Microchemical Journal 126 (2016) 121131

Contents lists available at ScienceDirect

Microchemical Journal
journal homepage: www.elsevier.com/locate/microc

Casting cores of French bronze statues of the 16th and 17th centuries:
Identication of regional practices and artistic choices
Manon Castelle a,b,, Yvan Coquinot a, David Bourgarit a,c
a
b
c

Centre de Recherche et de Restauration des Muses de France (C2RMF), Palais du Louvre, 14, quai Franois Mitterrand, 75001 Paris, France
Laboratoire DYPAC, EA 2449, Universit Versailles Saint Quentin en Yvelines, 47, boulevard Vauban, 78280 Guyancourt, France
Prhistoire et Technologies, CNRS-UMR7055, Maison Archologie & Ethnologie, Ren-Ginouvs (MAE), 21 Alle de l'Universit, 92 023 Nanterre cedex, France

a r t i c l e

i n f o

Article history:
Received 24 July 2015
Received in revised form 17 November 2015
Accepted 17 November 2015
Available online 22 November 2015
Keywords:
Casting core
Large bronze
French Renaissance
Petrography
Image analysis
PIXE

a b s t r a c t
What can the chemical and petrofabric study of casting cores of early modern large bronzes tell us about the
practices of sculptors and foundry men? The technical study of 23 major large French bronzes dated from the
16th and 17th centuries provides some answers. These include three prestigious royal commissions, namely
two of the six bronzes copied by Primaticcio from the marble models of the Belvedere for Francis I, two of the
six statues from the funeral monument of Henry II and Catherine de Medicis, and the three bronzes from the
heart monument of Constable Anne de Montmorency. Among the investigations carried out (alloy composition,
X-ray radiography, and so on) [1], the chemical and petrographic analysis of the refractory cores sampled from
inside the statues has proved to be crucial. Surprisingly, this approach is far from systematically applied in the
eld, despite some successful results [2] including the pioneering work of C. Reedy [3]. This paper introduces
the new methodology of casting core analysis developed at the C2RMF. Results obtained from a large corpus
of French bronze statues will be described, illustrating the complementary role of core analysis alongside the
informative potential of metal analysis and the characterization of casting technique through multivariate
technological studies. During the 16th and 17th centuries in France, casting cores were generally created from
a sandy clay material with a variable addition of organic. The results of the study cast light on two aspects of
the nature of the cores. On the one hand, the bulk chemical compositions of cores and mineralogical composition
of added sand show a relatively good homogeneity within the region of Paris and its surroundings, thus testifying
to a similar provenance of raw materials. Conversely, signicant variations were evident in compositions from other
regions. Consequently, the casting cores constitute an excellent regional marker. On the other hand, signicant
variations were observed in the size distributions of aplastic inclusions, thus witnessing local foundry know-how
through the use of several sandyclayey mixtures.
2015 Elsevier B.V. All rights reserved.

1. Introduction
Clay cores (complex mixtures of sands and clays) have long been
used in the production of hollow-cast bronze statues. The use of a core
reduces the metallic wall thickness, helping to minimize casting defects
that can be generated during the solidication of large volumes of metal.
Further benets are highlighted by the Italian metallurgist Vannoccio
Biringuccio (1480c.1539) in his Pirotechnia of 1540 [4], where he
describes that cores minimize the amount of metal used and thereby
reduce both the weight and the production cost of the statue. As an
important step of the casting process, the design of the casting core
is driven by various technical, cultural, and economical constraints
related to supply of raw materials. The nal material characteristics
Selected papers presented at TECHNART 2015 Conference, Catania (Italy), April 27
30, 2015.
Corresponding author.
E-mail address: manon.castelle@culture.gouv.fr (M. Castelle).

http://dx.doi.org/10.1016/j.microc.2015.11.032
0026-265X/ 2015 Elsevier B.V. All rights reserved.

result from a manufacturing process comprising ve steps, namely

the selection, possible modication, the mix of raw materials, the
shaping and the baking.
Although quite a number of technological studies [1] have been carried out on casting cores, these are still undertaken in a less systematic
way than, for instance, metal analysis of bronze statues, and few have
led to results that provide denitive or useful conclusions so far. Several
reasons for this may be put forward. First, most body of statues that
have been studied are either too heterogeneous [3] or conversely
restricted to monographic studies [516]. Second, the objectives
targeted as encountered for ancient ceramic studies [17,18] are
often restricted to the provenance of the raw material [2,3,5,6,19].
Characterization of workshop and/or related technical specicities is
rare [2]. Third, methodologies may be incomplete and/or hardly comparable with other studies. Whereas petrographic characterization is ideally
performed by thin-section analysis of casting core chip [2,57,15,20],
several studies were only performed on loose grains samples [3,9,10]. In
this case, observation and/or quantication of textural features, including

122

M. Castelle et al. / Microchemical Journal 126 (2016) 121131

grain size distribution, cannot be achieved, although such observation

may prove essential for casting core characterization. Qualitative petrographic studies have been undertaken [3,20] but a quantitative approach
using point counting [2,7,15] or image analysis [2,15] provides more
signicant data. Indeed, a quantitative approach has proved to be crucial
in order efciently to distinguish multi-source clays through the natural
sandy temper grain size distribution characterizing each clay and/or the
addition of sandy temper to clayey material for cultural objects such as
ancient ceramics [2125]. Chemical characterization is usually performed
using X-ray diffraction [2,5,6,11,12,15,19] even though detection is
generally limited to about 2%, sometimes more [26], Scanning Electron
Microscopy (SEM) coupled with Energy Dispersion Spectroscopy (EDS)
[2,5,7], Inductively Coupled Plasma Atomic Emission or Mass Spectroscopy [2,6] (ICP-AES or ICP-MS) neutron activation [6,19] or rare earth
element analysis [2]. Finally, very studies have focused on a consistent
body of objects while using a sufciently complete methodology, including petrographic characterization and quantication. In this respect, the
recent work by Mugnaini et al. deserves attention [2]. Within their Italian
Renaissance body of objects, the authors were able to distinguish several
groups according to the chemical composition of the core samples, thus
yielding promising clues for authentication. Our work aims to take a
step further in the characterization of casting cores through quantication
of their inclusions using image analyses and optical cathodoluminescence.
This will be enhanced by an integrated approach that includes metal analysis, the investigation of casting techniques and the study of historical
sources.
2. Materials and method
Thirty-one samples stemming from 23 French bronze statues,
most of them from the 1617th centuries, have been analyzed. This
body of objects includes major artworks, which were among the
rst examples of large-scale French bronzes of the early modern
period: the bronze copies of the Belvedere marbles by Francesco
Primaticcio (15401542), the Virtues from the funerary monument
of Henri II and Catherine de Medicis (1567), and the three Personications from the funerary monument of the Constable Anne de Montmorency (1571) (Fig. 1). The two Spiritelli (1439) attributed to Donatello
(13861466) were also analyzed, in order to investigate possible
variations in the founding practice between France and Italy. When
possible, samples were taken from several locations within the statue
in order to test the homogeneity of the core. However, all the statues
that were studied had been cast in one pour of metal, with no assembly
of different parts [27,28], thus limiting variations in the casting core
composition.
In order to investigate the provenance of the raw material and to
complete previous reference data [29] concerning ceramic objects,
15 clay and 3 sand reference samples were obtained and analyzed
from locations that would have been used by the foundry-men.
Those from Paris and its surroundings comprised 6 Sparnacian, 1
Stampian, 5 Baremian, 1 Albian, with 2 Hauterian clays and sands
from Fontainebleau and Auvers. For comparison, 3 Eocian clays from
Poitiers (400 km from Paris) were also analyzed.
Bulk chemical elemental composition was determined using
Particule Induced X-ray Emission (PIXE) at the AGLAE facility (C2RMF,
Paris, France) [30,31]. Surface scans were performed on 3 mm2 resinimpregnated polished sections. No variations in composition were
observed on those areas, indicating that they were representative of the
whole bulk. Principal Component Analysis (PCA) has been carried out
on 8 major elements to sort out the core composition. In order to compare
casting cores and reference clay compositions for provenancing, the ratios
of only four major and minor elements were considered, namely Mg, Sr, Ti
and Al. As a consequence, the chemical signature of clay may be severely
altered. Indeed, most of the other major elements can be found in several
minerals such as calcium which is contained in calcite, secondary gypsum
and plagioclase feldspars or potassium contained in potassic feldspar and

Fig. 1. Jean Bullant, Funerary monument of Constable Anne de Montmorency, 1571,

Louvre Museum, statue height c. 1m30). Peace (front) and Abundance (right) by Barthlmy
Prieur, and Justice (left) by Martin le Fort. C2RMF A.Maigret.

micas. Thus, Mg, Sr, Ti and Al can be considered as relevant chemical

markers of the clay minerals since they are present mostly in clayey
fraction, sand and silt grains corresponding essentially to quartz (N97%
of all minerals).
In order to characterize inclusions quantitatively for the casting
cores, petrographic characterization was carried out on few mm2 thin
sections (the size depends of the availability and accessibility of casting
core materials). As a rst step, an Optical Microscope (OM) was used.
Two complementary images were acquired using either the crossed
polar set-up or not, with a ten-fold magnication. Two different angles
of the crossed polar lenses were used, so as to maximize the number of
visible grains contrasting with the dark matrix on the nal image.
For each core sample, four pairs of OM images have been analyzed
using the AxioVision SE64 4.9.1 software. This allowed total control of
the analysis, from the primary treatment of the image to the nal detection and quantication of inclusions. Those under 20 m have not been
taken into consideration since they can hardly be dissociated from the

M. Castelle et al. / Microchemical Journal 126 (2016) 121131

clayey matrix. Moreover, they come from the natural silt fraction of the
clay chosen as raw material, rather than from the added sand. In
the absence of any clear standardized methodology for the OM image
analysis of cultural heritage ceramics (including casting cores), we
have set up our own methodology inspired by previous studies.
Several features were thus quantied:
the sandy temper area ratio,
the area ratio occupied by each grain on the total area occupied by
sandy temper (area to area sum) [24],
the size of each grain according to the maximum ferret values,
and nally the shape of the grains according to their degree of
sphericity and their ferret ratio.

The examination of the correlation between area to area sum values

[24] and the size of each grain has proved to be a determinant in order to
distinguish the mode(s) of grain size distributions. In contrast to what
has previously been proposed [2], the frequency of appearance of each
category of grain size is most of the time not, on its own, sufcient to
identify variations between samples of comparable sandy clay materials.
The graphic representation of grain size frequencies indeed led to systematic and similar uni-modal distributions for all samples. For shape parameters of the grain, the frequency of each category (from 0 to 1 with a step
of 0.05) of ferret ratio and sphericity degree has also been plotted. The
clayey matrix area ratio has been quantied by colorimetric segmentation
on binary transformed polarized non-analyzed OM images using ImageJ
software. The latter has proved efcient for this purpose and more userfriendly than AxioVision. Once the grain and the clayey matrix surface
ratio had been calculated, the total porosity could be inferred by considering that the casting core is mainly composed of sand, clayey matrix and
porosity. Note that no distinction has been made between porosity from
the shrinkage of the clay and porosity from the organic temper. Addition
of organic temper could be identied through prints left by the organic
materials after carbonization or devolatization on the sandyclayey
material. Those have been observed using OM with either the crossed
polars set-up or not (see Fig. 4 and Table 1).
Optical Cathodoluminescence (OCL) was carried out with the CITL
(Cambridge Image Technology Ltd) system mounted on a BH2 Olympus
microscope. Images were captured using a Spot CCD camera, with an
acquisition time of 7 to 10 s. Potassic feldspars have been quantied
for most of the cores from the analysis of cathodoluminescence images.
Feldspar shows a strong blue luminescence that allows their quantication by colorimetric segmentation through ImageJ analysis software.
3. Results
3.1. Preliminary observations
All samples are constituted of a sandy clayey material. All casting
cores were shaped around an iron armature. Bricks regularly placed
next to each other with a plastic sandyclayey material used as cement
were observed in a number of statues, namely in the Peace (1571) and
Diana the Huntress (1603) by Barthlmy Prieur (15361611), the
torch bearing Angel (1613) by Francesco Bordoni (15801654), Sleeping
Ariane and Laocon (both 1540) by Primaticcio (15041570) (Fig. 2). In
the following, the sandyclayey material used as cement or composing
the whole part of the casting core will be designated as plastic core, as
opposed to the fragile bricks. Note that the case of Prieur's Abundance
(1571) is less clear, but one may suspect the presence of bricks also.
Most samples exhibit a light to dark brown coloration, except those
from the two Italian Spiritelli, which present black and opaque coloration. When sampled just next to the metallic wall, the core has a typical
blackish color, caused by the local high temperature (due to the contact
with molten metal), as well as the relatively reducing environment at
this location.

123

3.2. Petro fabric characterization

3.2.1. Relative proportions of sand and porosity
In all French core samples the sand and porosity ratios are high
when compared to usual ceramic pastes [32,33]. They randomly vary,
respectively from 22 to 41% and from 16 to 50%. The clayey matrix
proportion varies from 20 to 50% (Fig. 3). As shown by multiple
sampling, a relative variation of this ratio within one statue may be
estimated around 15%. Interestingly, the bricks exhibit similar paste as
the plastic cores. Yet, no organic temper or remains of organic material
prints could be observed in the former, whereas it has been used in
various quantities for the plastic core (see Table 1). Martin le Fort's
Justice bears large amounts of organic material prints whereas they
are rare in Ponce Jacquiot's Prudence. For the Italian cores, image analyses were difcult to carry out, because of the small size of the samples
and the darkness and opaqueness of the clayey matrix. However, the
inclusions and porosity proportions are clearly smaller, respectively 9%
and 16% for Spiritello 1, 1420% and 2021% for Spiritello 2.
3.2.2. Silt and inclusions
In both French and Italian cores, silt and inclusions are mostly constituted of monocrystalline quartz (accounting for ~8090% of the inclusions volume). The inclusions, constitutive of the sandy temper, are
sub-angular to sub-rounded shaped. The ferret ratio value of all grains,
indistinctively of their nature, remains between 0.7 and 0.8 for both
the French and Italian cores. Low amounts (b 1%) of polycrystalline
quartz particles as well as feldspars (mostly b100 m) were systematically observed (Fig. 4 and Table 1). For French cores, the proportion of KFeldspars ranges from 0.54% to 2.8%. Note that in the three natural sands
investigated, the contents of K-Feldspars ranged from 0.12 to 0.25%. In
numerous French core samples (see Table 1) chert fragments were observed (b 1%), as well as very few glauconite pellets, micas, epidote and
calcite. Amphibole has also been observed occasionally. Conversely,
sandstone fragments were found in the two Italian core samples. The
petrographic characterization is synthesized in Table 1.
3.2.3. Grain size distribution
In the French cores, whereas the proportion of inclusions may vary,
grain size distribution remains unchanged within each statue. Five types
of grain size distributions were found (Fig. 5) leading to ve different
sandyclayey mixtures (see Table 1 for the detail of each statue):
A uni-modal grain size distribution (Mixture 1a), spreading between
70220 m, with a maximum around 110150 m. This relates to 6
plastic cores.
Another uni-modal grain size distribution is observed (Mixture 1b),
this time characterized by a 40200 m mode, with a maximum
around 7090 m, that is slightly but clearly lower than in Mixture
1a. This relates to 2 plastic cores.
A third uni-modal distribution (Mixture 1c), exhibiting small size
grains with a maximum reached around 30 m and decreasing softly
until 200 m. This relates to only one brick.
A bi-modal distribution (Mixture 2) has been identied with one
mode around 2060 m and the other mode around 70270 m.
This relates to 13 plastic cores and 1 brick.
A tri-modal distribution (Mixture 3) develops around three modes:
2060 m, 70220 m, and N250 m. This relates to 2 plastic cores.

As far as Parisian raw materials are concerned, grain size distribution

always appears uni-modal. For Parisian clay, grain with a size N100 m
is rare, whereas for sands, it is rare to nd grains b 100 m (Fig. 5).
Consequently, the lower grain size category (mostly b100 m) may
reasonably be associated with the natural sandy silt fraction of the
clay. Conversely, grains N100 m may be characteristic of added temper.

124

M. Castelle et al. / Microchemical Journal 126 (2016) 121131

Table 1
presentation of the studied statues and synthesis of main petrographic observation.

67911

Duplessis
Bronze copies of the
Bevedere marbles

Chteau de
Fontaine
bleau

MR 3284

Sleeping Ariane

Rustici

Jean Varin

Primatice

un.
LeSueur or
Perlan
Pierre

Primatice

Benoit Le

Fun. mon. Henri II

and Catherine de
Medicis

Denis

Basilique Saint

C2RMF

Ponce

Benot le

70493

Jacquiot

Boucher

C2RMF

Prudence

70492
MR 1681

Abundance

Ponce

Benot le

Fun. mon.

Jacquiot

Boucher

Barthlmy

Nicolas

Prieur

Pron

MR 1682
Martin le

Louvre, sculpture dept.

of Constable Anne de
Montmorency

Justice

45

35

1b

2.7

1.6

19

55

26

1542-1543

Plastic core

Light brown

47

30

23

42

30

28

Plastic core

and black

Dark brown
black surface

Rest of fibers

1.6

1542-1543
Brick

Light brown

26

40

34

Plastic core

Dark brown

30

48

22

Plastic core

Dark brown

32

35

33

1567

Plastic core

Dark brown

46

20

34

1.6

1571

Plastic core

Brown

22

50

28

1a

0.5

Plastic core

Brown

30

35

35
1a

1.5

Plastic core

Brown

42

30

28

Plastic core

Light brown

28

35

37

Plastic core

Light brown

35

30

35

Brick

Light brown

16

50

34

Dark brown

32

40

28

1a

1.3

50

31

1a

0.3

1567

1571
un.

Foraminifera
miliolidae

1a
MR 1683

Peace

Barthlmy

Nicolas

Prieur

Pron

1571

MR 1683

MR 1784

Morvillier
Funerary Genius

MR 1685

Fun.mon Thou

Germain
Pilon
Barthlmy
Prieur

un.

1578

Plastic core

un.

1583-1585

Plastic core

Brown

19

Brick

Light brown

40

35

25

1c

Plastic core

Brown

39

20

41

Plastic core

Brown

38

25

37

1603

Plastic core

Dark brown

28

50

22

un.

1603

Plastic core

Black

43

30

27

un.

1603

Plastic core

Black

31

35

34

un.

Early 17th c.

Plastic core

Brown

19

50

31

1.4

un.

th
Early 17 c.

Plastic core

21

45

34

2.5

Tourmaline

17

45

38

2.5

Biotite

23

40

37

RF261
Diana the

RF261

Huntress

Barthlmy
Prieur

un.

1603

RF261

MR 3440

Ewer
Dina fontain

Chteau de

Fontainebleau

MR 3401

Dog

Stag

Royal couple

Henri IV as

WB-38

Angel
MR 1668

chains
LL32

Bust of Louis XIII

Monument on
the Pont au
Change

MR 3232

Boy Louis XIV

MR 3231

Louis XIII
Louis XIV crushing the
Fronde

un.
Fun. mon.

Bordoni

Biard

Biard

OA 11055

Young Captive on

Bordoni
Pierre Ier

Pierre Ier

OA 11054

Torch bearing

Francesco

70873

Jupiter
Marie de Mdicis

Francesco

C2RMF

as Juno

Louvre, sculpture dept.

et

20

Dark brown

MR 1683

Bust J. de

Carnaval-

un.

Plastic core

un.

Fort
MR 1682

20

1642

Boucher

Temperance

60

Plastic core

and
MR 3290

20

1536

Beauchesne

Laocon

un.

fiber prints/

C2RMF

14

Black C

MR 1680

65

Other

Bust of Arman

21

Amphibole

Pie De Savoie

dept.

black
Light brown

un.

Chert

sculpture

black
Dark brown/

Clayey

Plastic core

Dark brown/

75

Ox.-Fe pellet

1439

16

inclusions

1773-2

black

Epidote

un.

Plastic core
MJAP-S

Spiritello (2)

Dark brown/

Musc.

Donatello

Plastic core

Glauconite

1439

Lim. fgt

un.

Cal. : c. / sp.

Donatello

Feld.*

1773-1

Quartz pol.

Cantoria of Santa
Maria del Fiore

MJAP-S

Spiritello (1)

Gisant D'Albert

Louvre,

Color

Quartz

Type of
sample

Sandy clay

Date

mixture

Caster

Sand (%)

Artist/craftman

Clayey

Ref

Porosity (%)

Title statue

matrix (%)

group

Jacquemart
-Andr

Monum./

Loc.

S3420

Barthlmy
Prieur
Barthlmy
Prieur
Francesco

Francesco

Bordoni

Bordoni

Bordoni

Francesco

Francesco

Bordoni

Bordoni

Francesco

Francesco

Bordoni

Simon

Simon

Guillain

Guillain?

Simon

Simon

Guillain

Guillain?

Gurin
Gilles

un.

1613

Plastic core

Brown - light
brown
Reddish
brown
Reddish

0.68

1618

Plastic core

1643

Plastic core

Light brown

35

25

40

1a

1.0

1647

Plastic core

Brown

30

40

30

0.6

1647

Plastic core

Brown

30

35

35

Plastic core

Brown

27

45

28

1b

th

18

c?

brown

Unknown
Funerary monument
plastic core

* Felpaths quantification is indicated in ratio % of the thin section surface when CL

image analyses have been possible. Otherwise visual quantification is indicated

Finally, the ve aforementioned sandyclayey core mixtures were

obtained using two different kinds of clay and at least two different
kinds of sand.

RECIPES

Presence of
hair

1.4

Unimodal type a

1a

7090%

Unimodal type b

1b

2010%

Unimodal type c

1c

105%

Bimodal

50%

Trimodal

Absence

For Mixtures 1a and 1b, a naturally low-silted clay has been added
with only one type of sand; the sand used in 1b having a slightly smaller
grain size than the sand used in 1a.

M. Castelle et al. / Microchemical Journal 126 (2016) 121131

125

Fig. 2. Francesco Primaticcio Laocon (15401542, Chteau de Fontainebleau, H. 1.91 m L, 1.50 m). (a) General view. (b) View from below. The core is lled with bricks (10 10 cm). See in
the middle of the image the extremity of the main iron armature of the statue C2RMF A.Maigret.

A naturally high-silted clay has been used for Mixtures 1c, 2 and 3. A
very low quantity of sandy temper (if any) has been added to the clay
for Mixture 1c. For Mixture 2, a sand with a uni-modal distribution
(similar to the one used for Mixture 1a) has been added to the clay.
The bi-modal distribution found in the sand of Mixture 3 may originate
either from one sand characterized by a bi-modal distribution or, most
probably, from a mix of two different kinds of sands.
The clayey matrix of the Italian cores contains a ner siltsandy
fraction: most of the grains are smaller than 150 m and grains bigger
than 200 m are almost nonexistent.

Trace element contents are similar between French and Italian cores,
except for Cr and Mn. The French cores contain lower amounts of Cr
(3092 vs 105185 ppm) and Mn (7603 vs 8361262 ppm). PCA
conrms the aforementioned observations of both narrow core compositions for all French cores analyzed and differences between French and
Italian productions (Fig. 6). Note that our data is consistent with
previous work on Italian productions [2].
Regarding the provenance of the clay, four clays/earths appear as
potential candidates (Fig. 7), namely both the Stampian and Sparnacian
clays from Paris, as well as the Barremian, and Albian clays from
Beauvais. This will be discussed further below.

3.3. Chemical characterization

4. Discussion
All studied casting cores are constituted of aluminasilicate based
sandyclayey materials. French cores exhibit a relatively narrow
composition when considering major elements (SiO2, Al2O3, Na2O,
MgO, P2O5, K2O, CaO and TiO2). This alone distinguishes the French
cores from those of the two Italian Spiritelli (Table 2).
French cores have higher contents of SiO2 (8085 vs 6974 wt.%)
and lower contents of Al2O3 (67 vs 1214 wt.%). K2O contents (0.9
2.8 vs 2.32.7%) as well as CaO (0.64,4 vs 0.73 wt.%) are more or
less similar in both French and Italian cores. Note that in general Ca
variations are mainly due to the presence of more or less calcite and
gypsum salt in the cores, whereas the clayey matrix systematically
exhibits low CaO contents. Na and Mg are present in low quantity in
both cases though always lower in French cores (0.150.6 vs 1.2
1.7 wt.% Na2O, and 0.31 vs 1.41.9 wt.% MgO). Note the exceptional
high amount of MgO in one of the sample providing from the French
Ewer by Bordoni (2.2 wt.%).

The high ratio of inclusions and porosity in the core are necessary
both to ensure proper refractory properties [34,35] and to promote
gas release during casting. However, these technical constraints allow
a wide-ranging ratio between clay, sand and porosity, as observed in
our study, as well as in all the casting cores studied so far [2,7,9,10].
Therefore, this ratio cannot be used alone neither as a chronological
nor cultural marker.
4.1. Bricks in the cores
Fired bricks were frequently observed as main constituents of the
16th- to 17th-century French cores studied here. From a chemical
point of view, bricks and the plastic material are similar, thus revealing
a possible common provenance of raw materials, namely sand and clay.
However two features prove to be different. First, unlike the plastic

Fig. 3. Volumic ratio (%) of sand, clayey matrix and porosity in plastic core and brick materials of the French and Italian bronze statues.

126

M. Castelle et al. / Microchemical Journal 126 (2016) 121131

Fig. 4. OM crossed-polarized light images (top), OM plain light images (middle) and OCL images (bottom) of a selection of three casting core characterized by three types of grain size
distribution.

cores, bricks do not contain any ber prints. Little, if any, organic temper
has been added. Second, except for Primaticcio's Laocon, the grain size
distribution in the brick and in the plastic material is different in each
statue. This indicates that the plastic material has very probably been
processed during the statue manufacture, within the workshop. The
intrinsic small variations of chemical and petrofabric compositions
observed within one statue tend to show that the core material has
been prepared simultaneously with the lling of the statue mould. At
each new step a freshly prepared clayey material might have been introduced into the mould, thus generating small composition variations. In
contrast, bricks would most probably have been purchased ready-made.
A comparison with bricks used for furnaces may prove interesting, such
as those excavated at the late 16th-century foundry workshop of SaintDenis near Paris [36]. Note, that two types of brick pastes were found in
the statues investigated here. The two bricks sampled in Primaticcio's
Laocon and Barthlmy Prieur's Peace are characterized by Mixture 2.
Interestingly, the corresponding plastic core materials are different.
Whereas the same Mixture 2 has been used for the plastic core of the
Laocon, Mixture 1a has been used for the Peace. Mixture 1c which is
only encountered in the brick from Diana the huntress by Barthlmy
Prieur follows a specic sandyclayey mixture.
It may be worth noticing that the structure of the brick cores refers
more to the sphere of building than to that of ne arts as one would
expect for sculpture. Notably, cores made with bricks are built up rather
than sculpted/shaped. The use of such material and technique may
pertain to the medieval tradition of bell casting [37]. The only other
coeval examples we know of bricks in casting core come from Italian
Renaissance bronze statues [38]. The perpetuation of the use of bricks
in French bronzes in the following late 17th18th centuries has still to
be investigated. Very few cores of large bronze statues have been
studied so far except for four 18th-century statues by Jean-Antoine
Houdon (17411828). There, an important quantity of plaster was
found, together with clay and sand used as a plastic material [39,40].

That said, bricks are still to be found in 17th and 18th-century monumental bronzes such as the equestrian monuments of Louis XIV by
Franois Girardon (16281715) and Louis XV by Edm Bouchardon
(16981762) [41].

4.2. Parisian casting cores

Given the large tolerance on the specicities of clayey and sandy
materials required for the fabrication of cores, the local supply of raw
material is inferred [2,3,5,6]. All the aforementioned petrographic
features are frequently found in casting cores, and in ceramic
materials in general. However, the presence of chert fragments and
glauconite pellets is potentially consistent with a provenance from
Paris and the surrounding area. The foraminifera Miliolidae observed
in the plastic core sample of Diana the huntress further conrms this
provenance. Conversely, sandstone fragments were found in the
Italian cores.
Regarding the source of raw clay, four materials can be considered,
namely the Parisian Stampian and Sparnacian clays, and the Barremian
and Albian clays from Beauvais. These clays were frequently used
for local Ancient [42] and Renaissance ceramics [29]. They are all
kaolinite-rich and contain sand and stoneware, thus providing good
refractory properties. However, despite the fact that the compositions
of those four materials make them good candidates to be used in the
fabrication of cores, none of them match perfectly what was observed
in the French Renaissance cores analyzed in this study. For example,
the K-content in the core materials is always higher than in Sparnacian
reference clays, whereas it is comparable to the levels observed in the
three other clays. Furthermore the Ca-content exhibits very small
random variations between samples from both the French and Italian
casting cores and in the reference clay samples.
These variations may have been generated by several factors.

M. Castelle et al. / Microchemical Journal 126 (2016) 121131

127

Fig. 5. Grain size distribution of core samples from Parisian and surrounding natural clays (the two right columns) and French statues.

First, as far as we know, most of these clay deposits exhibit a

large variety of compositions [29,43]. The measured core composition might possibly correspond to a not-yet investigated geological
layer.

Regarding the K-content variations, the latter cannot constitute

a valid criterion to rule out the Sparnacian clay as a possible source.
Indeed, the added sand-temper contains feldspar and thus might have
increased the K-content in the cores.

128

M. Castelle et al. / Microchemical Journal 126 (2016) 121131

Table 2
Elemental composition of French and Italian casting cores and reference clays (PIXE analysis on bulk sections results in oxide wt.% for major elements and ppm for minor elements).
Major elements (wt.%)

Reference clay samples

Casting core samples

Ref statue/sample
MJAP-S 1773-1
MJAP-S 1773-2
MR 1680
C2RMF 67911
MR 3284
MR 3290
MR 3290_brick
C2RMF 70493_c
C2RMF 70492_c
MR 1681
MR 1682
MR 1683
MR 1683_brick
MR 1784
MR 1685
RF261_brick
RF261
MR 3440
MR 3401
C2RMF 70873_e
OA 11054
OA 11055
WB-38
MR 1668
LL32
MR 3232
MR 3231
S3420
spar.IDF.Mureaux
spar.IDF.Mureaux
stam.IDF.Collegie
spar.IDF.Provins1
spar.IDF.Provins2
spar.IDF.Provins3
spar.IDF.Provins4
haut.BVS.ferriere1
bar.BVS.Landrons
bar.BVS.Landrons
haut.BVS.ferriere2
bar.BVS.Landrons
bar.BVS.Landrons
bar.BVS.Landrons
alb.BVS.Landrons
eoc.Poitiers1
eoc.Poitiers2
eoc.Poitiers3

Minor elements (ppm)

Na2O

MgO

Al2O3

SiO2

P2O5

K2O

CaO

TiO2

1.7
1.3
0.3
0.5
0.3
0.4
0.3
0.4
0.3
0.3
0.2
0.3
0.2
0.4
0.2
0.5
0.3
0.6
0.2
0.4
0.3
0.5
0.2
0.3
0.1
0.4
0.2
0.9
0.0
0.1
0.2
0.1
0.1
0.2
0.0
0.2
0.2
0.2
0.1
0.2
0.1
0.1
0.2
0.1
0.1
0.1

1.9
1.6
0.8
0.5
0.5
0.5
0.5
0.5
0.8
0.6
0.7
0.6
0.7
0.5
0.6
0.6
0.6
0.5
0.3
0.6
0.4
0.5
0.5
0.6
0.4
0.5
0.4
0.5
0.9
1.1
2.3
0.2
0.2
0.3
0.3
0.3
0.3
0.4
0.4
0.3
0.1
0.8
1.2
0.2
0.3
0.2

14.7
13.9
10.8
5.4
6.7
7.0
7.9
7.7
8.9
7.8
7.2
6.5
6.3
7.5
7.3
7.9
7.8
7.3
6.1
7.0
6.3
6.8
6.7
6.7
6.2
7.0
7.1
4.7
27.1
16.2
16.5
19.6
23.4
27.6
34.7
22.8
20.6
29.4
22.4
22.0
14.2
15.7
16.1
33.3
38.9
32.0

69.7
73.1
83.0
82.9
83.9
84.1
85.0
84.4
81.6
84.2
83.7
84.5
84.7
83.4
83.8
82.2
85.2
85.3
87.7
82.0
86.7
85.7
85.7
82.8
87.2
85.5
85.8
87.1
51.3
74.3
67.6
76.8
72.0
68.9
60.4
71.6
74.2
61.6
72.0
71.9
82.6
75.9
71.1
63.0
56.4
56.5

0.2
0.2
0.0
0.1
0.3
0.4
0.1
0.2
0.1
0.1
0.3
0.1
0.2
0.2
0.2
0.3
0.1
0.3
0.1
0.4
0.2
0.0
0.3
0.2
0.1
0.1
0.3
0.0
0.0
0.1
0.0
0.1
0.1
0.1
0.2
0.0
0.1
0.1
0.1
0.1
0.1
0.1
0.1
0.0
0.0
0.0

2.4
2.5
1.7
1.8
1.8
1.9
2.1
2.1
2.2
2.2
2.8
1.9
2.7
2.0
2.2
2.0
1.5
1.2
1.5
2.3
1.6
1.6
1.3
1.5
1.5
1.8
1.4
1.9
0.3
2.6
3.9
0.5
0.0
0.6
0.0
2.1
1.9
2.1
1.5
2.1
0.4
2.4
3.0
0.1
0.3
0.3

3.2
1.4
0.6
5.3
2.9
2.3
0.6
1.2
2.1
0.7
1.2
0.8
1.8
2.2
1.9
2.5
0.9
2.1
0.7
3.6
1.0
0.8
1.4
4.5
1.4
1.1
1.5
1.4
1.1
0.3
1.2
0.2
0.4
0.2
0.7
0.2
0.2
0.3
0.2
0.2
0.1
0.6
2.7
0.2
0.3
0.2

0.9
0.8
0.4
0.8
0.6
0.6
0.5
0.5
0.6
0.5
0.5
0.5
0.5
0.6
0.3
0.8
0.5
0.4
0.5
0.5
0.4
0.7
0.6
0.4
0.3
0.3
0.3
0.6
1.6
1.5
0.9
1.7
1.9
1.3
1.9
1.4
1.1
1.1
1.3
1.3
1.3
1.1
0.9
1.9
1.0
1.1

As for the Ca-content variations, these could be the result of the presence of calcite as observed in some samples (Table 1), or of the addition
of ash [44], horse dung [45] and/or other organic materials. Ash and
horse dung are reported as ingredients for casting core in several
Renaissance treatises [4,46]. Note that ash and horse dung may also
increase the P and also K content. This could also account for the higher
amount of P and K in the cores than in the reference clays.
That being said, the narrow composition observed for all Parisian
casting cores may reasonably point to a unique source of clay, located
in either Paris or Beauvais.
As stressed by Mugnaini et al. [2], the provenance of the sandy silt
part appears to be more difcult to identify since it may result from
the mix of both natural and added sandy temper. However, the relatively high amount of feldspars in French cores tends to rule out the
Soissonian, the Fontainebleau, and the Auvers deposits as sources for
natural sand. Moreover none of the uni-modal distributions that characterize those three sands ts the grain size distribution modes observed
for French casting cores. No other source can be suggested so far. As
for the clay, a unique provenance of sand may be inferred for all the Parisian cores studied, as suggested by the overall narrow composition of
cores.

Fe2O3 Total
5.3
5.1
2.4
2.8
2.9
3.0
3.1
2.9
3.4
3.5
3.5
4.8
3.0
3.3
3.4
3.2
3.0
2.4
2.8
3.2
3.0
3.2
3.5
3.1
2.9
3.3
3.0
2.8
17.6
3.9
7.3
0.7
2.0
0.9
1.7
1.4
1.5
4.8
1.9
1.9
1.1
3.2
4.7
1.2
2.6
9.5

100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100

Cr

Mn

Ni

Ga

Rb

Sr

Zr

u.l.o.d.
41
u.l.o.d.
u.l.o.d.
u.l.o.d.
u.l.o.d.
45
u.l.o.d.
u.l.o.d.
36
51
91
38
u.l.o.d.
38
u.l.o.d.
23
31
u.l.o.d.
31
u.l.o.d.
u.l.o.d.
u.l.o.d.
34
u.l.o.d.
44
40.5
u.l.o.d.
146
333
111
301
392
481
87
271
537
595
85
234
60
83
115
149
74
109

135
144
61
u.l.o.d.
55
65
58
58
75
69
64
74
69
44
66
120
57
30
77
53
79
u.l.o.d.
85
u.l.o.d.
54
47
29
196
135
117
117
213
185
178
95
78
123
131
98
71
90
98
158
155
134

985
1069
21
431
238
234
162
270
289
281
144
232
109
308
99
592
258
198
140
236
309
603
78
126
7
98
108
58
27
360
175
17
5
7
5
50
30
98
19
64
107
143
172
18
33
95

60
59
10
113
22
23
19
17
65
26
22
23
18
30
22
20
19
205
254
21
20
20
u.l.o.d.
23
28
25
36
16
23
61
51
16
25
21
42
17
100
33
28
202
11
65
37
42
49
25

14
16
5
5
8
7
9
8
7
u.l.o.d.
7
10
6
7
u.l.o.d.
9
6
u.l.o.d.
25
7
7
9
14
u.l.o.d.
9
7
5
24
37
34
23
21
27
29
24
29
22
43
29
25
14
20
20
33
34
31

u.l.o.d.
u.l.o.d.
60
64
63
73
73
u.l.o.d.
u.l.o.d.
79
u.l.o.d.

79
83
47
89
63
63
52
46
72
55
46
56
52
66
69
65
45
44
35
82
64
61
103
62
72
54
49
47
317
355
61
176
99
160
676
81
50
102
109
58
37
86
129
26
26
11

52
41
u.l.o.d.
63
31
32
14
34
45
u.l.o.d.
29
39
30
24
u.l.o.d.
19
25
u.l.o.d.
71
13
10
18
u.l.o.d.
u.l.o.d.
25
u.l.o.d.
25
21
30
28
58
21
28
22
38
62
78
59
77
81
26
69
50
32
28
23

173
262
498
222
142
239
223
227
177
190
85
190
114
475
474
252
341
53
90
86
150
184
178
182
46
185
35
626
282
196
175
249
423
195
381
291
200
248
344
775
841
373
376
422
220
218

u.l.o.d.
86
78
76
u.l.o.d.
38
76
68
67
75
57
74
69
72
60
63
na
na
na
34
2
32
2
na
na
na
na
na
na
na
na
na
na
na

Interestingly, the two Donatello statues studied as part of this

research, as well as the Renaissance Italian statues (including several
by Donatello) investigated by Mugnaini et al. [2] exhibit core compositions very different from the Parisian ones. This clearly promotes chemical composition of cores as a powerful tool to discriminate between the
production of different regions, even if it has proved difcult so far to
determine exact provenance.

4.3. Intra-regional considerations

Four modes of grain size distribution were uncovered in this study,
thus revealing four different patterns of core sandyclayey mixtures
within the Parisian corpus, see Table 1 (if one excepts Mixture 1c
which is only observed in the bricks from Prieur's Diana).
Mixture 2 is encountered in more than half of the studied cores (13/
23) and one brick. The corresponding statues or ensembles are attributed
to a variety of sculptors and/or foundry-men active across the time period
considered. For example, Primaticcio's Laocon and Sleeping Ariane of
1542, as well as the 1618 Bordoni Captive were cast using this sandy
clayey mixture. It therefore seems that a common mixture was used

M. Castelle et al. / Microchemical Journal 126 (2016) 121131

129

Fig. 6. Principal Component Analysis on major oxide contents SiO2, Na2O, K2O, CaO, Al2O3, Fe2O3, P2O5 and TiO2 in the casting cores. French statues are in blue, Spiritelli in red, and Italian
bronze statues (right, after Mugnaini [2]) are in green.

by a number of Parisian craftsmen over a long period of time, and therefore in a number of different foundries.
Such uniformity is particularly interesting regarding the sculptor
Francesco Bordoni, whose role as a foundry-man or as a supervisor is
questioned [28]. The similarity of the core composition of the Ewer
(1603), the Torch-bearing Angel (1613) and the Young captive in chains
(1618) corroborates the consistency of casting processes and alloys
previously shown within Bordoni's production [28]: a red brass alloy
is always used, as well as a lasagna-type casting process. This may, in
turn, demonstrate Bordoni's personal involvement in the casting
process. The fact that the Bust of Louis XIII (1643) by Bordoni does not

follow Mixture 2 is remarkable (Table 1). It may suggest Bordoni used

to delegate casting for small items (bust versus life-size statue or
fountain). One should also bear in mind that the bust is a much later
production (1643).
The production of a unique sandyclayey mixture for casting
purposes way thus testify for the existence of a long-living tradition
(at least in the 16th and 17th centuries) in the Parisian bronze sculpture
technical world through. This would somehow corroborate what has
already been suggested for alloys, although in this case, several types
of alloy are coexisting, namely low-alloyed copper, tinbronze, and
zinc-containing bronzes [47]. Hence, core and alloy are not

Fig. 7. Ratio of respectively MgO/Al2O3 versus SrO/Al2O3 and TiO2/Al2O3 versus SrO/Al2O3 for the French core samples and reference clay samples. The French bronze core samples appear
in red without label.

130

M. Castelle et al. / Microchemical Journal 126 (2016) 121131

unequivocally linked. For example, while both artists stick to core Mixture 2, Bordoni works exclusively with zinc-containing bronzes whereas
Primaticio prefers tin bronzes.
Three other sandyclayey mixtures (1a, 1b and 3) have been discovered which are much more marginal. Mixture 1b is encountered only in
Giovan Francesco Rustici's Albert Pie de Savoie (circa 1536), and in the
statuette of Louis XIV crushing the Fronde by Gilles Gurin, the date of
casting of which is debated (18th century?) [48]. Mixtures 1a and 3
are almost exclusively connected with Barthlmy Prieur, as discussed
below.
4.4. Barthlmy Prieur: a particular case
With respect to the wide use of Mixture 2 for the 16th to 17thcentury French bronze statuary casting cores, the sculptor Barthlmy
Prieur appears to be following quite original casting practices. Not
only do the cores used for his statues vary from the main stream, but
the sandyclayey mixture produced for the casting core varies according to the type and/or date of production. For his earliest sculptures
made in France, namely the Peace and the Abundance (1571), and for
the Funerary Genius (15831585), an original Mixture, 1a, has been
used. Interestingly, a commission contract [49] clearly states that the
foundry-man Nicolas Pron would undertake the casting of the Peace
and the Abundance. Is it, therefore, likely that the Funerary Genius, for
which nothing is known about the foundry-man, has been cast in the
same workshop? In any event, the casting techniques and the alloy
are the same [27]. For the late large casting of Prieur, Diana the HuntressI
(1603), the more common Mixture 2 has been used. Does this mean
that it was not cast in the same workshop? We know from archival
documents [49] that Barthlmy Prieur ran a small bronze workshop
in Paris from at least 1582 onwards [27]. Also, Diana the Huntress is
the only large bronze by Prieur to bear the artist's signature in the
form BP (in addition to the two royal statuettes of Henri IV and
Marie de Medicis). When Prieur turned to make bronzes of a smaller
scale in the two royal statuettes, a third core sandyclayey mixture
(Mixture 3) was used. Interestingly, the strong similarity of core pastes
for the two statuettes is mirrored in the similarity and specicity
between the alloy used (brass) and the casting technique (slush wax)
[27]. Was Mixture 3 specic to the commission of the royal pair of
statuettes [27] or was it routinely used for all Prieur's workshop statuettes? Unfortunately, no other core from Prieur's statuettes has been
investigated to date.
The almost exclusive use of Mixture 1a for Prieur's earliest large
bronzes, including the two statues for the Montmorency monument,
raises another important point. The mixture was also used for the
third statue of the monument, the Justice, whose model had been
executed by another sculptor, Martin le Fort (known at Paris between
1560 and 1570). Yet, unlike the two Prieur statuettes, the name of the
foundry-man involved in the casting of the Justice is not known. The
present study strongly suggests that the latter was also cast in Nicolas
Pron's workshop, as corroborated by the consistency of casting
techniques [27]. One feature remains puzzling: the Justice is made of a
very different alloy red brass instead of bronze. Notably, from our
study, the Bust of Jean de Morvilliers by Germain Pilon, cast during the
same period (1578) is the only bronze not designed by Prieur together
with the Justice for which Mixture 1a was used (the other exception
being the much later Bust of Louis XIII by Bordoni of 1643). One may
thus suggest that the Bust of Jean de Morvilliers might also have been
cast by Nicolas Pron's workshop. Interestingly, the alloy used for this
bust is quite similar to that of the Justice, being a ternary copperzinc
tin alloy. This raises the question as to whether Nicolas Pron used to
work with such ternary alloys. Only Prieur, because of his skills as
foundry-man he is often designated by scholars as a sculptor/
foundry-man and thanks to his renown, would have been able to
impose another alloy, a tinbronze. This may be the reason why
the commission contract states that Prieur should provide Nicolas

Pron with the metal. Such supremacy of sculptorfounders over

foundry-men regarding the choice of alloy might not have been an
exception during Renaissance, as shown for the Netherlandish sculptor,
Adrian de Vries (15561626) [14]. Contracts vary also in Italy with
sometimes the metal provided and sometimes the sculptor or founder
gets it.
5. Conclusion
Parisian bronze statues produced during the late 16th century to the
beginning of 17th century show a particular casting-core composition,
which may enable us to distinguish them from other contemporaneous
productions, notably those of Italy. Yet, although a limited number of
sources could be identied, more accurate identication of the provenance of the materials would require larger sampling and complementary investigation including rare earth analysis [2,17].
Moreover, the new methodology set out above has allowed us to
shed light on the use of a particular sandyclayey mixture for casting
cores (Mixture 2), which was adopted for the majority of the French
statues studied to date. Hence, while core composition may point with
a greater degree of condence to Parisian production, it does not enable
us to distinguish between different sculptors or workshops, as is also
generally the case for alloy composition and casting technique when
considered in isolation. Thus, in order to attempt to resolve those issues,
all the characteristics of a bronze statue must be taken into consideration, together with information provided by any archival documents
associated to a specic commission. However, despite the predominance of Mixture 2 and the marginal aspect of Mixtures 1b and 1c,
two other recipes point to two specic areas of production. On the one
hand, Mixture 1a may be linked to one specic foundry-man, namely
Nicolas Pron. On the other hand, Mixture 3 may have been used exclusively by one sculptor/foundry-man for specic productions, namely
Barthlmy Prieur, when producing statuettes in his own workshop.
Although these trends require more studies to be asserted with certainty, the present work demonstrates the efciency of the investigation of
casting core materials in helping to document production methods of
bronze sculpture. It has been possible in this paper to tackle the
relationships between founders and sculptors, such as Nicolas Pron
and Barthlemy Prieur, providing fresh insight into a fascinating period
in art history, when bronze sculpture was reintroduced as a major art
into French modern period France.
Acknowledgments
We wish to thank the Patrima Foundation for founding the PhD
project that enabled us to carry out the present study on casting cores.
We gratefully acknowledge the museum curators that have supported
us in this project by allowing us access to the bronze statues: Sophie
Jugie (Muse du Louvre), Genevive Bresc-Bautier (Muse du Louvre),
Vincent Droguet (Chteau de Fontainebleau), Valrie Carpentier
(Chteau de Fontainebleau), Nicolas Sainte Fare Garnot (Muse
Jacquemart-Andr) and Florian Meunier (Muse Carnavalet). Many
thanks go to our colleagues at the C2RMF, Thomas Calligaro and
Nathalie Gandolfo, for helping at different stages of this work. We
warmly acknowledge the three anonymous reviewers for their remarks
as well as Peta Motture (Victoria and Albert Museum) for the English
editing and fruitful suggestions.
References
[1] D. Bourgarit, J. Bassett, F.G. Bewer, G. Bresc-Bautier, P. Malgouyres, French Bronze
Sculpture: Materials and Techniques 16th18th century, Archetype, Paris, 2014.
[2] S. Mugnaini, M. Giamello, A. Pisani, S. Siano, Casting cores used to craft large bronze
masterpieces of the Florentine Renaissance and Mannerism, J. Archaeol. Sci. 47
(2014) 8598.
[3] C.L. Reedy, Petrographic analysis of casting core materials for provenancing studies
of copper alloy sculptures, Archaeomaterials 5 (1991) 121163.

M. Castelle et al. / Microchemical Journal 126 (2016) 121131

[4] V. Biringuccio, De la pirotechnia, Cambridge MA and London, MIT Press, 1966
(Venice, 1540).
[5] G. Lombardi, The casting core composition and provenance of the Goljamata
Kosmatka (Bulgaria) bronze head, J. Archaeol. Sci. 36 (2009) 520527.
[6] G. Lombardi, M. Vidale, From the shell to its content: the casting cores of the two
bronze statues from Riace (Calabria, Italy), J. Archaeol. Sci. 25 (1998) 10551066.
[7] B. Vincent, Searching for the Bronze Workshops of Angkorian Cambodia
Petrographic Study Applied to Casting Cores of 11th12th Post-doctoral
Fellowship, Smithsonian Institution History of Art Department of Conservation
and Scientic Research, Washington, 2014.
[8] F. Caro, Petrographic and mineralogical analysis of the casting core of a Chinese
bronze Buddha Maitreya, Metrop. Mus. J. 1 (2010) 155161.
[9] B.M. Weisman, C.L. Reedy, Technical studies on Renaissance bronzes, in: P.B.
Vandiver, M. Goodway, J.L. Mass (Eds.), Materials Issues in Art and Archaeology
VI: Symposium Held November 2630, 2001, Boston, Massachusetts, USA, Materials
Research Society, Warrendale, PA 2002, pp. 483495.
[10] B.M. Weisman, C.L. Reedy, C. Sussman, Technical Analysis of Renaissance Bronzes for
Provenance Studies: Pilot Project, Getty Conservation Institute, Malibu, CA, 1988.
[11] R.E. Stone, Antico and the development of bronze casting in Italy at the end of the
Quattrocento, Metrop. Mus. J. 16 (1981) 87116.
[12] A. Amarger, L. Lefevre, Etude de l'Hercule de Grenoble, 1990.
[13] B. Mille, I. Gajda, F. Demange, C. Pariselle, Y. Coquinot, E. Porto, O. Tavoso, A. Zink,
Hawtar'athat, ls de Radaw'il du lignage de Shalalum, une grande statue en bronze
du royaume de Saba' (Ymen), Monument et mmoire de la fondation Eugne Piot,
Acadmie des Inscriptions et Belles-Lettres, 892010 568.
[14] F. Bewer, The sculpture of Adriaen de Vries: a technical study, in: P. Debra (Ed.),
Small Bronzes in the Renaissance, National Gallery of Art and the Center for Advanced Study in the Visual Arts, Washington and New Haven, 2001, pp. 158193.
[15] R.C. Schmidtling, Core analysis, in: J. Bassett (Ed.), The Craftsman Revealed: Adriaen
de Vries, Sculptor in Bronze, Getty Conservation Institute, Los Angeles 2008,
pp. 3544.
[16] G. Lombardi, A petrographic study of the casting core of the Lupa Capitolina bronze
sculpture (Rome, Italy) and identication of its provenance, Archaeometry 44
(2002) 601612.
[17] Y. Waksman, Etudes de provenance de cramiques, in: P. Dillmann, L. Bellot-Gurlet
(Eds.), Circulation et provenance des matriaux dans les socits anciennes, Editions
des archives contemporaines, Paris 2014, pp. 195216.
[18] Y. Lon, La circulation des savoir-faire techniques des cramiques, in: P. Dillmann, L.
Bellot-Gurlet (Eds.), Circulation et provenance es matriaux dans les socits
anciennes, Editions des archives contemporaines, Paris 2014, pp. 195216.
[19] L.L. Holmes, G. Harbotte, Provenance study of cores from Chinese bronze vessels,
Archeomaterials 5 (1991) 165184.
[20] R. Newmann, Analysis of core and investment samples from some aquamanilia, in:
P. Barnet, P. Dandridge (Eds.), Lions, Dragons, and Other Beasts: Aquamanilia of the
Middle Ages, Vessels for Church and Table, Exh. cat. New York and New Haven, Bard
Graduate Center for Studies in the Decorative Arts, Design, and Culture in association with Yale University Press, New Haven and London, 2006, pp. 5763.
[21] C.L. Reedy, Review of digital image analysis of petrographic thin sections in
conservation research, J. Am. Inst. Conserv. 45 (2006) 127146.
[22] J. Riederer, Thin section microscopy applied to the study of archaeological ceramics,
Hyperne Interact. 154 (2004) 143158.
[23] P.C. Livingood, A.S. Cordell, Point/counter point: the accuracy and feasibility of
digital image techniques in the analysis of ceramic thin sections, J. Archaeol. Sci.
36 (2009) 867872.
[24] B. Velde, Use of image analysis in determining multi-source ceramic materials, in:
A.L. Smith, D. Bosquet, R. Martineau (Eds.), Pottery Manufacturing Processes: Reconstitution and Interpretation, BAR international series Universit de Lige, Belgique,
2005.
[25] G. Dal Sasso, L. Maritan, S. Salvatori, C. Mazzoli, G. Artioli, Discriminating pottery
production by image analysis: a case study of Mesolithic and Neolithic pottery
from Al Khiday (Khartoum, Sudan), J. Archaeol. Sci. 46 (2014) 125143.
[26] D.L. Bish, J.E. Post, Modern powder diffraction, reviews in mineralogy, Mineral. Soc.
Am. 20 (1989).
[27] R. Seelig-Teuwen, D. Bourgarit, F.G. Bewer, Barthlmy Prieur fondeur, son atelier,
ses mthodes de travail, in: D. Bourgarit, Bassett, F.G. Bewer, G. Bresc-Bautier, P.
Malgouyres (Eds.), French Bronze Sculpture: Materials and Techniques 16th18th
Century, Archetype, Paris 2014, pp. 1837.

131

[28] D. Bourgarit, F.G. Bewer, G. Bresc-Bautier, Francesco Bordoni: spcicits techniques

chez un fondeur sculpteur du XVIIme sicle, in: D. Bourgarit, J. Bassett, F.G. Bewer, G.
Bresc-Bautier, P. Malgouyres (Eds.), French Bronze Sculpture: Materials and
Techniques 16th18th Century, Archetype, Paris 2014, pp. 117.
[29] F. Ravoire, A. Bouquillon, La poterie de terre de la Renaissance en le de France
(XVeXVIe sicle): production et diffusion, Techn, no 20, 2004.
[30] T. Calligaro, J.-C. Dran, E. Ioannidou, B. Moignard, L. Pichon, J. Salomon, Development
of an external beam nuclear microprobe on the AGLAE facility of the Louvre
Museum, Nucl. Inst. Methods 161163 (2000) 328333.
[31] J.-C. Dran, T. Calligaro, J. Salomon, Particle-induced X-ray emission, Modern Analytical
Methods in Art and Archaeology, John Wiley, Chichester, 2000 135166.
[32] M. Schweissing, B. Velde, I. Druc, Archaeological ceramic materials, Origin and
Utilization. , E. Schweizerbart'sche Verlagsbuchhandlung, 2000.
[33] L. Moraru, F. Szendrei, Structural features of archaeological pottery, J. Eng. Stud. Res.
17 (2011) 7378.
[34] M.S. Tite, V. Kilikoglou, G. Vekinis, Strength, toughness and thermal shock resistance
of ancient ceramics, and their inuence on technological choice, Archaeometry 43
(2001) 301324.
[35] P.T. Craddock, Refractories with a purpose II: ceramics for casting, The Old Potter's
Almanack, Vol 192014.
[36] O. Meyer, N. Thomas, M. Wyss, Tmoins archologiques d'un atelier de bronzier
fonctionnant Saint Denis la n du XVIe sicle, in: D. Bourgarit, Bassett, F.G.
Bewer, G. Bresc-Bautier, P. Malgouyres (Eds.), French Bronze Sculpture: Materials
and Techniques 16th18th Century, Archetype, Paris 2014, pp. 215224.
[37] M. Castelle, D. Bourgarit, F. Bewer, The lasagna technique for casting large 16th c.
French bronzes: looking for origins, in: J. Plumier, N. Thomas (Eds.),Medieval Copper, Bronze and Brass Dinant-Namur 2014: History, Archaeology and
Archaeometry of the Production of Brass, Bronze and Other Copper Alloy Objects
in Medieval Europe (12th16th centuries), 2016 (to be published).
[38] L. Morigi, Considerations on the Use of Bricks for Renaissance Italian Bronze Statue,
2015 (Personal communication).
[39] J. Bassett, Technical Examination of the Apollo by Houdon, CALOUSTE Gulbekina
Museum (inv 552), J Paul Getty Museum, Decorative arts and sculpture conservation,
Los Angeles, 2005.
[40] J. Bassett, G. Scherf, Jean-Antoine Houdon: sculptor and founder, in: D. Bourgarit,
Bassett, F.G. Bewer, G. Bresc-Bautier, P. Malgouyres (Eds.), French Bronze Sculpture:
Materials and Techniques 16th18th Century, Archetype, Paris 2014, pp. 107124.
[41] A.-L. Desmas, Boffrand's and Mariette's Descriptions of the casting of Louis XIV and
Louis XV on Horseback, in: D. Bourgarit, J. Bassett, G. Bresc-bautier, P. Malgouyres,
G. Scherf (Eds.), French Bronze Sculpture Materials and Techniques 16th18th
Century, Archetype, Muse du Louvre, Centre de Recherche et de Restauration des
Muses de France, Londres 2014, pp. 234249.
[42] S. Dubois, B. Mille, E. Binet, La cramique pate blanche a quartz: contribution
l'tude d'un facies rgional, in: M. Truffeau-Libre, A. Jacques (Eds.),La cramique
du Haut-Empire en Gaule Belgique et dans les rgions voisines: facis et courants
commerciaux; Actes de la table ronde d'Arras, Nord Ouest Archologie Berck sur
mer 1994, pp. 4446.
[43] M.P. Aubry, M. Thiry, C. Dupuis, W.A. Berggren, The Sparnacian deposits of the Paris
Basin: Part I A lithostratigraphic classication, Stratigraphy 2 (2005) 65100.
[44] P. Colomban, A. Tourni, Les cendres vgtales, matires premires des verres et des
maux: un exemple, Les cendres de vgtaux utilises par Fr. D de Montmolin2010.
[45] C. Galibardy, C. Loustale, E. Arribarrouy, Etude de caractrisation des fumiers de
cheval issus de centres questres an daider la dcision sur les possibilits de
valorisation, APESA, 2009.
[46] A. Flibien, Des Principes de l'Architecture, de la Sculpture, de la Peinture et des
autres Arts qui en dpendent avec un Dictionnaire des termes propres chacun
de ces Arts, 1690.
[47] D. Bourgarit, J. Bassett, F. Bewer, Synthesis on copper alloys used for the casting of
bronze statues during the French 16th and the 17th c., Unpublished work, 2014.
[48] G. Bresc-Bautier, G. Scherf (Eds.), Gilles Gurin, Louis XIV crasant la Fronde, cat 53,
Bronzes franais de la Renaissance au Sicle des lumires, Muse du Louvre
Somogy, Paris, 2008.
[49] C. Grodecki, Documents du Minutier central des notaires de Paris, Histoire de l'art au
xvie sicle (15401600). Sculpture, peinture et gravure, broderie, faence et
maillerie, orfvrerie et joaillerie, armes et armures, tome II, Archives nationales,
1986.