Bulletin de la Société préhistorique française
Tome 111, numéro 2, avril-juin 2014, p. 325-338
Functions and contexts of Bronze Age
metal casting moulds from Poland
Justyna Baron, Beata Miazga and Kamil Nowak
T&
N F
ot P
fo roo
rd f
is
tri
bu
tio
n
Abstract: Metal casting moulds, designed mostly to produce only one type of artefact, are a fascinating group of artefacts that have been recorded in many parts of Bronze Age Europe. Despite some examples of moulds for ornaments,
the use of metal moulds is strongly connected with axe production. Moreover, most of the moulds were deposited
deliberately in bogs and rivers in a similar way as axes. A recent ind of such a mould from Gaj Oławski in southwestern Poland inspired us to bring together a small collection of bronze casting moulds from Poland. In our study,
we focus on the contexts and possible functions of metal casting moulds with a particular focus on the recent ind.
Metal casting moulds dated back to the Bronze Age are rare in Poland with only ive examples of such inds known
to date. Such moulds were discovered at Pawłowiczki, Brzeg Głogowski, Kiełpino and Rosko. Their distribution is
random even if the sites occur only in western Poland they are too few to speak of any pattern of deposition (ig. 1).
A complete, two-piece item from Pawłowiczki — probably part of a deposit — was found in 1869; however, the precise
location is unknown. Based on the axe’s characteristics, the mould can be dated to the 5th period of the Bronze Age
(i.e. ca. 900 – 700 BC). At Brzeg Głogowski, a complete two-piece mould was found, however no details of its discovery
are known. Together with the mould, a complete socketed axe was found which had been cast in this mould (ig. 4d). As
at Pawłowiczki, this mould can be dated to the 5th period of the Bronze Age (i.e. ca. 900 – 700 BC). The bog hoard from
Kiełpino was discovered in 1884, and among a number of bronze ornaments and elements of horse gear, an iron knife
and a lump of iron ore, there were also two complete two-piece moulds and two socketed axes that itted them. The
deposit from Rosko consists mostly of bronze socketed axes decorated with vertical ribs (67 items), and two, two-piece
bronze casting moulds (igs. 5, 6), which makes the deposit an excellent example of selected deposition. The deposit
contained mostly axes that were typical for large parts of south-western Poland and may be dated to ca. 900 – 700 BC.
According to an opinion that is common in Polish sources, metal casting moulds, due to their low resistance to high
temperatures, were used exclusively for making wax or lead models and not bronze products themselves. The archaeometric investigation’s aim was to identify both the chemical composition of the mould (including the alloy’s main elements and the molecular composition of the corrosion layers), and the original purpose of the item. There are two types
of data obtained for the mould from Gaj Oławski. One group refers to the mineralogical composition of the item and
proved the relative low tin content of it. The result corresponds with the opinion that the axes — regarding the alloys and
casting techniques — display relatively lower quality. This also its the view that the low tin content may indicate better
resistance against high temperatures and direct metal casting. The other group of data refer to the black residue from
the mould which is ancient beeswax remains. The analysed specimen yielded traits that correspond well with identiied
archaeological samples of beeswax. The discussed mould from Gaj Oławski displays properties allowing it both to be
used to cast metal (relected in the low tin content) and wax (beeswax remains), which makes it a good example of a
multi-purpose tool used in various conditions. On the other hand, the wax can represent the remnants of coating used
as a layer preventing the welding of mould with poured liquid metal, which would mean that the mould was covered
with coating but was not ired.
Considering the context and the interpretation, the discussion on bronze casting moulds is strongly connected with
the discussion on bronze deposits in general. The known locations allow us to consider the bronze casting moulds as
objects of deliberate, and usually selective deposition, mostly in what can be described as ‘offering contexts’, including
river, bog and hilltop inds. In his detailed study on depositional patterns and contexts, D. Fontijn argues that axes,
sickles and weapons were kept away from graves, deposited somewhere else, mostly in what may be broadly understood as ‘wet areas’ (Fontijn, 2002, tables 10.1 and 10.2). Although his study concerns the Netherlands, it seems to it
the situation from other areas as well. Similar conclusions with a particular focus on river and bog inds in Poland were
presented by J. Woźny (1996). He argues that the composition of water deposits in the Bronze and Early Iron Age in
Poland relects a clear opposition between weapons/tools and ornaments (Woźny, 1996, p. 89).
Recent publications on bronze deposits in Poland relect a move away from purely economic interpretations to more
complex views including signiicant examples of ‘votive offerings’ (Blajer, 2001, pp. 298 – 299), with many examples
connected with bog and river contexts (e.g., Woźny, 1996 and 1997). Among ive Polish sites with the moulds, three
Justyna Baron, Beata Miazga and Kamil Nowak
were bog deposits (Kiełpino, Gaj Oławski, and indirectly Rosko) while for the other two, no information is available
and thus similar contexts cannot be excluded. The three sites with known water contexts are located in a lowland area
where the number of tools deposited in water increase towards the end of the Bronze Age. When the contexts of the
moulds and axes are compared, it is clear that both were given similar signiicance, i.e. the moulds were treated like
tools (axes) and deposited in the same way.
Keywords: casting mould, Bronze Age, deposit, Poland, beeswax.
T&
N F
ot P
fo roo
rd f
is
tri
bu
tio
n
Résumé : Les moules en métal, élaborés essentiellement en vue de produire un seul type d’artefact, constituent un
groupe d’objets fascinant qui a été documenté dans beaucoup de régions de l’Europe à l’âge du Bronze. Bien qu’il existe
des exemples de moules ayant été destinés à la fabrication de parures, l’usage de moules métalliques est très fortement
lié à la production de haches. De plus, la plupart des moules ont été déposés intentionnellement dans des tourbières et
des cours d’eau, d’une manière similaire aux haches. La découverte récente d’un moule de ce type, à Gaj Oławski dans
le Sud-Ouest de la Pologne, nous a amenés à rassembler la petite collection de moules métalliques de Pologne. Cette
étude est centrée sur les contextes et les fonctions possibles des moules en métal avec un focus particulier sur la récente
découverte. Les moules métalliques datés de l’âge du Bronze sont rares en Pologne avec seulement cinq exemplaires
connus à ce jour. Ces moules ont été découverts à Pawłowiczki, Brzeg Głogowski, Kiełpino et Rosko. Leur répartition
est aléatoire. Même si les découvertes sont localisées dans l’Ouest de la Pologne, elles sont trop rares pour discuter d’un
schéma de déposition (ig. 1). Un exemplaire complet, avec ses deux valves, a été découvert à Pawłowiczki en 1869 et
faisait préalablement partie d’un dépôt ; sa localisation précise est toutefois inconnue. D’après les caractéristiques de la
hache, le moule peut être daté de la cinquième période de l’âge du Bronze (c’est-à-dire entre 900 et 700 avant notre ère).
À Brzeg Głogowski, un moule complet, bivalve lui aussi, a été découverte mais aucune information sur le contexte de
découverte n’a été conservée. Associée à ce moule, a été découvert une hache à douille complète, qui a été fabriquée
dans ce même moule (ig. 4d). Comme à Pawłowiczki, ce moule peut être daté de la cinquième période de l’âge du
Bronze (vers 900-700 avant J.-C.). Le dépôt en tourbière de Kiełpino a été découvert en 1884. Parmi de nombreux éléments de parure de bronze et de pièces de harnachement de cheval, un couteau en fer et un morceau de minerai de fer,
il y avait aussi deux moules complets, tous les deux bivalves, et deux haches à douille qui leur correspondaient exactement. Le dépôt de Rosko est principalement composé de haches à douille en bronze décorées de nervures verticales et
deux moules bivalves (ig. 5 et 6), ce qui montre très bien son caractère sélectif. Aucune des haches ne correspond aux
moules, qui ont du coup été interprétés comme ayant été utilisés pour la production de modèles en cire. Sur l’une des
pièces, l’empreinte d’une étoffe tissée a été identiiée et a été publiée par ailleurs (Sikorski, 2006). Le dépôt contenait
pour l’essentiel des haches typiques d’une grande partie du Sud-Ouest de la Pologne et peuvent être datées d’environ
900-700 avant notre ère.
Selon une opinion assez usuelle dans la littérature archéologique polonaise, les moules en bronze, en raison de leur
faible résistance aux hautes températures, étaient utilisés exclusivement pour produire des modèles en cire ou en plomb
et non pas pour fabriquer directement des produits en bronze. Le but de l’analyse archéométrique était d’identiier à la
fois la composition chimique du moule (ce qui inclut les éléments principaux de l’alliage et la composition moléculaire
des niveaux de corrosion), et la inalité originelle de l’objet. Deux types de données ont été obtenus pour le moule de
Gaj Oławski. Les premières correspondent à sa composition minéralogique et montrent un niveau d’étain relativement
bas. Ce résultat conforte l’opinion que les haches – du point de vue des alliages et des techniques – présentent une
qualité relativement basse. Ceci concorde également avec le point de vue selon lequel la faible proportion d’étain pourrait indiquer une meilleure résistance aux hautes températures et à la fusion directe du métal. Le deuxième groupe de
données concerne le résidu noir dans le moule, qui s’avère être un reste ancien de cire d’abeille. Le spécimen analysé
présentait des caractéristiques qui correspondent bien à des exemplaires archéologiques de cire d’abeille précédemment
identiiés. Le moule de Gaj Oławski présente des propriétés lui permettant d’être utilisé aussi bien pour la fonte du
métal (ce qui est relété par le faible niveau d’étain) que pour la mise en forme de la cire (restes de cire d’abeille), ce
qui en ferait un bon exemple d’outil multifonctions. D’un autre côté, la cire peut représenter les restes de l’enduit utilisé
comme une couche intermédiaire empêchant la fusion du moule lors du versement du métal liquide, ce qui voudrait dire
que le moule était couvert par l’enduit mais non chauffé.
Étant donné le contexte et son interprétation, la discussion sur les moules métalliques est fortement liée à celle sur les
dépôts de bronze en général. Les lieux de découverte connus nous permettent de considérer les moules métalliques
comme des objets ayant fait l’objet d’un dépôt intentionnel et généralement sélectif, principalement au sein de ce qui
peut être décrit comme des contextes d’offrande, incluant les découvertes en cours d’eau, tourbières et sites de hauteur.
Dans son étude détaillée sur les modalités et les contextes de dépôt, D. Fontijn argumente que les haches, les faucilles
et les armes étaient exclues des sépultures et déposées ailleurs, principalement au sein de ce qui peut être globalement
considéré comme des zones humides (Fontijn, 2002, tableaux 10.1; 10.2). Bien que cette étude concerne les Pays-Bas,
il semble qu’elle concorde avec la situation d’autres régions. Des conclusions similaires, avec un focus particulier
sur les découvertes en cours d’eau et tourbières en Pologne, ont été présentées par J. Woźny (1996). Il montre que la
composition des dépôts dans les eaux de l’âge du Bronze et du premier âge du Fer relète une opposition nette entre les
armes et les outils d’un côté et la parure de l’autre (Woźny, 1996, p. 89).
Les publications récentes sur les dépôts de l’âge du Bronze en Pologne relètent un abandon des interprétations purement
économiques et un transfert vers des visions plus complexes incluant des exemples signiicatifs d’offrandes rituelles (Blajer, 2001, p. 298-299), avec de nombreux exemples en lien avec des contextes en tourbières et cours d’eau (par exemple
Woźny, 1996 et 1997). Des cinq sites polonais où ont été découverts des moules, trois étaient des dépôts en tourbière
326
Bulletin de la Société préhistorique française
Functions and contexts of Bronze Age metal casting moulds from Poland
(Kiełpino, Gaj Oławski et indirectement Rosko) alors que pour les deux autres, aucune information n’est disponible et des
contextes semblables ne peuvent ainsi pas être exclus. Les trois sites à contexte humide connus sont localisés dans une
zone basse où le nombre d’outils déposés dans les eaux augmentent à la in de l’âge du Bronze. Lorsque l’on compare les
contextes des moules et des haches, il est clair que les deux ont eu la même signiication, c’est-à-dire que les moules ont
été traités comme les outils (les haches) et déposés de la même manière [traduction Karim Gernigon].
Mots-clefs : moule métallique, âge du Bronze, dépôt, Pologne, cire d’abeille.
M
beginning of the 20th century and two are considered to
have been lost during the Second World War.
THE MOULDS
D
uring an archaeological search for the remains of a
Second World War aircraft in the autumn of 2012,
which took place north of Oława — a town about 30 km
south-east of Wrocław in south-western Poland (ig. 1),
one of the archaeologists was given a half of a metal
T&
N F
ot P
fo roo
rd f
is
tri
bu
tio
n
etal casting moulds are a fascinating group
of artefacts that have been recorded in many
areas of Bronze Age Europe, and were designed mostly to produce only one type of artefact. Despite
some examples of moulds for ornaments, in Poland, the
use of metal moulds is strongly connected with axes production. Moreover, most of the moulds were deposited
deliberately in bogs and rivers in a similar way as axes.
A recent ind of such a mould from Gaj Oławski in
south-western Poland inspired us to bring together a
small collection of bronze casting moulds from Poland
from ive locations. Three of them were discovered at the
mètres
2000
1500
1000
500
200
100
0
Kiełpino
Rosko
Brzeg Głogowski
Gaj Oławski
Pawłowiczki
0
50
100 km
Fond Martin Sauvage
d’après GEOATLAS
Fig. 1 – Distribution of Bronze Age metal casting moulds and axes of Kowalewko type in Poland. Black spots are the axes (large ones
stand for three or more objects found together), white spots are the moulds and green star is the mould from Gaj Oławski (after Kuśnierz
1988, plate 43, modiied by J. Baron).
Fig. 1 – Répartition des moules métalliques de l’âge du Bronze et des haches de type Kowalewko en Pologne. Les points noirs représentent les haches (les plus grands représentent une découverte de trois objets ou plus), les points blancs correspondent aux moules,
l’étoile verte désigne le moule de Gaj Oławski (d’après Kuśnierz, 1988, pl. 43, modiié par J. Baron).
Tome 111, numéro 2, avril-juin 2014, p. 325-338.
327
Justyna Baron, Beata Miazga and Kamil Nowak
of deposition (ig. 1). In the next paragraphs, the moulds
from the other sites are briely presented.
Pawłowiczki, Kędzierzyn-Koźle district
A complete, two-piece item from Pawłowiczki — probably part of a deposit — was found in 1869; however,
the precise location is unknown. Both pieces were of
similar size: 12.7 cm long, 5 cm wide and 1.8 thick and
were designed to cast socketed axes decorated with three
vertical grooves (ig. 4a). The two elements were joined
together with three projections — two situated at the longer edges of the mould and one at the blade of the axe.
The axe cast in this mould could have measured 9.3 cm in
maximum length with a maximum blade width of 3.3 cm.
Despite the ornament, which is typical for these artefacts,
H. Seger considered it a rare type due to the low loop
location (Seger, 1909, p. 24). Based on the axe’s characteristics, the mould can be dated to the 5th period of the
Bronze Age (i.e. ca. 900 – 700 BC).
Brzeg Głogowski, Głogów district
T&
N F
ot P
fo roo
rd f
is
tri
bu
tio
n
casting mould. According to information provided by the
founder, the artefact was discovered in the course of a
surface survey in 2012, north-west Gaj Oławski village
in the Oława district, at the eastern edge of a ravine of
a small, nameless stream. The whole area is now relatively lat and wet with a number of small streams both of
natural and artiicial origin. No archaeological location
was previously recorded there, although some traces of
prehistoric settlements are known in the vicinity.
The paper aims to present several bronze casting
moulds with a particular focus on the mould discovered
recently at Gaj Oławski in south-western Poland. This
category of artefacts is rare, both in Polish and in Central
European contexts. To date, only six complete two-piece
moulds from ive Polish locations are known (ig. 1).
The discussed artefact is a half of a casting mould
designed for the production of socketed axes. This is an
example of a durable, two-piece wedge-shaped mould.
The surface is covered with a green patina and number of
corrosion marks (ig. 2). The mould measures 17.5 × 7.2
× 4 cm, while the axe was designed to be 15.6 cm long
with a socket diameter of 3.5 cm and a maximum blade
width of 4 cm (ig. 3). The weight of the item was 629.3 g.
On the inner side, at the negative edge, four pegs were
evident, which most likely itted into four holes in the lost
part of the mould. The pegs were used to join and stabilize
both elements during the casting. Also on the inner side,
two heavily corroded knobs of unknown function were
observed. Perhaps they helped with tying together both elements of the mould or were used to open the still hot mould
after the item had been cast. Such a function is sometimes
given to outer loops, which often occur on moulds found in
Western Europe (e.g., Armbruster, 2000, p. 41).
The negative proves that the mould was used to cast
socketed axes of the so-called ‘Lusatian type’ with a loop,
and an oval socket in cross section. The part between the
cylindrical socket and the wedge-shaped axe blade is one
of the typical characteristic of such axes (Kuśnierz, 1998,
p. 25). This type is the most frequent ones of socketed
axes known from Poland (Kuśnierz, 1998, p. 25). Due
to the axe’s size and its decoration, it may be identiied
as Kowalewko type, variant B. Such axes often occur
in deposits, and less frequently in graves and are dated
to the 4th period of the Bronze Age (ca. 1100 – 900 BC)
(Kuśnierz, 1998, p. 30). They occur mostly in Lower
Silesia (south-western Poland), with some examples from
West Pomerania and western Great Poland (west Poland).
A few were also found in Brandenburg, Bohemia and
Lusatia (Kuśnierz, 1998, p. 31).
OTHER BRONZE MOULDS FROM POLAND
M
etal casting moulds are very rare in Poland with
only ive examples of such inds known to date.
Their distribution is random even if the sites occur only
in western Poland they are too few to speak of any pattern
328
At Brzeg Głogowski, a complete two-piece mould was
found, however no details of its discovery are known
(Seger, 1927, p. 62). Both pieces were 13.8 cm long,
4.2 cm wide while their thickness is unknown (igs. 4b-c).
The mould elements were joined together with ive projections — four situated along the mould edge and one
at the blade. Together with the mould, a complete socketed axe was found — 10.8 cm long and 2.8 cm wide
with a socket outer diameter of 3 cm, which had been
cast in this mould (ig. 4d). As at Pawłowiczki, this
mould can be dated to the 5th period of the Bronze Age
(i.e. ca. 900 – 700 BC).
Kiełpino, Gryfice district
The bog hoard from Kiełpino was discovered in 1884, and
among a number of bronze ornaments and elements of horse
gear, an iron knife and a lump of iron ore, there were also two
complete two-piece moulds and two socketed axes that itted
them (Ebert, 1926, p. 14 – 15, plate 10). Both the moulds and
the axes were much smaller than the other artefacts presented in this paper. One of the moulds was 5.6 – 5.8 cm long
and 3.2 – 3.5 cm wide at the blade part (ig. 4e). The second
one was smaller: 4.9 – 5.1 cm long and 2.7 cm wide (ig. 4g).
The axes were also small — one was 4.4 cm long, 3 cm wide
with a socket outer diameter of 2.2 cm (ig. 4f) and the other
one was only 3.6 cm long, 2.5 cm wide and its socket outer
diameter was 1.9 cm (ig. 4h).
Unlike the other moulds, the moulds from Kiełpino
were very thin and precisely relect the shape of the cast
axes including the side loops. They were also provided with small handles (ca. 1.5 cm long in both cases)
which make them similar to inds from Western Europe
(e.g., Cordier, 1962, ig. 1; Butler and Streegsta, 20052006, p. 207–240, ig. 107). The deposit is dated to the
transitional period between the Bronze and the Iron Age
Bulletin de la Société préhistorique française
T&
N F
ot P
fo roo
rd f
is
tri
bu
tio
n
Functions and contexts of Bronze Age metal casting moulds from Poland
Fig. 2 – Metal casting mould from Gaj Oławski in south-western Poland (photo by Ł. Kapa).
Fig. 2 – Moule métallique de Gaj Oławski, Sud-Ouest de la Pologne (cliché Ł. Kapa).
Tome 111, numéro 2, avril-juin 2014, p. 325-338.
329
T&
N F
ot P
fo roo
rd f
is
tri
bu
tio
n
Justyna Baron, Beata Miazga and Kamil Nowak
Fig. 3 – Metal casting mould from Gaj Oławski in south-western Poland (drawning N. Lenkow).
Fig. 3 – Moule métallique de Gaj Oławski, Sud-Ouest de la Pologne (dessin N. Lenkow).
330
Bulletin de la Société préhistorique française
T&
N F
ot P
fo roo
rd f
is
tri
bu
tio
n
Functions and contexts of Bronze Age metal casting moulds from Poland
Fig. 4 – Metal casting moulds from Poland. a: Pawłowiczki; b – d: Brzeg Głogowski; e – h: Kiełpino (after Seger, 1909 and 1927; Ebert
1926; redrawn by J. Baron).
Fig. 4 – Moules métalliques de Pologne. a : Pawłowiczki ; b-d : Brzeg Głogowski ; e-h : Kiełpino (d’après Seger, 1909 et 1927 ; Ebert,
1926 ; redessinés par J. Baron).
Tome 111, numéro 2, avril-juin 2014, p. 325-338.
331
Justyna Baron, Beata Miazga and Kamil Nowak
(i.e. 8-7th century BC) with clear inspiration from the
Late Bronze Age style (Ebert, 1926, p. 15).
Rosko, Czarnków-Trzcianka district
T&
N F
ot P
fo roo
rd f
is
tri
bu
tio
n
Several inds from Rosko were reported as early as the
1930s, but probably large parts of this hoard were discovered in the course of various construction activities irst
in 1985, then in 2001. Due to this, the precise original
number of items originating from this deposit remains
unknown, and to date only 71 have been brought together (full publication in Machajewski and Maciejewski,
2006). None of the items bear traces of use. The hoard
location was very interesting as it was under a single
mound, 1.5 m high and 10 m in diameter, constructed of
earth and large stones. Both the stone construction and
the mound itself were destroyed in the 1930s, but a probable reconstruction was recently completed on the site
(Machajewski, 2005, photo on p. 16). Near the mound,
a small stream and a pond were located. The deposit
mostly consists of bronze socketed axes decorated with
vertical ribs (67 items), and two, two-piece bronze cas-
ting moulds (igs. 5 and 6), which makes the deposit an
excellent example of selected deposition (cf. Fontijn,
2002, p. 211 – 217). None of the axes it the moulds. One of
the moulds was 11.7 cm long, 3.6 and 3.9 cm wide and had
ive pegs (igs. 6a – c). The other one was 12.9 cm long, 3.7
and 4.3 cm wide, again with ive pegs (igs. 6d – f). Both
are interpreted as having been designed for the production of wax models (Machajewski and Maciejewski, 2006,
p. 139 – 140). On one piece, a print of a woven fabric was
identiied which was the subject of another publication
(Sikorski, 2006). The deposit contained mostly axes that
were typical for large parts of south-western Poland and
may be dated to ca. 900 – 700 BC. The composition both
of the moulds and the products were analysed and proved
that they had been made of the same type of alloy containing 86.42% – 91.35% of copper and 6.86 – 11.61% of tin
(Machajewski, 2005, p. 19; Sałat et al., 2006). However,
there is a signiicant difference in composition between
two halves of one mould (called ‘ON’ and ‘ONA’). ON
contained 90% copper, 5.2% tin and 2.3% antimony,
while ONA contained 87% copper, 11% tin and 0.44% of
antimony (Sałat et al., 2006, table 1).
Fig. 5 – Deposit from Rosko (courtesy of the Museum of Piła).
Fig. 5 – Dépôt de Rosko (avec l’aimable autorisation du musée de Piła).
332
Bulletin de la Société préhistorique française
T&
N F
ot P
fo roo
rd f
is
tri
bu
tio
n
Functions and contexts of Bronze Age metal casting moulds from Poland
Fig. 6 – Metal casting moulds from the Rosko deposit (after Machajewski, Maciejewski, 2006).
Fig. 6 – Moules métalliques du dépôt de Rosko (d’après Machajewski et Maciejewski, 2006).
ARCHAEOMETRIC INVESTIGATION
OF THE MOULD FROM GAJ OŁAWSKI
T
he aim of the archaeometric investigation was
to identify both the chemical composition of the
mould (including the alloy’s main elements and the
molecular composition of corrosion layers), and the original purpose of the item. All the analyses were done
in the Laboratory for Archaeological Conservation and
Archaeometry of the Institute of Archaeology, University of Wrocław. Detailed macroscopic and microscopic
Tome 111, numéro 2, avril-juin 2014, p. 325-338.
observations were performed to observe possible wax
remains, which might conirm the production of wax
models of axes. The chemical composition was investigated with the use of the XRF, while the analysis was
carried out with the use of a Spectro Midex (an X-ray
luorescence spectrometer with energy dispersion). The
spectrometer is equipped with a molybdenum X-ray
lamp of an excitation energy of 44.6 kV and a Si Drift
Detector (SDD) with a resolution of 150 eV. For calibration, references from Certiied Reference Material BCR
691 of the Institute for Reference Materials and Measurements were used.
333
Justyna Baron, Beata Miazga and Kamil Nowak
Sample
Cu
Sn
Sb
As
Pb
Ni
Sum
GOl 1
90.65
4.45
1.34
0.47
0.06
0.57
97.99
GOl 2
90.01
5.00
1.46
0.48
0.05
0.52
98.00
GOl 3
89.90
5.24
1.51
0.42
0.03
0.43
97.99
GOl 4
89.28
5.39
1.65
0.58
0.08
0.52
97.98
GOl 5
88.80
5.68
1.79
0.60
0.12
0.51
97.95
GOl 6
88.45
5.93
1.85
0.61
0.12
0.52
97.97
GOl 7
88.86
5.68
1.67
0.52
0.06
0.53
97.99
GOl 8
77.81
14.50
4.18
1.43
0.38
0.18
99.80
GOl 9
84.12
10.19
3.07
0.95
0.19
0.42
99.80
GOl 10
88.97
5.61
1.73
0.59
0.10
0.51
97.94
GOl 11
87.84
6.40
1.93
0.66
0.13
0.50
97.94
GOl 12
88.38
6.04
1.80
0.63
0.11
0.50
97.97
GOl 13
90.87
5.77
1.69
0.61
0.09
0.48
99.94
GOl 14
85.28
10.08
2.72
0.65
0.15
0.05
99.80
88.95
5.85
1.60
0.59
0.04
0.47
97.95
87.54
6.89
2.06
0.67
0.12
0.47
98.39
Standard deviation
4.11
3.30
0.95
0.32
0.11
0.12
0.80
T&
N F
ot P
fo roo
rd f
is
tri
bu
tio
n
GOl 15
Average
Table 1 – Elemental composition of the metal of the casting mould from Gaj Oławski. A number below 100% results from the lack of
quantitative analysis of light elements (e.g. S, P, Si, Al, and Mg).
Tabl. 1 – Composition élémentaire du métal du moule de Gaj Oławski. Le total inférieur à 100 % résulte de l’absence d’analyse quantitative des éléments-traces (par exemple S, P, Si, Al et Mg).
The data (15 measured points marked as GOl 1 – 15)
from the analysed object obtained on the composition (table 1) enabled us to consider the alloy as rather
homogenic (only three samples from the mould rim
showed higher tin content). Due to the artefact age and
observable corrosion, variability of metal content had
been expected, which is well illustrated by measurements of point GOl 8 (77.81% Cu, 14.5% Sn and 4.18%
Fig. 7 – The ED-XRF spectrum of the mould from Gaj Oławski
(sample Gol 11). Range 15-20 keV relects the XR lamp interference.
Fig. 7 – Spectre ED-XRF du moule de Gaj Oławski (échantillon Gol 11). L’intervalle 15-20 keV relète l’interférence de la
lampe XR.
334
Sb) and GOl 13 (90.87% Cu, 5.77% Sn and 1.69% Sb).
Regardless the diversity of particular metal content, the
mould was made of tin bronze (ig. 7), similar to the analysed mould from Rosko (called ‘ON’; Sałat et al., 2006).
Further analysis was done with the use of infrared
spectroscopy. The sample was prepared using the potassium bromide pellet method. After the removal of contamination from the sample, several milligrams of a black
substance was crushed and homogenized. Next, the 10 mg
of the sample was mixed with 200 mg of spectral KBr in
an agate mortar. The mixture of the sample and KBr was
pressed at a pressure of about 10 tons to form a transparent pellet. The pellet was transferred quickly to the
FT-IR spectrometer for analysis. FT-IR spectroscopy was
performed using a Thermo Nicolet 380 spectrometer with
Omnic software. Single beam spectra (4000 – 400 cm-1) of
the samples were obtained against air as a background, to
present the spectra in absorbance units at a resolution of 1
to 8 cm-1 and 16 – 64 scans.
The specimen containing products of oxidation of
bronze alloy exhibited traits typical for copper carbonates. Having in mind that the typical corrosion products
of bronzes are copper and tin oxides and carbonates:
cuprite, malachite, azurite, tin oxides: casserite and romarchite (Scott, 1994), the presence of malachite (hydroxycarbonate of copper Cu2CO3(OH)2) is not a surprise. Its
presence is also conirmed by chemical signatures of the
carbonic hydroxyl group constituting a series of characteristic peaks. The IR absorption spectrum of malachite
shows sharp and distinct bands at 818, 879, 1047, 1099,
Bulletin de la Société préhistorique française
Functions and contexts of Bronze Age metal casting moulds from Poland
Fig. 8 – The FT-IR spectra of specimen sampled from corroded
material from the mould (c.l. 2) and malachite (HR Inorganic).
Fig. 8 – Spectres FT-IR des spécimens échantillonnés sur du
matériau corrodé venant du moule et de malachite (HR inorganique).
T&
N F
ot P
fo roo
rd f
is
tri
bu
tio
n
1398, 1494, 3330 and 3406 cm-1 (Aliatis et al., 2009,
p. 532 – 538). Figure no. 8 shows the IR spectrum of the
specimen sampled from the mould with all the signatures present. Moreover, the spectrum was compared and
conirmed with a malachite spectrum pattern (HR Inorganic database).
Further analysis focused on the original purpose of the
mould i.e. to answer the question what kind of material
was poured into the mould. To achieve this, a specimen
was sampled from the area of a smooth, semi-glossy spot
(3 – 4 mm in diameter) situated by the parallel ribs and
contrasting with the oxidized mould’s inner surface (ig. 9).
The investigated residue was a black, soft paste with a
characteristic wax-like gloss. From the specimen, a pellet
(only 2 mg were sampled which — together with the age of
the material — inluenced the intensity of obtained signals
in the IR spectrum) was formed which then was investigated under a changing resolution, varying from 8 to 1cm-1.
In the course of analysis, a spectral IR image showing
signatures of mineral substances was obtained (ig. 10).
A search in the available databases (HR Inorganic, US
Geological Survey) allowed us to identify three minerals:
bentonite (Na, Ca)0.33(Al, Mg)2Si4O10(OH)2·nH2O (usually
indicated as a raw-material for pottery production), chalcosiderite CuFe6(PO4)4(OH)8·4H2O, and chrysocolla (Cu,
Al)2H2Si2O5(OH)4·nH2O. Apart from inorganic compounds, bands from the vibrations of hydrocarbons — possibly indicating natural waxes or resins were identiied
(Łucejko et al., 2012, p. 110 – 118). These signatures,
although very weak, are visible after subtracting the spectra in the following areas: ca. 3000, 1700, 1500 – 1350,
700 – 800cm-1. According to previous macroscopic observations, the focus was on possible ancient beeswax
remains. The available results of the analyses of archaeological waxes (e.g., Bakr et al., 2005; Latuatti-Derieux
et al., 2009; IRUG Database) allowed us to state that the
beeswax IR-signals are: 2955, 2918, 2849, 1462, 1737,
729 and 719 cm-1 (Kovala-Demertzi et al., 2012) and
3428, 2952, 2915, 2844, 1735, 1467, 1375, 1195, 1172,
729 and 721 cm-1 (Luo et al., 2012). J. Łucejko et al.
(2012) argues that wax content is conirmed with a signal
of 1460 cm-1. The analysed specimen gave such signals
as: 3411, 2948 – 2924 – 2853, 1733, 1464, 1388, 1165,
730 and 709 cm-1 corresponding well with the identiied
archaeological samples of beeswax (ig. 11).
DISCUSSION
Were these moulds used for wax
or metal casting?
There are two types of data obtained for the mould from
Gaj Oławski. One group refers to the mineralogical composition of the item and proved the relatively low tin
content (table 1) which is far from Bronze Age ‘standards’
of ca. 10%. This may correspond with the opinion that the
axes — regarding the alloys and casting techniques — display relatively lower quality than weapons or ornaments
Fig. 9 – Microscopic image of the mould. A shiny black area in the central part was sampled for FT-IR tests.
Fig. 9 – Image microscopique du moule. L’aire noire et luisante dans la partie centrale a été échantillonnée pour les tests FT-IR..
Tome 111, numéro 2, avril-juin 2014, p. 325-338.
335
Justyna Baron, Beata Miazga and Kamil Nowak
Fig. 10 – FT-IR spectra of investigated specimen and identiied
minerals.
T&
N F
ot P
fo roo
rd f
is
tri
bu
tio
n
Fig. 10 – Spectres FT-IR du spécimen analysé et minéraux identiiés.
while the irst one is considered to have been lost during
the last war. The data for Rosko proved that there is no
direct and simple correspondence between the moulds
and axes’ composition, as both types of artefacts are not
homogeneous and vary signiicantly in their particular tin
and antimony content — even in one, two-piece artefact
(Sałat et al., 2006, table 1).
Some experimental work proved that the metal
moulds could have been used successfully in the production of bronze artefacts, and when controlled in terms
of temperatures, did not get broken in the course of casting. The process has been already described in detail in
several papers and thus will not be discussed here (e.g.,
Mohen, 1978, p. 27 with further literature therein).
The mould from Gaj Oławski displays properties
allowing it both to have cast metal (low tin content) and
wax (beeswax remains), which makes it a good example
of a multi-purpose tool used in various conditions. It also
proves that bronzesmiths had great lexibility while working on metal or wax. It must be noticed here, bronze
moulds are considered to be designed rather for mass
production of axes which excludes use of single-use wax
models. Fired organic coatings, mostly charcoal, could
have been used as a layer preventing the welding of mould
with poured liquid metal, and the wax can represent the
remnants of such coating (which would mean that the
mould was covered with coating but was not ired). Alternatively, wax could have been used to protect the mould
surface from corrosion as metal moulds are considered to
be highly-valued items (Kujipers, 2008, p. 89).
Fig. 11 – FT-IR spectrum of investigated specimen. The squares
mark signals of beeswax.
Fig. 11 – Spectre FT-IR du spécimen analysé. Les carrés indiquent les signaux de cire d’abeille.
(e.g., Sałat et al., 2006, p. 150). From this perspective,
the moulds and the axes represent the same ‘kind’ of artefacts. This also its the opinion that the low tin content
may indicate a better resistance against high temperatures
and direct metal casting.
The other group of data refer to the black residue
from the mould which is beeswax remains. This appears
to conirm that bronze moulds were used in in the production of wax models, not metal items. Can these two set of
data and thus two opinions be combined? Alternatively,
was the mould a multi-purpose tool?
According to an opinion that is common also in Polish
literature, metal casting moulds, due to their low resistance to high temperatures, were used exclusively for
making wax or lead models and not for producing bronze
products themselves (e.g., Kostrzewski, 1953, p. 193;
Dobrzańska, 1959, p. 90; recently Machajewski and
Maciejewski, 2006, p. 142). The similar composition (the
tin content in particular) of both the moulds and products
are considered to preclude direct casting (Machajewski,
2005, p. 19). Although both moulds and axes were found
at Kiełpino, Brzeg Głogowski and Rosko, only the latter one was analysed in terms of the mineral composition
336
Distribution and context
Bronze moulds, mostly designed for production of axes,
have been discovered in large parts of Europe, but it is not
possible to identify any clear distribution pattern, although
one concentration may be noticed in north-east France
(Mohen, 1978, ig. 7). Three such objects for casting axes
(but none complete) are known from the Netherlands,
while two of them were found in a wet environment and,
based on the context were interpreted as votive offerings.
They are dated to the Late Bronze Age (Kuijpers, 2008,
p. 89, appendix 2). Similar moulds of mostly the same
chronology are known from Spain, France and the British
Islands (e.g., Cordier, 1962; Mohen, 1978; Fleury, 1991,
p. 269-270; Armbruster, 2000). A few have been found
in the Czech Republic, Slovakia and Hungary, however, their exact number and contexts are rather confused
(Kosařova, 2009, p. 75). Probably the best documented
are moulds from deposit II from Velim, ca. 60 km east
of Prague, where three bronze two-piece items for axes
were found (Hrala, 2000, p. 222-223). The location itself
is very interesting as the deposit was one of seven hoards
found at the cult site of ‘Skalka’ [The Rock] at Velim,
accompanied by a system of ditches illed with hundreds
of human bones (Hrala et al., 2000).
Considering the context and the interpretation, the
discussion on bronze casting moulds is strongly connected with the discussion on bronze deposits in general.
Bulletin de la Société préhistorique française
Functions and contexts of Bronze Age metal casting moulds from Poland
that both were given similar signiicance, i.e. the moulds
were treated like tools (axes) and deposited in the same
way. This convergence corresponds well with the results
of the archaeometric investigation that proved the similar mineralogical composition of moulds and axes from
Rosko — both types of artefacts are made of similar heterogeneous alloys (Sałat et al., 2006, table 1).
The last ind of the mould from Gaj Oławski and older
inds, some published over 100 years ago seem to be an
interesting category of inds based on various aspects
including their use and mode of deposition.
CONCLUSIONS
T
he bronze moulds from Poland complete the list of
previously known bronze casting moulds from other
parts of Europe. They seem to share the same characteristics including the destination of the moulds which was
the axe production.
Although very limited data are available, at least one
mould (from Gaj Oławski) can be called a multi-purpose
tool used both for metal casting (indirectly proven by the
low tin content) and for wax model production. If we
deny use of wax models for axes, the beeswax remains
may be considered as organic coating, which could have
been used as a layer preventing the welding or to protect
the mould surface from corrosion.
In terms of context, three of the ive locations were
bog deposits. Observations of contexts from other parts
of Europe allow us to say that the meaning and thus the
deposition pattern was the same for moulds and ready
axes and other tools.
T&
N F
ot P
fo roo
rd f
is
tri
bu
tio
n
The known locations allow us to consider the bronze
casting moulds as objects of deliberate, and usually
selective deposition, mostly in what can be described
as ‘offering contexts’, including river, bog and hilltop
inds. In his detailed study on depositional patterns and
contexts, D. Fontijn argues that axes, sickles and weapons were kept away from graves, deposited somewhere
else, mostly in what may be broadly understood as ‘wet
areas’ (Fontijn, 2002, tables 10.1 and 10.2). Although his
study concerns the Netherlands, it seems to it the situation from other areas as well. Similar conclusions with
a particular focus on river and bog inds in Poland were
presented by J. Woźny (1996). He argues that the composition of water deposits in the Bronze and Early Iron Age
in Poland relects a clear opposition between weapons/
tools and ornaments (Woźny, 1996, p. 89).
Recent publications on bronze deposits in Poland
relect a move away from purely economic interpretations
to more complex views including signiicant examples of
‘votive offerings’ (Blajer, 2001, p. 298 – 299), with many
examples connected with bog and river contexts (e.g.,
Woźny, 1996 and 1997). Among water deposits, 56% can
be dated to the Late Bronze, Early Iron Age (Blajer, 2001,
ig. 38) which is interpreted as relecting mostly ritual
activities (Blajer, 2001, p. 288).
Among ive Polish sites with the moulds, three were
bog deposits (Kiełpino, Gaj Oławski, and indirectly
Rosko) while for the other two, no information is available and thus similar contexts cannot be excluded.
The three sites with known water contexts are located
in a lowland area where the number of tools deposited
in water increase towards the end of the Bronze Age
(Woźny, 1997, p. 38; Blajer, 2001, ig. 38). When the
contexts of the moulds and axes are compared, it is clear
BIBLIOGRAPHY
Aliatis I., Bersani D., Campani E., Casoli A., Lottici P. P.,
Mantovan S., Marino I.-G., Ospitali F. (2009) – Green
pigments of the Pompeian artists’ palette, Spectrochimica
Acta Part A: Molecular and Biomolecular Spectroscopy,
73, pp. 532 – 538.
Armbruster B. R. (2000) – Goldschmiedekunst und Bronzetechnik. Studien zum Metallhandwerk der Atlantischen
Bronzezeit auf der Iberischen Halbinsel, Montagnac, Monique Mergoil (Monographies instrumentum, 15), 231 pp.
Bakr Mossa A., Kawiak T., Pawlikowski M., Sawlowicz Z. (2005) – Characterisation of 15th century red
and black pastes used for wall decoration in the Qijmas
El-Eshaqi mosque (Cairo, Egypt), Journal of Cultural Heritage, 6, pp. 351 – 356.
Blajer W. (2001) – Skarby przedmiotów metalowych z epoki
brązu i wczesnej epoki żelaza na ziemiach polskich, Kraków, Uniwersytet Jagielloński w Krakowie, 427 pp.
Butler J. J., Steegstra H. (2005-2006) – Bronze Age Metal
and Amber in the Netherlands (III:2). Catalogue of the socketed axes, part C, Palaeohistoria, 47-48, pp. 207 – 240.
Tome 111, numéro 2, avril-juin 2014, p. 325-338.
Cordier G. (1962) – Quelques moules de l’âge du Bronze provenant de la Touraine et du Berry, Bulletin de la Société
préhistorique française, 59, 11-12, pp. 838 – 849.
DobrzaŃska E. (1959) – Przyczynek do znajomości metalurgii
brązowej w epoce brązu i okresie halsztackim w dorzeczu
górnej Wisły i górnej Odry, Silesia Antiqua, 1, pp. 83 – 102.
Ebert M. (1926) – Reallexikon der Vorgeschichte, VII, Berlin,
Walter De Gruyter, pp. 14 – 15.
Fleury L. (1991) – Moules de l’âge du Bronze en France atlantique et dans les îles Britanniques : aspects techniques et
culturels, in C. Chevillot and A. Coffyn (eds.), L’Âge du
Bronze atlantique : ses faciès, de l’Ecosse à l’Andalousie
et leurs relations avec le Bronze continental et la Méditerranée, procceedins of the irst congress held in the Parc
archéologique de Beynac, Beynac-et-Cazenac, Association
des musées du Sarladais, pp. 269 – 270.
Fontijn D. R. (2002) – Sacriicial landscapes. Cultural biographies of persons, objects and ‘natural’ places in the Bronze
Age of the southern Netherlands, ca. 2300-600 BC, Leiden,
University of Leiden (Analecta Praehistorica Leidensia,
33-34), 392 pp.
337
Justyna Baron, Beata Miazga and Kamil Nowak
Hrala J. (2000) – Metal artefacts and artefacts of other materials, in J. Hrala, R. Šumberová & M. Vávra (eds.), Velim. A
Bronze Age fortiied site in Bohemia, Praha, pp. 222 – 223.
Scott D. A. (1994) – An Examination of the Patina and Corrosion Morphology of Some Roman Bronzes, Journal of the
Hrala J., Šumberová R., Vávra M. (2000) – Velim. A Bronze
Age fortiied site in Bohemia, Praha, Institute of Archaeology, Academy of Sciences of the Czech Republic, 347 pp.
Seger H. (1909) – Beiträge zur Vorgeschichte Schlesiens von
Hans Seger: 3. Gussformen, Schlesiens Vorzeit in Bild und
IRUG Database, http://www.irug.org/ed2k/search.asp
KosaŘova Z. (2009) – Odlévací formy mladší doby bronzové
z oblasti středního Podunají, masters thesis, Masarykova
univerzita v Brně, Brno, pp. 75.
American Institute for Conservation, 33, 1, pp. 1 – 23.
Schrift, N. F. V, 5, pp. 16 – 27.
Seger H. (1927) – Vermehrung der vorgeschichtlichen Sammlung des Schlesischen Museums für Kunstgewerbe und
Altertümer, Altschlesien, 2, 1, pp. 59 – 68.
Kostrzewski J. (1953) – Wytwórczość metalurgiczna w Polsce od neolitu do wczesnego okresu żelaznego, Przegląd
Archeologiczny, 9, pp. 177 – 213.
Sikorski (2006) – Wyrób sprangowy na brązowej formie do
odlewu siekierek z Roska, in H. Machajewski & J. Rola
(eds.), Pradolina Noteci na tle pradziejowych i wczesnośred-
Kovala-Demertzi D., Papathanasis L., Mazzeo R.,
Demertzis M. A., Varella E. A., Prati S. (2012) – Pigment identiication in a Greek icon by optical microscopy
and infrared microspectroscopy, Journal of Cultural Heritage, 13, pp. 107 – 113.
WoŹny J. (1996) – Symbolika wody w pradziejach Polski, Bydgoszcz, Wydawnictwo Uczelniane WSP w Bydgoszczy,
186 pp.
WoŹny J. (1997) – Symbolika depozytów wodnych z epoki
brazu na Nizu Polskim [Symbolik der Wasserdeponierungen
der Bronzezeit in der polnischen Tiefebene], in W. Blajer
(ed.), Beiträge zur Deutung der bronzezeitlichen Hort- und
Grabfunde in Mitteleuropa = Przyczynki do interpretacji
skarbów i znalezisk grobowych z epoki brązu w Europie
Środkowej, proceedings of the Instytut Archeologii Uniwersytetu Jagiellońskiego (Kraków, 1996), Kraków, Materialen
der archäologischen Konferenz „Bronzen und Menschen an
der Schwelle der Urnenfelderzeit im Östlichen Mitteleu-
T&
N F
ot P
fo roo
rd f
is
tri
bu
tio
n
Kuijpers M. H. G. (2008) – Bronze Age Metalworking in the
Netherlands (C. 2000-800 BC). A Research into the Preservation of Metallurgy Related Artefacts and the Social Position of the Smith, Leiden, Sidestone, 89 pp.
niowiecznych szlaków handlowych, Poznań, pp. 153 – 156.
KuŚnierz J. (1998) – Die Beile in Polen III (Tüllenbeile), Stuttgart, Franz Steiner (Prähistorische Bronzefunde, 9, 21),
pp. 25, 30 – 31.
Lattuati-Derieux A., Egasse C., Regert M., Chung YongJae, Lavédrine B. (2009) – Characterization and degradation pathways of ancient Korean waxed papers, Journal of
Cultural Heritage, 10, pp. 422 – 427.
Luo W., Li T., Wang C., Huang F. (2012) – Discovery of
Beeswax as binding agent on a 6th-century BC Chinese
Turquoise-inlaid Bronze sword, Journal of Archaeological
Science, 39, pp. 1227 – 1237.
Łucejko J.-J., Lluveras-Tenorio A., Modugno F., Ribechini E., Colombini M. P. (2012) – An analytical approach
based on X-ray diffraction, Fourier transform infrared spectroscopy and gas chromatography/mass spectrometry to
characterize Egyptian embalming materials, Microchemical
Journal, 103, pp. 110 – 118.
Machajewski H. (2005) – Brązowy skarb z Roska, Archeologia Żywa, 3, 33, pp. 16 – 19.
Machajewski H., Maciejewski M. (2006) – Skarb ludności
kultury łużyckiej z Roska nad Notecią, in H. Machajewski
& J. Rola (eds.), Pradolina Noteci na tle pradziejowych i
wczesnośredniowiecznych szlaków handlowych, Poznań,
pp. 127 – 146.
Mohen J.-P. (1978) – Moules en bronze de l’âge du Bronze,
Antiquités nationales, 10, pp. 23 – 32.
Sałat R., Kozakowski S., Krokosz J., Warmuzek M.
(2006) – Badania metalograiczne przedmiotów brązowych
pochodzących z Roska, gmina Wieleń, in H. Machajewski
& J. Rola (eds.), Pradolina Noteci na tle pradziejowych i
wczesnośredniowiecznych szlaków handlowych, Poznań,
pp. 147 – 152.
338
ropa”, pp. 33 – 43.
Justyna Baron
Institute of Archaeology, Wrocław University
ul. Szewska 48
50-139 Wrocław, Poland
justyna.baron@gmail.com
Beata Miazga
Laboratory for Archaeological Conservation
and Archaeometry
Institute of Archaeology, Wrocław University
ul. Koszarowa 3, bud. 28
51-149 Wrocław, Poland
beata.miazga@archeo.uni.wroc.pl
Kamil nowak
Institute of Archaeology, Wrocław University
ul. Szewska 48
50-139 Wrocław, Poland
akinakesy@gmail.com
Bulletin de la Société préhistorique française