of the Wilkawain-
Three subdivisions are found. Tiahuanaco style. Another Type 1 house which
6a. A large, hollowed-out room under a we opened at Ayapampa contained some poor
great, surface boulder. The walls are crudely material which also seemed to be related to the
made and the boulder serves as a roof. Wilkawain style. Tellol reports material of the
6b. A gallery-like subterranean room entered Tiahuanaco and Inca type in houses of Type 2
by a shaft at one end which opens in turn into at Katak. One Type 4 house was excavated by
an elongated, rectangular gallery. us at Wilkawain and contained refuse of the
6c. The specialized gallery type at Wilka- Wilkawain-Tiahuanaco style. Furthermore, this
wain, similar to 6b, but differing in detail. 1 Tello, 1930, 264.
house was stratigraphically over a gallery of great mixture of sherds from a number of styles,
Type 6b which contained Recuay material. The including some Recuay. Both the Santa-type
three-floor, above-ground temples of Type 5 Tiahuanaco ceramics and Recuay vessels are
belong to the Wilkawain-Tiahuanaco period, said to have been found at Copa Chica where
judging by their relation to other constructions the houses are all of Types 2 and 3, although in
in the region ofWilkawain and by the similarity our own survey of this site not even a sherd was
in architecture. Furthermore, stone puma heads discovered. Recuay materials have also come
on tenons were at one time inserted in the walls from the Katak region. The one house of Type
of one of the Wilkawain temples, which is again 8b which we excavated contained sherds of a
suggestive of Tiahuanaco. The Macedo Collec- Late Tiahuanaco style, but this does not prove
tion of Recuay ceramics in Berlin is said to have that the house type belongs exclusively to this
come from the Recuay region where practically period. Some of the house types may represent
all the houses are of Type 6b, although exami- the Inca period although confirming sherds are
nation of three houses revealed no Recuay not generally found around the ruins. Other
sherds. Type 6c is definitely Recuay period, construction types are mentioned in the litera-
according to our Wilkawain data. The Type 6a ture, but with no further evidence for identifi-
houses evacuated at Wilkawain contained a cations with chronological periods.
DISTRIBUTIOI0 OF CALLEJ6N HOuSE TYPES
Sites House Types
1 2 3 4 5 6a 6b 6c 7 8a 8b
Wilkawain . . . . . . 20 1 2 2 2 15
Copa Chica. . . . . . 33 2
Ayapampa .... 35 . .
Katak ... ....12

. 12 1 1
Recuay .......37 1 3
LATE PERIOD AND INCA

A great quantity of Late period material was Cuzco-Inca. It is noteworthy that there is a
found at Chavin, including construction work lack of distinctive Late material in local collec-
of rough stone facing walls with window- niches, tions in the Callej6n. The Inca are known to
stone-covered shallow canals, and stone-lined have inhabited this region, and some of the
pits. Late graves were encountered in the construction work at Copa Chica and at Copa
Chavin Pits 2 and 3. In the Chavin Pit 3, a Grande appears to be of a generalized Inca type
house was uncovered which had walls covered architecture, although identification is uncer-
with a clay-like plaster. The ceramic fragments tain. In fact, the absence of a well-defined Inca
for this Late period include a great mass of period or of any distinctive Late period in the
plainware, mostly from cooking vessels, as well Callej6n and at Chavin is surprising. It is, of
as some simple, painted ware. None of the styles course, possible that there is no Late period
is particularly outstanding. A pair of redware style as distinct as the Late Chimu on the north
goblets are the most characteristic Inca mate- coast, but rather that the whole era is marked
rial found. Various painted sherds at Chavin by a general decadence of the Tiahuanaco-
are similar to those found at San Jer6nimo near influenced Middle period. The results of our
Huaraz. These were described in the text as exploration do not warrant any definite state-
being of possible Inca origin, although the style ment on the status of the Late periods.
is definitely not that usually identified with the
TENTATIVE CHRONOLOGY FOR THE NORTH HIGHLANDS

Inca Period (San Jer6nimo style) on-red style is still uncertain. Recuay in turn is
Late Period (material uncertain) separated from the Wilkawain-Tiahuanaco by
Middle Periods stratigraphy at Wilkawain, but stylistically the
Coast Tiahuanaco (found in collections) two are far apart and one gets the impression
Wilkawain-Tiahuanaco that a good part of the intervening history is
Early Periods missing. One would expect some overlap in the
Recuay two styles, if one immediately followed the
White-on-red other. Likewise, no sound evidence exists for
Chavin the time relationship of the Wilkawain-Tia-
huanaco and the other Tiahuanaco-influenced
This seems to be the most logical arrange- material in local collections. Late periods are
ment of the periods on the basis ofour evidence. not clearly isolated, nor well defined, and even
However, it is not altogether satisfactory. the Inca identification is in some doubt, al-
While the relative positions of white-on-red and though the materials so designated certainly
Chavin are confirmed by stratigraphy at Cha- belong in a Late position. Far more detailed
yin and on the coast of Peru, the position of excavation is needed before the solution of the
Recuay in respect to Chavin and the white- chronology of the North Highlands is possible.
LITERATURE CITED
ANT6NEZ DE MAYOLO, SANTIAGO California Publ. Amer. Archaeol. and
1935. Las ruinas de Tinyash (Alto Marani6n). Ethnol., vol. 21, pp. 305-329.
Rev. de la Escuela Nac. de Artes y Oficios, GAYTON, A. H., AND KROEBER, A. L.
Chorrillos, no. 5. 1927. The Uhle pottery collections from Nazca.
BENNETT, WENDELL C. Univ. California Publ. Amer. Archaeol.
1934. Excavations at Tiahuanaco. Anthrop. and Ethnol., vol. 24, pp. 1-46.
Papers Amer. Mus. Nat. Hist., vol. 34, GONZALEZ DE LA ROSA, MANUEL
pp. 359-494. 1910. Les deux Tiahuanaco, leurs problWmes et
1936. Excavations in Bolivia. Ibid., vol. 35, pp. leur solution. Verhandl. des 16 Internatl.
329-507. Amerikanisten Kongr. Wien, 1908, pp.
1937. Chimu archaeology. Sci. Monthly, vol. 45, 405-428.
pp. 35-48. KINZL, HANS
1939. Archaeology of the north coast of Peru. 1935. Altindianische Siedlungsspuren im Um-
An account of exploration and excavation kreis der Cordillera Blanca. In Borchers,
in Viru and Lambayeque valleys. Anthrop. Philipp, Die Weisse Kordillere, Berlin, pp.
Papers Amer. Mus. Nat. Hist., vol. 37, 262-265.
pp. 1-153. KREUZ, M.
1942. Chavin stone carving. Yale Anthrop. Stud- 1934. Alte Kulturstatten in der Cordillera Blanca
ies, vol. 3. (Peru). Atlantis, pp. 599-600.
1943. The position of Chavin in Andean se- KROEBER, A. L.
quences. Proc. Amer. Phil. Soc. Phila- 1925a. The Uhle pottery collections from Moche.
delphia, vol. 86, pp. 323-327. Univ. California Publ. Amer. Archaeol.
BORCHERS, PHILIPP and Ethnol., vol. 21, pp. 191-234.
1935. Die Weisse Kordillere. Berlin. 1925b. The Uhle pottery collections from Supe.
CARMAND, A. C. DE Ibid., vol. 21, pp. 235-264.
1903. P6rou. Departement d'Ancachs. Lima. 1926a. Archaeological exploration in Peru. Part I.
GARRO, J. E. Ancient pottery from Trujillo. Field Mus.
1940. Notas arqueol6gicas. Una excursi6n al Nat. Hist. Anthrop. Mem., vol. 2, pp. 1-43.
norte del Peru. Rev. del Mus. Nac. Lima, 1926b. The Uhle pottery collections from Chan-
vol. 9, pp. 263-271. cay. Univ. California Publ. Amer. Ar-
GAYTON, A. H. chaeol. and Ethnol., vol. 21, pp. 265-
1927. The Uhle collections from Nieveria. Univ. 304.
1930. Archaeological explorations in Peru. Part SELER, E.
II. The northern coast. Field Mus. Nat. 1893. Peruanische Alterthiumer. Berlin.
Hist. Anthrop. Mem., vol. 2, pp. 47-116. SIEVERS, W.
LARCO HOYLE, RAFAEL 1914. Reise in Peru und Ecuador. Wiss. Verof-
1938. Los Mochicas. Lima, vol. 1. fentl. der Gesell. fur Erdkunde zu Leipzig,
1939. Los Mochicas. Lima, vol. 2. vol. 8, pp. i-xii, 1-411.
1941. Los Cupisniques. Lima. SORIANO INFANTE, AUOUSTO
LOTHROP, S. K. 1941. Monografia de Ancash: Nepefna (Provincia
1941. Gold ornaments of Chavin style from de Santa). Rev. del Mus. Nac. Lima, vol.
Chongoyape, Peru. Amer. Antiquity, vol. 10, pp. 263-277.
6, pp. 250-262. STRONG, WILLIAM DUNCAN
MARKHAM, CLEMENTS ROBERT 1925. The Uhle pottery collections from Ancon.
1910. A comparison of the ancient Peruvian Univ. California Publ. Amer. Archaeol.
carvings and the stones of Tiahuanacu and and Ethnol., vol. 21, pp. 135-190.
Chavin. Verhandl. des 16 Internatl. Ameri- TELLO, JULIO C.
kanisten Kongr. Wien, 1908, pp. 389-394. 1923. Wira Kocha. Inca, Lima, vol. 1, pp. 93-
MEANS, PHILIP AINSWORTH 320, 583-606.
1931. Ancient civilizations of the Andes. New 1929. Antiguo Peru. Lima.
York. 1930. Andean civilization: Some problems of
1934a. Des commentaires sur l'architecture an- Peruvian archaeology. Proc. 23d Inter-
cienne de la c8te P&ruvienne. Bull. Soc. natl. Congr. Americanists, New York,
Am6ricanistes Belgique, no. 14, pp. 75-110. 1928, pp. 259-290.
1934b. New clews to early American culture. New UHLE, MAX
York Times Mag., May 20, p. 12. 1903. Pachacamac. Philadelphia, Univ. Pennsyl-
MIDDENDORF, E. W. vania, Department of Archaeology.
1895. Peru. Vol. 3, Das Hochland von Peru. 1908. Ober die Friihkulturen in der Umgebung
Berlin. von Lima. Verhandl. des 16 Internatl.
MUELLE, JORGE C. Amerikanisten Kongr. Wien, 1908, pp.
1937. Filogenia de la Estela Raimondi. Rev. del 347-370.
Mus. Nac. Lima, vol. 6, pp. 135-150. 1913. Die Mulschelhiigel von Ancon, Peru. Proc.
MUELLE, JORGE CA, AND BLAs, CAMILIO 18th Internatl. Congr. Americanists, Lon-
1938. Muestrario de arte Peruano precolombino. don, 1912, pp. 22-45.
Rev. del Mus. Nac. Lima, vol. 7, pp. 163- 1920. Los principios de la civilizacion en la sierra
280. peruana. Bol. Acad. Nac. Hist. Quito,
RAIMONDI, ANTONIO vol.1, pp. 44-56.
1873. El departamento de Ancachs y sus riquezas VALCARCEL, Lufs E.
minerales. Lima. 1935. Litoesculturos y ceramica de Pukara. Rev.
ROOSEVELT, CORNELIUS VAN S. del Mus. Nac. Lima, vol. 4, pp. 25-28.
1935. Ancient civilizations of the Santa valley, WIENER, CHARLES
and Chavin. Geogr. Rev., vol. 25, pp. 21- 1880. P&ou et Bolivie. Paris.
42.
EXPLANATION OF ILLUSTRATIONS
PLATES e-g. Three statues from the wall at Casa Ghig-
1. View of three-story temple and deep stone-lined
lino, Aija
tombs at Wilkawain b-g. Photographs by Donald Collier
a. General view of three-story temple TEXT FIGURES
b. Detail of the doorway entering the second 1. Map of the Department of Ancash, Peru
floor Map showing the extent of the Rio Santa and
c, d. Views of deep stone-lined tombs the principal cities near which work was
2. Wilkawain-Recuay ceramic types done, as referred to in the text.
a. Incised brownware vessel from Site 7H-7 2. Groundplan of Wilkawain three-storied temple
(Museo Nacional, Lima) The plan is of the first floor and includes the
b. Blackware goblet from deep stone-lined tomb, supplementary terrace with three insert
Wilkawain (Museo Nacional, Lima) niches. The two rooms indicated by dotted
c. Blackware vessel with double spout from deep lines occur on the second and third floors
stone-lined tomb, Wilkawain (Museo Na- which correspond to the same groundplan.
cional, Lima) 3. Cross-section of Wilkawain temple
d. Recuay style conical-handled bowl from The figure shows the arrangement of rooms on
Shankaiyan (Museo Nacional, Lima) the three floors, supplementary terraces and
e, f. Tiahuanaco style vessels from Site 8H-1, niches on the first floor, and the cornice and
Wilkawain (Museo Nacional, Lima) eaves at the base of the roof.
g-j. Recuay style vessels in the American Mu- 4. Ceramic styles from above-ground house site,
seum of Natural History (B-9054, 8816, Wilkawain
8726, 8272) a. Redware double-spout, Site 6H-2 (41.1-3559)
3. Subterranean galleries, Wilkawain b. Fragmentary double-bowl, two-color negative
a. Covering slabs of gallery, Site lOH-1 design, Site 6H-2 (41.1-3557b)
b. Entrance to gallery, Site 7H-12 c. Fragment of thick black, white-on-redware,
4. Construction types in the Callej6n de Huaylas Site 6H-2 (41.1-3553)
a. Stone box tomb at Copa Chica d, e. Fragments of tripod bowls, Site 6H-2 (41.1-
b. Gateway of a wall at Copa Chica, House Type 3552a-c)
B f. Orangeware plate with annular base, Site
c. Houses on the hill at Copa Chica, House Type 6H-2 (41.1-3554a)
A g-k. Samples of decorated blackware, Site 6H-2
d. Two-story building at Gekosh near Recuay (41.1-3555a-e)
5. Subterranean houses at Katak 5. Groundplan and cross-section of deep stone-lined
a. Detail of doorway from the central chamber tombs
to a niche Tombs of Sites 7H-1 and 7H-3
b. Detail of wall construction 6. Blackware style of deep stone-lined tombs,
6. Chavin architecture Wilkawain
a. Detail of the Tello stairway at the Castillo a. Goblet-like vessel with slightly flaring rim,
b. Detail of dressed stone facing of Castillo Site 7H-3 (41.1-3630)
c. Stone cells of Pit 14. Photograph by Dr. b. Rimless bowl, Site 7H-1 (41.1-3581)
Augusto Soriano Infante c. Goblet with raised band (Museo Nacional,
d. Rough stone base wall of terrace Lima)
7. Views of Chavin d. Goblet with slightly bulging sides, Site 7H-3
a. General view of plastered clay house in Pit 3 (41.1-3631)
b. Detail of house doorway, Pit 3 e. Collar-jar, Site 7H-1 (41.1-3580)
c. View of river cut through the south plaza plat- f. Goblet with relief faces from the house, Site
form. Photograph by Donald Collier 7H-5A (41.1-3648)
d. View of the southwest corner of the Castillo. g. Double-bowl, Site 7H-3 (41.1-3628)
Photograph by Donald Collier h. Shallow plate, Site 7H-1 (41.1-3571)
8. Stone carving of Chavin and Aija i. Double-spout bowl, Site 7H-3 (41.1-3611)
a. Feline head carved in the round from the ter- 7. Redware style of deep stone-lined tombs, Wilka-
race of Chavfn wain
b, c. Carved heads with tenons for wall insertion a. Modeled monkey jar, Site 7H-1 (41.1-3570)
from Chavin b. Fragment of bowl which probably had a wide
d. Puma head with tenon from Aija flaring rim, Site 7H-3 (41.1-3600)
111
c.Open-bowl, Site 7H-1 (Museo Nacional, h. Open-bowl with rounded base, Site 7H-15
Lima) (41.1-3660f)
d. Constricted-collar jar, Site 7H-3 (41.1-3602) 13. Ceramic styles from subterranean house site
e. Goblet, orangeware, Site 7H-2 (41.1-3594) a. Annular-base bowl painted in special design
8. Negativeware plate from deep stone-lined tombs, style in reddish-brown on a white area plus
Wilkawain two orange bands, Site 9H-2 (Museo Na-
a. Top view, Site 7H-1 (41.1-3588) cional, Lima)
b. Side view b. Annular-base vase with relief face and special
9. Tiahuanaco style of deep stone-lined tombs, painted design in reddish-brown plus
Wilkawain orange on a white area, Site 9H-2 (41.1-
a. Collar of face-collar jar, Site 7H-3 (41.1- 3673a)
3608c) c. Orangeware conical handle suggesting Recuay
b. Face-collar jar, Site 7H-3 (41.1-3608b) ,type, Site 9H-2 (41.1-3676c)
c. Foot of modeled puma vessel, Site 7H-1 (41.1- d, e. Spoon fragments with cursive designs, Site
3586) 9H-2 (41.1-3672, 3671)
d. Face-collar jar, Site 7H-1 (41.1-3587) f. Loop handle spoon painted in black-on-orange
10. Tiahuanaco style of deep stone-lined tombs, lines, Site 9H-2 (41.1-3670)
Wilkawain g, h, j, k. Samples of white-on-redware, Site
a. Double-spout jar, Site 7H-3 (41.1-3612) 9H-2 (41.1-3675, 3676b, 3676d, 3676h)
b. Angular-bodied vessel, Site 7H-3 (41.1-3616) i. Thick orangeware open-bowl with modeled
c. Foot of modeled puma vessel, Site 7H-3 (41.1- animal head lugs, Site 9H-2 (41.1-3675)
3600) 14. Groundplan and cross-section of Recuay style
d. Constricted straight-collar jar with globular gallery, Wilkawain
body, Site 7H-3 (41.1-3604) Site 7H-12
e. Flask-shaped vessel, Site 7H-3 (41.1-3610) 15. Recuay style ceramics from galleries, Wilkawain
f. Brownware fragment with modeled face and a. Globular vessel of Recuay H-1 style painted
incised design of Tiahuanaco style from in white-on-red, Site 7H-10 (41.1-3656)
house at Site 7H-5A (41.1-3646) b. Globular vessel, grayware base with red daubs
11. Miscellaneous artifacts from deep stone-lined around the neck, Site 7H-10 (41.1-3655)
tombs, Wilkawain c. Annular-base open-bowl with red-on-orange
a. Copper pin with decorated animal head, Site design, Site 10H-1 (41.1-3680)
7H-3 (41.1-3613) d. Annular-base open-bowl with red-on-orange
b. Flint spearpoint, Site 7H-I (41.1-3577) cross-hatch design, Site 7H-13 (41.1-
c. Bowl of red-on-whiteware spoon, Site 7H-1 3657ab)
(41.1-3574) e. Conical handle bowl of Recuay B-1 style
d. Obsidian spearpoint, Site 7H-1 (41.1-3575) painted in white-on-red, Site 7H-8 (41.1-
e. Fragment of a high-collar jar with two-color 3654)
negative decoration, Site 7H-1 (41.1-3583) f. Dipper in two-color negativeware, Site 1OH-1
f. Fragment of an orangeware sieve, Site 7H-3 (41.1-3679)
(41.1-3626) 16. Groundplan and cross-section of house-gallery
12. White-on-redware from unlined grave, Wilka- stratification, Wilkawain
wain Sites 7H-5A and 5B showing the relationship
a. Open-bowl with straight sides decorated with ofgallery in Recuay style to two-story house
white angular scroll design on red base, of Wilkawain style
Site 7H-15 (41.1-3660h) 17. Recuay style artifacts from gallery at Shankaiyan
b. Flaring-sided open-bowl, Site 7H-15 (41.1- a. Fragment of annular-base open-bowl with
3660c) linear design, Site IH-A (41.1-3533a)
c. White-on-red double whistling jar (Museo b. Red-on-white open-bowl with band and dot
Nacional, Lima) design, Site 1H-A (41.1-3533e, f)
d. Tall brownware cylindrical neck jar, Site c. Red-on-white open-bowl painted with bands
7H-15 (41.1-36601) and wavy lines, Site 1H-A (41.1-3533)
e. Convex-sided open-bowl of redware, Site d. Fragment of irregular open-bowl, Site 1H-A
7H-15 (41.1-3660i) (41.1-3542)
f. Open-bowl with a flat base, Site 7H-15 (41.1- e. Bowl of dipper, handle missing, Site 1I-B
3660d) (41.1-3539)
g. Open-bowl with straight diverging sides, Site f. Red-on-orange design inside of bowl of dip-
7H-15 (41.1-3660e) per, Site 1H-B (41.1-3539)
g. Red-on-whiteware conical handle bowl of B4. Site Ch-15 (41.1-3781c)
Recuay B-2 style, Site 1H-B (41.1-3538) B-5. Site Ch-15 (41.1-3783d)
h. Orange open-bowl with two bulges on the B-6. Site Ch-15 (41.1-3781b)
body, Site 1H-A (41.1-3532) C-1. Site Ch-15 (41.1-3783c)
18. Recuay style artifacts from Huaraz galleries C-2. Site Ch-12 (41.1-3764)
a. Modeled figure of three-color negativeware C-3. Site Ch-11 (41.1-3762)
with complex geometric design, Site 1H-A C4. Site Ch-10 (41.1-3757b)
(41.1-3534c) D-1. Site Ch-16 (41.1-3781a)
b. Fragment of three-color negativeware, Site D-2. Site Ch-10 (41.1-3757a)
7H-14, Wilkawain (41.1-3659) D-3. Site Ch-15 (41.1-3781e)
c. Black-orange two-color negative piece from a D4. Site Ch-15 (41.1-3783g)
shallow open-bowl, Site 7H-5B (41.1-3651) E-1. Site Ch-10 (41.1-3754a)
d. Mouthpiece of orange clay trumpet, Site E-2. Site Ch-15 (41.1-3783)
1H-A (41.1-3682) E-3. Site Ch-15 (41.1-3781e)
e. Plain orangeware fragment with grooved A. Site Ch_3A (41.1-3724)
oval nubbins, Site IH-A (41.1-3536) B. Site Ch-8A (41.1-3740)
f. Modeled bird on a vessel decorated in three- C. Site Ch-3C (41.1-3725)
color negative design, Site 1H-A (41.1- 29. Chavin style decorated sherds from Chavfn
3535a) a. Incised circle with dot, Site Ch-15 (41.1-
g. Black-white-red painted figure from a vessel, 3781f)
Site 1H-A (41.1-3535b) b. Grooved circle and rectangular punch mark,
h. Black and red-on-orange clay spoon, Site Site Ch-15 (41.1-3780i)
1H-A (41.1-3531) c. Incised circles, Site Ch-15 (41.1-3784b)
i. Horn end of orange clay trumpet, Site iH-A d. S-shaped punch mark, Site Ch-15 (41.1-
(41.1-3682) 3780d)
j. Copper pin with nail-head top, Site 1H-B e. Circles between grooves, Site Ch-13 (41.1-
(41.1-3537) 3775a)
19. Groundplan and cross-section of house at Aya- f. Incised circles between grooves, Site Ch-15
pampa (41.1-3780c)
20. Groundplan and cross-section of house at Copa g. Incised circles, Site Ch-10 (41.1-3749)
Chica, Carhuaz h. Thumb-nail impression, Site Ch-15 (41.1-
House Type A 3780b)
21. Groundplan of house at Katak, Recuay i, j. Combinations of grooves and punch circles,
Site 1K-A. House type with central subterran- Site Ch-3 (41.1-3722c, 3722b)
ean chamber and several connected side k. Combination of groove and punch circles,
chambers Site Ch-15 (41.1-3787a)
22. Groundplan of house at Katak, Recuay 1. Feline incised design, Site Ch-10 (41.1-3746a)
23. Groundplan of house at Katak, Recuay m. Groove and dull gouge, Site Ch-2 (41.1-
24. Groundplan and cross-section of house at Katak, 3708ab)
Recuay, Site 4K-D n. Champlev6, Site Ch-10 (41.1-3752c)
25. Groundplan of ruins at Chavin o. Incised design, Site Ch-15 (41.1-3787b)
26. Groundplan and cross-section of Pit 3, Chavmn p. Simple stamped design, Site Ch-15 (41.1-
27. Groundplan of excavations of Pit 9 and Pits 3780h)
11-14, Chavin q. Circle and vertical groove, Site Ch-2 (41.1-
Shows the arrangement of six cell-like houses, 3700)
indicates those excavated by the expedition, r. Curvilinear relief, Site Ch-10 (41.1-3746b)
and locates the grave finds in Pit Ch-11 s. Grooved lines, Site Ch-15 (41.1-3787d)
28. Chavin style rim types and post-Chavln vessels
from Chavin 30. Chavin style decorated sherds from Chavfn
A-1. Site Ch-14 (41.1-3778a) a. Raised ridges with cross cuts, Site Ch-3 (41.1i
A-2. Site Ch-10 (41.1-3757) 3722)
A-3. Site Ch-1 (41.1-3693) b. Relief tabs with cross cuts, Site Ch-15 (41.1-
A-4. Site Ch-14 (41.1-3778b) 3780)
A-5. Site Ch-2 (41.1-3708) c. Rough scratch, Site Ch-10 (41.1-3752b)
B-1. Site Ch-15 (41.1-3783d) d. Incised dashes, Site Ch-15 (41.1-3780a)
B-2. Site Ch-9 (41.1-3742) e, f, g. Grooves and circle combinations, Sites
B-3. Site Ch-15 (41.1-3783e) Ch-8, Ch-15 (41.1-3734, 3784, 3787c)
h, k. Groove designs, Sites Ch-12, Ch-10 (41.1- e. Site Ch-16 (41.1-3792b)
3765, 3758a) f. Site 2H (41.1-3543a)
i. Complex groove design, Site Ch-10 (41.1- g. Site Ch-3 (41.1-3721)
3758a) h. Site Ch-2 (41.1-3703)
j. Complex groove design, Site Ch-14 (41.1- i. Site 2H (41.1-3543f)
3778b) j. Site 2H (41.1-3543e)
1. Suggested negative painted pieces, Site Ch-10 k. Site Ch-2 (41.1-3707)
(41.1-3755) 1. Site 2H (41.1-3543b)
m. Black and red positive painted with color m. Site 2H (41.1-3543g)
area separated by grooves, Site Ch-11 n. Site 2H (41.1-3543c)
(41.1-3762) o. Site 2H (41.1-3543d)
n. Raised band and incised grooves, Site Ch-14 p. Huaraz (41.1-3547n)
(41.1-3778a) q. Red-on-white annular-base bowl, Site Ch-5
o. Groove and punched hole, Site Ch-15 (41.1- (41.1-3734)
3790) r. White-on-red style vessel, Site Ch-13 (41.1-
p. Modeled feline face, Site Ch-16 (41.1-3791) 3770)
q. Grooved circles filled with red paint, Site 32. Recuay ceramic shapes
Ch-15 (41.1-3780e) 33. Stone carving styles in the Callej6n de Huaylas
r. Groove and gouge combination, Site Ch-13 (Huaraz Museum, Huaraz)
(41.1-3775b) a. Relief carved slab of Style C
31. Post-Chavfn ceramic styles from Chavin and San b. Relief carved slab of Style C
Jer6nimo c. Carved puma head with tenon for wall in-
a. Site Ch-16 (41.1-3792a) sertion, Style D
b. Site Ch-3 (41.1-3728) d. Elaborately decorated statue, a variant of
c. Site Ch-11 (41.1-3761) Style B
d. Site Ch-4 (41.1-3732) e. Simple cross-legged statue of Style A
PLATES 1-8
I
ANTHROPOLOGICAL PAPERS AMER. Mus. NAT. HIST. VOL. 39, PLATE I
14.
CU
0)
la
0
-0
4.'
0)
C)
CL)
CU
C)
04
C)
CL)
,IC;
0
C)
u
ANTHROPOLOGICAL PAPERS AMER. MuS. NAT. HIST. VOL. 39, PLATE 2
F
D
G H I J
Wilkawain-Recuay ceramic types

ANTHROPOLOGICAL PAPERS AMER. Mus. NAT. HIST. VOL. 39., PLATE 3
A
I
a
Subterranean galleries, Wilkawain
ANTHROPOLOGICAL PAPERS AMER. Mus. NAT. HIST. VOL. 39, PLATE 4
S4.1
Q -
I
C'
go
*. ...
e .v
.A
....
t ..
t -
is *
... 0
4 Cu
B
Subterranean houses at Katak
ANTHROPOLOGICAL PAPERS AMER. Mus. NAT. HIST. VOL. 39 PLATE 6
ct
ct
u
ANTHROPOLOGICAL PAPERS AMER. Mus. NAT. HIST. VOL.
V 39, PLATF. 7
0l
Ua)
u
A
r
F G
E
Stone carving of Chavin and Aija
ANTHROPOLOGICAL PAPERS
OF -THE
AMERICAN MUSEUM OF NATURAL HISTORY
Volume 39
North'Highlands
Huaylas and at Chavin de Huzntar. By Wendell-C. Bennett. Callejdn
PART 1. The of Peru. Excavations in the
8: plates, and 33 text figures. Price $1.75.
de
Pp. 1-116,
In preparation.
PART-2.

Highland: Excavations 'Huaylas..And Chavin DE Huantar

Uploaded by

Copyright:

Available Formats

Highland: Excavations 'Huaylas..And Chavin DE Huantar

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Highland: Excavations 'Huaylas..And Chavin DE Huantar

Uploaded by

Copyright:

Available Formats

TH11 HIGHLAND

EXCAVATIONS IN THE CALLEJON.

EXCAVATIONS IN THE CALLEJON DE HUAYLAS

FIG. 1. Map of the Department of Ancash, Peru.

THE WILKAWAIN SECTION

FIG. 2. Groundplan of Wilkawain three-storied temple.

FIG. 3. Cross-sectioz n off Wilkawain temple.

FIG. 4. Ceramic styles from above-ground house site, Wilkawain.

No. Per Cent Totals Per Cent

FIG. 5. Groundplan and cross-section of deep stone-lined tombs, Wilkawain.

refuse. A grouping by outstanding wares is One thousand twenty-nine fragments of ollas.

FIG. 7. Redware style of deep stone-lined tombs, Wilkawain.

FIG. 10. Tiahuanaco style of deep stone-lined tombs, Wilkawain.

Shape Fragments Complete aEstimate

Orangeware, 624, 1 complete .

Two-color, 1..................... (.12)

Clay Spoons long and 7 millimeters wide, and the other a

FIG. 14. Groundplan and cross-section of Recuay style gallery, Wilkawain.

FIG. 15. Recuay style ceramics from galleries, Wilkawain.

Recuay Descriptive Shape

(Recuay A-1) Plain open-bowl . . . . . . . . . 12 4.49 (2)

FIG. 16. Groundplan and cross-section of house-gallery stratification, Wilkawain.

OBSERVATIONS ON THE SHANKAIYAN GALLERY represented in the Wilkawain collection can be

Fic. 21. Groundplan of house at Katak, Recuay.

FIG. 22. Groundplan of house at Katak, Recuay.

1 .65 .10 .65 I

SIKETCH PLAN OF CHAVIN

ROUGH FACING WALL

FIe. 26. Groundplan and cross-section of Pit 3, Chavin.

FiG. 27. Groundplan of excavations of Pit 9 and Pits 11-14, Chavin.

quency sharp cut groove.

CHAVfN COMPARISONS AND DISTRIBUTION

examples. Although we examined many sites Tom6s La Rosa Sanchez Collection . . 51

A-I ='2 B-i

FIG. 32. Recuay ceramic shapes.

lMono- Two- Three- R Three- Two-

CALLEJON HOUSE TYPES

Katak ... ....12

LATE PERIOD AND INCA

TENTATIVE CHRONOLOGY FOR THE NORTH HIGHLANDS

Wilkawain-Recuay ceramic types

You might also like