ChapterPDF Available

Stone materials used for lintels and decorative elements of Khmer temples

Authors:

Figures

Content may be subject to copyright.
Metropolitan Museum
Studies in Art, Science,
and Technology
THE METROPOLITAN MUSEUM OF ART, NEW YORK
DISTRIBUTED BY YALE UNIVERSITY PRESS, NEW HAVEN AND LONDON
Volume 2 / 2014
Photographs of works in the Metropolitan Museum’s
collection are by e Photograph Studio, e Metro-
politan Museum of Art, unless otherwise noted.
Additional picture credits appear at the end of each
article.
Typeset in Warnock Pro
Printed on 150 gsm Galaxi Matte
Separations by Professional Graphics, Inc.,
Rockford, Illinois
Printed and bound by Verona Libri, Verona, Italy
Cover illustrations: Saint Michael, lower right panel of the
retable of Saint Anne with the Virgin and Child (1938.141d;
Figure 5, page 74): (left), detail, photograph by Andrew
Winslow; (right) infrared reectogram (Figure 19, page 80)
Frontispiece: Detail of Suzanne de Court, Mirror: Venus
Mourning the Dead Adonis (1975.1.1236; Figure 3a, page 159)
e Metropolitan Museum of Art endeavors to
respect copyright in a manner consistent with its
nonprot educational mission. If you believe any
material has been included in this publication
improperly, please contact the Editorial Department.
Copyright © 2014 by e Metropolitan Museum of
Art, New York
First printing
All rights reserved. No part of this publication may
be reproduced or transmitted in any form or by any
means, electronic or mechanical, including photo-
copying, recording, or any information storage and
retrieval system, without permission in writing from
the publishers.
e Metropolitan Museum of Art
1000 Fifth Avenue
New York, New York 10028
metmuseum.org
Distributed by
Yale University Press, New Haven and London
yalebooks.com/art
yalebooks.co.uk
Cataloguing-in-Publication Data is available from
the Library of Congress.
ISBN 978-1-58839-527-6 (e Metropolitan Museum
of Art)
ISBN 978-0-300-20439-1 (Yale University Press)
Editorial Board
Silvia A. Centeno
Research Scientist, Department of Scientic Research
Nora W. Kennedy
Sherman Fairchild Conservator of Photographs,
Department of Photographs
Marijn Manuels
Conservator, Sherman Fairchild Center for
Objects Conservation
Deborah Schorsch
Conservator, Sherman Fairchild Center for
Objects Conservation
Richard E. Stone
Conservator Emeritus, Sherman Fairchild Center for
Objects Conservation
Zhixin Jason Sun
Curator, Department of Asian Art
Mark T. Wypyski
Research Scientist, Department of Scientic Research
is publication is made possible by gifts from
Ludmila Schwarzenberg Bidwell and her late husband
Carl B. Hess. Further support was provided by the
members of the Visiting Committees of the Sherman
Fairchild Center for Objects Conservation, the
Sherman Fairchild Center for Paintings Conservation,
the Sherman Fairchild Center for Works on Paper
and Photograph Conservation, and the Department
of Scientic Research.
Published by e Metropolitan Museum of Art,
New York
Mark Polizzotti, Publisher and Editor in Chief
Gwen Roginsky, Associate Publisher and General
Manager of Publications
Peter Antony, Chief Production Manager
Michael Sittenfeld, Managing Editor
Robert Weisberg, Senior Project Manager
Edited by Ann Hofstra Grogg
Designed by Nancy Sylbert based on original design
by Tony Drobinski
Production by Paul Booth
Proofread by Sylvia Tidwell
Inquiries and other correspondence may be sent to:
metropolitan.museum.studies@metmuseum.org.
Contents
6 Director’s Foreword
7 Acknowledgments
11 Conservation in Context: e Examination and Treatment of Medieval
Polychrome Wood Sculpture in the United States
Lucretia Kargère and Michele D. Marincola
51 Stone Materials Used for Lintels and Decorative Elements of Khmer Temples
Federico Carò, Martin Polkinghorne, and Janet G. Douglas
69 e Retable of Saint Anne with the Virgin and Child at e Cloisters: Insights
into Fifteenth-Century Working Practices in the Kingdom of Aragon
Karen E. Thomas and Silvia A. Centeno
93 Petrography of Stone Used for Sculpture from the Buddhist Cave Temples
of Xiangtangshan Dating to the Northern Qi Dynasty
Janet G. Douglas and John T. Haynes
115 Lebanese Mountain Figures: Advances in the Study of Levantine
Anthropomorphic Metal Statuary
Deborah Schorsch
Research Notes
157 Silver Surface Decoration on Limoges Painted Enamels
Gregory H. Bailey and Robyn E. Hodgkins
167 A Closer Look at Red Pictorialist Photographs by René Le Bègue and
Robert Demachy
Anna Vila, Silvia A. Centeno, and Nora W. Kennedy
173 Tracing the Origins of Japanese Export Lacquer on an Eighteenth-Century
French Side Table
Christina Hagelskamp
181 A Technical Examination of Lake Como by omas Moran
Sarah Kleiner
189 A Technical Examination of the Lion Helmet
Edward A. Hunter
197 An Investigation of Tool Marks on Ancient Egyptian Hard Stone Sculpture:
Preliminary Report
Anna Serotta
50 ◆ ◆ 51
    , ,   
Stone Materials Used for Lintels and Decorative
Elements of Khmer Temples
Federico Carò, Martin Polkinghorne, and Janet G. Douglas

Intricately carved lintels occupy a privileged
and symbolic position in Khmer temples. eir
production was commissioned to highly special-
ized craftsmen with specic material and carv-
ing traditions. e stone of seven decorative
lintels in e Metropolitan Museum of Art has
been characterized by means of petrographic
and textural analyses, together with the stone
oftwenty-nine other lintels and ornamental
elements dating from the seventh to the thir-
teenth century now in the National Museum of
Cambodia, Phnom Penh; the Musée National
des Arts Asiatiques Guimet, Paris; and the
Arthur M. Sackler Gallery, Smithsonian
Institution, Washington, D.C. Results suggest
that particular stone types were selected by
Khmer carvers for certain elements in the temple
structure. Whereas blocks of sandstone from
the Terrain Rouge Formation of Cambodia
were used to build and clad the Angkor temples
and for decorative lintels and other ornaments,
quartz-rich sandstone was reserved exclusively
for these decorated elements. is sandstone
may originate from the Grès Supérieures
Formation, a sedimentary sequence that
extends into ailand, Laos, and Vietnam.
The Metropolitan Museum of Art holds
several stone decorative lintels produced
during the time of the Khmer Empire (sixth
fteenth centuries). ese objects belong to a
special variety of Khmer sculptural arts that
dier in several ways from free standing sculp-
ture. A true lintel is the upper horizontal com-
ponent in the framework of a doorway, which is
usually formed of four independent stone blocks
held together by mortise and tenon ttings. A
decorative lintel, on the contrary, is rarely load
bearing and is located above the true lintel, sup-
ported by two similarly decorated colonnettes.
Decorative lintels occupy privileged positions
above the entryways of Khmer temples, watch-
ing over all those who pass through from the
secular to the divine realms. In this role they
dene an important boundary. e forms and
icon ographies of the decorative lintels sought
to maintain the temple in a permanent state of
festival. Often they represent ephemeral deco-
rations of garlands and rinceaux that gave the
impression of a building alive with celebrations.¹
e lintels are thought to have been carved in
situ by specialized craftsmen employing a par-
ticularly time-consuming and expensive pro-
cess, following established artistic conventions.²
Carò, Polkinghorne, and Douglas
52 ◆ ◆ 53
Accession
Number Collection Description Provenience Date Q
(%)
F
(%)
L
(%)
Mean
(mm)
Median
(mm)
Sorting
(σ)
Stone
Typ e
. e Metropolitan
Museum of Art
Lintel ailand, exact
provenience unknown
st half th c. . . . . . .
.. e Metropolitan
Museum of Art
Lintel
(see Figure)
Cambodia or Vietnam,
exact provenience unknown
th c. . . . . . .
. e Metropolitan
Museum of Art
Lintel Cambodia or ailand,
exact provenience unknown
st half th c. . . . . . .
. e Metropolit an
Museum of Art
Lintel
(see Figure)
Cambodia or ailand,
exact provenience unknown
st quarter th c. . . . . . . 
Ka National Museum
of Cambodia
Lintel Wat Ang Kh th c. . . . . . .
Ka National Museum
of Cambodia
Lintel Exact provenience unknown th c. . . . . . .
Ka National Museum
of Cambodia
Lintel
(see Figure)
Wat Preah eat th c. . . . . . .
MG Musée Guimet Lintel Phnom Da beginning th c. . . . . . .
MG Musée Guimet Lintel Wa t Bas et end th –
beginning th c.
. . . . . .
MG Musée Guimet Colonnette Phnom Da beginning  th c. . . . . . .
S. Arthur M. Sackler
Gallery
Lintel Cambodia or ailand,
exact provenience unknown
mid-th c. . . . . . .
.. e Metropolitan
Museum of Art
Lintel
(see Figure)
Kôk Sla Ket mid-th c. . . . . . .
. e Metropolitan
Museum of Art
Lintel Cambodia or ailand,
exact provenience unknown
mid-th c. . . . . . .
. e Metropolitan
Museum of Art
Lintel
(see Figure)
Cambodia, exact
provenience unknown
nd half th c. . . . . . .
.. e Metropolitan
Museum of Art
Colonnette Cambodia, exact
provenience unknown
mid-th c. . . . . . .
. e Metropolitan
Museum of Art
Antex C ambodia, exact
provenience unknown
rd quarter th c. . . . . . .
MG  Musée Guimet Lintel
(see Figure)
Wat Kralanh mid-th c. . . . . . .
MG Musée Guimet Pediment Banteay Srei late th c. . . . . . .
Ka National Museum
of Cambodia
Lintel Sambor Prei Kuk th c. . . . . . . 
Ka National Museum
of Cambodia
Lintel Preah Kô th c. . . . . . .
Ka National Museum
of Cambodia
Lintel
(see Figure)
Kompong Cham Province,
exact provenience unknown
mid-th c. . . . . . .
Ka National Museum
of Cambodia
Lintel Preah Vihear Province, exact
provenience unknown
late th c. . . . . . .
Ka National Museum
of Cambodia
Lintel Preah Khan of Kompong
Svay / Bakan
end th –
beginning th c.
. . . . . .
MG Musée Guimet Frieze Exact provenience unknown last quarter th
beginning th c.
. . . . . . b
MG Musé e Guimet Pilaster Beng Mealea mid-th c. . . . . . .
MG Musée Guimet Colonnette B eng Mealea mid-th c. . . . . . . 
MG Musé e Guimet Pediment Preah Khan of Kompong
Svay / Bakan
rd quarter th c. . . . . . .
MG Musé e Guimet Lintel Bayon end th –
beginning th c.
. . . . . .
MG Mus ée Guimet L intel Kapilapura last quarter th c. . . . . . .
MG Musée Guimet Lintel Sambor Prei Kuk st half th c. . . . . . .
MG Musé e Guimet Lintel Prasat Prei Khmeng nd half th c. . . . . . . b
MG Musée Guimet Lintel Prasat Koki e arly th c. . . . . . .
MG Musée Guimet Colonnette Prasat Prei Khmeng nd half th c. . . . . . . b
MG Musée Guimet Colonnette Kulen early th c. . . . . . .
MG Musée Guimet Pediment Prasat Sok Kraup end th –
beginning th c.
. . . . . .
MG Musée Guimet Colonnette Wat Choeng Ek st half th c. . . . . . .
Table  Key Compositional and Textural Parameters of the Lintels and Carved Architectural Elements. e samples are listed in order
ofaccession number and stone type (Type, white to light brown quartz arenite; Type, reddish quartz arenite; Type and b, greenish-
gray feldspathic arenite). Q = quartz; F = feldspar; L = rock fragments. Sorting (σ): see note .
Stone Materials Used for Lintels and Decorative Elements of Khmer Temples
52 ◆ ◆ 53
In support of this hypothesis, previous tech-
nical studies suggest that specic stone materials
were purposefully chosen for the production of
these architectural elements, which are char-
acterized by deeply carved, intricate details.³
However, given the complex history of Khmer
temples and the vast production of decorative
elements, it is dicult to draw any conclusion
about patterns of stone choice and usage
without considering a comprehensive and rep-
resentative database of provenienced objects.
Toward this end, this study aims to character-
ize by means of scientic analyses the stone
materials used in the production of a selection
of decorative lintels and other architectural
ornamental elements, and ultimately to help
unveil connections between Khmer artistic
production and the geological sources of con-
struction materials.
e twenty-four lintels and twelve decora-
tive elements (colonnettes, friezes, and pedi-
ments, as well as one antix and one pilaster)
examined for this study are in the collections
of e Metropolitan Museum of Art, New
York; the National Museum of Cambodia,
Phnom Penh; the Musée National des Arts Asi-
atiques Guimet, Paris; and the ArthurM.
Sackler Gallery, Smithsonian Institution,
Washington, D.C. ey were produced over a
range of dates consistent with the pre-Angkor
(sixth eighth centuries) and Angkor (ninth f-
teenth centuries) periods. Although rarely dated
on the basis of epigraphic evidence, decorative
lintels are particularly respected chronological
markers for Khmer art historians and
archaeologists.
e studied lintels and decorative elements
originate from various locations in Cambodia,
ailand, and Vietnam. ose in the Musée
Guimet and National Museum of Cambodia
come with certain provenience, and most can
be associated with specic sites (Table). From
a stylistic perspective alone it is dicult to be
precise about the origins of the lintels and dec-
orative elements in e Metropolitan Museum
of Art and the Arthur M. Sackler Gallery, but
they are logically associated with temple sites
north of the Tonlé Sap (Great Lake) and may
even have come from what today is ailand.
 
 
Five lintels from e Metropolitan Museum of
Art have been selected for discussion, as they
illustrate the iconography and decorative
motifs common to classical Khmer sculptures.
ey range in dates from the seventh century
to the early eleventh century.
e lintel with a mask of Kāla (..;
Figure) is of exceptional interest; its motifs
indicate that it likely dates to the seventh cen-
tury, making it the earliest included in this
st udy. Its central and dominating motif is
theKāla or Kīrtimukha (face of glory), an
extremely common central motif on decorative
lintels thought to represent an aspect of Shiva
Figure  Lintel with a
mask of Kāla. Cambodia or
Vietnam, exact provenience
unknown, th century.
White to light brown quartz
arenite (Type sandstone),
H..cm ( ½in.),
L..cm (in.),
D..cm (in.). e
Metropolitan Museum of
Art, Gift of Margery and
Harry Kahn, 
(..)
Carò, Polkinghorne, and Douglas
54 ◆ ◆ 55
as time, who devours himself and destroys all
things. Kāla in this role is known by the Khmer
as Rahu. Rahu is associated with funerary and
cremation rites and also possesses a bivalent
nature as the rst step toward new life. In this
manifestation, Rahu is regarded as the demon
of the eclipse, causing the darkness to make
new light appear.
In , Victor Goloubew of the École
Française d’Extrême-Orient (EFEO) proposed
that the Kāla motif was adopted by the Khmer
from Java.¹ Subsequent scholars have com-
mented on this possible connection¹¹ but point
to many examples of Khmer architectural
decoration that predate those from Java.
Mireille Bénisti argues that the indigenous
inuence from early Khmer art on the appear-
ance and composition of the Kāla has been
underestimated at the willing acceptance of a
Javanese inspiration and cites seventeen exam-
ples that predate the Javanese examples,¹² par-
ticularly those at Borobudur mentioned by
Gilberte de Coral-Rémusat.¹³ e Metropolitan
Museum’s Kāla lintel corroborates the theory
that the motif as it developed in both Cambodia
and Java most likely derives from a common
source of inuence in India.¹
e Kāla motif on the Metropolitan Muse-
um’s lintel is distinctive in that it dominates
the entire surface. Although dierent in repre-
sentation, this Kāla is analogous in size to that
on a lintel from Sala Prambei Lveng, in ala
Borivat, Stung Treng Province.¹ Between the
Kāla’s eyes is a euron emblem common to
central motifs in other pre-Angkorian lintels.¹
e conscious transformation of motifs from
simple to complex, or vice versa, may be dened
as a form of stylization and was part of the
Figure  Lintel with
carved gures. Kôk Sla
Ket, mid-th century.
Reddish quartz arenite
(Type sandstone),
H..cm ( ¼ in.),
L..cm (in.). e
Metropolitan Museum of
Art, Fletcher Fund, 
(..)
Figure  Lintel depicting
Indra riding Airāvata.
Kompong Cham Province,
exact provenience
unknown, mid-th
century. Greenish-gray
feldspathic arenite (Type
sandstone), H..cm
( ¼in.), L.. cm
( ⅞in.), D..cm
( ⅜in.). National
Museum of Cambodia
(Ka)
Stone Materials Used for Lintels and Decorative Elements of Khmer Temples
54 ◆ ◆ 55
repertoire of classical Khmer artists.¹ e
interplay between the Kāla and euron sug-
gests that the motifs were interchangeable and
perhaps share a similar or double meaning.
Below the Kāla mask are series of motifs com-
mon to pre-Angkorian lintels, including colon-
nette capitals with euron medallions, lotuses
of the species Nelumbo nucifera, and hanging
pendants that establish a seventh-century date
for this work.¹
e Khmer collection of e Metropolitan
Museum of Art includes three examples of
decorative lintels dating to the second half of
the tenth century. One of these (..;
Figure) comes from the Angkorian site of
Kôk Sla Ket, located  kilometers south of the
West Baray. e sculpture is dated stylistically
to the mid-tenth century, which is consistent
with an inscription found at the site.¹ Accord-
ing to photographs taken by the EFEO in ,
the lintel is from the east facade of the north
tower.² After being removed to the Conserva-
tion dAngkor,²¹ the lintel was acquired by e
Metropolitan Museum of Art in February 
with six additional Khmer sculptures, includ-
ing a bust of Hevajra (..) found outside
the Gate of the Dead atAngkor om.²²
is lintel’s central motif of a caryatid lion
anked by additional lions issuing foliage
branches from their mouths is reminiscent of
the early tenth century, and it is possible that
the lintel was carved by artists who had also
worked at the early tenth-century capital
northeast of Angkor.²³ e top frieze captures
seven r
.s
.i or ascetics in prayer. Lintels from the
mid-tenth century display a consistency in the
division of decoration on their surface. Arched
foliage branches terminating in leafy curls, a
regular and symmetrical number of oshoots
above and below the branches, and the distinc-
tive frieze are components also observed on a
lintel from the National Museum of Cambodia
(Ka; Figure).
A second lintel (.; Figure) from the
second half of the tenth century in the Metro-
politan Museum is in the style of Banteay Srei.
e ne carving and complexity of decorative
motifs from this period have led numerous
scholars to consider the period between approx-
imately  and  as among the pinnacles of
Khmer art.² e central motif depicts Yama,
the god of death, atop his Vāhana, the bualo,
holding a danda, or club. In this context Yama
is designated as a Lokapāla or Dikpāla, both
directional deities. is identication conrms
that the lintel originally faced south in the
conguration of the temple tower.
A third tenth-century lintel in the Metro-
politan Museum collection (.) depicts a
Kāla face ridden by Indra holding a vajra
(thunderbolt).² Of particular interest are the
motifs of celestial beings praying in eurons
forming the lintel frieze. Each of these small
representations depicts the torso of a crowned
gure with hands together in a gesture of pray-
ing, positioned within a triangular vegetal
Figure  Lintel with Yama
on bualo. Cambodia,
exact provenience
unknown, second half of
the th century. Reddish
quartz arenite (Type
sandstone), H..cm
( ½in.), L ..cm
(in.). e Metropolitan
Museum of Art, Gift of
e Kronos Collections,
 (.)
Carò, Polkinghorne, and Douglas
56 ◆ ◆ 57
frame. Nearly identical celestial beings praying
in eurons make their rst appearance in lin-
tels from Koh Ker, the Khmer capital from
about  to , approximately  kilometers
northeast of Angkor. ese motifs recur for
the next fty years, continuing even after the
court returned to Angkor (Yaśodharapura)
from Koh Ker. e same artistic workshop that
contributed to the foundations of the king
JayavarmanIV at Koh Ker followed the new
sovereign back to Angkor, suggesting that
despite political change or instability, the arti-
sans shadowed the center of power and were
not tied to a particular administration.²
e last decorative lintel at e Metropoli-
tan Museum of Art under consideration here
(.; Figure) dates to the rst decades of
the eleventh century and can be associated
with lintels from the Khleangs and the Royal
Palace of Angkor om. e principal motif is
again the Kāla, which issues from its mouth
two garlands that terminate in curls of foliage.
e monster supports a seated brahmanic
divinity, perhaps the most common motif on
decorative lintels but whose details are often
confused and indistinct and therefore dicult
to identify. From their seated position known
as rājalilāsansa (Sanskrit, meaning “pose of
royal ease”), the gures are usually associated
with Dikpāla and are identied with a mon-
arch, as can be seen on the lintel in Figure.
Iconographically, this lintel is similar to a lintel
from the Musée Guimet (MG; Figure),
especially in the Kāla face with extended arms
and the frieze of pendants andeurons. Varia-
tion of the central motif (Umāheśvara) and
Figure  Lintel with a
mask of Kāla and Dikpāla.
Cambodia or ailand,
exact provenience
unknown, rst quarter of
the th century. White to
light brown quartz arenite
(Type sandstone),
H..cm (in.),
L..cm (in.). e
Metropolitan Museum of
Art, Gift of R.Hateld
Ellsworth, in honor of
Florence and Herbert
Irving,  (.)
Figure  Lintel depicting
Umāheśvara on Kāla. Wat
Kralanh, mid-th century.
Reddish quartz arenite
(Type sandstone),
H..cm ( ⅛in.),
L..cm ( ½in.),
D..cm ( ½in.).
Musée National des Arts
Asiatiques – Guimet
(MG)
Stone Materials Used for Lintels and Decorative Elements of Khmer Temples
56 ◆ ◆ 57
additional depth of carving suggest an anity
to the Baphuon style and a slightly later date in
the mid-eleventh century.
   

e addition of petrographic analysis to the
suite of interpretive tools applied to Khmer
sandstone sculpture has considerable potential
to answer questions raised by art-historical
studies. Petrographic analysis requires that a
small fragment of stone be removed for thin-
section preparation, that is, mounted on a glass
slide and polished to a thickness of  microns.
For this study fragments were removed from
already damaged areas, generally located at the
bottom edges of the lintels.
For each thin section, the nature, dimen-
sions, and abundance of the various constitu-
ent grains were assessed using a polarized light
microscope.² e number of grains analyzed
varied with each sample according to its size
and also depended on grain size and sorting,²
but usually at least  points were counted.
e classication scheme adopted is that pro-
posed by Paolo Gazzi and by WilliamR. Dick-
inson, which uses the relative abundance of
quartz (Q), feldspar (F), and rock fragments (L)
to assign specic names to dierent species of
sandstone.² Further information regarding the
texture of the rock was recorded, including the
size, sorting, shape, and arrangement of the
constituent elements. With few exceptions, the
objects studied were carved from three main
types of sandstone, designated herein as Types
, , and  (Figures, ; see also Table).
    

Type sandstone varies in color from white to
light brown, depending on the amount of iron
oxides present, and has a crisp appearance.
Using petrography, this sandstone is classied
as a ne- to medium-grained, moderately well-
sorted to well-sorted quartz arenite. Grains are
subrounded to subangular in shape and are
cemented by abundant authigenic quartz.
Kaolinite cement arranged in coarse stacks of
pseudohexagonal plates is also abundant and
postdates the quartz overgrowth. e shapes
of the original grains are often revealed by a
thin hematite coating on their surfaces.
Monocrystalline undulose quartz grains are
the most abundant constituent. e quartz
often appears cloudy from inclusions. Poly-
crystalline quartz is subordinate and mostly
strained and foliated, with sutured contacts.
Also present but rare are grains of crypto-
crystalline quartz. Feldspar content is gener-
ally low but can reach  percent of the total
grains. Feldspars are often weatheredor
kaolinized.
Rock fragments do not exceed  percent
ofthe framework grains. Among them, most
characteristic are aphanitic volcanic rock
fragments, low-grade metamorphic rocks
Figure  Micrographs of
the three main identied
lithotypes (crossed Nicols):
(a) medium-grained, white
to light brown quartz
arenite (Type sandstone);
(b)very ne-grained, red-
dish quartz arenite (Type
sandstone); (c)very ne-
to ne-grained, greenish-
gray feldspathic arenite
(Type sandstone)
ab c
Carò, Polkinghorne, and Douglas
58 ◆ ◆ 59
(phyllite), and fragments of argillaceous mud-
stone. Some of the sedimentary rock frag-
ments, such as the argillaceous component,
have been squeezed to form pseudomatrix.
Heavy minerals³ are rare and include grains
ofrutile, tourmaline, zircon, and epidote. A
ne-grained matrix, often pigmented by the
abundant iron oxides, is present in variable
amounts between the framework grains.
When the sandstone is extremely quartz-
rich and devoid of iron oxides, opaque miner-
als, and matrix, it is white to light gray, as in
the lintel from the National Museum of Cam-
bodia depicting Indra riding Airāvata (Ka;
Figure). With the increase of the above-
mentioned accessory minerals, the sandstone
attains a reddish-brown color, like the Metro-
politan Museum’s lintel with a mask of Kāla
and Dikpāla (.; see Figure) and the
Musée Guimet’s lintel depicting scenes from
the Rāmāyana (MG).
  
Type  sandstone has a characteristic reddish
color. Petrographically, this rock is classied as
very ne-grained, moderately well-sorted quartz
arenite. Grains are subrounded to rounded in
shape and are cemented by ne-grained
kaolinite and authigenic quartz in variable pro-
portions. Characteristic is the abundant hema-
tite cementation that occurs as grain-coating
and pore-lling and accounts for the color of
this sandstone.
Undulose and nonundulose monocrystal-
line quartz grains dominate the framework,
while feldspar is subordinate and usually
weathered. Polycrystalline strained quartz and
cryptocrystalline quartz are also present. Rock
fragments are sparse and include aphanitic
volcanics, phyllite, siltstone, and mudstone.
Fragments of igneous rock with micrographic
texture are present but very rare. Accessory
muscovite is often bent between the grains.
Heavy minerals are rare and include ilmenite,
rutile, tourmaline, and zircon. A ne, brown
matrix of clay-size particles rich in iron oxides
is occasionally deposited between the grains,
possibly as a result of mechanical compaction
of preexisting rock fragments.
-  
Type  sandstone is a greenish-gray feldspathic
arenite; it constitutes almost half of the sam-
ples. is sandstone is composed of very ne
to ne, moderately well-sorted to well-sorted,
subangular to rounded grains, cemented
Figure  Petrographic
and textural characteris-
tics of the lintels studied:
(a)framework grain
composition grouped
according to the relative
abundance of quartz (Q),
feldspar (F), and rock
fragments (L) following
Gazzi  and Dickinson
; (b)classication
ofthe sandstone types
according to sorting and
grain size of the constitu-
ent grains following
Folkand Ward .
WS= wellsorted;
MWS= moderately
wellsorted; MS=
moderately sorted
a b
Stone Materials Used for Lintels and Decorative Elements of Khmer Temples
58 ◆ ◆ 59
predominantly by chlorite. Calcite and authi-
genic quartz and feldspar can be also present
in variable amounts. e degree of compaction
and cementation varies.
Mono-, poly- and microcrystalline quartz
grains, both undulose and nonundulose, are
the most abundant framework constituents.
Feldspar makes up about  percent of the
framework grains and is dominated by plagio-
clase, with minor alkali feldspar. Both fresh
and altered feldspars are present.
e abundance of rock fragments is typically
about percent and only rarely exceeds per-
cent. Characteristic lithic fragments are volcanic
(andesite), metamorphic (phyllite, quartzite,
micaceous schist), and sedimentary (argillite,
shale, siltstone) in origin. e heavy minerals
assemblage consists mostly of hematite, mag-
netite, ilmenite, rutile, titanite, garnet, epidote,
zircon, apatite, monazite, and tourmaline, in
varying proportions. is assemblage reects a
mixed provenience with strong metamorphic
and felsic igneous inuence, as other authors
have pointed out.³¹ Within the Type  group it
is possible to distinguish a small subset (Type
b) of very ne-grained, quartz-rich feldspathic
arenite characterized by abundant biotite.
 
is study of the mineral composition and
texture of the sandstones used for lintels and
other decorative architectural elements helps
to correlate the lithotypes identied to specic
geological formations³² and, in turn, to more
narrowly dened geographic regions within
the territory of the Khmer Empire.
All three lithotypes have strong anities
with sedimentary rocks constituting part of the
Khorat Group,³³ a nonmarine Mesozoic sedi-
mentary sequence that outcrops in the Khorat
Plateau of ailand and extends into Cambodia
and Laos. In Cambodia the sandstones are well
represented by the Dangrek Range in the north,
where its steep anks form a natural border
with ailand; in the central region by Phnom
Kulen and Phnom Tabeng (in Khmer, phnom
means “mountain” or “hill”), in Siem Reap and
Preah Vihear Provinces respectively; and in the
southwest by the Cardamom Mountains
(Figure). Numerous scattered outcrops of
sandstones and conglomerates with similar
characteristics are exposed in subhorizontal
beds in isolated hills of low altitude and limited
extension in central and eastern Cambodia. is
sequence was described in detail by various
Figure  Lintel depicting
Indra riding Airāvata. Wat
Preah eat, th century.
White to light brown
quartz arenite (Type
sandstone), H..cm
( ¾in.), L..cm
( ⅝in.). National
Museum of Cambodia
(Ka)
Carò, Polkinghorne, and Douglas
60 ◆ ◆ 61
geologists in the late s during early attempts
to complete a systematic geological mapping of
Cambodia. e overall stratigraphy has been
variously interpreted, but the most often cited
Mesozoic formations of the Khorat Group in
ailand from the Late Triassic to the Middle
Cretaceous are Nam Phong, Phu Kradung,
Phra Wihan, Sao Khua, Phu Phan, and Khok
Kruat (Figure).³ In Cambodia, the last four
formations of the Khorat Group are known as
Figure  Simplied
geological map of
Cambodia based on
United Nations , with
locations of major Khmer
archaeological sites
Stone Materials Used for Lintels and Decorative Elements of Khmer Temples
60 ◆ ◆ 61
lintel production cannot be identied with pre-
cision. However, geological and petrographic
data conne the source of these litho types to
the upper members of the Khorat Group, cor-
responding to the Grès Supérieures Formation
of Cambodia. e medium-grained, white to
light brown quartz arenite (Type sandstone)
may originate from the Phra Wihan and the
Phu Phan Formations. ese formations are
known to include white to light-bu, very ne-
to coarse-grained sandstones, tightly cemented
by authigenic quartz and characterized by
extensive kaolinitization.³ Furthermore, these
sandstones are often iron stained and dier-
ently colored, as can be seen in modern quar-
ries in Banteay Meanchey Province, as well as
in the ancient quarries of Ban Khao Luk
Chang, Ta Phraya, and Preah Vihear, scattered
along the Dangrek Range.
e mineralogy of Phra Wihan sandstones
is dominated by monocrystalline, mostly
Figure  Mesozoic stratig-
raphy of Cambodia (based
on Sotham ) correlated
to the Khorat Group forma-
tions of ailand. Two pub-
lished interpretations of the
stratigraphy inailand are
presented (Workman ;
DMR ), with the most
representative lithologies
for each formation. Modi-
ed after Racey etal. ,
p.. A wavy line indicates
an unconformity; a dashed
line indicates an uncertain
sedimentary boundary.
the Grès Supérieures Formation, a designation
introduced by French geologists to indicate the
subhorizontal continental sediments occupy-
ing the highlands of western and northern
Cambodia.³ ese rocks, mostly quartz-rich
sandstones and conglomerates, are considered
separately from the Lower Middle Jurassic
uvial and lacustrine sequence known as the
Terrain Rouge Formation, roughly correspond-
ing to the Phu Kradung Formation, and from
the Indosinian Triassic sequence that occupies
the lowermost unit of the Khorat Group and is
known in ailand as the Nam Phong Forma-
tion. In this framework, the well-known quar-
ries at the foot of Phnom Kulen are located at
the upper portion of the Terrain Rouge Forma-
tion, close to the contact with the Grès Supéri-
eures Formation, which is in turn quarried on
top of Phnom Kulen, close to Prasat Rong Chen.³
At the present time, a geological provenience
of the quartz arenite (Types  and ) used for
Carò, Polkinghorne, and Douglas
62 ◆ ◆ 63
undulose quartz, while feldspars and rock frag-
ments are sparse. Heavy minerals, including
rutile, tourmaline, and zircon, are also rare.
Similar medium- to coarse-grained, quartz-
rich sandstone that is light in color is also
abundant in the Phu Phan Formation and is
dicult to distinguish from Phra Wihan
lithotypes even at the microscopic scale.
e very ne-grained, hematite-rich, reddish
sandstone (Type ) identied in this study could
originate from either the Phra Wihan or the
Sao Khua Formations. Similar lithotypes domi-
nated by quartz grains, scarce feldspar, and
rock fragments cemented by abundant hematite
and kaolinite occur in both these formations;³
they could potentially be found on top of
Phnom Kulen, although there are currently no
petrographic data to conrm this possibility.
Finally, petrographic analysis indicates that
the greenish-gray feldspathic arenite in this
study (Type  sandstone) belongs to the upper
portions of the Terrain Rouge Formation, con-
sidered equivalent to the Lower Middle Juras-
sic Phu Kradung Formation of ailand. is
subcontinental sequence is characterized in its
upper section by very ne to ne, moderately
well-sorted sandstone, intercalated with mud-
stones and calcrete horizons. e sandstone is
mineralogically and texturally quite homo-
geneous and comparable to sandstone found in
the provinces of Siem Reap and Preah Vihear,
where this formation constitutes extensive
portions of the foothills of Phnom Kulen and
crops out in isolated hills and numerous river-
beds of northern Cambodia.³
Because Terrain Rouge sandstones with
similar petrographic characteristics are widely
distributed throughout the regions, it is almost
impossible to identify the geographic prove-
nience for any of the Type  lintels studied. One
of the possible sources of feldspathic arenite
(Type sandstone) is the well-known and well-
studied quarry district active in the Angkor
period, located at the eastern foothills of
Phnom Kulen, about  kilometers northeast
of Angkor. is area is scattered with open
quarries of variable size and geometry that
follow the sandstone attitude and form a com-
plex system of stone exploitation. Stepped sur-
faces with clear chisel marks, wedge holes, and
channels indicative of the removal of sand-
stone blocks are evidence of quarrying activity
that most likely relates to the construction of
the Angkor temples. Several quarries of feld-
spathic arenite are known also in Koh Ker¹ and
close to Prasat Kdak,² and it is highly probable
that others exist in locations still unknown.
   
  
e current article reports results for a corpus
of twenty-four lintels and twelve decorative
elements of various dates and origins and pro-
vides the basis for further research on Khmer
practices of stone sourcing and usage. is
pilot study shows that the sandstones used
belong to Mesozoic sandstone formations
readily available in Cambodia, especially in
thenorthern part of the country. Half of the
analyzed objects are made with the same sand-
stone (Type), a greenish-gray feldspathic
arenite, extensively used in the construction
ofAngkor temples.³ e remaining objects
arecarved from other lithotypes (Types  and
sandstone) rarely employed in the temples
forstructural purposes, suggesting that these
stones were deliberately selected by Khmer
carvers for the production of decorative lin-
tels. e ndings for the twelve ornamental
architectural elements included in the study
are similar.
Numerous technical studies of Angkor
temples indicate that the majority of the stone
blocks were quarried from the Terrain Rouge
Formation (Type  sandstone in this study).
Conversely, the use of quartz-rich sandstones
from the Grès Supérieures Formation (Types
and  sandstones in this study) as the predomi-
nant building material is reported for monu-
ments situated in ailand close to the Khorat
Plateau, where the choice of stone seems to be
inuenced by the surrounding geology.
e occurrence of quartz arenite (Types
and  sandstones) lintels in many temples in
Stone Materials Used for Lintels and Decorative Elements of Khmer Temples
62 ◆ ◆ 63
present-day Cambodia, where other lithotypes
were more accessible and abundant or where
no sandstone at all was available in the vicin-
ity, clearly demonstrates that the choice of a
specic building material is not necessarily
based on its local availability. Understanding
how and why specic lithotypes have been
used in lintel production and whether the
selections dier from standard Khmer temple
building practices is a complex task that needs
to take into account a number of variables.
One fruitful approach may be to investigate
the organization of stone workshops and arti-
sans in charge of the ornamentation of the
temples. Recognizing similarities and dier-
ences in large quantities of decorative material
might allow some of it to be attributed to indi-
vidual sculptors or workshops. For instance,
decorative lintels and colonnettes were usually
xed in place before being carved. e cor-
relation of these elements from specic temple
sites to a specic lithotype (e.g., Phnom Da:
MG, MG, Type  sandstone; Beng
Mealea: MG, MG, Type sandstone;
Prasat Prei Khmeng: MG, MG,
Typeb sandstone) suggests that both were
sourced from the same quarry and supplied at
the same time. By comparing stone type, carv-
ing methods, and nishing techniques of deco-
rated elements from various sites, it is possible
to infer some general trends of stone usage and
learn about frequent practices of stone
workshops.
ree lintels and a colonnette from the
Musée Guimet and the National Museum of
Cambodia (MG, MG, MG,
Ka) were created in close succession in the
ninth century. ese objects and others from
the period are demonstrative of a rise in artis-
tic standardization at a time of continuity in
stone provisioning. An increase in artistic
standardization is evident in materials produced
at the time when the administrative center of
the region shifted from Mahendrapavarta
(Phnom Kulen) back to Hariharālaya (Roluos).
ere were concurrent settlements at Mahen-
drapavarta and Hariharālaya, but the lintels
and temples appear to have been produced only
when each city was the abode of the reigning
monarch and the focus of the burgeoning
empire. During this period lintels show
increased usage of the same repertoire of
motifs across the city and throughout Khmer-
controlled territory. Two dierent sites in
Hariharālaya even appear to have identical lintel
designs. Although there were developments
in iconography, the uniform lithotype (Type 
sandstone) used for lintels from Mahendrapavarta
and Hariharālaya is suggestive of an ongoing
workshop tradition using the same stone
sources as Phnom Kulen.
During the early to mid-tenth century the
number of monuments constructed in durable
materials (brick and sandstone) increased
markedly, and a proliferation of homogeneous
lintel designs emerged that identied these
foundations. Uniformity of motifs and compo-
sition are observed on the lintel friezes, which
illustrate the implementation of a design rep-
ertoire that had likely been committed to
memory in a workshop environment. From the
combination of three or four distinctive motifs
that characterize the work of a particular
workshop, one motif in particular the celes-
tial being praying in a euron was perhaps
the most idiosyncratic.¹ is and associated
motifs of polylobe eurons and romyuol owers
appear on the friezes of three of the lintels in
the Metropolitan Museum (.; ..,
see Figure; ., see Figure).² at
these lintels, likely originating from sites sepa-
rated by large geographical distances, share the
same lithotype (Type  sandstone) supports
the suggestion that they are the product of a
tenth-century tradition characterized by a
preference for a specic type of stone.
Among the lintels that were studied, those
from the early and mid-eleventh century pos-
sibly originating in present-day ailand and
now in the Metropolitan Museum (.;
.; ., see Figure) and the Arthur
M. Sackler Gallery (S.) are rendered in
white to light brown quartz arenite (Type
sandstone). is sandstone is consistent with
Carò, Polkinghorne, and Douglas
64 ◆ ◆ 65
sedimentary formations widely exposed in
northern Cambodia, along the Dangrek Range,
and on the Khorat Plateau, which is the region
the lintels are thought to come from. In this
case, workshops may have been inuenced by
the surrounding geology in the choice of the
stone for their various commissions.³
Undoubtedly, a large dataset on stone mate-
rials from provenienced objects can oer valu-
able support to current studies on material
usage and building practices during the Khmer
Empire. However, given the fragmentary infor-
mation available regarding the provenience of
many of the lintels in museum collections, as
well as the large spatial distribution of the
sandstone outcrops, conclusions for now
remain speculative.
It is clear that the three identied types of
sandstone were used simultaneously across the
Khmer Empire and within the same temple
complexes, and it is likely that their concur-
rence was determined by several concomitant
factors, including geography, geology, techni-
cal knowledge and skill, and, more generally,
building traditions. A better understanding of
building materials and traditions, as well as the
specic organization of workshops, can best be
achieved by focusing the investigation on indi-
vidual structures in situ for which building
phases and dates have been established. Such
an approach would reveal the distribution of
the stone materials in the temple structure and
help establish the relationship among stone
types, typology of surface nishing, functions
within the temple, and building phases.
e study of the stone materials used for
sculpture has signicant potential to enhance
understanding of the provisioning, production,
and distribution of classical Khmer art. Stone
characterization is useful for addressing spe-
cic questions about the authenticity and prove-
nience of carved architectural elements and
can complement archaeological studies that
consider networks of control and acquisition
ofresources, the relationships among the state,
temples, and artistic workshops, and aspects of
the economy of the Khmer Empire.
 
Associate Research Scientist
Department of Scientic Research
e Metropolitan Museum of Art
 
Australian Research Council
Postdoctoral Fellow
Department of Asian Studies
University of Sydney
 . 
Conservation Scientist
Department of Conservation and
Scientic Research
Smithsonian Institution, Freer Gallery of Art /
Arthur M. Sackler Gallery

is research was supported by an AndrewW.
Mellon Fellowship at e Metropolitan
Museum of Art, by a Forbes Fellowship at the
Freer Gallery of Art and Arthur M. Sackler
Gallery, Smithsonian Institution, and by an
Australian Research Council Post Doctoral
Fellow ship (DP). e authors are
grateful to their colleagues in Cambodia: HE
Chuch Phoeurn, secretary of state, Ministry of
Culture and Fine Arts; HE Bun Narith, direc-
tor general of the Authority for the Protection
and Management of Angkor and the Region of
Siem Reap (APSARA); HE Khuon Khun Neay
and HE Seung Kong, deputy directors general
of APSARA; Im Sokrithy, Ea Darith, and all the
APSARA sta who contributed directly to the
eldwork in Cambodia; HE Hab Touch, direc-
tor general for Museums, Antiquities and Mon-
uments, Ministry of Culture and Fine Arts,
and former director of the National Museum
of Cambodia; Kong Vireak, director of the
National Museum of Cambodia, and to the
museum for granting permission to use photo-
graphs of works from the collection; Bertrand
Porte, l’École Française d’Extrême-Orient,
which provided the samples from the National
Museum of Cambodia; John Guy, Florence and
Herbert Irving Curator of the Arts of South and
Southeast Asia, and Donna Strahan, Sherman
Stone Materials Used for Lintels and Decorative Elements of Khmer Temples
64 ◆ ◆ 65
Fairchild Center for Objects Conservation, e
Metropolitan Museum of Art; Anne Bouquil-
lon, Centre de Recherche et de Restauration
des Musées de France (CRMF), and Pierre
Baptiste, Musée National des Arts Asi-
atiques Guimet; and the Freer Gallery of Art
and Arthur M. Sackler Gallery, Smithsonian
Institution.

Dagens , p..
Polkinghorne a; Polkinghorne b,
pp. – ; Polki ng ho rne .
Baptiste et al. , pp. ; Carò ; Douglas,
Carò, and Fischer .
Polk i ng horne b, pp. – .
Coral-Rémusat ; Stern .
e lintel with a mask of Kāla is unique and warrants
separate study for its relevance to issues relating to the
dating of Khmer art and decorative lintels between the
seventh and ninth centuries, a subject of considerable
academic discourse. Presently, the principal issue is
the chronolog y and duration of decorative elements
attributable to the Kompong Preah style (see especially
Bénisti ; Boisselier ; Woodward b). A
new synthesis is required that critically appraises dec-
orative material linked to epigraphic material, recent
absolute dates from archaeological contexts (e.g., Pottier
and Bolle ), and analogous sculpture in the round.
Similarly it must consider archaism, innovation, and
peripheral and parallel artistic developments. ough
the stylized euron emblem between the Kāla’s eyes
on the Metropolitan Museum lintel (..) is
associated with the central motif of Kompong Preah
style, the majority of motifs can be ascribed to the
Prei Khmeng style; therefore it is likely the lintel was
produced in the rst half of the seventh century.
Snodgrass , pp.,  .
Marchal , p.; Ang , pp. .
A ng , pp. – .
 Coral-Rémusat , p..
 Bosch ; Coral-Rémusat , p.; Coral-
Rémusat , pp. ; Coral-Rémusat ,
pp. ; Marchal ; Stern , p.; Stern
b, p..
 B én is ti , pp. – .
 Coral-Rémusat , p.; Coral-Rémusat ,
pp. – ; C ora l-Rémus at , pp.  – .
 More recently Woodward (a, pp.,  n.)
suggested decorative inuences from China.
 anks to Heng Piphal, student at the University of
Hawai‘i at Manoa, for alerting the authors to
thissimilarity. See Carte Interactive des Sites
Archaéologiques Khmers (CISARK), http://www.
site-archeologique-khmer.org/core/showsite.
php?id= (accessed September , ).
 For example, at Kuk Roka (see CISARK, http://www.
site-archeologique-khmer.org/core/showsite.php?id=
[accessed September , ]), Olok (see Boisselier
, gs. ; Stern a, pl.LVII), Prei Khmeng
(see Christian Pottier, Alexandrine Guerin, Heng
an, Im Sokrithy, Koy Tchan, and Eric Llopis, “Mission
Archéologique Franco-Khmère sur l’Aménagement
du Territoire Angkorien [MAFKATA],” Campagne
 Rapport, EFEO, Siem Reap; Angkor National
Museum, N.), Sambor (see National Museum of
Cambodia, Ka; Boisselier , g.), Yeay Poan
(S, Sambor Prei Kuk, see Bénisti , g.).
 Polkin ghor ne b, pp. – .
 For example, at Ak Yum (see Boisselier , gs. ;
Stern a, pl. LVI), Ampil Rolum (see Parmentier
, g.), Kompong Preah (see Boisselier ,
g.), Kuk Roka (see CISARK, http://www.site-
archeologique-khmer.org/core/showsite.php?id=
[accessed September , ]), Tuol Kuhear (see
Boisselier , g.), Yeay Poan (S, Sambor Prei
Kuk, see Bénisti , g.).
 Inscription K.. See Cœdès , p..
 EFEO photo cliché fonds Cambodge INVLU.
 “Chronique” , p.; see also EFEO photo cliché
fonds Cambodge INVLU.
 Singaravélou , pp. ; see also Sharrock
, pp. – .
 Polkinghorne .
 For example, Giteau , pp. .
 Included in this study but not illustrated. For an
image, see http://www.metmuseum.org.
 Polkinghorne .
 A Zeiss Axioplan  polarized light microscope was
used at e Metropolitan Museum of Art, and a Nikon
Eclipse E polarized light microscope was used at
the Freer Gallery of Art and Arthur M. Sackler Gallery.
 e sorting (σ) of a sandstone refers to the distribu-
tion of the grain-size values around the mean grain
size. e higher the sorting (σ), the more dispersed
are the values around the mean.
 Gazzi ; Dickinson .
 Heavy minerals are detrital minerals having a specic
gravity greater than about ..
 Contri , p.; Kučera et al. , p..
 A formation is a body of rock strata that can be distin-
guished from others by specic physical characteristics.
 A group includes two or more formations that are
related to one another.
 Dating such sedimentary formations is problematic
and has been the subject of much debate. It is impor-
tant to bear in mind that evidence for their reported
ages is not conclusive. However, this study follows
ages and stratigraphy reported in the available
: geological map of Cambodia published by
the Bureau de Recherches Géologiques et Minières
and later used by other authors (United Nations ;
Sotham ).
 Contri , pp. ; Dottin , p.; Alabouvette
, p..
 See CISARK, http://www.site-archeologique-khmer.
org/core/showsite.php?id= (accessed September,
). Other quarries in the Grès Supérieures Forma-
tion can be seen at Phnom Santuk (see CISARK,
http://www.site-archeologique-khmer.org/core/
showsite.php?id= [accessed September , ]),
and Phnom Batheay (see CISARK, http://www.site-
archeologique-khmer.org/core/showsite.php?id=
[accessed September , ]).
 Racey et al. , pp. .
 Ibid., pp. ; Uchida, Ito, and Shimizu , p..
 Contri , p.; Alabouvette , p..
 Delve rt , pp. – .
Carò, Polkinghorne, and Douglas
66 ◆ ◆ 67
 Evans ; Carò and Im .
 See CISARK, http://www.site-archeologique-khmer.
org/core/showsite.php?id= (accessed September,
).
 Delvert , pp. ; Uchida, Ito, and Shimizu
, pp. – .
 For example, Saurin , p.; Delvert ,
pp. ; Uchida et al. , pp. ; Kučera et
al. , p..
 Uchida, Ito, and Shimizu , pp. .
 e many unnished and partially nished lintels in
temples throughout the Khmer world suggest that this
carving was one of the nal tasks of temple decoration.
See Polkinghorne b, pp. ; Polkinghorne
, pp. – .
 Penny et al. ; C. Pottier,A. Bolle, E. Llopis,
D.Soutif, C. Tan, J. B. Chevance, V. Kong, S.Chea,
S.Sum, F. Demeter, A.-M. Bacon, N. Bouchet,
C.Souday, and M. Frelat,“Mission archéologique
Franco-Khmère sur l’aménagement du territoire
angkorien (MAFKATA),” Campagne  Rapport,
EFEO, Siem Reap.
 Polkinghorne .
 O Ka-aek (IK.) has a lintel design identical to
four examples from Preah Kô temple. It depicts Aśvin
(cavaliers) with weapons riding three-headed crowned
Nāga (mythical serpents). e distinctiveness of this
design indicates that the same artist, from the same
workshop, who worked upon Preah Kô lintels also
fashioned the O Ka-aek lintel. See Polkinghorne a.
 e increase in construction during this period is
marked by an increase in inscriptions, which are often
placed on doorjambs (see, e.g., Lustig , esp. g.).
 Polkinghorne a.
 Romyuol is a traditional Khmer oral decorative
motif (Kbach) similar to a water lily (Nyphaea lotus)
or a hibiscus ower (Hibiscus sagittifolius).
 Uchida, Ito, and Shimizu , p..

Alabouvette 1973. Bruno Alabouvette. “Notice
explicative sur la feuille Stung Treng. Carte
Géologique de Reconnaissance à 1 / 200000. Paris:
Editions du Bureau de Recherches Géologiques et
Minières, .
Ang 2004. Chouléan Ang. “La mort: Renaissance en
abstraction iconographique. ” Udaya: Journal of
Khmer Studies  (), pp. .
Baptiste et al. 2001. Pierre Baptiste, Catherine
Chevillon, Clémence Raynaud, Anne Bouquillon,
Sandrine Pagès, Alain Leclaire, and Philippe
Recourt. “La restauration des sculptures khmères
du musée Guimet. Technè  –  (), pp. – .
Bénisti 1973. Mireille Bénisti. “Recherches sur le
premier art khmer V: La face de monstre. Arts
Asiatiques  (), pp. .
Bénisti 1977. Mireille Bénisti. “Recherches sur le
premier art khmer VII: Le problème de Sambor
S.. ” Arts Asiatiques  (), pp. .
Bénisti 2003. Mireille Bénisti. Stylistics of Early
Khmer Art.  vols. New Delhi: Indira Gandhi
National Centre for the Arts and Aryan Books
International, .
Boisselier 1968. Jean Boisselier. “Les linteaux khmers
du VIIIe siècle nouvelles données sur le style de
Kompong Prah. ” Artibus Asiae , no.   (),
pp. – .
Bosch 1960. Frederik David Kan Bosch. e Golden
Germ: An Introduction to Indian Symbolism.
Indo-Iranian Monographs . New York: Humani-
ties Press, .
Carò 2009. Federico Carò. “Khmer Stone Sculptures:
A Collection Seen from a Material Point of View.
Metropolitan Museum of Art Bulletin, no.
(), pp. – .
Carò and Im 2012. Federico Carò and Sokrithy Im.
“Khmer Sandstone Quarries of Kulen Mountain
and Koh Ker: A Petrographic and Geochemical
Study. ” Journal of Archaeological Science , no.
(), pp. – .
“Chronique” 1933. Bulletin de l’École Française
d’Extrême-Orient, no. (), pp. .
Cœdès 1953. George Cœdès, ed. and trans. Les inscrip-
tions sanskrites estampées par... [Aymonier]
maisnon encore traduites... [et] lesinscriptions
découvertes après 1900 par l’École française
d’Extrême-Orient et restées inédites, vol., Inscrip-
tions du Cambodge. École Française d’Extrême-
Orient, Collections de textes et documents sur
l’Indochine . Paris: E. de Boccard, .
Contri 1972. Jean P. Contri. “Notice explicative sur
lafeuille Tbeng-Meanchey. ” Carte Géologique de
Reconnaissance à 1 / 2000 0. Paris: Editions du Bureau
de Recherches Géologiques et Minières, .
Coral-Rémusat 1933. Gilberte de Coral-Rémusat.
“Inuences javanaises dans l’art de Rolûoh
(IXesiècle) et inuences de l’art de Rolûoh sur le
temple de Banteay Srei (n du Xe siècle). Journal
Asiatique  (), pp. .
Coral-Rémusat 1936. Gilberte de Coral-Rémusat.
Animaux fantastiques de l’Indochine, de l’Insulinde
et de la Chine.” Bulletin de l’École Française
d’Extrême-Orient , no.  (), pp. .
Coral-Rémusat 1940. Gilberte de Coral-Rémusat.
Lart khmer: Les grandes étapes de son évolution.
Études d’art et d’ethnologie asiatiques . Paris:
Éditions d’Art et d’Histoire, .
Coral-Rémusat 1951. Gilberte de Coral-Rémusat.
Lart khmer: Les grandes étapes de son évolution.
nded. Études d’art et d’ethnologie asiatiques .
Paris: Vanoest, Éditions d’Art et d’Histoire, .
Dagens 2002. Bruno Dagens. Les Khmers. Guides
belles lettres des civilisations . Paris: Les Belles
Lettres, .
Delvert 1963. Jean Delvert. “Recherches sur l’érosion
des grès des monuments d’Angkor. Bulletin de
l’École Française d’Extrême-Orient , no. (),
pp. – .
Dickinson 1970. William R. Dickinson. “Interpreting
Detrital Modes of Greywacke and Arkose. Journal
Stone Materials Used for Lintels and Decorative Elements of Khmer Temples
66 ◆ ◆ 67
of Sedimentary Petrology , no.  (),
pp. – .
DMR 1992. Department of Mineral Resources.
1:2.500.000 Scale Geological Map of ailand.
Bangkok, ailand: Department of Mineral
Resources, .
Dottin 1972. Olivier Dottin. “Notice explicative sur
la feuille Siem Reap. Carte Géologique de Recon-
naissance à 1 / 200000. Paris: Editions du Bureau
de Recherches Géologiques et Minières, .
Douglas, Carò, and Fischer 2010. Janet Douglas,
Federico Carò, and Christian Fischer. “Evidence
ofSandstone Usage for Sculpture during the
Khmer Empire in Cambodia through Petro-
graphic Analysis. ” Udaya: Journal of Khmer
Studies  (), pp. .
Evans 2009. Damian H. Evans. “Towards a Land-
scape Archaeology of Koh Ker: Methods, Issues
and Recent Research. ” In Jaya Koh Ker Project:
Annual Report 2009, edited by János Jelen, Róbert
Kuszinger, and I. Farago, pp. . Budapest:
Royal Angkor Foundation, .
Folk and Ward 1957. Robert L. Folk and WilliamC.
Ward. “Brazos River Bar (Texas): A Study in the
Signicance of Grain Size Parameters. Journal of
Sedimentary Research , no.  (), pp. .
Gazzi 1966. Paolo Gazzi. “Le arenarie del ysch
sopracretaceo dell’Appennino modenese:
Correlazioni con il Flysch di Monghidoro.
Mineralogica and Petrographica Acta  (),
pp. – .
Giteau 1965. Madeleine Giteau. Khmer Sculpture
and the Angkor Civilisation. Translated by Diana
Imber. London: ames & Hudson, .
Kučera et al. 2008. J. Kučera, J. K. Novák, K.Kranda,
J. Poncar, I. Krausová, L. Soukal, O.Cunin, and
M.Lang. “INAA and Petrological Study of Sand-
stones from the Angkor Monuments. Journal of
Radioanalytical and Nuclear Chemistry , no.
(), pp. – .
Lustig 2009. Eileen Lustig. “Power and Pragmatism
in the Political Economy of Angkor. ” PhD diss.,
Department of Archaeology, University of Sydney,
.
Marchal 1938. Henri Marchal. “e Head of the
Monster. ” Journal of the Indian Society of Oriental
Art  (), pp. .
Marchal 1951. Henri Marchal. Le décor et la sculpture
khmers. Études d’art et d’ethnologie asiatiques.
Paris: Vanoest, .
Parmentier 1927. Henri Parmentier. L’ art khmer
primitif. Publications de l’École Française
d’Extrême-Orient  . Paris: EFEO, .
Penny et al. 2006. Dan Penny, Christophe Pottier,
Roland Fletcher, Mike Barbetti, David Fink, and
Quan Hua. “Vegetation and Land-Use at Angkor,
Cambodia: A Dated Pollen Sequence from the
Bakong Temple Moat. Antiquity , no.
(), pp. – .
Polkinghorne 2007a. Martin Polkinghorne. “Artists
and Ateliers: Khmer Decorative Lintels of the
Ninth and Tenth Centuries. Udaya: Journal of
Khmer Studies  (), pp. .
Polkinghorne 2007b. Martin Polkinghorne. “Makers
and Models: Decorative Lintels of Khmer Temples,
th to th Centuries. ” PhD diss., Department of
Art History and Film Studies, Department of
Archaeology, University of Sydney, .
Polkinghorne 2008. Martin Polkinghorne. “Khmer
Decorative Lintels and the Allocation of Artistic
Labour. ” Arts Asiatiques  (), pp. .
Polkinghorne 2013. Martin Polkinghorne. “Decora-
tive Lintels and Ateliers at Mahendraparvata and
Hariharālaya. ” In Materializing Southeast Asia’s
Past: Selected Papers from the 12th International
Conference of the European Association of South-
east Asian Archaeologists, edited by MarijkeJ.
Klokke and Véronique Degroot, : .
Singapore: NUS Press, .
Pottier and Bolle 2009. Christophe Pottier and
Annie Bolle. “Le Prasat Trapeang Phong à
Hariharâlaya: Histoire d’un temple et archéologie
d’un site. ” Aséanie  (), pp. .
Racey et al. 1996. A. Racey, M. A. Love, A.C.Canham,
J.G.S. Goodall, S.Polachan, and P.D. Jones. “Stra-
tigraphy and Reservoir Potential of the Mesozoic
Khorat Group, NE ailand. Journal of Petro-
leum Geology , no.  (), pp. .
Saurin 1954. Edmond Saurin. “Quelques remarques
sur le grès d’Angkor.Bulletin de l’École Française
d’Extrême-Orient , no. (), pp. .
Sharrock 2007. Peter. D. Sharrock. “e Mystery of
the Face Towers. ” In Bayon: New Perspectives,
edited by Joyce Clark, pp. . Bangkok: River
Books, .
Singaravélou 1999. Pierre Singaravélou. L’École
française d’Extrême-Orient ou l’institution des
marges (1898 1956): Essai d’histoire sociale et
politique de la science coloniale. Paris: Éditions
L’Harmattan, .
Snodgrass 1985. Adrian Snodgrass. e Symbolism
of the Stupa. Ithaca, N.Y.: Southeast Asia Program,
Cornell University, .
Sotham 1997. Sieng Sotham. “Geology of Cambo-
dia. ” Coordinating Committee for GeoScience Pro-
grams in East and Southeast Asia Technical
Bulletin  (), pp. .
Stern 1927. Philippe Stern. Le Bayon d’Angkor et
l’évolution de l’art khmer: Étude et discussion de la
chronologie des monuments khymers. Annales du
Musée Guimet, Bibliothèque de vulgarisation .
Paris: Paul Geuthner, .
Stern 1934. Philippe Stern. “Évolution du linteau
khmer.Revue des arts asiatiques , no. (),
pp. – .
Stern 1938a. Philippe Stern. “Hariharalâya et Indra-
pura.” Bulletin de l’École Française d’Extrême-
Orient , no. (), pp..
68 ◆ ◆ 69
Carò, Polkinghorne, and Douglas
68 ◆ ◆ 69
Stern 1938b. Philippe Stern. “Le style du Kulên.
Bulletin de l’École Française d’Extrême-Orient ,
no. (), pp. – .
Uchida et al. 2007. Etsuo Uchida, Olivier Cunin,
Chiyuki Suda, Akyio Ueno, and Takeshi Kakagawa.
“Consideration on the Construction Process and
the Sandstone Quarries during the Angkor Period
Based on the Magnetic Susceptibility. Journal of
Archaeological Science , no.  (), pp. .
Uchida, Ito, and Shimizu 2010. Etsuo Uchida, K. Ito,
and N. Shimizu. “Provenance of the Sandstone
Used in the Construction of the Khmer Monuments
in ailand. ” Archaeometry , no.  (),
pp. – .
United Nations 1993. United Nations. Economic
and Social Commission for Asia and the Pacic:
Atlas of Mineral Resources of the Escap Region,
vol., Cambodia, Explanatory Brochure. New
York: United Nations, .
Woodward 2010a. Hiram Woodward. “Bronze
Sculptures of Ancient Cambodia.” In Gods of
Angkor: Bronzes from the National Museum of
Cambodia, edited by Louise A. Cort and Paul Jett,
pp. . Washington, D.C.: Arthur M. Sackler
Gallery, Smithsonian Institution, .
Woodward 2010b. Hiram Woodward. “Dvaravati,
Siep, and Wendan. Bulletin of the Indo-Pacic
Prehistory Association  (), pp. .
Workman 1977. David Richard Workman. Geology
of Laos, Cambodia, South Vietnam and the East-
ern Part of ailand. Overseas Geology and
Mineral Resources . London: Institute of
Geological Sciences, .
 
Unless otherwise indicated, photographs of works in the
Metropolitan Museum’s collection are by the Photograph
Studio, e Metropolitan Museum of Art.
Department of Scientic Research, e Metropolitan
Museum of Art: Figures, , , 
Musée National des Arts Asiatiques Guimet, Paris:
Figure
National Museum of Cambodia, Phnom Penh: Figures, 




  
 
 
   

 
 Metropolitan 

 



Musée National des Arts Asiatiques–Guimet
 Arthur M. Sackler
Gallery, Smithsonian Institutio
Washington, D.C
  
  

 
(from the Terrain Rouge Formation)

  

Grès Supérieures Formation


 
e Metropolitan Museum of Art
 
University of Sydney
. 
Smithsonian Institution, Freer Gallery of Art /
ArthurM. Sackler Gallery
    
 
... com.). Although the western Cambodian Grès Supérieurs may be chronostratigraphic equivalents of at least some of the formations of the Khorat Group [10], it seems impossible to find clear lithostratigraphic equivalents of the Thai formations in that part of Cambodia. Fortunately, palynological and palaeobotanical evidence from the Bokor area yields important stratigraphic information about the age of the local Grès Supérieurs. ...
Article
Full-text available
The first discovery of a dinosaur bone from the Kingdom of Cambodia is reported in this paper. It consists of a sauropod fibula from a sandstone layer on Koh Paur island, in Koh Kong province, in south-western Cambodia. The dinosaur-bearing bed belongs to the non-marine Grès Supérieurs series and is apparently of Early Cretaceous age. On the basis of various characters, notably the development of the anteromedial crest, the dinosaur fibula from Koh Paur is referred to a euhelopodid titanosauriform. This first dinosaur discovery in Cambodia suggests that the thick non-marine formations which cover vast areas in the south-western part of the country are potentially an important source of continental Mesozoic vertebrates.
Article
From ∼ 900 to 1450 CE the Angkorian Khmer state extended control over much of inland Southeast Asia. This economically complex polity supported a large number and a diverse range of specialist producers across its territory. Given the limited types of textual data, our understanding of the Angkorian Khmer economy is heavily reliant on archaeological data to address key issues such as the relationship between specialist producers and the Angkorian elite. Here our focus is on the production side of this question through elemental characterization (Neutron Activation Analysis) of a large sample of high temperature ceramics - stonewares - from 15 Angkorian Khmer complexes in Cambodia and Thailand. In the majority of cases we can identify kiln-specific compositional signatures as the groundwork for the next stage of analysis: studying the consumption-side dynamics of Angkorian stoneware distribution. The study also highlights the spatial resolution of stoneware NAA possible in this context. Together, defining patterns of stoneware production, distribution and consumption will allow a new bottom-up perspective on the operation of the Angkorian Khmer political and ritual economy.
Article
Full-text available
Detailed magnetic susceptibility measurement was conducted on the sandstones used for the Angkor monuments constructed in the period spanning the Preah Ko and Angkor Wat styles, and the construction process of the buildings and quarries of the sandstones was considered. Combined with the previous study on the sandstones of the Bayon style [Uchida, E. The construction process of the Angkor monuments elucidated by the magnetic susceptibility of sandstone, Archaeometry 45, 221e232], the magnetic susceptibility measurement revealed that there were 7 sandstone quarries corresponding to Stages I, II, III, IVa, V, VII and VIIIb during the Angkor period. The sandstones used for the monuments belonging to Stage I show average magnetic susceptibility values ranging from 1.1 to 2.3 Â 10 À3 SI units. In the Bakheng style period (Stage II), the average magnetic susceptibility of the supplied sandstones decreased gradually from around 10 to 1 Â 10 À3 SI units over time. The magnetic susceptibility range of the sandstones used for the monuments of Stage III was from 2.3 to 3.0 Â 10 À3 SI units. In the early Angkor Wat style period (IVb), the average magnetic susceptibility of the supplied sandstones increased over time, reflecting the supply of the sandstones from two different quarries, that is, the quarry corresponding to the Khleang and Baphuon style monuments (Stage IVa), with low average magnetic susceptibilities of 1.1 to 2.4 Â 10 À3 SI units, and the quarry corresponding to the main Angkor Wat style monuments (Stage V), with high average magnetic susceptibilities ranging from 2.8 to more than 4.3 Â 10 À3 SI units. The sandstones of Stage VII show low average magnetic susceptibility around 1 Â 10 À3 SI unit. The sandstones of Stage VId is a mixture of sandstones of Stages V and those of Stage VII. The sandstones with high magnetic susceptibilities are found in the monuments belonging to Stage VIIIb, reflecting the supply from the new sandstone quarry.
Article
Full-text available
Sandstone from several ancient quarries and natural outcrops located at the eastern foothill of Kulen Mountain and in Koh Ker, northern Cambodia, has been characterized by means of petrographic analysis, scanning electron microscopy, and geochemical analysis. The samples have been collected during a series of field surveys organized jointly by The Metropolitan Museum of Art and the Authority for the Protection and Management of Angkor and the Region of Siem Reap (APSARA). The data show how the sandstone quarried from these sites is characterized by an overall petrographic and geochemical homogeneity, and is consistent with the upper terms of the Lower–Middle Jurassic Terrain Rouge Formation.This result suggests that petrography and geochemistry of sandstone belonging to the Terrain Rouge Formation might not be unambiguously characteristic of single quarry sites, but rather representative of vast portions of the sedimentary sequence.
Article
Full-text available
We investigated the sandstone used in the construction of the Khmer monuments situated upon and around the Khorat Plateau in north-east Thailand in order to clarify the provenance. The sandstones of the 22 investigated Khmer monuments can be classified into three groups. The sandstone of Group 1 is lithic and is derived mainly from the Khok Kruat Formation. This group includes the sandstone used at Phimai, Phnom Wan, Muang Khaek etc. The sandstone of Group 2 is siliceous and can be subdivided into three further groups. The sandstone of Group 2 is considered to have been derived from the Phu Phan, Phra Wihan or Sao Khua Formations. The sandstone used at Muang Tam, Phnom Rung, Sdok Kok Thom, Preah Vihear (Khao Phra Wihan), Narai Jaeng Waeng etc. belongs to Group 2. The sandstone of Group 3 is feldspathic and is correlated with the grey to yellowish-brown sandstone that is commonly used in the Angkor monuments in Cambodia. This sandstone is used at Wat Phu and Hong Nang Sida in Laos. The above results reveal that the choice of sandstone used for the Khmer monuments, including the Angkor monuments, was dictated by the surrounding geology.
Gilberte de Coral­Rémusat. L' art khmer: Les grandes étapes de son évolution. Études d'art et d'ethnologie asiatiques 1
  • Coral-Rémusat
Coral-Rémusat 1940. Gilberte de Coral­Rémusat. L' art khmer: Les grandes étapes de son évolution. Études d'art et d'ethnologie asiatiques 1. Paris: Éditions d' Art et d'Histoire, 1940.
La restauration des sculptures khmères du musée Guimet
  • Baptiste
Baptiste et al. 2001. Pierre Baptiste, Catherine Chevillon, Clémence Raynaud, Anne Bouquillon, Sandrine Pagès, Alain Leclaire, and Philippe Recourt. " La restauration des sculptures khmères du musée Guimet. " Technè 13 – 14 (2001), pp. 131 – 40.
Gilberte de Coral­Rémusat Influences javanaises dans l'art de Rolûoh (IXe siècle) et influences de l'art de Rolûoh sur le temple de Banteay Srei (fin du Xe siècle)
  • Coral-Rémusat
Coral-Rémusat 1933. Gilberte de Coral­Rémusat. " Influences javanaises dans l'art de Rolûoh (IXe siècle) et influences de l'art de Rolûoh sur le temple de Banteay Srei (fin du Xe siècle). " Journal Asiatique 223 (1933), pp. 190 – 92.
24 For example 25 Included in this study but not illustrated. For an image, see http
  • Polkinghorne
23 Polkinghorne 2008. 24 For example, Giteau 1965, pp. 79 – 84. 25 Included in this study but not illustrated. For an image, see http://www.metmuseum.org. 26 Polkinghorne 2008.
Notice explicative sur la feuille Tbeng­Meanchey Carte Géologique de Reconnaissance à 1
  • Contri Jean
  • P Contri
Contri 1972. Jean P. Contri. " Notice explicative sur la feuille Tbeng­Meanchey. " Carte Géologique de Reconnaissance à 1 / 20000. Paris: Editions du Bureau de Recherches Géologiques et Minières, 1972.
1933, p. 1134; see also EFEO photo cliché fonds Cambodge INVLU16242
  • Chronique
Chronique " 1933, p. 1134; see also EFEO photo cliché fonds Cambodge INVLU16242.