GRAVE GOODS FROM EARLY HUNGARIAN CONTEXTS: TECHNOLOGICAL IMPLICATIONS OF DEBASED ALLOY COMPOSITIONS WITH ZINC, TIN AND LEAD

Römisch-Germanisches ZentralmuseumForschungsinstitut für Archäologie

Verlag des Römisch-Germanischen Zentralmuseums Mainz 2012

SONDERDRUCK AUS:

RGZM – TAGUNGENBand 17

Bendeguz Tobias (Hrsg.)

DIE ARCHÄOLOGIE DER FRÜHEN UNGARN

CHRONOLOGIE, TECHNOLOGIE UND METHODIK

Internationaler Workshop des Archäologischen Instituts

der Ungarischen Akademie der Wissenschaften

und des Römisch-Germanischen Zentralmuseums Mainz

in Budapest am 4. und 5. Dezember 2009

Redaktion: Evelyn Garvey (New York); Reinhard Köster, Bendeguz Tobias (RGZM)Satz: Hans Jung (RGZM)Umschlaggestaltung: Reinhard Köster (RGZM)

Bibliografische Information der Deutschen Nationalbibliothek

Die Deutsche Nationalbibliothek verzeichnet diese Publikation inder Deutschen Nationalbibliografie; detaillierte bibliografischeDaten sind im Internet über http://dnb.d-nb.de abrufbar.

ISBN 978-3-88467-205-1ISSN 1862-4812

© 2012 Verlag des Römisch-Germanischen Zentralmuseums

Das Werk ist urheberrechtlich geschützt. Die dadurch begrün detenRechte, insbesondere die der Übersetzung, des Nach drucks, derEntnahme von Abbildungen, der Funk- und Fernsehsen dung, derWiedergabe auf photomechanischem (Photokopie, Mikrokopie)oder ähnlichem Wege und der Speicherung in Datenverarbei -tungs anlagen, Ton- und Bild trägern bleiben, auch bei nur auszugs-weiser Verwertung, vor be halten. Die Vergü tungs ansprüche des § 54, Abs. 2, UrhG. werden durch die Verwer tungs gesellschaftWort wahrgenommen.

Herstellung: Strauss GmbH, MörlenbachPrinted in Germany.

Falko Daim

Vorwort . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . IX

Vor der Landnahme

Attila Türk

Zu den osteuropäischen und byzantinischen Beziehungen der Funde des 10.-11. Jahrhunderts im Karpatenbecken . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3

Charlotte Hedenstierna-Jonson

Traces of contacts: Magyar material culture in the Swedish Viking Age context of Birka . . . . . . . . . . . . . . 29

Chronologische Fragen zum Fundmaterial des Karpatenbeckens

Péter Langó

Notes on the dating of Byzantine coin finds from 10th century context in the Carpathian Basin . . . . . . . . 49

Péter Prohászka

Bemerkungen zum byzantinischen Münzverkehr der ungarischen Landnahmezeit und der Staatsgründung im Karpatenbecken . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67

Gabriel Fusek

Chronologische Fragen der Nitraer Gräberfelder des 10.-11. Jahrhunderts:das Fallbeispiel Nitra-Šindolka . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89

Gabriel Nevizánsky · Jiří Košta

Die Ausgrabung eines frühungarischen Reitergräberfeldes in Streda nad Bodrogom (okr. Trebišov/SK) in den Jahren 1926 und 1937 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113

Miklós Takács

Die Chronologie der Siedlungen und besonders der Siedlungskeramik des Karpatenbeckens des 8.-11. Jahrhunderts im Spannungsfeld zwischen den verschieden Datierungsmöglichkeiten und ihren Einwänden . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145

Zwischen Ost und West? »Fremde« Schwerter in »lokalem« Kontext

Naďa Profantová

Examples of the most important results of technological analyses of swords in the Czech Republic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169

III

INHALT

IV

Ádám Bíró

Dating (with) weapon burials and the »Waffenwechsel«. A preliminary report on new investigations of the so-called Viking-Age swords in the Carpathian Basin from a chronological point of view . . . . . . . 191

Valeri Yotov

The Kunágota sword guard and the dating of two bronze matrices for hilt manufacturing . . . . . . . . . . . 219

Beiträge technologischer und naturwissenschaftlicher Untersuchungen zu archäologischen Fragestellungen

Adam Bollók

Chronological questions of the Hungarian Conquest Period: a technological perspective . . . . . . . . . . . . 229

Susanne Greiff

Silver grave goods from early Hungarian contexts: technological implications of debased alloy compositions with zinc, tin and lead . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 241

Nataša V. Eniosova

Tracing the routes of silver procurement to the early urban centre Gnëzdovo in the 10th/early 11th centuries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 261

Mariela Inkova

A contribution to the problem of producing the Old Bulgarian belt-fittings from the 10th century . . . . . 277

Naďa Profantová

Ein tauschierter Steigbügel aus der Umgebung von Dobruška (okr. Rychnov nad Kněžnou/CZ) . . . . . . . . 295

Verzeichnis der Autorinnen und Autoren . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 309

Hinweis für den Leser:Kyrillische Buchstaben wurden wissenschaftlich transliteriert. In Ausnahmefällen wurde bei Eigennamen aufeine wissenschaftliche Transliteration verzichtet. Bei den Fundortangaben in den Gebieten der ehemaligenSowjetunion werden die heutigen Ortsnamen angegeben. Bei Kulturen bezeichnenden Fundorten wurdevon einer Änderung abgesehen.

SUSANNE GREIFF

SILVER GRAVE GOODS FROM EARLY HUNGARIAN CONTEXTS:

TECHNOLOGICAL IMPLICATIONS OF DEBASED ALLOY

COMPOSITIONS WITH ZINC, TIN AND LEAD

RESEARCH BACKGROUND

The investigation of the chemical composition of an alloy enables us to answer many questions concern-

ing technical achievements, metal resources and trade contacts. The present work is mainly based on the

Micro-X-Ray Fluorescence (Micro-XRF) analysis of early Hungarian grave goods dating to the 10th century

from different archaeological sites that were investigated at the archaeometry laboratory of the Römisch-

Germanisches Zentralmuseum in Mainz (RGZM). The objects were studied in the context of silver alloy com-

positions in relation to manufacturing techniques and the silver-based monetary system of that period.

The grave goods are magnificent examples of the early mediaeval craftsmanship comprising an extensive

use of fire-gilded silver (fig. 1). The study initially focused on silver alloys used for grave goods that were

presented in a special exhibition at the RGZM1. Having assembled so many important finds of early Hun -

garian nomadic horsemen at one and the same location, a comprehensive study of those objects with

respect to their chemical composition was performed. Some pieces turned out to be made of unusual zinc-

rich silver alloys, and it was decided to extend the study to other early Hungarian silver objects in order to

241Die Archäologie der frühen Ungarn

Fig. 1 The personal equipment of early Hungarian warriors was richly decorated with gilded silver, such as the mounts on this sabrefrom Karos cemetery III grave 11. These were typical items analysed within the project. – (Photo V. Iserhardt, RGZM).

find out whether this phenomenon of debased silver is typical of this period and this region. Some time

after this first set of analyses, several silver objects from the Zemplín site (Trebišov dist./SK) were investigat-

ed by the same method. The newest data set obtained from the site of Szeged-Kiskundorozsma Hosszúhát

(Csongrád county/H) is discussed in another paper2.

As a general result, different groups of alloys could be distinguished, and the work demonstrated signifi-

cant correlations between alloy composition and manufacturing typology. Apart from this, the origin and

function of different alloying elements observed in the context of ancient silver alloys will be reviewed. The

development of debased silver alloys was also studied based on published data from other periods and geo-

graphic sources. The composition of early mediaeval silver coins is discussed as a possible source material

for Hungarian silver objects3.

33 objects were obtained from several archaeological sites. Most of the objects are related to male grave

goods customs and include sabres, swords, spears, belt plaques, purses, quivers, parts of horse harnesses,

stirrups etc. One group of objects came from archaeological excavations in Hungary, Romania (Muşca) and

Austria (Gnadendorf) 4; they were all investigated in the course of the Mainz exhibition mentioned above:

sabre, Geszteréd-Kecskelátó dűlő (Szabolcs-Szatmár-Bereg county/H); belt plate and purse mount, Buda -

pest- Farkasrét (Pest county/H); quiver and sabre, Karos-Eperjesszög (Borsod-Abaúj-Zemplén county/H) cem -

etery III grave 11; sabre, ibidem cemetery II grave 52; sword, Szob-Kiserdő (Pest county/H); stirrup (loca lity

unknown); stirrup, Muşca (Hung. Muszka; Arad county/RO); snaffle bit, Muşca.

Another assemblage analysed two years later was retrieved from a rich burial at Zemplín5. Among the 29

objects were different strap-ends, belt plaques and small decorative silver ornaments.

The analytical equipment used here allowed for a minimal intervention with the objects, as the huge sam-

ple chamber of the Micro-XRF machine is able to accommodate artefacts of more than 100 cm in length,

50 cm in width and a height of up to 40 cm. While the artefacts may be large, the spot size of the beam is

limited to 0.3 mm, permitting the accurate analysis of small areas and manufacturing details 6. The data

were acquired from freshly prepared metal surfaces only.

242 S. Greiff · Silver grave goods from early Hungarian contexts

Fig. 2 Belt from Zemplín with the original leather strip still attached to the mount. – (Photo R. Müller, RGZM).

MICRO-XRF RESULTS

The first data set: objects from Hungary, Romania and Austria

33 silver objects (belt plaques, small ornaments from quivers, stirrups, swords etc.) were analysed in the

RGZM archaeometry laboratory; other objects have been analysed before via SEM/EDAX at the Vienna

Institute for Archaeological Sciences (VIAS)7. For the purpose of comparing coins and object compositions,

26 silver coins found in early Hungarian tombs near Szeged were also analysed but not included in figure2 (tab. 1). Many objects were found to be made from an unusual silver alloy rich in copper (Cu), tin (Sn),

zinc (Zn) and lead (Pb), with contents up to 50 % Cu, 7 % Sn, 4.5 % Zn and 11% Pb. Bismuth (Bi) and Gold

(Au) were also found in most of the objects, with Bi ranging from 0.05% up to 0.9 % and Au from 0.04%

up to 5 %. There is a strong correlation between silver and copper (fig. 3), whereas other inter-elemental

correlations are much less pronounced.

It is important to note that all Micro-XRF analyses had been performed on freshly prepared metal surfaces,

cleaned under the microscope in order to eliminate the uppermost layers. This means that copper migrat-

ing to the surface due to corrosion processes or silver-enriched surfaces created by ancient craftsmen by

pickling were not included in the tables meant for interpretation and comparison. Some samples were

analysed before and after the cleaning in order to reveal the mentioned effects of surface alteration (tab.2). For the purpose of comparison with other published data sets it is important to know by which analyt-

ical methods, particularly with regard to sample preparation and standard reference materials, the results

were obtained. If this information is not given in a publication, its values must be interpreted with great

caution.


Fig. 3 Graphical presentation of the copper vs. silver content for the objects of the first data set. Both copper-rich and objects with a»normal« level of copper have been found. – (S. Greiff, RGZM).


object inv. no./collection Ag Au Bi Cu Pb Sn Zn

1 sabre, 61.201.1-2; 61.201.9; 96.22 1.33 0.3 1.9 0.23 0 tr.Geszteréd 61.201.15-16/Nyíregyháza,

JAM2 sabre, 61.201.1-2; 61.201.9; 95.88 1.24 0.29 2.4 0.19 0 tr.

Geszteréd 61.201.15-16/Nyíregyháza, JAM

3 sabre, 61.201.1-2; 61.201.9; 96.02 1.15 0.26 2.29 0.24 0 tr.Geszteréd 61.201.15-16/Nyíregyháza,

JAM4 belt, Budapest- 104/1909.6-33/Budapest, 79.44 2.74 tr. 7.33 1.61 8.23 0.61

Farkasrét HNM5 belt, Budapest- 104/1909.6-33/Budapest, 68.15 2.65 tr. 20.16 1.16 7.08 0.68

Farkasrét HNM6 belt, Budapest- 104/1909.6-33/Budapest, 79.29 2.76 tr. 6.8 1.88 7.81 1.38

Farkasrét HNM7 purse mount, 104/1909.66-72/Budapest, 85.23 0.56 0.53 7.41 2.38 0 3.9

Budapest- HNMFarkasrét

8 purse mount, 104/1909.66-72/Budapest, 80.34 0.52 0.57 11.61 2.69 0 4.28Budapest- HNMFarkasrét



11 stirrup, 16/1922/Budapest, HNM 48.35 0.83 0.26 49.08 0.96 0 0.42unknown

12 stirrup, 16/1922/Budapest, HNM 45.94 0.77 0.19 51.59 0.95 0 0.44unknown

13 sword, 2/1937.20/Budapest, HNM 94.13 0.47 0.29 4.42 0.19 0 tr.Szob-Kiserd

14 stirrup, 45/1898.3/Budapest, HNM 92.18 1.19 0.13 5.56 0.9 0 tr.Musca

15 stirrup, 45/1898.3/Budapest, HNM 91.93 1.26 tr. 5.64 1.04 0 tr.Musca



18 stirrup, 45/1898.4/Budapest, HNM 91.29 1.28 0.14 5.9 0.97 0 tr.Musca

19 snaffle bit, 45/1898.9/Budapest, HNM 85.18 1.07 0.13 6 0.98 6.61 tr.Musca

20 snaffle bit, 45/1898.9/Budapest, HNM 85.41 1.09 tr. 5.89 0.85 6.63 tr.Musca

21 sabre, 94.82.5/Miskolc, HOM 87.13 0.55 0.46 8.57 1.34 0 1.93Karos III/11




25 sabre, 94.82.5/Miskolc, HOM 82.71 0.67 0.42 13 1.28 0 1.83Karos III/11


Tab. 1 First data set for Hungarian and Romanian objects: quantitative analysis results given in weight per cent (values for Gnadendorfcan be found in Mehofer / Greiff 2006). – Karos = Karos cemetery. – JAM = Jósa András Múzeum. – HNM = Hungarian National Museum.– HOM = Herman Ottó Múzeum. – tr. = traces.


When presenting the copper content vs. the silver content of the objects from the Hungarian and Romanian

sites exhibited in Mainz in a binary diagram (fig. 3), it becomes obvious that silver alloys with quite variable

compositions were used. Copper concentrations can reach levels higher than 30 %. The relative distribu-

tion of objects with high copper contents and objects with low copper contents has been found to be con-

nected to the manufacturing technique. This can be revealed by a suitable graphical interpretation. If we

object inv. no./collection Ag Au Bi Cu Pb Sn Zn

27 sabre, 94.82.5/Miskolc, HOM 95.33 1.27 0.31 2.71 0.34 0 tr.Karos III/11







34 quiver, 94.86.2/Miskolc, HOM 63.36 0.59 0.38 28.24 1.55 5.48 0.29Karos III/11

35 quiver, 94.86.2/Miskolc, HOM 74.61 0.75 0.41 15.3 1.67 7.05 0.1Karos III/11

36 quiver, 94.86.2/Miskolc, HOM 77.73 1.19 0.22 18.62 1.78 0 0.4Karos III/11



39 quiver, 94.86.2/Miskolc, HOM 94.9 0.45 0.55 3.28 0.77 0 tr.Karos III/11

40 sabre, 94.46.4-5/Miskolc, HOM 93.27 2.61 0.29 3.17 0.54 0 tr.Karos II/52

41 sabre, 94.46.4-5/Miskolc, HOM 95.31 0.69 0.31 2.98 0.66 0 trKaros II/52 .






47 sabre, 94.46.4-5/Miskolc, HOM 93.07 1.22 0.1 4.52 1 0 tr.Karos II/52



50 sabre, 94.46.4-5/Miskolc, HOM 92.77 4.82 tr. 2.17 0.09 0 tr.Karos II/52


Tab. 1 (continued).


use open symbols for chased sheet metal objects and bold ones for cast objects, it becomes evident that

the copper content of objects made from sheet metal never exceeds 5 %, while cast objects may contain

up to 30 % copper and more. The elemental composition as a function of the production technique is fur-

ther illustrated by other components such as zinc, tin and/or lead. These elements are associated with cast

objects only8 and can be interpreted as the chemical remnants of the addition of copper alloys such as

bronze, leaded bronze, brass or gunmetal to an originally purer silver alloy.

Cast objects show copper contents between 5 % and 32 %, while sheet metals have less than 5 % copper.

For lead, the line separating cast objects from hammered ones lies at a value of ca. 1.2 % (fig. 4). The max-

imum value is 2.8 % Pb (purse mount from Budapest-Farkasrét, inv. no. 104/1909.66-72). A few objects

show elevated amounts of tin between 5% and 8 % Sn (belt plaques from Budapest-Farkasrét [inv. no.

104/1909.6-33] and parts of the Karos quiver decoration [inv. no. 94/86.2]), whereas others have been

alloyed with a compound containing zinc. The cast objects discussed here have zinc in amounts of 0.3%

up to nearly 4.5 %.

If we recalculate the zinc as being an addition to a silver melt in the form of brass with a moderate zinc

content, then a fraction of 1/10 to 1/30 in mass must have been added. Of course, the variability of brass-

es9 and the loss of zinc by evaporation have to be taken into account, but these numbers should give an

approximate indication of the possible range. Some of the analysed alloys have both tin and zinc, and some

have an elevated content of lead so that the addition of gunmetal (ternary copper alloy with tin, zinc and

analysis RGZM object Ag Au Bi Cu Zn Pb no. no.

6 – small fitting (belt/horse harness?) 86.98 1.72 0.39 8.88 0.89 0.947 – small fitting (belt/horse harness?) 89.28 2.99 0.60 5.45 0.46 1.115

21 – small fitting (belt/horse harness?) 94.17 2.83 0.54 1.44 0.15 0.758 07/205 fragments of belt mounts 88.18 3.52 0.86 4.47 1.31 1.329 07/205 strap-end 90.82 1.50 0.31 5.76 0.75 0.71

10 07/205 fragments of belt mounts 90.87 2.49 0.66 3.42 1.168 1.2422 07/205 fragments of belt mounts, rivet 82.05 1.97 0.60 2.43 1.03 11.7223 07/205 strap-end, rivet 92.3 1.85 0.93 2.83 1.09 0.7511 07/206 decorative ornament 88.13 1.71 0.20 7.8 0.38 1.6812 07/206 decorative ornament 89.56 2.10 0.20 5.83 0.44 1.7524 07/206 decorative ornament, rivet 87.78 1.45 0.24 8.22 0.37 1.8125 07/206 decorative ornament, rivet 90.22 1.61 0.18 5.80 0.37 1.731 07/207 fitting of horse harness 49.85 tr. 0.46 46.05 2.51 1.022 07/207 fitting of horse harness 70.22 0.33 0.45 25.73 2.19 0.923 07/207 fitting of horse harness 62.78 0.15 0.43 32.45 2.75 1.394 07/207 fitting of horse harness 48.23 tr. 0.60 41.35 3.4 5.005 07/207 fitting of horse harness 53.38 tr. 0.60 42.06 2.44 1.43

18 07/207 fitting of horse harness 88.74 1.47 0.11 6.77 0.7075 1.9419 07/207 fitting of horse harness 87.31 1.16 0.62 6.54 2.595 1.2426 07/207 fitting of horse harness, rivet 95.75 1.21 0.74 0.63 0.54 0.5327 07/207 fitting of horse harness, rivet 92.12 0.93 0.60 4.81 0.56 0.8615 07/208 small fitting 55.91 2.17 tr. 41.06 0.21 0.5616 07/208 small fitting 90.36 1.56 0.11 5.03 0.75 2.0417 07/208 small fitting 87.13 1.38 0.12 8.52 0.665 1.8520 07/208 small fitting, rivet 89.07 1.77 0.42 7.33 0.61 0.7128 07/208 small fitting, rivet 62.90 0.30 0.15 31.97 3.25 1.3329 07/208 small fitting, rivet 86.37 0.47 0.23 7.93 3.48 1.3913 07/295 large decorative plaque 84.98 0.63 0.78 9.60 3.89 0.0714 07/295 large decorative plaque 74.7 0.81 0.43 19.34 4.58 0.07

Tab. 2 Second data set for objects from Zemplín (Trebišov dist./SK): quantitative analysis results are given in weight per cent (tin con-tents [Sn] were beneath the detection limit of the met).

also lead) is another possibility. The belt plaques from Budapest-Farkasrét are a good example of this gun-

metal usage.

The same pattern has been found regarding the objects from Gnadendorf (Lower Austria/A) where most

cast objects are rich in copper and some also in zinc (not more than 2.4 %)10. Some obviously cast objects

were, however, made from a silver alloy with an amount of copper not higher than usual for chased objects.

The maximum copper content for Gnadendorf is ca. 26 %, which is similar to the range observed for the

other objects discussed here. Tin was not observed but in one case (reverse side of the sword chape cat.

no. 29), but this could be related to remnants of soft-soldering. Sometimes an increased lead content

accompanies copper in cast objects.

If one considers silver, copper and zinc only, the data points of some Gnadendorf items (suspension ring of

cat. no. 28 and sword chape cat. no. 23-2) lie close to the points representing the composition of a sabre

and a quiver from Karos. As the Gnadendorf objects were analysed with different analytical equipment, no

values for minor elements are available to verify this relationship in more detail.

Within the data set obtained by Micro-XRF, there are different small groups of items (all made by chasing)

with a very similar composition for all elements analysed. Three groups can be distinguished: several silver

decorations from the sabre from Karos cemetery III (inv. no. 9482.4) and some parts of the sabre from

Geszteréd (inv. nos 61.201.1-2; 61.201.9; 61.201.15-16) define one group; this could evoke the idea of a

specialised sabre workshop. In the second group, the decorative wire inlays of both stirrups from Muşca

(inv. nos 45/1898.3-4) are closely associated with each other, proving that they were indeed manufactured

as a pair. Some parts from the Karos cemetery II sabre also belong to this Muşca cluster. Other parts of this

sabre decoration, however, define a third group of data points. These are extremely similar in their alloy

compositions so that a production from the same melt batch can be deduced with fair certainty. For the

sword from Szob (inv. no. 2/1937.20) and the quiver from Karos no such clear tendencies and relationships

can be deduced from the available data.


Fig. 4 Graphical presentation ofthe silver vs. lead content for theobjects of the first data set. Therange of lead contents that are dueto the cupellation process do notexceed 1.2 wt. %. Higher valuesrepresent a deliberate addition oflead (for symbols see fig. 3). –(S. Greiff, RGZM).

The second data set: objects found at Zemplín

A choice of 29 pieces from eight objects coming from the site of Zemplín was in vestigated 11. The chemi-

cal composition of the analysed Zemplín objects is mainly characterised by silver and copper (fig. 5), and

the data are arranged in two groups: there is one group with copper concen trations between 0.5 % and

nearly 10 % and another with copper values between 20 % and 45 %. The latter group also shows elevat-

ed levels of zinc up to 4.5 %, with two exceptions. One rivet (no. 29) and two small plaques (nos 13 and

19) have an elevated zinc level without having an elevated copper level. Their analytical counterpart is

no. 15, with a copper content of more than 40 % not accompanied by an elevated zinc level. This piece

must have been produced by adding pure copper instead of brass to the original alloy. Another peculiar

feature of this object is a comparably high gold content of 2.2 %, while all other debased silver objects

have gold contents below 1% due to dilution effects. The original silver alloy either must have been

extremely rich in gold, or the copper has been associated with gold in some way12.

Some objects show remarkably similar alloy compositions. Several smaller groups with closely related alloy

compositions can be distinguished among the Zemplín data set. Samples 3 (a belt plaque, PNM-479) and

28 (a rivet from object PNM-492-01, a circular mount) are extremely close in their compositions, despite

their coming from different objects. They can be identified as coming from the same melt.

Another chemically related group consists of three ornamental decorations from object PNM-497 (nos 1, 4

and 5). They are similar, although not as close as nos 3 and 28.

Tin is not present in any of the samples, so that a debasement by adding bronze can be definitely exclud-

ed. The silver and copper data define a set of three groups with a similar silver/copper ratio but with dif-

ferent amounts admixed (fig. 3), resulting in a more or less pronounced debasement of the silver.


Fig. 5 Graphical presentation ofthe copper vs. silver content forthe objects from Zemplín. –(S. Greiff, RGZM).

DISCUSSION

General discussion of alloying elements in ancient silver objects

The chemical data summarised above can be interpreted by having a look at how copper, tin, zinc and other

elements control the properties of mixed silver alloys.

Copper

Copper in amounts of about a few per cent usually is deliberately added to silver because it confers hardness

and strength to an alloy which otherwise would be too soft for the use in jewellery or decorative elements 13.

Beside copper, lead, zinc and tin have been identified as important metal impurities. While copper-enriched

alloys are quite commonly observed in silver objects, the use of several per cent of lead, zinc and tin is less

often recorded; their possible origin and benefits will be discussed later. Values for antique objects typically

lie within a range from 2-5% to sometimes 10% Cu14. For the current project, we were able to identify

objects with additions of copper-tin-lead, copper-zinc-lead as well as binary copper-lead combinations.

Lead

The results show a distinct relationship between lead and copper. Whenever copper is reduced to the nec-

essary minimum, lead is also reduced. As a consequence, some of the observed elevated lead contents

(above 1 %) seem to be related to a clearly intentional alloying of the silver with lead or a leaded copper

alloy. A review of the corresponding literature supports this hypothesis 15. Most ancient silver objects con-

tain concentration levels of lead of up to 1% or 1.2 %. This is owing to the fact that native silver is rare in

nature, and the majority of this precious metal has to be obtained from lead ores. What is even more impor-

tant is that lead is incorporated into silver during the process of cupellation. This high-temperature opera-

tion can be used for different purposes in the metallurgical chain of silver production, refinement or assay-

ing. By the technique of cupellation precious metals can be extracted from lead obtained from silver-rich

lead ores or from recycled impure scrap metal mixtures. In all cases, the procedure involves the addition of

excess lead which is oxidised by a well-directed air flow. The unwanted impurities are oxidised together with

the lead, forming a layer floating on top the melt. These oxide layers are either removed mechanically or

absorbed by suitable crucibles or pots made from special porous and reactive materials called cupels16.

Repeating this procedure several times will gradually enhance the fineness of the treated alloy and finally

lead to rather pure silver with minor impurities such as lead, gold and bismuth. The lead contents of cupel-

lated silver are typically ca. 0.5% to up to 1% or 1.2 %. This amount has only a minor influence on the

material properties, whereas higher lead levels such as, e. g., in object no. 4 from Zemplín which are accom-

panied by copper cannot be regarded as the result of cupellation alone without any impact on the alloy

behaviour. Contamination with an inadequate amount of lead is usually avoided by silversmiths because of

its undesirable tendency to evoke hot brittleness17. The relationship observed for some objects between

copper and lead contents here indicates the addition of an alloy containing both elements such as leaded

bronze or gunmetal. This will be further discussed in the section dealing with the physical characteristics.

Tin and zinc

A review of the published analytical results of silver alloy compositions dating to the Middle Ages reveals

that most silver items have been produced by using quite pure silver alloys. Among the early Hungarian


objects analysed here, we find a set of objects with quite pure silver (having only the usual level of im pu -

rities such as lead, bismuth and gold) and alloy compositions that deviate from this scheme in that lead,

zinc and in some objects tin are present in amounts clearly exceeding the quantities of natural im purities

and cupellation additives. On the contrary: any contents of tin and zinc present in the original material

would be largely driven off during the process of cupellation. The levels found in some of these early

Hungarian examples exceed any amount that would be explainable as natural impurity18. There is a grad-

ual increase but rather a pronounced gap between low-copper and high-copper alloy compositions.

Bismuth and gold

Gold and silver geochemically behave similar, so that natural occurrences often contain both metals. As a

noble metal, gold follows the path of silver during cupellation and is not driven off together with the lead

oxide. Thus, most antique silver objects contain gold in amounts reaching several per cent. This is also the

case for the objects in this study. The majority of the analysed samples show gold contents between 0.5%

and 3.5%, but one belt from Zemplín features several parts with extremely low values (belt 07/207: strap-

end analysis no. 4, a rectangular plate analysis no. 1 and a rosette-shaped ornament analysis no. 5), while

one part of another belt exhibits the highest value of 3.5 % (belt 07/205 analysis no. 8). The low contents

are probably due to the extreme degree of debasement with copper. Ratios of silver and copper nearly

reach 1:1, so that any minor elements in the primary silver would be considerably diluted. However, even

taking this dilution into account, the silver alloy must have had a comparatively low gold content already

before the debasement.

Bismuth can be found in silver mines19. It is not as noble as gold or silver but still not so easily oxidised as

to be transferred in total into the lead oxide mixture during cupellation20. It is distributed among the cupel-

lated silver and the lead oxide mass. All objects from Zemplín regularly show bismuth contents higher than

the detection limit of the instrument of ca. 0.1 % w/w. This is not the case for the first set of objects

analysed while on display in Mainz, coming from different localities. Some of them have bismuth contents

below the detection limit, while for others the bismuth values are ca. 0.5 %. The amounts of these two

metals are not high enough to cause noticeable changes in the alloy’s mechanical or thermal behaviour. But

bismuth reveals that Islamic coins may have played a crucial role in the production of these items21. It has

to be investigated if the mines under caliphate control were the only ones producing bismuth-rich silver

alloys. Bismuth-rich silver has also been reported for the Spanish Río Tinto mining district 22. The content of

bismuth can exceed the content of silver in some galenite occurrences, as is observed for some mines of

the Middle Black Forest hydrothermal vein systems23.

Physical characteristics of debased silver

What is the advantage of adding copper, zinc, tin or lead materials to a silver alloy? The most important

benefit is the lowering of the melting point. If one mixes a metal with a small amount of another metal

with a lower melting point, the resulting alloy will start to melt at lower temperatures than the original

metal. By adding ca. 28 % copper to silver, the melting point is lowered from 961°C to 797°C. There is,

however, a drawback of this method: the silver alloy will have a yellowish colour compared to purer silver.

But this can be overcome by adding zinc which acts as a sort of »whitener« brightening up the colour. Thus

it is not astonishing that adding brass (a mixture of copper and zinc) instead of pure copper was sometimes

preferred by ancient craftsmen when they wanted to produce a silver alloy with higher amounts of copper.

This colour enhancement would have particularly played a role in jewellery and ornamental purposes. Zinc


also acts as a deoxidiser, preventing the molten silver from »spitting«. Zinc additions enhance tarnish-proof-

ness, and the objects take on a much nicer polish24. Brass was used, up to modern times, preferably for

the production of cast silver alloys and silver soldering materials 25; this was obviously a long-lasting tradi-

tion, as can be deduced from the objects made by early Hungarian craftsmen.

The addition of tin has a similar effect when it comes to the improvement of colour and tarnish-proofness

in a copper-rich silver alloy. The mechanical properties of tin-enriched silver alloys make them more bene-

ficial for the production of sheet metal 26, as there is an enhancement of ductility when several per cent of

tin are added; a good example might be the chased snaffle bit from Muşca (inv. no. 45/1898.9 ) with 6 %

tin and only ca. 5 % copper. Other examples, such as a plaque from the belt from Budapest-Farkasrét (inv.

no. 104/1909.6-33) and the quiver from Karos (inv. no. 94/86.s), are, however, cast. The plaque from the

quiver from Karos obviously follows a different tradition, with a high copper content (tin ca. 6 % and more

than 20% Cu). For the usage of tin, the correlation between manufacture and alloy composition is much

less straightforward than for zinc-rich silver.

Lower contents of a few per cent copper, on the other hand, are important in terms of mechanical alloy

properties. Pure silver shows a Vickers hardness (HV)27 of 26, while a 3 % copper addition to pure silver

yields values of 73 HV, and with 28 % Cu (eutectic composition) a Vickers hardness of even 97 is ob -

tained28. Furthermore, by specific heat treatment cycles of low-alloyed compositions silversmiths may have

systematically created desired properties in terms of hardness, elasticity and malleability.

Silver in the 10th century: coins, ingots and scrap metal as possible sources for debased silver

After having discussed the question of material properties, we have to look for the sources these alloys

might have come from. Silver was the most important monetary metal in that period, but payment trans-

actions were also done involving weighing pieces of metal such as ingots or the Nordic »hack-silver« and

scrap metal collections. How was the fineness of these alloys? Could we explain the peculiarly mixed

Hungarian silver alloys as simply being part of the broad spectrum of silver available at that time? Or were

they specific products purposely manufactured by silversmith workshops?

From the 9th to the 11th century, Europe was divided, in terms of its silver economy, into two regional sys-

tems: one with a dominance of a monetary coin system in the western part and a weight-based system in

the eastern and northern parts, connecting the region east of the Elbe with Scandinavia29. Whereas the

»Nordic« silver metal distribution has been investigated in some detail for many decades now30, the silver

market in the heart of the continent and its southern parts is less well examined. Much of the silver that

was used east and north of the Elbe river arrived in the form of Islamic coins because of a well-developed

network of trade and commerce between the Islamic caliphates, Byzantium and the regions along the

Volga up to Scandinavia. However, the coins were weighed and not valued according to their nominal

amount. Slaves were undoubtedly the most important commodity the silver was exchanged for31. Other

sources were diplomatic exchange, raids and tributes that were paid more or less voluntarily as an answer

to the menace of Viking and Hungarian intrusions.

The most substantial data set for 9th and 10th century silver alloys has been collected based on coins, ingots

and scrap metal from around the Baltic Sea and the North Sea in order to elucidate issues of trade and com-

merce. Looking at the immense data set comprising 5975 analyses of 10th-13th century Islamic and Euro -

pean coins as well as hack-silver published by Illisch and others32, we find that only few belong to the multi-

metal type with elevated levels of copper, zinc, tin and lead. Exploring only those objects with zinc levels of

more than 1%, we find the following distribution: among 2771 Islamic coins, only six have an elevated zinc


concentration33 (corresponding to 0.2% of this coin type and even to as low as 0.1% if we count only the

three 10th century pieces); the 557 analysed hack-silver objects only feature 29 analyses containing zinc

(= 5.0%34 with only one Otto-Adelheid-Pfennig from Waterneversdorf/D that can be counted to the 10th

century = 0.2 % of all analysed hoard items); and for the European coins, only 3 35 of 1629 pieces analysed

belong to this type (= 0.2 %). The enrichment with copper is most pronounced in Islamic coins issued after

the middle of the 10th century due to a »silver crisis« evoked by political upheavals in Middle Asia36, but

the addition of zinc remains rare.

Other published data sets of this period also contain some zinc-rich compositions as well. English coins

of the 9th and 10th century, especially those emitted in the South of England by Mercian and West Saxon

kings and the Archbishop of Canterbury, show elevated levels of both copper and zinc. The debasement

was most distinct during the third quarter of the 9th century37, with an annual inflation of ca. 20 %. These

low-silver coin issues are discussed in the light of the immense payments of »danegeld« by which English

kings tried to buy peace from the »heathen army«. One explanation could be that low-value silver coins

were given to the invaders to pay them out. Another explanation points out that the Vikings received the

»better« money while the English homeland currency was debased to compensate for the silver loss38.

Either way, it was a procedure under official control by the Royal Treasury. The debasement was achieved

by adding brass (earlier in the 9th century) 39 and later bronze alloys. In other parts of England, it took place

the other way round. Archaeological evidence of these coins excavated from early Hungarian tombs is, in

fact, rare40. More direct evidence comes from Italic, German and, of course, Islamic coins. Byzantine silver

examples are also rare among the grave goods. A small set of 23 coins excavated from different tombs at

the cemetery of Szeged-Kiskundorozsma Hosszúhát-halom and from other Hungarian archaeological com-

plexes mainly from the eastern Carpathian Basin have been analysed in the context of the current project

(tab. 3). Only three of them41 show elevated copper contents, but not one of them comprises zinc in the

levels detected for some of the jewellery items discussed above. Despite the small number of coins from

early Hungarian find contexts that have been analysed so far, the dirhams and European coins analysed by

Illisch and others point to the same direction, namely that coins with zinc and copper were extremely rare.

Copper-rich (without other base metal additions) emissions from the second half of the 10th century, how-

ever, are more numerous, corresponding to the late Samanid »silver crisis«42.

Beside coins, ingots and rods were another form of traded silver. Some Viking Age examples from Swe -

den43, Scotland and North-West England44 have been analysed for their alloy compositions. Those analysed

by Arrhenius and others all have silver contents greater than 90%, the rest being copper. One has 10 %

zinc, but its composition is more or less gunmetal with a low silver content of 5 %. Among the 133

analysed pieces from the British Isles, there are three with suspiciously high zinc levels from Chester (Cu 20-

24 %, Zn 1.6-2.9%, ca. 965 AD) and one from Burray with 5.2 % Zn and 24.4 % Cu (ca. 1000 AD). These

fit into the general pattern of English coins of the period.

There is some more evidence for zinc-rich silver in the 11th century in coins from France45 from different

important mints, but this phenomenon was also observed regarding the treasure of Fécamp (dep. Seine-

Maritime/F) lain down to ca. 980. There seems to be a sudden drop in the zinc contents – to a level near

zero – during the second half of the 11th century46.

Precursor alloys predating our Hungarian finds are rare. This might be partly due to the lack of a systemat-

ic silver analysis published for Roman or Migration Period objects. One set of samples with higher zinc con-

tents are coins from the hoard of Goting-Kliff on the Isle of Föhr/D47, but these date basically to the 8th

century. Among the coins found in early Hungarian contexts, we often encounter older emissions but typ-

ically nothing as early as the 8th century. The earliest coin from Deszk-Újmajor, grave 10 for example, e. g.,

is a coin from Charles the Bold (840-874).



RG

ZM

inv.

no

./A

gA

uB

iC

uZn

Pbn

o.

colle

ctio

n

10/8

5M

FM 5

3.43

.47

92.3

91.

28tr.

4.51

0.25

1.28

Des

zk-Ú

jmaj

or,

Am

brus

halo

mW

est

Fran

kish

Kin

gdom

, ki

ng a

nd m

int

unde

term

ined

(K

om.

Cso

ngrá

d/H

)gr

ave

1010

/86

KK

JM 7

3.1.

2883

.70

0.82

tr.13

.92

tr.1.

56K

isku

nfél

egyh

áza-

Kin

gdom

of

Italy,

Kin

g Ru

dolf

II of

Bur

gund

y (9

22-9

26),

Radn

óti M

ikló

s u.

48

min

t of

Pav

ia(K

om.

Bács

-Kis

kun/

H)

10/8

7K

KJM

73.

1.45

95.4

41.

10tr.

2.47

tr.0.

90id

emW

est

Fran

kish

Kin

gdom

, K

ing

Odo

(88

8-89

8),

min

t of

Tou

lous

e10

/88

KK

JM 7

3.1.

2193

.55

0.77

tr.4.

23tr.

1.27

idem

Kin

gdom

of

Italy,

em

p. B

eren

gar

(915

-924

), m

int

of P

avia

10/8

9M

FM 2

002.

18.4

89.6

90.

600.

148.

48tr.

1.01

Kis

kund

oroz

sma-

Kin

gdom

of

Italy,

Kin

g H

ugh

of A

rles

(926

-931

), m

int

of V

enic

eH

ossz

úhát

-hal

om(K

om.

Cso

ngrá

d/H

)gr

ave

100

10/9

0M

FM 2

002.

18.7

85.8

20.

560.

1212

.37

tr.1.

14id

emK

ingd

om o

f Ita

ly, K

ing

Hug

h of

Arle

s (9

26-9

31),

min

t of

Ven

ice

10/9

1M

FM 2

002.

18.5

74.4

60.

54tr.

23.7

5tr.

1.19

idem

Kin

gdom

of

Italy,

Kin

g H

ugh

and

Loth

aire

II (

931-

947)

, m

int

of P

avia

10/9

2M

FM 2

002.

18.8

93.4

70.

77tr.

4.86

tr.0.

84id

emK

ingd

om o

f Ita

ly, K

ing

Loth

aire

II a

s so

le k

ing

(946

-950

), m

int

of

Mila

n10

/93

HO

M 9

4.46

.194

.68

1.16

tr.3.

53tr.

0.58

Kar

os-E

sper

jess

ögG

erm

any,

Kin

g O

tto

I-III

(962

-996

), m

int

of M

agde

burg

(?)

Grä

berf

eld

II(K

om.

Bors

od-A

baúj

-Ze

mpl

én/H

) gr

ave

5210

/94

MFM

55.

106.

197

.79

0.16

0.42

1.20

tr.0.

44Sz

eged

-Kirá

lyha

lom

Sam

anid

es,

Ism

a'il

b. A

hmad

, Ba

lkh

(293

)(R

ívói

erdő)

(Kom

. C

song

rád/

H)

10/9

5M

NM

59.

1.26

.A96

.97

0.24

tr.2.

06tr.

0.68

Sáro

spat

ak-B

aksa

hom

okSa

man

ides

, Is

ma'

il b.

Ahm

ad,

And

arab

a (2

95)

(Kom

. Bo

rsod

-Aba

új-

Zem

plén

/H)

grav

e 1

10/9

6M

NM

59.

1.27

.A99

.12

0.59

tr.0.

17tr.

tr.id

emSa

man

ides

, Is

ma'

il b.

Ahm

ad,

Sam

arka

nd (

288)

10/9

7M

NM

59.

2.8.

A97

.63

tr.0.

481.

52tr.

0.28

Sáro

spat

ak-B

aksa

hom

okSa

man

ides

, A

hmad

b.

Ism

a'il,

al-S

hash

(29

9)(K

om.

Bors

od-A

baúj

-Ze

mpl

én/H

) gr

ave

210

/98

MN

M 5

9.2.

10.A

98.3

20.

75tr.

0.64

tr.0.

25id

emSa

man

ides

, N

asr

b. A

hmad

, M

arw

(30

6)10

/99

MFM

55.

83.1

97.4

60.

670.

111.

47tr.

0.29

Kun

ágot

a (K

om.

Béké

s/H

)Ro

man

os I,

Kon

stan

tin V

II an

d St

epha

nos,

Mili

ares

ion

10/1

00M

FM 2

004.

3.2

98.1

91.

16tr.

0.38

tr.0.

19Sz

eged

-Cso

ngrá

di ú

tK

onst

antin

VII.

and

Rom

anos

II.,

Mili

ares

ion

BMC

67-

69(K

om.

Cso

ngrá

d/H

)(9

48-9

59)

grav

e 1

10/1

23RI

I.119

8895

.58

0.24

0.29

3.89

tr.tr.

Már

mar

os/U

A h

oard

Sam

anid

es,

Nas

r b.

Ahm

ad,

Balk

(31

4)10

/124

RII.1

2055

95.1

00.

240.

254.

42tr.

0.24

idem

Sam

anid

es,

Nas

r b.

Ahm

ad,

Sam

arka

nd (

320)

10/1

25RI

I.120

9496

.92

0.24

0.17

2.61

tr.tr.

idem

Sam

anid

es,

Nas

r b.

Ahm

ad,

al-S

hash

(32

1) (

fig

. 6)

10/1

26RI

I.121

1895

.79

0.21

0.33

3.21

tr.0.

47id

emba

rbar

ized

–al

-am

ir…10

/127

RII.1

2119

95.4

80.

260.

213.

74tr.

0.31

idem

barb

ariz

ed –

al-a

mir…

10/1

28RI

I.121

2195

.75

0.29

0.13

3.42

tr.0.

43id

emba

rbar

ized

–al

-am

ir…,

al-S

hash

(30

8)10

/133

MN

M 5

9.1.

21.A

98.4

40.

330.

160.

71tr.

0.37

Sáro

spat

ak-B

aksa

hom

okSa

man

iden

, N

asr

b. A

hmad

, Sa

mar

kand

(30

2)(K

om.

Bors

od-A

baúj

-Ze

mpl

én/H

) gr

ave

1

Tab

. 3C

oins

from

the

east

ern

Car

path

ian

Basi

n. T

he n

umis

mat

ic s

tudi

es o

f the

Sam

anid

coi

ns w

ere

done

by

Nik

olau

s Sc

hind

el (A

cade

my

of S

cien

ce, N

umis

mat

ic C

omm

issi

on, V

ienn

a), t

he w

este

rnco

ins

by H

uber

t Em

mer

ig (

Inst

itut

for

Num

ism

atic

s an

d M

onet

ary

His

tory

, V

ienn

a),

for

the

Byza

ntin

e co

in s

ee t

he a

rtic

le o

f P.

Pro

hász

ka in

thi

s vo

lum

e.

Going back further in time and looking at the Roman coinage, it can be stated that it was often debased

with copper alone48, without substantial additions of lead, tin, zinc or any other material. There are some

examples cited in the literature, but they remain scarce49.

Coin finds from early Hungarian contexts

Compared to the many examples of hoards detected in the countries surrounding the Baltic Sea, hoard finds

from early Hungarian contexts are rare. Beside hack-silver material, the more north-eastern hoards boast of

dirhams and to a lesser extent of English and German coins. From the 10th century onwards, Byzantine coins

become more numerous50. Coins from early Hungarian contexts are mainly confined to grave finds and coin

hoards. Most often they occur as perforated ornaments meant to decorate textiles, horse fittings, belts,

shoes or as a sort of obolus (fig. 6) 51. From literary evidence we know about the original provenance of

these coins, as they were mainly assembled from raids, as diplomatic gifts and tributes from German and

other western European territories. Coin emissions from Burgundy, Lorraine and North Italian cities are reg-

ularly identified among the grave goods, with comparably few German examples. This is astonishing, bear-

ing in mind that in 904 Bavaria, Saxony, Thurin gia, and Swabia suffered from numerous Hungarian raids,

with the consequence that a lot of silver coins changed hands52, but without leaving many traces among

the remains found in Hungarian graves. It can be supposed that part of the material was melted into ingots

before it was transported to the heartland.

Beside the English coins mentioned above, there are no coins from the 9th and 10th century reported show-

ing elevated zinc contents; however, copper-rich coins (analyses indicate values as high as 45 % Cu) were

struck, e. g., in Salzburg by the Duke of Bavaria in the period between 985 and 995 AD53.

Zinc-rich silver used for ornamental purposes

There is yet another important object type to investigate in order to elucidate how especially zinc-rich sil-

ver alloys were used and distributed. The evidence from jewellery and ornamental work is more numerous

than from coins or ingots, as we have already seen from the analytical data collected from the early Hun -

garian graves. The use of zinc in the form of brass as an addition to silver alloys is recorded for ornamen-

tal objects predating our 9th and 10th century finds.

C. Mortimer analysed Roman silver rings regarding their fineness and found appreciable amounts of zinc

from 1% to 5 %54. Ostrogothic brooches and belt buckles55 feature high zinc contents, but the objects

analysed from several other Roman silver hoards most often lack zinc-bearing silver items. An exception is

a set of rings and other cast items found in the Snettisham (Norfolk/GB) Roman treasure56 with up to 3.5%

zinc. Other pieces, such as a 12th century ring brooch from the hoard of Mörtträsket (Norrbotten coun-

ty/S) 57 that has been addressed as being of eastern origin, account for a late use of zinc-rich silver alloys.


Fig. 6 Dirhams are often recorded among early Hungarian findcontexts, such as this Samanid coin by Nasr b. Ahmad, al-Shash(321) (analysis see tab. 3 no. 10/125). – (Photo R. Müller, RGZM).

But there are many more examples that can be assigned to the 9th and 10th century, and it seems that mix-

ing copper alloy scrap metal into silver alloys became quite a popular procedure. The picture may be, how-

ever, incomplete, as there are only few published analyses of West and Central European silverware com-

pared to the more northern Viking silver finds.

One example is a Slavic sword58 from the settlement of Rostock-Dierkow (Mecklenburg-Western Pome -

rania/D) showing copper contents between 13% and 35% and zinc contents between 1.5% and 5.8 %.

A 10th century warrior’s belt from a Middle Volga burial ground also seems to be made from copper and

zinc-rich silver59. Among the objects from the Gnezdovo archaeological complex analysed intensively by

N. V. Enio sova60 are also some containing zinc, but the values do not seem to exceed 1.4 %. G. Nevizánsky

and J. Košta61 found that, among the objects from the early Hungarian cemetery of Streda nad Bodrogom

(Trebišov dist./SK, hung. Bodrogszerdahely), only few had zinc and high copper contents, with a maximum

value of 2.5 % zinc. The analysed silver objects from the National Historical Museum in Sofia do not fea-

ture any zinc-rich silver alloy62. Among the 55 analysed objects from the Bol’shiye Tigany (Rep. Tatarstan/

RU) burial ground at least half contain zinc between 1% and 5 %63.

CONCLUSIONS

Reviewing the results obtained from analysing several early Hungarian silver-based grave goods, it can be

stated that craftsmen of the 10th century were aware of how to prepare the ideal silver alloy for every pur-

pose. Copper-rich and in particular zinc-rich alloys were preferably used for making cast objects, while ham-

mered sheet metal, wires etc. were made from purer silver. Adding brass had the advantage of producing

an alloy appropriate for casting without blurring the bright silver colour. As long as the data basis for

analysed silver items from western territories remains comparably limited, there is no evidence that coins

with such a peculiar zinc-rich composition could have represented the main source for the silversmiths. Both

our own analytical results as well as data taken from the literature show that merely English coins have a

composition matching (theoretically) the zinc-rich Hungarian artefacts. Neither dirhams nor coins from

Burgundy, Bavaria, North Italian urban centres etc. show elevated zinc contents. The number of English

coins found in early Hungarian contexts, however, is too limited to account for a regular occurrence of

these alloys in finished ornamental objects. Zinc-rich silver alloys were made on purpose and with techno-

logical expertise, thus it is more a question of technology than of an increasing acceptance of debased sil-

ver by customers as argued in the literature64.

For copper-rich alloys without zinc the case might have been different. There were several relatively debased

coin issues from different territories that could have been transformed directly into an object without fur-

ther chemical interventions65. The contents of bismuth, on the other hand, might indicate the intensive use

of dirhams. Among these many are made from copper-rich silver. Beside contemporary coins among grave

goods and in hoards we often encounter much older emissions, so we have to consider a material that has

been minted long before the object in question was produced66.

The idea of making cast objects from copper-rich silver was not new. Cast objects from the Snettisham

Roman jeweller’s hoard, e. g., are more debased than those produced by hammering67.

The preferential use of zinc-rich silver for ornamental objects rather than for making coins is emphasised

further by the hoards containing pieces of hack-silver, such as in the find of Waterneversdorf (Schleswig-

Holstein/D)68. In the treatise »De diversis artibus« by Theophilus Presbyter, there are several recipes men-

tioning the use of scrap brass when making a silver object, such as in section 3069 (on casting the handle

of a chalice): »Do melt the silver immediately and add a small portion of Spanish brass [...]« or section 2370


(on purifying silver): »In case you see the silver start to boil and bubble, you can be sure that tin or brass

has been mixed into it«.

This 12th century text connects analytical results and technological object information in a very direct way

and proves how useful scientific analysis can be if used properly. Many questions, however, still remain

unanswered, such as a detailed regional distribution pattern of different alloy types. In this context we need

valid results, and it has to be stressed how important it is to use minimally invasive techniques that produce

fully quantitative analyses instead of semi-quantitative non-destructive techniques that only touch the sur-

face in the true sense of the word.

Acknowledgments


I would like to acknowledge the generosity of the followingmuseums and institutions, making available some of the mostsplendid objects of early Hungarian archaeology: Hungarian Natio-nal Museum (Budapest), Jósa András Múzeum (Nyíregyháza), Her-man Ottó Múzeum (Miskolc), Ponitrianske múzeum and SlovakAcademy of Sciences (Nitra), Hungarian Academy of Sciences

(Budapest) and Móra Ferenc Múzeum (Szeged). For their help I amgrateful to my colleagues Dr. László Révész, Dr. Miklós Takács, Dr.Matej Ruttkay, Gábor Lőrinczy and Dr. Attila Türk. For much of theanalytical measurements I am indebted to my colleague SonngardHartmann. I gratefully thank Dr. Bendeguz Tobias and Dr. FlorianStröbele for providing me with information on useful publications.

APPENDIX

Analytical details

Micro-X-Ray-Fluorescence (μ-XRF)

μ-XRF allows for the qualitative and quantitative determination of most of the elements (heavier than sodi-

um) in a sample. Depending on the sample material, the method is non-destructive or nearly non-destruc-

tive. Only corrosion layers have to be removed in the area to be investigated. Due to a small analytical spot

size (0.3 mm), this injury is minimised. The small diameter of the primary beam allows the analysis of tiny

structures. During analysis the sample is exposed to X-ray radiation via a thin glass fibre capillary. This pri-

mary radiation excites a secondary radiation from within the sample material, the so-called X-ray fluores-

cence. Each element present in the sample emits X-rays of a characteristic energy. The intensity depends

on the concentration of the element in the sample. A detector analyses energy and intensity of the emit-

ted fluorescence radiation and displays the series of intensity maxima in a spectrum. By comparing this

spectrum with calibration samples of a known composition, the exact quantitative composition of the

object can be calculated for element contents of more than 0.01 weight per cent.

Sample preparation and measurement

The alloys were analysed after removing any possible corroded surface with a small scalpel tip or a micro

drill in order to obtain results from the bulk fresh metal. The fire gilding was left untouched because cor-

rosion phenomena are restricted to the very surface of gold alloys. In general, three points on the same

object were examined for each analysis; the mean values are presented in the tables.

Instrumentation and measurement details

Eagle III XXL, Roenalytic GmbH, Taunusstein (Germany)

Rhodium Tube, max. 40 kV, 1 mA, Oxford Instruments

Si(Li) detector, EDAX, resolution 148 eV for MnK

sample chamber: 75 x 75 x 135 cm

X-ray optics: monocapillary, 0.3 mm focus (analysed spot size)

EDAX analytical system cooled with liquid nitrogen

Analytical parameters

atmosphere: air

tube voltage: 40kV

tube current: 300μA; coins from Szeged-Kiskundorozsma were analysed with a

Ti25 filter

acquisition time: 200-500 Lsec

amplification time: 35μs

quantification: combination of fundamental parameters and standard aided method

with calibrated results recalculated to 100 %; standard reference

materials: RAgA1, RAgA3, RAgA5 (certified by Rand Refinery, South

Africa) and 207, 214 non-certified dental alloys

For inter-laboratory comparison, the results for the international reference silver standard RAgA1 are as

follows: Ag certified 97.02 wt %; Ag measured 96.96±0.03; Cu certified 3.030 wt %; Cu measured

3.04±0.03.

Detection limit for Fe, Ni, Cu, and Zn is ca. 0.2 % and ca. 0.1% for Pb, Au and Bi.

Notes


1) »Heldengrab im Niemandsland – ein frühungarischer Reiteraus Niederösterreich«, September 14th-November 19th 2006,Römisch-Germanisches Zentralmuseum Mainz (see Daim2006).

2) Greiff 2011; Lőrinczy / Türk 2011, 438-440. The followingstudy is part of a joint project between the Römisch-Germani-sches Zentralmuseum and the Móra Ferenc Múzeum (Szeged)titled »Coins and prosperity: Multidisciplinary studies on richfemale burials dating to the 10th century in the CarpathianBasin«. The scientific investigation in the archaeometry labora-tory of the RGZM were performed on coins from importantearly Hungarian graves such as four of the most prominentfemale burials boasting with silver items, but also on otherHungarian 10th century grave finds (Tab. 3).

3) Langó / Türk 2004, 450-455.

4) The Gnadendorf objects were partly analysed by MathiasMehofer from the Vienna Institute of Archaeological Sciences

(VIAS) and partly by the author. Analytical tables are presentedelsewhere (Mehofer / Greiff 2006) and will not be reproducedhere. They are, nevertheless, part of binary diagrams discussedin the present paper.

5) These analyses have been performed in the RGZM archaeome-try laboratory by Dipl.-Ing. Sonngard Hartmann.

6) Analytical details can be found in the appendix.

7) See note 4.

8) At least for concentrations exceeding those discussed belowas natural or which have been introduced during the primaryprocesses of smelting and recovery (cf. Greiff / Mehofer /Révész 2007).

9) The zinc content of ancient brass alloys may vary considerably,from a few per cent up to ca. 30 %.

10) Mehofer / Greiff 2006, 183 tab. 1.


11) Nitra, Ponitrianske Múzeum and Academy of Science dating:end of 9th/beginning of 10th century.

12) Such as a gilded copper object.

13) Brepohl 1994, 49.

14) The database of the RGZM archaeometry laboratory comprisesanalyses of about 300 silver objects from different periods andgeographical sources.

15) Wanhill 2002, 20; 2003, 4.

16) A description of the process can be found in Craddock 1995,221-231.

17) Wolters 1981, 63.

18) Mehofer / Greiff 2006, 183 tab. 1 no. 29 with 10 % Sn; or thequiver from Karos cf. tab. 1.

19) Steuer 1998, 121; Craddock 1995, 221.

20) McKerrell / Stevenson 1972, 198.

21) Ibidem 209, tab. V; Steuer / Stern / Goldenberg 2002, 151 fig.10; N. V. Eniosova in this volume.

22) Craddock 1995, 221.

23) Unpublished diploma thesis of A. Dorn, Tübingen 2008, page156; I thank Dr. Florian Ströbele for making available a copy ofthe thesis.

24) Sterner-Rainer 1930, 94.

25) »Legierungen mit 2 bis 5 % Zink verwendet man gerne alsGussmetalle [...]« (ibidem).

26) Ibidem 96.

27) Vickers hardness values are obtained by a device with a minutediamond crystal fixed to an indenter that is pressed onto themetal surface. The diameter of the resulting impression ismeasured and converted into hardness values. The softer thematerial, the deeper the indenter penetrates, thus producinga wider sized mark.

28) Wolters 1981, 63.

29) Steuer 1998, 111.

30) Steuer 1998; 2004; Brather 2007.

31) Steuer 2004, 138 f.

32) Illisch et al. 2003; for this copious study, the objects were ana-lysed on the surface without preparation, with all drawbacksresulting from this procedure.

33) A coin issued at Bukhara (287 AH = 900 CE), one from Anda-rabah (296 AH = 908 CE), one from Khust (760 AH = 1358CE), one from Balkh (960 AH = 1553 CE), one from Ghazna(411 AH = 1020 CE) and one Iranian coin (310 AH = 922 CE)imitating a coin issued at the Baghdad mint.

34) Four objects from a hoard find from Vastäde/S (t.p.q. 1075/ 1100), one object from a hoard find from Leissow/PL (t.p.q.beginning of the 11th century), seven objects from List/D (t.p.q.1000/1003), nine objects from Pöschendorf/D (t.p.q. 790/ 794), five objects from Wangels/D (t.p.q. 1038/1045), onepiece of hack-silver from Waterneversdorf/D (t.p.q. 976/1009)and two objects from Westerland/D (t.p.q. 1040/ 1042); theonly item that can be counted to the 10th century is the Otto-Adelheid-Pfennig that started to be coined either in 983 or in991 (see discussion in Leimus 2005, 1206).

35) Among these, there is one Otto-Adelheid-Pfennig Hatz type 5.

36) Steuer 2004, 125.

37) Metcalf / Northover 1985, 150 f.; McKerrell / Stevenson 1972,200 fig. 2; 201 fig. 3.

38) Metcalf / Northover 1985, 160.

39) Ibidem 162.

40) Wolfgang Hahn mentions an Anglo-Saxon penny issued underKing Æthelstan (924-939) in his review of László Kovács’»Münzen aus der ungarischen Landnahmezeit« (Arch. Austri-aca 75, 1991, 316); Kovács mentions among the 748 coinswith secure identification four English coins.

41) Rudolph of Burgundy (922-926), Hugh of Provence (926-925)and Hugh of Provence/Lothar II (931-947).

42) See note 35.

43) Arrhenius / Linder Welin / Tapper 1972/73, 154.

44) Kruse / Tate 1992, 320-322 tab. 1.

45) Bompaire / Guerra 2000, annexe; analyses 881ff.

46) Ibidem 878.

47) I have to thank Ernst Pernicka for making available his contri-bution published in connection with Hatz 2001, 95 f.

48) Butcher / Ponting 1998, 321-325 tabs 1-5.

49) cf. examples cited by Mortimer 1986, 240.

50) Buko 2007, 438.

51) Révész 2006, 134.

52) Ibidem 135.

53) Gresits / Gedai 2000, 905.

54) Mortimer 1986, 236 tab. 1.

55) Riederer 1976, 233 tab. 1; 236 tab. 2; LaNiece / Cowell 2008,155 tab. 2.

56) Pike et al. 1997, 54 tab. 5.

57) In southern Lappland near Lycksele/S; Zachrisson 1984, 56tab. 8.

58) Richter 1992 / 93, 241.

59) Saprykina / Zelentsova / Mitoyan 2009, 4 tab. 1; high valuesof iron reveal that the analyses were performed on non-prepa-red object surfaces so that the results are merely semi-quanti-tative.

60) N. V. Eniosova in this volume, 272.

61) G. Nevizánsky and J. Košta in this volume, tab. 1; all analyseswere carried out on restored but not prepared surfaces.

62) M. Inkova in this volume, tab. 1.

63) Valiullina / Khramchenkova 2001, 273 f. tab. 1.

64) Mortimer 1986, 241.

65) In certain cases, even a direct relationship between silver alloysfrom coins and objects from neighbouring tombs can bededuced from detailed analyses, such as for coin no. 7 fromSzeged-Kiskundorozsma (Greiff 2011, 489 tab. 1).

66) In the inventory of some graves from Szeged-Kiskundorozsma-Hosszúhát-halom, e. g., we encounter several coins from the9th or 10th century (Bende / Lőrinczy / Türk 2002, 367)..

67) Pike et al. 1997, 54.

68) See note 33.

69) Brepohl 1987, 102 f.

69) Ibidem 85.

References


Arrhenius / Linder Welin / Tapper 1972/73: B. Arrhenius / U. S. Lin -der Welin / L. Tapper, Arabiskt silver och nordiska vikingasmyk-ken. Tor 15, 1972 / 73, 151-160.

Bende / Lőrinczy / Türk 2002: L. Bende / G. Lőrinczy / A. Türk, Hon -foglalás kori temetkezés Kiskundorozsma-Hosszúhát-halomról.(Eine landnahmezeitliche Bestattung von Kiskundorozsma-Hoss-zúhát-Hügel). Móra Ferenc Múz. Évk. Studia Arch. 8, 2002, 351-402.

Bompaire / Guerra 2000: M. Bompaire / M. Guerra, Analyses demonnaies françaises du XIe siècle, le problème du zinc. In: Kluge /Weisser 2000, 876-884.

Brather 2007: S. Brather, Counted and Weighed Silver: the Frag-mentation of Coins in Early Medieval East Central Europe. In:Henning 2007, 451-471.

Brepohl 1987: E. Brepohl, Theophilus Presbyter und die mittelalter-liche Goldschmiedekunst (Wien, Köln, Graz 1987).

1994: E. Brepohl, Theorie und Praxis des Goldschmieds (Leipzig,Köln 1994).

Buko 2007: A. Buko, »Tribal« Societies and the Rise of the EarlyMedieval Trade: Archaeological Evidence from Polish Territories(Eight-Tenth Centuries). In: Henning 2007, 431-450.

Butcher / Ponting 1998: K. Butcher / M. Ponting, Atomic Absorp-tion Spectrometry and Roman Silver Coins. In: A. Oddy / M.Cowell (eds), Metallurgy in Numismatics 4. Royal Num. Soc. Spe-cial Publ. 30 (London 1998) 308-334.

Craddock 1995: P. Craddock, Early Metal Mining and Production(Cambridge 1995).

Daim 2006: F. Daim (ed.), Heldengrab im Niemandsland. Ein frü-hungarischer Reiter aus Niederösterreich. Mosaiksteine. Forsch.RGZM 2 (Mainz 2006).

Daim / Lauermann 2006: F. Daim / E. Lauermann (eds), Das frühun-garische Reitergrab von Gnadendorf (Niederösterreich). Monogr.RGZM 64 (Mainz 2006).

Greiff 2011: S. Greiff, A Szeged-Kiskundorozsma, hosszúhátiezüstleletek ötvözéstechnikai vizsgálata. Adatok a 10. századifém mellékletek és ezüstpénzek lehetséges összefüggéseiről (Sil-berfunde aus Szeged-Kiskundorozsma, Hosszúhát. Eine legie-rungstechnische Diskussion über den möglichen Zusammen-hang zwischen Schmuckwaren und Münzsilber im 10. Jahrhun-dert). Móra Ferenc Múz. Évk. Studia Arch. 11, 2011, 481-491.

Greiff / Mehofer / Révész 2007: S. Greiff / M. Mehofer / L. Révész,Gezielte Nutzung zinkreicher Silberlegierungen an frühungari-schen Silberfunden. Archäometrie und Denkmalpflege – Kurzbe-richte 2007, 60-62.

Gresits / Gedai 2000: I. Gresits / I. Gedai, Chemische Zusammen-setzung von Münzen des Frühmittelalters. In: Kluge / Weisser2000, 904 f.

Hatz 2001: G. Hatz, Der Münzfund vom Goting-Kliff/Föhr (miteinem Beitrag von E. Pernicka). Num. Stud. 14, 2001, 95-109.

Henning 2007: J. Henning (ed.), Post-Roman Towns, Trade andSettlement in Europe and Byzantium 1. The Heirs of the RomanWest (Berlin, New York 2007).

Illisch et al. 2003: L. Illisch / M. Matzke / F. Schwarz / W. B. Stern /H. Steuer, Dirham und Rappenpfennig. Mittelalterliche Münz-prägung in Bergbauregionen. Zeitschr. Arch. Mittelalter Beih.17(Bonn 2003).

Kruse / Tate 1992: S. Kruse / J. Tate, XRF Analysis of Viking Age Sil-ver Ingots. Proc. Soc. Ant. Scotland 122, 1992, 295-328.

Kluge / Weisser 2000: B. Kulge / B. Weisser (eds), Akten des XII.Internationalen Numismatischen Kongresses Berlin 1997 (Berlin2000).

Langó / Türk 2004: P. Langó / A. Türk, Honfoglalás kori sírok Mind -szent-Koszorús-dűlőn. (Adatok a szíjbefűzős bizánci csatok és aDélkelet-európai kapcsolatú egyszerű mellkeresztek tipológiájá-hoz). (Landnahmezeitliche Gräber in Mindszent-Koszorús-dűlő[Angaben zur Typologie der trapezförmigen byzantinischenSchnal len und einfachen Kreuzanhänger mit Südosteuropäi-schen Beziehungen]). Móra Ferenc Múz. Évk. Studia Arch. 10,2004, 365-457.

LaNiece / Cowell 2008: S. LaNiece / M. Cowell, Crimean Metal-work: Analysis and Technical Examination. In: D. Kydd / B. Ager(eds), The Berthier-Delagarde Collection of Crimean Jewellery inthe British Museum and Related Materials (London 2008) 151-160.

Leimus 2005: I. Leimus, Crux, Köln Häv. 34 / 67 und Otto-Adel-heid-Pfennige. Ihr Vorkommen in den Funden aus dem Ende des10. Jh. In: C. Alfaro / C. Marco / P. Otero (eds), XIII CongresoInternational de Numismática, Madrid 2003 2 (Madrid 2005)1205-1215.

Lőrinczy / Türk 2011: G. Lőrinczy / A. Türk, 10. századi temető Sze-ged-Kiskundorozsma, Hosszúhátról. Újabb adatok a Maros-tor-kolat Duna–Tisza közi oldalának 10. századi településtörténeté-hez. (Gräberfeld des 10. Jhs. in Szeged-Kiskundorozsma, Hosszú-hát. Neue Ergebnisse zur Siedlungsgeschichte des 10. Jhs. derRegion zwischen Donau und Theiß gegenüber der Maros-Mün-dung). Móra Ferenc Múz. Évk. Studia Arch. 12, 2011, 419-479.

McKerrell / Stevenson 1972: H. McKerrell / R. B. K. Stevenson, SomeAnalyses of Anglo-Saxon and Associated Oriental Silver Coinage.In: E. T. Hall / D. M. Metcalf (eds), Methods of Chemical andMetallurgical Investigation of Ancient Coinage. Royal Num. Soc.Special Publ. 8 (London 1972) 195-209.

Mehofer / Greiff 2006: M. Mehofer / S. Greiff, ArchäometrischeUntersuchungen an Metallgegenständen. In: Daim / Lauermann2006, 181-188.

Metcalf / Northover 1985: D. M. Metcalf / J. P. Northover, Debase-ment of the Coinage in Southern England in the Age of KingAlfred. Num. Chronicle 145, 1985, 150-176.

Mortimer 1986: C. Mortimer, Early Use of Brass in Silver Alloys.Oxford Journal Arch. 5, 1986, 233-242.

Pike et al. 1997: A. Pike / M. Cowell / J. Lang / C. Cartwright, TheScientific Examination of the Hoard. In: C. Jones, The SnettishamRoman Jeweller’s Hoard (London 1997) 50-68.

Révész 2006: L. Révész, Auswertung der Funde. In: Daim / Lauer-mann 2006, 119-158.

Richter 1992/93: E.-L. Richter, Der Hort eines Edelmetallschmiedesaus der frühslawischen Siedlung Rostock-Dierkow, analytischeUntersuchung der Schwertgefäßteile. Offa 49/50, 1992/93,241f.

Riederer 1976: J. Riederer, Metallurgische Untersuchung ostgoti-schen Trachtzubehörs. In: V. Bierbrauer, Die ostgotischen Grab-und Schatzfunde in Italien (Spoleto 1976) 231-238.

Saprykina / Zelentsova / Mitoyan 2009: I. Saprykina / O. Zelent so -va / R. Mitoyan, Warrior’s Belt from the Middle Volga BurialGround X A.D. – Technology and Origin. In: 2nd Latin-American


Symposium on Physical and Chemical Methods in Archaeology,Art and Cultural Heritage Conservation and Archaeological andArt Issues in Materials Science (Cancún 2009) 1-6.

Sterner-Rainer 1930: L. Sterner-Rainer, Die Edelmetall-Legierungenin Industrie und Gewerbe (Leipzig 1930).

Steuer 1998: H. Steuer, The Ore Deposits of Middle Asia and VikingAge Silver in Scandinavia: Introduction. Hammaburg NF 12,1998, 111-124.

2004: H. Steuer, Münzprägung, Silberströme und Bergbau um dasJahr 1000 in Europa – wirtschaftlicher Aufbruch und technischeInnovation. In: A. Hubel / B. Schneidmüller (eds), Aufbruch inszweite Jahrtausend. Innovation und Kontinuität in der Mitte des Mittelalters. Mittelalter-Forsch. 16 (Stuttgart 2004) 117-149.

Steuer / Stern / Goldenberg 2002: H. Steuer / W. B. Stern / G. Gol-denberg, Der Wechsel von der Münzgeld- zur Gewichtswirt-schaft in Haithabu um 900 und die Herkunft des Münzsilbers im9. und 10. Jh. In: K. Brandt / M. Müller-Wille / Ch. Radtke (eds),

Haithabu und die frühe Stadtentwicklung im nördlichen Europa.Schr. Arch. Landesmus. 8 (Neumünster 2002) 133-167.

Valiullina / Khramchenkova 2001: S. I. Valiullina / R. Kh. Khram -chen kova, The Chemical Composition of the Articles of Non-Fer-rous Metals of the Bolshe-Tiganskiy Burial Ground. In: V. Zavya-lov / A. Ivanov (eds), The Ancient Craftsmen of Ural (Ishevsk2001) 271-275.

Wanhill 2002: R. J. H. Wanhill, Archaeological Silver Embrittlement:a Metallurgical Inquiry. National Aerospace Laboratory NLRReport NLR-TP-2002-224 (Amsterdam 2002).

2003: R. J. H. Wanhill, Ancient Silver Embrittlement: Significancesof Copper, Lead and Cold-Deformation. National AerospaceLaboratory NLR Report NLR-TP-2003-617 (Amsterdam 2003).

Wolters 1981: J. Wolters, Der Gold- und Silberschmied 1 (Stuttgart1981).

Zachrisson 1984: I. Zachrisson, De Samiska Metalldepåerna år1000-1350. Archaeology and Environment 3 (Umeå 1984).

Abstract / Zusammenfassung

Silver grave goods from early Hungarian contexts: technological implications of debased alloy compositionswith zinc, tin and leadThe paper describes the results from a programme of alloy analyses of early Hungarian silver performed at thearchaeometry laboratory of the Römisch-Germanisches Zentralmuseum in Mainz. A selection of 9th and 10th centurysilver ornaments and coins from different archaeological complexes were analysed by minimal-invasive Micro-XRF.Some ornaments turned out to be made of unusual zinc-rich silver alloys. The objects were studied in the context ofsilver alloy composition in relation to manufacturing techniques and the silver-based monetary system of that period.The results are compared to other analyses of contemporary finds. As a general result different groups of alloys couldbe distinguished and the work demonstrated significant correlations between alloy composition and manufacturingtypology. Besides this, origin and function of different alloying elements observed in the context of ancient silver alloysare reviewed. In addition, the development of zinc-rich debased silver alloys was studied based on published data fromother periods and geographic sources. The composition of early mediaeval silver coins is discussed as a possible sourcematerial for Hungarian silver objects.

Untersuchung frühungarischer Fundobjekte aus Silber: technologische Aspekte von Silberlegierungen mit hohen Gehalten an Zink, Zinn und BleiAm Archäometrielabor des Römisch-Germanischen Zentralmuseums in Mainz wurden in den letzten Jahren systema-tisch frühungarische Silberobjekte des 9. und 10. Jahrhunderts mithilfe der Mikro-Röntgenfluoreszenzanalyse unter-sucht. Einige der hier vorgestellten Schmuck waren und Münzen zeigen erstaunlich hohe Gehalte an Zink. Die Legie -rungen werden in Zusammenhang mit den für die Objekte angewandten Herstellungstechniken diskutiert und auchentsprechenden Daten aus der Literatur gegenübergestellt. Es zeigt sich, dass es eine enge Korrelation zwischen guss -technisch hergestellten Zierwaren und der Verwendung zinkreicher Silberlegierungen gibt. Darüber hinaus werden hierUrsprung und Wirkungsweise anderer Legierungselemente zusammenfassend dargestellt. Die Verwendung zinkreicherSilberlegierungen in verschiedenen zeitlichen und regionalen Kontexten wird ebenso angesprochen wie die möglicheVerwendung von Münzen als Aus gangsmaterial für die Herstellung dieser Legierungen.

Monographien des RGZMBand 64 (2. Auflage 2006)

315 S. mit 197 Abb., 12 Farbtaf.ISBN 3-88467-094-8

48,– €

Falko Daim · Ernst Lauermann (Hrsg.)

Das frühungarische Reitergrab von Gnadendorf (Niederösterreich)Das Reitergrab von Gnadendorf ist aus verschiedenen Gründen außer -gewöhnlich. Zu nächst wurde es außerhalb des damaligen ungarischen Sied -lungsgebietes angelegt, weiters handelt es sich bei dem Bestatteten umeinen 14-jährigen, kampf erfahrenen Jungen, und drittens verfügt das Grabüber eine vorzügliche Ausstattung. Das Grab wirft einige grund legende Fra-gen auf, denn sämtliche Fundge genstände scheinen lange in Ge brauch ge -wesen zu sein. Außerdem haben zwei 14C-Datierungen einen Bestattungs-zeitpunkt erst um das Jahr 1000 ergeben. Treffen die natur wissenschaft -lichen Datie rungen zu, stellt sich die Frage, warum man den Knaben weitweg von den ungarischen Siedlungen mit wertvollen, aber teils sehr altenSachen bestattet hat. Bedenkt man, dass der ungarische Stämme bund umdie Jahrtausendwende die Umstruktu rie rung zu einem »modernen« mittel-alterlichen Staat auf christlichen Grund lagen erlebte, könnte es sein, dassdie Bestattung von Gnadendorf als Demonstration gegen diese Verände-rung gedacht war.Das vorliegende Buch enthält neben einer detaillierten Fundvorlage zahl -reiche Studien, die »den Fall Gnadendorf« aus unterschiedlichen Perspek -tiven beleuchten.

AUS DEM VERLAGSPROGRAMM

Mosaiksteine. Forschungen am RGZMBand 2 (2., verbesserte Auflage 2007)

68 S. mit 49 Farb- u. 29 sw-Abb.ISBN 3-88467-101-4

16,50 €

Falko Daim (Hrsg.)

Heldengrab im NiemandslandEin frühungarischer Reiter aus Niederösterreich

»Heldengrab im Niemandsland« erscheint anlässlich der gleichnamigen Ka -binettausstellung des RGZM im Kurfürstlichen Schloß Mainz (14. Septem-ber bis 19. November 2006). Das aufwändig ausgestattete Werk fasst inmehreren Bei trägen die Forschungsergebnisse zum Grab von Gnadendorfsowie zum historisch-archäologischen Umfeld zusammen. Ein umfassenderArtikel von Mecht hild Schulze-Dörrlamm thematisiert darüber hinausge-hend die archäologischen Belege für die frühungarischen Raubzüge in derersten Hälfte des 10. Jahrhunderts. Die lange Zeit fast unbesiegbaren Reiterge langten bis nach Oberitalien, an die Atlantikküste und die heutige däni-sche Grenze, bis sie 955 vom Heeresaufgebot König Ottos I. bei Augsburgvernichtend geschlagen werden konnten.

Verlag des Römisch-Germanischen Zentralmuseums, MainzErnst-Ludwig-Platz 2 · 55116 Mainz · Tel.: 0 6131/ 91 24-0 · Fax: 0 6131/ 91 24-199E-Mail: [email protected] · Internet: www.rgzm.de · www.shop.rgzm.de


Hajanalka Herold

Zillingtal (Burgenland)Die Awarenzeitliche Siedlung und die Keramikfundedes Gräberfeldes

Die Bearbeitung der frühmittelalterlichen Siedlung (7.-8. Jahrhundert n.Chr.)sowie der Keramikfunde des zugehörigen Gräberfeldes konzentriert sich aufdrei Schwerpunkte: awarenzeitliche Siedlungsbefunde und Siedlungsstruk-turen im Karpatenbecken, Keramikproduktion und Keramik gebrauch in derAwarenzeit sowie awarenzeitliche Traditionen in Zillingtal bei der Beigabevon Keramikgefäßen ins Grab.Bei den Siedlungsbefunden interessiert vor allem die frühmittelalterlicheWie derverwendung der römischen Ruinen. Die Auswertung des Fund mate -rials konzentriert sich auf die Keramikfunde, mit denen zusammen auch dieKeramikgefäße des awarenzeitlichen Gräberfeldes untersucht werden. Dazudienen archäologische und archäometrische Analysen sowie Methoden derexperimentellen Archäologie. Die gewonnene Chronologie der Grabgefäßeund die anthropologischen Daten der Bestatteten bilden die Basis für dieAnalyse der awarenzeitlichen Traditionen bei der Beigabe von Keramikgefä-ßen in die Gräber.

Monographien des RGZM, Band 80,1-22 Bände, zus. 438 S., 120 Abb.,

240 Farbtaf., 4 Beil.ISBN 978-3-88467-133-7

272,– €


RGZM – Tagungen, Band 131. Auflage 2012, 262 S.mit 127 z. T. farb. Abb.

ISBN 978-3-88467-191-737,– €

Lutz Grunwald · Heidi Pantermehl · Rainer Schreg (Hrsg.)

Hochmittelalterliche Keramik am RheinEine Quelle für Produktion und Alltag des 9. bis 12. Jahrhunderts

Durch die Tagung »Hochmittelalterliche Keramik am Rhein« gelang es, fürdas 9. bis 12. Jahrhundert eine Bilanz des derzeitigen Forschungsstands zudiesem »Leitfossil« der archäologischen Wissenschaft zu ziehen. Der vor -liegende Band bietet mit seinen 21 Beiträgen nicht nur einen wichtigenÜberblick über den aktuellen Forschungsstand zur entlang des Rheins anzu-treffenden hochmittelalterlichen Keramik. Ausgehend von den unterschied -lichsten, in der Schweiz, Frankreich, Deutschland und den Nieder landenangesiedelten Forschungsvorhaben erweitert er darüber hinaus den Blickvon einzelnen Fundstellen und Töpferregionen auf überregionale Betrach-tungen und Zusammenhänge hinsichtlich der Warenarten, ihrer Produktionund des Handels mit keramischen Gütern. Einige Beiträge liefern für be -stimmte Regionen am Rhein zudem erstmals eine Beschreibung der dort indieser Zeit vorhandenen Tonwaren. In der Zusammenschau der Einzeldar-stellungen ergeben sich neue Einblicke sowohl in die regionale Wirtschafts-geschichte als auch in die großräumigen Entwicklungstendenzen, die in die-ser Epoche das Leben und den Alltag der Menschen entlang des Rheinsprägten.



Monographien des RGZM, Band 92268 S. mit 270 meist farbigen Abb.ISBN 978-3-88467-172-6 (RGZM)

76,– €

Monographien des RGZM, Band 98288 S. mit 89 Abb., 32 Taf.

ISBN 978-3-88467-188-7 (RGZM)72,– €

Ljudmila Pekarska

Jewellery of Princely KievThe Kiev Hoards in the British Museum and TheMetropolitan Museum of Art and Related Material

In the capital of Kievan Rus’, princely Kiev, almost 70 medieval hoards havebeen discovered to date. The hoards contained gold and silver jewellery ofthe ruling dynasty, nobility and the Christian Church. They were unique toKiev and their quantity and magnificence of style cannot be matched by any-thing found either in any other former city of Rus’, or in Byzantium. Most ofthe objects never had been published outside the former Soviet Union.During the 17th-20th centuries, many medieval hoards were gradually un -earthed; some disappeared soon after they were found. This book providesa complete picture of the three largest medieval hoards discovered in Kiev:in 1906, 1842 and 1824, and traces the history and whereabouts of otherlost treasures. Other treasures took pride of place in some of the world’stop museums.This publication highlights the splendid heritage of medieval Kievan jew-ellery. It illustrates not only the high level of art and jewellery craftsmanshipin the capital, but also the extraordinary religious, political, cultural andsocial development of Kievan Rus’, the largest and most powerful EastSlavic state in medieval Europe.

Aleksandr I. Ajbabin

Archäologie und Geschichte der Krimim FrühmittelalterObwohl die Archäologie und Geschichte der byzantinischen Krim ein gutuntersuchtes Thema ist, wurden die Forschungsergebnisse jenseits des rus-sischen Sprachraums nur schwach rezipiert. Die hier vorgelegte Monographie des international renommierten Archäo-logen Aleksandr I. Ajbabin, die aus einem gemeinsamen Projekt des RGZMund der Ukrainischen Akademie der Wissenschaften hervorgegangen ist,soll dabei helfen, diesen wesentlich vom Spannungsverhältnis von Steppen -völkern und Byzantinischem Reich geprägten Raum neu und verstärktwahr zu nehmen. Die gründlich überarbeitete und erweiterte Übersetzung des erstmals 1999in russischer Sprache erschienenen Werkes präsentiert dem deutschenPublikum eine umfassende Übersicht über das teilweise schwer zugänglichpublizierte Fundmaterial und seine Chronologie.

GRAVE GOODS FROM EARLY HUNGARIAN CONTEXTS: TECHNOLOGICAL IMPLICATIONS OF DEBASED ALLOY COMPOSITIONS WITH ZINC, TIN AND LEAD

Documents